Datasets:

chemNLP

/

ncbi-text

like 0

Invasive fungal infections (IFI) have significantly increased due to advances in medical care in the at risk immunocompromised population. Fungal species are widely distributed in soil, plant debris and other organic substrates, and make up approximately 7 per cent (611,000 species) of all eukaryotic species on earth, although only about 600 species are human pathogens. Major risk factors for IFI include neutropenia <500 neutrophils/ml for more than 10 days, haematological malignancies, bone marrow transplantation, prolonged (>4 wk) treatment with corticosteroids; prolonged (>7 days) stays in intensive care, chemotherapy, HIV infection, invasive medical procedures, and the newer immune suppressive agents. Other risk factors are malnutrition, solid organ transplantation, severe burns or prolonged stays in intensive care (>21 days), systemic corticosteroids for >7 days, and major surgery. There are also reports of the presence of infection in immunocompetent patients without signs or symptoms of conditions associated with immunocompromised status. Infection can be transmitted by the inhalation of spores (aspergillosis, cryptococcosis, histoplasmosis), percutaneous inoculation in cutaneous and subcutaneous infections (dermatophytosis, madura foot), penetration into the mucosa by commensal organisms such as , and the ingestion of a toxin in contaminated food or drink (gastrointestinal disease). Infections may be mild and only superficial or cutaneous (. dermatophytosis and ) or may cause life-threatening, systemic illness (. candidiasis, aspergillosis and mucormycosis). The clinical manifestations of the disease caused by a given fungal agent can be highly variable and related to host immunity and physiological condition. For example, spp. can invade a local site (mucocutaneous or cutaneous candidiasis, onychomycosis) or cause systemic infections (renal, liver abscess, lung and nervous central system). Allergic symptoms were reported in infections with other fungi such as spp. (allergic bronchopulmonary aspergillosis). The isolation of these organisms from clinical samples may indicate colonization, infection or disease; consequently interpreting the results of diagnostic tests may be challenging for clinicians who treat these patients. Treatment requires early suspicion and is difficult because only a few antifungal agents are available, most usually have side effects, and some organisms have developed resistance. Clinicians need drugs that are highly effective but have low toxicity. In this review opportunistic fungal infections, diagnostic methods and the management of these infections are discussed. The standard definition of IFI was developed by members of the European Organization for Research in the Treatment of Cancer–Invasive Fungal Infection Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group. Among immunocompetent hosts, keratitis and onychomycosis are the most common infections. Other infections in immunocompetent patients include sinusitis, pneumonia, thrombophlebitis, peritonitis, fungemia, endophtalmitis, septic arthritis, vulvovaginitis and osteomyelitis. In immunocompromised patients, any fungus present in the environment may be potentially pathogenic. and spp. are the main organisms isolated most frequently from immunocompromised patients. The other most relevant aetiologic agents are spp., spp., , fungi and opportunistic yeast-like fungi. Colonization of the mucosal surfaces by spp. is the first step in the process of systemic candidiasis. is isolated from 84-88 per cent of mucocutaneous surfaces in hospitalized patients or even healthy adults. Because of its stronger adherence capacity, is found more often than other species. Central nervous system candidiasis may present with scattered brain microabscesses or larger lesions, vasculitis or thrombosis. species are reportedly the third or fourth most frequently isolated organisms in nosocomial bloodstream infections, the fourth most common cause of hospital-acquired systemic infections in the USA in nosocomial candidemia, and the fifth most common cause of bloodstream infections in paediatric intensive care units. The crude mortality rates in the USA were reported to be up to 50 per cent. There are many species of in the environment but the most frequently recovered species depending on the geographical region are , , or . Clinical manifestations can range from colonization in allergic bronchopulmonary disease to active disseminated infection, with mortality rates ranging from 40 to 90 per cent depending on the site of infection, host immune status, and the treatment regimen. The crude mortality rate has been reported to be as high as 95 per cent in bone marrow transplant recipients. Allergic bronchopulmonary aspergillosis (ABPA) is a hypersensitivity reaction to mycelia, which colonize the bronchi with a prevalence rate of 1 to 2 per cent in people with asthma, 7 to 14 per cent in steroid-dependent asthma patients, and 2 to 15 per cent in patients with cystic fibrosis. Other clinical manifestations of aspergillosis are invasive pulmonary aspergillosis, cutaneous and wound infections, keratitis, and sinusitis. is a major cause of infection in immunocompromised patients. Primary infections due to are common and most are asymptomatic in the lung. can cause a severe form of meningitis and meningo-encephalitis in patients with AIDS. Disseminated cryptococcal infection can cause clinical manifestations in the skin, ocular, soft tissue or bones and joints. Zygomycosis is the most lethal opportunistic fungal infection particularly among patients with diabetes mellitus, haematological malignancies, and patients receiving deferoxamine treatment. , , and species account for up to 75 per cent of mucormycosis cases. Infection with species has been reported in immunocompetent patients. species cause a broad spectrum of infections in humans, including superficial and disseminated infections, the latter with a mortality rate that approaches 100 per cent. Members of this genus may also cause allergic diseases and mycotoxicosis following the ingestion of toxin-contaminated food. Many factors are important in the virulence and pathogenic capability of microorganisms. Enzymes secreted from colonizing organisms (especially when is involved), such as extracellular phospholipases, proteinases and hydrolytic enzymes, contribute to host tissue invasion. Dimorphism in the ability of some species to grow as a unicellular yeast at room temperature and in pseudohyphal and filamentous forms in the host's body or at 37 °C is another virulence factor, because the hyphal forms are able to release hydrolytic enzymes and can specifically invade epithelial and endothelial cells. Conidial size and count are factors that affect virulence in some aetiologic agents. , with small conidia, is the main agent in invasive pulmonary aspergillosis, whereas , with large conidia, is an important aetiologic agent in sinusitis, cutaneous and wound aspergillosis. A low conidia count may not produce infection. According to Pasqualotto , species caused death when as few as 102 spores were inoculated. Other pathogenic factors detected in patients with invasive aspergillosis are sensory deprivation (environmental H), mutational restriction of nutrient acquisition, siderophore and amino acid biosynthesis, extracellular elastolytic proteases, gliotoxin and hyaronic acid. and can support growth across a broad range of environmental H values, and the biosynthesis of specific materials protects virulence factors involved in fungal pathogenicity. The outer cell wall layer plays a major role in an organism's pathogenicity. Chitin can influence immune recognition by blocking dectin-1, leading to significant reductions in cytokine production. In species the polysaccharide capsule protects the organism against the host immune system. Other pathogenicity factors in this organism include melanin production, mannitol secretion, superoxide dismutase, proteases and phospholipases. The synthesis of melanin and other conidial pigments in fungal organisms such as reduces complement opsonization by camouflaging binding sites, and reduces the ability of C3 to bind with conidia, although this property is still controversial. Rapid growth and affinity to the blood stream, heat tolerance, the production of efficient proteolytic enzymes such as lipases, proteases, glycosidic and lipolytic extracellular enzymes, siderophore production and an efficient iron transport system are pathogenic agents in members of the Zygomycetes family. Toxin production is another pathogenic factor in and species, which can contaminate food and cause mycotoxicosis. The first-line anatomical barriers of defense are the skin and mucosal surfaces, which protect the human body with an acidic H, enzymes, mucus and other antimicrobial secretions. When these barriers are broken by surgery or indwelling catheters, radiotherapy, burns or chemotherapy, fungal agents can reach the deep tissues. The clinical forms and severity of the disease manifestations depend on the host's defenses and immune response. An active host immune response against fungal spores or hyphae leads to allergies and asthma, and to the collapse of immune system defenses in infection, which results in the invasive form of infection. Both acquired and congenital immunodeficiency may be associated with increased susceptibility to IFI, which frequently occurs in patients with phagocytic and cellular immune defects. Once fungi enter, opsonization promotes fungal uptake and innate immune activation by a wide range of phagocytic and signaling receptors which include C-type lectin receptors and Toll-like receptors by host cell pattern recognition receptors. In healthy and immunocompetent individuals, the innate immune system (neutrophils and macrophages) is an efficient sentinel that provides protection from thousands of fungal species through phagocytosis of the invading pathogens by cell surface receptors. Numerous cytokines, . interleukin (IL)-1β, IL-12, IL-17, IL-23 and tumour necrosis factor (TNF)-α, are important in directing innate and adaptive responses to fungal pathogens. Cell-mediated immunity through Th1-based responses to fungal infections, characterized by TNF-α, IL-12 and interferon (IFN)-γ production, is a protective factor, and Th2-based responses, characterized by IL-4, IL- 6 and IL-13 production, are maladaptive and deleterious factors. Uncontrolled Th2 responses lead to chronic infections or allergic responses; in this connection, IL-4 and IL-13 play important pathogenic roles in allergic bronchopulmonary mycosis. The clinical manifestations of fungal infection are not specific, and like other infective diseases, a high degree of suspicion is required for the early diagnosis and optimal management of these infections. Systemic fungal infections, according to standard criteria, are established when histopathologic examination with special stains confirms fungal tissue involvement or when the aetiologic agent is isolated from clinical sterile specimens by culture. Radiological evidence from X-rays and high-resolution computed tomography is useful for the diagnosis of fungal infections. Pulmonary fungal infections such as aspergillosis, fusariosis, scedosporiosis or zygomycosis are characterized by central cavitation of pulmonary lesions, infiltration, pulmonary nodules, and halo or air-crescent signs. Documentation of the diagnosis of infection requires serial high-resolution computed tomography; however, the risk of radiation exposure in children must be considered. The specificity of these methods is lower in children than in adults. Conventional mycological methods include direct microscopic examination and the culture of samples in the mycology laboratory. Pathological examination and direct smears of samples with potassium hydroxide by an expert is the most rapid, cost-effective and sensitive method for the diagnosis of fungal infections. An expert can identify some genera of aetiologic agents such as yeasts or molds (with septated or nonseptated hyphae) and thus help ensure prompt, effective therapy. However, it may not be possible to identify fungal strains by this method, and fungal growth or the use of complementary methods such as hybridization in paraffin-embedded tissue may be necessary. Culturing clinical specimens (tissue, sputum, urine, wound, or blood) to isolate the aetiologic fungal agent is the gold standard for the diagnosis of IFI, provided that the samples are from sterile sites such as blood, tissue or cerebrospinal fluid, in which fungal infection can be documented. The sensitivity of culture differs among published studies. In cases with a specific fungal infection such as zygomycosis, the aetiologic agent loses its viability during tissue homogenization before culture. In addition, this method can produce false negative results if the clinical specimens are obtained after treatment with antifungal agents. Collecting appropriate tissues and other sterile specimens for culture or histology from patients whose condition is unstable, especially when neutropenia is present or platelet count is low, is difficult because it requires invasive procedures. The sensitivity of blood culture for the diagnosis of fungal infection is controversial. Multiple or repeated blood cultures should be performed to increase the likelihood of detecting candidemia, and filamentous fungi are rarely isolated from blood. Culture can be useful to determine the sensitivity of the isolated fungi to antifungal agents and identify resistant species, and are thus needed to optimize patient management. However, given the limitations of conventional methods of diagnosis for IFI, negative result on direct or pathologic smears and cultures do not rule out infection, so it is essential to use other suggested methods. There are many serological methods for the diagnosis of fungal infections, and the results of these tests become available sooner than culture. Skin tests and serum IgE level are suitable methods for the diagnosis of disorders such as ABPA. It should be noted that antibody assays are often negative in immunosuppressed patients. The detection of antigens in serum or cerebrospinal fluid is recommended for the diagnosis of infections. In severely ill patients the detection of antigens by enzyme immunoassay or the latex agglutination test in blood, bronchoalveolar lavage (BAL) fluid or urine can be useful for a rapid diagnosis. Cross-reactions are seen in some infected patients. The cut-off values for the detection of antigens in different populations may be different, therefore, it is important to interpret the results appropriately. The serial measurement of antigen titres is not a reliable indicator for the evaluation of antifungal therapy. The detection of fungal cell wall markers in serum has been reported for galactomannan (GM),-beta-D-glucan (BDG) and mannan. Galactomannan is relatively specific for species, and can be detected in urine, bronchoalveolar lavage fluid cerebrospinal fluid and other specimens with enzyme immunoassay. Various sensitivity rates from 30 to 100 per cent, and similarly wide-ranging specificities from 38 to 98 per cent have been reported for GM. Factors that limit the specificity of this test are immune reactivity with other fungi such as spp. and spp., false positive results with antibacterial agents such as beta-lactam antibiotics, particularly piperacillin–tazobactam and amoxycillin with or without clavulanate, and dietary GM in pasta, cereals and milk. 1,3-Beta-D-glucan (BDG) is present in the cell wall of most pathogenic fungi, including , , and , and is not species-specific or genus-specific for each organism. However, , and contain relatively small amounts of cell wall BDG; therefore, these assays may not be completely reliable in patients infected with these organisms. Sensitivities of 55 to 100 per cent, specificities of 71 to 93 per cent, positive predictive values of 40 to 89 per cent and negative predictive values of 73 to 100 per cent have been reported with various cut-off values for positivity ranging from 6 to 120 pg/ml for this assay. Depending on the type of assay (Fungitell or Fungitec) and population, the recommended cut-off levels are 7 pg/ml, 20 pg/ml or 60-80 pg/ml. False-positive BDG findings occur in the patients with fungal colonization or mucositis who have received empirical antifungal therapy. The combined use of a BDG assay (Glucatell) and a GM enzyme immunoassay (Platelia ) improves the specificity of diagnosis. Mannan is mainly found as a characteristic cell wall component in yeasts. The detection of circulating mannan and anti-mannan antibodies has been used as a diagnostic marker for invasive candidiasis or candidemia caused by the most pathogenic species of in adult patients with neutropenia and after myeloablative chemotherapy. The overall sensitivity of mannan antigen detection in patients with candidemia has been reported to be between 69 and 90.9 per cent, and specificity between 89 and 46.2 per cent compared to culture as the gold standard. In neonates, this test has yielded promising results particularly for ruling out candidiasis, considering its high negative predictive value of 98 per cent. The polymerase chain reaction (PCR) assay may serve as a powerful non-culture method for the diagnosis of systemic fungal infection in high-risk patients. Qualitative methods are sensitive in detecting fungal DNA in human blood samples, tissues, bronchoalveolar lavage and other body fluids. The sensitivity and specificity vary according to type of the tests. For example, nested PCR has a sensitivity of 92.8 per cent and a specificity of 94 per cent , and panfungal PCR has sensitivity 80 per cent and a specificity of 95.6 per cent. Quantitative methods include PCR ELISA, with a sensitivity of about 83.3 per cent and a sensitivity of about 91.7 per cent, and real-time PCR, with the sensitivity about 100 per cent and a specificity about 97 per cent. Molecular methods, which are rapid and can yield results within 6 h, have revolutionized the diagnosis of fungal infections because these enable diagnosis during the incubation period and early stage of infection, and prior to bone marrow transplantation. The diagnosis of infection based on molecular and serologic techniques can provide powerful tools for the early diagnosis of IFI. Because the fungi are common in the environment and opportunistic fungi in immunocompromised patients can cause high morbidity and mortality, the interpretation of positive or negative results with different laboratory methods is difficult for clinicians, so more than one method should be used for early diagnosis. #text The timely initiation of antifungal treatment is a critical component in the outcome for the patient. Unfortunately, patients with fungal infection often die of complications attributed to the infection despite antifungal therapy. Delays in antifungal treatment in candidemia infections are associated with a 20 per cent increase in mortality if more than 12 h have elapsed after a positive blood culture result, and the mortality rate increases significantly on each of the following three days. Localized infection is usually treated with topical antifungal agents, whereas disseminated infection requires the use of systemic agents with or without surgical debridement, and in some conditions immunotherapy is also advisable. Fungal infections are difficult to treat because antifungal therapy in infections is still controversial and based on clinical grounds, and in molds, the fungus isolated from the culture medium must be assumed to be the pathogen because these organisms are saprophytic in the environment. Some molds such as , and are intrinsically resistant to antifungal agents; therefore, mortality rates in the patients infected with these organism are high. The management of fungal infections is different depending on the type of infection and aetiologic agents. Antifungal agents have varying spectrums of activity, dosing, safety profiles and costs. Furthermore, many confounding factors such as the aetiologic agent, age, underlying diseases and surgical complications can influence the outcome. For example, for invasive aspergillosis voriconazole is superior to deoxycholate amphotericin B as the primary treatment in most patients. Many studies have shown an association between drug dose and outcome. Therapeutic drug monitoring (TDM) is necessary to ensure that therapeutic levels are achieved. Evidence supports TDM to optimize clinical efficacy when the drug is used for prophylaxis and therapy in IFI. The criteria for treatment response include the disappearance of clinical and radiological symptoms attributed to the infection. In some cases, the interpretation of radiologic images may be problematic because lesions may remain in the lungs and sinuses. Because fungal pathogens are eukaryotes and many of their biological processes are similar to those in humans, many antifungal drugs can cause toxicity when used at therapeutic doses. Monitoring the clinical response to individualized dose regimens can be helpful in interpreting the patient's condition with regard to treatment efficacy and toxicity, disease progression during therapy, drug absorption in patients with suspected poor oral absorption, signs or symptoms of significant toxicity or improper compliance with therapy. Serum drug concentration is influenced by many factors. Fluconazole and flocytosin in most preparations are excreted by the kidney as active drugs, and itraconazole and voriconazole metabolism in the liver involves specific enzymes, predominantly CYP3A4, CYP2C9, CYP3A4 and CYP2C19. Genetic polymorphisms of the enzyme involved and complications in the liver or kidney can reflect the serum level of these antifungal agents. The absorption of some antifungal agents requires special conditions. For example, absorption of itraconazole from capsule formulations is pH-dependent and requires an acidic environment, which is favoured by a full meal or a cola drink. In contrast, absorption of the oral solution is enhanced in the fasted state. Several patient-specific factors have been demonstrated to affect azole absorption including food (especially fat), gastric H and the use of proton pump inhibitors, diseases associated with chemotherapy, and the frequency of administration (because of saturable absorption). Differences in serum drug levels between healthy volunteers and patients have been reported in connection with these factors. Another important pharmacokinetic variable is drug interaction between azoles and rifampicin, carbamazepine, long-acting barbiturates, ritonavir, efavirenz and rifabutin, and most notably with cytochrome P450-inducing drugs. The clinical use of flucytosine is limited due to its gastrointestinal, haematological and neurological toxicity, the rapid development of resistance when used as monotherapy, and the lack of parenteral formulations. Accordingly, serum flucytosine concentration should be monitored to identify clinical indications, drug interactions and toxicity. Fluconazole is primarily eliminated via renal excretion, with approximately 80 per cent of the unchanged drug appearing in the urine. Thus, dosage adjustments are warranted in patients with a creatinine clearance <50 ml/min. When the susceptibility pattern of the drug is dose-dependent and in patients with renal dysfunction, monitoring can help improve the response. Routine monitoring methods are available for lipophilic triazoles, itraconazole, voriconazole and posaconazole. High performance liquid chromatography and bioassays have been developed to monitor the serum concentrations of triazoles, although the former method is the assay of choice when available. Additional studies are needed to measure serum levels and determine the best timing to achieve optimum concentrations of different antifungal agents in different populations. The efficacy and safety of antifungal agents are influenced by serum concentrations, clinical factors and the patient's physiological condition; all these factors can therefore, provide clinically useful information to monitor steady-state concentrations in patients with serious IFI. e b e s t a p p r o a c h t o t h e o p t i m a l m a n a g e m e n t o f f u n g a l i n f e c t i o n i s e a r l y d e t e c t i o n a n d i d e n t i f i c a t i o n o f t h e c a u s a l a g e n t , s o t h a t a p p r o p r i a t e t r e a t m e n t c a n b e i n i t i a t e d a s s o o n a s p o s s i b l e , e s p e c i a l l y i n i m m u n o c o m p r o m i s e d p a t i e n t s . C l i n i c i a n s s h o u l d b e f a m i l i a r w i t h t h e u s e o f d i a g n o s t c m e t h o d s a n d s u i t a b l e a n t i f u n g a l a g e n t s , b e c a u s e t h e s e a r e t h e f a c t o r s w i t h t h e g r e a t e s t i m p a c t o n t h e o u t c o m e f o r p a t i e n t s . T h e s e r u m c o n c e n t r a t i o n o f a n t i f u n g a l a g e n t s s h o u l d b e m o n i t o r e d t o r e c o r d b o t h t h e e f f i c a c y a n d t o x i c i t y o f t h e s e d r u g s .

italic xref #text There was no organized programme for malaria control in India in the pre-independence era; but there are records of epidemics and their control by the then Indian Medical Service. In 1912, a special malaria department was created in Mumbai (then Bombay). The department, apart from various surveillance and vector control activities, also distributed quinine and febrifuge free of cost. Large epidemics, and their classic investigations, were reported from Punjab, Bombay, and Bengal. Quinine was the treatment of choice for malaria and distribution measures for prophylaxis and treatment existed in several areas. In 1917, the Bengal Nagpur Railway and the East India Railways formed separate malaria control organizations for controlling malaria in and around stations. Similar programmes were undertaken in tea plantations of Assam and in Mysore by the Rockefeller Foundation. The first organized national programme in health - the National Malaria Control Programme was launched in 1953. In view of its initial successes, it was rechristened the National Malaria Eradication Programme (NMEP) in 1958 and developed organized surveillance for active case detection and treatment in 1961. A single dose of any 4-aminoquinoline was recommended as the presumptive treatment to all fever cases, while 8-aminoquinoline was added as the radical treatment to achieve gametocytocidal cure in falciparum and hypnozoiticidal cure in vivax malaria. By 1965, only 99,667 malaria cases were reported, but the situation deteriorated in subsequent years in the face of administrative, political, and technical challenges (). Hence, the Modified Plan of Operations was introduced in 1977 which emphasized the reduction of disease burden in a cost-effective and integrated manner. Fever treatment depots (FTDs), which obtained blood smears prior to presumptive treating, and drug distribution centres (DDCs), which did not, were established at the village level to ensure the availability of antimalarials in remote and inaccessible areas. Chloroquine resistant malaria was first reported in 1973 from the State of Assam in the northeast of the nation. Under the modified plan, the emphasis on chemotherapy was also supported by measures to strengthen operational research by mapping areas with chloroquine resistant strains. In 1978, NMEP created six regional monitoring teams to routinely conduct therapeutic efficacy studies of antimalarials drugs which expanded to 13 teams by 1985. : During the early days of the malaria programme in the 1950s-1970s the reduction of transmission occurred through vector control, primarily indoor residual spray operations. Case detection was geared towards identifying foci of transmission and not providing health care . The treatment aspect of eradication work sought to reduce morbidity among detected cases with little emphasis on radical cure until the latter maintenance phase of the programme as re-infection was though likely. No formal drug policies existed but the treatment was a 4-aminoquinoline (chloroquine or amodiaquine 10 mg/kg single-dose) for presumptive therapy with the addition of five days of primaquine (0.25 mg/kg for five days) regardless of the species present. For mass treatment in special situations, such as temporary labour camps, pyrimethamine (50 mg adult dose) was added for its sporontocidal action. : The first antimalarial drug policy was drafted in 1982 following the initial report of chloroquine resistance and the documentation of its presence in other States. The policy recommended different regimens for different areas depending on the species prevalent and the chloroquine resistance status. Areas were designated as chloroquine-resistant based on the proportion of RIII cases (early treatment failure) found during sensitivity studies. In chloroquine sensitive areas, presumptive treatment was recommended in the form of single dose of chloroquine (10 mg/kg) for malaria cases detected by active case detection (ACD), DDCs, and FTDs. After confirmation of the diagnosis by microscopy, radical treatment in the form of single dose primaquine (0.75 mg/kg) was recommended for falciparum malaria with the use of sulphalene-pyrimethamine (SLP) (adult single dose 1000/50 mg) in cases where the patient did not respond to chloroquine. In chloroquine resistant areas, amodiaquine (10 mg/kg single dose) was recommended for presumptive treatment in patients detected through ACD, DDCs and FTDs while patients detected through passive case detection (PCD) were presumptively treated with SLP. In migrant labour, a single dose of primaquine would be added during presumptive treatment. Radical treatment for falciparum malaria was SLP plus single dose of primaquine. In all areas the radical treatment for vivax malaria was chloroquine (10 mg/kg) and primaquine (0.25 mg/kg for five days). The five day regimen of primaquine was developed by the NMEP for its operational ease and reduced toxicity compared to the 14 days course and early reports of its comparable efficacy. The summarizes the revisions in the National Drug Policy for malaria in India. : The number of reported malaria cases dropped from 2.2 million in 1982 to 1.6 million in 1987 but again increased to 3 million by 1995. In light of several large epidemics of malaria with substantial mortality, the policy underwent a major revision in 1995. The NMEP stratified primary health centres (PHCs) into high and low risk areas based on the proportion of falciparum malaria cases, focus of chloroquine resistance in , slide positivity rate, and recorded malaria deaths. In low risk areas, presumptive and radical treatment and primaquine continued as recommended in the earlier policy. In high risk areas, the full dose of chloroquine (25 mg/kg over three days) as opposed to the single dose of chloroquine (10 mg/kg), along with single dose of primaquine was recommended as radical treatment for all fever cases. Additional primaquine (0.25 mg/kg for five days) was provided for all confirmed vivax malaria cases. In chloroquine-resistant areas, a single dose of sulphalene/sulphadoxine-pyrimethamine (SP) (adult single dose 1500/75 mg) was recommended for the treatment of falciparum malaria. The SP dose was increased from the two-tablet adult dose (1000/50 mg) recommended earlier to the three tablet adult dose (1500/75 mg) after studies suggesting higher efficacy of the latter. Amodiaquine was withdrawn from the drug policy since it possessed no advantage over chloroquine due to cross-resistance and was considered more toxic. The World Health Organization (WHO) also recommended the withdrawal of amodiaquine at the time because of reported side effects. The policy also approved the use of mefloquine in the country but only by a registered medical practitioner in cases of confirmed with ring stages and in chloroquine resistance areas. Finally, a review of the national drug policy was recommended every two years to keep up with the complex scenario and changing patterns in the country. : In 1998, the NMEP became the National Anti-Malaria Programme (NAMP) acknowledging the change of emphasis in the goals of control efforts. The 2001 review of the drug policy continued the recommendations of 1995 policy. The criteria for the designation of chloroquine-resistant areas, more than 25 per cent treatment failure (RI-RIII) in at least 30 patients of one PHC, were stated in the policy. In 2003, NAMP acquired additional responsibilities and emerged as the National Vector Borne Disease Control Programme (NVBDCP). In 2003, the short follow up (7 day) drug resistance studies were also ended. : The WHO technical advisory group, while meeting in India in 2004, recommended the use of combination antimalarial therapy, particularly with artemisinin derivatives, in member countries for treating to delay the emergence of drug resistance. Artemisinin combination therapy (ACT) consists of an artemisinin derivative combined with a long acting partner antimalarial drug. In the 2005 drug policy, in light of SP monotherapy resistance and WHO recommendations, artesunate (AS) + SP replaced SP alone in the national drug policy for the treatment of confirmed falciparum malaria cases in chloroquine resistant areas in 2005. Injection artemisinin was to be restricted to severe malaria cases only but oral artemisinin could be used in cases which were resistant to chloroquine and SP. The use of artemisinin related compounds was not recommended in infants. In 2007, several major changes occurred in the malaria drug policy. First, presumptive treatment, that is single dose chloroquine, was no longer recommended and the use of clinical diagnosis alone was rejected. The policy recommended investigating all suspected malaria cases by microscopy or with rapid diagnostic kits (RDK). In situations where diagnosis was not possible or the delay would be great, clinical treatment should use the full-dose, three days, of chloroquine until diagnosis was obtained. Second, the cut-off for designating an area as chloroquine-resistant was now only 10 per cent treatment failure given the recognition of the rapid spread of drug resistance as well as new cost-effectiveness analysis. Furthermore, clusters of PHCs, with a high (>30%) proportion of falciparum cases, around the resistant focus became the unit used for adopting second-line drug. Third, the anti-relapse treatment for was extended to 14 days of therapy after definitive studies demonstrating the poor efficacy of the five day course. Other notable points were for cases in whom chloroquine and AS+SP failed, oral quinine plus tetracycline or doxycycline would be used. The policy also dictated the disuse of single dose of primaquine along with AS+SP given that artesunate itself reduces gametocyte carriage. Another revision in 2008 added the treatment of patients negative by RDK with full-dose chloroquine as the NVBDCP kits are monovalent and only detect . The policy expanded the use of AS+SP to 117 districts across India which represented more than 90 per cent of the reported burden. The policy also recommended avoiding the use of mefloquine alone or in combination with artesunate in cerebral malaria. A flow diagram of the case management process was included for the first time to facilitate interpretation of the policy. Therapeutic efficacy studies continued to demonstrate a high prevalence of chloroquine resistance in falciparum malaria. In 2010, the drug policy was further reviewed and revised with the use of AS+SP for treating falciparum malaria cases made universal all across the country. For the first time the sulpha component of SP was specified as sulphadoxine instead of sulphalene/sulphadoxine. Single-dose primaquine was added to AS+SP, on day two, to reduce gametocyte carriage post-treatment since artesunate only acts against the immature forms. In 2013, there was another policy change in the seven North Eastern States (Arunachal Pradesh, Assam, Manipur, Meghalaya, Mizoram, Nagaland and Tipura) in view of the resistance to partner drug SP. The combination was replaced by artemether lumefantrine in these States. : Initially, only parenteral chloroquine and quinine were recommended for the treatment of severe malaria cases. Parenteral artemisinin derivatives were introduced in the national drug policy in 1995 for treating severe and complicated malaria in addition to quinine, particularly in areas of chloroquine resistance or during quinine shortages. Chloroquine was no longer recommended. Similarly, quinidine, under cardiac monitoring, was also recommended when quinine was not available. The 2002, the policy re-recommended injectable chloroquine for severe malaria, with precaution in children, in situations where injectable artesunate or quinine were unavailable. In 2005, the doses used for the artemisinin derivatives (artesunate, artemether, arteether, and artemisinin) were indicated, the minimum duration of treatment was seven days, followed by a full-course of ACT. In 2008, artemisinin was removed from the list of recommended derivatives. Till recently, quinine was the drug of choice for falciparum malaria in pregnancy though the emphasis of the national policy was on the drugs which were contraindicated rather than which were recommended. In 2001, the drug policy warned against the use of artemisinin derivatives in pregnant women. The present national drug policy recommends AS+SP in second and third trimesters though quinine is to be used in the first trimester until safety data for the artemisinin derivatives in the first trimester become available. For malaria, chloroquine has been recommended. The national programme recommends chemoprophylaxis only for select groups from non-endemic areas (travelers, and military personnel) exposed to malaria in highly endemic areas. Among the population in endemic areas, chemoprophylaxis is only recommended in pregnant women. The 1995 drug policy recommended weekly chloroquine prophylaxis in chloroquine sensitive areas. In chloroquine resistant areas, besides weekly chloroquine, daily proguanil was recommended. Since 2008, the drug policy recommends daily doxycycline for short term prophylaxis (less than six weeks) and weekly mefloquine for long term prophylaxis with treatment beginning two days or two week before and ending after four weeks of return, respectively. Among migrant labourers, weekly case detection instead of chemoprophylaxis was recommended on operational grounds. The maximum duration for chloroquine treatment was limited to three years because of concerns of toxicity. : Artemisinin monotherapy was banned in India in 2009. The drug policy recommends antimalarial therapy only after parasitological confirmation of the diagnosis which will reduce drug pressure for resistance, prevent side-effects, decrease drug costs, and improve the management of other causes of febrile illness. The current first-line therapy for , AS+SP, showed 98.8 per cent treatment success across 25 sites in India during 2009 and 2010 over 28 days of follow up. The programme has changed the ACT in North Eastern States to artemether lumefantrine in view of the resistance to partner drug SP. Chloroquine continues to be recommended for malaria. Though there were reports of chloroquine resistance in , the therapeutic efficacy studies showed a 100 per cent efficacy. The joint NIMR-NVBDCP National Drug Resistance Monitoring System conducts both widespread and longitudinal measurement of the treatments used in both species through simultaneous and molecular methods. The policy process is now well-defined, consultative, and evidence-based in addition to expert opinion. outlines the policy process for the formation of National Drug Policy for Malaria in India. The frequency of drug policy updates has also increased with three policy changes in the last five years. Finally, the policy has been translated into easy to follow case management guidelines for use by clinicians. : The present ACT (AS+SP) being recommended all over India except North Eastern States is a blister pack. Compared to fixed-dose combinations (FDCs), blister packs where the individual drugs are co-packaged may have poorer adherence, the potential for monotherapy use, and even poorer bioavailability. Another challenge for the drug policy is access to the delivery systems used for malaria diagnosis and treatment in India. Citizens living in remote, inaccessible, or disturbed areas may have to undergo considerable hardship to reach publicly provided care and turn to self-treatment or the formal and, more often, informal private sector for care. Community-based care, while introduced in some places, is not available everywhere. On the provider side, there is lack of awareness of the National Drug Policy, and best practice in general, among the private sector. A host of available therapies () shows a wide variation in treatment choice along with dose, duration, and co-administered drugs such as antibiotics. Physician and patient compliance to radical treatment (primaquine) is poor and may be contributing unnecessary burden in terms of additional transmission or relapses. : New ACTs have recently completed or are undergoing phase III studies and some are now registered. Phase III clinical trials have been completed for fixed dose ACTs including artesunate + mefloquine, dihydroartemisinin + piperaquine, arterolane + piperaquine, and pyronaridine + artesunate. Arterolane is a synthetic analogue of artemisinin and has the potential to replace plant-derived artemisinin. Trials are underway for combinations of current ACTs like artesunate + lumefantrine, artesunate+piperaquine, . Pharmacovigilance of antimalarial drugs is generating data on adverse events in patients which will help improve future policy. The case management of malaria has been extended to the village level in many areas through the use of community-based health workers. This should help promote more access and quicker treatment for suffering patients. The bivalent RDKs have recently been introduced and will improve the diagnosis. : Emerging resistance to antimalarial drugs poses the greatest threat to the National Drug Policy on malaria. While the results of sensitivity testing of antimalarial drugs in India have not shown any evidence of decreased sensitivity to artemisinin derivatives, clinical resistance to artemisinin drugs has emerged along the nearby Thai-Myanmar and Thai-Cambodia borders. The spread of resistance westwards, as happened with chloroquine, could jeopardize the most effective class of compounds we have for malaria treatment today. There is considerable evidence (clinical, , and molecular) of drug resistance to the partner drug used in the first-line ACT. Studies suggested the presence of double mutations in and single/double mutations in . Changes in these drug resistance markers are currently being monitored among patients enrolled in therapeutic efficacy studies in sentinel sites across the country. The spread and increase in SP resistance, which is likely inevitable, may decrease the present high efficacy of AS+SP in India and necessitate the switch to a different combination therapy. Though data on the efficacy of AS+SP on mixed infections are sparse, we know that SP is not very effective against vivax malaria. Finally, the emergence of chloroquine resistance in , as has happened elsewhere in the not too distant Western Pacific region, would complicate the control of the species responsible for half of the national malaria burden. : A key transition from the malaria eradication era towards a modern malaria control programme is moving from drug distribution to case management. The former is concerned with an output, supplying drug, while the latter is an entire process from diagnosis to care to referrals and is concerned with quality. The change to the case management can be challenging where activities are influenced by many interconnected factors. While the process has been long initiated, and strengthened by policy changes such as the end of presumptive treatment, quality has room to improve. To begin with, indicators such as the time from fever to diagnosis and treatment need to be monitored. Another goal should be increasing the proportion of malaria cases from passive detection, which is better suited to quality care, than from active detection, which is needed when health systems are not available or accessible. Finally, at present there are no protocols for the management of malaria-negative fever patients who seek care. : The universe of malaria treatment practices in India is wide and diverse. The National Drug Policy for malaria seeks to be evidence-based best practice. However, the adherence of the private sector to correct treatment of malaria, according to species or severity, is generally poor though more extensive surveys are needed. In 2008, private sector treatment was the largest risk factor for receiving artemisinin monotherapy in a six State survey. In 2009, the Drugs Controller General of India has banned the use, manufacture, sell and export of oral artemisinin monotherapy in the country. However, injectable artemisinin derivatives remain a preferred antimalarial treatment in rural areas for treating uncomplicated malaria. There is a need to rationalize the use of injectable artemisinin derivatives by limiting to severe malaria. While at present 80 per cent of medical care in India is privately provided, household survey data suggest that in rural areas of malaria endemic States only half of patients with fever seek private sector care. This is still a substantial proportion. Strategies for communicating and promoting the quality of care, including print media, workshops, and even one-to-one interaction, in the private sector are needed. : The choice of optimal ACT for future use is not clear. Artesunate amodiaquine has the disadvantage of cross-resistance with chloroquine whose sensitivity is decreased nationwide in . Artemether+lumefantrine is effective in India but has to be administered twice daily and can have erratic absorption. Arterolane+piperaquine is promising as a treatment for both species and has a long half-life but more data need to be generated. AS+mefloquine was also effective, India is largely mefloquine naïve from a resistance point of view, but has the disadvantage of neuropsychiatric complications and a higher cost than other ACTs. Evolutionary-epidemiological modeling suggests that the use of multiple first-line therapies may slow the spread of resistance although there is no empirical validation of the idea. Switching to multiple ACTs, or region-wise ACTs, in the public sector may be beneficial, but there are several operational barriers for doing so from procurement and supply chain difficulties to training multiple levels, including community-based staff. One step regarding the regional policy has been taken by the programme by replacing AS+SP with artemether lumefantrine. : Current policy recommends a single dose of primaquine on the second day in falciparum and for 14 days in vivax malaria. For the former, the efficacy, optimal day of administration, dose, and safety are not well known though these are being evaluated in an on-going randomized controlled trial (CTRI/2012/12/003273). For the latter, the course is long and compliance, by both provider and patient, is not well-known though suspected to be poor. It is important to improve compliance to antirelapse therapy since upto 40 per cent infections are known to relapse. Strategies to improve anti-relapse primaquine treatment could include directly observed therapy or administering the same total dose over a short duration. Tafenoquine, a long half-life 8-aminoquinoline resulting in a quicker treatment course, could become an alternative choice of drug and is in clinical development. Finally, there is a need to assess both the risks and benefits of primaquine therapy given its haemolytic potential. While glucose-6-phosphate dehydrogenase (G6PD) deficiency is rare in the general population, studies have documented its prevalence in up to 10-27 per cent of certain ethnic groups including tribal populations at higher risk for malaria. However, primaquine is being used since several decades and no significant adverse events have been documented till date though these are not well monitored either. Tools for G6PD testing at the primary healthcare level could help address this challenge. : In the present National Drug Policy on malaria, personal protection measures are recommended for preventing malaria during pregnancy. There is a need to assess other methods of preventing malaria in this vulnerable group, particularly in regions where the burden may be high. Strategies for evaluation include intermittent screening and treatment, intermittent preventive treatment, and other protection measures during antenatal care. The first strategy is currently being evaluated (CTRI/2012/08/002921). Finally, more data on the safety and efficacy of different drugs are also needed. Trials are underway in India to compare two ACTs (AS+SP versus AS+mefloquine) for treating malaria during pregnancy. Data from these efforts will be useful for future revisions. : Counterfeit and substandard antimalarials may pose a risk to patient health and antimalarial drug resistance in the country, with the North Eastern States near the China and Myanmar borders being particularly vulnerable. In a limited study of chemist shops in two sites of India, 12 per cent of essential drugs, including antimalarial drugs, were of substandard quality. Pre-procurement quality checks of antimalarial drugs are conducted by the procuring agency for public sector supply, but similar monitoring does not exist in the general retail market. Routine monitoring of the quality of drugs available on the market should be conducted, ideally by the drug regulatory agencies, in India. Even for public sector drugs, there is a need to check drug quality after dispatch and storage in field conditions where temperature, humidity, and physical placement may be adverse. #text

The variation observed in rates of incidence as well as mortality due to breast cancer, is due to a number of contributing factors like age, race, socio-economic status, life style, reproductive history, family history, . According to GLOBOCAN 2008 cancer fact sheet, incidence of breast cancer was approximately 1.38 million (23% of all neoplasms). Developed countries (except Japan) have a higher incidence (more than 80 for every 100,000 persons) as compared to developing nations (less than 40 for every 100,000 persons). As a consequence of advancements in diagnostic procedures and treatments available, the rate of survival of patients has increased. Hence, it is expected that the population susceptible to develop pain as a complication would increase. It has been estimated that in developing nations 70 per cent of new breast cancer cases would be seen by 2020. Pain arising in advanced stage of breast cancer can cause emotional suffering and affects quality of life of patients. As per the estimates of the International Association for the Study of Pain (IASP) the prevalence of pain in breast cancer ranges from 40-89 per cent. It has been found that persistent pain after surgical treatment is quite common and is higher among young patients, those undergoing radiotherapy and axillary lymph node dissection, and about 20-50 per cent women are affected by persistent neuropathic pain after their surgical treatment. #text V a r i o u s f o r m s o f p h a r m a c o l o g i c a l a n d n o n - p h a r m a c o l o g i c a l t r e a t m e n t s a r e b e i n g d e v e l o p e d t o a i d i n c a n c e r p a i n r e l i e f . S o m e o f t h e s e a r e d e s c r i b e d b e l o w : T h e l i f e e x p e c t a n c y o f b r e a s t c a n c e r p a t i e n t s i s i n c r e a s e d d u e t o e f f e c t i v e t r e a t m e n t o p t i o n s a v a i l a b l e t o d a y . N o n e t h e l e s s , p e r s i s t e n t c h r o n i c p a i n o f o n c o l o g i c o r i g i n h a s d e p r e c i a t e d t h e q u a l i t y o f l i f e i n a d v a n c e d s t a g e b r e a s t c a n c e r s u r v i v o r s a f t e r t r e a t m e n t . A r a n g e o f a n a l g e s i c s a n d a d j u v a n t m e d i c a t i o n s a r e a c c e s s i b l e t o t h e p a t i e n t s . T h e s e m e d i c i n e s p r o v i d e s a t i s f a c t o r y a n a l g e s i a b u t a r e a l l i e d t o a n u m b e r o f s i d e e f f e c t s . H e n c e , m o r e e f f e c t i v e w a y s f o r m a n a g i n g b r e a s t c a n c e r p a i n a r e n e e d e d . H o w e v e r , f u r t h e r s t u d i e s a r e n e e d e d f o r t h e n o v e l t h e r a p i e s a n d a g e n t s t o a s s u r e f a s t a n d a d e q u a t e p a i n r e l i e f w i t h m i n i m u m s i d e e f f e c t s .

Clinical pharmacology, a bridge discipline between basic sciences and clinical disciplines, was started in India in the 1960s. The development of pharmaceutical industry, clinical trials, accreditation issues in hospitals, and the commitment of the Government to provide essential medicines have necessitated a sea change in the role of the clinical pharmacologist. Capacity building with training in various skills for roles and responsibilities of the clinical pharmacologist is therefore needed. The Indian Council of Medical Research (ICMR) in its centenary year 2011-2012 gave a major boost to development of clinical pharmacology in India and planned workshops in thrust areas. One of the major thrust areas was that of antibiotic use, resistance and infection control. This area is unique in that it is truly an integrated area covered by the disciplines of clinical pharmacology, microbiology and infectious diseases. Bacterial infections are a major contributing factor to mortality in the world, and bacterial resistance to antibiotics has become an all pervading problem throughout the world, and in India. One of the major contributing factors towards resistance has been antibiotic use. Antibiotic use has been seen to be high in India with one recent surveillance study indicating 40 per cent of patients in the community on antibiotics, and an even higher rate of inappropriate use. One of the strategies to counter this would be to initiate antibiotic stewardship programmes. Antibiotic stewardship is a multidisciplinary programme with interventions and strategies to encourage appropriate use of antibiotics. Stewardship programmes aim to restrict inappropriate use of antibiotics, optimize selection, dose, route and duration of treatment for best outcomes, minimizing detrimental adverse events, excessive costs and emergence of resistance. The team involved in Antibiotic Stewardship is multidisciplinary and consists of an infectious disease physician, clinical pharmacologist/pharmacist, clinical microbiologist, infection control nurse and hospital administrator. These programmes have been shown to have both clinical and economic impact in hospitals. Strategies such as development and implementation of treatment guidelines, audit and feedback, and parenteral to oral conversion have been shown to be effective. There have however, been barriers also such as lack of awareness and support from physicians, lack of trained manpower and sometimes administrative shortsightedness. In India, while infection control guidelines and training programme had started relatively early, Antimicrobial Stewardship programme, guidelines and training have lagged behind. Publications from India on prevention of infection and control are only a few and on antibiotic stewardship are even fewer (). It was, therefore, deemed important by the ICMR that the discipline of clinical pharmacology takes on this void through a training programme not just to build capacity in antibiotic stewardship and infection control and impact hospitals, but also to initiate research projects in different hospital centres which would help in baseline studies and improve the situation with interventional strategies. The ASPIC programme was initiated through the collaboration of the Office of the National Chair of Clinical Pharmacology, ICMR, and the Christian Medical College, Vellore. This report outlines the content of the Antibiotic Stewardship, Prevention of Infection and Control (ASPIC) programme, the primary workshop conducted in 2012 in Vellore, Tamil Nadu, India, the issues in capacity building in Antibiotic Stewardship (AS) and the way forward. The overall programme was of one year with two contact sessions (workshops) and a project. The workshop was planned to provide training to participants to equip them with skills and understanding required for infection prevention and control practice; knowledge and skills required for development and implementation of antimicrobial policy guidelines for rational use of antibiotics to curb antibiotic resistance; and ability to plan and conduct research projects in antibiotic policy, infection prevention and control practice. : Twenty participants were selected on the basis of their qualification, past experience and quality of the concept research paper on antimicrobial stewardship and infection control. Among the participants selected, 15 were microbiologists, four were pharmacologists and one was a physician (). : The first contact session of five days (April 16-20, 2012) consisted of 40 hours of lectures, site visits, hands on practical training, demonstrations, and project discussion. The main topics covered and types of training assignments given during the workshop were: Patient safety - Lectures, practical demonstration and site visits; Health care associated infections - Definitions, preventions, demonstration and visits; Health care workers health and audits in infection control- Lectures and case discussions; Antimicrobial stewardship- Lectures and case discussions; and Project discussion. Resource persons for the workshop were faculty from various specialties (Infectious disease, Clinical microbiology, Clinical pharmacology, Nursing, Intensive care, Surgical specialties, and Hospital infection control). : By the end of the session on patient safety the participants were able to describe the standard precautions to be used to provide patient safety, principles and methods of, hand hygiene, biomedical waste management and its ultimate disposal, and sterilization of instruments. After the session on health care associated infections (HCAI), the participants were able to describe the definitions, surveillance and prevention of the following HCAI: Catheter related blood stream infection (CR-BSI); Ventilator associated pneumonia (VAP); Skin and soft tissue infection (SSI); and Catheter associated urinary tract infection (CA-UTI). By the end of the session on health care workers (HCW) health and infection control related audits the participants were able to describe the dos and dont's of needle stick injury prevention, protection of HCW by immunization and post exposure prophylaxis, main steps of outbreak investigation, different audit on various aspects of infection control, and disinfection of various areas and equipment in the hospital. By the end of the session on the antimicrobial resistance and stewardship the participants were able to describe the following: Surveillance of antimicrobial resistance and antibiotic consumption in their hospital; Principles of planning standard treatment guidelines and antimicrobial stewardship for their hospitals; and Principles of epidemiological investigation of antimicrobial resistance in the laboratory. During the first contact session, research projects to be undertaken at the participant's institution were discussed. Two projects were selected based on the common problems of the centres and the need for improvement. These projects were aimed at assessing the knowledge and practices in two areas in the hospital, . use of antimicrobials in surgical prophylaxis, and use of carbapenems/3 generation cephalosporins in intensive care units. After assessment, each centre would develop focused learning modules as an intervention strategy which would be pilot tested and then used as part of the stewardship practice of that hospital. After dissemination of these modules their effectiveness would be assessed by determining the practices once again. : All participants found the workshop productive and useful, and they enjoyed the style of teaching, the visits to see various laboratories, processes and facilities. The content of the workshop was felt to be adequate. However, some of the participants were of the opinion that the microbiological aspects could be reduced. The most appreciated session was on antimicrobial stewardship. Many participants wanted more time for project discussion and finalization. : At the end of the programme, it was hoped that each participant's institution would have a comprehensive systematic Infection Control and Antibiotic Stewardship programme, an Antibiotic Guidelines manual based on the local antimicrobial resistance pattern, and completed research projects and papers for publication. The programme will be evaluated using the following parameters: Participant attendance for workshops; Pre- and post-test workshop scores; Workshop feedback; Training others in participant's institution and region; Implementation of antibiotic stewardship in participant's institution; Implementation of infection control practices in participant's institution; Development of antibiotic policy manual in participant's institution; Research project initiation and completion; and Dissemination of findings through publication. : The project is expected to be completed within 6-8 months after initiation. The second contact workshop was held in July 2013. The projects were presented, discussed and evaluated. The improvements made in the area of antimicrobial stewardship and infection control in each site hospital were discussed. The commitment of the participants during the first workshop was commendable. The programme had applications from microbiologists and pharmacologists. The mix of personnel from different disciplines also created the need for a balance of topics between the disciplines in the short span of workshop time. The participants were from both private and government medical colleges and hospitals. The presence of an infection control team, quality assured microbiology laboratory, and access to various groups of antibiotics, were highly variable across the institutions. This was a challenge when designing projects that would be able to capture the practices and data uniformly from all institutions. The variation in approach of ethics committees in granting permission for projects was time consuming. It was also important to generate projects which were feasible and doable within the short span of time. Planning for collaborative projects when there is a great variation between participating centres, is challenging. Individual institutional projects may be another approach but the comparison as to which is better between both approaches is debatable. i s w o r k s h o p w a s t h e b e g i n n i n g o f a p r o g r a m m e i n i t i a t e d b y t h e I C M R , N a t i o n a l C h a i r f o r C l i n i c a l P h a r m a c o l o g y a n d C h r i s t i a n M e d i c a l C o l l e g e , V e l l o r e , t o t r a i n p h a r m a c o l o g i s t s a n d m i c r o b i o l o g i s t s t o a d d r e s s t h e n e e d f o r A n t i b i o t i c S t e w a r d s h i p a n d I n f e c t i o n C o n t r o l i n v a r i o u s i n s t i t u t i o n s a n d h o s p i t a l s t h r o u g h o u t I n d i a . T h e f o c u s w a s t o b r i n g t o g e t h e r d i f f e r e n t d i s c i p l i n e s a n d f a c u l t y i n w o r k i n g t o w a r d s a c o m m o n c a u s e , a s s e s s i n g t h e g r o u n d l e v e l r e a l i t i e s i n a n t i b i o t i c u s e a n d h o s p i t a l i n f e c t i o n s a s w e l l a s d e v e l o p i n g s t r a t e g i c i n t e r v e n t i o n s t h r o u g h a c o l l a b o r a t i v e a p p r o a c h t o i m p r o v e i n f e c t i o n c o n t r o l a n d r a t i o n a l a n t i b i o t i c u s e . I t i s h o p e d t h a t t h i s p r o g r a m m e w i l l r a i s e a w a r e n e s s a b o u t t h e s e i s s u e s a n d e m p o w e r t h e p a r t i c i p a t i n g c e n t r e s t o t r a i n o t h e r c e n t r e s s o t h a t a n t i m i c r o b i a l r e s i s t a n c e a n d h o s p i t a l i n f e c t i o n s c a n b e o p t i m a l l y c o n t a i n e d .

Stakeholders in public health research are researchers, funding agencies, organizations hosting research activities, policymakers, health managers, professionals in the health care system, patients and the community as well as the healthcare industry. It is important to understand the perceptions and views of various stakeholders on priorities in public health research to maximize the benefits of research. Involvement of various stakeholders apart from researchers is crucial to align the research initiatives with policies and public health programmes. It is particularly important in view of low public health research funding and low public health research output. The need for better collaboration between researchers and policymakers to enhance the use of research has been reported. In UK, these efforts were successful in addressing needs of various stakeholders. In Canada, the study that evaluated the interactions between researchers and decision makers identified three models of decision maker involvement namely formal supporter, responsive audience, and integral partner. There is a lack of data regarding involvement of policymakers in the public health research prioritization in India. Health is a State subject in India, therefore, public health programmes are delivered by the State health systems. State level stakeholders who are actively involved in the implementation of public health programmes are one of the most important stakeholders in identifying public health research agenda that would lead to direct programmatic benefits. In this context, it is important to understand the public health priorities as perceived by these stakeholders. In India, Department of Health Research (DHR) was constituted in 2007 with the mandate to serve as an apex Department for medical, allied basic sciences, clinical and public health research in the country. The DHR mandate is to translate the innovations into products/ processes and to introduce these innovations into public health service through health systems research, and to strengthen the coordination between various stakeholders to increase the use of research findings in practice of public health. A two day consultative meeting was organized by the DHR in September 2011 to bring the State government officials and public health researchers together on the same platform. A survey was conducted prior to the two day consultation in August, 2011 to obtain perspectives of State government health system stakeholders regarding public health research priorities in various Indian States. The survey also covered public health researchers from leading public health organizations in India to understand their views. An exploratory study was conducted to generate information that could serve as a basis for discussion in the meeting. The major objective of the meeting was to identify the key public health research priorities as perceived by State officials and researchers. A cross-sectional survey was conducted by National Institute of Epidemiology (NIE), Chennai, India in June-July, 2011 among stakeholders of various State government and national level programme stakeholders as well as researchers from public health research organizations and academic organizations. The State officials included National Rural Health Mission (NRHM) directors, Directors of Health Services or State level national programme managers. Public health researchers from leading national public health research and academic organizations in various States were also approached. : At least two officials from all 35 States were contacted. These officials were informed that they could take the feedback from State level programme managers for the survey. Nearly 30 leading public health research institutions in India were listed and at least one renowned public health researcher was contacted from each of these institutions. : A self-administered semi-structured questionnaire was used for data collection. The study questionnaire included structured questions to capture demographic profile and work experience of the respondents and open ended questions requiring them to write five leading public health research priorities. Additionally, the participants were requested to rank the six pre-identified public health research domains namely reproductive/maternal health including family planning, child health problems for under-five age group, adolescent health, undernutrition including micronutrient deficiencies, infectious diseases and non-communicable diseases, from one to six. The questionnaires were sent to the individuals by post and email and follow up was done by telephone and email. Data collection was facilitated by graduates of epidemiology and public health training programmes of NIE working in various State governments. They contacted the respective State officials, explained the background of the survey, handed the questionnaires, and made the required follow up to encourage the State officials to complete the survey. The study was approved by the NIE ethics committee, and informed consent was obtained from all the participants. : The proportions for the quantitative data were computed. The data were coded from the open ended questions regarding five leading public health research priorities. MS Excel and SPSS version 16.0 (SPSS Inc., USA) were used for data analysis. Overall 35 State officials from 15 Indian States and 17 public health researchers responded to our request and participated in the study (). In response to an open ended query regarding public health research priorities, 153 responses (multiple responses from each respondent) were obtained from 35 State officials and 64 from 17 public health researchers. Five leading priorities that were identified included maternal and child health (24%), non-communicable diseases (22%), vector borne diseases (6%), tuberculosis (6%) and HIV/AIDS/STI (5%) (). In addition to open ended query, a list of five broad areas was separately ranked for urban and rural areas by the participants. Reproductive/ maternal health including family planning was ranked one or two by 23 (66%) State officials and 12 (71%) public health researchers in rural areas. Child health problems were ranked one or two by 18 (51%) State officials and 12 (71%) public health researchers. Infectious diseases were ranked one or two by higher proportion of State officials as compared to public health researchers [12 (34%) vs. 5 (29%)]. In contrast, undernutrition was ranked one or two by higher proportion of public health researchers as compared to State officials in rural areas (). In urban areas, higher proportion of State officials ranked reproductive/maternal health at one or two as compared to public health researchers [22 (63%) vs. 4 (24%)]. Child health problems for under-five age group were ranked higher by nearly half of the State officials and public health researchers. At least one fourth of the State officials and public health researchers ranked adolescent health as one or two for urban areas. Non-communicable diseases were given higher ranking by 10 (59%) public health researchers and only 9 (26%) State officials for urban areas (). In addition to disease specific priorities, health systems research, community involvement and environmental issues were also identified among the leading ten priority areas (). Various other broad research areas such as impact assessment/translational research, adolescent health, economic impact of public health programmes, technology, human resources, immunization and vaccine preventable diseases, gender empowerment, urban health, de-centralized planning and new emerging infectious diseases were among the other identified priorities. This survey was an effort to include the State level policymakers and public health managers in the process of planning relevant public health research. Maternal and child health research continued to remain the leading priority; however, public health researchers gave more emphasis on need for research in the emerging public health challenges such as non-communicable diseases and adolescent health. These priorities were aligned to health research agenda of 12 Plan. Involvement of non researchers in the process of research to define the problems better can increase the use of research by practitioners and society. Traditionally it is believed that researchers should be able to determine the research agendas on their own, however, the need for collaborative research with other stakeholders’ involvement in setting the agenda has also been emphasized. Several countries initiated structured interventions in 1990s to increase the collaborations between researchers and policymakers. National Health System Research and Development Strategy-UK, Agency for Health Care Policy and Research in US and Prime Minister's National Forum on Health in Canada are some of the examples of such initiatives. In India, the DHR proposes to establish special linkages with central and state public health services to fulfill the mandate of translating the research findings to public health programmes and policies. This survey provided an insight into the perceived priorities of important stakeholders based on the experiences in the respective States. It needs to be emphasized that these are initial steps and the survey has to be expanded and also extended to other States for getting a comprehensive picture. Maternal and child health was perceived as the leading priority by researchers as well as State officials. It has been a leading area of public health research in the past decade. However, in context of the need to achieve millennium development goals, this should continue to be the focus of research. This is also consistent with the health research agenda of 12 Plan and goals of National Rural Health Mission, an initiative of the central/federal government to strengthen the primary care and achieve the millennium development goals for maternal and child health. Non-communicable diseases related research was reported higher in the priority, more so for urban areas. This is consistent with rising burden of NCDs in India and is one among the ten health research priorities identified in the 12 Plan. Lower priority given to research related to under-nutrition by the state officials was not consistent with the data that indicated high rates of malnutrition including micronutrient deficiencies in most of the States. This may be because nutrition is perceived beyond the health system functions and the interventions involve programmes and initiatives from other social sector departments of the government as well. However, 12 Plan calls for need for convergent action on nutrition and need for monitoring the health impact of programmes and policies of non-health sector. Health systems research was identified as an important priority area only by a small proportion of participants probably due to low exposure to health systems research. Health systems research is cross-cutting and may help in strengthening the health systems and implementation of various programmes. Among the ten key priorities for health research in 12 Plan, there are three priority areas of health systems research namely health financing, health information systems, and public health systems strengthening. There is a need for sensitization of the public health professionals regarding scope and need for health systems research. It is possible that some respondents could not have clearly differentiated between public health priorities from public health research priorities though it had been clarified in the invitation letters. Another limitation of the study was that we received the responses from 15 (42%) States even after repeated efforts to contact using various modes of communication. For every State at a particular level of progress and development which was not represented in our survey, there was a comparable State with similar geographic location which was represented in the survey. Hence it is appropriate to conclude that in spite of non-representation of some States, overall national views have been captured. However, it is important to explore the views and options of the health programme managers and policymakers from the States not represented in the survey at some later time point for consensus building. Special emphasis will have to be given to consider the viewpoints of stakeholders in the less developed States to consider their unique problems. Strength of our study was bringing together the views of both researchers and non-researchers. In conclusion, structured initiatives are needed to promote interactions between policymakers and researchers at all stages of research starting from defining problems to the use of research to achieve the health goals as envisaged in the 12 Plan over next five years. There is a need to identify specific research questions in the priority research areas that are most relevant and useful to the policymakers. Special focus to promote health systems research will be required to address the cross-cutting challenges in the implementation of public health programmes.

The study was conducted by the department of Psychiatry, St. John's Medical College and Hospital, Bangalore, Karnataka, India, in six villages covered under a primary health centre (PHC). There were 33 villages under this PHC covering a population of 29,117; of which, six villages were selected based on their proximity to the PHC. A house-to-house survey was conducted in the selected villages by a trained research investigator to screen for rural women with depression after obtaining a written informed consent. The institutional ethics committee approved the protocol of this study. The study was conducted from August 2006 till September 2009. CONSORT guidelines with extension to cluster randomised trials were followed. General Health Questionnaire (GHQ) -28 item version was administered to all adult women (>18 yr) in each of the selected villages. GHQ has been used in previous community studies in India and a standardized translated version in the local Kannada language was used in the present study. Only treatment naive women diagnosed with depression, who did not receive any treatment in the last six months were included in the study. Whole sampling frame was considered for this study. A demographic profile of the women was obtained using a questionnaire designed for the purpose of the study. Age, marital status, education, family size, type of family, employment, past history and family history of mental illness and occupational history were obtained. Socio-economic status of the women was obtained using Standard of living Index (SLI). This index was used in the National Family Health Survey 1998-1999 to compare the standards of living between rural and urban areas in India. The scores were tabulated and residents were classified into three categories: Low SLI, Medium SLI and high SLI. Women who obtained a score of more than or equal to 5 on GHQ were interviewed on the Mini International Neuropsychiatric Interview (MINI), a structured interview schedule to confirm a psychiatric diagnosis of major depression according to DSM-IV TR (fourth edition of Diagnostic and Statistical Manual of Mental Disorders - Text Revision) criteria. Any axis I co-morbid psychiatric diagnosis was noted. The severity of depression was assessed using Hamilton Depression Rating Scale (HDRS-17 item version). HDRS is available in the local Kannada language and has been used in previous studies in India. Scores obtained on the HDRS were used as the outcome variable. HDRS was administered at baseline and later at six months. WHO Quality of Life (Brev) scale was used to assess the quality of life of the women with depression. This questionnaire is available in local languages and has been used in a previous study of CMD in rural populations in India. The WHOQOL scale consists of 26 items and has four domains that measure physical, psychological, social and environmental components of quality of life. WHOQOL was administered at baseline and at six months. : The survey team consisted of four CHWs and a research assistant (RA). The CHWs were all women from the local community, had studied upto 10 standard and had a previous experience of working in community mental health programmes. The research assistant was trained in the administration and scoring of various questionnaires. He was blind to the two groups and was lead through the villages by the CHWs who did not reveal which villages belonged to the intervention group. The participants were reviewed by a trained physician at the PHC, and was blind to the treatment randomization. : A house-to-house survey was done in the six villages by the team of health workers and research investigator. All women aged 18 to 65 yr were interviewed after obtaining informed written consent. Initially, socio-demographic details were collected by the CHWs. The research assistant administered the GHQ. Women obtaining a score of more than or equal to five on GHQ and diagnosed to have major depression on MINI-Plus were referred to the weekly clinic in PHC. MINI-Plus has earlier been used in the Indian settings. : The six villages covered by the PHC were randomized into two groups of three villages each namely ‘Treatment as usual (TAU)’ and ‘Treatment intervention (TI)’ groups. Cluster randomized analysis was used. Village was taken as the unit of randomization and the analysis was done at the participant level. In the TI group, patients were monitored regularly by the CHW. Patients diagnosed with depression needed to visit the primary health centre once a month to consult with the physician. In the TI group, the CHWs visited patients immediately following the first medical consultation, educated the patient and her family members about depression and its treatment. They also emphasised taking antidepressant medication and continuing the treatment regimen. This was followed by another visit in the subsequent week to enquire about any possible side effects of medication and clarification of any doubts concerning the medical treatment of depression. This pattern of visits was maintained after every monthly consultation with the physician in the TI group. In addition, CHWs visited those patients who discontinued medication and / or those who did not visit the PHC for an initial consultation and encouraged them to resume treatment in the intervention group. In the TAU group, patients diagnosed with depression were encouraged to seek help from the physician at PHC with no additional input from the CHW. Treatment completers were defined as those patients who had come for at least four visits or have taken antidepressants for at least 16 wk. A dropout in this study was defined as any patient who missed at least two consecutive appointments with the primary care physician. : Those women with depression who were referred to the PHC and visited the clinic were registered. They were given an identity card containing their name, survey number, village and a specific registration number. Separate case sheets were maintained for each of them. After evaluation by a trained physician at the PHC, they were offered treatment with antidepressant medication. The antidepressants selected included either amitryptiline or fluoxetine and the physician made the choice based on the symptom cluster of depression. Those with predominantly anxiety related symptoms were treated with selective serotonin uptake inhibitors (SSRIs) whereas those with predominantly depressive symptoms were treated with amitriptyline. These medications were not available in the essential drug list of the PHC. : Adherence to treatment recommendations was measured using the total number of PHC clinic related visits and the total number of weeks the subjects took antidepressant medication. In addition, during home visits the CHW performed pill counts to ensure that patients took medication as prescribed by the PHC doctor. : Adjustments were made in the statistical tests to account for the clustering effect of village for the primary outcomes. Intra-cluster correlation . ICC (ρ) was calculated on the log transformed variable. Design effect was calculated by 1+ (n-1) ρ. Descriptive statistics were reported using mean ± SD or number and percentages as appropriate. Independent test was used to assess the difference between the TAU and the intervention group. Chi-square test or Fisher's exact test as appropriate, was used to find the association between the categorical variables. Mann-Whitney U test was used to compare the number of visits and duration of treatment between the groups. ANCOVA was used to assess the efficacy of intervention at endpoint (the post intervention quality of life and severity of depression) between the study groups adjusting for the baseline values. In ANCOVA, post intervention measures were the dependent variables, study group was the fixed factor and baseline measure were covariates. Intention to treat analysis was performed for the primary objective. The data were analysed using SPSS Inc Chicago, USA. Probability value of less than 5 per cent was considered as significant. A total of 814 houses were surveyed in six villages. Of the 1055 women (18-65 yr) who were interviewed; 859 (81.4%) consented for the study; 260 subjects who had a diagnosis of major depression as per DSM-IV TR criteria using MINI-Plus formed the final sample for the study (). Among the study participants, majority were married (87.3%), came from nuclear families (57.3%), were not formally educated (56.16%) and most were not employed in formal / informal sector (94.5%). At baseline, there were no significant differences between the TAU and the TI groups in terms of socio-demographic data, clinical characteristics and co-morbid diagnosis (Tables & ). A significantly (=0.001) greater number of depressed women from the TI (n=28) completed the treatment compared to the TAU (n=3). The rest of the participants had dropped out of the study. The reasons for the same were not collected. The number of clinic visits and weeks of treatment on antidepressant medication was significantly greater in the TI group compared to TAU group in participants who came for at least one visit to the clinic (). The number of participants who did not come for any visits was higher in the TAU (n=86, 70.5%) compared to TI (n=73, 53%). After adjusting for the major depressive disorder, recurrent type, participants in the TI group were 2 times more likely to visit the clinic as compared to TAU group. Intention to treat analysis revealed significantly more number of visits in the TI group then TAU group (<0.001). The choice of antidepressants did not influence the treatment adherence. Based on ICC, the estimated design effect was 2.25. After accounting for the clustering effect, the number of clinic visits and weeks of treatment on antidepressant medication was significantly greater in the TI group compared to TAU group (<0.05). There was no significant difference in the outcome measures at six months on HDRS (11.73 ± 7.24 in TAU group vs 11.30 ± 6.22 in TI group) and quality of life using WHO-QOL (Brev) between the groups after controlling for baseline HDRS score (). Patients in both the groups had improved significantly at six months. The mean dose of amitryptiline was 54.27 ± 23.23 mg (95% CI=25 to 100) and of fluoxetine was 22.40 ± 6.63 mg (95% CI= 20 to 40), respectively. Additionally, in the entire study population, there was no significant difference in the outcome measures at six months on HDRS between treatment completers and treatment dropouts (11.58 ± 6.06 vs 11.48 ± 6.80) after controlling for baseline HDRS score (). The present study was an attempt to improve treatment adherence to antidepressant medication in treatment naïve rural women diagnosed with major depression. Greater number of depressed women from the TI group contacted the physician compared to women from the TAU group. The dropout rate was higher in the TAU group when compared to the TI group. In addition, the number of clinic related visits and the duration of treatment (as measured by the number of weeks that subjects took antidepressant medication for) was significantly greater in the TI group compared to TAU group. It was difficult to compare findings from the present study to earlier studies on treatment adherence to antidepressant medication, owing to the fact that in majority of the earlier studies treatment adherence was estimated among subjects seeking help from a primary health center; whereas our study catered to women with major depression living in the community who had not sought any medical help. Nevertheless, findings from the present study were in broad agreement with earlier studies that noted better adherence and treatment compliance among patients with major depression who received a collaborative or stepped up care approach. Treatment adherence to the recommended protocol has been identified as one of the major factors in the effective treatment of depression. While there have been various attempts to increase treatment retention using collaborative or stepped up care approach in industrialized countries, only a few studies have examined the application of such approaches in resource poor countries. Araya undertook a study with a resource limited and socio-economically deprived population in a developing country and found a significant improvement in the severity of depression (HDRS) and functional impairment (SF-36) in patients who received the stepped care approach compared to the usual care. The stepped-care programme where trained primary care physicians provided pharmacotherapy had several components such as patient education, behavioural activation, problem solving and structured follow up by non-medical health care workers. Rickles in their study used a pharmacist-guided education and monitoring programme to enhance medication adherence. They found that the rate of missed doses at the end of the study was significantly lower in the pharmacist-guided education and monitoring (PGEM) group than the control group, hence suggesting the role of collaborative staff that could play a vital role in improving treatment adherence. Despite providing enhanced support through trained CHWs a large number of women living in the community diagnosed with depression did not seek help from the PHC. While other studies from India have made similar observations, it highlights the fact that there are considerable barriers to treatment of depression in rural women in India. In an earlier study on the outcomes of depression in a rural community, it was noted that financial problems, poor access to health care facilities and inability to take time off from work acted as significant barriers to seeking help from the PHC. In addition, among women, factors such as interpersonal difficulties, heavy drinking in the spouse, and economic difficulties linked to depression, are embedded within the family set up and may act as additional barriers to help seeking. There is a tendency for women to seek lay help due to both the lack of well-developed services in developing countries and the lack of awareness of mental health services. Stigma was found to be more in depressed women in primary care and adherence to treatment was poor in women with high stigma. Hence, these factors need to be considered for a comprehensive treatment of women in developing countries. While there was a significant difference in the treatment adherence pattern between the two groups, there was no significant difference in the outcomes of depression as measured by HDRS and WHO-QOL at six month follow up. There was a significant reduction in the severity of depression and an improvement in the quality of life in both the groups. In addition, there were no significant differences in outcome measures between treatment completers and dropouts. Chishlom followed up patients who received standard primary health care and mental health care incorporated with standard primary health care and found that there was a significant improvement in both groups with respect to the outcome domains such as depression, disability and quality of life. They suggested that it could either be due to spontaneous remissions or that the act of interviewing individuals and advising them to seek care could itself have served as an intervention. This could have been the case in our study as well. We did not do formal sample size estimation and power of the study was not calculated. In addition, a significant limitation of the study was that we did not ascertain the reasons for the large number of women with depression who either did not seek help and or dropped out of treatment. To conclude, enhanced community support provided by trained CHWs to rural women with major depression resulted in greater number of women adhering to treatment with antidepressant medication and also better treatment retention. However, a significant number of women with major depression still defaulted from treatment recommendations and did not seek help from a local PHC. The finding that a high proportion of subjects with depression did not access locally available health services has important public health implications. In resource poor countries it is also important to examine whether trained community level workers who have far easier access to the community can provide psychosocial interventions. Finally, and importantly this programme helped in enhancing the skills of CHWs in the domain of diagnosis, early identification and counselling of patients with depression.

This questionnaire-based study was conducted in the Department of Psychiatry, PGIMER-Dr RML Hospital, New Delhi, India, during 2009-2010. All psychiatrists and psychologists who were members of the Delhi Psychiatric Society (DPS) and listed with full contact information in the DPS Directory (n= 121) were invited to participate in this study. A little more than half (n=62) responded and were contacted by phone. Details of the study were explained to them, and they gave initially oral informed consent to participate. Two questionnaires were developed. One assessed knowledge of informed consent guidelines and attitudes toward obtaining informed consent, including from individuals with mental health conditions. The second assessed knowledge of guidelines regarding confidentiality protection and attitudes relevant to their interpretation and implementation. Each section contained ten questions (one section contained eleven), for a total of 41 closed-ended questions. True/false questions were used to assess knowledge, while agree /disagree/do not know were the options in the attitude assessment sections. At the end of each questionnaire, an open ended question was included to invite participants to express their attitudes towards informed consent and confidentiality. The questionnaires were given to the participants in person and at that time written informed consent was obtained. Some completed the questionnaires immediately and returned to the investigators while others said they would complete the questionnaire at their convenience. Ultimately, only half of the 62 professionals who had expressed initial willingness to participate completed the questionnaire (n=31; psychiatrists: n=26, psychologists, n=5). : The mean age of participants was 40.2 ± 10.9 yr, and mean duration of professional experience was 13.2 ± 10.9 yr. There were 24 male and 7 female participants. The participating psychologists and psychiatrists were comparable on demographic variables and range of years of experience; however, statistical comparisons between the responses of members of these groups were not feasible due to the small number of participating psychologists. : For the knowledge assessment questions, each correct answer was given a score of one with a maximum of 10. All scores were added to calculate final scores regarding participants’ knowledge of ethics guidelines and informed consent. The mean knowledge scores regarding informed consent and confidentiality were 8.16 ± 1.29 and 8.55 ± 1.46, respectively. There was no relation between duration of professional experience and knowledge about informed consent guidelines among these participants. : All were of the opinion that informed consent should be a mandatory document in every research project. All but one (who was uncertain) agreed that informed consent should have accurate information about a research protocol. On all other questions opinions varied (). Approximately a quarter did not think that informed consent was a safeguard against harm to participants. : Correct answers for individual questions received one mark, and the scores were summed up to yield the final knowledge score. The mean knowledge score on the confidentiality questionnaire was 8.65 ± 1.45. Knowledge of confidentiality was not influenced by duration of experience of the mental health professional in the field. : All agreed that “confidentiality should be maintained whenever possible with the exception of situations where there is a risk of harm to others.” The majority (90%) agreed that “doctors are patients’ representatives and, therefore, should not be expected to release information about a patient to a third party without the patient's proper informed consent”. About 60 per cent of participants agreed that “research using individually identified health information is important to the development of medical care”. Some (26%) felt that “researchers should be able to use unidentifiable personal health information without a person's consent.” Eight (26%) participants did not agree with the statement that there should be a witness during the informed consent process. Participants were also asked to provide detailed opinions regarding confidentiality and informed consent. Some of them opined that confidentiality should be maintained in all situations, particularly with regard to HIV infection. Some felt that confidentiality could be breached in exceptional situations, but most did not elaborate what these situations would be. However, one participant wrote that “confidentiality is a right of the patients and must be maintained in all situations - clinical or research except in a few conditions like risk of harm to others”. Some participants expressed the view that spouse and family members should be told about a patient's illness especially in case of psychiatric illness. A participant stated that “confidentiality should be maintained at all costs but the information can be given to others without patients’ consent when such information can be helpful to the community”. This study demonstrated that participating mental health professionals knew about informed consent and confidentiality guidelines and issues in research, and had positive attitudes towards fulfilling these important ethical requirements. However, only approximately a quarter of those professionals who were eligible to participate and only half of those who expressed initial willingness to do so-completed the study. This might have introduced some bias. With regard to attitudes towards informed consent, all participants agreed that the process of obtaining informed consent should be mandatory and accurate information should be presented, but opinions were divided on the amount of information that should be provided to potential participants. It is worth noting that participants considered informed consent to be a form or document, rather than a process that is documented by execution of a consent form, a mistake or a limited view that various commentators note. Regarding attitudes towards confidentiality there was relatively little variability. All were of the opinion that confidentiality is important and should be maintained. Several respondents felt that right to confidentiality ends where the safety of others begins, for instance, in the case of HIV infection, but generally did not specify what magnitude of risk of harm to others could justify breaching confidentiality. There was a great divergence of opinion regarding the permissibility of research employing identifiable and unidentifiable information. This divergence indicates a need for additional training about existing guidelines that prohibit use of identifiable participant information without participants’ consent, limit the use of identifiable information, and explain consent requirements and procedures for using de-identified or anonymized research data and biological samples. A few responses to the open-ended question also demonstrated confusion or seemingly contradictory opinions regarding confidentiality, for example, stating that “confidentiality should be maintained at all costs,” but also that confidentiality may be breached for the benefit of “the community”. In the open-ended comments section, respondents offered reasons for maintaining confidentiality. Some stressed that confidentiality was a right of the patients. Some emphasized the benefits resulting from maintenance of confidentiality, such as avoiding stigma and discrimination associated with health conditions, including psychiatric conditions. The participants revealed views about the scope of patient or research subject privacy protected by a professional's duty of confidentiality. Some were of the opinion that family members who are primary caretakers should be told about the nature of illness that their relatives have. The apparent rationale for breaching an individual patient's confidentiality with regard to the family was to benefit and provide support for the patient. This view reflected the belief that doctors, families and patient should collaborate for better treatment. This study had several limitations. Only knowledge and stated opinion were assessed, not the actual practice. This difference between knowledge or opinion and actual practice is important, as it is actual practice that affects the rights and welfare of patients and research participants. Atac reported that all physicians in his study on attitudes towards consent thought consent should always or generally be obtained, but less than 50 per cent obtained it in practice. This was because they thought that patients might not understand the informed consent form. Similarly, Yousuf studied doctors in India and Malaysia and observed that though awareness of informed consent was high in India (Kashmir), physicians practiced medical paternalism in clinical decision making by ignoring their patient's autonomy. In summary, the participating mental health professionals appeared to have adequate knowledge of basic ethical principles and guidelines concerning informed consent and confidentiality. Most respondents were aware of ethical issues in research. As more research studies, especially clinical trials are initiated in India, it is necessary to study whether professionals and researchers really attend to these issues in practice. Though knowledge itself is a critical prerequisite, it is important to ensure that actual practice reflects stated knowledge and attitudes.

This was a cross-sectional population survey using multi-stage cluster randomized sampling conducted from April 2008 to June 2009 in Chandigarh, north India, involving 2227 subjects. Chandigarh city is divided into three zones by two main roads. Two sectors from each of the three zones were selected by simple random sampling. The first house was selected from within each selected sector by simple random sampling. Starting from that house, all the eligible people ≥20 yr of age were screened from the consecutive houses till a sample size of at least 375 was reached in that sector. Subjects having any acute illness like fever and/or on medications likely to increase plasma glucose such as glucocorticoid, and pregnant females were excluded from the study. The study protocol was approved by Institutes Ethics Committee. The procedure was explained to the participants at least a day prior to the study and informed written consent was obtained from each. Detailed history regarding age, gender, education, occupation, any chronic illness in the participants or their family was recorded. History regarding smoking, alcohol consumption and diet was also recorded. Physical activity of the participants was recorded in a proforma adapted from Global Physical Activity - Questionnaire 2 (GPAQ-2) of World Health Organization (WHO), and classified as high, moderate or low physical activity. Subjects having low physical activity were labelled to have sedentary lifestyle. The socio-economic status (SES) scale as described by Kuppuswamy which takes into account the education of the head of the family, occupation of the head of the family, and monthly income of the family was followed. Height, weight and waist circumference (WC) were measured thrice and mean was noted. Height was recorded on a stadiometer to the nearest mm. Weight was measured by a digital weighing machine to the nearest 100 g and was calibrated using standard weight every day. Waist circumference was measured with a non-stretchable tape at the midpoint between lower border of rib cage and upper border of iliac crest. High WC was defined based on criteria modified for Asian Indians (WC ≥90 cm in men and ≥80 cm in women. BMI ≥ 23 kg/m was defined as overweight). Blood pressure was measured twice with mercury sphygmomanometer in sitting position in the right arm after 5 min of rest, to the nearest 2 mmHg and average of systolic and diastolic blood pressure was recorded. Hypertension was defined as blood pressure ≥ 140/90 mmHg. After an overnight fast of 8-14 h, capillary plasma glucose estimation was done with a glucometer using glucose-oxidase method (One Touch Ultra 2, Johnson and Johnson, Mumbai). The glucometer had in-built calibration to convert whole blood capillary glucose values to plasma glucose values. The coefficient of variation for this glucometer was 3.2 per cent. To strengthen the validity of the study external quality control was ensured by every 10 sample sent to the reference laboratory in an oxalate sodium fluoride vial, for glucose estimation using Hitachi auto analyzer 902 (Tokyo, Japan) based on the glucose oxidase-peroxidase method. The agreement between capillary plasma glucose and laboratory glucose values was calculated using Bland Altman methodology. The mean difference for fasting plasma glucose was 5.2 mg⁄ dl (0.29 mmol⁄ l) (95% CI 3.7-6.7) and for 2 h plasma glucose post-glucose load was 3.4 mg ⁄ dl (0.19 mmol⁄ l) (95% CI 2-4.8), with capillary glucose being higher than laboratory glucose values. Triglycerides and HDL-C was measured in venous blood by enzymatic colorimeter method using commercial kits (FAR Srl, Verona, Italy). The blood samples were transferred from the place of collection to the laboratory in insulated containers packed with ice bags and were processed within 4-6 h. HbA1c was estimated from venous blood on National Glycohaemoglobin Standardization Program-certified Bio-Rad D-10 system (Bio-Rad Laboratories, Hercules, CA, USA) based on ion-exchange high performance liquid chromatography. For the diagnoses of diabetes and prediabetes, the1999 WHO criteria for capillary plasma glucose were used. Diabetes was defined as fasting plasma glucose (FPG) ≥126 mg/dl (≥7 mmol/l) or 2 hPG ≥220 mg/dl (≥12.2 mmol/l), or both. Prediabetes was diagnosed as the presence of isolated impaired fasting glucose (IFG), defined as FPG ≥110 mg/dl (≥6.1 mmol/l) and <126 mg/dl (<7 mmol/l), and 2 hPG <160 mg/dl (< 8.8 mmol/l) or isolated impaired glucose tolerance (IGT), defined as 2 hPG≥160 mg/dl (≥8.8 mmol/l) and <220 mg/dl (<12.2 mmol/l) and FPG<110 mg/dl (<6.1 mmol/l), or both IFG and IGT (as defined above). : In the sampling design employed, the probability of selecting a sector within each of the three zones varied, and so did the probability of selecting a study subject within each selected sector. Consequently, the probability of selecting each study subject from the total population was not uniform - . the design was not a ‘self-weighing’ one. The inverse of the probability of selection was, therefore, employed as a weight for that subject, and all subsequent statistical estimations were undertaken using STATA 9.0 (Texas, USA). Quantitative characteristics were summarized by arithmetic mean and standard deviation. Numerical trends in various cardiovascular risk factors across different age groups were assessed using regression analyses. Regression coefficients, standard deviations, standardized regression coefficients (beta), and values were estimated. Multivariate logistic regression analysis was performed to determine the significance of association of different cardiovascular risk factors with age, BMI and waist circumference. All statistical tests were two-sided and performed at a significance of α=0.05. A total of 2368 subjects aged ≥20 yr, were approached based on multistage cluster randomized sampling in different sectors of urban, Chandigarh. Of them, 123 were non-responders. However, these subjects were similar to the study subjects, in terms of age, gender and body mass index (BMI). Of the remaining 2245 subjects, 18 subjects were excluded since HDL-C/TG levels were not available. Finally, 2227 subjects were evaluable in the study with a response rate of 94 per cent. The study population included 1068 men and 1159 women (1:1.08) with mean age of 42.7±16.6 yr (range 20-94 yr). The baseline demographic characteristics of the study population are shown in . The prevalence of cardiovascular (CV) risk factors among the study population in various age groups is shown in . Among the parameters studied, the most prevalent CV risk factor in the age group of 20-29 yr was sedentary lifestyle (63.2%), while from fourth decade and onwards, overweight/obesity was the most common risk factor with a prevalence varying from 58.9 to 84.7 per cent. The second most common cardiovascular risk factor in the age group of 20-29 yr was overweight/obesity, in 30-49 yr sedentary lifestyle, in 50-69 yr hypertension and in subjects ≥70 yr, it was hypertriglyceridemia (). On gender-wise analysis, in women in the age group of 20-39 yr, the most prevalent cardiovascular risk factor was sedentary lifestyle, however, in the age group of 30 yr and onwards overweight/obesity predominated (). The second most frequent cardiovascular risk factors in women was low HDL-C in all age groups except in the third and forth decade where sedentary lifestyle and obesity/overweight were more frequent. However, in men, the trend of prevalence of the two most common cardiovascular risk factors was similar to that of the whole population (). Low HDL-C were characteristically more common in women compared to men and this gender-wise differential trend was observed throughout the various age groups (). Hypertension was more frequently recorded in men younger than 60 yr of age compared to women, however, this gender difference was significant only between 30 to 39 yr. The prevalence of all the studied CV risk factors other than sedentary lifestyle and smoking significantly increased with age (<0.001). Similar trend was observed on analyzing separately for both the genders (). The prevalence of the studied CV risk factors including overweight/obesity, hypertension, dyslipidaemia, prediabetes and smoking was double in subjects in fourth decade of life, as compared to subjects in third decade of life, whereas prevalence of diabetes showed an abrupt increase by 20-folds during fifth decade (0.8 to 20.7%) followed by a progressive rise in the later part of life. Sedentary lifestyle was observed in two-third of the subjects between 20 to 39 yr and a decline in its prevalence was noted between 40 to 69 yr followed by a rise again at ≥70 yr. Smoking was significantly (<0.001) more prevalent in men than in women. None of the subjects had personal history of coronary artery disease during 20 to 29 yr. However, 2 per cent of subjects had history of coronary artery disease (CAD) during fifth decade and its prevalence significantly increased with age (<0.001). Multivariate regression analysis showed that age, BMI and central obesity were significantly and positively associated with the presence of hypertension, hypertriglyceridaemia and diabetes mellitus and low HDL-C (). Waist circumference was more strongly associated with the presence of hypertension and low HDL-C as compared to BMI. Adjustment for age subdued the association of various CV risk factors with waist circumference and BMI (). This study showed a high prevalence of cardiovascular risk factors including sedentary lifestyle, obesity and low HDL-C in the third and fourth decade of life and the prevalence of these risk factors progressively increased with advancing age. The prevalence of central obesity, hypertension, dysglycaemia and smoking was almost double in subjects in the fourth decade of life, as compared to those in the third decade of life. Throughout all age groups, sedentary lifestyle, overweight/obesity and low HDL-C were more frequent in women as compared to men. Various epidemiological studies have shown that physical inactivity increases the risk of CVD. Overall, 61 per cent of the subjects in the present study followed sedentary lifestyle. Similar observations have been made from other parts of the country like in Kolkatta, 59 per cent of the subjects including 61 per cent men and 56 per cent women were following sedentary lifestyle. Sedentary lifestyle was more prevalent in women as compared to men at all ages as shown by others. This can be attributed to prevalent socio-cultural factors. The present study showed that sedentary lifestyle was opted by almost 63.2 per cent of the population even in the third decade of life and this was higher compared to that observed from Rajasthan where 38.4 per cent of men and 33.6 per cent of women adopted sedentary lifestyle in the third decade of life. This suggests that there is a need for targeting the population as early as in the third decade of life, so that the incidence of other CV risk factors can be reduced. Overweight/obesity is an established risk factor for CVD and diabetes as shown previously. Our study showed a high prevalence of overweight/obesity as compared to previous studies and it was the most prevalent CV risk factor during the fourth decade of life and onward. The varying prevalence of obesity in different studies partly can be attributed to different criteria used to define it. In the present study, low HDL-C was a common prevalent CV risk factor in all age groups including the subjects in the third decade of life and was more common in women as compared to men. This observation was similar to another study conducted in north India. The possible reasons for low HDL-C may be sedentary lifestyle, obesity and ethnicity as shown in previous studies and in migrant Asian Indians. As the age advanced, hypertension became one of the most prevalent CV risk factors both in men and women and the prevalence of hypertension was higher in the present study than that reported previously in other studies from India. The high prevalence of sedentary lifestyle and central obesity in the study population possibly would have contributed to high prevalence of hypertension. Smoking is a well established risk factor for CVD. In the present study, prevalence of current smoking was 7.5 per cent, which was low compared to that reported earlier (32%). The prevalence of almost all the studied cardiovascular risk factors progressively increased with age as has been shown earlier. The prevalence of central obesity, hypertension, dysglycaemia and smoking was almost double in subject in the fourth decade of life as compared to the subjects in the third decade of life. Previous studies from India have also reported similar finding. The prevalence of diabetes was 20-folds higher in subjects in fifth decade of life compared to those in fourth decade of life. This has also been shown in our previous study where the prevalence of diabetes was maximum in the fifth decade of life, which is similar to the observation in other South-east Asian countries but is a decade earlier than the western population. Increasing age and central obesity are associated with accumulation of multiple metabolic abnormalities, our study showed a strong association of increasing age, and obesity with prevalence of hypertension, diabetes, low HDL-C, and hypertriglyceridaemia. The strengths of the study included multistage-cluster randomized study involving a large number of subjects with good response rate from a city with a high literacy rate in India. Limitations of the study included lack of history regarding diet and family history of premature cardiac events, single estimation of glucose values by glucometer and lack of estimation of serum cholesterol. Being a cross-sectional study, it did not provide an opportunity to explore the cause and effect relationship between various risk factors and cardiovascular events. In conclusion, sedentary lifestyle, obesity and low HDL-C were found to be the most prevalent CV risk factors in subject in the third and fourth decade of life in this north Indian population. A mandate is required to target this population to prevent this epidemic of cardiovascular diseases.

: From March 2011 to March 2012, a total of 240 cirrhotic patients (174 men, 66 women) were enrolled in the study conducted at Tongji Hospital, a tertiary care hospital in Shanghai, PR China. Patients with hepatocellular carcinoma, other malignancies, known haemostatic disorders other than liver disease, bacterial infection, renal dysfunction, clinical history of peripheral venous thrombosis, Budd-Chiari syndrome, spleen resection, liver transplantation, endoscopic treatment, or anticoagulation therapy were excluded. Consequently, 162 patients were found to be eligible for inclusion in this study. HBV cirrhosis was defined by a positive diagnosis of HBV-related cirrhosis and the presence of the HBV surface antigen (HBsAg) in the absence of a history of alcohol consumption or other co-existing viral infections. Decompensated liver cirrhosis was diagnosed by clinical findings or morphological features and its severity was scored according to the Child-Pugh classification. To explore the risk factors associated with PVT, the patients were classified into two groups: the PVT group ( = 40) and the non-PVT group ( = 122). All patients, including the 40 with PVT underwent computed tomography portal angiography (CTPA) and colour Doppler ultrasonography (CDUS) to rule out underlying hepatocellular carcinoma and to confirm PVT. The study protocol was approved by the Tongji Hospital ethics committee. Written informed consent was obtained from each patient. : After fasting for at least 12 h, 20 ml of blood was collected from each patient and haemoglobin (Hb) levels and blood platelet counts (BPCs) were determined using a Sysmex XE- 2100 automated analyzer (Sysmex, Kobe, Japan). Prothrombin time (PT), activated partial prothrombin time (APTT), and fibrinogen levels were determined by routine coagulation methods, using a Sysmex CA6000 automated analyzer (Sysmex, Milton Keynes, UK). Total bilirubin (TBIL) and albumin levels were determined using a bromocresol green assay and spectrophotometric method, respectively. The platelet aggregation rate (PAR) was determined by turbidometric platelet aggregometry (SC-2000; Beijing SUCCEEDER Technology, Beijing, China). Anticardiolipin antibodies (ACAs) were detected using an enzyme-linked immunosorbent assay (Byk Gulden, Milano, Italy), and D-dimers and high sensitivity C-reactive protein (hs-CRP) were detected using kits from Sun Biotech (Shanghai, China). Levels of intercellular adhesion molecule-1 (ICAM-1), alpha-interferon (IFN-α), and tissue necrosis factor alpha (TNF-α) were measured using enzyme-linked immunosorbent assays; IgA, IgM, and IgG were detected by turbidometric methods. : CDUS examinations were performed using a colour Doppler ultrasound scanner with a 3.5-5 MHz convex probe (Mylab-90, Philips Electronics, Amsterdam, The Netherlands). The PV system was examined following current guidelines. The PV diameter and portal flow velocity were calculated automatically by the instrument. Splenic thickness was measured perpendicular to the long axis of the spleen. All the patients were examined by CTPA to further confirm the presence of thrombosis, especially if thrombosis was not confirmed by CDUS. : The SPSS software package for Windows 2000 (ver. 11, IBM SPSS, Armonk, NY, USA) was used for all statistical analyses. All quantitative data were expressed as mean ± standard deviation (SD). Categorical variables were shown in terms of frequencies (percentages). The differences between the two groups were evaluated using the chi-square (χ) or test. The continuous variable (age) was examined by a normality test. The Pearson chi-square test was performed for categorical data and the Pearson bivariate correlation test was subsequently applied. Multivariate binary logistic regression was performed and the model was estimated using the stepwise backward method (Wald). The coefficients obtained from the logistic regression analyses were also expressed in terms of odds ratio with 95% confidence intervals. PVT was found in 40 patients [26 (65%) male patients (age range, 39-88 yr) and 14 (35%) female patients (age range, 61-82 yr)]; the overall presence was 24.7 per cent in this study. PVT occurred in 34 cirrhotic patients following HBV hepatitis, in two following HCV hepatitis, and in two patients each with either autoimmune or cryptogenic cirrhosis. However, PVT was not observed in either schistosomal or alcoholic cirrhosis patients. Viral hepatitis was, therefore, the aetiological factor in 90 per cent of the cirrhotic patients demonstrating PVT; the percentages of patients with these aetiological factors did not significantly differ between the two groups. Univariate analysis showed that the ratios of both mean age and gender were similar between the two groups and that the percentages of patients who smoked, consumed alcohol, had hypertension, and had diabetes mellitus did not significantly differ between the two groups. Among the patients with PVT, four (10%) were in Child-Pugh class A, 20 (50%) in class B, and 16 (40%) in class C. Child-Pugh classes were not significantly different between the patients with and without PVT. GI bleeding, abdominal pain, abdominal distention, ascites, jaundice, and hepatic encephalopathy were common clinical presentations among patients in the PVT group. Three of the 40 PVT patients died within 12 months of hospital admission; two died of gastrointestinal haemorrhaging and one died of surgical complications. Additionally, 17 of the 40 PVT patients were re-admitted due to gastrointestinal bleeding or hepatic encephalopathy within the 12 months following their initial admission; the remaining 20 PVT patients demonstrated stable conditions over the follow up period. Careful examination of the clinical characteristics of the PVT patients indicated differences between the characteristics of patients with isolated PV trunk thrombosis and those with PV trunk and superior mesenteric vein or splenic vein thrombosis. For instance, ascites was most commonly observed in patients with isolated PV trunk thrombosis. However, gastrointestinal bleeding and abdominal pain were most common among PVT patients with superior mesenteric vein or splenic vein thrombosis. Jaundice was also commonly observed in patients with thrombosis in the branches of the PV (). PVT occurred in the PV trunk in 12 (30%) patients, in the PV trunk and superior mesenteric vein in 20 (50%) patients, and in both the PV trunk and splenic vein in two (5%) patients. In five (12.5%) patients, thrombosis was found in a branch of the PV. PVT occurred in the PV trunk, splenic vein, and superior mesenteric vein at the same time in only one patient (2.5%; ). The thrombotic risk factors and the clinical and biochemical characteristics of patients with and without PVT are shown in . The levels of fibrinogen, TBIL, PT, IgA, IgG, and IgM were lower in the PVT group than in the non-PVT group but this difference was not significant. The levels of albumin, APTT, ICAM-1, hs- CRP, and PAR were higher in the PVT patients than in the non-PVT patients, but the difference was not significant. The levels of IFN-α, TFN-α, and D-dimer were also not significantly different between the two groups. Univariate analysis showed that the PVT patients had significantly lower levels of Hb and PBC than the patients in the non-PVT group (=0.003 and <0.001, respectively). All the plasma samples were negative for ACA. The diameters of the PVs and the splenic thicknesses showed significant differences between the two groups (=0.041 and =0.001, respectively). However, the portal flow velocities were not significantly different between the groups (). A logistic regression analysis model was applied to the variables associated with the presence of PVT. This analysis showed that decreased levels of BPC and Hb and the presence of splenic thickening were the variables independently associated with PVT (). A significant positive correlation was observed between PAR and BPC, as shown in the . The value of R was 0.533 (<0.01). PVT was found in 24.7 per cent patients in this study, confirming the increased occurrence of PVT among cirrhotic patients compared with the general population. Among the patients in both groups, HBV was observed to be the single major cause of cirrhosis. These results support those of a previous report indicating that HBV is the major risk factor for PVT in Southeast Asian populations. PVT was primarily observed to occur in the PV trunk, with the superior mesenteric vein being the second most common site. This suggests that extension of the thrombosis beyond the PV occurred preferentially towards the mesenteric vein. In the PVT patients, the most common clinical presentations were gastrointestinal bleeding, abdominal pain, abdominal distention, fever, jaundice, and hepatic encephalopathy. However, the signs were not identical for the patients with PVT in different locations. Further studies need to be done to show whether the site of PVT affects the clinical characteristics of the disease in a patient. Some of the previously hypothesized risk factors of PVT, such as age, gender, smoking status, alcohol consumption history, hypertension, and diabetes mellitus, were not associated with PVT in this study. Furthermore, various laboratory markers (fibrinogen, TBIL, PT, IgA, IgG, IgM, albumin, APTT, ICAM-1, D-dimer hs-CRP, IFN-α, and TNF-α), portal flow velocity, and the presence of ACAs did not appear to affect the risk of developing PVT; Child-Pugh classes were also similar between the groups with and without PVTs. The absence of ACA in the plasma of the patients in this study was particularly remarkable, given that the presence of these antibodies has been previously associated with thrombotic events. Theoretically, Hb levels are expected to be an independent risk factor for PVT in this study. In fact, because splenomegaly decreases the Hb level and hypersplenism is more severe in PVT patients, the lower Hb level may be more appropriately attributed to splenomegaly. Thus, although Hb was associated with PVT, it was not an independent risk factor of PVT. Similarly, inconsistent conclusions have been drawn from different studies regarding the association of BPC with PVT. Lower BPCs can also be attributed to splenomegaly since splenomegaly is more evident in PVT patients than in non-PVT patients. In addition, the positive correlation between BPC and PAR suggests that the defect associated with the platelets may be the result of their decreased numbers. Extrahepatic portal venous obstruction has previously been shown to result in a decreased PAR in 83 per cent of the patients in one study. Therefore, the qualitative changes observed in BPCs were speculated to contribute to the formation of PVT, and these were considered more important than the quantitative changes in BPCs during the development of PVTs, although BPCs might not be a true independent risk factor for PVT formation. Portal flow velocities of <15 cm/sec have been previously identified as an independent risk factor of PVT formation. However, the present findings were dissimilar from previous reports in that the portal flow velocities for the patients in the two groups were greater than 15 cm/sec and no significant difference was observed in the velocities for the groups. Thus, PVT may not occur if the only alteration that has occurred is a change in portal flow velocity. Another study showed that preoperative splenic vein diameter was a risk factor for postsplenectomy portal or splenic vein thrombosis, which is supported by the present results. A widened portal diameter might be involved in the development of PVT. However, multivariate analyses failed to detect a correlation between portal diameter and PVT. The present results indicated that splenic thickness was associated with PVT occurrence. Splenomegaly has been associated with portal hypertension, with greater splenic enlargement resulting in more aggravated portal hypertension. Further, the resultant risk increases during PVT development due to the presence of aggravated portal hypertension. Thus, the occurrence of splenic thickening may contribute to the formation of PVT. The current study had some limitations. First, all the data for this study were obtained from a single center, involving a relatively small number of subjects with or without PVT. A larger sample needs to be investigated to understand the features of PVT. Second, small numbers of patients with aetiologies other than hepatitis B was also a limitation of this study, and more cirrhotic patients with aetiologies other than hepatitis B should be enrolled into the study in future clinical practice. In conclusion, this single centre study showed PVT in 24.7 per cent cirrhotic Chinese patients, with PVT mainly occurring in patients with post-hepatitis B liver cirrhosis. The lower levels of haemoglobin and BPCs as well as splenic thickening were no underline associated with PVT. Splenic thickening is likely to increase the risk of PVT. Further studies are required to evaluate the relevance of these parameters in a larger population and to identify additional risk factors.

This prospective study included 32 consecutive women with 51clinically symptomatic uterine fibroids (excessive and irregular menstrual bleeding, pelvic pain, pressure, urinary or bowel problems and anaemia) who attended GSL general hospital, Rajahmundry, Andhra Pradesh, India, from February 2011 to October 2011. They were first clinically examined by a gynecologist followed by ultrasound scan of the pelvis and based on the clinical symptoms and ultrasound findings they were subjected to screening MRI to assess the feasibility for MRgFUS treatment. The MR images of the pelvis were used to diagnose the presence of fibroids, their blood flow (perfusion volume) location, size and volume, characterize the tissue, and diagnose other uterine or pelvic conditions. These women were assessed for clinical symptoms and health related quality of life at the time of recruitment for the study and after treatment at one, three and six month follow up using a validated uterine fibroid symptom and quality of life questionnaire (UFS-QOL). The questionnaire uses eight questions assessed on a 5-point Likert scale to assess both bleeding and bulk-related symptoms due to uterine fibroids. Thus, the maximal raw score for the symptoms severity score is 40 points. However, the transformed score is typically reported on a single 100-point scale, with 100 points indicating maximal symptomatology. In validation studies normal women had an average transformed symptom score of approximately 20 points, and women with uterine fibroids had an average score of 40 points. The women with a symptom severity score (SSS) of more than 41 points on UFS-QOL questionnaire qualified for the study. A 20 point reduction in SSS was considered as significant improvement. Immediately after the procedure and at six month follow up contrast enhanced MRI scan of the pelvis was performed to assess the nonperfused volume (NPV) or treated area and reduction in fibroid size. NPV ratio (NPV ratio = nonperfused volume/perfused volume expressed as percentage) was calculated and its relation to symptom relief as per UFS-QOL was assessed at 6 months follow up. The primary outcome of the study was symptom relief at one, three and six months and secondary outcome was reduction in the fibroid size at 6 months follow up and its relation to NPV ratio. The inclusion criteria were age above 18 yr, weight less than 110 kg, not more than four fibroids, fibroid size not more than 10 cm, location not more than 12 cm from anterior abdominal wall and at least 4 cm away from bone and nerve bundles. The exclusion criteria were pregnancy, massive scarring over lower abdomen, pelvic or systemic disease, contraindications for MRI like cardiac pacemaker, calcified fibroids and pedunculated fibroids. Written informed consent of the patients was obtained. This study protocol was approved by the institutional ethics review committee. Selected patients reported to treatment centre in fasting state. An intravenous line and Foley's catheter were inserted. Conscious sedation was administered with fentanyl citrate or pentazocine. During MR guided focused ultrasound patients lied in the prone position on a modified MR gantry. The FUS machine integrates fully with a 1.5 T MR scanner (GE, Milwaukee, USA) and the transducer is situated within a bath of degassed water in the mid-section of the table. This allows a direct acoustic pathway from the transducer into the target fibroid, which was directly positioned above. To carry out MRgFUS safely an adequate acoustic window for the sound wave pathway was created which did not traverse abdominal wall scars or any loops of bowel. The goal of the therapy was to ablate maximum volume of the fibroid without damage to surrounding structures. The treatment parameters like fibroid volume, number of sonications, acoustic power and sonication frequency were initially determined by ultrasound ablation system (EXABLATE-2010, Insightec Haifa Israel). These were modified later on the basis of temperature mapping and feedback from patients regarding sonications related discomfort. On an average it took about 3-4 h for treating a 6 cm size fibroid. : Statistical analysis was performed using SPSS software trail version 16.0, USA. Symptom severity scores (SSS) were analyzed using ANOVA with Turkey's test. Student's paired t-test was used to compare the fibroid volumes before procedure and at six months follow up. Pearson coefficient of correlation was used to study the correlations between NPV ratio and reduction in fibroid size immediately after the procedure and at six months follow up. For all statistical analyses <0.01 was considered significant. italic sup #text The outcome of the MRgFUS treatment in this study was on the basis of improvement in symptom severity scores and improvement in quality of life after the procedure. The results showed significant improvement in the SSS after the procedure at one, three and six months compared to symptom scores at the time of initial presentation. Although the relief of symptoms and improvement in the quality of life is promising about the effectiveness of this new non invasive ablative treatment in short term, its long term efficacy and sustainability is yet to be established. The reductions in SSS and fibroid volumes at the end of six months follow up after the MRgFUS treatment correlated positively with the NPV ratio. These observations were in accordance with the findings of other studies conducted outside India. The common less serious adverse effects included leucorrhoea, fever, localized pain, erythema and swelling which disappeared in a few days. Serious adverse effects included, necrosis of non targeted tissue like bowel and bladder with perforation, nerve damage or haemorrhage in the treatment area, and skin burns resulting in ulceration and scar formation. various studies reported adverse effects in 1-3 per cent cases. Some investigators have correlated the therapeutic efficacy of MRgFUS with fibroid intensity in the pretreatment MRI images. Some other studies revealed that the treatment outcome was not influenced by the phase of menstrual cycle. But in this study these aspects were not taken into consideration. Long term efficacy of this treatment is yet to be tested and compared with other available alternative minimally invasive treatment options. So far 24 months follow up studies are available which show lower recurrence rates (20%) when the treated volumes were around 60 per cent of the fibroid volume. The other treatment options available for uterine fibroids include hysterectomy, open myomectomy, laparoscopic and hysteroscopic myomectomy. Surgically guided thermoablative techniques like myolysis and cryoablation have shown variable response. Uterine artery embolization though widely used and less invasive can cause post embolization infection, fever, chronic vaginal discharge, fibroid extrusion and ovarian failure and requires expensive equipment and expertise. The advantages of MRgFUS over other therapies is that it is a day care procedure, can be conducted under conscious sedation, patient can communicate with the treating physician during treatment, can stop treatment herself and resume her daily activities from the next day of treatment. The other advantages with this treatment are uterus is conserved with chances of future pregnancy and it can be carried out even in anaemic patients and those who cannot withstand surgery. Spontaneous conception after treatment with MRgFUS for uterine fibroids has been reported. Though this study included five subjects with uterine fibroids and infertility, so far there was no positive pregnancy outcome. MRgFUS is also emerging as a treatment option for various other medical conditions like carcinoma breast, prostate, liver and palliation of bone metastasis. The limitating factors for widespread application of this modality of treatment are less than 50 per cent of subjects with uterine fibroids are eligible for treatment with this procedure, high cost of equipment, need for experienced radiologists and gynaecologists well versed with the procedure, and need for more comparative studies with other modalities of treatment to establish its long term efficacy. To conclude, MRgFUS is a novel non invasive treatment option for treating uterine fibroids. It is relatively safe and effective in selected patients with minimal serious adverse effects and can be repeated if necessary. The long term efficacy and durability of this treatment is yet to be tested and compared with other presently available minimally invasive treatment options. This study included a limited number of subjects and studies involving more number of subjects are required in future to establish its safety and efficacy.

: Patients with EH (n=170) and normotensive controls (n=154) were consecutively selected from hypertension outpatient clinic and medical center, respectively, affiliated to the hospital of Zhejiang Medical College, Hangzhou, PR China from February to August 2010. The normotensive controls were the people who received common physical examination in the hospital. All subjects gave written informed consent for participation in the study and the study protocol was approved by the medical ethics committees of the Zhejiang Medical College. The EH group consisted of 97 male and 73 female patients with mean age of 57.4 ± 24 yr. Subjects with a history of diabetes mellitus and renal failure were excluded. Patients on antihypertensive drugs were excluded. The control group consisted of 81 males and 73 females with mean age of 56.9 ± 9.0 yr. These individuals came for routine physical examination in the hospital and had no family history of EH. The blood samples (3 ml) were collected after overnight fasting at morning without stasis in EDTA vacutainers. The patients were selected according to the Seventh Report of Joint National Committee on Detection, Evaluation and Treatment of High Blood Pressure (2003 JNC7) (. systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mmHg were used as indication of hypertension). Patients with secondary hypertension, diabetes, abnormal liver and kidney function were excluded. All subjects were asked about smoking status. Body mass index (BMI) was calculated with the formula, weight (kg)/height (m). : Plasma high-sensitivity C-reactive protein (hsCRP) levels were measured within 2-3 h after collection of blood samples by high sensitivity enzyme immunoassay (Dade-Behring, Marburg Germany). Low-density lipoprotein (LDL) and high-density lipoprotein (HDL) were assessed using Flex Reagent Cartridges (Dade-Behring, Marburg Germany). : Genomic DNA was extracted from peripheral blood using Blood Genome DNA Extraction Kit (TAKARA Biotechnology Co. Ltd., Dalian, Japan). The genotyping of G614T polymorphism of was done by PCR-RFLP technique. DNA fragments were amplified in a total volume of 50 μl PCR reaction mixture containing 10 × buffer 5 μl, 2.5 mM dNTP 4 μl, forward primer (5’-ctcctttgctagtgacggtgattc-3’) 0.5 μl, reverse primer (5’-gacttggcactgcttccattcgc-3’) 0.5 μl, double distilled water (DDW) 37.75 μl, Taq polymerase (TAKARA Biotechnology Co. Ltd., Japan) 0.25 μl and DNA 2 μl. Amplification was carried out under the following conditions: one cycle of 5 min at 95 °C, 35 cycles of 35 sec at 95 °C, 35 sec at 56 °C, and 35 sec at 72 °C, followed by 5 min at 72 °C. A mismatch, which introduces a Sau96I restriction site, was placed in the 3’ region of reverse primer to enable genotyping via restriction digest (the mismatched nucleotide is underlined and in heavier version. Amplified products were digested with Sau96I enzyme (NBE Inc., US) at 37 °C for 16 h. All products were loaded onto 3 per cent MS-6 agarose (TAKARA Biotechnology Co. Ltd., Dalian), and electrophoresed. Bands were visualized and typed after GelRed staining (Biotium Company, US). The length of PCR amplification product with G614T was 147 bp. The Sau 96I restriction enzymes were used to distinguish 614G/T, resulting in 122 bp and 25 bp fragments in the presence of the G allele. The polymorphism analysis was performed by two persons independently in a blind fashion. More than 10 per cent of the samples were randomly selected for confirmation, and the results were 100 per cent concordant (). : The SPSS 16.0 software (SPPS Inc. USA) was used in this study. The expected frequencies of the G614T genotypes were tested for the Hardy-Weinberg Equilibrium. Statistical differences for the distribution of genotypes G614T between EH and control groups were assessed by χ2 test. The relationships of the genotypes with the clinicopathologic parameters of patients were tested by -test. Logistic regression analysis was performed to assess the independent effect of each risk factor on the occurrence of EH. The odds ratio (OR) was calculated as estimators of relative risk, together with their 95% confidence intervals (95% CI). The difference in the clinicopathologic parameters according to G614T genotype distributions was compared using ANOVA test. No significant differences were observed with respect to age, sex, smoking status, serum HDL, CRP levels between hypertensive subjects and the controls. However, the EH group had a significantly (<.05) higher blood pressure (), BMI and the serum LDL concentration compared to the control group. shows the genotype distributions and the allele frequencies of G614T polymorphism in the two groups. A significant difference in genotype distributions between hypertensives and controls was noted with the observed power of 0.899. The TT genotype was significantly (<.05) higher in hypertensives as compared with controls (TT: 31.18 vs. 16.88%). From the “Genetic Power Calculator”, the genotypic risk for TT was 12.27 per cent. For G and T allele frequencies, there was a significant (<.05) difference between hypertensives and controls as well (G: 41.76 vs. 57.47%, T: 58.24 vs. 42.53%). The frequencies of the genotypes were significantly different between the patient and control groups (<0.05). Both genotypes and alleles frequencies in control and EH groups were in Hardy-Weinberg equilibrium. By univariate analysis, it was found that EH was influenced by BMI, LDL and gene 614T allele. By multivariate logistic regression, T allele (OR=2.217, 95% CI: 1.243-3.953, =0.007<0.05/3=0.017) was the independent risk factor of essential hypertension (). The demographic and clinical characteristics in EH group were studied according to the G614T genotype distributions. It was found that the concentration of plasma LDL differed significantly between the three genotypes in EH group (). But the level of LDL was not associated with gene G614T polymorphism in the control group (data not shown). When subdivided according to gender, no significant association was observed with respect to clinical characteristics in female patients, while LDL was significantly elevated in the male patients (<0.05). Though there was lack of statistical significance, the plasma CRP concentration was lower in patients who were T allele carriers than in those with GG genotype. G614T polymorphism plays a potential biological role in the development of high blood pressure. Several previous studies revealed that this variant was a likely candidate for studying association with hypertension status. In this study, there was a significant difference in α-adducin genotype between EH and control groups. The levels of LDL were significantly increased in persons with GT and TT genotypes in EH group as compared with GG genotypes. The frequency of α-adducin gene G614T varies in different populations. In our study, the frequency of α-adducin gene 614T in south China was 42.53 per cent in normotensives and 58.24 per cent in hypertensives. The frequency of α-adducin gene 614T allele in hypertensives and controls were 46 and 48 per cent in Chinese, 54 and 60 per cent in Japanese, 59 and 61 per cent in Koreans and 53 and 33 per cent in Americans. Several studies have demonstrated the potential involvement of α-adducin in the pathogenesis of EH, but no definite conclusion could be drawn (). Factors like ethnic diversity, sample sizes are expected to be the cause for these inconclusive results found. A sample of 904 African Americans (from Jackson, Mississippi) was examined for α- adducin gene 614T association with hypertension. The results showed that African Americans not only have a higher prevalence of hypertension, but also the condition strikes at an early age, with greater severity, often ending in death when compared with whites in the United States. No association was found between hypertension status and the G614T polymorphism in American Blacks, white population of USA, Australia. A case-control study conducted in a large population from Sassari, Italy, did not find any association of the α-adducin T allele with hypertensives while the study in a large population from Milan, Italy confirmed a positive association. No significant association was found in a well characterized Japanese population and Korean population that was also Asian descent. G614T polymorphism in Chinese Han population of Shang Hai also reported absence of association with hypertension. Adjusted for the conventional risk factors of hypertension, alpha-adducin polymorphism has been shown to play an independent role on systolic blood pressure in Indians living in Car Nicobar Island. Our study showed that there was significant association between G614T substitution of gene and hypertesion phenotype of EH patients in South China. In our study, the concentration of plasma LDL was higher in hypertensives carrying at least one 614T allele than in GG homozygotes. Furthermore, no association of LDL concentration and the mutation was observed in female hypertensives. Males with 614T allele had higher LDL, suggesting risk for cardiovascular diseases. Castejon reported that the concentration of plasma LDL was significantly different in the GG and GT healthy groups. However, because of the small sample size (n=90), TT homozygotes were not detected. In our study, TT homozygotes were detected and EH patients with T allel had higher plasma LDL level. The mechanism by which polymorphism influences LDL levels is not clear and further study is needed to investigate underlying factors in detail. In conclusion, our results indicate towards genetic association between α-adducin gene G460T polymorphism and hypertension. Further studies need to be done on the association of this polymorphism with hypertension in different ethnic groups with larger samples.

: Data for this retrospective study were retrieved from PHPT registry of the department of Endocrinology, Postgraduate Institute of Medical Education & Research (PGIMER), Chandigarh, India. All cases of histopathological proven PHPT from January 2001 to December 2009 were included in the study. PHPT was defined as inappropriately elevated PTH in presence of hypercalcaemia. Preoperatively parathyroid adenoma was localized in most with either ultrasonography and/or sestamibi scintigraphy. Patients with secondary or tertiary hyperparathyroidism were excluded from the study. The study protocol was approved by the ethics committee of the Institute. : Subjective symptoms like anorexia, nausea/vomiting and abdominal pain were reported as present if patient reported these symptoms at the time of presentation. All PHPT patients were evaluated for the presence of gallstone and pancreatitis. Gall stone(s) was defined as either imaging evidence of gallstones and/or a history of cholecystectomy. Acute pancreatitis was diagnosed based on the presence of typical symptoms, elevation of pancreatic enzymes more than three times and positive radiological evidences. Renal stone disease was defined as either imaging findings consistent with renal stone disease or history of surgery for renal stone(s). Anaemia was defined as haemoglobin <13 g/dl in male and less than 12 g/dl in female. : Serum calcium (8.6-10.2 mg/dl), inorganic phosphate (2.7-4.5 mg/dl), albumin (3.4-4.8 mg/dl), and alkaline phosphatase (40-129 IU/l) were measured by autoanalyzer (Roche diagnostics, Modular P 800, Indianapolis, USA). Serum intact PTH (15-65 pg/ml) was measured by chemiluminescence assay (Roche, ELYCYS 2010) and 25 (OH) D (11.1-42.9 ng/ml) was measured by radioimmunoassay using commercially available kits (DiaSorin Inc., Stillwater, MN, USA). : The statistical analysis was carried out using SPSS 15 (Chicago, USA). Continuous variables were described as mean ± SD and categorical variables were described as frequencies and proportions. Chi square test was applied to assess significant difference in symptoms between men and women. Paired -test was applied to find out the difference in biochemical parameters in presence of different symptoms. Binary logistic regression analysis was performed to calculate odd ratio (OR) for development of gallstone and pancreatitis in relation to high calcium, low phosphate and elevated PTH levels. All statistical tests were two-sided and performed at a significance level of <0.05. One hundred and fifty three patients with PHPT were analyzed in the present study; 46 were men (30%) and 107 (70%) were women. The mean age of patients with PHPT was 39.2 ± 13.9 yr (range; 11-74 yr, 36.4 ±13.7 yr for men and 40.4 ±13.8 yr for women). The lag time between the symptoms suggestive of PHPT and diagnosis was 46.3 ± 39.9 months (Median 36, 37-54; 95% CI). The mean (±SD) preoperative serum calcium, phosphate, 25 (OH)D and PTH levels were 11.5± 1.6 mg/dl, 3.1±0.9 mg/dl 27.6 ± 28.3 pg/ml and 735.2 ± 622.9 pg/ml (Median- 526 pg/ml), respectively. The most common location of parathyroid adenoma in our study was left inferior parathyroid region (36%) followed by right inferior parathyroid region (22%). One patient had parathyroid carcinoma and none of these patients had parathyroid hyperplasia. The mean weight of parathyroid adenoma was 6.2 ± 5.9 g (Median-4.0, 4.2-8.2; 95% CI). : The most common symptom noted at the time of diagnosis was fatigue 105 (69%) followed by bone pain 89 (58%), renal stones 75 (49%) and fracture 60 (39%) while anaemia was present in 89 (58%) of the PHPT patients. Eighty per cent of PHPT patients had at least one symptom or sign related to gastrointestinal system. The most common gastrointestinal manifestation was abdominal pain 66 (43%). Other symptoms pertaining to gastrointestinal system were constipation 55 (36%), nausea and/or vomiting 46 (30%), anorexia 40 (26%), and dyspepsia 37 (24%). Thirty four (22%) patients had history of either imaging evidence of gallstone and/ or cholecystectomy. Twenty two (14%) patients had imaging evidence of gallstones and 27 (18%) had history of cholecystectomy for symptomatic gall stone disease. Twenty seven patients (18%) had imaging and biochemical parameters consistent with pancreatitis. Of the 27 patients with pancreatitis, none had history of alcohol abuse or other risk factor known to cause pancreatitis, four had gallstone disease but underwent cholecystectomy prior to pancreatitis. Twenty six of these (96.29%) had two or more episodes of pancreatitis before the diagnosis of PHPT was established. : Gastrointestinal symptoms were comparable in both sexes except anaemia (<0.05) and gallstones (<0.05) were more prevalent in women while abdominal pain (<0.05) and pancreatitis (<0.05) were more common in men. History of cholecystectomy was higher but statistically insignificant in women compared to men. Pancreatitis was more common in men compared to women despite the prevalence of gallstones being greater in women (). : Patients of PHPT with nausea/vomiting had higher serum phosphate (<0.05) but comparable serum calcium and PTH levels than those who did not have nausea/vomiting. Constipation in patients with PHPT was associated with low phosphate (<0.01) but not with hypercalcaemia. Pancreatitis was more common in those who had low PTH levels <0.01 (). : Using binary logistic regression analysis, serum calcium [odds ratio (OR)-0.88 (0.6-1.1, 95% CI), =0.4], phosphate [OR= 0.81 (0.4-1.4, 95% CI), =0.5] or PTH levels [OR=1 (0.9-1.1, 95% CI), =0.6] were not found to be risk factors for development of gallstones, however, higher calcium levels [OR=1.3 (1.0-1.9, 95% CI), =0.04] but not the low phosphate [OR=1.1 (0.6-1.5, 95% CI), =0.6] were associated with high chance of pancreatitis while high PTH levels [OR=0.9 (0.90-0.99, 95% CI), =0.02] was found to be protective factor for pancreatitis. : Seven days post curative parathyroidectomy, mean serum calcium, phosphate and PTH levels were 8.9±1.4 mg/dl, 2.9±0.8 mg/dl and 84.7± 11.9 pg/ml, respectively. Three months post curative parathyroidectomy, persistent or recurrent symptoms and signs were as follows; abdominal pain, anorexia and acid peptic disease like symptoms were noted in two cases each, nausea and/or vomiting in four. Of the 27 patients with pancreatitis and PHPT, 23 had no recurrence of pancreatitis after mean follow up of 3.7 years of successful parathyroidectomy, two patients had recurrence of pancreatitis; probably attributed to severe calcification involving the pancreatic ducts as well as parenchyma and two patients died due to obstructive uropathy leading to end stage renal dysfunction. This study reports higher prevalence of gastrointestinal symptoms among patients with symptomatic PHPT. Gallstone disease was more common in women while pancreatitis was more common in men. In majority of patients, gastrointestinal symptoms resolved within three months after successful parathyroidectomy. Pancreatitis did not re-occur after successful parathyroidectomy in majority (85%) of patients. The women: men ratio was 7:3 with mean age of 39.2 ± 13.9 yr at the time of diagnosis. All PHPT patients were symptomatic. These findings are in consonance with other studies from India suggesting that Indian PHPT patients are younger at the time of diagnosis and have symptomatic disease compared to PHPT patients in western countries. The possible reason for more severe and symptomatic PHPT in our study may be due to a delay in diagnosis as evident from lag time (between symptoms suggestive of PHPT and diagnosis of PHPT) of 46.3 ± 39.9 months. The most common gastrointestinal symptoms were abdominal pain and constipation in this study. High incidence of abdominal pain was noted by us and others previously in small number of PHPT subjects. The most common digestive symptom noted was constipation in earlier studies. However, one study from western India reported absence of gastrointestinal symptoms. These differences in symptomatolgy are due to difference in number of subjects, population and retrospective design of these studies. Various mechanisms have been proposed for abdominal pain in patients with PHPT like nausea and/or vomiting, dyspepsia, referred pain due to renal colic, symptomatic gallstone and pancreatitis. Anorexia and nausea/vomiting were present in nearly one third of patients with PHPT. The exact mechanism for nausea and vomiting in PHPT is not known but animal experimentation suggested that it could be because of increased calcium ion concentration in the sympathetic ganglia, impeding transmission of afferent stimuli and diminishing efferent discharges. Hebert have shown that the Ca-sensing receptor is expressed along the entire gastrointestinal tract and its function was suggested to regulate acid secretion in stomach and fluid secretion in colon. However, we did not find significant difference in the level of serum calcium between patients with and without anorexia, nausea/vomiting and constipation. High occurrence of gallstone was seen in patients with PHPT and it was more common in women. Gallstones as such are more common in women compare to men and that may be the possible reason for higher prevalence of gallstones among women in the present study. Pancreatitis was observed in 18 per cent patients in the study. The prevalence of pancreatitis was reported to be greater in Indian patients with PHPT than those in western countries. Cholelithiasis is a common cause of pancreatitis. However, in this study, gallstones were seen more in women and pancreatitis was higher in men. The reason for this paradox is not known, but this finding implicates that factors other than gallstone are more important for development of pancreatitis in patients with PHPT. High calcium has been implicated in the pathogenesis of pancreatitis but exact mechanism is yet to be elucidated. Plausible mechanisms include calcium-phosphate deposition in the pancreatic ducts; calcium-dependent conversion of trypsinogen to trypsin; increased permeability of pancreatic duct due to hypercalcaemia; and an apparent direct toxic effect of PTH on the pancreas. A large US population-based study revealed no increase in pancreatitis among PHPT patients. We noted the disappearance of gastrointestinal signs and symptoms within three months in majority of patients after successful parathyroidectomy. Chan also reported significant reduction in symptom rates after parathyroidectomy. The persistent symptoms of nausea, vomiting, fatigue and abdominal pain in a few patients in our study might be explained by concurrent renal stone disease and or renal failure. In conclusion, this study reveals high frequency of gastrointestinal symptoms in patients with symptomatic PHPT. There was not much gender difference in gastrointestinal symptoms except higher occurrence of gallstones in women and pancreatitis in men. The biochemical profile was also similar in those who had and did not have gastrointestinal symptoms.

: The partnership between NGOs and government health facilities to provide STI treatment to FSWs was implemented in 13 inland districts of Andhra Pradesh where the adult HIV prevalence has been estimated to be nearly 1 per cent. Unlike the coastal districts, these inland districts are mainly rural, and the population is dependent on subsistence agriculture. There were considerable variations in economic and development indicators across these districts. The HIV prevention programme in these 13 districts was started in 2004 providing services to about 15000 FSWs initially. The programme expanded its coverage to include another 13,000 FSWs by 2006. These FSWs were located across 139 (sub-districts) including 46 rural . Mapping and needs assessment exercise in these districts highlighted the lack of access to STI treatment facilities as a major barrier in HIV prevention efforts, particularly in rural areas with the existing model of static clinic and network of preferred providers. : The aim of initiating the partnership was to provide effective STI services to FSWs in the underserved rural areas and strengthen the institutional capacity of health providers. The availability of government health facilities, mainly PHCs, in the rural areas provided an opportunity to establish partnerships between government facilities and local NGOs. Partnerships with government health facilities were piloted in six headquarters of a district. Two NGOs were selected as implementing partners and were trained in outreach activities and the provision of STI clinical services. National AIDS Control Organization (NACO) intervention guidelines for FSWs were followed for selecting health facilities and service providers. Discussions were held with local sex worker groups to obtain their opinions and reactions regarding locations and the acceptability of the health facilities. While the process of building partnerships with doctors from government health facilities was underway, local NGOs negotiated with district health officials to obtain permission for engaging doctors from these facilities. The team from State level NGO also met with officials from the State AIDS Control Society (SACS), the government agency managing the State HIV prevention programme to seek help in advocating with the State Directorate of Health Services for partnership arrangement approvals. Auxiliary nurse mid-wives (ANMs) were recruited at each facility to support the doctors in providing STI services. ANMs were trained to conduct routine clinic operations, including drug procurement, clinic system management, infection control, documentation, and risk-reduction counselling. The STI technical officer from the State level NGO visited the government health facilities to orient doctors about syndromic case management and STI clinic operational guidelines. Efforts were also made to sensitize doctors and ANMs on issues related to FSW, in particular, the need for confidentiality and respect for patients’ rights. NACO recommended guidelines on management of STI was used to train medical doctors and ANMs in the study areas to ensure uniformity in trainings. Doctors were provided with appropriate treatment algorithms and other written materials on syndromic case management. : Soon after permissions were obtained from the State and district health officials for the government health facilities to engage in this initiative, the model was upscaled to all the rural areas with similar protocols as followed in the pilot phase. In addition to regular trainings for doctors and nurses, two annual reorientation training sessions were conducted in every six months. : The primary role of local level NGOs was to negotiate with the district authorities to obtain permission for engaging doctors from government health facilities and allow them time to attend training sessions. They were also responsible for recruiting ANMs, procuring drugs and supplies, and reporting to the State level NGO. The peer educators and outreach workers at these NGOs met groups of sex workers at the places of solicitation (hotspots of sex work activity) and provided information on STI/HIV, motivated them to seek services from government health facilities, provided free condoms, and conducted verbal screening for tuberculosis. Staff at the State level NGO were responsible for training the doctors, supporting the establishment of clinic systems according to the operational guidelines, facilitating the supply of drugs, testing kits, and other materials, and providing supportive supervision. The Health department at the State and district approved the use of the clinic facilities and the services of doctors. District health officials were responsible for ensuring doctors’ availability at the health facilities and giving them permission to attend trainings organized by State level NGO. : The basic minimum services provided in the partnership clinics included quarterly medical check-ups, presumptive treatment, STI treatment using the syndromic approach, follow up for STI cases, risk-reduction counselling including condom promotion, verbal screening and testing for tuberculosis, biannual syphilis testing and referral for HIV testing. All these services and necessary medicines were provided free of cost to FSWs. : Weekly meetings were organized by NGO project coordinators to ensure coordination between clinic and outreach activities. If any FSW was diagnosed with an STI, the ANM discussed the case with the peer educator to ensure compliance with treatment and follow up visits to the clinic. : Technical officers from the State level NGO visited the health facilities every quarter to monitor the quality of clinical services and provide technical support. The monitoring team made regular assessment of the clinic performance and systems using a validated standardized quality-monitoring tool. The performance of a clinic was assessed based on the number of FSWs accessing clinical services, diagnosed with STI, provided presumptive treatment, referred to other facilities for services such as HIV testing and syphilis testing. Two data sources were used to examine the accessibility and utilization of the partnership model for STI service delivery: programme monitoring data, and behavioural tracking survey. : The State level NGO developed and defined programme monitoring indicators and established monitoring and information system (MIS) to gather data periodically on different indicators related to programme inputs and outputs. The MIS designed for this programme was independent of the structure of NACO's MIS. Programme monitoring data were collected monthly at the local level NGO which were aggregated in paper formats and consolidated by the State level NGO. FSWs who accessed services at the programme clinics were assigned an unique identification number to track their service utilization during the programme. The staff from State level NGO provided technical support to local level NGO to ensure integrity and quality of data as well as flow of data and maintenance of MIS. : Independent of the programme monitoring data, a cross-sectional survey, named as behavioural tracking survey (BTS), was conducted among FSWs in five districts: Ananthpur, Warangal, Khamam, Kurnool and Medak. The objective of BTS was to monitor the key components of the HIV prevention programme: safe sex behaviour, STI treatment seeking behaviours and community mobilization. Data were collected using a two-stage sampling design, wherein FSWs’ hotspots were selected during the first stage and FSWs were selected randomly from each selected hotspot during the second stage. A sample size of 400 FSWs was estimated for each district, allowing for detection of an absolute difference of 15 per cent or more from the assumed value of 50 per cent for consistent condom use with all clients, with 95 per cent confidence, 90 per cent power and a design effect of 1.7. Detailed sampling procedures for selection of sites and sex workers are discussed elsewhere. At the end of the survey, 2389 FSWs were approached, of whom 403 either refused to participate or withdrew during the interview. This resulted in a total analytical sample of 1986 FSWs with a response rate of 83 per cent. Sample weights were calculated to account for the unequal selection probability of respondents and non-response rates. The institutional review boards of Family Health International (FHI) and Karnataka Health Promotion Trust (KHPT) reviewed and approved the protocols of BTS. A comprehensive informed consent process was followed and no names or identifying information was recorded. : Indicators were developed to assess the utilization and acceptability of services from the partnership model. Six indicators from the programme monitoring data were used and these were monitored annually from 2007 to 2010. The indicators used to assess service utilization from programme monitoring data were: number of FSWs who visited clinic, FSWs who visited clinic with STI-related symptoms, FSWs who were provided with presumptive treatment, FSWs who underwent speculum examination, FSWs who were screened for tuberculosis, and number of FSWs who were tested for and diagnosed with syphilis. BTS collected wide range of information on FSWs’ behaviour including experience of STI-related symptoms, service utilization and attitude towards health facilities. Single item questions were asked to understand if they had experienced one of the following STI-related symptoms in the six months prior to survey: genital ulcer/sore, genital discharge, or lower abdominal pain. To understand service utilization in the last six months, single item questions on number of times visiting STI clinics (recoded into “0” if less than two visits; else coded as “1” to indicate at least two visits), and HIV test conducted were asked. FSWs were asked single item questions to understand their attitude towards utilization of government health services. Responses were as follows: not at all confident, somewhat confident, very confident, and completely confident. FSWs who responded either ‘very confident’ or ‘completely confident’ were combined to represent confident attitude and coded as ‘1’, and the remaining were coded as ‘0’. The FSWs were grouped into three groups based on information on the type of clinic providing services in the survey area. These groups were: area covered by partnership clinics (coded as ‘2’), area covered by non-partnership clinics (coded as ‘1’), and area covered by other clinics (coded as ‘0’). ‘Non-partnership clinics’ included either static clinics or preferred-provider clinics; and ‘other clinics’ were clinics functioning without support from State level NGO. This variable was considered as the key independent variable for multivariate analysis. : Programme monitoring data are presented in terms of absolute numbers and percentages. Bivariate and multivariate analyses were conducted to present the results of the behavioural tracking survey. A series of multiple logistic regression models were generated to examine the differences in programme outcomes, and attitude towards government clinic and service utilization. The logistic regression models were adjusted for age (continuous), educational status (no formal education versus some formal education), marital status (currently/formerly/never married), duration in sex work, solicitation location (home, brothel, street, and phone), and residential location (rural, semi-urban, and urban). All the analyses were performed using Stata 11.1 (Stata Corp, USA). Programme coverage and utilization of STI services from government health facilities under the partnership model during 2007-2010 are presented in . The data indicated that the number of FSWs visiting government health facilities increased sharply after the initiative was scaled-up in 2008. During 2008-2010, there was a decline in reporting of STI symptoms (from 54 to 18%) and presumptive treatment (from 43 to 21%) among FSWs. Notable increases were observed in the proportion of FSWs who underwent speculum examination, TB screening, syphilis and HIV testing. The effects of partnership model clinics on experience of STI-related symptoms, STI treatment seeking behaviour and other clinical outcomes among FSWs are presented in . Experience of any STI-related symptom, including genital ulcers/sores, genital discharge, and lower abdominal pain, were similar whether FSWs belonged to areas with partnership clinics or non-partnership clinics. Among those who experienced at least one STI symptom in the last six months, FSWs from areas with a partnership clinic were three times more likely to seek treatment from the government health facilities than those belonged areas where clinics were not supported by State level NGO (71 vs. 41%, AOR: 3.23, 95% CI: 1.72–5.88). There were no significant differences in frequency of clinic visits by FSWs in areas with or without a partnership clinic. Moreover, the odds of HIV testing were approximately two times higher among FSWs in areas with a partnership clinic than in areas with areas with clinics without State level NGO support (47 vs. 34%, AOR: 1.79, 95% CI: 1.20-2.63) (). Further, FSWs were four times more likely to receive STI services from clinics despite being ill treated by health workers in areas with partnership clinics than FSWs from areas with other clinics (60 vs. 29%, AOR: 3.57, 95% CI: 2.33-5.26) (). Compared to FSWs from areas with non-partnership clinics, those from areas with partnership clinics were more likely to be confident in accessing services from government facilities, even if they were identified as sex worker (57 vs. 49%, AOR: 1.28, 95% CI: 1.05-1.56) or even if the project services were terminated (63 vs. 55%, AOR: 1.41, 95% CI: 1.15-1.69). Compared to FSWs from areas with programme supported statis clinics, those from areas with partnership clinics were more likely to perceive fair treatment at government hospitals (10 vs. 24%, AOR: 3.03, 95% CI: 1.72-5.56) (). The study findings indicated that providing STI treatment services through partnership with government health facilities improved the utilization of such facilities among female sex workers in Andhra Pradesh. This also showed that the outreach activities of NGOs and peer educators successfully motivated FSWs to visit the partnership clinics. These findings were consistent with results from another study in Andhra Pradesh, which indicated that FSWs in intervention areas had a positive attitude to seeking services from government health facilities. The monitoring data indicated that certain services were better utilized as compared to certain other indicators. For example, proportion of FSWs who underwent speculum examination was higher than those who were screened for tuberculosis in 2007 and 2008. This can be due to the fact that the programme has emphasized on FSWs’ speculum examination since the inception; however, efforts for screening of tuberculosis were generated much later. Although no specific information on cost involved was collected in these partnership clinic model, information from secondary sources suggested that the cost invested per sex worker was lower in these clinics as compared to that in static clinics. Analysis of costing data suggests that the State level NGO spent around ₹ ≅ 600 INR (12 USD) per sex worker of which 14 per cent was spent on STI services. As the partnership clinics required less investment in developing infrastructure, it would have cost lesser than the average cost spent indicating cost efficiency of this model of STI service delivery. Although the data suggest that the partnership with government health facilities has been relatively successful, the effectiveness of such partnerships is highly context specific. The criticism of partnership between private and public facilities is well known, and it is noteworthy that in rural under-served areas, only a few alternative approaches beyond the partnership model exist for the delivery of health services. In resource-constrained settings, innovative actions are required to overcome deficient health facilities, and in many of those resource-poor rural areas, government health facilities are the only available resources. In a partnership mechanism in rural Karnataka, the government had assigned NGOs to manage staff and operate some PHCs. Another example of successful partnership programme was the program implemented by the Society for Education, Welfare, and Action (SEWA) rural in Gujarat. The private-public partnership clinics described here were located within the government-operated PHCs or CHCs and utilized government doctors and infrastructure; these facilities were strongly supported by the NGOs in terms of training, outreach and monitoring including data collection activities. This arrangement was more flexible than those adopted in the ‘conventional models’, such as SEWA rural and other NGO-dominated arrangements. The study findings indicate scope for further experimentation with flexible partnerships, which does not require an NGO to assume complete control over the operation and management of the clinics. No strict single formula needs to be followed to build partnerships with government health facilities. Further, building a partnership with the government health facilities for STI service delivery was not always easy. Alongside the challenges faced by the programme due to frequent transfers of doctors, some doctors were unwilling to continue the partnership arrangements, which resulted in the loss of a few clinics in 2009. In some ‘lost clinic’ cases, FSWs complained that the clinic provider had a negative attitude towards them. While service utilization by sex workers in areas with partnership clinics was improved, there were several challenges to this approach. These included: doctors’ attendance at the training sessions was irregular; frequent transfers of doctors across health facilities requiring a significant number of trainings; and frequent absence of doctors, particularly during senior government officials’ visits. These issues need to be taken into consideration if the programme needs to be scaled-up. Overall, the HIV prevention initiative in Andhra Pradesh has utilized three different models for STI service delivery based on the special geographical and social characteristics of the areas with FSW populations. It is not particularly useful to compare the static clinic, preferred private provider, and partnership clinic models, as each was designed to meet the requirements of particular social and geographic locations. In settings like Andhra Pradesh where sex work is prevalent in both rural and urban areas, it is likely that all three types of clinic facilities are necessary and useful. Irrespective of the level of STIs among sex workers, the partnership model can serve as an alternative mechanism to provide STI and other general health services to sex workers in areas where sex workers are scattered and establishing static clinic may not be cost effective. Moreover, STI prevalence ranging from 3-10 per cent has been reported among FSWs in the area where the partnership clinics were established. This study also observed that FSWs who belonged to the areas with partnership clinic models were less likely to experience STI and more likely to receive STI treatment and undergo HIV test than those residing in area where clinics were not supported by the State level NGO. This indicates that the support of State level NGO in building the capacity of partnership clinics and providers have resulted in better service provision and ultimately resulting in better utilization of services as compared to areas where such support systems were not in place. Although the current study findings offered important recommendations on the usefulness of partnership with government health clinics over other models, the study findings should be interpreted in the light of certain limitations. No biological data were collected on STI and HIV and only self-reported STI symptoms were used as the study outcomes. Future studies should make an attempt to include the STI and HIV incidence/prevalence as outcomes to assess the behaviour change. Further, it can be argued that improved service utilization over time may not necessarily reflect behavioural modification/change. However, studies conducted in this geographical area have demonstrated a significant improvement in the safe sex behaviour of FSWs and hence to some extent the decline in STI symptoms as well as improvement in treatment seeking behaviour could be linked to the change in behaviours. In summary, the HIV prevention initiative in Andhra Pradesh, which used government health facilities to provide STI treatment to FSWs, offers a sustainable approach to provide timely and accessible services. Such partnerships may not only promote HIV prevention services, but also promote the utilization of other health services from the government health facilities by marginalized populations. The side effects of such a partnership model have not been carefully assessed through well-designed operations research; however, upscaling partnership clinics to provide sex workers STI treatment and other services could help reduce institutional and individual level stigma and provide a one-stop shop for comprehensive and accessible health services.

: The study was carried out in the 10 southern districts of Odisha State during September 2010 - February 2012. Most of the districts are hilly and forested. Dry summer (March-June), wet rainy (July-September) and dry winter (October-February) are the three prevailing seasons. The districts have been hyper-endemic for malaria for many decades. is the predominant species, contributing to >90 per cent of the total malaria cases. Malaria incidence peaks during two seasons, one during July to September and the other during November to December. has been incriminated as the major malaria vector. Streams and terraced paddy fields are the major breeding habitats of . , which is a secondary vector, prevalent during summer and rainy seasons, prefers to breed in riverbed pools, terraced paddy fields and ponds. There are altogether 115 community health centres (CHCs) in the 10 districts. Among these, 20 CHCs, two from each district, were randomly selected. In each CHC, three villages were selected randomly for collection of the vector mosquitoes to determine their susceptibility status. The study protocol was approved by the Ethical Committee of Vector Control Research Centre (VCRC), Puducherry, India. : The required number of female mosquitoes of and were collected from cattle sheds and human dwellings in the morning hours using mouth aspirator and flash light in the selected villages. Susceptibility tests were performed on the wild caught blood-fed females using WHO kits. The field collected mosquitoes were provided with 10 per cent glucose solution soaked in cotton pads and brought to the camp laboratory in 1ft mosquito cages wrapped with a wet towel. The temperature and relative humidity in the camp laboratory was maintained at 25 ± 2 °C and 70-85 per cent, respectively. Insecticide impregnated papers of DDT 4 per cent, malathion 5 per cent and deltamethrin 0.05 per cent were obtained from the University Sains Malaysia, Penang, Malaysia. Female mosquitoes were exposed for one hour in 3 to 4 replicates, each replicate with 15 to 20 mosquitoes, to the diagnostic dosage of the insecticides. Parallel controls for comparison were maintained. Number knocked down was recorded after one hour exposure and after the exposure the mosquitoes were maintained for 24 h with glucose food at the same temperature and relative humidity. Mortality was scored after 24 h holding period and if the control mortality remained within 5-20 per cent, the test mortality was corrected using Abbott's formula and expressed as corrected per cent mortality. In case, the control mortality is >20 per cent, the tests were discarded. According to the WHO criteria, a corrected mortality of >98 per cent is ‘susceptible’, <80 per cent is ‘resistant’ and 80-98 per cent is ‘verification required’. The corrected mortality of on exposure to DDT 4 per cent, malathion 5 per cent and deltamethrin 0.05 per cent are given in . Adequate number (minimum of 100) of could not be exposed to each of the three insecticides, because of its relatively lower density in the study area. The results indicated that was susceptible to DDT, malathion and deltamethrin in all the 10 southern districts. Tables summarizes the susceptibility status of to DDT, malathion and deltamethrin. The corrected mortality of this species ranged between 9.5 and 16.7 per cent against DDT 4, 63.5 and 86.7 per cent against malathion 5 and 81.7 per cent and 100 per cent against deltamethrin 0.05 per cent. The results thus showed that was resistant to DDT and malathion in all the 10 districts except in Kalahandi and Gajapati, where the response of this species against malathion was under ‘verification required’ category and further monitoring at periodical intervals could confirm the susceptibility/ resistance status of this species against malathion in these districts. To deltamethrin, was susceptible in two districts . Nuapada and Koraput, while, in the remaining eight districts, its status was under ‘verification required’ category. Overall, the results indicated that developed resistance to DDT and malathion, and showed an increased tolerance to deltamethrin, as the corrected mortality of this species fell under ‘verification required’ category in most of the districts. Chemical control of vectors continues to be the mainstay of the malaria control programme in India. Monitoring vector susceptibility to the insecticides at regular interval has become imperative to ensure judicious and effective use of insecticides in the control programme. The present findings highlight the current status of insecticide susceptibility / resistance status of malaria vectors in the southern districts of Odisha State, which are predominantly inhabited by tribes and hyperendemic for malaria. The primary role of and the secondary importance of in transmission of malaria in the study area were established during nineties. The earlier studies have reported that was susceptible to DDT, malathion and deltamethrin in Koraput and Sundergarh district of Odisha State. Sharma also reported its susceptibility to the three insecticides in five other districts of Odisha State including Kalahandi and Phulabani (presently called as Kandhamal) which are among the 10 southern districts covered by the current study. However, resistance to DDT in this vector species has been reported from Puri and Balasore districts of Odisha. Kumari reported development of resistance by this species to DDT in 11 districts from eight States in India. Recently, DDT resistance in has been reported from Jharkhand. In the current study, was found to be still susceptible to DDT, malathion and deltamethrin in all the 10 southern districts of Odisha. Further, is mainly distributed in hill-top, foot-hill and forested villages in India and its preferential breeding habitat is streams, which are less likely to be exposed to agriculture pesticides. was resistant to DDT and malathion in all the 10 southern districts except in two, where its mortality against malathion fell under ‘verification required’ category. To deltamethrin, this species was found susceptible in two districts while in the other eight districts, its response was under ‘verification required’ category indicating that this species was tending to develop resistance to deltamethrin. The earlier studies in Koraput district showed that was resistant to DDT but susceptible to malathion 5 per cent and deltamethrin 0.025 per cent. Subsequent studies carried out during 2004 in eight districts of Odisha State including the five districts covered in the current study reported that was resistant to DDT in all the eight districts, to malathion in four districts (Mayurbhanj, Bolangir, Nuapada and Kalahandi) and was showing signs of development of multiple resistance to DDT, malathion and deltamethrin in three districts (Bolangir, Nuapada and Kalahandi). Double or triple resistance to DDT, dieldrin and malathion was reported in from 30 districts of Maharashtra. Malathion resistance in this species was first observed in the adjoining State of Gujarat. Subsequent report of resistance to malathion came from Andhra Pradesh. In Dhanora taluka of Gadchiroli district in Maharashtra, was resistant to DDT, but found susceptible to malathion and deltamethrin. In Murumgaon PHC area of Gadchiroli district in Maharastra was found resistant only to DDT, while it was tolerant to malathion and deltamethrin. There are also reports of showing resistance to synthetic pyrethroids in Tamil Nadu and Gujarat, indicating the possibility of widespread resistance to other related compounds of this group. Synthetic pyrethroids are the potent insecticide most commonly used for indoor residual spraying, space spraying and for impregnating bednets under vector control programme. Synthetic pyrethroids are highly effective, if optimally applied, but development of resistance to these insecticides reduces their impact. Although, is only a secondary vector in the study area, the sign of development of resistance by this species to deltamethrin may pose certain amount of threat to the ongoing vector control programme. Therefore, regular monitoring is required for early detection of development of resistance by this species to deltamethrin and for assessing its epidemiological impact. The density of in the study area was low and, therefore, as per the WHO criteria, the minimum number of 100 mosquitoes of this species could not be exposed to the insecticides in each district due to non-availability of adequate number in the field. Two important conclusions could be derived from the current study. The first one is that , the major vector of malaria in the study area was susceptible to DDT and synthetic pyrethroids, the presently used insecticides; while DDT has been used for indoor residual spraying, synthetic pyrethroids are used both for indoor residual spraying and impregnating bednets under public health programme in these districts. Therefore, indoor residual spraying with DDT could be continued but by ensuring adequate coverage and quality through strengthening the advance information system. Also, compliance of bednet use by the community should be enhanced. While advocating use of bednets, it is necessary to insist upon regular use and to focus malaria prevention as the benefit of using nets, because, people who used bed nets as protection against mosquito bites were more likely not to use these when mosquitoes were few than those who used bed nets for malaria protection. The second conclusion is that in view of resistance developed by to DDT and malathion and of its ‘verification required’ status to deltamethrin, monitoring susceptibility of this species to synthetic pyrethroids, which are currently being used in the malaria control programme, is essential for a rationalized use of insecticides for vector control.

The study was conducted in the Department of Pathology, NRI Medical College and General Hospital, Chinakakani, Andhra Pradesh, India, from May 2009 to April 2011. One hundred and twenty nine HIV seropositive patients with lymphadenopathy were included. Fourteen patients with lymphadenopathy of <0.5 cm in diameter and known haematological disorders were excluded from the study. One hundred and fifty three lymph node aspirates from 129 cases (34 cases had lymph nodes at multiple sites) using 23G needle and 5 ml disposable syringe were obtained. Smears were alcohol-fixed. Hematoxylin & eosin (H&E) and Ziehl-Neelsen (Z-N) staining were done in all cases. Acid fast bacilli (AFB) grading was done on Z-N stain positive smears. Grading of AFB was done as proposed by Kumar . Grade 1+ = AFB was found after a careful search; Grade 2+ = AFB were singly scattered; Grade 3+ = AFB were found in large numbers arranged in faggots and singly. Bacilli could be detected even under 10X magnification. Other special stains like Gram's stain, perioidic acid schiff stain (PAS), Grocott methanamine silver stain (GMS) and mucicarmine were used to rule out any other offending pathogen. Culture and sensitivity were performed from the lymph node aspirates whenever necessary. Histological study was done only in one case. Laboratory details (including complete haemogram, CD4 count) were compared with lymph node cytomorphology and revised WHO clinical staging system for adults and adolescents. The study protocol was approved by the institutional ethics committee. The age group of patients ranged from 13-65 yr with a mean age of 32.4 ± 10.9 SD yr. The male: female ratio was 1.3:1; 98 (76%) patients were between 21-40 yr. The most common symptom was fever in 109 patients (84.49%). Pallor was the most common sign in 79 patients (61.24%). The increased occurrence of signs (pallor, oral thrush) and symptoms (fever, fatigue, weight loss) possibly was due to severity of the illness as majority of the patients were in WHO clinical stage IV. The most common site of lymphadenopathy was cervical (left posterior cervical group) in 124 patients (81.04%), followed by axillary lymphadenopathy in 19 (12.4%). The size of the lymph nodes ranged between 0.5 to 6 cm. Majority (80.62%) were of <2 cm size. Of the lymph nodes of size <2 cm, 41.34 per cent were diagnosed as reactive lymphadenitis. Lymph nodes with larger sizes (>2-6 cm) included five cases of lymphomas with a mean size of 4.5 cm, one case of metastatic deposits of 5 cm size and the remaining were cases of suppurative lymphadenitis and cold abscesses with a mean size of 3.2 and 3.6 cm, respectively, indicating that increased lymph nodel size was associated with clinically significant and malignant lesions. The distribution of cytological diagnosis of various HIV lymphadenopathies included tuberculous lymphadenitis in 54 cases (41.8%), which was the most common cytological diagnosis. The three cytological patterns of tuberculous lymphadenitis diagnosed were caseous necrosis- epithelioid cell granulomas (CN-ECG) in 28 (51.85%) of patients, caseous necrosis (CN) in 20 (37.04%) patients and epithelioid cell granulomas (ECG) in 6 (11.11%) patients (). Cytological features were associated with AFB grading on Z-N stain positive smears. Acid fast bacilli (AFB) were abundant in smears with caseous necrosis and decreased in cases with granulomas. This difference in staining pattern was assumed to be due to compromised immunity in HIV infected individuals. The present study showed grade 2+ as the predominant pattern (62.96%). Z-N grade 1+ was seen in 6 (11.11%) and grade 3+ in 14 (25.92%) patients. CD4 counts in tuberculous lymphadenitis patients were found to be decreased with increased bacillary load. Mean CD4 counts in patients with CN were 99.6-103.7, 105.2 in CN-ECG and 139.5 cells/μl in ECG (). Reactive lymphadenitis was seen in 46 patients (35.6%). This was the second common cytological diagnosis with a mean CD4 count of 283.3 cells/μl. Non-specific chronic granulomatous lymphadenitis was seen in only one patient. The routine culture of this lymph node aspirate was sterile and no fungus was demonstrated on special stains. Sixteen cases (12.4%) of suppurative lymphadenitis were diagnosed. Gram's stain, PAS and GMS staining was done on these smears. Smears were negative for PAS and GMS stains but nine patients showed either Gram negative/positive bacilli in the lymph node aspirates indicating secondary bacterial infections. Mean CD4 count in this group of patients was 181.65 cells/μl. One patient was diagnosed with cryptococcal lymphadenitis. was confirmed by PAS, GMS and mucicarmine stains. When compared to other lesions, CD4 counts in cryptococcal lymphadenitis showed lowest value of 48 cells/μl and the patient was in WHO clinical stage IV (). Of the five cases of lymphomas, four were reported as non-Hodgkin's lymphoma (NHL) and one was a Hodgkin's lymphoma. One case of papillary adenocarcinomatous deposits from lung was observed. All these malignant lesions had a CD4 count <100 cells/μl (). Of the 129 patients, lymph node aspirates from five (3.9%) patients were inadequate to report. Repeated aspirations in these cases yielded only blood and blood cellular elements. Among the haematological alterations, anaemia was the most common presentation in 79 patients (61.24%). Most common type among females was microcytic hypochromic anaemia (n=28, 58.22%) with WHO clinical stage IV disease and a cytological diagnosis of cold abscess. Normocytic normochromic anaemia was reported in 45patients (34.9%). Most of them had a cytological diagnosis of reactive lymphadenopathy due to a better immune status, anti retroviral therapy and high CD4 counts. Macrocytic anaemia was seen in a case of NHL with increased cell turn over utilizing B12 and folic acid stores. Biochemical analysis in this patient showed decreased B12 and folic acid levels. Of the 79 patients presenting with anaemia, 66 patients (83.54%) with Hb% <11 g% had CD4 count <200cells/μl and the remaining 13 patients (16.46%) had CD4 counts >200 cells/μl reflecting variable immune status and varied clinical staging. The total leukocyte counts ranged from 1900 to16300 cells/μl. Relative neutrophilia was noted in 32 (24.8%) patients, neutrophilic leucocytosis in 28 (21.7%) with toxic granulations in 19 patients (14.7%). Neutrophilic alterations were seen in reactive and suppurative lymphadenitis. Lymphocytic preponderance was seen mostly in tuberculous lymphadenitis constituting 23 patients (17.82%), lymphocytosis in four (3.10%) and five patients (3.87%) showed reactive lymphocytes. Peripheral eosinophilia was seen in three cases with skin and oral infections but no fungal elements were isolated from the lymph node aspirate (). Leucopenia (in 4.65% patients) and thrombocytopenia (in 3.10% patients) were seen in cases with NHL undergoing chemotherapy and in a case with secondary deposits. Reactive thrombocytosis was seen in 3(2.4%) patients and 9 (7%) patients had a normal blood picture. The WHO clinical staging was done in all cases. It included eight (6.2%) patients in WHO clinical stage I, 37 (28.68%) in stage II, 24 (18.6%) in stage III and 60 (46.51%) in stage IV disease. The WHO clinical staging of HIV seropositive patients was associated with cytological diagnosis and haematological alterations. CD4 counts showed a descending pattern with progression of WHO clinical staging (). Lymphoid tissues are one of the prime targets in HIV/AIDS. Regardless of the portal of entry of HIV, these are the major anatomic sites for establishment and propagation of HIV infection. The commonest opportunistic infection among HIV seropositive cases is tuberculosis. Extrapulmonary involvement occurs frequently and earlier than the other opportunistic infections, especially in individuals dually infected with HIV and tuberculous bacilli. Occurrence of extrapulmonary tuberculosis has increased by 20 per cent as compared to 3 per cent increase in cases of pulmonary tuberculosis and is believed to be due to more severe immunodeficiency in the HIV-infected patient. In the present study, maximum number of cases . 98 patients (76%) were seen in the age group of 21-40 yr. Similar findings have been reported earlier. In the present study, males (57.3%) were affected more commonly then females (42.6%) with a male: female ratio of 1.34:1. Similar findings were reported by Deshmukh (1.7:1) and Guru (2.3:1). In contrast, female predominance was noted by Narang (4:5). The incidence of lymphadenopathy decreased with advancing age after 50 years as in a study by Agravat . The most common site of lymphadenopathy was cervical in 124 patients (81.04%). Similar findings were recorded by others. The most common cytological diagnosis was tuberculous lymphadenitis(41.8%); similar to observations made in studies by others. The diagnosis of tuberculous lymphadenitis was rendered only when smears were positive for AFB. A similar criterion is considered by others. CN was the predominant cytological pattern in the present study among tuberculous lymphadenitis. Gupta had made similar observations. Other groups have recorded CN -ECG as predominant pattern of tuberculous lymphadenitis. Association of grading of AFB staining pattern with cytomorphological features of tuberculous lymphadenitis demonstrated the presence of AFB in greater number in smears showing CN pattern and their number decreased with appearance of granulomas. Similar observations have been made by others. Mean CD4 T lymphocyte count decreased progressively with increase in AFB load and appearance of necrosis. Three cases were found to be multi drug resistant tuberculosis by culture and sensitivity. In the present study, highest CD4 counts were observed in patients with reactive lymphadenitis. Similar observations have been reported by others. Diagnosis of non-specific chronic granulomatous lymphadenitis was considered when granulomas were observed in the absence of AFB in the smears studied. Jayaram and Chew and Kumar Guru considered a similar criterion. This lesion needs further evaluation such as biopsy and serological investigations to establish the aetiological agent. In the present study, cryptococci were seen in sheets in the lymph node aspirate of a seriously immuno compromised HIV seropositive patient. Saikia observed two cases and Jayaram and Chew observed one case of cryptococcal lymphadenitis in their studies. One patient with metastatic disease was also observed . papillary lung carcinoma metastatic to a lymph node. Squamous cell carcinoma deposits and poorly differentiated carcinoma deposits have been seen and reported in earlier studies. Haematological complications are a common cause of mortality in HIV infected patients. Cytopenias are most frequent during the advanced stages of the disease. Of all the cytopenias, anaemia was most common in the present study as reported earlier also. HIV infected individuals with anaemia are at increased risk for progression to AIDS and its associated high mortality. The multifactorial origin of anaemia in HIV disease complicates its differential diagnosis and treatment. In the present study, increased prevalence of microcytic hypochromic anaemia in females was due to periodic blood loss during regular menstrual cycles, parity, along with irregular iron metabolism and reutilization due to chronic HIV infection. The present study demonstrated the utility of lymph node cytology in the diagnosis and segregation of HIV lymphadenopathies with various haematological alterations, CD4 counts and WHO clinical staging. FNAC of HIV lymphadenopathy is a valuable tool for identification of opportunistic infections, neoplastic and non-neoplastic lesions and also for segregating cases that need further evaluation. Non specific chronic granulomatous lymphadenitis needs biopsy for aetiological work up. Neoplastic lesions like NHLs need biopsy and immunohistochemistry for further typing. Correlation of cytological lesions with CD4 counts, haematological parameters and WHO clinical staging provides information about immune status and stage of the disease.

This study was conducted in laboratory of School of Pharmaceutical Sciences, and Institute for Research in Molecular Medicine both at Universiti Sains Malaysia (USM), Penang, Malaysia. The study period was from September 2010 to December 2011. The study protocol was approved by the USM Human Research Ethics Committee. : A 100 ml culture of serovar Icterohaemorrhagiae strain RGA (L44) was prepared using commercial liquid Ellinghausen-McCullough-Johnson-Harris (EMJH) medium (Difco, USA). The medium was prepared according to the manufacturer's instruction and placed in an incubator shaker at 30°C for seven days. Cells were harvested and washed twice with phosphate-buffered saline, H 7.2 (PBS) by centrifugation at 10000 × g for 10 min at 4°C. The washed pellet was suspended in 40mM tris (pH 7.8) at a ratio of 1:2 and vortexed vigorously for 5 min. The cell suspension was then subjected to three cycles of freeze-thawing using liquid nitrogen and 37°C water bath for 5 and 10 min, respectively. The cell lysate was centrifuged and the pellet was mixed with 2ml of commercial sample buffer (Solution A, Agilent 3100 OFFGEL Fractionator, USA) containing urea, thiourea, dithiothreitol (DTT) and glycerol. The mixture was vortexed, centrifuged and the resulting supernatant was used as the protein antigen for subsequent analysis. Protein content of the leptospiral sequential protein extract (SEQ) antigen preparation was determined using a reducing agent and detergent compatible RC DC™ assay (BioRad, USA), aliquot and stored at -80°C. : All serum samples used in this study were those stored at USM serum bank and coded to maintain anonymity. A total of 42 serum samples collected from cases of acute leptospirosis within 10 days from onset of symptoms, were used. In addition to strong clinical diagnosis, these serum samples had been tested positive by three serological tests namely, microscopic agglutination test (MAT), latex agglutination (Tek Dri-Dot™,BioMerieux, France) and IgM lateral flow test (Omega Diagnostics, United Kingdom). These were performed to ensure that the samples had demonstrable IgM antibodies to . Based on MAT, the serum samples used in this study reacted with serovars of the following serogroups: Icterohaemorrhagiae, Autumnalis, Pyrogenes, Bataviae, Grippotyphosa, Canicola, Australis, Pomona, Javanica, Sejroe, Djasiman, Tarassovi, and Hebdomadis. Control samples comprised 42 serum samples, 24 were from patients with other febrile diseases [dengue (n=11), malaria (n=5), typhoid (n=3), toxoplasmosis (n=4) and amoebic liver abscess (n=1)], and the remaining 18 samples were taken from healthy blood donors. The control serum samples were also tested with the same three tests to ensure that these had no IgM antibodies to . Prior to use, all serum samples were pre-absorbed with (RF) absorbent (Virion-Serion, Germany) to remove rheumatoid factors, thus excluding non-specific reactions. : SDS-PAGE was performed using Mini-PROTEAN 3 apparatus (Bio-Rad, Hercules, CA, USA) and Laemmli's standard protocol. Briefly, 16 μl of the liquid protein preparation was mixed in 4 μl of 5× final sample buffer, placed in 37°C incubator for 10 minutes, then centrifuged. The supernatant was loaded onto 10 per cent resolving gels at 100 volts. For gel staining, Coomassie brilliant blue (BioRad, USA), or mass spectrometer-compatible silver stain were used (Thermo Fisher Scientific, USA). : The first dimension electrophoresis was performed using 3100 OFFGEL fractionator (Agilent Technologies, USA) and run according to the manufacturer's instructions. Two mililiters of the leptospiral SEQ antigen preparation containing 2 mg protein was loaded onto 12 well-frame using immobilized H gradient (IPG) strips of H 4-7. Fractionation process was started at 500V, 50 μA and 200 mW followed by separation at maximum 8000 V, 100 μA and 300 mW until 50 kVh was reached. Each fraction was then concentrated to half its volume using Speed Vac (Thermo Fisher Scientific, USA). The second dimension electrophoresis by SDS-PAGE was performed as described above. : The electrophoresed SDS-PAGE gel was transferred onto a nitrocellulose paper (NCP) with 0.45μm pore size (GE Osmonic, USA) using a semi-dry transblot (BioRad, USA). The NCP strips were blocked by Super Block® (Pierce, USA) and incubated overnight with diluted serum (1:50) at 4°C. After a washing step, the strips were incubated with monoclonal anti-human IgM conjugated with horseradish peroxidase (HRP) (Zymed, USA) at 1: 8000 or monoclonal anti-human IgG-HRP dilution (Zymed, USA) at 1:2000. The reactivity was detected using chemiluminescence substrate (Roche Diagnosrtic, Germany) and developed on X-ray film (Kodak, USA). The molecular weights (MW) of the antigenic bands were determined using molecular weight markers (BioRad, USA) and image analyzer (SynGene, USA). Sensitivity was determined based on the number of leptospirosis serum samples (n=42) that were reactive with an antigenic band. Specificity was based on the number of serum sample from other infections and healthy controls (n=42) which were reactive with the antigenic band. : The identified silver-stained protein band was manually excised and subjected to destaining by transferring it into a 1.5 ml microfuge tube containing 100 mM sodium thiosulphate and 30 mM potassium ferricyanide (1:1 ratio) for 15-20 min. After a brief centrifugation, the supernatant was removed, 100 μl of 200 mM ammonium bicarbonate was added and left for 15-20 min at room temperature and the supernatant was removed. The washing and de-staining steps were repeated twice until the silver stain was completely removed. The sample was sent to the Proteomic Center, National University of Singapore for mass-spectrometry analysis using ABI 4800 Proteomic Analyzer MALDI-TOF/TOF mass-spectrometer (Applied Biosystems, Foster City, USA). Identification and characterization of the 72kDa protein was performed by comparing their peptide mass against the built-in MASCOT search engine with Protein Pilot proteomic software (4800 Proteomics Analyzer, Applied Biosystems, USA), a computer generated database of tryptic peptides of known proteins. : The nucleotide coding sequence of highest scoring protein identified from the antigenic band was sent to a scientific company (GenScript, USA) for custom production of recombinant protein by gene synthesis, cloning and expression using their proprietary expression system. SDS-PAGE, protein transfer and IgM immunoblot were performed as described earlier using serum samples as the primary antibody and anti-human IgM-HRP as the secondary antibody. Determination of sensitivity and specificity of the recombinant protein was performed using 20 serum samples from leptospirosis patients and 32 randomly selected control serum. The latter comprised five from patients with dengue, three with malaria, two with typhoid and the rest were from healthy individuals. : The protein content of the leptospiral SEQ antigen preparation was determined to be 2.8-3.4 mg per 100 ml bulk culture. SDS-PAGE profiles were observed in gels stained with Coomassie blue and mass spectrometry-compatible silver stain. Most bands were not clearly seen with the former, while the latter showed a profile of complex pattern of bands with molecular weights ranging from <15 to >150 kDa. The immunoblot assay using the panel of positive and control leptospirosis serum samples showed strong consistent seroreactivity to a 72 kDa band in the IgM immunoblot assay (), however, no reactivity was observed in IgG immunoblot (data not shown). Although other reactive bands could be seen in the IgM blots, they were not consistently present when probed with the panel of sera from leptospirosis pateints. The overall sensitivity and specificity of the 72 kDa band to detect specific anti-leptospiral IgM antibodies were found to be 83.3 per cent (35/42) and 95.2 per cent (40/42), respectively. : The 72kDa protein bands mainly appeared in fractions 6-8, which corresponded to the H 5.25 - 6.00. The fractions were then subjected to Western blot assay using confirmed leptospirosis and control serum samples. Intense reactivity at 72 kDa was observed in IgM immunoblot of fraction 6. : Analysis of the 72 kDa protein by mass-spectrometer revealed a significant homology match (<0.05) of a top scoring protein to be heat shock protein DnaK of (score of 124; cut-off score =85), gi 2735761. : Evaluation using serum samples from 20 leptospirosis patients and 32 controls (remaining serum from the panel previously used for the native protein) showed that r72SEQ was sensitive (85%, 17/20) and specific (81%, 26/32). shows representative IgM immunoblot results using patients’ serum samples containing heterologous anti-leptospiral antibodies, and control serum. Despite the availability of a number of serological methods of diagnosing leptospirosis, the definitive serological investigation remains the microscopic agglutination test (MAT). It is highly specific but is not adequately sensitive for clinical management of early infection. The diagnostic value of MAT is strongly linked to the time of the sample collection and the serogroups in the MAT panel used for each locality. DNA-DNA hybridization and PCR have also been used for detection but are laborious, expensive and time consuming, thus unlikely to be implemented for routine work particularly in developing countries. There are many surface exposed proteins such as OmpL1, LipL32, LipL21, LipL36, and LipL41 that have been reported to stimulate specific antibody response in animal models. Most of these protein antigens have shown a high sensitivity and specificity for detection of leptospirosis with convalescent-phase serum, however, these performed poorly with acute-phase serum. Thus, there is a need to identify new protein antigen(s) with improved sensitivity for acute leptospirosis. In this study, protein preparation from one of the most prevalent pathogenic serovars in Malaysia was tested on a set of patients’ serum samples from various parts of Malaysia who were infected with a variety of serogroups. In a previous report, heat shock protein DnaK has been reported as one of seven proteins that were reactive in immunoblots using 150 serum samples from Brazil and Barbados. However, there are several differences noted between their study and the present findings. One difference was in the method of antigen preparation whereby in the previous study organisms were directly added to sample buffer before electrophoresis. Further, the previous study used IgG instead of IgM immunoblot and the molecular weight of the band was reported to be 76 kDa. In addition, identification of the band as heat shock protein DnaK in that study was based on its reactivity with antisera to DnaK, and not by mass-spectrometry. By fractionation of the proteins, the authors reported that 76 kDa protein was found in the cytoplasm and cytoplasmic membrane of the bacteria. In this study heat shock protein DnaK was detected in the hydrophobic fraction of the protein preparation, thus it was probably derived from the membrane protein of the cytoplasm. The patients serum samples used in the present study were from acute leptospirosis cases (≤ 10 days of symptoms), while 42 control samples were from patients with other related infections and from healthy individuals. Therefore, specific IgM antibody recognition of heat shock protein DnaK by the patients’ serum is a strong indication that this protein is specifically expressed during early stage of the infection. This is consistent with results reported by Guerreiro . Pol and Bharadwaj utilized a heat extracted antigen from non-pathogenic leptospires and purified it by high-performance liquid chromatography (HPLC). This 50 kDa protein fraction was reported to have a specificity and sensitivity of 93.3 and 85.0 per cent, respectively. In another study, MPL17 and MPL21 surface protein antigens were evaluated with serum samples from patients in the early and convalescent phases of leptospirosis. The prevalence of total IgG antibodies against MPL17 and MPL21 were 38.5 and 21.2 per cent, respectively. However, IgM-ELISA was positive with only MPL21 and the antibody level was not significantly different from that of MAT. Recombinant protein-based serological test can attain consistent high sensitivity and specificity because of high concentration of effective antigen, and the absence of non-specific moieties which are present in whole-cell preparations. In the present study, recombinant protein to heat shock protein DnaK (r72SEQ) showed 85 per cent sensitivity and 81 per cent specificity. The sensitivity of the native protein (83%) is comparable to the recombinant protein. However, the former seemed to have higher specificity than the latter, and the reason is not clear. It is possible that post-translational modification, which has been reported in , may be a reason for the higher specificity of the native protein. More accurate sensitivity and specificity rates can be obtained by testing a large number of samples. There are some reported proteins antigens in the outer membrane of with good diagnostic value for human infection. The notable ones are LipL32, LipL41 and OmpL1 and the first two are surface exposed proteins expressed only in pathogenic spp. Flannery evaluated five recombinant antigens using IgG-ELISA for serodiagnosis of human leptospirosis namely LipL32, LipL41, HSP58, LipL36 and OmpL1, and their results showed that rLipL32 was the most useful antigen. The recombinant LipL32 IgG ELISA showed the highest sensitivity when tested with serum samples from acute and convalescent leptospirosis patients; and highest specificity with serum of healthy individuals and those with other diseases with similar symptoms. However, it is interesting to note that IgM antibodies to rLipL32, rHsp58, and rOmpL1 were not detected. This is in contrast to the study by Luo which detected IgM antibodies to rLip32 in more than 90 per cent of the samples. Another study showed improved sensitivity and specificity of recombinant Lig-based immunoblot assay as compared to other recombinant protein-based assays. Further studies on the r72SEQ need to be performed using a larger number of patients and control serum samples, and also including serum samples from convalescent patients. The diagnostic value of the protein would be high if there is IgG but no IgM reactivity when tested with serum from convalescent patients. In conclusion, the findings of this study showed that leptospiral 72kDa protein is a potential marker for acute infection, and the recombinant protein may be feasible to be used in developing a rapid test.

xref #text From the news reports in the past 2 years, we learn that the search for the Holy Grail of Regeneration Medicine, , the creation of whole working organs, is moving forward at full speed. What medical scientists have been able to achieve with decellularized ‘scaffolds’ and ‘spray-on’ stem cells is amazing. Moreover, the recent Nature “Reprogramming ” article from Spain ]21[ may indeed have a strong impact on ‘organ replacement’ soon. Similar to my 2011 Commentary [], I wish to conclude this article with a few words on the issue of ‘government laws and regulations.’ It is true that there are still countries that need to impose better regulation on unproven stem cell treatments carried out by unscrupulous individuals. Fortunately, there are countries in which their government agencies are working hard to establish laws and restrictions to regulate such unproven stem cell therapies. First, the US is an example. In 2008, the US Food and Drug Administration (FDA) began investigating Regenerative Sciences Inc. in Broomfield, CO for treating orthopedic problems using the company’s stem cell product called ‘Regenexx’ []. In July 2012, after 4 years of investigation and legal argument (United States of America Regenerative Sciences), a US Federal Court ruled that the ‘stem cell culture product is a drug,’ since the stem cells extracted from the patient’s own bone marrow were more than minimally manipulated using reagents that were transported across state lines []. As mentioned in my 2011 Commentary [], the stem cell businesses in China were described as ‘Stem Cell China. Wild East or Scientific Feast.’ It sounded good that in January 2012, the Chinese Ministry of Health announced “All medical research and clinical practices of stem cell therapy without approval from the ministry and the State Food and Drug Administration will be put to an end after the overhaul.” However, this 2012 Health Ministry announcement has fallen on deaf ears. It was thus reassuring that in March 2013, the Chinese Ministry of Health (MOH) and the State Food and Drug Administration (SFDA) issued a joint announcement, detailing the draft regulations and requirements for drug and stem cell clinical trials and indicating non compliant centers would face severe penalties []. At this time, l can report that the Chinese MOH and SFDA’s crackdown on unapproved stem cell centers seems to be working. I have learned that a few stem cell centers have indeed completely stopped their operations in China since the early part of 2013. The above regulatory news from the US and China is encouraging. However, a word of caution is worth repeating. In view of the increasing interest in ‘stem cell tourism’ and the huge profits from marketing ‘stem cell therapy,’ I suppose unproven stem cell therapies will continue in countries where regulations are less stringent. I sincerely hope that I do not have to hear more news about patients who have died from such unproven treatments! Finally, I wish to end this article on a high note with one more inspirational story. A recent Los Angeles television news and video in August 2013, reported the happy and emotional first-time meeting between a Stem Cell Donor (a young German woman living in Dresden, Germany) and a Stem Cell Recipient (an American woman living in Orange County, CA) 2 years after the PBSC from the former had saved the latter from leukemia []. I sincerely hope that stories like this will inspire more individuals to register as stem cell donors to

Coronary heart disease (CHD), also called coronary artery disease, is a complex and heterogeneous cardiovascular disease (CVD). It belongs to a group of atherosclerotic CVD that is defined as a chronic disorder which develops insidiously throughout life and usually progresses to an advanced stage by the time symptoms occur []. The critical underlying process of pathogenesis is atherosclerosis (AS) that, in itself, is a multifactorial and peculiar condition. There are a number of known controllable and uncontrollable factors, one of the last-mentioned is genetic, named as strong family history of premature CVD []. There are many genome-wide association studies (GWAS) performed worldwide to determine the main genetic factors that could be used for CVD identification and creation of useful tests for effective diagnosis, prognosis and treatment. Regrettably, the genetic factors and their importance are not yet sufficiently applied in clinical practice []. Moreover each population may have some exceptional genetic characteristic that does not necessarily correspond with results from other studies. The background of our study is from the previous Linkoping-Vilnius CHD risk assessment study [], which demonstrated the differences of atherosclerotic process between Lithuanian and Swedish male individuals. Subsequently, other study aimed to identify potential genetic markers associated with AS and CHD in the Lithuanian population []. The results lacked significant values for strong association of single nucleotide polymorphisms (SNPs) and disease. Novel genotyping techniques and platforms provided an improved opportunity for a more precise analysis of whole genome variation associated with human complex diseases. Thus, in this study we performed the GWAS in 32 families of Lithuanian ethnicity in search of significant genetic markers (SNPs) of CHD that may elucidate the underlying specificity of AS in this population. l s t u d y p r o t o c o l s w e r e a p p r o v e d b y t h e V i l n i u s R e g i o n a l B i o m e d i c a l R e s e a r c h E t h i c s C o m m i t t e e ( N o . 1 5 8 2 0 0 - 1 1 - 2 5 5 - 0 6 7 L P 2 ; 2 0 1 0 - 1 1 - 0 5 ) . I n f o r m e d c o n s e n t w a s o b t a i n e d f r o m a l l i n d i v i d u a l s w h o p a r t i c i p a t e d i n t h e s t u d y . According to the Illumina Inc. protocol guidelines, all of the samples except one, were of good quality and had been properly processed (call rate >98; LogRDev <0.3; coincidental sex list file created). At the beginning of the analysis there were 731,412 SNPs genotyped in the group of 96 individuals. After the data filtering procedure, two individuals were removed from further analysis for low genotyping (MIND >0.05); 25,293 heterozygous genotypes were excluded from analysis because the second allele of the genotype was missing; 298 SNPs were excluded based on the Hardy-Weinberg equilibrium test ( >0.0005); 2528 SNPs failed missingness test (GENO >0.1); 82,552 SNPs failed frequency test (MAF <0.01); 591 SNP were not used because of homogeneity over all individuals. After the final frequency and genotyping pruning, 646,445 SNPs in 31 patient and 63 parents were included for further association analysis. Twelve SNPs were found to be significantly associated with CHD phenotype with values smaller than 0.0001. The SNPs annotation (transmitted allele, chromosomal position, gene, gene function) along with the χ, value, OR and empirical power based on the sample size calculations, are presented in . The SNPs are annotated according to the National Center for Biotechnology Information (NCBI) dbSNP and Gene databases []. The acceptable power values were greater than or equal to 0.65, and thus fell partly into the desired range between 0.8 and 0.95 []. Only the power value of the significant SNP rs1321936 diverged and was excluded from further evaluation. The OR values in show the size of the effect. The greater the deviation of OR is from the value of 1, the more significant the test is. As can be seen from the Manhattan plot (), there are three significant markers (rs12734338, rs3883013, rs3853444) that do not have the significant adjacent SNPs (according to the nucleotide’s position), , the correlation is absent. Thus, these markers could be artefacts. We could also suspect that not all of the adjacent SNPs were genotyped. This is more likely to happen with rs3853444, as there were two adjacent SNPs that were excluded from the analysis. It was previously mentioned that the study group included only male patients and their parents. A male could possess the SNP allele on either his X or Y chromosome and this affects the analysis algorithm. Each transmission from a heterozygous mother to a male offspring should be given twice the weight of a transmission to a female offspring []. Thus, the standard TDT appears to be unsuitable for the analysis of SNPs in sex chromosomes and eventually sex chromosomes were excluded from the analysis. Our aim was to identify the particular genetic factors for Lihuanian CHD patients. In order to avoid population stratification we planned the familial GWAS and subsequently performed TDT analysis. Out of 12 significant SNPs, at least two (italicized and bold rows in ) had promising OR values in addition to power and values. Despite relatively wide OR intervals, caused by a modest sample size, these SNPs may indicate the potential genes that could be involved in CHD pathogenesis. The SNP rs17046570 is located in an intron of the reticulon 4 coding gene () on chromosome 2. Retic-ulons are associated with endoplasmic reticulum, and are involved in neuroendocrine secretion or membrane trafficking and apoptotic processes. In particular, reticulon 4 has been identified as a potential inhibitor of central nervous system regeneration by means of the inhibition of neuron outgrowth [,]. Common variants that are associated with schizophrenia in the Japanese population [] and also blood lipid phenotypes [], are cited. is a candidate gene associated with vascular cell apoptosis and AS modulation []. It is thought to participate in vascular remodeling and is a considerable new factor for atherogenesis process [–]. It was also stated that reticulons may be factors that mediate between the apoptosis and AS processes []. Thus, our results are consistent with these findings. Another SNP on chromosome 5, rs11743737, is located in an intron of the F-box and leucine-rich repeat protein 17 coding gene . The FBXL17 protein has an F-box that is a 40 amino acid motif typical for F-box containing proteins. F-box containing proteins together with culin and SKP1 (S-phase kinase-associated protein 1) make up the SCF complex (SKP1, cullin, F-box containing complex) that is a protein ubiquitine ligase []. The SCF is a key complex in the ubiquitine-proteosome system (UPS) that is involved in 70.0–90.0% of protein degradation processes including the degradation of a number of proteins important for the cardiovascular system. The UPS is also important in the regulation of endothelial cell cycle. The effect of oxidative stress on the SCF complex may disrupt the function of UPS and in turn the function of the endothelium that is regulated the by UPS []. According to the NCBI Gene database review of association results [from National Human Genome Research Institute (NHGRI) Catalogue and association results submitted to the database of Genotypes and Phenotypes (dbGaP)] there are many SNPs in the gene region associated with the various phenotypes including cholesterol, high-density lipoproteins, body mass index. These findings do not compromise our findings either. Moreover, these summarized results might show us the complexity and universality of the FBXL17 protein function in the pathogenesis of different diseases. It is possible that other SNPs that are in linkage disequilibrium with the identified CHD associated SNPs were not identified during this TDT analysis but may also be involved in the development of the disease.

xref #text During the period from 2002 to 2012, 3800 prenatal samples [2556 amniotic fluids and 1244 chorionic villus samples (CVS)] were referred to the Department of Diagnostic Laboratories (Cytogenetic Laboratory), Clinical Hospital Acibadem Sistina, Skopje, Republic of Macedonia (ROM). Referral reasons for prenatal diagnosis were advanced maternal age, abnormal ultrasound findings, history of chromosomal abnormalities, positive maternal serum triple test (). Amniotic fluid samples and CVS were cultivated in Amniogrow complete medium (Cytogen, Sinn, Germany). Peripheral blood samples of the parents (prenatal detection of structural chromosomal abnormality) were cultivated in Lymphogrow medium (Cytogen). At least 15 metaphases were analyzed for each case and 10 metaphases were karyotyped using Bandview software from Applied Spectral Imaging (Carlsbad, CA, USA). The results were reported according to the recommendations of the International System for Chromosome Nomenclature 2009 []. During 10 years prenatal diagnosis and 3800 samples provided by amniocentesis and chorion biopsy, we detected seven Robertsonian translocations, eight autosomal reciprocal translocations and one sex chromosome translocation in balanced and unbalanced states (). Referral reasons for prenatal diagnosis for all cases are represented in . An overall frequency of all Robertsonian translocations was 0.18%; all four were in balanced state, three of which originated from maternal Robertonian translocations and only one case of paternal origin. Maternal carriers of this Robertsonian translocation did not have reproductive problems or pregnancies with unbalanced rob translocations. The translocation with paternal origin resulted from oligoastenoteratozoospermia in the paternal carrier of rob and conception was achieved after applying techniques of assisted reproduction. There were two rob translocations detected in an unbalanced state, both of them of maternal origin. In one of the patients with familial rob the microscopic analysis of the curetted placental tissue showed abnormalities that suggested chromosomal abnormalities consistent with trisomy. The chorionic villi had irregular villus contours (shapes) with mucinous or edematous stroma (). The villous blood vessels were diminished and nucleated erythrocytes were absent. The trophoblast on the villous surface showed trophoblastic proliferations in the form of sprouts ().There is one case with 46,XY, +13,der with detected ultrasound abnormalities. The autopsy of the fetus showed multiple anomalies. The fetus had cheilognathopalatoshisis, hexadactily on both toes. The visceral organs did not show any abnormalities. These findings were consistent with a Patau’s syndrome phenotype. The frequency of autosomal reciprocal translocations was 0.21% (eight cases). Five conceptions with reciprocal translocations of autosome chromosomes in a balanced state (paternal origin) were achieved normally. There was one double translocation 46,XX tt with normal ultrasound parameters from maternal origin, mother was the carrier for both translocations, without phenotypic abnormalities (). There was only one 45,XY,t(p11;q11),18p- case associated with ultrasound abnormalities detected after a pregnancy was achieved with the assistance of IVF. One case, 45,X, t(p10;p10) of unknown origin, was represented with ultrasound hydrops fetalis that was associated with Turner Syndrome phenotype. There was one case of apparently balanced sex chromosome translocation [46,X,t(X;10) (p11.23; q22.3)] in a single pregnancy achieved after transfer of three thawed embryos (). The karyotypes of the parents were normal. Chromosomal translocations are represented in our study with an overall frequency of 0.42%. The frequency of translocations in a balanced state were 0.29% and of translocations in an unbalanced state was 0.13%. There were seven Robertsonian translocations detected in our study, six of them being inherited from a parent who was a carrier of a Robertsonian translocation. All four detected rob were in a balanced state. Three of them were of maternal origin, and in all cases there was no evidence of reproductive problems. One case of rob was of paternal origin, and the analyzed pregnancy was achieved by assisted reproduction because of the father’s severe oligoastenozoospermia. The two cases of rob detected in our study were in an unbalanced state and of maternal origin. Our results correlate with the European collaborative study [], where all karyotypes of 280 prenatal samples [parent rob carrier] were in a balanced state. It was noted by several investigators that meiotic segregation products in male carriers of all Robertsonian translocations result mostly from alternate segregation mode (>75.0%) [–]. Analysis of meiotic prophase cells in heterozygous carriers of different Robertsonian translocations showed that the predominance of a preferential -configuration of the meiotic trivalent structure could promote alternate segregation [,]. The risk for translocation trisomy 21 (Down’s syndrome) at amniocentesis in female heterozygote was estimated to be about 15.0% [,]. The risk of having a live-born child with translocation trisomy 21 was around 10.0%. There was an about 1.0% risk for paternal transmission of translocation trisomy 21 []. In our study, there was one case of unbalanced homologous translocation with karyotype 46,XY, rob(q10;q10)+13. The histopathological findings of this terminated pregnancy confirmed a Patau’s syndrome phenotype. According to the consulted references, 90.0% of cases with t are and estimated mutation rate for t is 0.5% per 10 gametes at conception []. Bugge [] used 20 polymerase chain reaction (PCR)-based DNA polymorphisms to determine whether trisomy 13 due to rea(13q;13q) in six cases is caused by translocation (13q;13q) or isochromosome (13q;13q), and the determine the parental origin of the rearrangements and the mechanisms of formation. In five cases, isochromosomes with two identical q arms were revealed, one of maternal origin and four of paternal origin. Only one case had a Robertsonian translocation of maternal origin []. Reciprocal translocations of different autosome chromosomes were presented in our study with six cases in a balanced state, five of them of paternal origin and one double translocation inherited from the mother. Only the cases of 46,XY,t(p21;q26) pat. and double translocation 46,XX,tt mat. were associated with reproductive problems. All other cases were of paternal origin and did not report reproductive abnormalities; they have other children with normal phenotypes. It was noted that if the same (balanced) karyotype found in the carrier parent was detected at prenatal diagnosis, there was no increased risk of phenotypic abnormality in the child []. However, there are mechanisms where apparently balanced translocation may have phenotypic consequences in the progeny of translocation carriers. These are: cryptic unbalanced defect [], post zygotic loss of a derivative chromosome in one cell line [], position effect [], and uniparental disomy. Gametogenesis of reciprocal translocation carriers is affected by different segregation modes at meiosis and unbalanced gametes lead to infertility, recurrent miscarriages and fetal multiple malformations. Genetic counselling for prenatal cytogenetic diagnosis in all future pregnancies of a parent heterozygous for reciprocal translocation is required. Genetic counselling for a double translocation is the same as for a single translocation, although the risk for future pregnancies is increased []. We found two unbalanced karyotypes with reciprocal translocations. One unbalanced state was the result of monosomy X (Turner syndrome) associated with translocation of unknown origin. If there is a parent carrier of the same translocation, this karyotype may be the result of ICE (interchromosomal effect) []. The other one is a 45,XY,t (p11;p11) 18p- (IVF pregnancy) in parents with normal karyotypes. Random error in parental gametogenesis seems to be the reason for this translocation and the recurrence risk is low. Regarding the normal phenotype of the girl with the apparently balanced sex chromosome translocation 46,X,t(X;10)(p11.23;q22.3), we can assume that there was early replication of the translocated X chromosome in all cells. However, reproductive failure and recurrent miscarriages can be expected later in life and genetic counselling as well as prenatal diagnosis is required. Sex chromosome translocations (X-autosome) are distinct from autosome translocations because of transcriptional silencing of an extra X chromosome in the female []. Inactivation pattern is crucial for phenotypes of affected female carriers. Silencing of a normal X chromosome is required for a normal phenotype. Even so, there is a risk of gene disruption or position effect in X-autosome female carriers. In the review of 122 cases of balanced X-autosome translocations in females [], with respect to the X inactivation pattern, the position of the X breakpoint and the resulting phenotype, there were 77.0% of the patients where translocated X chromosome was replicated early in all cells analyzed. The breakpoints in these cases were distributed all along the X chromosome. Most of these patients were either phenotypically normal or had gonadal dysgenesis, some had single gene disorders, and less than 9.0% had multiple congenital anomalies and/or mental retardation. In the remaining 23.0% of the cases, the translocated X chromosome was late replicating in a proportion of the cells. In these cells, only one of the translocation products was reported to replicate late, while the remaining portion of the X chromosome showed the same replication pattern as the homologous part of the active, structurally normal X chromosome. The analysis of DNA methylation in one of these cases confirmed non inactivation of the translocated segment. Consequently, these cells were functionally disomic for a part of the X chromosome. #text

xref italic #text This study was approved by the Ethical Review Board at Neilein University, Khartoum, Sudan and informed consent was obtained from all patients. Cross-sectional design was used in this study. Twentynine meningioma tissue samples () were collected during surgeries from the Alshaab Teaching Hospital located in Khartoum State, Sudan. Thirteen patients were male (44.8%) and 16 female (55.2%), aged 15 to 55 years, thus being much younger than meningioma patients in Western countries. Samples were collected in sterile containers with sterile RBMI-1640 media to be processed for tissue culture within two hours. After long-term culture, chromosomes were prepared and GTG-banding was done using standard procedures []. Interphase fluorescence hybridization (I-FISH) was performed using a two-color FISH approach: DNA derived from BAC-probe RP11-551L12 was gene-specific (22q12.2) and labeled with Texas Red; RP11-172D7-DNA located in 22q11.21 served as an internal control and was labeled in SpectrumGreen. The FISH-procedure was done according to standard protocols []. For microscopic evaluation, 100 interphase nuclei were examined for each specimen. In 10 of the 29 meningioma samples, cell cultivation led to successful karyotyping. The banding cytogenetic results are shown in . Representative I-FISH results are shown in . In all but one sample, deletions of NF2 were observed, in 60.0–100.0% of the interphase nuclei studied. In six of the cases the signal pattern indicated an interstitial or terminal deletion of 22q including the gene region (), in the other 22 cases the signals indicated the complete loss of a chromosome 22 (, ). Statistical analysis of the gene deletion against age and gender did not reveal any correlation (results not shown). In agreement with the literature, the banding cytogenetic approach led to more comprehensive results but was less successful in terms of cell cultivation and growth in almost 70.0% of the cases [,]. Besides chromosome 22, involvement of chromosomes 5, 6, 8, 10, 11 and 14 was observed in the present study. Even though involvement of chromosomes 8 and 10 are rather unusual findings [], chromosome 5, 6, 11 and 14 are known to be involved in menigiomal chromosomal rearrangements []. In the present study, statistically there was no significant variation between gender/age and gene deletion as initiator in tumorigenesis. This might be explained by small samples size, as previous studies demonstrated clear female predominance []. However, it may also be an influence of genetic background of Sudanese patients; a comparable rate of 97.0% of meningiomas with NF2 deletions has not yet been found in other ethnicities. Two points that would need more comparison studies for clarification, as comparable studies found loss of NF2 only in ∼60.0% of the studied patients []. This may have different reasons such as small sample size. However, neither an influence of ethnic background nor of young age of the studied patients (on average ∼37 years) can be neglected. This study shows that even in known clinical entities more studies especially from African countries are necessary.

Rhesus D (RhD) alloimmunization still remains the major cause of severe hemolytic disease in fetuses and newborns (HDFN), which may lead to anemia, hydrops fetalis and intrauterine fetal death. The incidence of HDFN has been reduced by anti-RhD prophylaxis at the 28th and 34th week of gestation. Therefore, the demonstration of fetal D status is very important in the management of HDFN. Accurate prediction of the fetal RhD type provides prenatal prophylaxis in RhD-negative women with an RhD-positive fetus (about 40.0%) from unnecessary administration of anti D. The most common cause of RhD negativity is the absence of the gene []. Consequently, most genotyping strategies are based on detecting the presence or absence of the gene. An RhD status of the fetus can be detected by invasive methods of prenatal diagnostic tests such as amniocentesis and chorionic villus sampling (CVS) that require fetal tissue but may result in miscarriage or risk of increased maternal sensitization because of complications attributed to CVS or amniocentesis. Recent studies have focused on new non invasive prenatal diagnostic techniques such as circulating fetal nucleic acids in maternal plasma to develop reliable non invasive tests for clinical prenatal diagnosis for RhD status of the fetus [–]. In this study, we assessed the feasibility of fetal gender and RHD genotyping in the plasma samples of RhD-negative pregnant women by using primers and probes targeted toward the gene and exons 7 and 10 of the gene. Blood samples (9 mL), collected in EDTA vacutainers, from 30 RhD-negative Turkish women between 9 and 39 weeks of gestation, who were referred to us for invasive testing because of advanced maternal age, increased maternal serum screening test, fetal sonographic abnormality and previous history of chromosomal or single gene disorder. Routine assay for ABO and RhD typing and testing for unexpected antibodies were performed to include RhD negative women in the study. The positive control for the and genes was a heterozygous -positive man, while the negative control for both genes was an RhD-negative non pregnant woman. The gene served as an internal control marker to confirm the presence of male fetal DNA. All analyses were performed blind, that is, the fetal RHD genotyping was performed without knowing the fetus RhD status, which was confirmed by serological methods postpartum. Nine mL of maternal blood was collected in EDTA vacutainers and sent to the laboratory at room temperature. The blood was centrifuged at 2840 rpm for 10 min., the plasma was transferred without disturbing the buffy coat and recentrifuged again at 3600 rpm for 20 min. and the supernatants were collected and stored at −80 °C before DNA extraction. Written informed consent was obtained from all the families. The study was approved by the Faculty Ethics Committee of Ege University Faculty of Medicine, Izmir, Turkey. Non invasive fetal RHD genotyping using cffDNA in maternal plasma was performed in 30 pregnancies. The median gestational age at the time of blood sampling was 25.4 weeks (range 9–39 weeks). Serological tests on the infant’s red blood cells (RBCs) were performed, and fetal gender was confirmed after delivery. One sample (case 12, whose blood group was found to be AB Rh [+]) was excluded because of controversial results from repeated serological analyses. Among 29 plasma samples, one was in her first trimester, 15 were in the second trimester and 13 were in the third trimester. When compared to postpartum serological results, an accuracy rate of 100.0% was achieved in our prenatal prediction of fetal RhD status and gender determination from the maternal plasma. No false-negative or false-positive results were obtained. Out of 29 cases, 21 were RhD [+] and eight were RhD [−]. The results are summarized in . Out of all cases, 12 were male and 17 were female. Nine out of 12 male cases had an SRY-positive PCR amplification in each of the three separate reactions. Of the remaining three male cases, two had SRY-positive amplifications in three separate reactions. Two male fetuses had chromosomal aneuploidies at amniocentesis (47,XXY and 47,XY,+21, respectively). The SRY amplifications were all negative in three separate reactions in 17 female fetuses. The presence of fetal DNA in 17 female fetuses was shown by establishing a PCR reaction specific for the human FOXP1 DNA sequence. Sex detection of the fetus according to the SRY amplification from maternal plasma showed 100.0% sensitivity. The earliest gestational week for sex identification was 9 weeks in this study. The fetal RhD status was correctly confirmed by amniocentesis or CVS in 29 of 29 cases with the diagnostic accuracy of 100.0%. It has become a common practice to offer routine antenatal anti-D prophylaxis, usually at 28–34 weeks of gestation, or within 72 hours after delivery to prevent anti-D immunization in RhD-negative pregnant women. Anti-D prophylaxis is not indicated in about 40.0% of all RhD-negative cases because the fetus is also RhD-negative. We have shown that real-time PCR analysis of cffDNA in maternal plasma is a feasible and reliable technology after analyzing 29 maternal plasma samples which resulted in sensitivity and specificity rates of 100.0%. These high sensitivity and specificity rates were also achieved in the second and third trimesters. Due to the high complexity of the Rh system and the possibility of false results, more than one region of the gene is suggested to be examined for RhD typing [–]. Therefore, two RHD-specific exons (7 and 10) were amplified, which is importantly below the mean size of circulating cffDNA in maternal serum (range 145–201 bp) []. The sizes of the two amplicons are 89 and 73 bp for exons 7 and 10, respectively. In this study, two regions of the gene was examined for RhD typing and elimination of false-negative results. There are several distinctive results concerning the quantification cycle (Cp) values of RHD-positive fetuses. It is accepted that RHD-positive fetus gives Cp values in the range of 35–40, and no Cp values are observed when the fetus is RhD-negative. Rouillac-Le Sciellour [] highlighted the presence of a silent variant gene such as the Ψ (pseudogene) in the maternal genome when Cp values are in the range of 26–30 cycles. They also pointed out different levels of expression of exon 7 and exon 10 of the gene depending on the gestational age. If amplification of exon 7 is [+] but exon 10 is [−], it was suggested that exon 7 PCR was more sensitive than exon 10 PCR in which the result was considered as RHD-positive. This is usually indicative for a sample collected during early pregnancy (less than 10 weeks of gestation) when the level of fetal DNA is low in the mother’s plasma. In a completely opposite condition, if exon 7 is [−] but exon 10 [+] and Cp >35, they emphasized the presence of a Rh variant that could be either RhD-negative or weak or partial D type. The amplification of only exon 7 was suggested to be useful for the determination of the fetal RHD genotype []. The discrepancies between the results of exons 7 and 10 might be solved by the third RHD-specific PCR. In our study, Ct values of exons 7 and 10 in RhD [−] cases did not show amplification, however, Ct values of both exons were >35 in RhD [+] cases. Insufficient amplification of exon 10 was detected in only two cases by the confirmation of PCR analysis; therefore, amplification solely of exon 7 PCR has been accepted to be indicative of RhD status. Towards the end of the third trimester, there is a rapid increase in the amount of circulating cffDNA in maternal plasma [,]. In this study, exon 7, exon 10 and SRY DNA copy numbers were significantly increased in parallel with the increasing gestational age as pointed out in the literature. The existence of the and genes definitely indicate the presence of fetal genetic material without considering maternal genome in Rh [−] pregnant women. In SRY [+] and RhD [+] cases, the copy number of the gene was significantly correlated with the copy number of exons 7 and 10. In addition, the correlation between the exons 7 and 10 in RhD [+] female fetuses showed that the method used in this study is highly reliable to determine fetal genetic material. The sensitivity of genotyping fetal DNA from maternal plasma varies in the literature. Bischoff [] observed a sensitivity of 70.0% in 20 sensitized RhD [−] pregnant women. Fetal RHD genotyping in another study correctly predicted fetal Rh status in 92 of 98 (94.0%) cases []. By combining amplification of three exons, the concordance rate of fetal RHD genotypes in maternal plasma and newborn with RHD phenotypes at delivery was 100.0% (99.8% including one unusual false-positive) in the Belgian group []. Since then, numerous groups have reported similar results for fetal RHD genotyping in RhD-negative mothers [–] as in the presented study, which indicated a sensitivity and specificity of 100.0%. Our results have shown that non invasive fetal RHD genotyping can be performed rapidly and reliably using cffDNA in maternal plasma with TaqMan real-time PCR assay with a sensitivity and specifity of 100.0%. The weakness of our study is the low number of tested samples, which could be the reason for the 100% sensitivity and specificity. A further prospective study with a larger number of samples will be performed in the future to confirm the reliability of this protocol.

Hypospadias is a congenital hypoplasia of the penis, with displacement of the urethral opening along the ventral surface, often associated with dorsal hooded foreskin and chordee []. The anatomical location of the misplaced urethral meatus determines the severity of this anomaly with the severity increasing from distal to proximal []. Hypospadias has also been reported to be one of the most common congenital anomalies, occurring in approximately 1:250 to 1:300 live births []. Although, the etiology of hypospadias remains unknown, a genetic component in the transmission of this birth defect has been suggested so it seems to be multifactorial []. In addition, 30.0% of severe hypospadias can be attributed to defects in the synthesis of testosterone or adrenal steroid hormones, receptor defects, syndrome-associated hypospadias, chromosomal anomalies, and/or defects in other genetic factors []. Thus, hypospadias may be a highly heterogeneous condition subject to multiple genetic and environmental factors []. With regard to molecular biology and microarray technology, it appears that hypospadias is potentially related to disrupted gene expression []. Previous studies revealed candidate genes including , , , , , , , , , , , , , , , , , , , , , , , and []. Also, in order to study the epigenetic modification of DNA methylation in hypospadias, genome-wide DNA methylation profiling was performed, and the and genes have been reported to be involved in the etiology of hypospadias []. As hypospadias is reported to be an easily diagnosed malformation at the crossroads of genetics and environment, it is important to study the genetic component in order to elucidate its etiology []. Thus, in this study, our aim was to study the gene expression profiles both in human hypospadias tissues compared with that in normal penile tissues. A total of eight patients with isolated distal (subcoronal) hypospadias (mean age 6.8; range 2–10 years) and five healthy circumcised controls (mean age 6.5; range 2–10 years) were enrolled in this study. The penile skin tissue specimens obtained at surgery during hypospadias repair or elective circumcision were divided into two groups: children with hypospadias ( = 8), and normal controls ( = 5). Informed consent was obtained from the parents/guardians of all the children. Samples were distrupted and powdered under liquid nitrogen with pestle and mortar. Tissues were transferred into microcentrifuge tubes and lysated with 1 mL of TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). All samples were homogenized using Qiashredder (Qiagen, Valencia, CA, USA); 0.2 mL chloroform was added to homogenized samples and centrifuged at 12,000g for 15 min. Upper aqueous phase was transferred into a new microcentrifuge tube and isolation of the high-purity total RNA was perfomed using the RNeasy® Mini Kit (Qiagen) following the manufacturer’s specifications. The quantity and purity of RNA was determined using the NanoDrop 1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA) at 260 nm. Genechip® Prime-view™ Human Gene Expression Array (Affymetrix) was used for gene expression studies. Five hundred ng of total RNA was reverse transcribed, amplified and biotin-labelled with Genechip 3’IVT ( transcription) Express Kit (Affymetrix) according to manufacturer’s instructions. aRNAs were then purified with magnetic beads and after fragmentation of purified biotinylated aRNAs, samples were loaded to Genechips for subsequent hybridization. Afterwards, Genechips were washed and stained on the Fluidics station with specified protocol. Signal intensities were acquired by Genechip Scanner 3000 7G (Affymetrix) to generate cell intensity files (CEL). Statistical analysis was performed using Partek Genomics Suite software (Partek Inc., St. Louis, MO, USA). Robust multi-array average (RMA) algorithm was used for data normalization. One-way analysis of variance (ANOVA) was used when >2 groups were compared, followed the by -test. The statistical significance level was set at false discovery rate ((FDR) <0.05 to minimize false identificaion of genes. Greater than 2-fold changes were analyzed for up or down regulated genes. Hierarchical clustering based on genes and samples was performed with Partek Genomics Suite Software (Partek Inc.). A total of 24 genes were found to be upregulated (). The upregulation of the and genes that were previosuly reported have been detected in hypospadias patients ( <0.05). Other genes that were not previously reported, were also found to be upregulated: , , , , , , , , , , , , , , , , , , , , and ( <0.05). The upregulated genes between hypospadias samples and normal controls were selected. Cluster analyses revealed several patterns of genes and included a number of transcription factors, signal pathways, cell cycle, metabolism, nuclear receptor family and structure proteins as well as growth factor receptors (). Hypospadias has multifactorial origins that involve the actions of environmental factors with a genetic background []. The previous microarray studies indicated that, activating transcription factor 3 (), connective tissue growth factor () and cysteine-rich, angiogenic inducer 61 () genes were upregulated in hypospadias and all three genes were also estrogen-responsive [–]. The gene is upregulated in the skin of patients with hypospadias compared to normal prepuce. Also, expression at the mRNA level in fetal mouse tissues demonstrated that its mRNA is expressed significantly more in genital tubercles from fetal mice exposed to estrogens than in those of unexposed fetal mice [,]. This gene has a role in suppression of cell cycling; therefore, it had been hypothesized that its role in hypospadias might be inhibition of cell growth in urethral formation. is upregulated in human and mouse hypospadiac tissues compared with control tissues, at both the mRNA and protein levels []. It has been suggested that may play a role in development of hypospadias as a result of exposure to estrogenic compounds []. Sequence variants of the gene may be involved in the genetic risk for hypospadias []. These genomic variants of have been reported to be present in 10.0% of patients with hypospadias []. In our study, we detected an upregulation of the gene by 13-fold in hypospadias tissues with respect to the controls. The other genes that have been identified from a human microarray analysis study were and . These genes were both members of the cyclin gene family and might have roles in matrix remodeling through the activation of metalloproteinases [,]. Our study only revealed an upregulation of the gene by 5.8–6.0-fold. Among the other 22 upregulated genes, several patterns of genes including apoptosis (), apoptosis and signalling (), metabolism (), protein binding (), receptor activity (), signalling (, , , , , , ), transcription (, , , , , , , ), translation () and transporter activites () were also assessed (). With regard to the top upregulated genes, and , were shown to induce apoptosis (). Such expression of the gene has been associated with apoptotic cell death, whereas the NR4A1 gene has also been reported to induce apoptosis [,]. These two apoptotic genes (, ) have not been reported before. It has been reported that apoptosis may induce external genitalia defects in fetal mouse []. The events leading to hypospadias formation had also been demonstrated to be associated with apoptotic and proliferative events in dorsal urethral epithelia and sinus cord []. However, Baskin [] indicated that hypospadias resulted from an arrest in urethral seam formation or seam remodeling but not by an epithelial apoptosis. Thus, the apoptotic genes need to be studied in a larger population. In this study, we found a relation between hypospadias and the previously reported and genes. We also detected an upregulation of 22 genes in hypospadias patients that have not been reported before. Further studies including GWAS with expression studies in a larger patient group will help us to identify the candidate gene(s) in the etiology of hypospadias.

Ageing happens due to the accumulation of mutations in the genome of somatic cells. It results in tissue atrophy, development of neoplasia and decreased functions of cells and tissues []. A combination of both genetic and environmental factors can affect the process at the cellular level []. As ageing affects resistance against diseases and speeds up the end, new data and investigations suggest that the ageing process can be slowed down at the molecular and cellular level, thereby increasing ones’ life-span []. Cytochrome P450 (CYP) enzymes are involved in phase I of xenobiotic metabolism to oxidize these compounds. Such enzymes can metabolically produce activated substances from chemicals that may act as highly reactive mutagenic metabolites []. There are specific forms of CYP450 enzymes that are major susceptibility biomarkers to activate mutagens and their activities associated with a variety of socio-demographic factors and genetic characteristics []. The cytochrome P450 1A2 () gene polymorphism is one of CYP450 enzymes family involved in metabolic activation of many of chemicals. The reactive oxygen species (ROS) generated by the CYP1A2 enzyme activity can lead to the oxidative DNA damage and mutagenesis in cells. Apparently, the enzyme activity is an important tool to assess the risk of mutagenesis from chemical exposure. The polymorphism of the gene can affect the levels of enzyme activity. The polymorphism type can cause a higher activity of enzyme from exposure to different environmental factors such as smoking and caffeine []. Using different molecular and cytogenetic techniques in various studies of toxicology helps to reflect the risk of exposure to mutagenic agents such as environmental and occupational exposures. In addition, biomarkers are used in the studies to evaluate exposure, effect and susceptibility in individuals. As a susceptibility biomarker, the CYP genetic polymorphism can affect the activation or inactivation of xenobiotics and determine the risk of DNA damage at the exposure to genotoxic agents. Such biomarkers, and the biomarkers of early biological effects, help to identify the risk of genome damage in cells []. These data can serve as an early warning to show the potential risk of health damage from long term chemical exposure. Thus, using biological parameters increased our ability to study the effects of exposure and determine the spectrum of DNA damage []. This study was carried out to identify whether occupational exposure had any effect on DNA damage in the cells carrying the gene polymorphism that enhanced the risk of early ageing. Permission and approval for the study were obtained from the ethical committee of the Medical and Health Sciences Faculty, University Putra Malaysia (UPM), Serdang, Selangor, Malaysia [Reference Number: UPM/ FPSK/PADS/T7-MJKEtikaPer/F01 (JSB-Aug05]. The samples were epithelial cells of buccal mucosa. For this project 120 mechanical workshop workers were selected. The exposed group included males aged 18 years and above. The considered duration time of working in the workshops was at least 1 year or more. Furthermore, 120 people who were not exposed to petrochemical products such as fruit sellers, textile shop keepers, sellers in mobile phone shops, restaurant workers, sundry shops workers, bank staff, photography shop workers, supermarket staff, workers in computer centers, electronic centers and optical examination centers, were selected as a control group. Subjects were interviewed about their health status, educational level, smoking habits, alcohol consumption, work history, duration of working at one occupation and other aspects relevant to the study. In addition, workers were divided into two groups, those with 5 or more years in one group and those with less than 5 years in another group. Respondents were asked to rinse their mouth with water before collection of the buccal samples. The cells were collected by scraping the inner part of the cheeks both sides with a cytology brush. Then, the cells were gently mixed with 0.9% sodium chloride and phosphate buffered saline (PBS) in separate micro-centrifuge tubes and brought to the laboratory. The cells were treated for micronuclei (MN) test, comet assay, real-time polymerase chain reaction (RT-PCR) and restriction fragment length polymorphism (RFLP). The effect of polymorphism on the samples was assessed by MN formation, comet tail length and telomere length shortening as the biological parameters. The methods of the MN test and comet assay were performed according to a pattern described in []. In the current study, genomic DNA was extracted from the cells using QIAamp DNA blood MiniKit (Qiagen, Courtaboeuf, France) and then was quantified by Nanodrop™ 1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA). Meanwhile, the extracted DNA was run on a 0.7–1.0% agarose gel. Genomic DNA was used to run RT-PCR and RFLP. The gene polymorphism was amplified by PCR followed by the RFLP method using I and I restriction enzymes. Using both restriction enzymes helped to fix the determination of the gene polymor-phisms. , and show the images of PCR and RFLP products of the gene. The gene polymorphism was amplified by PCR and the product was 596 bp (). The RFLP products resulting from the I enzyme digestion were the wild (WW) and mutant (WM, MM) genotypes. The size of the WW genotype was 596 bp; the fragment sizes for the heterozygous mutant (WM) genotypes were 596, 464 and 132 bp, and for homozygous mutant (MM) genotypes were 464 and 132 bp (). The RFLP products after digestion with the I enzymes were MM (475 and 121 bp), MW (475, 343, 132 and 121 bp) and WW (343, 132 and 121 bp) genotypes (). The RFLP products of CYP1A2 using the I enzyme digestion were referred as the wild and mutated genotypes. The WW genotype was placed in the wild group and WM and MM were in the mutated group. The respective frequencies of the WW, WM and MM genotypes were 49.2, 46.3 and 4.6% in all individuals. Furthermore, the results showed that the gene polymorphism did not significantly influence ( >0.05) the individual biomarkers (). Differences in biomarkers between workers and controls was evaluated in wild (WW) and mutated (WM and MM) genotypes. It was found that difference in MN frequency between workers and controls was statistically significant in both wild ( = 0.001) and mutated ( = 0.001) genotypes. In addition, the results showed that the mutated genotype significantly affected the relative telomere lengths ( = 0.002) in workers. No statistically significant effect on comet tail length ( >0.05) was found in wild or mutated genotypes (). The results showed that the gene polymorphism had no significant effects on the biomarkers in workers and control groups below 30 years old or above. However, the wild genotype significantly affected comet tail length in workers below 30 years of age ( = 0.047) (). The findings of socio-demographic factors indicated that ethnicity had a significant effect on MN frequency ( = 0.004). Furthermore, duration time of 5 years or more significantly affected MN frequency ( = 0.001), comet tail length ( = 0.001) and relative telomere length ( = 0.001). It was found that smoking, alcohol consumption and educational levels showed no statistically significant effect on each of the biomarkers ( >0.05). DNA damage can occur due to the effects of the gene and environmental factors []. The subjects showed a shorter telomere length on the mutated genotype yet a higher MN frequency in both wild and mutated genotypes. Such effects suggest the possible influence of gene polymorphism on DNA damage mediation in the cells [], which could be due to the increased enzyme activity as well as sensitivity of cells to genotoxic effects []. Despite the paradoxical reports, our result confirmed a correlation between the different genotypes and MN frequency []. Such correlation is probably due to more sensitivity of MN than the other biomarkers to express the effects of genotypes on genetic material damage from occupational exposure. Contrary to previous reports indicating DNA damage in the cells [,,], no statistically significant influence was observed on comet tail length in our research either in wild or mutated genotypes. In the current study, age did not contribute to enhance the influence of the gene presenting in higher MN, shorter telomere length and greater comet tail length in workers and controls, except greater comet tail length in the younger group of workers. Lack of age effects suggest a protection effect of gene against ageing. Apparently, the gene prevents the modulation of DNA damage [–] by transcriptional activation and resistance to changes, which can be interfered with lifestyle factors [,], different genotypes and genes interactions []. Among all lifestyle factors, only ethnicity significantly affected MN frequency regardless of occupational exposure, which indicates the possible influence of gene polymorphism on the cell protection against genome damage []. Meanwhile, this research was a confirmation of the studies [,] indicating the association of DNA damage with working duration time. It seems that good interpretation of results depends on a suitable sample size in each group of various genotypes and age. However, the study was cross-sectional, therefore, finding effective and non effective factors were difficult. Another limitation of the study was the difficulty of determining and isolating the exact effects of gene polymorphisms. Despite the above limitations, this study can serve as a base to address the effects of these genotypes and surrounding risk factors on early ageing. Accordingly, CYP1A2 genotypes contributed to DNA damage from occupational exposure, hence, further investigations are needed to evaluate the exact effects of different genotypes on a subject’s premature aging.

Interleukin-6 (IL-6) is a pleiotropic inflammatory cytokine that has been implicated in the development of Alzheimer’s disease, cardiovascular diseases and many different types of cancers [–]. The role of IL-6 in mediating humoral and cellular immune response relating to inflammation and tissue injury has been well established []. Several studies have demonstrated that elevated IL-6 level is associated with vascular smooth muscle growth and increased production of acute phase protein, thereby contributing to possible development of cardiovascular disease as well as Alzheimer’s disease [,,,]. The IL-6 –174G allele had been demonstrated to be associated with higher IL-6 production []. This polymorphism affects the circulating serum IL-6 level and IL-6 gene transcription. There have been extensive studies on IL-6 gene polymorphisms in different diseases and interestingly, there is also significant variation in the frequencies of this polymorphism among different ethnic groups. It was reported that frequency of the –174C allele is much lower in the Japanese, Africans and Asian Indians compared to European Caucasians [–]. It appears that the majority of Asian populations carry the GG genotypes, ranging from 75.0–100.0%, while Caucasians in the West had a higher frequency of CC genotypes, ranging from 18.0–32.0% [,,,–]. Within the Chinese communities in China, there is also a difference in IL-6 allele frequency. A recent study of the Chinese Han population found that 99.57% carried the GG genotype and none were found to carry the CC genotype []. Malaysia is a multiethnic country where the three major ethnic groups are Malays, Chinese and Indians. The Malays are the major population group (70.0%) and are made up of a mixture of people extant in Southeast Asia as early as 3000 years ago []. The Chinese account for about 20.0% of the total population of Malaysia and majority originating from Southern China; the Indians account for about 10.0% of the total population and were mainly immigrants from Southern India. There have not been any published reports on the IL-6 polymorphism in the Malay or the local Indian populations. A study published by Chua . [] in Malaysia which studied the gene frequency of IL-6 in systemic lupus erythematosus (SLE) patients found that the homozygous G genotype was significantly higher in SLE patients compared with healthy control subjects []. However, there was no racial breakdown of these subjects. In view of the functional implications of the IL-6 gene, we proposed to study the prevalence of the IL-6 –174 (G/C) polymorphism in the Malaysian population. We hope these results can be used as a reference for further studies in determining the disease risk including coronary artery disease among these groups of patients. Blood was obtained from healthy blood donors residing in Kuala Lumpur, capital city of Malaysia. The study was approved by the local institution ethics committee and informed consent was obtained from all individuals. DNA was extracted from the blood samples by a standard phenolchloroform method and QiAmp DNA Blood Mini Kit (Qiagen GmbH, Hilden Germany). DNA concentration was measured using an ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA) to assess the quantity of the product. The –174 (G/C) (rs1800795) polymorphism in the 5′ regulatory region of the IL-6 gene was performed by a custom TaqMan® single nucleotide polymorphism (SNP) genotyping assay (Applied Biosystems, Foster City, CA, USA) system on the LightCycler® 480 Real-Time polymerase chain reaction (PCR) 384 well plate system (Roche Applied Science, Mannheim, Germany). It discriminates the (SNP) by detecting differences in the melting temperatures of the products (m). The forward primer was 5′-CGA CCT AAG CTG CAC TTT TCC-3′ and reverse primer was 5′-GGG CTG ATT GGA AAC CTT ATT AAG ATT G-3′; the probes for the –174C allele was 5′-CCT TTA GCA TGG CAA GAC-3′ and the –174G allele was 5′-CCT TTA GCA TCG CAA GAC-3′. The 5′ nuclease assay was performed using 10 to 30 ng genomic DNA, 2X TaqMan® GTXpress™ Master Mix (Applied Biosystems), and 20X TaqMan® genotyping assay. The PCR cycle consisted of hold for 10 min. at 95 °C, 40 cycles of denaturing for 15 seconds at 92 °C and annealing for 1 min. at 60 °C. Negative, non template controls and known positive controls were included in each experimental run. Allele and genotype frequencies of the three ethnic groups were compared using the χ contingency table. The data were tested for Hardy-Weinberg equilibrium. A value of <0.05 was considered as statistically significant. All statistical tests were performed using the Statistical Package for the Social Sciences (SPSS) version 17 (SPSS Inc., Chicago, IL, USA). A total of 348 blood samples were available for analysis. The median age for the subjects was 31 years (ranges from18–62 years). There were a total of 245 males and 103 females. The sample characteristics are shown in . Among the 348 samples studied, 85.0% carried the GG wild type (–174G/C), 14.0% carried the GC genotype and only 2.0% carried the CC genotype. The GG genotypes were observed in 100.0% Chinese, 93.0% Malays and 66.0% Indians. The CC genotype was absent in both Chinese and Malay individuals and 4.0% of individuals of Indian descent carried the CC genotype (). These differences were statistically significant between the three races ( <0.01). The total allele frequencies for the G and C alleles were 91.0 and 9.0%, respectively. Among the Malays, the allele frequency of the C allele was 4.0% compared to 19.0% in the Indians. shows the distribution of the IL-6 –174 (G/C) genotype and allele frequencies in the three different ethnic populations. The genotype distribution among these populations was in Hardy-Weinberg equilibrium. Cytokine gene polymorphism has been demonstrated in many articles resulting in interindividual variation of transcriptional regulation and subsequently different serum level of cytokines. It has been well established that frequency of the different cytokine gene allele varies among the different populations. The frequency of the IL-6 gene polymorphism had also been extensively studied in different populations. Ethnic differences were well demonstrated among all major populations. It has been reported that –174C allele was completely absent in Japanese and significantly lower in Koreans and Chinese compared with Caucasians [,,,]. Similar findings have also been demonstrated in Africans and the remaining Eastern Asian populations [,,–,,,–] (). It is also interesting to note the absence of the homozygous C genotype among the Chinese, Korean and African populations [,,,,]. In this study, the allele frequency of –174C was only 9.0%, much lower than those reported in the European population [,]; 85.0% of the subjects carried the GG genotypes and only 2.0% carried the CC genotype. However, there were significant differences in the distribution of IL-6 –174 (G/C) between the three ethnic groups. There was a complete absence of the –174C polymorphism in the Chinese subjects, which is similar to that reported in the Singapore Chinese populations []. This is of no surprise as Malaysian Chinese and Singapore Chinese are very closely related due to geographical and historical reasons. In contrast, the C allele frequency was significantly higher (19.0%) in the Indians in this study compared with the other two ethnic groups. The IL-6 –174 (G/C) polymorphism was relatively more evenly distributed in Indians: 66.0% carried the GG genotype, 30.0% carried the GC genotype and 4.0% carried the CC genotype. Study of Indian populations in India had demonstrated the difference in the IL-6 –174 (G/C) polymorphism among the North and South Indians [,]. There was a higher frequency of the C allele in North Indians compared to South Indians, 32.4 15.5%, respectively. The majority of the South Indians studied (72.0%) carried the GG genotype and only 2.9% carried the CC genotype []. Our study reported similar findings with the South Indian rather than the North Indian populations. The most likely reason could be due to the fact that most of the Indians in Malaysia were of South Indian ancestry. The –174C allele was only detected in 4.0% of Malays. There had not been any reports on the frequency of IL-6 –174 (G/C) polymorphism in the Malay population and hence this will be the first report. Interleukin-6 has been extensively studied and has been implicated in many diseases, including coronary artery disease, cancers and Alzheimer’s disease. In a meta-analysis reported by Liu . [], the IL-6 –174 (G/C) polymorphism was associated with cancer risk in African populations but not in Caucasian population. Similar meta-analysis by Xu [] had also demonstrated that there was a possible association of the IL-6 –174 (G/C) polymorphism with cancer risk among Asians and Africans. A recent meta-analysis of 20 studies suggests that the IL-6 –174 (G/C) polymorphism was associated with increased risk of coronary heart disease among Asians []. In the subgroup analysis, it was suggested that individuals with the C allele might have a higher risk of coronary heart disease. Interestingly, studies involving Asian populations had demonstrated that Asian Indians were at greater risk of coronary heart disease compared to other ethnic populations [,]. After adjusting for the other risk factors of coronary artery disease, the Indian population remained one of the risk factors []. Hence, it is of interest that a higher frequency of the C allele was found in Indians in this study cohort. Therefore, this study serves as an important reference study for the Malaysian population and also to provide data for the least studied populations, , Malays and Indians. The result of this study can be used as a reference point for future studies in determining risk factors and association of the mentioned diseases in the Malaysian population.

Deep neck space infections are defined as infections that spread along the fascial planes and spaces of the head and neck []. They can arise from various head and neck regions. The most common etiology is pharyngitis, tonsillitis, odontogenic infections, upper respiratory infections, otitis media or trauma. The deep neck space infections produce significant morbidity and mortality, particularly when associated with the predisposing factors that impair a functional immunological response []. Even in the era of antibiotics, these infections have been potentially life-threatening conditions due to the airway obstruction, jugular vein thrombosis, descending mediastinitis, sepsis, acute respiratory distress syndrome and disseminated intravas-cular coagulation []. Tumor necrosis factor-α (TNF-α) is a multi-functional, inducible cytosine that is produced in response to infection, inflammation and injury. Tumor necrosis factor-α can be produced by lymphoid cells, mast cells, endothelial cells, fibroblasts and neuronal tissue []. It is mainly produced by the macrophages in response to activation of membrane-bound pattern-recognition molecules, which detect common bacterial cell surface products such as polysaccharides, carbohydrates and lipopolysaccharides (LPS). It is the main mediator in response to Gram-negative bacteria and concentration of TNF-α correlates with the amount of bacteria and the phase of inflammation []. Transforming growth factor-β1 (TGF-β1) is a pleio-tropic cytokine with a variety of effects on a wide range of cells in the immune system, playing an important role in cell differentiation, growth, matrix formation, and regulation of immune and inflammatory responses []. It is also a very potent stimulator of monocyte, lymphocyte, neutrophil and fibroblast migration []. The genetic control of inflammatory response in humans has been extensively studied, including the investigation of TNF-α and TGF-β1 responses. Numerous studies have shown that the variations in production and activity of cytokines influence the susceptibility and/or resistance to a range of infectious agents, autoimmune diseases, cancer and other disorders []. Differences in the production of cytokines between individuals are often caused by single nucleotide polymorphisms (SNPs) in the promoter or coding regions of cytokine genes that directly affect the transcription and synthesis of mRNA []. A biallelic polymorphism at the position −308 within the promoter region of the TNF-α gene is one of the most investigated. The presence of the polymorphic TNF-α −308A allele is considered to be associated with the higher TNF gene transcription and TNF-α overproduction []. This substitution leads to 2- to 3-fold higher transcriptional activity of the TNF-α upon stimulation with bacterial LPS []. The TGF-β1 gene also has several polymorphisms, including C-988A, G-800A and C-509T. The cytosine (C) to thymine (T) base exchange at position −509 relative to the first major transcription start site of the TGF-β1 gene was found to be differentially related to transcription factor binding to the TGF-β1 promoter, transcriptional activity of TGF-β1, and TGF-β1 plasma concentration []. Genetic variations within the cytokine genes may be critical in understanding individual predisposition and susceptibility to different clinical conditions. To the best of our knowledge, there are no available studies examining the distribution of TNF-α G-308A and TGF-β1 C-509T polymorphisms in patients suffering from deep neck space infections. Thus, the aim of this study was to analyze the distribution of these polymorphisms and their correlation with the values of inflammatory markers [C-reactive protein (CRP) and white blood cell (WBC) count] in patients suffering from infections of deep neck spaces. Blood samples were collected from 41 patients admitted at the Department of Maxillo-facial Surgery, Dental Clinic, Niš, Serbia. The patients with deep neck infections were classified into three groups: abscess, phlegmon and others (, cellulitis, suppurative parotitis, ), where the particular diagnoses that were included in “others” did not have a sufficient number of patients to perform statistical tests. Forty-four randomly selected healthy individuals, without known acute or chronic disease, were included in the study as a control group. An informed consent was obtained from all participants and the study was approved by the Ethical Committee of the Medical Faculty, University of Niš, Niš, Serbia. At the moment of admission to the hospital, venous blood samples were obtained from the median cubital vein and collected into EDTA vaccutainer tubes. Two-hundred microliters of blood were used for DNA isolation and the rest for biochemical analysis. An automatic hematology analyzer (MEK 6318K; Nihon Kohden, Tokyo, Japan) was used for WBC count determination. The CRP levels were measured using nephelometric immunoassay (Dade Behring, Marburg, Germany). Genomic DNA was isolated from whole blood samples using QIAamp DNA Blood Mini Kit (Qiagen GmbH, Hilden, Germany). The biallelic polymorphisms within TNF-α and TGF-β1 genes were determined using the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) technique. Polymerase chain reaction was performed in a final volume of 25 μL containing 20 ng of DNA, 12.5 μL KAPA2G Fast HotStart ReadyMix (Kapa Biosystems Inc, Boston, MA, USA) and 20 pmol of each primer. Primer sequences used in this study are summarized in the . The PCR conditions were as follows: initial denaturation at 95°C for 2 min., followed by 35 cycles of denaturation at 95°C for 15 seconds, annealing at 60°C for 15 seconds and extension at 72°C for 15 seconds, ending with a final extension at 72°C for 1 min. The PCR products were electrophoresed on a 2.0% agarose gel, stained with ethidium bromide and visualized under UV light. The amplification products were digested using I and F3I restriction enzymes (Fermentas GmbH, St. Leon-Rot, Germany). Restriction enzyme digestion was carried out at 37°C overnight and analyzed by 8.0% polyacrylamide gel electrophoresis (PAGE). The gel was stained with ethidium bromide and visualized under UV light ( and ). The interpretation of the obtained results was performed according to . The allele and genotype frequencies were determined in patients and controls. They were compared with the values predicted by the Hardy-Weinberg equilibrium using the χ test. The two-tailed Fisher’s test was used when the number of expected cases was small. Genetic risk magnitudes (effect size) were estimated by calculating odds ratios (ORs) with 95% confidence intervals (95% CI). C-reactive protein levels and WBC counts were expressed by a median (25th to 75th percentiles). The correlation of TNF-α -308 and TGF-β1 -509 genotypes with CRP and WBC count values were determined using the Mann-Whitney U test. One-way analysis of variance (ANOVA) was used to compare the mean CRP levels and WBC counts between the groups. Probability values less than 0.05 ( <0.05) were considered statistically significant. Statistical analyses were performed using the SPSS version 13.0 statistical software package (SPSS Inc, Chicago, IL, USA). Forty-one patients and 44 unrelated controls were involved in this study. The demographic characteristics of the study groups are summarized in . Genotype frequencies for the SNPs in the study groups were in Hardy-Weinberg equilibrium ( >0.05). As the TNF-α -308 AA genotype was present in only a small number of subjects (only one in the control group and two in the patients), it was analyzed together with individuals heterozygous for the TNF-α -308 polymorphism. The observed TNF-α -308 genotype distribution in the patients’ group did not show significant differences compared to controls (). Moreover, no differences in the distribution of TNF -308G and TNF -308A alleles were observed between the patient and control groups (). The genotype and allele distribution of the TGF-β1 C-509T gene polymorphism did not show significant differences compared to controls ( and ). Also, no association of the particular genotype or allele of the TNF-α G-308A and TGF-β1 C-509T polymorphisms was obtained after the classification of the samples by diagnosis (). Furthermore, in order to evaluate the common association of polymorphic alleles, we investigated the association of the combination of high producing TNF -308A and TGF -509T alleles. The data regrouping was as follows: A/T (high TNF-α/high TGF-β1), A/T (high TNF-α/low TGF-β1), A/T (low TNF-α/high TGF-β1), A/T (low TNF-α/low TGF-β1). However, no statistically significant differences were observed (χ = 1.069, df = 3, = 0.784). C-reactive protein levels and WBC counts, as well as their relation to TNF-α G-308A and TGF-β1 C-509T genotypes were determined in patients with deep neck space infections. The CRP levels were found to be 5- to 60-fold over the base line. The obtained results show no correlation of CRP levels and WBC counts with TNF-α G-308A and TGF-β1 C-509T genotypes (). Furthermore, after the classification of the patients by diagnosis, neither CRP levels ( = 0.699) nor WBC counts ( = 0.787), showed significant difference between the groups. Deep neck infections are less common in the antibiotic era but they often have a rapid onset and can progress to life-threatening complications, especially in the elderly and patients with systemic diseases associated with impaired functional immunologic response. The most common source of inflammation of deep neck spaces in adults are odontogenic infections with the involvement of the submandibular space []. The presence of the functional polymorphisms in cytokine genes affect cytokine expression, and thus may have an important role in the genetic regulation of the inflammatory response and resistance or susceptibility to infections []. The gene for TNF-α is located in chromosome 6 (region p21.3) within the class III region of the major histocompatibility complex. The substitution of guanine (G) with adenine (A) at the -308 site of the TNF-α gene generates two alleles, TNF -308G and TNF -308A. The less common TNF -308A allele is considered to be associated with higher TNF gene transcription and TNF-α overproduction []. A number of studies indicate that the TNF-α G-308A polymorphism is associated with the higher susceptibility for a variety of inflammatory and autoimmune diseases [–]. In oral and maxillofacial pathology, the TNF-α G-308A polymorphism has been studied in patients with burning mouth syndrome, aph-thous stomatitis and periodontal disease. Some investigators observed a higher TNF-α production in the carriers of the TNF -308A allele, while others found no functional significance of this SNP [,]. To the best of our knowledge, there are no reported studies concerning the association of the TNF-α G-308A polymorphism with infections of deep neck spaces. However, our study did not confirm significant differences in the genotype and allele frequency distribution of the patient and control groups. Even though it is well known that TNF-α is a potent chemotactic factor for WBCs, our study did not show any association of the TNF-α -308 polymorphism with WBC count. Moreover, proinflammatory response of TNF-α results in its increased secretion, and releasing of the messenger cytokine, interleukin-6 (IL-6), that stimulates the liver to secrete CRP, which is reliable marker of the acute phase response to infectious burdens and/or inflammation []. In healthy adults, the TNF AA genotype is associated with increased plasma CRP levels in Caucasian and Black men and in Caucasian women, suggesting that this polymorphism contributes to variability in plasma CRP levels []. Our results showed a 5- to 60-fold over the baseline rise of CRP levels in patients with deep neck space infections, but without significant differences in CRP values in the presence of the TNF -308A allele. This can partially be explained by the very low number of homozygous TNF -308A allele carriers, reflecting the low frequency of the AA genotype in this study population. These results are in accordance with those previously reported in the literature that approximately 60.0–70.0% of the Caucasian populations are homozygous for the wild type TNF -308G allele, 30.0–40.0% are heterozygous, and only 1.5–3.0% are homozygous for the variant TNF -308A allele []. The TGF-β1 polymorphism provides chemotactic stimuli for leukocyte migration, but in contrast to its che-motactic effects, it also shows anti-inflammatory effects [,]. The TGF-β1 gene is located on chromosome 19 (q13.1–13.3). A C>T SNP at position -509 relative to the first major transcription start site was found to be differentially related to transcription factor binding to the the TGF-β1 promoter, transcriptional activity of TGF-β1, and TGF-β1 plasma concentration []. This polymorphism was previously studied in asthma, chronic obstructive pulmonary disease, hepatocelluar and gastric cancer [–]. In oral pathology, the TGF-β1 C-509T promoter polymorphism was mostly studied in chronic periodonitis [,]. To the best of our knowledge, this is the first study to examine the association of the TGF-β1 C-509T polymorphism with deep neck infections. Our results suggest that this polymorphism is not associated with deep neck space infections. Additionally, no association of the TGF-β1 C-509T polymorphism with WBC counts and CRP levels was observed in patients with deep neck space infections. This study also showed no association of the TNF-α G-308A and TGF-β1 C-509T polymorphisms with certain diagnoses such as abscess or phlegmon. No difference between CRP levels and WBC counts was obtained after the classification of the samples by diagnosis. Since the cytokines act in a highly complex coordinated network, it would be of great importance to investigate the common influence of the genetic polymorphisms that regulate their production. Particularly, TGF-β1 is known to have a potent immunosuppressive activity, downregulating the transcription of other proinflammatory cytokines, including TNF-α []. In order to evaluate the common association of polymorphic alleles, we have investigated the association of the combination of high producing TNF -308A and TGF -509T alleles. However, no statistically significant differences were observed. Generally, the discrepancies in observed results, besides the genetic heterogeneity of the study populations, might also be explained by population stratification and population bias. In conclusion, this is the first study examining the association of the SNPs of the TNF-α and TGF-β1 genes in patients with deep neck infections. The present study did not confirm the specific role of the TNF-α G-308A and TGF-β1 C-509T polymorphisms in patients with the infections of deep neck spaces. However, further studies are needed to examine genetic markers that can be used for following the disease progression and early identification of individuals at high risk of developing complications.

italic xref #text The patient is a 20-year-old female, the first child of non consanguineous, healthy Caucasian parents (mother was 24 years old and father was 27 years old when the proband was born). There was no family history of ID, congenital anomalies or psychiatric disorders. The pregnancy was uneventful; she was born at term by normal delivery with a birth weight of 2,200 g (below the 3rd percentile), length and head circumference were not recorded. All developmental milestones were delayed: she achieved head control at 6 months, walked without support at 2 years, spoke first clear words at 1.5 years. The girl was referred for genetic evaluation at the age of 5.5 year old, due to ID and single central maxillary incisor. Her growth parameters were: height 95.5 cm [−3.68 standard deviation (SD)], weight 13 kg (−2.48 SD), and head circumference 47 cm (−2.76 SD). She had a triangular face, horizontal palpebral fissures, blue sclera, short, slightly protruding philtrum and upper lip, blunted Cupid’s bow, slightly everted lower lip, mild microretrognathia, bilateral preauricular sinus (). Oral cavity examination showed absent maxillary and mandibular frenulum and single central maxillary incisor. Her language was limited to single words (she could not produce sentences). She had nocturnal and diurnal enuresis, for which she received therapy. No hearing impairment has been identified. Echocardiography showed an atrial septal defect. Abdominal ultrasound, routine biochemical and hematological tests, endocrine investigations [growth hormone (GH), free thyroxine (T4), thyroid-stimulating hormone (TSH)] were normal. No metabolic tests have been performed. Cranio-cerebral computed tomography (CT) scan and magnetic resonance imaging (MRI) of the spine did not show any changes. When examined at the age of 20, her height was 147 cm (−2.89 SD), weight was 43 kg [−1.76 SD, body mass index (BMI) 19.9 kg/m), and head circumference was 53 cm (−1.75 SD). We have noticed her standing and walking position (leaning slightly forward, with widened base of support), as well as slowness in motion and action. The face became elongated, mature for age, with slightly coarse features (). She had mild webbed neck, broad chest and narrow hips, normal posterior hairline, kyphoscoliosis, pectus excavatum, short and wide hands and feet (below the 3rd percentile), with mild brachy dactyly. Puberty was normal, but she developed asymmetric mammary glands and excessive hair growth in pre-sternal, circumareolar and subumibilical regions (Ferri-man-Gallwey score of 2). Psychological testing established a moderate ID (IQ 45), with impaired speech and language skills, difficulties with interpersonal relationships and oppositional behavior. She presented giggle incontinence. Endocrine investigations were as follows: normal levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol, prolactin, elevated plasmatic levels of testosterone, dehydroepiandrosterone (DHEA)-sulfate, and 17-hyroxyprogesterone (OH). The chromosome analysis was performed for the patient and her parents using the G-banding technique on metaphase chromosomes from peripheral blood lymphocytes, according to standard protocol. Chromosome C-banding was performed by the standard BSG (barium hydroxide/saline solution/Giemsa) method [] with slight modifications. Genomic DNA was purified from peripheral blood using Wizard Genomic DNA Purification Kit (Promega Corp., Madison, WI, USA). The SNP array was performed using Human CytoSNP-12 v2.1 BeadChip platform (Illumina Inc., San Diego, CA), containing approximately 300,000 SNPs per sample, according to the manufacturer’s instructions. The data were processed using Genome Studio V2010.1 software (Illumina). Genomic positions were defined according to the GRCh37/hg19 Assembly of the Human Genome (February 2009). italic #text We describe a female patient with some signs of Turner syndrome, mild dysmorphic face, minor features of holoprosencephaly (HPE), small hands and feet, excessive hair growth on anterior trunk and ID. The karyotype showed an unbalanced translocation between chromosomes 14 and 18 resulting in the formation of a dicentric derivative chromosome. Single nucleotide polymorphism array analysis revealed three abnormalities: an 18p deletion flanked by a duplication, and a 16p11.2 duplication. The translocation is as both parents had a normal karyotypes. Non Robertsonian dicentric autosomes are rare findings, reported in only 26 cases in a review by Lemyre []. The majority of cases involve the acrocentric chromosomes, with a short arm breakpoint, followed in frequency by chromosome 18. Most of the heterodicentric autosomes have only one primary constriction on metaphase chromosomes, and the constriction is noticed mostly at the site of the non-acrocentric centromere [], as in our case. Deletions of p arms of acrocentrics containing nucleolus organizer regions (NOR) regions are not known to be associated with phenotypic anomalies, and have therefore probably not contributed to the phenotype. Also, the 1.15 Mb 18p duplication is not likely to have contributed to the phenotype, since most patients with trisomy 18p have normal or mild phenotypes, and may or may not have ID []. Our patient displays some of the features of 18p- syndrome, such as ID, features of the HPE spectrum (mild microcephaly, single central maxillary incisor), and features evocative of Turner syndrome (short stature, mild webbed neck, pectus excavatum, broad trunk and narrow hips) (). Facial dysmorphism (triangular face, blue sclera, bilateral preauricular sinus) is different from that described for 18p deletion, excepting the oromandibular region. Facial appearance has changed over time, becoming elongated (), as described also by Tsukahara []. Congenital cardiac defects, present in our case, have been observed in 10.0% of cases of 18p- []. Although the phenotype described above is not characteristic for monosomy 18p, the standing position with widespread legs and leaning slightly forward as well as marked slowness in motion and action are very suggestive for this chromosomal syndrome. Recurrent 16p11.2 microduplications were initially associated with phenotypes ranging from normal to ID, autistic spectrum disorders and psychiatric problems [–]. Other studies showed that these duplications can manifest with dysmorphic features without a recognizable pattern, microcephaly, congenital anomalies (including torticollis, cleft lip and palate, pectus excavatum, pectus carinatum, mild scoliosis, hypospadias, phimosis, tethered cord, pes planus), and seizures []. Jacquemont [] showed that 16p11.2 duplication is associated with a BMI <18.5 kg per m in adults and <-2 SD from the mean in children. Among the features mentioned above, our patient exhibited mild microcephaly, pectus excavatum, mild scoliosis and ID, but these features are also described in 18p deletion. She was underweight during childhood, but recovered later, her BMI being within normal range as an adult. Considering that empiric estimate for penetrance of proximal 16p11.2 duplication established a penetrance of 27.2%, and the likelihood of a normal phenotype is ∼73.0% [], we cannot clearly conclude how this copy number variation (CNV) influences the phenotype. More recently, a patient with thoracolumbar syringomyelia and a 16p11.2 duplication has been described []. Although our patient presented kyphoscoliosis and nocturnal enuresis, MRI of the spine showed no changes. In a study of three patients with 18p deletion, Portnoi [] suggested that there might be a critical region for GH deficiency between 18p11.23 and 18pter. Our patient has a deletion which includes that region, but the level of GH is normal and the craniocerebral CT did not show any pituitary gland anomalies. The critical region for ID has been tentatively mapped between 18p11.1 and 18p11.21 []. Our patient has a deletion distal to this point and moderate ID, but this feature may be due to the 16p11.2 microduplication. Brenk [] proposed round face to map to the distal 1.6 Mb of 18p, and post-natal growth retardation and seizures to the distal 8 Mb. Our patient has a terminal deletion larger than 10 Mb, but she had no history of seizures, and the face was triangular in childhood and elongated in adulthood. Considering that a pointed chin can be noticed in five out of 13 patients with 16p11.2 duplication for which the facial features were presented [,], we appreciate that the triangular aspect of the face may be due to this rearrangement. Ptosis and short neck, frequently associated with 18p- [], were attributed by Brenk [] to the proximal half of 18p. These features were absent in our patient, in whom the proximal 5.1 Mb of 18p was not deleted. Thus, haploin-sufficiency of genes located in this region may be responsible for these features. Our patient has a microform of HPE, although only 10.0% of patients carrying an 18p deletion (including the gene) present HPE []. Holoprosencephaly is a complex developmental disorder in which multiple genetic and environmental factors can affect the severity of the phenotype []. A recent array CGH study of a large group of HPE patients demonstrated a high frequency of submicroscopic anomalies involving known but also novel HPE loci, including 16p11.2 []. Therefore, the 16p11.2 micro duplication present in our patient can be a second genetic event contributing to HPE manifestation. In conclusion, we report a female patient with a pseudodicentric 14;18 chromosome that carries two additional CNVs. These CNVs confer phenotypic variability to 18p- syndrome, leading to difficulties in establishing the contribution of each abnormality to the phenotype. Although the phenotype of 18p-syndrome is not as typical as for other syndromes, HPE microform and Turner stigmata associated with characteristic posture and marked slowness in motion and action is very suggestive for this syndrome. Microarray analysis of our patient allowed us to define precise molecular characterization of the translocation breakpoints and to uncover two unsuspected cryptic abnormalities, improving genotype-phenotype correlations and management.

Pericentric inversions of the human Y-chromosome [inv(Y)] are rather common and show an estimated incidence of 0.6–1:1,000 in males in the general population []. Most of the reported cases with inv(Y) are familial [] and may include progeny with aneuploidies, preferentially +21,XXY and other chromosomal syndromes. For the carriers of inv(Y) the risk of mental retardation, multiple abortions or phenotypic abnormalities is not apparently increased [,]. It makes sense that inv(Y) neither impedes the production of normal sperm nor does it predispose non disjunction of other chromosomes in the progeny [–], , it is normally considered as a chromosomal heteromorphism [–]. However, inv(Y) has also been reported in association with infertility [] and in patients with either concomitant minute Yq11 deletions [] or a breakpoint in the ‘deleted in azoospermia’ () gene cluster region []. In this report, we present a detailed molecular and molecular cytogenetic characterization of a family having an inv(Y) (p11.2q11.221∼q11.222) over two generations and different clinical outcomes. A 37-year-old male, his unrelated 32-year-old wife and the two sons of the family are reported, aged 15 and 17, respectively. The mother was healthy and had a karyotype 46,XX. Father and both sons had varying features of mental retardation but otherwise normal phenotypes. The parents did not have any history of miscarriages. Cytogenetic analysis using GTG-banding was performed according to standard procedures []. A minimum of 20 metaphases analyzed from stimulated peripheral blood cultures were analyzed. Karyotypes were described according to the International System for Human Cytogenetic Nomenclature []. Fluorescence hybridization (FISH) was carried out on the metaphases using commercially available probes (LSI SRY), subtelomeric for Xp/Yp and Xq/Yq and centromeric for Y (DYZ3) (Abbott Molecular/Vysis, Des Plaines, IL, USA). Additionally centromere-near probes of the previously reported subcentromeric multicolor-FISH (subcenM-FISH) mix were applied []. A minimum of 20 metaphases spreads were analyzed. The results were evaluated on a fluorescence microscope (AxioImager.Z1 mot; Zeiss, Jena, Germany) equipped with appropriate filter sets to discriminate between a maximum of five fluorochromes and the counterstain DAPI (4’,6-di-amino-2-phenylindole). Image capturing and processing were carried out using an Isis imaging system (MetaSystems, Altlussheim, Germany). The azoospermia factor (AZF) microdeletions on the Y-chromosome were detected according to the procedure of Al-Achkar []. p Constitutional chromosomal abnormalities are an important cause of miscarriage, infertility, congenital anomalies, and mental retardation in humans. The frequency of structural chromosomal abnormalities has been estimated as 0.25% in live-born infants []. Chromosomal polymorphisms of the constitutive heterochromatin regions of chromosomes 1, 9, 16, and the Y-chromosome have been reported []. Mental retardation results from a defect in the structure and function of the neuronal synapse. Its worldwide incidence [intelligence quotient (IQ <70)] is ∼2.0–3.0%. Males are found to be more affected than females. The risk of mental retardation is higher in children with congenital structural defects []. The cause of mental retardation may be genetic (30.0%) or environmental, congenital or acquired. Chromosomal aberrations account for 15.0% of mentally retarded individuals. Several types of structural aberrations are also known to cause mental retardation, the common ones being deletions, duplications, inversions, translocations and/or isochromosome formation []. Isodicentric Y chromosomes [idic(Y)] are formed by homologous crossovers between opposing arms of palindromes on sister chromatids. The authors propose that intrapalindrome sequence identity is maintained non crossover pathways of homologous recombination. DNA double-strand breaks that initiate these pathways can be alternatively resolved by crossovers between sister chromatids to form idic(Y) chromosomes, with clinical consequences ranging from spermatogenic failure to sex reversal and Turner syndrome []. In all inv(Y) chromosomes cases previously described, which appear metacentric after banding analysis, the inversion breakpoints on the short arm in Yp11.2 fall in a gene-poor region of X-transposed sequences proximal to the pseudo autosomal regions () on the X- and Y-chromosomes at the end of the short (p) arm SRY [,]. However, in our familial cases, the long arm inversion breakpoint maps proximal to the fertility genes and in Yq11.223, resulting in our familial inv(Y)-type II. A similar familial inv(Y) case has been published []. In conclusion, we present a detailed molecular-cyto-genetic characterization of a family who had an inv(Y)p11. 2q11.221∼q11.222 with mental retardation features but an otherwise normal phenotype. However, molecular analysis of some genes implicated in mental retardation and detection of small gains or losses of genetic material using appropriate new high-resolution test methods and/or genome sequencing should be considered for future studies.

Treacher Collins syndrome (TCS, OMIM 154500), also known as mandibulofacial dysostosis, is a rare developmental disorder of the craniofacial region and is one of the most severe forms of mandibulofacial dysostosis syndromes []. The disease was named after the report of Treacher Collins in 1900 [] and reviewed in detail by Franceschetti and Klein in 1949 []. The estimated incidence is about 1/50,000 live births []. Treacher Collins syndrome is transmitted in an autosomal dominant manner, but only 40.0% of cases are familial [,]. Mutations of the gene, which is located on 5q32-q33.1, have been found to be responsible for most of the cases. The gene encodes the treacle protein, a serine/alanine rich protein predicted to have an important role in ribosome biogenesis []. More than 130 mutations have been reported in the gene. Deletions ranging from 1 to 40 bases are the most common causes of TCS []. Clinical manifestations of TCS include downward slanting of palpebral fissures, hypoplasia of the zygomatic complex and the mandible, complete or partial cleft palate, coloboma of the lower eyelids, and atresia of external ear canals with abnormalities of the external ears accompanied by conductive hearing loss [,]. The phenotypic variability of the clinical symptoms in TCS is a major obstacle for its diagnosis. Although there is as yet no clear explanation, genotype and phenotype discordance have been reported in some studies [,]. Phenotypic expression of the syndrome is so mild in some individuals that they cannot be distinguished physically, while some others experience sudden death after birth due to the respiratory distress []. The patient was a 9-day-old female with multiple congenital anomalies. She was the first child of a non consanguineous couple of Turkish origin. The mother was 33 years old and the father was 35 years old. There was no family history. Mother was under treatment for hypothyroidism during pregnancy, and a history of polyhydramnion existed. The infant was born Cesarean section at term. She was 3100 g and 49 cm tall; head circumference was 34 cm. Downward slanting of palpebral fissures with bilateral absence of zygomatic bones were accompanied by coloboma of both lower eyelids. The cystic appearance of the left upper eyelid was examined (). Eyelashes were absent in the medial part of both lower eyelids. The patient’s nose was broad with a wide nasal bridge and a flattened root. Microtia and external ear abnormalities with conductive deafness and choanal atresia were noted. She had preauricular tags. There was complete cleft palate with retrognathia and micrognathia. A tracheostoma was done at the age of 2.5 months. On examination at the age of 4.5 months, motor developmental delay was observed. Transcranial ultrasonography was normal. Heparinized and EDTA blood samples from the parents and from the patient were taken after informed consent forms were obtained. The GTG-banded chromosomes of the cultured lymphocytes revealed a 46,XX karyotype. DNA samples were extracted from EDTA-blood of the family members. Sanger sequencing of the gene of the patient revealed a heterozygous c.1021_1022delAG deletion in exon 7 of the gene (NG_011341.1). The deletion causes a frameshift mutation and a premature stop codon at position 348 (p.Ser341Glnfs*7; NM_001135243.1) of the treacle protein. After sequencing of both parents for the mutation site, we determined the mutation arose as a mutation in our patient (). Primer sequences and polymerase chain reaction (PCR) conditions can be shared upon request. Clinical manifestation of TCS is widely variable, even in cases with the same mutation. Hypoplasia of zygomatic bones and downward slanting of palpebral fissures are the most common findings, but in some cases, these minimal diagnostic criteria may not be distinguished or evident during physical examination. No clear explanation exists for the different presentations of TCS. The wide variability has been attributed to modifier genes, epigenetic factors and the role of the non mutated alleles [,]. The patient described in this report was diagnosed on physical examination; the TCS diagnosis was subsequently confirmed by molecular analysis. She has a severe phenotype with a score of 18/20 when we evaluate her clinical characteristics according to the scoring system developed by Teber []. In the literature, a previous report described four further TCS cases in which the same mutation was present []. The clinical characteristics of these cases are summarized in . As in our patient, one of the four previous cases is also of Turkish origin (Case 1, ). The similarity of the clinical characteristics of the TCS in these two Turkish cases is dramatic. The only criteria that could not be assessed with respect to its similarity is speech development, which could not be evaluated in our patient due to her age. In the prenatal period, both mothers had a history of polyhydramnion, which is a common finding in pregnancies where a newborn is diagnosed with TCS [–]. For this reason, in pregnancies in which the mother experiences polyhydramnion and findings on ultrasound imaging are abnormal for the facial anatomy, the obstetrician may take the possibility of TCS into consideration. The remaining three cases also reported to have the c.1021_1022delAG mutation on the gene possess the minimal diagnostic criteria. Two of them are father and daughter from the same family; among five cases with the c.1021_1022delAG mutation, the mildest one is the father. As mentioned in previous reports [,], the wide variability in the clinical spectrum among cases in patients who carry the c.1021_1022delAG mutation on the gene reflects the differences in clinical presentation for TCS () [,]. While all five cases possess at least the minimal diagnostic criteria for TCS, and the high degree of similarity in clinical features for the two severe Turkish cases is obvious, it still seems difficult to suggest the phenotype from the mutation status as usual in TCS.

italic xref #text Dermatitis as a presenting symptom of CF seems rare and has been previously reported in 24 other patients (). However, the incidence of this presentation of CF is likely higher because of unreported cases. The pathophysiology of the cutaneous symptoms of CF is unclear but it is thought that lack of protein, zinc and essential fatty acids (especially linolenic acid) play a role (,–). The hypo-proteinemia, zinc and essential fatty acid deficiency are due to malabsorbtion in CF and secondary to pancreatic failure. The rash typically appears in infancy, from age 2 weeks to 15 months. The eruption usually begins in the diaper area, has a predilection for the perioral area and the extremities, and can extend to cover the entire body. The initial eruption consists of erythematous, scaly papules and progresses within 1–3 months to extensive, desquamating plaques (,,,). Vesicles, bullae and pustules can also occur. In our patient, the rash appeared at 2 months of age as an eczematous dermatitis on the lower extremities. Because of the history of atopy in our patient’s family, her rash was diagnosed as atopic dermatitis and treated with H1-receptor antagonists, neutral cream and hypoallergenic infant formula. Despite the therapy, skin lesions extended on to the face, arms and trunk and desquamating plaques in the inguinal and popliteal regions. Regarding hair depigmentation as a presenting symptom of CF, the hair of patients in two reviewed cases (,) was completely gray. In a third case (), there was hypo-pigmentation of the proximal 1 cm of scalp hair as in our patient, together with the depigmentation of the eye lashes at the proximal parts. The hypopigmentation of hair probably results from a combination of nutritional deficiencies. It has been shown that lack of the amino acid tyrosine and coenzymes required for the synthesis of pigments in the hair and skin, result in changes in the hair color and hyper-pigmentation of skin in protein-energy-malnourished children (). Elements of the rash observed in CF resembled the skin lesions found in protein-calorie malnutrition, essential fatty acid deficiency, and acrodermatitis enteropathica (AE), an autosomal recessive defect causing primary zinc deficiency. Severe protein malnutrition can lead to skin changes consisting of erythematous plaques, desquamation, stomatitis, glossitis, thinning nails and alopecia. Deficiency in fatty acid, especially linoleic, is associated with periorificial cutaneous eruptions consisting of dry, thickened, erythematous and desquamating plaques (). In the reviewed literature (–,,), all patients had normal mucous membranes and nails, like our case, contrary to what is seen in AE patients. The differential diagnosis of this type of rash beside AE, protein-energy malnutrition, biotinidase deficiency, also includes psoriasis, seborrheic dermatitis, atopic dermatitis, Langerhans cell histiocytosis, epidermolysis bullosa and immunodeficiency syndromes such as Wiscott-Aldrich syndrome, Netherton’s syndrome and severe combined immunodeficiency (–,–).

xref italic #text A 13-hour-old male infant, the first-born of a non consanguineous marriage to a 23-year-old father and a 21-year-old mother, presented cyanosis half an hour after birth. The baby was delivered by Cesarean section at a local hospital at 38 weeks’ gestation because the ultrasound assessment showed the amniotic fluid was less than normal. He was gravida 3, para 1, and born without asphyxia history. His birth weight was 2750 g with an Apgar score of 10 at 1 min. Half an hour after birth, he exhibited cyanosis of the lips and face when taking a bath. He was immediately administered oxygen inhalation using head hood and medicine treatment through an intravenous injection. He was transferred to our hospital due to the fact that his symptoms did not improve after therapy. On the way to our hospital, he was administered oxygen inhalation through a nasal catheter, and the cyanosis was relieved. Physical examination showed that the child was 50 cm in height and his head circumference was 33.5 cm. The boy’s anterior fontanelle was patent and flat without broadening cranial sutures. The genitalia were normal immature male. On admission, he presented with tachypnea, cyanosis and slight hypertonia. The features of DS including hypertelorism, slightly lowset ears with protruding pinna, were obvious. Chest radiography showed exudative lesions in the lungs. Two-dimensional echocardiography indicated complex CHD with the presence of an ostium secundum atrial septal defect (diameter 0.6 cm, bidirectional shunt flow), enlarged right ventricle and mild tricuspid valve regurgitation (). Cytogenetic study performed on peripheral blood samples using standard procedures revealed a complement of 48 chromosomes with two extra chromosomes in the G group. Fifty metaphases from PHA-stimulated peripheral blood lymphocytes demonstrated a karyotype of 48,XXY,+21 according the International System for Human Cytogenetic Nomenclature (ISCN) (). There was no evidence of mosaicism and the diagnosis of double aneuploidy involving chromosome 21 and X was made. Chromosomal karyotypes of the parents were unknown due to their refusal to be tested, and they were counseled accordingly. Double aneuploidy, the existence of two chromosome anomalies in the same person, is rare, which can involve autosomes (chromosome 13, 18 or 21) and sex chromosomes. The causes of aneuploidy are not well-documented, however, it is known that the most common chromosomal mechanism is meiotic non disjunction. The cause of non disjunction is also uncertain. Non disjunction can occur during meiosis I when the chromosome pairs fail to separate or during meiosis II when chromatids fail to separate. Generally, the mother contributes the extra chromosome 21 in 85.0% and the father in 15.0%, of cases. However, the extra X chromosome is contributed by the parents contribution in 50.0% of cases. Our patient is an infant, who exhibited typical DS features with a 48,XXY,+21 karyotype, born at term to a 21-year-old mother and 23-year-old father. It is evident that the risk for trisomy 21 in offspring increases with maternal age, and the maternal age effect was also demonstrated in the 47,XXY aneuploidy []. A previous study also suggests women aged 20 through 24 years have the lowest prevalence rate of DS (1/1400 births), and for women aged 35 years, the rate is approximately 1/350 births, and for women over 45 years old, the rate rises to 1/25 births [–]. Thus, maternal age-related factors seem to play a more important role in the etiology of 48,XXY, +21 than genetic predisposition. Kovaleva and Mutton [] reported that the double aneuploidy of 48,XXY,+21 is age-dependent, with a mean maternal age of 33 and a mean paternal age of 37.9. However, in our case, the parents were very young. From the published cases, we can conclude that the paternal ages are remarkably different in patients with a double aneuploidy of 48,XXY,+21 associated with CHD () [–]. The exclusion of advanced maternal age as risk factor for chromosomal non disjunction in the present case suggests the existence of other risk factors. The occurrence of double aneuploidy of DS combined with KS is unclear, not to mention the double aneuploidy associated with CHD. Approximately 65 cases of double aneuploidy of XXY and trisomy 21 have been published since 1959, and there are only eight cases associated with CHD [–], including our case (). It is well known that 40.0–50.0% of patients with DS and half the patients with KS have CHD []. The incidence of cardiovascular anomalies in patients with 48,XXY,+21 karyo-type is not clear. To the best of our knowledge, only eight case reports of CHD in these patients have been published (). These patients have less vascular anomalies than the general population, probably because of an increased inhibition of vascular endothelial growth factor, whose genes are located on chromosome 21 []. In conclusion, DS-KS syndrome is an extremely rare condition. We present a case of 48,XXY,+21 karyotype with typical features of DS, whose parents were very young. Together with the other seven cases of double aneuploidy associated with CHD, the maternal ages are different, which do not support the maternal age effect.

Balanced reciprocal translocations result from exchange of fragments between two chromosomes, without any gain or loss of genetic material, and are a common form of chromosomal abnormalities, occurring in about 1 in every 625 newborns [–]. Although, these translocation carriers usually do not exhibit any particular phenotypes, there is a balanced complement of genes. These are responsible for a high incidence of infertility, pregnancy loss, mental retardation, behavioral abnormalities, morbidity and mortality. Carriers of reciprocal translocations have reduced fertility and thus form an increased risk of having a spontaneous abortion or an unbalanced karyotype in their offspring [,,–]. Pure trisomy due to non disjunction of chromosome 21 is responsible for 96.0% of Down Syndrome with a recurrence risk of less than 1.0%. Parental karyotype is not required in non disjunction type of trisomies [–]. The population risk for trisomy 21 is 1 in 700 births but some couples are at a much higher risk owing to parental translocation or mosaicism []. Trisomy 21 due to reciprocal translocations are caused by exchange of euchromatic regions of chromosome 21 with the euchromatin regions of different autosomes or gonosomes. In addition to trisomic regions in various lengths and location of chromosome 21, unbalanced forms also show partial mono-somy for the exchanged regions of the other translocation chromosome. As a result, the phenotype in the carriers of an unbalanced translocation is not consistent []. Since these translocations rarely occur, there are no reliable data for their incidence, but their frequency is assumed to be less than 1:1000 in standard trisomy cases. According to reports in previous literature, the most common partners for a reciprocal translocation seem to be chromosomes 18 and 22 []. The true mechanisms responsible for structural rearrangement at segregation remain unknown. There is some evidence that considers chromosomal translocations as a risk factor for aneuploidy, therefore, translocations have to be considered in combination with aneuploidy analysis []. In one of our recent studies we assessed a family in which the translocation between chromosomes 12 and 16 segregates; one of the eight progenies carried the 47,XY, +21,t(q24;q24) karyotype and presented with Down Syndrome []. His mother was phenotypically normal, one brother and one sister also carried the same translocation. Apparently, this rearrangement occurred due to the unbalanced chromosome segregation of the mother [t(q24;q24)mat]. Here, we present a segregation of a balanced translocation between chromosomes 3 and 21 [t (q21; q22] of a phenotypically normal mother that led to one offspring with a viable balanced translocation and Down Syndrome, one child with a normal karyotype and one phenotypically normal fetus with the translocation. This rearrangement apparently originated from the mother [t(q21;q22)mat]. A 30-year-old woman with a history of two previous pregnancies was referred by the Department of Obstetrics and Gynecology, Çukurova University, Adana, Turkey, to our genetics laboratory for prenatal diagnosis due to having a terminated pregnancy because of a fetus with Down Syndrome and a risk of positive triple test screening at 16 weeks of gestation. The biopsy of the terminated fetus was performed by the pathology department of our faculty for confirmation. The woman and her 35-year-old husband were healthy and phenotypically normal. Both parents were subjected to chromosomal analysis, based on standard blood lymphocyte culture and G-banding techniques. Twenty metaphases were microscopically analyzed for the parents. They were not consanguineous and the mother had become pregnant a total of three times, as can be seen in the pedigree of the family (). Of these three pregnancies, only one resulted in the birth of a phenotypically and karyotypically normal male baby. Karyotypes of the fetuses of the second generation of the family’s pedigree were performed on fetal cells that were obtained from the amniotic fluid sample by using a long-term cell culture. After adequate growth, cultures were harvested after an average 8 to 10 days. Karyotyping was routinely performed by G-banding using the trypsin-Giemsa staining technique. At least 20 metaphases were analyzed. Chromosome analysis confirmed that all cells of the mother had a translocation between chromosomes 3 and 21; her karyotype was 46,XX,t(q21;q22) (). The father’s karyotype was normal. In the mother’s first pregnancy, the fetus was a carrier of the translocation between chromosomes 3 and 21 with trisomy 21; the karyotype of these fetus was 47,XX,+21,t(q21;q22) (). The family decided to terminate the pregnancy in the second trimester. The diagnosis of Down Syndrome was verified at the clinical examination of the terminated fetus. Bilateral telecanthus (+), nose origin flatness (+), back of neck was short and abundant nuchal skin with cystic hygroma residues (). Brachydactylia was also seen on the right hand’s third finger (+) and clinodactyly of both right and left hands’ fifth finger (+) (). Simian line was observed on both right and left hands (). The fetus was phenotypic-ally female. Of these pregnancies, only the second resulted with the birth of phenotypically and karyotypically normal male baby. In the third pregnancy, the fetus was a carrier of the translocation between chromosomes 3 and 21; the karyotype of this fetus was 46,XX,t(q21;q22) (). This pregnancy was also terminated in the second trimester at the family decision. Autosomal reciprocal translocations have been proposed as the most common chromosomal changes in couples who have recurrent pregnancy loss (RPL). In parallel, reciprocal translocations were the most common abnormalities (2.9%) in our series as reported in the literature, and all the chromosomes that are involved in these reciprocal translocations were found in autosomes []. Although reciprocal translocations are balanced rearrangements, they are important for the offspring of carriers who have increased risk of a chromosomal imbalance during gameto-genesis due to unequal meiotic segregation. Especially when one of the parents is a carrier of a balanced reciprocal translocation, pregnancy may result in one of the three different types of offspring: a child with a normal karyotype, a child with a balanced reciprocal translocation, or a conceptus with an unbalanced karyotype that may lead to a spontaneous miscarriage or a live-born child with malformations and mental retardation. Cytogenetic findings do not only lead to RPL but also increase the frequency of bearing a malformed child, therefore, genetic counseling for subsequent pregnancies of couples who have balanced translocation is important [,]. It is generally accepted that balanced rearrangements lead to increased non disjunction of other chromosomes during meiosis. Carriers of balanced translocations may be apparent because of recurrent miscarriages with or without healthy and/or affected children []. Prenatal diagnosis is essential if there are recurrent miscarriages for couples with a known chromosome rearrangement and if there is advanced maternal age. Maternal age is the only well-established risk factor for Down Syndrome, and the associated risk increases exponentially at age 35 years and over [,,]. As a result, amniocentesis is regularly recommended for women at age 35 and over. It is also recommended that case reports of clinically normal subjects with balanced karyotypes should be published so that an informed decision can be made prenatally when a similar rearrangement is identified. In the rare group of parents carrying a balanced chromosome rearrangement affecting whole or partial 21q, the risk of having a child with a complete or partial trisomy 21 is relatively high, with about 20.0% in females and 10.0% in males. If the translocation chromosomes and their normal homologous show pairing difficulties in meiosis I, the additional risk of a 3:1 segregation has to be taken into account, leading to a recurrence risk of up to 30.0% []. We concluded that carriers of reciprocal translocations including chromosome 21 are at increased risk of having offspring with trisomy 21. Thus, the t could promote formation of trisomy 21 in the offspring. Recent studies quite clearly indicate that interchromosomal effects (ICEs) do exist [,–]. It has been hypothesized that a familial translocation frequently induces errors of pairing of het-erologous chromosomes in the prophase of meiosis I leading to an aneuploid gamete (ICE). Recent investigations could not prove this hypothesis. Kovaleva [] found that carriers of balanced reciprocal translocations or inversions but not a Robertsonian translocation, are at increased risk of bearing a trisomy 21 offspring. According to this researcher, these data do not support the existence if ICE in its common sense, , as an effect of rearrangement on another chromosomes’ segregation at the carrier’s meiosis. Nowadays, it is assumed that the reduced fertility of the translocation carriers is the reason for pregnancies at increased maternal age that leads to an elevated risk for pregnancies with trisomy 21 [,,]. Translocation of chromosome 21 (4.0% of Down Syndrome) recurrence risk varies between 10.0–25.0%, if one parent is a carrier of a translocation comprising chromosome 21 []. The risk of unbalanced translocation in the offspring will depend on both the type of translocation in the parents, and which parent is affected and whether the translocation is between homologous or non homologous chromosomes. If the parents are carriers of balanced translocation, risk for unbalanced translocation in the fetus is high and all subsequent pregnancies require prenatal sampling. Once an unbalanced translocation in the fetus/ child has been identified, parental karyotype is essential. More than 50.0% of the translocations in a fetus are . So if parents have a normal karyotype, no matter what type of translocation in the fetus, recurrence risk is minimal <1.0% []. We may then expect carriers of balanced parental structural rearrangements to be at a greater risk of having an offspring with a distinct aneuploidy.

Neuroendocrine tumors (NET) represent a diverse group of malignancies occurring throughout the body, the estimated incidence is 5.25/100,000 and is increasing (), though the reason for this phenomena remains unknown. NETs are the most common small bowel tumor () and overall the incidence of NETs in the gastrointestinal system is common (). They share a typically indolent growth pattern and manifest often symptoms related to tumor-induced hormonal secretion. NET patients frequently suffer from neuroendocrine tumor liver metastases (NETLM). It has been estimated that 46–93% of NET patients have NETLMs at the time of diagnosis (). Although these tumors progress slowly, the 5-year survival of NET patients with NETLMs is 40% compared to that of 75–99% in patients free of liver metastases (). Surgery is considered as the only potentially curative treatment method for the NETLMs. Surgical removal is typically considered if the disease is restricted to the liver, although surgical tumor debulking can be considered to control carcinoid syndrome in selected cases. Unfortunately, only <20% of patients with NETLMs are candidates for hepatic resection (,). Liver transplantation for metastatic NETs remains controversial (). The preferred primary treatment of NETLMs is surgical management, followed by liver-directed therapies, or a combination of these procedures (). Liver-directed therapies include thermal ablation, hepatic artery (chemo)-embolization, and selective intra-arterial radiation therapy. It is known that ablative techniques may have curative potential when small liver tumors are treated. Of thermal ablation techniques, the radiofrequency ablation is most widely used (,). Other modalities that can be used to achieve local ablation are laser-induced thermal therapy (LITT), cryoablation, microwave therapy, electroporation, and high intensity focused ultrasound (HIFU) but there are very few, if any, reports with any of these methods in conjunction with treating NET tumor liver metastases. All ablative techniques are based on the cytotoxic effects of non-physiologic temperatures that are focally induced within the treated tumor by percutaneously or perioperatively placed probes (apart from HIFU, which is totally non-invasive). Ablation techniques can be applied in the setting of inoperable disease, or, at the surgeon’s discretion as a complement to resection (). Indications for image-guided ablation are recommended as follows: adjunct intraoperative ablation; ablation in non-surgical patients; ablative debulking for symptom relief; and ablation of metastatic relapse after surgery (). Other NET therapy is mainly systemic and not specifically targeted to the liver but rather towards universal tumor volume in the body. The number of therapy options is voluminous, and their utilization partially depends on whether the tumor is hormonally an active functioning tumor or an inactive non-functioning tumor (). The target is to suppress the symptoms and the disease progression. Treatment is palliative, typically applied in a situation where there is systemic disease involvement and possible disease progression according to RECIST criteria (). These therapies include somatostatin analogs, proton pump inhibitors, systemic peptide receptor radionuclide therapy (I-mIBG, Y-DOTATOC, Y-DOTATATE, Lu-DOTATATE), chemotherapy, interferon-α, targeting vascular endothelial growth factors (sunitib), targeting mTOR pathway and micro RNA-regulated pathways (everolimus) (). Laser (Light Amplification by Stimulated Emission of Radiation) has been investigated and used in medicine since the 1960s. It currently permeates nearly every area of modern medicine from early diagnostic to therapeutic uses (). Laser-induced thermotherapy has been used successfully to treat tumors in the brain, lung, prostate, kidneys, and liver (,). Magnetic resonance imaging (MRI) provides excellent soft tissue contrast resolution and can be used guide percutaneous ablative therapy (,). MRI also is the only imaging modality that allows for noninvasive, real-time temperature monitoring during ablation procedure using a visualization of relative temperature values of the tumor and surrounding healthy tissue, here the proton resonance frequency (PRF) is the most widely used method but the thermal T1 effect can also be utilized (–). It has been shown that laser-induced interstitial thermotherapy can help to achieve survival rates similar to those seen with surgical resection in liver metastases from colorectal and breast cancer, as well as in other abdominal tumors, although randomized studies are scarce (). The largest published series of any percutaneous ablative technique is from laser ablation, the series consisted of liver metastases mainly from colorectal carcinoma (). A few small series and case reports of local ablative treatments have shown good response in liver metastases from NETs (,). Specific MRI-guided laser ablation reports on NETMLs management are still absent. The purpose of this report is to describe technique for MRI-guided laser ablation of NET liver metastases and present the treatment results of two patients. o p a t i e n t s w i t h h e p a t i c m e t a s t a s e s f r o m N E T w e r e t r e a t e d . I n f o r m e d c o n s e n t f r o m b o t h p a t i e n t s a n d i n s t i t u t i o n a l e t h i c a l s t a t e m e n t w e r e a c q u i r e d . I n a l l , t h r e e t u m o r s w e r e t r e a t e d . A p r e p r o c e d u r a l ( 1 . 5   T ) a b d o m i n a l M R I w a s p e r f o r m e d f o r b o t h p a t i e n t s p r i o r t o t h e l a s e r a b l a t i o n . A l o w f i e l d 0 . 2 3   T C - a r m M R I s c a n n e r w a s u s e d f o r i m a g i n g a n d p r o c e d u r a l g u i d a n c e . P o s t o p e r a t i v e i m a g i n g w i t h M R I ( 1 . 5   T ) o r c o m p u t e d t o m o g r a p h y ( C T ) w a s p e r f o r m e d a t i n t e r v a l s o f 6 m o n t h s t o 1 y e a r d u r i n g t h e f o l l o w - u p p e r i o d . Both patients were treated successfully with complete tumor ablation. The procedural times were 80 min and 126 min. There were no significant complications, but one patient (Patient 1) developed slight pleural effusion, segmental hepatic tissue, and capsular edema at a treatment site, which did not require further treatment. The energies delivered to tumors were: Patient 1, 54,000 Joules (J); and Patient 2, 12,000 J and 21,600 J at segments II and III, respectively. During the follow-up, Patient 1 has remained disease-free with normal gastrin and GgA levels (disease-free survival, 10 years; total survival, 21 years) (). He has been receiving medical treatment (omeprazol) to prevent gastric symptoms but is currently without medication. Patient 2 had increasing CgA levels 3 years after the laser ablation and a concomitant octreotide scan revealed two new small metastatic lesions in liver and also one small extrahepatic, paraspinal lesion. Due to the extrahepatic disease, further invasive treatment was not chosen and treatment with somatostatin analogue medicinal therapy was initialized (Sandostatin LAR, long-acting repeatable, Novartis, Nürnberg, Germany). Since that time, patient CgA levels have remained stable but he has developed two additional metastatic lesions to the liver at new locations (5 years after initial laser ablation). The disease has remained stable for 5 years and the patient was free from carcinoid syndrome symptoms for 9 years from the initial laser ablation. Total survival time is 13 years. Due to lack of symptoms and currently stable tumor situation, no immediate local ablative therapy has been planned. Treatment options for hepatic metastases of neuroendocrine tumors are numerous and several treatment algorithms have been published (,,,). In general, the treatment must be tailored specifically to the patient by a multidisciplinary group of oncologists, surgeons, interventional radiologists, and endocrinologists. Depending on the tumor grade, which is determined by the cellular proliferation (Ki-67, mitotic index), the NET tumors have a variable progress rate and thus patient survival is also capricious. Advances in loco-regional disease control and pharmacotherapy have led to better management and survival of NET patients. It is evident that as the survival times of these patients increase, and the prevalence of NET patients increases, the health system needs increased awareness of how to manage these patients. Here especially, the symptoms present a significant, quality of life (QoL) issue. This is especially pertinent should the patient have carcinoid syndrome. Symptoms can often be effectively controlled by medication but sometimes tumor mass removal or debulking through surgery or ablation is warranted for QoL amelioration. In our report, institutional multidisciplinary team of surgeons, oncologists, endocrinologists, and radiologists made treatment decisions and utilized these methods for patients benefit. Both of our patients show long survival: one is disease-free and the other is symptom-free although with new hepatic metastases. The first of our patients undoubtedly would not be disease-free without the ablation. The second patient has two new metastases and it is impossible to estimate which part of his survival is contribution of the ablative treatment or pharmacological therapy and which is due to the indolent nature of the disease. It is remarkable that neither of our patients developed local recurrences at the primary operation or ablation sites. As previously addressed, thermoablative treatments are effective in treatment of different types of hepatic tumors and MRI is feasible in monitoring thermal effects in tissue during and after the treatment (–). MRI is particularly well suited for thermal therapy monitoring as it enables multiplanar thermal monitoring during the ablation using either T1 effect or PRF-method. Also the novel development of miniaturized instruments (,) to achieve thermal laser ablation make the procedure less invasive and comparable to other thermal therapy methods, such as radiofrequency ablation, cryotherapy, interstitial electroporation, microwave, and brachytherapy. It must be acknowledged that most of these methods, apart from cryotherapy, do not facilitate seamless utilization of intraoperative MRI monitoring. We had no significant complications associated with the LITT and it is generally acknowledged that percutaneous ablative therapies are safe. There were no local recurrences at the treated liver site in our patients and it seems that MRI guidance combined with laser ablation may be more accurate in lesion obliteration than radiofrequency ablation when liver metastases are considered (). Concerning NETs, our experience shows, and as others have noted (), that in the treatment of NETLMs a multidisciplinary approach is of paramount importance. In conclusion, laser-induced thermotherapy implemented with MRI, MRI guidance, monitoring, and careful clinical follow-up seems to be the safe and feasible method to manage endocrine tumor liver metastases in a situation where other treatment paradigms are not possible.

In the domain of pulmonary surgery, advances have been made in thoracoscopic surgical techniques for diagnostic excisional biopsies of pulmonary nodules as well as for therapeutic resection of peripheral lung malignancies (). For small and deeply situated pulmonary nodules, however, a major factor limiting success of thoracoscopic resection is the difficulty in locating the target nodule because it cannot be palpated digitally. Fluoroscopy-assisted thoracoscopic resection of a small lung nodule marked with Lipiodol, which is generally used as a contrast medium for lymphatic vessels, has been reported to be useful in these cases (,). Recently, we experienced a patient who underwent fluoroscopy-assisted thoracoscopic resection after marking of nodules with Lipiodol and thereafter developed pneumonia. A 33-year-old man with multiple metastases to both lungs from a testicular tumor was referred to our hospital. After a high orchiedectomy, he was given chemotherapy. Even though the pulmonary lesions decreased in size, complete remission was not achieved. Therefore, fluoroscopy-assisted thoracoscopic resection was attempted for 12 pulmonary lesions in the left lung, of which four were very small Two of the small lesions were located in the lower lobe and two in the upper lobe. Because these four nodules would be too small to detect by fluoroscopy during the operation, Lipiodol marking was done under local anesthesia before thoracoscopic resection. Under CT fluoroscopic guidance (X Vigor Laudator; Toshiba Medical System, Tokyo, Japan), a 21-gauge needle (PEIT needle; Hakko Medical, Chikuma, Nagano, Japan) was advanced to the area adjacent to each lesion (). When the needle tip was confirmed in a position adjacent to a lesion, 0.3 mL of Lipiodol (Laboratoire Guerbet, Roissy, France) was injected (). For three of the lesions, the procedure was performed with the patient in the prone position. For the remaining lesion in the upper lobe, the patient was placed in the right lateral recumbent position. No complication related to marking with Lipiodol occurred. After the marking procedure the patient was taken to the operating room, and thoracoscopy was performed under general anesthesia with the patient in the right lateral recumbent position. A fluoroscopic unit with a C-shaped arm was used to identify the radiopaque nodules, which were grasped with a ring-shaped forceps during fluoroscopy using multiple projections. Successful resection was confirmed by viewing each radiopaque nodule within the resected specimens under the C-arm fluoroscope. Sixteen days after surgery, the patient experienced general fatigue and fever of 39℃. White blood cell count was 7230 and C-reactive protein was elevated to 19.55 mg/dl. Chest CT showed multiple ground-glass opacities (GGOs) distributed at the upper, middle, and lower lobes of the right lung (). Within the GGOs in the upper and middle lobes, high density spots corresponding to Lipiodol were seen. The patient was then treated for pneumonia, although information from a sputum culture, procalcitonin testing, and polymerase chain reaction testing was not obtained. After intravenous infusion of antibiotics, the symptoms disappeared. On CT 1 month later, the GGOs had markedly disappeared (). Fifty-two days after the fluoroscopy-assisted thoracoscopic resection in the left lung, the pulmonary lesions in the right lung were resected by the same procedure. Several methods are used for localizing small pulmonary nodules before thoracoscopic resection. These include percutaneous injection of water soluble dye () or Lipiodol (,,), percutaneous insertion of a hook wire (,), and barium marking via bronchoscopy or percutaneous injection (). Because of rapid diffusion of dye in lung tissue after injection, the general dye method has drawbacks. Marking must be performed within 3 h before thoracoscopy to enable dye detection and diffusion sometimes blurs the injection site (,). With water soluble dye, dye-marked lesions cannot be localized after formalin fixation (). Barium appears as a lesion in hematoxylin and eosin-stained sections and can cause inflammatory changes in lung tissue, possibly making a pathologic diagnosis difficult (,). With the hook-wire technique, dislocation of the hook wire has caused varying degrees of failure and was reported to cause massive air embolism (). Lipiodol, as used in the present case, offers the following advantages: (i) over-resection of normal lung tissue can be avoided because Lipiodol marks nodules as clear spots; (ii) Lipiodol can remain for up to 3 months after marking; (iii) Lipiodol does not affect pathologic findings; and (iv) because Lipiodol diffuses only to a small extent, leaving a clear spot, even deeply situated nodules can be easily localized (). A 100% success rate for the use of Lipiodol for marking prior to thoracoscopic resection was reported (,). Interstitial pneumonia induced by I-labeled Lipiodol administered from the hepatic artery as treatment of hepatocellular carcinoma was reported (). Earlier, pneumonia following chemoembolization for liver tumors in which Lipiodol was used was described (). Suggested mechanisms for such pneumonia were Lipiodol embolism by arteriovenous shunting or hepatic vein invasion, an immunoallergic phenomenon, and/or radioactive-induced lesions in patients after I-labeled Lipiodol infusion (). To our knowledge, however, pneumonia occurring after Lipiodol marking has not been reported. We hypothesize the following to explain the occurrence of pneumonia in this patient. Part of the Lipiodol injected for marking the pulmonary nodules in the left lung distributed in the alveoli and bronchioles. The Lipiodol then moved to the contralateral lung transtracheally by gravity during the thoracoscopic resection procedure during which the patient was placed in the right lateral recumbent position. Thereafter, Lipiodol distributing in the entire right lung was involved in causing pneumonia. The occurrence of this complication might be avoided by minimizing the time that a patient spends with the opposite lung dependent. In conclusion, care must be taken to avoid the occurrence of pneumonia after Lipiodol injection to localize pulmonary nodules before fluoroscopy-aided thoracoscopic resection.

Cerebral arteriovenous malformations (AVMs) are congenital vascular abnormalities in which arteriovenous shunting occurs through an abnormal vascular network (nidus) in the parenchyma. These lesions typically present with cerebral bleeding, seizures, headache, or neurological deficits; however, they are sometimes found incidentally (). As this abnormality has a high risk of rupture (2–4% per year), intensive therapeutic interventions are generally indicated, including surgical resection, transcatheter embolization, and stereotactic radiosurgery (SRS). For AVMs that are unresectable due to size or location, SRS can avoid the potential risks associated with traditional surgery; however, complete elimination of the nidus occurs more slowly with this technique than with conventional surgery/embolization, taking 2–4 years (). Although the potential for hemorrhage remains during this latency period, proliferation of the vascular endothelium is induced by irradiation, slowly but progressively obliterating the nidus and subsequently decreasing the volume of blood shunted (). Evaluation of hemodynamic changes in and around the AVM after SRS is therefore crucial for appropriate patient management, and is generally achieved by conventional digital subtraction angiography (DSA) or contrast-enhanced computed tomography (CT)/ magnetic resonance imaging (MRI). However, it is difficult to perform these evaluations with optimal frequency due to factors such as side-effects of contrast material, cumbersome procedures, radiation exposure, and expense. Arterial spin-labeling MRI (ASL-MRI) is a brain perfusion imaging method that can generate images without injection of contrast material (). Its clinical utility has been established in several cerebral conditions such as brain tumors (), infarction, and vascular lesions (). In this case report, we present a long-term sequential series of ASL-MRI as well as conventional MRI of an AVM treated with SRS. ASL-MRI clearly demonstrated hemodynamic changes after SRS, in a non-invasive manner and without administration of contrast material. It may, therefore, be useful in evaluating the therapeutic response of AVMs treated by SRS. A 42-year-old man presented to our hospital with a 3-year history of recurrent seizures. Contrast-enhanced MRI was performed with a 3.0-T clinical MRI scanner (Signa Excite HD, GE Healthcare, Milwaukee, WI, USA) and revealed a Spetzler-Martin grade III AVM in the right temporal lobe (). The AVM was fed by branches of the right middle and posterior cerebral arteries, and it drained to the dilated sphenoparietal sinus and other superficial veins. Along with conventional contrast-enhanced MRI, ASL-MRI was performed with a pseudo-continuous ASL sequence (label duration, 1.5 s; post-label delay, 1.5 s; spiral acquisition with 8 arms × 512 points; image matrix, 128 × 128; field of view, 240 mm; locations, 38; number of excitations, 3). These imaging techniques showed high signal intensity in the both nidus and distal portion of the draining veins (). These MRI findings were also confirmed by DSA (). Stereotactic radiosurgery was performed, in which the nidus and marginal areas were irradiated with 22 Gy and 20 Gy, respectively (). The post-radiosurgical course was uneventful. Subsequently, follow-up contrast-enhanced and ASL-MRI were performed at 3, 14, 24, and 30 months after SRS (). At 3 months, the nidus had shrunk slightly on conventional MRI, but the high signal intensity on ASL-MRI and dilatation of the draining veins were unchanged. At 14 months, the nidus had shrunk further and the draining veins were less dilated. Although the high signal intensity of the draining veins was still present on ASL-MRI, it was more proximal than in pretreatment images obtained with exactly same ASL parameters. Finally, at 24 months, when the high flows in the nidus and dilatation of the draining veins became unclear on T2-weighted (T2W) images and time-of-flight MR angiography, the high signal intensity on ASL-MRI had also completely disappeared. However, contrast-enhanced T1-weighted (T1W) images showed ring-like enhancements, probably due to radiation necrosis. These MRI findings remained unchanged at 30 months after SRS (not shown). Consistent ASL-MRI findings of high signal intensity in both the nidus and draining veins of cerebral AVMs have been reported by several authors (,). Hence, the present ASL-MRI findings of a cerebral AVM were in keeping with previous reports. Regarding response of these lesions to treatment, a few studies have focused on hemodynamic changes evaluated by ASL-MRI after radiosurgery/embolization (,). In a study of 21 patients with AVM, Pollock et al. demonstrated quantitatively lower regional blood flow and a progressive decrease in nidal perfusion after SRS when compared with untreated AVM (). However, a detailed visual assessment of hemodynamic changes using ASL-MRI in a single patient has not been well described. In the present case, unchanged dilatation of the draining veins and high signal intensity of this vessel on ASL-MRI indicated unchanged shunting at 3 months. At 14 months, when further shrinkage of the nidus was achieved, the high signal intensity of the draining veins moved more proximally on ASL-MRI obtained with exactly same ASL parameters. This change may represent a decrease of shunting due to partial obliteration of the nidus, since the draining veins also became less dilated. Finally, when no abnormal signals were indentified on ASL-MRI, high flow in the nidus also disappeared on conventional MRI, indicating complete obliteration of the nidus. The remaining ring-like enhancement was probably caused by radiation necrosis. ASL-MRI has several advantages over conventional MRI in assessing treatment response of AVM after SRS. First, ASL-MRI can be performed without contrast material. Thus, it can avoid the potentially lethal side-effects of contrast materials, and it is also less expensive. Second, ASL signals are reported not to be profoundly affected by vessel membrane permeability (). Hence, ASL-MRI is not influenced even when radiation necrosis causes breakdown of the blood–brain barrier, whereas contrast-enhanced MRI shows ring-like enhancements like those observed in the present case 24 months after SRS. This feature might be useful in detecting recurrent high flows in a nidus after SRS during MRI follow-up. In conclusion, the gradual hemodynamic changes in the present AVM after SRS were clearly visualized in a longitudinal ASL-MRI series. Although more cases need to be evaluated, this non-invasive perfusion-weighted MRI method may be useful not only for detecting AVMs, but also for precise assessment of the response of these lesions to SRS.

Xanthogranulomatous pyelonephritis (XGPN) is an atypical form of chronic pyelonephritis characterized by the destruction of the renal parenchyma and replacement with a chronic inflammatory infiltrate of lipid-laden macrophages, known as xanthoma cells (,). XGPN is usually classified in diffuse and focal forms, with the diffuse form accounting for >90% of cases (,). It usually affects middle-aged women and is extremely uncommon in children (,), accounting for 0.6% of histologically documented cases of chronic pyelonephritis (–). Its exact etiology remains unknown. However, it usually occurs in association with nephrolithiasis, urinary tract obstruction, and/or chronic urinary infection, with common pathogens such as , , , , and (). This disease has been called as the “great imitator” because the clinical and radiological findings closely resemble other pathological entities such as renal cell carcinoma (). The preoperative distinction between XPGN and malignant kidney tumors is often difficult. We report an unusual case of a 73-year-old woman presenting with asymptomatic right renal mass, incidentally discovered in a post-traumatic screening ultrasound. Right open radical nephrectomy was performed and the final histopathologic examination, despite the absence of symptomatology (lumbar pain, fever, anorexia, and weight loss) was consistent with the diagnosis of XGPN. xref fig #text XGPN is a severe form of chronic pyelonephritis, characterized by the destruction of the renal parenchyma and replacement by granulomatous tissue. Its name is derived from yellow color on gross pathology and a granulomatous reaction histologically. XGPN is usually divided into three stages: stage I (nephric XGPN), the inflammation is confined to the kidney; stage II (perinephric XGPN), the inflammation involves both the kidney and peri-renal fat; stage III (paranephric XGPN), the inflammation involves the kidney, peri-renal fat, and the retroperitoneum (). The etiology is still unclear, but appears to be multifactorial. It is clearly related to a combination of renal obstruction and chronic bacterial infection and usually occurs in association with nephrolithiasis, urinary tract obstruction, and chronic urinary infection (,). Other causes of obstruction include congenital abnormalities such as uretero-pelvic junction obstruction and tumors that occur mainly in the adult population (renal cell carcinoma, ureteral carcinoma, bladder carcinoma) (,). Other factors implicated in the etiology of XGPN include altered immune response and intrinsic disturbance of leukocyte function, alterations in lipid metabolism, lymphatic obstruction, malnutrition, arterial insufficiency, venous occlusion and hemorrhage, and necrosis of the pericalyceal fat (,,,,). The most commonly reported symptoms are fever, abdominal and/or flank pain, weight loss, malaise, anorexia, and lower urinary tract symptoms. Pyuria is present in 60–90% of patients. Common findings at physical examination are a palpable mass and flank tenderness. Rarely, in 5% of patients, a draining renal cutaneous fistula in the flank may be present (,). Laboratory tests include leukocytosis, anemia, and increased elevated sedimentation rate in the majority of patients. Urine cultures are usually positive at the time of diagnoses. The most common pathogens are , , and rarely , , and . Although the urine cultures may be negative, cultures of renal tissue at surgery are often positive for these pathogens. The US pattern of XGPN corresponds to that of a solid mass with inhomogeneous echoes US can show enlargement of the entire kidney with multiple hypoechoic areas representing hydronephrosis and/or calyceal dilatation with parenchymal destruction, as well as calculi. US may also help to differentiate the two forms of XPGN as focal and diffuse: in the diffuse form, generalized renal enlargement with multiple hypoechoic areas representing calyceal dilatation and parenchymal destruction is seen; in the focal form, a localized hypoechoic mass, often misdiagnosed as renal tumor, may be found (–). CT scan has been shown as one of the best preoperative diagnostic tests for the evaluation and confirmation of XGPN. Features that have been considered characteristic (but not pathognomonic) for diffuse XGPN are renal enlargement, perinephric fat strand, thickening of Gerota’s fascia, and water density rounded areas in renal parenchyma representing dilated calyces and abscess cavities with pus and debris, described as “bear paw sign”. CT may also reveal an obstructing urinary stone (mostly they are staghorn calculus) in the renal collecting system and absence of excretion of contrast medium, showing loss of function of the affected kidney, in 80% of patients. There may also be enlargement of the hilar and para-aortic lymph nodes. In the focal form, CT usually shows a well-defined localized intra-renal mass with fluid-like attenuation (–). Several reports have described a possible role of MR in the diagnostic evaluation of patients with suspicious XGPN; in particular, Cakmakci et al. () have shown that in the focal form of XGPN the mass has slightly low signal intensity on T2-weighted (T2W) images and is isointense with the renal parenchyma on T1-weighted (T1W) images. These findings suggest a fluid with very high protein content. The different signal intensity of the solid component of XGPN on T1W images, compared with the renal parenchyma, depends on the amount of xanthoma cells involved in the granulomatous process. The T2W sequences are very useful for accurate differentiation between XGPN from tumors. Although MR imaging (MRI) is inferior to CT in demonstrating renal calcifications and ureteral stones, contrast-enhanced MRI can easily demonstrate infiltration of the inflammatory mass into adjacent tissue structures and better demonstrates the anatomical relationship of the XGPN on coronal and sagittal planes, as well as the fat component within the mass and the compressed renal parenchyma (). The differential diagnosis of XGPN include neoplastic diseases such as clear-cell carcinoma, lymphoma, leukemia, Wilms’ tumor, neuroblastoma, and inflammatory processes (renal or peri-renal abscess, pyonephrosis, renal tuberculosis, focal and diffuse nephritis, and fungal infection) (,). The treatment of choice for diffuse XGPN, which is the most frequent form, is surgery and consists of nephrectomy with resection of all other involved tissues, with or without antibiotic therapy. Drainage of peri-renal or renal abscess with adjunctive antibiotic treatment is strongly recommended before definitive surgery, to decrease the complications in the diffuse form of the disease. In the localized form of the disease, segmental resection of the affected kidney is effective. Partial nephrectomy is also recommended in extremely rare bilateral cases (–). Macroscopic appearance of XGPN include an enlarged kidney with a thickened capsule, yellow nodules with or without central necrosis in the renal parenchyma, while the renal pelvis may be dilated and filled with stones, debris, or purulent fluid. Microscopic pathological examination of the yellow areas shows a large number of lipid-laden macrophages (foam cells) with extensive areas of inflammation and fibrosis (,). Misinterpretation of “foam cells” as “clear cells” consistent with renal adenocarcinoma, is the most important diagnostic challenge at histology. In conclusion, the unusual findings of this case report suggest a careful evaluation of patients with a renal cystic mass, especially in case after blunt abdominal trauma, that can be misdiagnosed with a renal cell tumor. A combined CT and MR evaluation together with laboratory and clinical findings are mandatory for a correct differential diagnosis of this rare renal entity.

T e t r a l o g y o f F a l l o t ( T O F ) i s o n e o f t h e m o s t c o m m o n c o n g e n i t a l h e a r t m a l f o r m a t i o n s c o m p r i s i n g a v e n t r i c u l a r s e p t a l d e f e c t , r i g h t v e n t r i c u l a r o u t f l o w t r a c t o b s t r u c t i o n , r i g h t v e n t r i c u l a r h y p e r t r o p h y , a n d o v e r r i d i n g a o r t a . A r a r e v a r i a n t i n c l u d e s p u l m o n a r y a t r e s i a a n d m a j o r a o r t o p u l m o n a r y c o l l a t e r a l a r t e r i e s . We report a case of a 39-year-old female patient with uncorrected TOF and pulmonary atresia presenting to our emergency department with fever, elevated infection parameters, and hemorrhagic macular lesion on the middle finger of the right hand. Patient history revealed a Blalock–Taussig shunt performed for palliation in 1983 due to pulmonary atresia with subsequent spontaneous closure. Corrective surgery failed in 1998. Coronary angiography performed in 1997 was normal. On admission, a contrast-enhanced multidetector computed tomography (CT) of the chest and abdomen was performed to rule out focal thoracic or abdominal infection. The CT study demonstrated an absent pulmonary artery, a perimembranous ventricular septal defect (VSD) of 1.8 cm diameter, an atrial septal defect (ASD), and right ventricular hypertrophy (RVH). Furthermore, major aortopulmonary collateral arteries (MAPCA) arising from the descending aorta, an obliterated ductus arteriosus, and a dilatation of the aortic root (diameter 5.0 cm) were observed (). The patient demonstrated multiple pulmonary and splenic emboli (). Echocardiography revealed an ostium secundum type ASD, a VSD, a moderately reduced left ventricular ejection fraction (45%), and two large vegetations on the tricuspid valve. Blood cultures were positive for staphylococcus aureus. Targeted antibiotic treatment was initiated immediately, and the patient was referred to the cardiac surgery department for tricusipid valve repair which was performed 2 days after admission. Due to a protracted clinical course and neurologic deterioration after surgery a cranial CT was performed. Several hypoattenuating cerebral and cerebellar lesions were predominantly localized in the right hemisphere consistent with thromboembolic events (). The further course of the patient was unfavorable, showing progressive cardiorespiratory insufficiency and increasing demand for vasoactive agents. Eighteen days after initial admission the patient passed away, due to cardiopulmonary failure. TOF is one of the most common congenital heart malformations. It consists of an interventricular communication, a biventricular connection of the aortic root, which overrides the muscular ventricular septum an obstruction of the right ventricular outflow tract, and RVH (). The variants of TOF include TOF with a patent foramen ovale/ASD (pentalogy), TOF with absent pulmonary valve, and TOF with pulmonary atresia (). TOF with pulmonary atresia is clinically and radiologically distinct from regular TOF. It comprises 5–10% of all tetralogy complexes (). The anatomy of central pulmonary arteries is often abnormal (,). MAPCA are systemic-to-pulmonary collateral arteries representing remnants of the embryonic ventral splanchnic arteries and provide an alternative pulmonary blood supply in patients with TOF and pulmonary atresia. These embryonic vessels normally regress concomitantly with the formation of the normal pulmonary arterial system in the first weeks of gestation, whereas in patients with pulmonary atresia they persist (). Survival rates in TOF with pulmonary atresia without surgical repair reported in the literature are as low as 50% at 1 year of age and 8% at 10 years (). Our case shows an adult survivor of uncorrected TOF with pulmonary atresia, the oldest survivor ever reported in the literature being 59 years old (). As postulated by Fukui et al. the survival might by primarily dependent on the adequacy of pulmonary blood flow derived from MAPCAs (). TOF with pulmonary atresia may lead to a more severe RVH than in non-variant TOF due to a functional single-ventricle circulation. Altered hemodynamics within this functional single-ventricle results in turbulent flow and may be more dominant in the right chamber than in the left due to the complete absence of the physiologic right ventricle outflow. Turbulent blood flow predisposes to endocardial vegetation formation on the tricuspid valve which may consequently lead to thromboembolic events. Usually emboli and vegetations from the right heart are filtered by the lung. This is not the case in TOF variants with systemic shunts, where dangerous systemic emboli to the brain and other vital organs may occur. In conclusion, TOF with pulmonary atresia is a rare entity of congenital heart disease presenting typical findings on cross-sectional imaging including VSD, right ventricular outflow tract obstruction, right ventricular hypertrophy, overriding aorta, and MAPCA. Uncorrected blood flow in patients with palliative surgery or failed correction predisposes to endocardial vegetations and consecutive thromboembolic events within the systemic circulation.

M a l i g n a n t p e r i p h e r a l n e r v e s h e a t h t u m o r ( M P N S T ) i s a m a l i g n a n t s o f t t i s s u e t u m o r a r i s i n g f r o m p e r i p h e r a l n e r v e s h e a t h c e l l s . M P N S T i s u n c o m m o n a n d c a n b e a s s o c i a t e d w i t h n e u r o f i b r o m a t o s i s t y p e I ( N F - I ) . M P N S T i n v o l v i n g b o n e i s v e r y r a r e , a n d t h i s i s t h e f i r s t r e p o r t e d c a s e o f M P N S T i n v o l v i n g t h e d i s t a l p h a l a n x o f t h e f i f t h t o e i n a p a t i e n t w i t h o u t N F - I . A 76-year-old man with no personal or family history of NF-I presented with a 1-year history of a swollen, red, and slightly painful right fifth toe. He had consulted the Dermatology Clinic at Kobe University Hospital because of pain and occasional bleeding. He had a 7-year history of diabetes mellitus that was being treated with insulin. Examination revealed no signs of NF-1. The right fifth toe was swollen, indurated, and erythematous, with nail destruction and a small ulcer. High-resolution magnetic resonance imaging (MRI) was performed using a 5-inch diameter microscopy coil, showing a tumor occupying the whole distal phalanx, with irregular thinning of the cortical bone and extension into subcutaneous soft tissues. The lesion was hypointense to muscle on T1-weighted (T1W) imaging and homogeneously hyperintense on T2-weighted (T2W) imaging and short-inversion-time inversion-recovery imaging. Heterogeneous enhancement was observed after gadolinium administration, mainly in the marginal regions (). Histological examination of a skin biopsy specimen showed massive cell proliferation in a storiform pattern extending from the superficial dermis to the subcutis, predominantly comprising spindle-shaped clear cells with abundant cytoplasm and atypical hyperchromatic nuclei (). Immunohistochemical staining for S-100 showed moderately positive results in the atypical spindle cells (), whereas staining for desmin, α-smooth muscle actin, factor XIIIa, CD34, cytokeratin (AE1/AE3), HMB-45, and melanA yielded negative results (). Malignant melanoma and clear cell sarcoma were therefore excluded as possible diagnoses. A diagnosis of MPNST involving the bone was made, and the right fifth toe and fifth metatarsal bone were amputated with no adjuvant chemotherapy. At 1 year after surgery, local recurrence was detected, and further investigation revealed multiple pulmonary metastases. MPNST is a malignant neoplasm of the connective tissues and nerves. NF-I is the most significant risk factor for MPNST. Approximately 50% of MPNSTs develop in patients with NF-I, and 4.6% of NF-I patients develop MPNST (). MPNSTs in non-NF1 patients occasionally arise from major nerves such as the sciatic nerve, sacral nerve, or brachial plexus. The most frequent site of MPNST involving the bone is the mandible (). It has been proposed that the long course of the alveolar nerve through the mandible may predispose to development of the tumor. However, this view is questioned, as other nerves with extensive intraosseous courses show low frequencies of MPNST. Other sites of MPNST involving bone include the vertebrae (), femur (), and calcaneus (). MPNST involving the bones of the digits is very rare. To the best of our knowledge, only one case of MPNST of the digits has been reported, in the distal phalanx of the thumb of a 6-year-old girl who did not have NF-I (). The incidence of MPNST is considerably higher in non-NF1 children than in non-NF1 adults (). The present case is unique in that the MPNST arose in the subcutaneous tissue of the fifth toe of a non-NF1 patient, where there are no major nerves. MPNST involving bone is rare because MPNST usually arises in the soft tissues. There are three possible mechanisms for bone involvement in MPNST. First, the tumor may arise from within the bone, as intraosseous MPNST. Second, the tumor may invade the bone through a nutrient foramen, producing a dumbbell-shaped lesion as it enlarges. Third, an extraosseous tumor arising in the soft tissues may invade the adjacent bone (,). In this case, the bone was not involved via a nutrient foramen and there was no dumbbell shape to the lesion. The lesion involved both the bone marrow of the distal phalanx and the subcutaneous fat around the bone, and determining whether the tumor originated in bone or soft tissue was difficult. From the perspective of MRI findings, the greater bulk of the tumor was intraosseous, the bone marrow seemed to be totally infiltrated, and the lesion seemed to have arisen within the bone. However, based on the histopathological features, the tumor is thought to have originated in soft tissues () and involved the adjacent whole bone through penetration. Most tumors arising in the digits are small, and there are technical limitations to the spatial resolution that can be achieved when evaluating such tumors using conventional MRI. In this case, high-resolution MRI using a 5-inch-diameter microscopy coil allowed us to evaluate the features and extent of the tumor. Heterogeneous enhancement of the marginal region reflected internal degeneration of the tumor (). Use of a microscopy coil yields a high signal-to-noise ratio, and high-resolution MRI using a microscopy coil is therefore useful for evaluating small lesions of the hand and foot. Preoperative evaluation of the location and extent of the lesion are very important for surgical planning (). The imaging findings of MPNST involving bone are non-specific. Relative to muscle intensity, MRI usually shows an isointense lesion on T1W imaging and a hyperintense lesion on T2W imaging, with various patterns of enhancement depending on the degree of degeneration (,), as in our patient. Cortical thinning of the bone and invasion of the soft tissues around the bone are usually accepted as signs of malignancy, but have also been described in benign schwannoma (). Differential diagnoses for the lesion in our patient included osteolytic lesions such as bone metastasis, lymphoma, plasmacytoma, and malignant fibrous histiocytoma, and soft tissue tumors such as metastasis, synovial sarcoma, and dermatofibrosarcoma protuberans. In conclusion, diagnosis of MPNST with bone involvement may be difficult in small bones because of the rarity of the condition and the non-specific radiological findings, and definitive diagnosis requires histopathological examination.

The term “Marjolin’s ulcer” describes the formation of malignant tumors in chronically inflamed skin such as non-healing ulcers or previously traumatized, burned, or scarred skin. Epidermoid malignant tumors, such as squamous cell carcinomas, basal cell carcinomas, and malignant melanomas account for over 90% of Marjolin’s ulcers (). Usually the patient sustains tissue damage with scarring or chronic ulcer for several decades prior to diagnosis. “Acute” transformation just a few months after primary injury is also described, however significantly less frequent (,). The pathophysiology, of Marjolin’s ulcer is hypothetically multifactorial (,,). A chronic ulcer or scar suddenly changing in characteristics is highly suspicious of cancer and should lead physicians to perform a diagnostic survey. Symptoms such as fever, malaise, and weight loss are typically absent (). Marjolin’s ulcers are generally recognized as very aggressive tumors with high mortality rate when regional lymph node metastases or distant metastases are present (). The metastatic rate varies between studies, but is generally reported to be approximately 30% (,). The diagnosis may be difficult to establish due to necrotic areas and areas without malignant transformation (). In a recent retrospective study, fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) was found to be helpful in the differentiation between malignant and benign ulcers (). A 56-year-old man was admitted to the hospital with a 20 × 25 cm giant ulcer on the right upper quadrant of the abdomen. Destroying underlying subcutaneous tissue and muscles, the ulcer was covered by malodorous necrosis and surrounded by a discrete peritumoral inflammation (). Fifteen years earlier a small pedunculate skin lesion had been removed from the area without performing histological annalysis on the removed specimen. After 3–4 years without symptoms, a small ulcer appeared in the same area with slow increase in size. One year prior to admission tumor size accelerated rapidly with increase in exudate and malodor, but no pain was reported. He had a high administrative position in a large company. His medical history was uneventful. He was in good general condition at the time of admission. Blood analysis showed mild anemia with a hemoglobin level of 6.6 mmol/L (ref 8.0–10.0) and mild hypoalbuminemia of 26 g/L (ref 37–48). White blood cell count was 11.9*10-9/L (ref 3.0–10.0) and C-reactive protein was 199 mg/L (ref <10). Initial punch biopsies were performed showing low-differentiated basosquamous cell carcinoma with numerous mitoses and perineural growth. Based on the PET/CT scan (including low-dose CT without contrast enhancement) () there were FDG-avid ipsilateral axillary ( and groin lymph nodes, suspicious of metastases. Furthermore a few lymph nodes less suspicious of metastases were identified in the contralateral axilla () and in the parasternal region. At contrast-enhanced CT, performed 1 week earlier, the parietal peritoneum seemed free of tumor invasion, which was essential to surgery planning (). During surgery a simultaneous resection of lymph nodes in the ipsilateral axilla and groin was performed, because of the findings at PET/CT. The entire ulcer was excised with a wide margin of 3 cm including the underlying costa and peritoneum, leaving a 10 × 10 cm peritoneal defect (). Reconstruction was executed with a large free latissimus dorsi musculocutaneous flap enforced with a Vipromeche and a combined Monocryl and Prolene mesh. The thoracodorsal vessels were anastomosed micro-surgically to the deep inferior epigastric artery and vein, with immediate good perfusion of the flap. The donor site was covered with a split thickness skin graft. The tumor was completely removed with clear margins of resection and surprisingly no metastasis in the resected lymph nodes. Postoperative recovery was uneventful (). PET/CT scan was repeated after 4 months, showing no signs of recurrence. The diagnostic reference standard of Marjolin’s ulcer is the punch biopsy. However, in the case of a giant tumor like the one in question, it is unlikely a single biopsy would be diagnostically representative for the entire lesion and biopsy should be guided by metabolism imaging like FDG-PET/CT (). In cases of deep dermal invasive limb lesions, amputation has been suggested rather than wide local excision (,). This emphasizes the need for imaging assistance in staging and surgery planning. In the assessment of the tumor site, PET/CT was highly suggestive of involvement of the abdominal muscles, but at contrast-enhanced CT the parietal peritoneum seemed unaffected. Tumor invasion into the parietal peritoneum would obviously exclude the patient from radical surgery. Although regional lymph nodes are the most frequent site of metastasis, liver lung, brain, kidney, and other distant metastases can also be observed (). In a tumor of the size and location presented in the current case, sentinel node biopsy would be pointless. Not only would engagement of bilateral axillary and groin lymph nodes be expected, but could potentially also involve lymph node regions unavailable to biopsy, for instance in the parasternal region. Whole body PET/CT was useful in the evaluation of lymph nodes, and in guiding lymph node resection selectively. Ipsilateral lymph nodes metastases could not be ruled out due to the FDG uptake and the appearance on CT which showed marginally enlarged nodes with an eradicated fatty hilus. On the contrary the contralateral lymph nodes were less FDG-accumulating and had a preserved fatty hilus on CT, as observed in inflammatory lymph nodes (). A drawback is, of course, that PET/CT may not be able to detect micrometastases (). FDG-PET/CT was very useful to exclude distant metastases in the patient, which is a recognized feature of PET/CT in skin cancer (). Moreover PET/CT was excellent to follow-up and monitor the patient after surgery, when lymph node metabolism had normalized. In conclusion, the limitation of FDG-PET is, being an unspecific modality, the uncertainty of differentiating between metastatic and inflammatory lesions. Of the resected lymph nodes, none proved to be metastatic at biopsy. The reason of the positive PET/CT was most likely infection and inflammation in the giant ulcer.

xref #text A 33-year-old man injured in a motorcycle accident had sustained compound open fractures of the left tibia and fibula with remarkable backward dislocation of the bone fragments (). Distal pulses were not detected after external fixation of the below-the-knee fractures, so emergency angiography was conducted to identify arterial injury. A left lower extremity angiogram showed complete occlusion from the popliteal artery to the tibioperoneal trunk and the origin of the anterior tibial artery (). Because the occluded portion was at the same level as the dislocation site, we deduced that intimal injury, and not a thrombus, was the primary cause of the occlusion. The patient had large skin defects and deep muscular lacerations, which we diagnosed as contaminated wounds. Surgical repair such as replacement of the artery with venous or synthetic grafts was considered to have a high risk of infection; therefore, endovascular recanalization was attempted. A 4-French sheath (Medikit, Tokyo, Japan) was inserted into the left common femoral artery via the antegrade approach. A 0.035-inch guidewire (Radifocus; Terumo, Tokyo, Japan) was passed through the true lumen of the occluded portion and exchanged with a 0.018-inch guidewire (Platinum Plus; Boston Scientific, Natick, MA, USA). The lack of extravasation of the contrast medium and disruption of the vessel wall suggested the absence of transection or laceration, so endovascular recanalization was continued. Heparin was administered intravenously (5000-U bolus, 1000 U/h) and a balloon catheter system (SLALOM; Cordis, Bridgewater, NJ, USA) was introduced co-axially. Then, stepwise inflation with three balloon catheters of 3-, 4-, and 5-mm diameters and 4-cm length was performed 10 times for 3 min (). During the procedure, we observed gradual joining of the intimal flap to arterial wall and sufficient peripheral blood flow through the repaired true lumen (). Satisfactory peripheral blood flow was maintained with anticoagulation therapy (continuous administration of intravenous heparin, 15,000 U/day). However, 14 days after the first endovascular treatment, the patient developed distal ischemia following internal fixation. Emergency angiography showed re-occlusion of the popliteal artery caused by formation of a thrombus adhering to the intimal flap (). Bending of the popliteal artery was also observed. We assumed that the re-occlusion was caused by distortion of the vessel, which probably occurred during the internal fixation. As a secondary treatment, balloon inflation of the true lumen and aspiration thrombectomy with a 5-French guiding catheter (Launcher; Cordis, Bridgewater, NJ, USA) were performed. The thrombus was removed, and recanalization and sufficient peripheral blood flow were again achieved. After the two sessions of endovascular treatment, adequate blood perfusion was maintained (), and the patient was moved to a rehabilitation hospital 55 days after admission. Ischemia did not recur during the 15-month follow-up period. A 74-year-old woman with knee osteoarthritis underwent left total knee replacement. Around 3 h postoperatively, she developed calf pain, swelling, and loss of the distal pulses in the foot. Compartment syndrome of the lower extremity caused by acute arterial occlusion was suspected, and the patient was transferred to our hospital 12 h after the initial operation. Emergency angiogram of the left leg showed focal thrombotic occlusion and dissecting pseudoaneursym of the popliteal artery (), which we diagnosed as iatrogenic intimal injury. The patient continued to suffer from compartment syndrome and experienced vasospasm in the leg; as surgical reconstruction was considered to be difficult, endovascular recanalization was attempted. The thrombus was removed by aspiration thrombectomy with an 8-French guiding catheter (Launcher; Cordis, Bridgewater, NJ, USA) inserted via the antegrade approach from the left common femoral artery; however, the intimal flap and dissecting pseudoaneurysm remained. Moreover, severe vasospasm was noted in the distal arteries of the leg (). The primary goal was to prevent limb loss. Therefore, a bare stent of 6-mm diameter and 4-cm length (E-Luminexx; BARD, Tempe, AZ, USA) was placed to cover the injured site. Immediately after recanalization, remarkably high pressures of 90 mmHg in the lateral compartment and 83 mmHg in the anterior compartment were confirmed. Therefore, fasciotomy was performed in the lateral, anterior, posterior, and deep compartments, and satisfactory peripheral blood flow was achieved (). The patient was moved to another hospital for rehabilitation 28 days after admission. She could stand and walk unaided after rehabilitation, and ischemia did not recur during the 12-month follow-up. In the cases presented here, we performed balloon angioplasty with long inflation time, aspiration thrombectomy, and stent placement to manage traumatic popliteal artery occlusion. The procedures were successful, and promising clinical outcomes were achieved. With regard to the reported risk factors for limb loss (), site of injury (popliteal artery) and associated injury (compound fracture) were risk factors in Patient 1, and site of injury (popliteal artery) and clinical status (compartment syndrome) were risk factors in Patient 2. Despite the high amputation risk and difficult surgical procedures, we successfully salvaged the lower limbs of both the patients. Acute limb-threatening ischemia after blunt arterial trauma is usually managed by surgical bypass of the affected artery. However, in trauma cases, open reconstruction is difficult because of lower-extremity edema and vasospasm of the injured vessels (). Patient 1 had compound open fractures with contaminated wounds, so surgical bypass was considered to have a high infection risk. Patient 2 had compartment syndrome and severe vasospasm in the lower leg; therefore, surgical bypass was a difficult option. Endovascular recanalization was assessed to be a suitable alternative to surgery. Advancements in endovascular and imaging techniques have allowed successful treatment of peripheral arteries and helped to overcome some of the obstacles in surgical repair. In particular, balloon angioplasty in combination with surgery, transcatheter embolization of hemorrhagic lesions, and stent or stent graft placement for dissecting or hemorrhagic lesions have been introduced (,). However, reports of treatment of occlusive lesions are limited (). Balloon angioplasty with long inflation time, the technique we employed to reduce the intimal flap, is used for treating infra-inguinal stenotic or occlusive lesions induced by obliterating arteriosclerosis (). In Patient 1, we re-approximated the intimal flap and achieved and maintained sufficient peripheral blood flow through the repaired true lumen. Aspiration thrombectomy is used to treat thrombotic arterial occlusion (peripheral, intracranial, or coronary arteries) or peripheral venous and pulmonary arterial thromboembolism (). We used this method to achieve sufficient removal of the thrombus in both cases. In Patient 2, we placed a bare metal stent into the P2-3 segment of the popliteal artery. Although stent placement up to the P3 segment is associated with a high risk of stent fracture and re-occlusion (), this procedure was necessary for our patient. She had compartment syndrome of the lower leg, so immediate recanalization was very important in her case. Non-conformance of the stent was not observed during the 12-month follow-up period. In conclusion, the combination of balloon angioplasty, aspiration thrombectomy, and stent placement is an alternative method to surgical bypass for treating traumatic popliteal artery occlusion even in cases with high amputation risk.

Intimal sarcoma, especially arising in the lumen of the pulmonary arteries, is an extremely rare and highly malignant tumor of the vessel walls. In general, primary neoplasms of the cardiovascular system are already rare. The prevalence of these tumors ranges from 0.001% to 0.28% within the literature (). The imaging method of choice is contrast-enhanced chest computed tomography (CT) of the pulmonary arteries, similar to suspected pulmonary embolism. Magnetic resonance imaging (MRI) or positron emission tomography (PET)-CT might also be helpful to detect the enhancement in the luminal mass that indicates a vessel tumor. We present a rare case of a patient with suspected pulmonary embolism that, in fact, was found to be an intimal sarcoma. A 40-year-old woman was admitted to our institution presenting with slowly progressing dyspnea for the last 6 months. She also exhibited non-specific thoracic pain. The discomfort and subjective shortness of breath were initially attributed to musculoskeletal symptoms. Lung function tests demonstrated no significant restriction of the lung volume. A chest CT scan with an intravenous contrast agent was performed to first rule out interstitial lung disease and further pathology of the chest (128 row multislice scanner, Somatom Definition Flash, Siemens Healthcare, Erlangen, Germany; 70 mL of Iomeron 400 i.v. Bracco, Milano, Italy; 4 mL/s flow; reconstructions with 1 and 5 mm axial slice thickness, 5 mm maximum intensity projection in the axial and coronal planes). Chest CT revealed a left-sided incomplete occlusion of the left main pulmonary artery and peripheral complete occlusions of the segmental braches on the same side. On the right side, the main pulmonary artery was nearly completely occluded by a hypodense structure ( and ). The filling defect started at the level of the pulmonary valve, across the bifurcation of the pulmonary trunk, and showed progression in the central and segmental pulmonary arteries on both sides. In the absence of any obvious enhancement, the obstruction was first interpreted as an embolus. Because of the inconsistency of the CT findings and the clinical presentation, concerns arose regarding the potential for another entity, such as a primary tumor of the heart. Next, a high resolution PET-CT scan of the chest was performed (tracer 371 MBq F-18-FDG; Swan Isotopen AG, Bern, Switzerland). The luminal lesions exhibited peripheral FDG uptake, which can indicate tumor tissue but also may be due to granulations in the periphery of a thrombus. The patient was submitted to surgery. After resection of the pulmonary trunk, the surgeons delivered a myxoid tumor adherent to the pulmonary valve ( and ). Intraoperative histological evaluation revealed an intimal neoplasia arising from the vessel wall (). During the same procedure, the surgeons removed the affected valve, the pulmonary trunk, and both pulmonary arteries. A valve-bearing conduit connecting the right outflow tract to the distal pulmonary artery on the right was placed. The left lung was then resected. At the 1-year follow-up examination, we found a tumor recurrence along the right cardiophrenic angle and multiple pleural metastases on chest CT (). Despite operative revision, tumor recurrence was evident on a follow-up MRI at 6 months (3 T MRI scanner, Skyra, Siemens Healthcare, Erlangen, Germany; chest and abdomen; 7.5 mL MultiHance i.v., Bracco Suisse SA; protocol consisted of Haste axial/coronal; Trufi axial; Flash 2 axial fat sat axial; T1 Vibe fat sat coronal; Flash 3D dynamic acquisition coronal) with a recurrence of the involvement of the right cardiophrenic angle. An infradiaphragmatic tumor that had spread through the phrenic hiatus along the inferior caval vein into the liver was observed. Furthermore, there was concern of a new pericardial implantation with continuous invasion of the pericardial sac ( and ). Because of the extent of the local tumor and its inoperability at this stage, palliative care was carried out. Theoretically, neoplasms of the pulmonary arteries can be grouped by their primary location: (i) extraluminal tumors; (ii) intraluminal carcinomas; and (iii) mixed types (). Histologically, these neoplasms can be divided into angiosarcomas, which arise from the intimal cell layers, and soft tissue sarcomas with mesenchymal descent (myofibroblastic, muscular, or osseous and chondral differentiation). Diagnostic clues to angiosarcomas include the expression of the factor VIII antigene or evidence of the so-called Weibel-Palade bodies (). Involvement of the P53 gene in the evolution of angiosarcomas is also discussed. However, angiosarcoma can be induced by exogenous stimulation as former use of Thorotrast (a former contrast agent) and mediastinal radiation for other tumors. The most common locations for sarcomas of the vessel wall include the inferior caval vein, the pulmonary veins, and the heart. However, manifestations in the pulmonary arteries are rare (). Epidemiologic data indicate that the first diagnosis peaks at the age of 50 years. There is no gender predominance. The clinical symptoms are strongly dependent on the site of tumor manifestation. Occlusion of the pulmonary arteries mimics a pulmonary embolus. Symptoms such as dyspnea, shortness of breath, chest pain, and hemoptysis are often present. Tumor involvement of the superior caval vein may manifest with superior congestion, soft tissue edema, or headache. In contrast, tumor occlusion of the inferior caval vein can spread into the liver parenchyma and manifest as a Budd-Chiari appearance. The mode of dissemination is primarily hematogenous. Metastases are found in the lungs, brain, bones, and pleura (). Depending on the location, surgical resection is the main therapeutic option. Stenting and conduit placement offer further possibilities for therapy. Imaging is mostly performed using contrast-enhanced CT to visualize the vascular pathology. The most characteristic finding is a vascular filling defect with varying enhancement of the so-called “pseudo-clot”. Otherwise, enhancement in a filling defect virtually excludes thrombus. Further CT findings include the following: lobulated filling defects, extension beyond the vessel lumen and metastases (most often to the lung and bones). In some cases, MRI may better depict the enhancement pattern of the neoplasm (). The radiographic findings in this case were non-specific; enlargement of the pulmonary arteries or lung nodules can be seen. A recent publication by Attinà et al. investigated the role of PET-CT in the differentiation of chronic pulmonary embolism from pulmonary intimal sarcomas. Based on increases in radiopharmaceutical uptake, PET-CT is able to reliably distinguish between chronic arterial filling defects and tumor tissue. A PET-CT revealing tracer uptake at the level of the arterial filling defect that exceed values of standardized uptake value (SUV) suggest malignancy. In contrast, thrombi generally do not exhibit increased tracer activity. In the case of a chronic embolism, a slight increase in activity may be observed (). In conclusion, one should always consider vascular neoplasms when pulmonary filling defects are suspected after observing enhancement within the clot or if the clinical setting is inappropriate. Vascular filling defects that expand the vessel lumen or grow outside the vessel wall are almost certainly aggressive carcinomas. Invasion of the adjacent structures indicates a local tumor that extends beyond the vessel wall.

xref #text A 72-year-old male patient underwent magnetic resonance imaging (MRI) of the head and neck for follow-up, 1 year after resection of a metastasis located in the right soft palate. The patient had initially a Merkel cell carcinoma of the right elbow with ipsilateral axillary lymph node metastases. The metastasis of the right soft palate mentioned above occurred 1 year after initial diagnosis. MRI was performed on a 1.5 T system (Avanto; Siemens Medical Solutions, Erlangen, Germany). The MRI examination was performed according to the standard protocol for head and neck examination used at our institution, which includes T1-weighted (T1W) images in axial orientation, turbo inversion recovery magnitude (TIRM) images in axial and coronal orientation, T2-weighted (T2W) images in coronal orientation, and contrast-enhanced T1W images with fat saturation in axial and coronal orientation. No metastasis recurrence or new metastases were detected on MRI. However, two relatively symmetrical, irregular protuberances were identified on the lingual aspect of the mandible. The protuberances did not show contrast enhancement and were isointense to compact bone, demonstrating very low signal intensity in all sequences (). The lesions did not show signs of contrast media uptake. The maximum thickness of the lesions was 10 mm. Retrospectively, they were unchanged in comparison to a previous MR, which was performed for follow-up 6 months earlier. A CT scan performed 5 months after the MRI clearly identified the protuberances as solid bony structures (), with densities as high as 1450 Hounsfield units (HU). They were also seen on a photograph made during a subsequent clinical examination (). Based on these cumulative findings a diagnosis of mandibular tori was made. The patient reported no symptoms caused by the tori, which thus had no therapeutical consequence. While the etiology of tori is not well understood, some authors assume that there is a strong hereditary component (,). The reported prevalence of mandibular tori varies greatly between ethnic groups (–). Tori are found almost exclusively in adults. In more than 90% of cases, mandibular tori are bilateral (). Although tori may grow slowly, they are usually asymptomatic, except for some edentulous patients, in whom tori may hinder the fit of dental prostheses. Most tori do not require therapy. Large tori may be removed, especially if they are an obstacle for prosthetic treatment (). Tori are often incidentally identified on computed tomography (CT) scans. On CT, mandibular tori present as bony protuberances, isodense to compact bone and typically located on the lingual aspect of the mandible. Choi et al. reported a thickness range of 4.3–11.3 mm for mandibular tori on CT (). To our knowledge, the MRI characteristics of tori have not been described previously. As described above, the tori proved to have very low signal intensity on all MR sequences, as they consist of compact bone. It may be assumed that smaller tori are difficult to identify on MR because of their low signal. Furthermore, the detection of tori on MR may be problematic due to metal artifacts, which were nearly absent in the patient described in this case report. In our opinion, lesions which are isointense to compact bone on MRI and are located on the medial aspect of the mandible can be classified as tori without additional CT scans. Most benign (e.g. neurofibroma, ameloblastoma) and malignant (e.g. lymphoma, squamous cell carcinoma, multiple myeloma) masses of the mandible are easy to distinguish from mandibular tori on CT or MRI because of bone destruction and lack of osteoblastic component. Chondrosarcoma or osteosarcoma of the mandible may have an osteoblastic component, but they also typically feature partial bone destruction and are unilateral. An important differential diagnosis of mandibular tori are exostotic osteoma found in patients with Gardner’s syndrome (). However, in contrast to tori these osteoma are typically numerous and asymmetrical, and often located on the buccal aspect of the mandible. In addition, patients with Gardner’s syndrome often have tooth impaction and odontoma. In conclusion, our case report summarizes imaging features of mandibular tori, with special emphasis on MRI. The characteristic MRI features of compact bone and the typical location of tori on the medial aspect of the mandible allows differentiation of mandibular tori from other jaw lesions.

T h e d i s s e m i n a t e d f o r m o f s p o r o t r i c h o i d d i s e a s e , d u e t o e r y t h e m a t o u s s p r e a d a n d s o f t - t i s s u e s w e l l i n g w i t h o u t a d e n o p a t h i c m a s s e s , c a n m i m i c a d e e p s o f t - t i s s u e i n f e c t i o n . C o m p u t e d t o m o g r a p h y ( C T ) f i n d i n g s s u c h a s c o n t r a s t - e n h a n c i n g s u b c u t a n e o u s n o d e s a n d t u b e s ( l y m p h a t i c c h a n n e l s ) i n a p a t i e n t w i t h h a n d n o d e s s t r o n g l y s u g g e s t e d t h i s r a r e e n t i t y . F u r t h e r m o r e , C T c a n h e l p i n r u l i n g o u t a i r b u b b l e s o f n e c r o t i z i n g f a s c i i t i s ; t h i s t e c h n i q u e i s m o r e s e n s i t i v e t o d e t e c t a i r b u b b l e s a n d f a s t e r i n e m e r g e n c i e s t h a n m a g n e t i c r e s o n a n c e i m a g i n g ( M R I ) . U n d e r t h e s e c i r c u m s t a n c e s , C T i m a g i n g c o u l d b e o f h e l p i n d i a g n o s i s . italic fig #text italic xref #text

xref #text italic #text Based on the standard of reference, mastoiditis was confirmed in 20 (87%) of 23 patients (), and a consecutive subperiosteal abscess was identified in 12 (52%) of 23 patients. The left mastoid was affected in 12 (52%) of 23 patients and the right mastoid in eight (35%) of 23 patients. The remaining three patients included one case of cholesteatoma, one with otitis media, and one with a solid mass of the mastoid, which was proven to be a manifestation of acute lymphatic leukemia by mastoid biopsy. MRI identified one case of thrombosis of the transverse and sigmoid sinus (), which was confirmed by surgical findings. Furthermore, MRI detected an infratentorial epidural abscess () in two patients, which was also confirmed by surgery. Sensitivity for mastoiditis was 100%, specificity 66%, and accuracy was 86%. Both readers agreed regarding the presence or absence of mastoiditis in 22 (96%) of 23 cases and had differing results in one case (κ = 0.78). Sensitivity for subperiosteal abscesses was 100% and accuracy 100%. All (12 of 12, 100%) subperiostal abscesses were identified by both readers (κ = 1). The results of the current study show that MRI has excellent accuracy for mastoiditis and concurrent subperiosteal abscesses. It also demonstrates the potential of MRI for assessing mastoiditis complications. To our knowledge, no systematic evaluation of the accuracy of MR in suspected mastoiditis has been published previously. At first sight, our findings differ greatly from earlier works, which investigated the use of incidental MR findings for diagnosing mastoiditis. Polat et al. () and Meredith et al. () found that fluid signal intensity in the mastoid should not be interpreted as a sign of mastoiditis. Based on our own clinical experience we fully agree that fluid retention in the mastoid is a common incidential finding of little clinical consequence, if signs of inflammation are otherwise absent. However, Polat et al. and Meredith et al. only evaluated the role of T2-weighted (T2W) images, while the role of DWI or contrast-enhanced MRI is not mentioned. In contrast, our results are based on multiparametric evaluation, as T2W, T1W images with without contrast enhancement, and DWI were used. Our choice of imaging protocol is influenced by previous studies, which has demonstrated the usefulness of both DWI () and contrast media (–) for diagnosing pyogenic infection. While gadobutrol is approved for use in adults and children aged 2 years and older, we used gadobutrol for all patients included in the study, including 13 patients younger than 2 years, because of the potentially life-threatening nature of possible intracranial complications of mastoiditis and because of previous experience with contrast agents in infectious disease. The studies by Polat et al. and Meredith et al. also completely differ from our study in regard to inclusion criteria. These studies retrospectively evaluated patients who received MRI of the skull and the skull base, regardless of symptoms. On the other hand, only patients with clinically suspected mastoiditis were included in our study. The current study shows a rather low specificity of MRI for acute mastoiditis (66%). The specificity of MRI is limited by the fact that fluid collections in the mastoid cells are not necessarily pathological and that contrast enhancement of the mucosal lining of mastoid cell may be caused by hyperemia due to otitis media, even if the mastoid is not infected. However, the estimate of specificity is based on just three patients who did not have mastoiditis. The assessment of specificity is limited by the preselection of patients based on clinical criteria, as most patients without clinical signs of mastoiditis do not receive an MRI scan. A larger sample would be needed for more accurate assessment of specificity. While CT is often used for imaging suspected mastoiditis, a comprehensive evaluation of the accuracy of CT in mastoiditis is lacking. Antonelli et al. have shown that CT is able to distinguish between coalescent and non-coalescent mastoiditis with high sensitivity and specificity (). In this case the reported sensitivity was lower in comparison with our study (67% 100 %), while the specificity was higher (90% 67%). These differences may be caused by differing inclusion criteria used in the studies. While only patients with acute mastoiditis were enrolled in our study, Antonelli et al. also included patients with suspected chronic mastoiditis. Migirov et al. showed that CT has a high sensitivity for subperiostal abscesses (96%), similar to our results (). Pediatric MRI, including MRI of the mastoid, has three important disadvantages in comparison to CT: longer scan times, more limited availability, and the eventual need for general anesthesia. While the logistical challenges presented by pediatric emergency MRI are considerable, MRI in suspected mastoiditis also has considerable potential advantages. Besides the high accuracy demonstrated in the current study, these also include absence of radiation exposure and better sensitivity for the detection of brain abscesses in comparison to CT (). The current study has several limitations. Above all, no comparison with CT, which is routinely used for imaging in cases of suspected mastoiditis by many centers, is available. In MRI, the diagnosis of mastoiditis is based on signs of pyogenic infection and thus coalescent mastoiditis, i.e. a fluid collection, restricted diffusion, and surrounding contrast enhancement. However, MRI is not suited for evaluation of small bony structures like the mastoid septa, whose destruction is an important sign of mastoiditis on CT. This may be a disadvantage of MRI in less pronounced cases of mastoiditis. The use of an echo-planar sequence for DWI can also be considered a limitation. Several authors recommend the use of turbo spin echo diffusion-weighted sequences instead of echo-planar sequences for evaluation of the mastoid because they show fewer susceptibility artifacts (,). One patient did not receive surgical treatment or biopsy and thus no surgical proof for the presence or absence of mastoiditis was available. In this case, diagnosis was based on clinical and imaging follow-up, which is however presumed to be less reliable than surgical findings. The frequency of intracranial complications observed in the current study is low, and because of the relatively low overall number of patients the study is not suited for a conclusive evaluation of such complications. In conclusion, multiparametric MRI has high sensitivity and diagnostic accuracy in suspected mastoiditis. Comparison with CT is needed to decide if MRI can replace CT as the primary method for mastoiditis imaging.

Aggressive fibromatosis or desmoid tumor is an infiltrating fibroblastic proliferation arising from the musculo-aponeurotic structures (,). It either arises in musculoskeletal sites, including the paravertebral musculature and the anterior abdominal wall, particularly in relation to surgical scars, or within the abdomen, involving the mesentery, the retroperitoneum, or pelvis (–). Intraabdominal fibromatosis are rare but those arising from the retroperitoneum are even rarer. Several etiologies have been proposed, which include trauma, abdominal surgery, irradiation, drugs, genitourinary infection, Gardner’s syndrome or familial adenomatous polyposis (FAP). Therefore, the diagnosis of intra-abdominal fibromatosis should be strongly considered when an abdominal mass is detected in patients with a history of previous abdominal surgery or hereditary diseases (). We report a case of retroperitoneal fibromatosis presenting as a presacral mass in a young female patient with no significant medical or surgical history and describe its imaging findings. fig #text Aggressive fibromatosis has been defined as an infiltrating fibroblastic proliferation without evidence of inflammation or definite neoplasia (,). Generally, these tumors occur more frequently in women, particularly in women of childbearing age. They may occur at any age but are seen most commonly in the third and fourth decades (). The cause of this disease is not clear, but several etiologies have been proposed. In our case, the patient did not have any history of trauma, drugs, surgery, or hereditary disease. Our case therefore exhibits the sporadic or primary form of fibromatosis. The primary form is extremely rare and presents fibroblastic proliferation with no connection to the patient’s medical or surgical history (). The diagnosis is difficult to establish preoperatively, especially in the case of a patient with no history of abdominal surgery or injury, drug medication, Gardner’s syndrome, or FAP, such as in our case. Fibromatosis either arises in musculoskeletal sites, including the paravertebral musculature and the anterior abdominal wall, particularly in relation to surgical scars, or within the abdomen involving the mesentery, the retroperitoneum or pelvis (–). Fibromatosis arising from the retroperitoneum are extremely rare and most of the data in the literature are from isolated case reports. In a large study of 189 cases of fibromatoses over 30 years, only eight (4%) were located in either the retroperitoneum or the mesentery (). In another study of 166 desmoid tumors complicating FAP, 83 tumors (50%) were intra-abdominal but only one (0.6%) arose from the retroperitoneum (). The clinical presentations of fibromatosis vary depending on the size of the tumor and the surrounding involved anatomical structures. These tumors tend to invade or surround muscles, tendons, nerves, vessels, the ureter, and bowel as a result of their infiltrating nature. Therefore, ureter or small bowel obstructions occur frequently in the presence of these tumors (). In our case, the inferior vena cava, aorta, and right common iliac artery were invaded. However, there was no involvement of the ureter or bowel perhaps due to the midline location of the tumor which tends to be locally invasive (,,). Fibromatosis often recurs after resection and the reported recurrence rate is in the range of 39–79% (). Therefore, these tumors impose significant morbidity on patients who require a larger subsequent re-excision (). Unlike the fibrosarcoma, fibromatosis exhibit normal mitosis and does not metastasize, but there is no correlation between its clinical behavior and its histologic appearance. The radiologic appearance of fibromatosis depends on the relative amounts of fibroblast proliferation, fibrosis, collagen content, and the tumor vascularity (,). On US, this tumor has a variable echogenicity with a smooth, well-defined margin. On contrast-enhanced CT scans, fibromatosis generally has high attenuation (relative to muscle) and has either an ill- or well-defined margin. On MRI, this tumor has low or iso-SI relative to muscle on T1W images and variable SI on T2W images. The difference in the SI of T2W images appears to be determined by cellularity rather than collagen content (). In our case, the tumor showed iso-SI to muscle on T1W images, low SI on T2W images, and strong enhancement on FS CE-T1W images. In a previous report (), the low SI on T1W and T2W images might have been characteristic for aggressive fibromatosis and therefore significant enhancement might have been expected as these lesions are frequently hypervascular on the previous arteriography and on contrast-enhanced CT studies. In conclusion, in our patient, retroperitoneal fibromatosis appeared as a presacral mass showing infiltrating nature, i.e. vascular encasement and invasion, low SI on T2W imaging, and relatively strong enhancement on FS CE-T1W imaging. This diagnosis is difficult to establish preoperatively, especially in patients without a significant medical or surgical history. Nevertheless, it is preferable to include aggressive fibromatosis in the differential diagnosis, when the presacral soft-tissue mass has an infiltrating nature, low SI on T2W imaging, and significant enhancement on CE-T1W imaging.

xref #text T h e r a d i o l o g y , p a t h o l o g y , a n d i n f e c t i o u s d i s e a s e s d a t a b a s e s o f t w o h o s p i t a l s , o n e u n i v e r s i t y a n d o n e g e n e r a l ( L U M C N L a n d K o n s t a n t o p o u l e i o H o s p i t a l i n A t h e n s , G r e e c e ) , w e r e s t r u c t u r a l l y s e a r c h e d f o r a c t i n o m y c o s i s . B e t w e e n N o v e m b e r 2 0 0 1 a n d F e b r u a r y 2 0 1 1 , 1 8 p a t i e n t s ( 1 5 w o m e n , 3 m e n ; a g e r a n g e , 2 5 – 7 5 y e a r s ) w i t h a b d o m i n o p e l v i c a c t i n o m y c o s i s w e r e i d e n t i f i e d . C o n t r a s t - e n h a n c e d a b d o m i n a l C T h a d b e e n p e r f o r m e d i n a l l p a t i e n t s , o n 1 6   - a n d 6 4 - s l i c e m u l t i d e t e c t o r s c a n n e r s . T h e c l i n i c a l d a t a i n c l u d i n g a g e , s y m p t o m , m a s s s i z e , p r e s e n c e o f i n t r a u t e r i n e d e v i c e ( I U D ) , a n d p r e o p e r a t i v e d i a g n o s i s w e r e r e t r o s p e c t i v e l y a n a l y z e d . B o w e l s i t e , w a l l t h i c k n e s s a n d e n h a n c e m e n t d e g r e e , i n f l a m m a t o r y i n f i l t r a t i o n , a n d f e a t u r e s o f p e r i t o n e a l o r p e l v i c m a s s , w e r e e v a l u a t e d a t C T b y t w o r e a d e r s i n e a c h c e n t e r . In our databases we found 18 patients, 15 women and three men. The clinical symptoms and signs in these patients included abdominal pain ( = 18), fever ( = 11), changed bowel habits (= 3) and palpable mass ( =  2). The duration of these symptoms and signs ranged from 5 days to 8 months. Laboratory results revealed leukocytosis in 16 patients (12.5–30.5/mm) and inflammatory markers (BSR and CRP) were elevated. Eleven female patients had a history of using IUDs for an average of 7 years (range, 2–14 years). Six women carried hormone-containing IUDs and five women had inert IUDs. At the time of the symptoms nine women had the IUD in place, while in the other two it was removed 2 and 4 months before. In the other seven cases (out of 18) there was a history of appendicitis (one male patient), diverticulitis (one female patient), inflammatory bowel disease (two male patients), and open or laparoscopic surgery (three female patients). No patients were immunocompromised. CT findings confirmed the infiltrative nature of the disease, which tended to invade across tissue planes and boundaries. In 11 patients an inflammatory mass involving the uterus and ovaries was revealed (). The main differential diagnoses proposed for all the patients were the following: tubo-ovarian abscess ( = 6), Crohn’s disease ( = 3), complicated diverticulitis ( = 2), colon cancer ( = 2), ovarian cancer ( = 2), prostatic cancer ( = 1), endometriosis ( = 1), and uterine cancer ( = 1). The most commonly involved sites in the gastrointestinal tract were the sigmoid colon in five patients, the appendix and the distal ileum in three cases (). Most patients showed concentric bowel wall-thickening (0.5–1.5 cm) while the length of the involved bowel was 5–15 cm (). The thickened bowel enhanced homogeneously in most patients and perirectal, pericolic, or perienteric infiltration was observed in all patients (). In 17 patients, a peritoneal or pelvic mass was seen adjacent to the involved bowel. It appeared to be predominantly cystic and heterogeneously enhanced. In only one case, the mass contained solid components showing marked contrast enhancement. The diameter of the masses was 1–5 cm and the margins were irregular and indistinct. Small bowel dilatation was noticed in one case. Infiltration into the abdominal wall was seen in three cases with a large abscess formation in one patient (). Lymphadenopathy was noted in five patients but it was minimal and involved the para-aortic, mesenteric, and pelvic lymph nodes. In two cases abscesses were found in the liver and in one case there was an abscess in the prostatic gland. Finally in one case there was thoracic dissemination. Actinomycosis was first described by Israel in 1879. It is a rare infectious disease caused by Actinomyces Israelii, a Gram-positive anaerobic saprophyte bacterium. The organisms are indigenous in the oral cavity, gastrointestinal tract, and genital track. The destruction of the mucosal barrier by trauma related to endoscopic procedures, operations, or chronic inflammatory disease, is recognized as predisposing factors (,,). The three main clinical forms of this disease are cervicofacial (50–65%), thoracic (15%), and abdominopelvic (20%). Pelvic actinomycosis has recently become more prevalent and it is associated almost exclusively with women who use IUDs (–) which is confirmed in our study as we found that 11 of our 15 female patients used an IUD. Clinical findings are variable, depending on the involved organ and the duration of the disease (,). Common symptoms and signs include abdominal pain with or without palpable mass, body weight loss, fever, vaginal discharge, constipation, or diarrhea. In laboratory analyses the dominating sign are leukocytosis, positive inflammation markers, and anemia as we confirmed in our results. High dose intravenous penicillin injection is the treatment of choice. Tetracycline, clindamycin, and erythromycin can alternatively be used for patients allergic to penicillin. Therefore, early diagnosis is important to minimize the morbidity of this disease and avoid unnecessary surgery (,). In our study 12 of 18 patients responded to conservative therapy and only six patients were treated surgically. Abdominopelvic actinomycosis may be the most indolent and latent of all the clinical forms of actinomycosis and diagnosis may be delayed for months after the inciting event. As we confirmed in our study actinomycosis may involve the abdominal wall, segments of the colon, uterus, ovaries, bladder, liver, gallbladder, and pancreas (,). The portions of the gastrointestinal tract commonly involved are sigmoid colon, rectum, cecum, appendix, distal ileum, and ascending colon (,). In our study the most commonly involved sites of the gastrointestinal tract were the sigmoid colon, appendix, and distal ileum. The common occurrence at the rectosigmoid colon contributes to the high frequency of pelvic involvement. One of the important radiologic characteristics of abdominopelvic actinomycosis is the aggressive nature of the infiltration. This disease’s infiltrative nature, and its tendency to invade normal anatomic barriers, was confirmed in most of our patients. Such a pattern may be the result of proteolytic enzymes produced by Actinomyces. This results in extensive inflammatory fat infiltration with abscesses formation in the abdominal wall (,). The organism in actinomycosis usually does not spread via lymphatic or hematogenous routes and regional lymphadenopathy is not a common finding. If lymphadenopathy is present, it is usually minimal as in five of our cases (,). It should be noted that despite the extensive inflammatory infiltration in the perirectal, pericolonic, or perienteric spaces, the disease process does not appear to spread into the whole peritoneal cavity and ascites is absent or minimal. The radiology findings in a barium study include mural invasion with structure formation, mass effect with tapered narrowing of the lumen, and thickened mucosal folds. Such radiology findings are not specific for abdominopelvic actinomycosis. The disadvantage of barium studies is that it does not examine the abdominal wall and in general it is no longer considered a mandatory study in these cases. On the other hand, the use of CT in abdominopelvic actinomycosis is essential for the diagnosis and for establishing the location and the extent of the disease. In our study and in other studies from the literature the most common findings are concentric bowel wall-thickening, enhancing homogeneously, and forming fistula (,,–). These radiologic findings are non-specific and are quite similar to those in Crohn’s disease, intestinal tuberculosis, or sometimes excavated malignant tumor. CT-guided fine needle aspiration may be not only diagnostic in equivocal cases but also therapeutic in cases of large abscesses. The most important CT feature for the correct diagnosis in our study was a large mass adjacent to the involved bowel. These masses appeared to be predominantly cystic or solid (pseudotumor) with contrast enhancement in the walls or the solid components of the masses (,,). Most cases of rectosigmoid colon involvement show predominantly cystic masses, whereas cases involving the transverse colon or appendix demonstrate predominantly or purely solid masses. These findings reflect the histologic features of actinomycosis: central suppurative necrosis surrounded by granulation tissue and intense fibrosis. In conclusion, actinomycosis is a rare disease that is not exclusively related to long-term use of IUDs. Actinomycosis should be included in the differential diagnosis when cross-sectional imaging studies show concentric bowel wall-thickening, intense contrast enhancement, regional pelvic or peritoneal masses, and extensive inflammatory fat infiltration with abscesses formation, especially in the absence of lymphadenopathy.

Birds present dazzling ecological diversity, with species differing in their climatic requirements, the habitats they use for feeding and breeding and the food resources they consume (). Ecological diversity originates when these characteristics of evolving lines of species repeatedly diverge during or between speciation events (). Beyond this fundamental fact, we do not understand for any large avifauna which characteristics have diverged little from ancestral conditions, that is, which display phylogenetic niche conservatism (PNC; ) and which have diversified substantially and account for most ecological diversity. The importance of PNC for explaining patterns of morphological and ecological characteristics is increasingly recognized (), to the extent that it could potentially serve as a null expectation for the diversification, or lack thereof, of avian ecological characteristics. Nonetheless, several studies of the ecological diversification of birds (), mammals () and amphibians () suggest that some ecological aspects of species could undergo diversification that does not conform to the expectations of PNC. It remains unclear whether the ecological characteristics of birds differ in aspects of historical patterns of diversification, and whether or not the ecological diversification that is represented in a large bird assemblage generally conforms to an expectation of PNC. An important debate has emerged over the last years on how PNC should be quantified or diagnosed relative to neutral expectations (; ). Perhaps the most commonly used neutral evolutionary model for this purpose is Brownian motion (BM), in which the ecological characteristics of lineages diverge at a constant rate across an isotropic phenotypic space and with which many metrics can be compared (). Two important properties have been widely used in comparative biology to formulate tests of deviations from the patterns expected under BM. First, phylogenetic signal expresses the case in which related species tend to share similar ecological characteristics; second, the degree of evolutionary gradualism expresses whether niche divergence increases gradually over time or whether niches diverge punctually, that is, independently of time. These two properties can be quantified with the Pagel metrics lambda and kappa (; ), which provide insightful descriptions of phylogenetic patterns of species niches. Nonetheless, recent simulation work suggests the limited usefulness of these metrics alone in diagnosing PNC because different rates of niche evolution (different degrees of niche conservatism) can yield a similar phylogenetic signal, while a low phylogenetic signal can also emerge under strong niche conservatism (; ). Another line of evidence can arise through examination of the temporal course of multivariate niche diversification, in a way that is independent of constraints imposed by evolutionary models, by conducting an analysis of multivariate disparity through time (DTT; ). This analysis compares the observed course of ecological diversification to expectations from a Brownian motion model in a multivariate niche space. Our objectives in this paper are to examine whether phylogenetic patterns of ecological niches of European birds support the hypothesis of PNC and to compare degrees of ecological niche diversification in the climatic requirements, habitat use and trophic habits of these birds. While patterns of species diversification in birds are becoming increasingly clear (), comprehensive ecological data to evaluate and compare ecological diversification broadly across many ecological characteristics are generally lacking for regional avifaunas. The European avifauna provides a notable exception because sufficient observational data exist to evaluate multiple components of Hutchinson's realized environmental niche (). In particular, the members of this avifauna can be distinguished using existing data to describe three niche components: the habitat niche (often called the Grinnellian niche; ), the climate niche () and the trophic niche (often called the Eltonian niche; ). Although other niche concepts could be considered (), we place our work in the conceptual framework recently proposed by , which allows depiction of all niche realms based on the same multivariate Hutchinsonian space. We focus on the habitat and trophic niches of birds because of their historical importance and prevalence in the ecological literature, their continued persistence in ongoing discussions regarding niche and their straightforward operationalization for use in quantitative analysis of observational data. We also include the climate niche because, in the case of birds at least, the distinction between physical habitat and climate plausibly reflects the larger geographical scale of the latter, in which variation in the former may be nested. Interestingly, these three components of species niches have never been investigated jointly in a comprehensive framework. The literature on avian ecology and evolution provides alternative empirical hypotheses regarding the relative contribution of the three niche components to the evolution of avian ecological diversity. Potentially, phylogenetic lability of any one of the three niche components, climate, habitat or trophic, could be primarily responsible for the ecological diversity of European birds, resulting in four testable hypothesis (Table ). Vicariant speciation and climate cycling have frequently exposed forming avian species to different climate regimes and may play a large role in diversification (), suggesting a principal role for climate niche lability in ecological diversification. However, feeding generalization has been linked to avian diversification rates (), which suggests a leading role for variation in the trophic niche in ecological diversification. Similarly, divergent habitat use for breeding and feeding between closely related avian species has long been recognized (), implying an important role for the diversification of habitat use in overall ecological diversification. A fourth hypothesis arises directly from the claim of the pervasiveness of PNC (). In this case, all three niche components should diversify in accordance with (or slower than; ) a BM model of trait diversification (Table ). In this paper, we compare phylogenetic patterns of trophic, habitat and climatic niches expressed during the breeding season in an assemblage of 405 birds that breed in Europe, west of the Urals. First, we test for phylogenetic signal and gradualism in each of the three niche components (, ). Second, we reconstruct past diversification of niches based on the computation of multivariate disparity over time (see Fig. S1 in Appendix S1 in the Supporting Information). We then compare the results for each niche component with the diversification that would be expected under BM. These two approaches allow us to finally examine the four alternative hypotheses for the primary origin of ecological diversity in this avian assemblage. We selected all European breeding bird species for which global breeding distribution data were available, a total of 405 species. We obtained data on the global breeding distribution of these species from a global database of avian species distributions (). These data represent the best available information on the global breeding distribution of this avian assemblage. We converted polygons to species presence and absence on a 10-arcmin resolution grid. We used the entire global distribution of each species in order to avoid missing portions of the species realized climate niche. We focused on ‘The Complete Birds of the Western Palearctic’ CD-ROM () as the data source for the habitat and trophic niches of these avian species because of its extensive treatment and literature review of habitat use, foraging behaviour, foraging location, food items and nesting characteristics. We addressed niche components only during the breeding season because distribution and ecological data regarding this season are much more robust than wintering data, especially for tropical migrants. We excluded from consideration all species feeding exclusively in the pelagic zone during the breeding season because data are lacking to meaningfully distinguish trophic differences among species. We decomposed patterns of diversification for all niche axes separately for each niche component (climatic, habitat and trophic). We applied multivariate analyses to continuous and discrete ecological characters to produce species-specific values on orthogonal niche axes. We selected ordination methods appropriate for analysis of categorical or continuous variables, or a mixture of the two, and which produced continuous numeric values that were compatible with the phylogenetic comparative methods we employed (detailed below). We included in the analysis a number of axes from each niche ordination sufficient to describe a uniform threshold percentage (90%) of the total ecological variation in each niche. Thus, the number of ordination axes varied among niches, but we needed to ensure that any differences among niches during phylogenetic comparative analysis were not due to consideration of arbitrary and variable percentages of ecological variation among niches. Use of a standard proportion of ecological variation for each set of variables enabled us to make consistent and comparable estimates of variability for the analysis of ecological divergence of species in terms of the three niche components. All ecological ordinations were conducted in R () with the package ade4 (). Together, our analyses consistently show that patterns of ecological niches of European birds diverge from those expected under PNC. Although patterns tend to vary among the three niche components, estimated values of phylogenetic signal and gradualism are systematically lower than expected under a Brownian model. Operative niche conservatism should have generated values indistinguishable from or greater than those expected under the BM assumption. In particular, we find a recurrent tendency towards punctualism for the three niches, suggesting that niche shifts have been largely decoupled from evolutionary time. Consistent with the low values of these indices, within-clade disparity for each of the niche components exceeds expectations under neutral (Brownian) niche divergence. Our results suggest that PNC, though potentially widespread (), is not an operative principle in the ecological diversification of European birds. This has an important implication for understanding the evolution and adaptation of species to changing environments. The gradualist view of evolution suggests that natural selection is often very weak, causing little change to accumulate over time, and that evolution ends up being uncoupled from ecology (for a recent review see ). Our results suggest that niche evolution deviates from gradualism in birds and can sometimes be rapid and independent of time and thus emphasize the potential for detectable relationships between macroevolutionary patterns of niche evolution and ongoing diversification of ecological traits (). Four potentially conflicting hypotheses attribute the primary origin of avian ecological diversity to: greater divergence of the climatic niche arising from global climate cycling and geographical barriers (; ); greater divergence of habitat use (; ; ); greater divergence of trophic preferences (; ); or a level of divergence consistent with neutral evolutionary drift (Table ; ; ). Our results suggest that neutral (BM) evolution alone is not adequate to explain the variety of among-niche patterns of disparity we observe. However, each of the other hypotheses has some support in the literature and differences in degree of diversification of the three niche components could exist. Ecological diversification of the climatic niche, as shown in the DTT analysis, is somewhat greater than that of the other two niches. This suggests the hypothesis that the primary evolutionary driver of ecological diversity held in these birds arises from greater divergence in species climatic requirements, which have sometimes occurred quite abruptly (hence the evidence for strong punctualism). One potential mechanism for this involves global climate cycling, which has probably driven species repeatedly in and out of geographically isolated refugia in the Northern Hemisphere and promoted vicariant speciation (). Such processes could have operated since the Pliocene () and could be common compared with mechanisms of sympatric speciation (). Nonetheless, ecological diversity in European birds is the product of ecological diversification in many different clades, many members of which are not present in Europe. Knowledge of these species would be necessary for an unbiased comparison of rates of niche evolution. Relatively rapid diversification of the climatic niche, if so confirmed, would contrast with rates of niche evolution in tropical avian assemblages, where the effects of climate cycling on niche diversification could be less important (). In contrast, our results suggest that evolutionary divergence of habitat preferences could contribute to ecological diversity, secondarily to diversification of the climatic niche. This finding contrasts with previous findings that divergent preferences for feeding and breeding habitat exist among many closely related avian species (), and may be the dominant contributor to ecological diversification in certain avian groups (; ). The analyses suggest that there is greater phylogenetic signal in the trophic characteristics of the species in this assemblage than in the other two niche components, and that trophic niches tend to diversify in a somewhat less punctuated (or more gradual) manner than other niches. The limited diversification of trophic habits in comparison with the other two niche components could be due to relatively slow evolutionarily rates. For example, constraints might occur from slow or infrequent divergence of specialized morphology that is associated with resource capture and feeding, and which may have diversified relatively early in the history of different lineages. Our results are partially consistent with earlier studies that report evidence for phylogenetic signal in the trophic niche of the assemblage of European birds. use Mantel tests on an ecological dissimilarity matrix and a matrix representing phylogenetic distance among 151 species to find phylogenetic signal in both diet and breeding habitat. Similarly, examines taxonomic levels to find that significant phylogenetic conservatism exists in the dietary niche breadth of 142 European bird species at the level of family and genus. Nevertheless, one potentially testable hypothesis is that low levels of disparity of trophic preferences (compared with the other niches) arise because macroecological sampling preferentially ‘selects’ lineages having particular trophic characteristics. Such sampling, and not variation in evolutionary rates, could be the cause of this pattern. Several processes other than simple heterogeneity in evolutionary rates and degree of niche diversification may influence the patterns we report. For instance, we focus on disparity as arising from variation in niche position and ignore possible effects of changes in niche breadth. The development of methods to address niche breadth in a comparative analysis (e.g. ) may shed additional light on differences among niches in phylogenetic patterns and underlying processes. Observed differences in DTT among niches could also involve heterogeneity in rates of niche evolution within the complete clade from which the European avifauna is drawn (). Such non-stationarity in rates of niche diversification, in combination with incomplete and biased macroecological sampling from within the encompassing clade, might create the appearance of heterogeneity among niche components in their DTT values. However, this heterogeneity might not be a good representation of any rate heterogeneity in the full tree of birds. Another possibility is that effects of macroecological sampling bias could arise when the birds of the European assemblage are not a random sample from the inclusive clade of birds of which they are members. As macroecological sampling would affect which species are found in Europe, the climatic niche characteristics of the assemblage might be notably influenced. Macroecological sampling of the lineages in this assemblage could involve additional evolutionary processes that are not identified as niche evolution per se. These might include range evolution during vicariance events and the development of migratory behaviour (; ; ), which may eventually precipitate evolutionary shifts in resource use or habitat selection. One effect of the evolution of migratory behaviour on niche characteristics is that the environmental niche of migratory birds during the breeding season may be less conserved than in the winter range (). This suggests that species niche characteristics during residency in the breeding range may become labile during the evolution of migratory behaviour. Change in migratory patterns could also contribute to geographical variation within species in their niche characteristics (), as in the case of partial migration (). Such potential effects of changes in migration strategies and geographical range on niche evolution have been virtually unexplored from a macroevolutionary perspective. e r t h e h i s t o r y a n d d i v e r s i t y r e p r e s e n t e d b y t h e E u r o p e a n a v i f a u n a , w e f i n d n o s u p p o r t f o r p h y l o g e n e t i c c o n s e r v a t i s m i n a n y o f t h e t h r e e m u l t i v a r i a t e n i c h e c o m p o n e n t s w e i n v e s t i g a t e d . W h i l e P N C h a s b e e n p r o p o s e d a s a n e v o l u t i o n a r y p r i n c i p l e , t h i s p e r s p e c t i v e m a y a r i s e f r o m f o c u s o n a s m a l l n u m b e r o f e c o l o g i c a l c h a r a c t e r i s t i c s o r a b i a s e d s a m p l e t h e r e o f , w h i l e o u r e c o l o g i c a l d e s c r i p t i o n o f s p e c i e s n i c h e s i s l i k e l y t o c a p t u r e m o s t a s p e c t s o f c l i m a t i c r e q u i r e m e n t s , h a b i t a t s e l e c t i o n a n d t r o p h i c h a b i t s . A l t e r n a t i v e l y , c o n s e r v a t i s m o f s o m e a v i a n n i c h e c o m p o n e n t s m a y b e g r e a t e r i n o t h e r r e g i o n s o f t h e w o r l d , s o m e o f w h i c h ( e . g . t r o p i c a l a r e a s ) a r e m u c h m o r e s p e c i e s r i c h t h a n E u r o p e o r h a v e b e e n d i v e r s i f y i n g u n d e r s t r o n g e r e n v i r o n m e n t a l s t a b i l i t y . S e v e r a l p r o c e s s e s m a y c o n t r i b u t e t o t h e p a t t e r n s w e o b s e r v e h e r e , i n c l u d i n g t h e p r o c e s s e s o f m a c r o e c o l o g i c a l s a m p l i n g a n d r e l a t i v e r a t e s o f e c o l o g i c a l d i v e r s i f i c a t i o n o f n i c h e c o m p o n e n t s . H y p o t h e s e s r e g a r d i n g t h e i m p o r t a n c e o f v a r i a t i o n i n u n d e r l y i n g e v o l u t i o n a r y r a t e s f o r v a r i a t i o n i n e c o l o g i c a l d i s p a r i t y , p h y l o g e n e t i c s i g n a l a n d g r a d u a l i s m m a y b e t e s t a b l e w i t h g r e a t e r e c o l o g i c a l a n d p h y l o g e n e t i c i n f o r m a t i o n . F i n a l l y , t h e a p p l i c a t i o n o f a d d i t i o n a l m e t h o d s o f c o m p a r a t i v e a n a l y s i s t o l a r g e a s s e m b l a g e s i s l i k e l y t o r e v e a l n e w i n s i g h t s t h a t c a n d r i v e h y p o t h e s i s g e n e r a t i o n a n d t e s t i n g .

Hypertrophic cardiomyopathy is a pathologic hypertrophy of the heart due to an increase in the size of myocytes in various heart diseases including long-term hypertension, myocardial infarction, chronic pressure overload, valvular defects and endocrine disorders (; ; ; ). Myocardial hypertrophy is an adaptive response of the heart to increased workload. However, increased myocyte size, increased left ventricular (LV) mass and decreased fractional shortening (FS) are risk factors of cardiac morbidity and mortality in the general population (; ; ). Previous studies have demonstrated that dyslipidemia, hypercholesterolemia and cardiac lipotoxicity are associated with cardiac hypertrophy (; ; ; ; ; ; ; ; ). Recently, we have observed that feeding a high fat and cholesterol diet to apoE mice results in marked increase in the level of GSL, e.g. glucosylceramide (GlcCer) and LacCer in heart tissue accompanied by an increase in the activity of glycosphingolipid (GSL) glycosyltransferases (GTs) () (submitted for publication). The association of marked atherosclerosis and cardiac hypertrophy with these biochemical changes has been confirmed by physiologic studies (LV mass, FS) and up-regulation of genes for brain natriuretic peptide (BNP), atrial natriuretic peptide (ANP) and alpha skeletal actin—all are well-known markers of cardiac hypertrophy (; ; ; ; ; ). Treatment of mice with -threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol (-PDMP), an inhibitor of GSL synthesis, not only reversed atherosclerosis but also markedly reduced cardiac hypertrophy () (submitted for publication). Regression in LV mass is known to be accompanied by reduced cardiovascular complications during hypertrophy (; ; ). Hence, decreasing GSL load in the myocardium seemed to reverse LV mass which is widely accepted as a desirable treatment goal in cardiovascular diseases. However, these studies conducted in experimental animal models could not establish clearly whether one or more GSLs take part in cardiac hypertrophy. Herein, using cultured cardiomyocytes, we demonstrate that LacCer specifically induces cardiac hypertrophy by way of generating reactive oxygen species (ROS) to transduce a signal transduction pathway leading to this phenotype. The incorporation of [H]-leucine into cell protein has been one method used widely to determine the rate of protein synthesis. Among all different glycolipids, LacCer specifically stimulated protein synthesis (2-fold) to a similar extent as phenylephrine (PE) in these cells (Figure ). In contrast, the other classes of GSL. e.g. sulfatides, complex gangliosides and other neutral GSLs, failed to increase protein synthesis in these cardiomyocytes, respectively. A time and LacCer concentration (50–100 μM)-dependent increase in protein synthesis in H9c2 myotubes and neonatal rat ventricular myocytes (NRVM) is shown in Figure A and B. H9c2 cells were induced to undergo differentiation (2% horse serum) to form myotubes. LacCer-induced protein synthesis was significantly increased (2- to 3-fold) at and above the concentration of 50 μM. The maximum increase in protein synthesis was also observed following 48 h of incubation with 100 μM of LacCer in H9c2 myotubes and in NRVM, respectively. Thus, LacCer alone increases protein synthesis in cardiomyocytes at a concentration similar to PE. [H]-Thymidine incorporation was measured to assess the proliferative effect of LacCer in these cells (Figure C). The cells were maintained in complete medium (10% fetal bovine serum (FBS)). At a lower concentration of LacCer (10 μM) and after 48 h of incubation, a ∼5-fold increase in [H]-thymidine incorporation into DNA was observed. Further increasing the concentration of LacCer did not significantly alter this observation. ANP and BNP genes have been reported to serve as bonafide biomarkers in the development of cardiac hypertrophy (; ). So, we examined the temporal profile of ANP and BNP mRNA expression in cardiomyocytes incubated with and without LacCer (Figure D and E). LacCer dose dependently and significantly (2- to 3-fold) up-regulated the levels of ANP and BNP mRNAs starting at a concentration of 50 μM (Figure D and E) and reaching a maximum increase at 100 μM. One of the hallmarks of cardiac hypertrophy is an increase in myocyte size. Therefore, we examined whether LacCer can induce an increase in cell size associated with cardiac hypertrophy (Figure ). NRVMs and H9c2 cells were treated with LacCer at different concentrations (10, 20, 50 and 100 μM) for 24 and 48 h. PE served as a positive control. The cells were fixed and stained with anti-α-actinin antibody to distinguish myocytes from other cell types in the cultures. Immunofluorescent and bright field images were taken. Treatment with 100 μM of LacCer for 48 h increased the cell size ∼40 ± 1%. The levels of intracellular ROS in H9c2 cells were measured using a fluorescent probe, 2′,7′-dichlorofluorescin diacetate (DCFH2-DA) () (Figure A). The intensity of DCF fluorescence was observed to increase slightly in normal cells exposed to DMSO. In comparison, a significant increase was observed in the intensity of DCF fluorescence in LacCer treated H9c2 cells. LacCer-induced increase in DCF fluorescence was time and concentration dependent. LacCer (10 μM) was sufficient to increase ROS levels within 10 min of treatment in H9c2 cells. We also studied the effect of antioxidant -acetylcysteine (NAC), a scavenger of ROS and diphenylene iodonium (DPI), an inhibitor of NAD(P)H oxidase on LacCer-induced ROS generation. The culture medium of H9c2 cells was replaced by a medium containing DPI (200 μM) and NAC (15 mM). Cells were then treated with DCFH2-DA (10 μM) for 15 min prior to the addition of LacCer. NAC and DPI prevented LacCer-induced increase in the intensity of DCF fluorescence (Figure B). As shown in Figure C, a 20 ± 5% decrease in cell viability is found to be associated with a concentration of 100 μM LacCer. Similarly, cells treated with 100 μM PE also showed ∼11 ± 3% decrease in cell viability. Thus, hypertrophy and cell viability both are affected by these diverse compounds. To determine whether ROS generated upon treatment with LacCer is critical in cardiac hypertrophy, we examined the effects of NAC and DPI on hypertrophy in H9c2 cells. Both NAC and DPI mitigated the increase in ANP and BNP gene expression induced by LacCer (Figure ). Thus, LacCer-mediated increase in ROS is critical for cardiac hypertrophy. The treatment of cardiomyocytes with LacCer significantly increased the c-jun mRNA level as detected by semi-quantitative real-time-polymerase chain reaction (RT-PCR) (Figure A) and real-time quantitative PCR (data not shown). To determine whether an increase in the expression of the transcription factors c-jun was of specific nature, we also examined the expression of several other transcription factors using real-time quantitative PCR. No changes in NFAT3, NFkB, c-myc, MEF2C and MEF2D mRNA levels were observed except for the upregulation for c-fos mRNA level (Figure B). PKC and ERK1/2 have been reported to be involved in the regulation of c-fos and c-jun gene expression in neonatal cardiomyocytes (; ; ; ). We have previously shown that in human arterial smooth muscle cells, PKC and ERK1/2 are involved in the LacCer-mediated increases in cell proliferation (; ). The possible role of PKC in LacCer-mediated hypertrophy was examined by pretreating cardiomyocytes for 8 h with bisindolylmaleimide (Bis) (100 and 200 nM), an inhibitor of PKC activity, prior to the addition of LacCer (100 μM). It can be seen from Figure C and D that inhibition of PKC attenuated the LacCer induced increases in ANP/BNP mRNA levels. Figure A–D shows that inhibition of PKC also attenuated the LacCer-induced increase in c-jun and c-fos mRNA levels. Pretreatment of cardiomyocytes with PD98059 (20 nM), an ERK1/2 signaling inhibitor, for 8 h prior to the addition of LacCer prevented the increases in ANP, BNP mRNA levels (Figure A and B) and c-jun, c-fos mRNA levels (Figure E–H) induced by LacCer. Bis or PD98059 alone did not affect c-fos and c-jun gene expression. The culture medium of H9c2 cells was replaced by a medium containing DPI (200 μM) and NAC (15 mM). Eight hours after the addition of DPI and NAC, cells were treated with and without LacCer for 48 h. c-fos and c-jun mRNA expressions were analyzed by real-time quantitative PCR (Figure A and B) and RT-PCR analysis (Figure C and D). NAC and DPI treatments have significantly downregulated the c-fos and c-jun expression induced by LacCer. The following major findings emerged from the present study. (i) Only LacCer, among a host of other GSLs and sphingolipids decorating the mammalian cell membrane, exerted a time- and concentration-dependent increase in hypertrophy in cardiomyocytes from neonatal rat heart and rat H9c2 cells. (ii) Mechanistic studies revealed that LacCer induces cardiac hypertrophy by an “oxygen-sensitive” signaling pathway by way of activating NAD(P)H oxidase to generate superoxides, protein kinase-C activation, p44 mitogen activated protein kinase (MAPK) phosphorylation and nuclear factor c-fos and c-jun expression (Figure ). Cardiac hypertrophy in vivo involves the enlargement of the myocardial cells due to an overload of blood volume and increased blood pressure. In contrast, cardiac hypertrophy in vitro is induced by the use of agonists such as PE which binds to its cognate receptors and transduces downstream components to eventually induce hypertrophy. In this study, we used PE as a positive control and demonstrated that at a similar concentration (100 μM), LacCer independently could serve as a bonafide agent to induce cardiac hypertrophy in H9c2 cells and freshly cultured primary rat cardiomyocytes. At the cellular level, hypertrophy is characterized by an increase in the size of cells. This increase in cell size is mainly accompanied by an increase in protein synthesis. During hypertrophy, cells grow in size without further cell division. Proliferation on the other hand is the division of cells accompanied by DNA synthesis and nuclear division. Treatment with 10 μM LacCer was found to be most effective in inducing cell proliferation, which is similar to the results reported in arterial smooth muscle cells (). However, when the cells were stimulated with 50–100 μM of LacCer, there was a significant increase in cell volume and cell size, which is a hallmark of hypertrophy. Increase in DNA synthesis was much smaller relative to increase in RNA and protein synthesis at a LacCer concentration of 50–100 μM. LacCer concentration-dependent phenotypic effect on hypertrophy was also reproduced in NRVMs, which are terminally differentiated cells. Our studies employed multiple criteria to assess hypertrophy in these cardiomyocytes, e.g. increased cell volume, increased protein synthesis using [H]-leucine as a precursor and the measurement of mRNA levels of ANP- and BNP-established biomarkers of cardiac hypertrophy. Next, we investigated the effects of a pool of GSLs and sphingolipids on cardiac hypertrophy and observed that they did not significantly affect this phenotype. Therefore, we conclude that an intact molecule of LacCer is required to induce cardiac hypertrophy. Importantly, the catabolic or anabolic products of LacCer failed to induce this phenotype. These studies suggest that LacCer specifically induced cardiac hypertrophy. The potential role for free oxygen radicals in cardiac hypertrophy has been elucidated in many earlier studies (; ; ). Therefore, we examined the effects of LacCer on superoxide production and its mitigation on cardiac hypertrophy. As shown in Figure , LacCer induced the generation of superoxides in a time- and concentration-dependent manner (Figure A) and this was mitigated by the use of -acetylcysteine (a scavenger of free oxygen radicals) and diphenylamine iodonium (DPI), an inhibitor of NAD(P)H oxidase (Figure B) (; ; ). Use of these inhibitors also mitigated LacCer-induced cardiac hypertrophy biomarkers mRNA levels, e.g. ANP and BNP (Figure ). This observation suggests that, by activating NAD(P)H oxidase, LacCer generates superoxide radicals which in turn activate a downstream signaling cascade leading to cardiac hypertrophy. At lower concentrations (10–20 μM), LacCer stimulated DNA synthesis to facilitate proliferation by producing small quantities of superoxide. 10 μM LacCer concentration was found to be most effective in inducing proliferation, which is similar to the results reported in arterial smooth muscle cells (). At higher concentrations (50–100 μM), LacCer is generating a large amount of superoxides that could affect cell viability and stimulate a pathways leading to hypertrophy. We have previously shown in cultured normal human fibroblasts () that [H]-LacCer associated with LDL bound to the cell membrane at 4°C. However, when the cells were warmed to 37°C, [H]-LacCer rapidly internalized, degraded to GlcCer and converted to GbOse3Cer and Gbose4Cer within 30 min. In contrast, in human fibroblasts lacking functional LDL receptors, LacCer did not metabolize rapidly, rather it entered the cells via an LDL receptor-independent pathway. Thus exogenously added LacCer should first incorporate to the plasma membrane, activate NAD(P)H oxidase to generate superoxides and then affect cardiac hypertrophy. The immediate early genes activated during hypertrophic stimulus include c-jun, c-fos and c-myc. Previous studies have shown that PE induces immediate early genes such as c-fos and c-jun leading to cardiac hypertrophy (; ). In our study, we found that LacCer-induced hypertrophy also involved the upregulation of both c-fos and c-jun genes (Figure A and B). Our study also demonstrates that the activation of these immediate early genes involves oxidative stress (Figure ). Furthermore, we investigated the effects of LacCer on PKC activation and cardiac hypertrophy. The involvement of PKC in cardiac hypertrophy has been reported previously (; ; ). For example, PE is also known to induce hypertrophy via PKC activation (). Using bisindolylmaleimide, an inhibitor of PKC, we observed a marked inhibition of LacCer-induced ANP and BNP mRNA levels in cardiomyocytes (Figure C and D), suggesting that PKC plays a central role in LacCer-induced hypertrophy (Figure ). Previous studies have also placed p44 MAPK activation as a central component in agonist-induced cardiac hypertrophy (; ; ; ; ; ; ). For example, PE and angiotensin II is known to induce p44 MAPK activation (; ; ; ; ; ; ). Also transforming growth factor--β1 induces hypertrophy and fibrosis via activation of p44 MAPK (). We therefore examined the effects of LacCer on p44 MAPK and cardiac hypertrophy. We observed not only that LacCer induced the rapid phosphorylation of p44 MAPK but also that this activation process was required to induce cardiac hypertrophy, as the use of p44 MAPK inhibitor mitigated LacCer-induced upregulation of ANP and BNP mRNA expression (Figure A and B). Also, LacCer-induced upregulation of c-jun and c-fos mRNA involves PKC and p44 MAP kinases activation (Figure ). Recently, LacCer has been implicated to directly bind to phospholipase A-2 to induce arachidonic acid production (; ). Therefore, we examined the effects of a c-PLA2 inhibitor -bromophenacyl bromide (20 μM) on LacCer-mediated cardiac hypertrophy. Since the results were negative (data not shown), it implies that LacCer does not recruit c-PLA2 to induce cardiac hypertrophy. Our findings are in agreement with previous work, showing that c-PLA2 has no significant/direct role in cardiac hypertrophy (). Hypertrophy induced by fat diet intake is fast becoming one of the primary cause of myocardial infarction, morbidity, stroke and contributes to ∼50% mortality in western countries. It is a major clinical concern in cardiovascular medicine. Increased levels of fatty acids from fat diets can impact the heart harmfully due to the formation of noxious derivatives of glucose and lipid metabolism. A close association between GSL level and cardiac hypertrophy in vivo in apoE mice fed a western diet was suggested by us recently () (submitted for publication). We found that the activity of several GSL glycosyl transferases and the level of GSLs, particularly LacCer, were markedly increased in mice fed a western diet alone when compared with apoE mice fed regular mice chow alone. Cardiac hypertrophy (LV mass measurement) in this transgenic mouse model of hyperlipidemia was prevented and interfered by the use of -threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol (-PDMP), an inhibitor of GlcCer synthase and LacCer synthase () (submitted for publication). However, these in vivo studies did not elaborate whether one or more GSLs were implicated in cardiac hypertrophy. The present study using cultured cardiomyocytes suggests that LacCer alone can induce hypertrophy and therefore exposes both LacCer and LacCer synthase as novel drug targets to mitigate this phenotype. Dulbecco's modified Eagle's medium (DMEM), trypsin-ethylenediaminetetraacetic acid, phosphate-buffered saline (PBS), -glutamine, penicillin/streptomycin, fluorescent probe DCFH2-DA, TRIzol reagent, cDNA synthesis kit, SYBR Green PCR Master Mix, fluorescein isothiocyanate (FITC)-conjugated antibody and DAPI nuclear stain were purchased from Life Technologies, Grand Island, NY. FBS, normal horse serum (HRS), DPI, NAC, PE, trichloroacetic acid, NaOH and monoclonal anti-α-actinin (Sarcomeric) (A7732) were purchased from SIGMA Chemical Co., St. Louis, MO. Neonatal rat NeoMyts Kit was obtained from Cellutron Life Technology, Baltimore, MD. 3-(4,5-Dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide (MTT) assay kit was obtained from Promega, Madison, WI. [H]-Leucine and [H]-thymidine were obtained from American Radiolabeled Chemicals. Scintillation cocktail was obtained from RPI, Mount Prospect, IL. Anti-phospho p44/42 MAP kinase antibody was obtained from Cell Signaling, Danvers, MA. Lactosylceramide (LacCer) (palmitoyl) and other GSL stocks were purchased from Matreya LLC, Pleasant Gap, PA. All other chemicals used in this study were of the highest grade available from commercial suppliers. H9c2 rat cardiomyoblasts were obtained from the American Type Culture Collection (Rockville, MD) and were grown in high-glucose DMEM medium supplemented with 10% FBS, 100 units/mL penicillin and 100 μg/mL streptomycin and maintained at 37°C in a humidified-atmosphere incubator (Thermo Fisher Scientific, Pittsburgh, PA) with 5% CO. The cells were seeded onto six-well plates at a density of 3 × 10 cells/well containing 3 mL culture medium or 10 per well in 24-well plates containing 1 mL medium. To induce differentiation of H9c2 myoblasts into myotubes, growth medium was replaced with differentiation medium (DMEM containing 2% horse serum) and allowed to grow for 48 h before any treatment. Ventricles were dissected from 1-day-old Sprague-Dawley rats. NRVMs were isolated using the Cellutron Neomyocytes isolation system (Cellutron Life Technology, Baltimore, MD) following the manufacturer's instructions (; ). Cells were pre-plated for 2 h to separate adherent fibroblasts from nonadherent cardiomyocytes. Myocytes were resuspended in DMEM containing 10% FBS, 2 mM -glutamine and penicillin/streptomycin (P/S) and plated onto either gelatin-coated cover glasses or culture dishes. Myocytes were plated at 700 cells/mm for cell size assays and at 1100 cells/mm for other experiments. [H]-Leucine was measured essentially by the method of . The embryonic rat-heart derived H9c2 cells American type culture collection were maintained in a growth medium comprising DMEM supplemented with 10% FBS. H9c2 cells were plated at a density of 5000 cells/cm and allowed to proliferate in the growth medium. When cells had reached near confluence, the growth medium was replaced with a differentiation medium (DMEM containing 2% horse serum) for 48 h to induce differentiation of H9c2 myoblasts into myotubes (, ). Cells were then stimulated with a single dose of 100 μM PE. Cells were also stimulated with different glycolipids (100 µM) or different doses of LacCer (10, 20, 50, 80 and 100 μM) for 24, 36, 48 and 72 h and co-incubated with [H]-leucine (5 Ci or 142 Ci/mmol) at indicated time points. At the end of the incubation period, cells were washed twice in PBS and proteins were subsequently precipitated with ice-cold 10% trichloroacetic acid. After dissolving the precipitates in 0.5 M NaOH, 5 mL scintillation cocktail was added and radioactivity was measured by liquid scintillation spectroscopy (; ). [H]-Leucine incorporation experiments were repeated five times with triplicate measurements for each experiment. H9c2 cells were plated (10 per well) in 24-well plates, and were made quiescent for 24 h in DMEM without serum. Quiescent cells were then stimulated for 48 h with a single dose of 100 μM PE. Cells were also stimulated with different glycolipids (100 µM) or different doses of LacCer (10, 20, 50, 80 and 100 μM) for 24, 36, 48 and 72 h and co-incubated with [H]-thymidine (5 mCi/mL media) at indicated time points. Cells were washed with phosphate-buffered saline and dissolved in 0.5 M NaOH. The incorporation of [H]-thymidine was measured as described above. [H]-Thymidine incorporation experiments were repeated five times with triplicate measurements for each experiment. Cell viability was determined using an MTT assay kit (; ). Briefly H9c2 cells were plated (10 per well) in 96-well plates, and were made quiescent for 24 h in DMEM without serum. Quiescent cells were incubated with LacCer and MTT assay reagent. The reaction was stopped at different time points and absorbance was recorded at 540 and 690 nm with a microplate reader (Thermo Scientific Multiskan Spectrum). Viability was determined when compared with control cells (normal H9c2 cells incubated with DMSO). Briefly the cells were fixed with 1% paraformaldehyde and permeabilized with 0.1% Triton X-100. For staining of α-actinin, cells were incubated with a mouse monoclonal antibody to α-actinin (1:200 in 1% BSA) for 2 h. Cells were washed with 1% BSA in PBS and were treated with FITC-conjugated goat anti-mouse IgG (green fluorescence) secondary antibody. Images were captured using an Olympus Invert scope microscope; Olympus, Tokyo, Japan. Photographic images were taken from five random fields (). NRVMs and H9c2 cells were treated with PE (100 μM) and LacCer (100 μM) for 24 or 48 h. Cell area was quantified from manually outlined cells in digitized microscopic images (recorded by an Olympus Invert scope microscope; Olympus, Tokyo, Japan) of randomly chosen cell fields using the Image-Pro Plus 6.0 software (Media Cybernetics, Silver Spring, MD). A minimum of 50 cells were measured for each independent experiment, repeated ≥3 times (). Total RNA was isolated from H9c2 cells and NRVMs using TRIzol reagent (Life Technologies, Grand Island, NY). Two micrograms of RNA was reverse-transcribed with SuperScript II (Life Technologies, NY) using random primers. Gene-specific primers were designed (Table .) and used for amplifying ANP, BNP, c-fos and c-jun mRNAs. Total RNA was isolated from H9c2 cells and NRVMs, using TRIzol reagent. Two micrograms of RNA was reverse-transcribed with SuperScript II using random primers. Real-time PCR was performed using SYBR Green PCR Master Mix (Life Technologies) in an Applied Bio system's Step one RT-PCR system with the following thermal cycling conditions: 10 min at 95°C, followed by 40 cycles at 95°C for 15 s and at 60°C for 1 min for denaturation, annealing and elongation. Relative mRNA levels were calculated by the method of 2 (). Data were normalized to GAPDH mRNA levels. To determine the specificity of amplification, melting curve analysis was applied to all final PCR products. All samples were performed in triplicate. The expression suite software (Applied Bio-systems) was used to analyze the data. Intracellular ROS levels were assessed using DCFH2-DA (). Cells were treated with DCFH2-DA (10 μM) for 15 min prior to the addition of LacCer. After incubation with LacCer for different time points (5, 10, 15 and 30 min, 1, 2, 4, 6, 8, 12 and 24 h) the cells were washed with PBS and analyzed using a fluorescent microplate reader (Beckman Coulter, Fullerton, CA). The relative intensity of DCF fluorescence was determined at a wavelength of 535 nm when compared with control cells (normal H9c2 cells incubated with DMSO). Data represent the mean intensities of DCF fluorescence, SD relative to the DCF fluorescence intensity of the control of four independent experiments. The culture medium of H9c2 cells was replaced by one containing DPI (200 μM) and NAC (15 mM). Cells were then treated with DCFH2-DA (10 μM) for 15 min prior to the addition of LacCer for different time points (5, 15 and 30 min, 1, 8, 12 and 24 h). Cells were washed with PBS and analyzed using a fluorescent microplate reader at a wavelength of 535 nm (; ). Eight hours after the addition of DPI and NAC, cells were treated with and without LacCer for 24 h. ANP and BNP mRNA expression levels were measured by real-time quantitative PCR method. Data represent the mean ± SD of three independent experiments. Proteins were separated on gradient sodium dodecyl sulfate–polyacrylamide gel electrophoresis (7.5–15% acrylamide) and electrophoretically transferred to polyvinylidene difluoride membrane. Membranes were blocked using 5% skim milk in PBS plus 0.05% v/v Tween-20 before incubation with anti-phospho p44/42 MAP kinase antibody (9101; Cell Signaling) and anti-GAPDH antibody (sc-47724; Santa Cruz Biotech, Dallas, TX). HRP-conjugated secondary antibodies and ECL plus kit (Amersham Life Sciences, Piscataway, NJ) were used to detect proteins of interest. All values are expressed as mean ± SEM. Comparison between groups was performed by one-way analysis of variance with Bonferroni's multiple comparison tests. Comparisons between the two groups were performed using nonpaired two-tailed Student's -test. A value of < 0.05 was considered significant. GraphPad Prism and MS-Excel statistical software were used for aforementioned statistical analysis. award-id #text N o n e d e c l a r e d . #text

A pair of papers from the same group (; ) has identified a heteromeric calcium-permeable TRP channel in primary cilia and defined these organelles as a specialized calcium signaling compartment. Solitary nonmotile structures known as primary cilia project from most vertebrate cells to act as sensory organelles; these specialized structures have been implicated in hedgehog signaling pathways, and ciliary defects are associated with various human disorders, including polycystic kidney disease. Patch-clamp analysis of primary cilia visualized with targeted, genetically encoded fluorophores revealed an outwardly rectifying noninactivating current (), with current density substantially greater than that in the cell body; permeability to calcium of the ciliary channel was estimated as six times that of sodium or potassium (). was activated by extracellular uridine or adenosine phosphates (presumably acting through a purinergic GPCR), and by cell-permeable calmodulin antagonists, but was inhibited by Gd and ruthenium red. siRNA-mediated knockdown of the polycystin proteins PKD1L1 or PKD2L1 decreased , and the much-diminished apparent in mice lacking PKD2L1 was linear and insensitive to a calmodulin antagonist. Heterologously expressed PKD1L1 and PKD2L1, which could be coimmunoprecipitated from HEK293 cells, yielded whole-cell currents with a single-channel conductance similar to that in primary cilia; moreover, like , these currents were activated by calmodulin antagonists and blocked by Gd and ruthenium red. In the second paper, used a ratiometric calcium sensor targeted to cilia (SMO-mCherry-GCaMP3) to measure ciliary [Ca] simultaneously with that in the cytoplasm (monitored with Fluo-4). Rupturing the membrane at the ciliary tip elicited a rapid increase in ciliary [Ca] that traveled down the cilium, with little effect on cytoplasmic [Ca], even at the cilia–cell body junction. Ca moved readily from the cytoplasm to the cilium, indicating that the lack of effect of an increase in ciliary [Ca] on cytoplasmic [Ca] did not result from diffusion barriers at the ciliary base, but rather from the difference in volume between the two compartments (a ratio of ∼1:30,000). Resting ciliary [Ca] was substantially higher than resting cytoplasmic [Ca] and ciliary membrane potential was substantially more positive than that in the cell body, indicating that cilia represent a functionally distinct ionic compartment. Mice lacking PKD2L1 showed defects in hedgehog signaling as well as intestinal malrotation (a phenotype consistent with defects in the hedgehog pathway). The authors thus conclude that the high [Ca] in cilia is maintained through their small volume and density of Ca influx pathways, with PKD1L1-PKD2L1 acting as a ciliary Ca channel to modulate ciliary [Ca] and thereby hedgehog signaling. #text xref sup sub #text

Ca signals initiate diverse responses in a cell, and Ca can regulate its own intracellular concentration. Ca influx through voltage-dependent Ca channels (VDCCs) can be modulated by cell-membrane repolarization through activation of Ca-activated K channels such as the small-conductance Ca-activated (SK2) K channel (). A major physiological role of SK2 channels is to restore an excited cell back to its resting state in response to increases in local intracellular Ca concentration. The SK2 protein binds Ca through a ubiquitous cell regulator, calmodulin (CaM; ). CaM, which can bind Ca at four sites, participates in essential Ca sensing roles for many diverse proteins (). The energetics of SK activity are controlled by the dynamic coupling between Ca binding to CaM, CaM binding to SK, and the open probability of SK. It has been shown by x-ray crystallography that CaM can bind to a recombinant SK2 fragment with a somewhat extended conformation, having a C-shaped configuration (; ). Several structures have been obtained with CaM bound to various targets (; ), yet the conformation of CaM bound to SK is unique. In the crystal, antiparallel CaMs and antiparallel SK peptides form a dimeric complex, with a stoichiometry of 2 SK peptide (SKp) to 2 CaM, which we will denote as 2SKp/2CaM. Ca ions are only bound to the N-lobe binding sites, and the C-lobe sites are unoccupied even at high Ca concentrations. Recently, the 2SKp/2CaM stoichiometry was supported by new crystal structures of the SKp–CaM complex. One structure contains an alternatively spliced SK variant (), and other structures were solved with either phenyl urea or 1-ethylbenzimidazolinone (1-EBIO) as CaM binding agonists at the protein–protein interface (). It has been argued that this 2SKp/2CaM configuration represents the conformation that the SK C termini assume in the full-length SK channel when activated (). The structure inspired a mechanistic hypothesis for channel gating involving two CaM molecules that bridge two SK fragments upon Ca binding to the N-lobe to lock the channel in a conformation that favors opening. We will refer to this as the 2/2 gating model. If all subunits are functioning during gating, SKs are proposed to assemble as dimers of dimerized subunits with twofold symmetry (). This contrasts with other homotetrameric K channels that generally assemble with fourfold symmetry (). Because CaM binds in both the absence and presence of Ca, CaM has been considered to be constitutively associated with SK (; ; ; ). However, no studies have shown that either the SK affinity for CaM or the structure of the complex are invariant with Ca concentration. The open probability of the channel will depend on Ca occupation of CaM, the affinity of CaM for SK at various Ca occupancies, and the Ca-dependent conformations of CaM, SK, and their complex. At present, the crystal structures provide only one possible conformation in the presence of Ca. We used multiple independent approaches to measure the stoichiometries and the molar masses and affinities of the complexes that form between SKp and CaM in high and low Ca. Multiangle light scattering (MALS) quantifies the scattered light intensity of a molecule at many angles to measure the molar mass (; ; ). Composition gradient MALS (CG-MALS) includes a recent advance in the technique to precisely mix different protein compositions and to measure the apparent molar mass at various molar ratios. Performing analyses as a function of composition increases the capability of MALS so that stoichiometry and protein affinities can be accurately measured (). Sedimentation velocity measurements using analytical centrifugation have the potential to measure both sedimentation and diffusion rate constants to provide information on the shape or hydration state in addition to estimating the mass of the molecule (; ; ; ; ). These methods provide interpretable conclusions about the stoichiometries and the affinities of SKp and CaM, an important step in characterizing and measuring the energetics of SK gating. The protein expression clones for the mammalian CaM constructs were provided by S. Hamilton (Baylor College of Medicine, Houston, TX). J. Adelman (Vollum Institute, Portland, OR) provided the clone for the SKp of SK2, which contains a C-terminal histidine tag (His-SKp). The human protein sequence of SK2 is identical in this region. Reagents used for buffers were minimal American Chemical Society quality and purchased from Sigma-Aldrich unless otherwise indicated. The SKp sequence is identical to the sequence used in previous studies, minus the histidine tag (). The sequence includes residues R396 to Q487, with extra methionine and glycine residues inserted at the amino terminus. PCR mutagenesis was used to insert a stop codon in the cDNA immediately before the histidine tag to create a construct that could express the SK2 peptide without a tag (SKp). For protein expression, BL21 (DE3) cells (Merck) were chemically transformed with CaM cDNA alone or with a 2:1 CaM/SKp molar ratio of cDNA together. The CaM construct confers carbenicillin resistance as a selection marker, whereas SKp confers kanamycin resistance. To select for BL21 (DE3) cells with both CaM and SKp, colonies were selected from plates that contained both antibiotics. A single colony was selected for a shaking overnight starter culture in Luria broth containing both 100 µg/ml carbenicillin and 50 µg/ml kanamycin. The next day, the starter culture was diluted to 1.5 liters and was grown to an optical density of 0.6–0.8 at 37°C, maintaining 250 rpm, then induced with 0.5 mM isopropyl β-D-1-thiogalactopyranoside and incubated in the shaker for 3 h before harvesting by centrifugation. The pellets were fully suspended at room temperature in 25 ml B-Per (Thermo Fisher Scientific) containing 100 U/ml Benzonase (Merck), 15 mM dithiothreitol (Roche), 0.5 mg/ml lysozyme (MP Biomedicals), Complete Mini, EDTA-free protease inhibitor tablets (Roche), and 2 mM MgCl. All subsequent steps were performed at 4°C. The suspension was centrifuged at 30,000 for 30 min. The pellet was saved on ice. The pellet with SKp was homogenized and solubilized in 6 M Guanidine-HCl followed by centrifugation for 30 min at 50,000 . The supernatant contains SKp, and it was dialyzed against 4 liters of 20 mM HEPES and 150 mM NaCl, pH 7.0, for 4 h. The sample was centrifuged again at 50,000 for 30 min. The supernatant was loaded onto a 20 ml HiPrep CM FF 16/10 (GE Healthcare) cation exchange column. Elution was performed by a gradient to 2 M NaCl and 20 mM HEPES, pH 7.0. CaM was purified according to established procedures that use ammonium sulfate precipitation and phenyl Sepharose chromatography (). As a final step in both the CaM and SKp procedures, the samples were concentrated and 5 ml of the protein solutions were separately passed through a HiLoad 16/60 Superdex 75 pg column (GE Healthcare). Purified proteins were stored at −80°C and thawed the same day of experiments. Before experiments, SKp and CaM were filtered through 0.45-µm pores in a centrifuge at 10,000 for 10 min to remove particulates. SKp was dialyzed against 1 mM HEPES, pH 7.0, and CaM was dialyzed against 1 mM Tris, pH 7.4, and 1 mM EDTA. A total of 1.5 nmol of each protein sample was sent to The University of Texas protein core facility for analysis. High protein purity for both SKp and CaM samples was confirmed by liquid chromatography–electrospray ionization/multiple-stage mass spectrometry (LC-ESI/MS/MS) in a hybrid triple quadrupole-linear ion trap (QTrap; ABI Sciex 4000) with Shimadzu Prominence analytical HPLC at the Protein and Metabolite Analysis Facility at The University of Texas at Austin. SKp appears as a single peptide having a mass of 11,015 D in LC-ESI/MS/MS. These data agree with the predicted mass of SKp, assuming nominal N-terminal methionine processing, of 11,014 D. The spectra suggest that up to three sodium atoms may bind, creating minor peaks at 11,037 and 11,060 D. Additionally, small but significant peaks are observed at 11,194 D and 11,274 D, which suggests a small amount of protein modification. The differences in mass are 179 and 258 D. These masses are consistent with gluconoylation at the N terminus of the peptide (; ). These studies suggest that gluconoylation occurs during the overexpression of recombinant peptides in bacteria, especially for histidine-tagged proteins. The SKp peptide has no histidine tag, as verified by mass spectrometry. If the modification is at the N terminus, it would be far enough from the putative CaM-binding residues of SKp that it would most likely be solvent exposed (). SKp has a theoretical isoelectric point of 10.6, perhaps suggesting that positive charges augment gluconoylation. The amount of gluconoylation appears to vary a little among different protein preparations, so greater numbers of experiments were performed across different batches of purification. Although not directly tested, we assume that any effect gluconoylation has on our studies is small. CaM in 1 mM EDTA and 1 mM Tris, pH 7.4, was also evaluated by similar LC-ESI/MS/MS analyses. There is a dominant peak at 16,708 D, which is virtually the same as the predicted mass of 16,706 D for monomeric CaM. Multiple minor peaks corresponding to adducts of Na, K, and/or Ca were observed. The Edelhoch method for measuring protein concentration was followed (). Dry protein was weighed on a balance that is accurate to ∼0.01 mg. CaM peptides with the sequences GYISAAE and VNYEE were synthesized and purified on Sep-Pak (Waters). Full-length CaM was dialyzed for 2 d with six exchanges against 4 liters water/exchange. Samples were frozen and then lyophilized under 2 × 10 mbar vacuum at −115°C and then heated to 100°C for 24 h to enhance the removal of water. Immediately upon completion of drying, samples were cap sealed to prevent moisture from being absorbed by samples. Masses were measured on an XP105DR balance (Mettler Toledo). Experimental variability for mass measurements was determined to be 0.04 mg. 8 M urea (Macron Chemicals) was used to dissolve all samples, as pure water was less effective. Dissolved samples were weighed. The solution was then ready for UV spectroscopy to determine the extinction coefficient of a known quantity of protein. Spectra from 240–350 nm were collected on a spectrophotometer (DU640; Beckman Coulter). The extinction coefficient (ε) was determined using the relation: ε = A/c, where A is the maximal absorbance near 280 nm with a path length of 10 mm and c is the protein concentration. The measured extinction coefficient for CaM we used for our studies is 3,020 ± 120 Mcm at 276 nm. Propagation of error was used for determining experimental variability. SKp has a tryptophan and its extinction coefficient was measured to be 7,200 ± 300 Mcm. CG-MALS was performed using Calypso II (Wyatt Technology), a triple syringe pump system, to deliver sample and solvent to a HELEOS II MALS detector (Wyatt Technology) and an inline UV/Vis dual wavelength detector (2487; Waters). Samples were dialyzed into 9 mM HEPES, 100 mM NaCl, and 2 mM CaCl, pH 7.0, or into 5 mM HEPES, 100 mM NaCl, and 5 mM EGTA. The free Ca in a solution of 5 mM EGTA was estimated to be <5 nM. We define 2 mM Ca as saturating Ca and <5 nM to be zero Ca. The free Ca in solutions containing EGTA was estimated using MaxChelator version 1.3 (). Buffers were filtered through 0.1 µm porous membrane. Proteins were diluted to predetermined stock concentrations and filtered to 0.02 µm with a syringe-tip filter (Whatman Anotop; GE Healthcare) before loading into Calypso II, which has 0.03-µm polyethersulfone inline filters (Sterlitech). After each injection, the flow was stopped for 60–180 s to allow for any association or dissociation kinetics. Two gradient protocols were used for proteins in Ca buffer, whereas just one protocol was followed for proteins in EGTA buffer. The “self+hetero-association” method consists of two single-component concentration gradients to assess self-association and a dual-component “crossover” composition gradient to assess the hetero-association behavior (). For each composition, 0.7 ml of protein solution at the appropriate composition was injected into the UV and MALS detectors. The flow was stopped for 60–180 s to allow the solution to come to equilibrium in the flow cell. During the hetero-association gradient, the concentration of CaM was increased while the concentration of SKp was simultaneously decreased to sample 13–15 different SKp/CaM ratios. The “dual crossover” protocol consisted of two mirror-image hetero-association gradients, each with 15 different concentrations of CaM and SKp, to replicate the binding conditions in the “self+hetero” method and to rule out hardware error (). The analysis of CG-MALS data has been previously described (). For a system of two interacting species A and B in a dilute solution, the light scattering intensity as a function of composition can be written aswhere R is the excess Rayleigh ratio, and M and M are the molar mass of each monomer. K* is an optical constant whereconsisting of the laser wavelength , refractive index of the solution , Avogadro’s number , and the differential refractive index of the molecule, (). The is how much the refractive index of a solution varies for a given increment of concentration and is assumed to be 0.185 ml/g (). The indices i and j represent the stoichiometric numbers of A and B in the AB complex, with AB and AB representing the competent monomers of A and B, respectively, and [AB] represents the molar concentration of the AB complex. The terms and refer to “incompetent” protein concentrations that are incapable of binding due to artifacts of protein purification, modification, or misfolding (). The concentration of each AB species is related to the total competent molar concentration of A and B ([A] and [B]) by the equilibrium association constant, K, and the conservation of mass as follows: If we let f equal the fraction of A that is competent and f equal the fraction of B that is competent, then we also have the following relations in the absence of ligand: The concentrations for [B] and [B] have the same relations. For each composition, the concentration of each monomer ([A] or [B]) is known from the UV measurement. The fraction of competent protein that participates in the interaction and the fraction of incompetent protein was determined through fitting. Iterative nonlinear curve fitting in the Calypso software (Wyatt Technology) was used to determine the monomer molar mass and K for each complex formed. Calypso uses the Levenberg-Marquardt algorithm for nonlinear curve fitting. All buffers were filtered through a 0.02-µm filter (Anodisc 47, catalog No. 6809-5002; Whatman). 100 µM CaM or 100 µM SKp-CaM complex were dialyzed against 20 mM HEPES, either 100 mM or 500 mM NaCl, and 2 mM CaCl, pH 7.0. Aliquots of samples (∼20 µl) or buffer were fractionated at room temperature by a TSK-GEL G3000PW SEC (300 mm × 7.8 mm [inside diameter], 14 ml column volume; TosoHaas Bioscience LLC) before measurements at 25°C in a DAWN EOS static light-scattering instrument (Wyatt Technology) as described previously (; ). To normalize the light scattering detectors, 2.5 mg/ml of BSA monomer (A1900; Sigma-Aldrich) was also fractionated on the G3000 SEC. Chromatography was conducted at a flow rate of 0.4 ml/min at room temperature, requiring 40 min to complete the elution of the protein sample. Insoluble components were removed via centrifugation for ∼30 s before injection from a 20-µl loop. Molar masses and peak concentrations were determined using Astra 4 and 5 software (Wyatt Technology). In 500 mM NaCl, CaM was used to elute SKp, which was bound to the column. When CaM removes SKp from the column, two maxima were observed when monitoring the differential refractive index, one maximum for the complex of CaM and SKp and one for excess CaM. For measuring the mean and variation of the measurement of mass, trials in both 100 mM NaCl and 500 mM NaCl were combined. This led to weaker but more generalizable conclusions for measurements of mass. Density and specific gravity of solutions were measured on a DMA 5000 density meter at 20.001°C (Anton Paar). Proteins were dialyzed with six exchanges against 4 liters of water over three days before measurements. After measurements, solutions were weighed, dried, and reweighed on a balance (XP105DR; Mettler Toledo). Specific volume (inverse of specific gravity) was plotted against percent protein mass (weight protein/weight solution) similar to what was done with BSA (), but as suggested by . The partial specific volume is extrapolated at 100% protein from either a linear fit or a tangent to the curve if the data are nonlinear, but a partial specific volume is desired for a specific range of protein concentration. With the molecules we studied, the data are linear through attainable concentrations. The errors were determined by propagation of error for density measurements and mass measurements. Partial specific volume of the SKp/CaM complex was measured at each concentration with fourfold molar excess Ca. Corrections and controls for added Ca were applied for mass and density measurements. Measurements are included in . Data were collected on an analytical centrifuge instrument (XL-I; Beckman Coulter) with a two-channel epon centerpiece (Beckman or Spin Analytical) inside a rotor (AN60-Ti; Beckman Coulter). Data were collected using Data Acquisition version P4.5 (Beckman Coulter). 450-µl samples were centrifuged at 42,000 or at 55,000 rpm at 20°C. The buffers at pH 7.0 contained 100 mM NaCl, 2 mM CaCl (saturating Ca), or 5 mM EGTA (<5 nM or zero Ca) and 5 mM of one of the following: Tris, HEPES, or sodium cacodylate. One experiment required adding Ca to a total of 7 mM to obtain 2 mM free Ca in the presence of 5 mM EGTA. Since adding Ca to an EGTA solution releases H that can drop the pH, HEPES was added to 25 mM to help stabilize the pH near 7.0. Protein concentrations varied among experiments from 15 µM to 200 µM. To obtain strong signals at different protein concentrations, different UV wavelengths were used to measure absorbance, including 230 nm, 242 nm, 280 nm, 287 nm, 290 nm, 292 nm, and 295 nm. Data were analyzed with the UltraScan III software package (UltraScan version 1.0-1206, a comprehensive data analysis software package for analytical ultracentrifugation experiments; The University of Texas Health Science Center at San Antonio, Department of Biochemistry; ). Time-invariant noise and meniscus position are modeled simultaneously, followed by corrected models for radial-invariant noise. The software also corrects for the density of water at 20°C. UltraScan III was used for time-derivative analysis and both van Holde-Weischet analyses and 2DSA Monte Carlo analyses. For time derivative, or dC/dt, analyses, a small subset of consecutive scans in each experiment was evaluated without removing time-invariant noise. The apparent sedimentation coefficient, , is determined by taking the time derivative of the selected scans. The result of dC/dt is plotted as a distribution for , which is defined as G*(). G*() was normalized to the maximum value at its peak so that all data with the same molar ratio of protein could be overlaid and compared with distributions at other molar ratios. Box plots of the median, first, and third quartiles were created from the sample means. The magnitudes of the whiskers correspond to the smaller of the data range or the value of (third quartile − first quartile) × 1.5. Van Holde-Weischet analysis (vHW) was performed for each experiment (; ). For statistical analysis, the integrated, binned frequencies of apparent sedimentation rates for each molar ratio were combined from all trials for comparative statistics with other molar ratios. Because the data do not show a symmetric distribution, a nonparametric permutation test was used to estimate a p-value (). The goal is to determine if the data for sample X is different from sample Y with sizes Nx and Ny, respectively. The observed means of X and Y were calculated and the difference in their means was stored. Next, test sets of size Nx and Ny were generated from random values from pooled collection of both X and Y. The means of test sets were calculated with 10,000 permutations of observed values. The difference in means of each test set was calculated, stored, and used to determine the proportion of test sets that have a greater difference in means than the observed set. If the null hypothesis is that the means of X and Y are the same, the distribution of mean differences should not depend on whether values were generated from X or Y. For example, if the experimental mean of the differences is not within 99% of the resampled sets, the null hypothesis can be rejected with a p-value of 0.01. The same raw data were fitted by 2D spectral analysis (). Error in the fitting was estimated through 50 Monte Carlo iterations, with simulated data having similar noise distributions as the original dataset. Global fitting and Monte Carlo analysis of each sedimentation velocity dataset was performed on the UltraScan LIMS cluster operated by The University of Texas Health Science Center at San Antonio. Modeled results from a single sample can include multiple solutes, i.e., molecular species. The total concentration of all solutes combined is normalized to one. The fitted parameter, partial concentration, is plotted as contoured lines on a two-dimensional plot. The axes of the two-dimensional plot are the fitted parameters frictional ratio and either the sedimentation coefficient or molar mass. Normalized results from similar experiments were pooled and rebinned for making contour plots of combined data using IGOR Pro version 5.0.5.7 (WaveMetrics Inc.). A complication with the crystallographic results () is that a histidine tag is present in the crystallographic studies with the SK peptide (). The histidine tag in the crystallographic model (see Protein Data Bank accession no. ) binds at the SK peptide interface and the amino acid chain of CaM that connects the N- and the C-lobes. The position of the histidine tag in the structure could complicate interpretations of both the 2/2 structure and our CaM binding measurements. A recent report suggests that the histidine tag at the interface of CaM and SK in the prior structure could have been modeled incorrectly, as a new, similar structural model more clearly shows SK residues (). To avoid complications we proceeded to use an SK construct without the histidine tag, which we call SKp, and purified it for binding measurements with CaM. Although a 1SKp/1CaM complex in Ca has not been reported to our knowledge, a 2SKp/2CaM complex was previously observed by determining structures using x-ray crystallography (; ,, ). Our initial goal was to determine the stoichiometry of CaM with SKp in Ca saturating solutions to compare with the complex that forms in a crystal environment. lists the molar masses of both CaM and SKp, and the predicted molar masses of possible complexes that could form between the two proteins. Throughout this paper, the forward slash symbol (/) is used to describe the stoichiometry of a complex formed from multiple proteins. A colon is used to describe the molar ratio of protein concentrations in a solution. Our goal was to study SKp/CaM complexes in solution. One way to determine the Ca-dependent stoichiometry of SKp to Cam is to measure the molar mass of the complexes that form as a function of molar ratio. CG-MALS combines precision composition gradients with a measurement of absolute molar mass from first principles; i.e., mass is not extrapolated from standards. Two protocols were followed for collecting data: self+hetero-association and dual crossover. Both protocols measure light scattering of proteins at equilibrium after mixing at different molar concentrations of SKp:CaM. The magnitude of the light scattering signal depends on the molar mass and concentrations of the molecules in solution. If complexes form, the light scattering signal will be greater than that of either of the individual proteins being titrated. If no complex forms, the light scattering signal increases relative to the weight-average signal for the individual components. This powerful method simultaneously quantifies monomer molar mass, stoichiometry, and equilibrium association constants for self- and hetero-association complexes formed in solution. We used CG-MALS to measure the complexes that form between SKp and CaM at zero and at saturating Ca (). We define zero Ca as <5 nM Ca in the presence of 5 mM EGTA, and saturating Ca as 2 mM Ca. In , light scattering signals are plotted at zero Ca, but at different protein concentrations. likewise shows the corresponding experiments at saturating Ca. A simple relation is shown for complex formation between SKp and CaM at zero Ca, which shows a single peak that has a maximum signal close to where SKp and CaM are equimolar (). At saturating Ca, but low protein concentrations (), a maximum occurs at concentrations where [SKp] > [CaM]. There is also a shoulder where [CaM] > [SKp], which indicates formation of at least one additional stoichiometric complex. At higher protein concentrations (), there is a clear minimum between two separate peaks. A dip in amplitude is not expected for simple 1/1 interactions between two proteins. The light scattering signal at each molar ratio of SKp:CaM depends on the contributions of each molecule or complex present in solution. We attempted to fit all of the data in simultaneously using different stoichiometric models for binding. The adjustable parameters in fitting include the following: the molar masses of SKp and CaM, percentage of SKp or CaM that is “incompetent,” the concentrations of each allowed stoichiometry in a model, and the for the formation of each species. Incompetent protein is defined as protein in solution that does not participate in the interaction, i.e., that is inactive, chemically modified, or misfolded. The incompetent protein fraction of either molecule is usually ∼10% ( and ). Although not tested, incompetent protein may represent an effect from SKp gluconoylation (see mass spectrometry in Materials and methods). In , the modeled light scattering contributions of each competent and incompetent molecule, and each complex, are shown. A simple 1/1 model for a 1SKp/1CaM complex (or 2/2) with no other stoichiometries would obviously fail to fit in all conditions because it does not account for more than one peak, nor can an exclusive 1/1 model account for shoulders. Fewer stoichiometries can be resolved at zero Ca than at saturating Ca because there are fewer features at zero Ca. Therefore, there are multiple possibilities that could adequately explain the zero Ca data. In all cases, both a 1SKp/1CaM and a 2SKp/1CaM complex were needed to suitably fit the data; however, other situations provided fits that were indistinguishable from the simplest case. In one example we included a 2SKp/2CaM complex in the model for fitting at zero Ca, but only an insubstantial fraction of 2SKp/2CaM complex was compatible with acceptable fits to the data, i.e., the maximum contribution is <5% of the light scattering signal. Acceptable fits could also be obtained with unacceptably large fractions of incompetent CaM and SKp, with >50% incompetent protein. Therefore, we present the simplest solutions with the best fit. The light scattering signal at zero Ca has at least two components, as shown by the modeled contributions of individual species to the light scattering fit (). Because other stoichiometries have larger masses and because larger masses produce more scattering, most SKp and CaM must be combined as 1SKp/1CaM to produce the observed light scattering signal. The calculated (see Materials and methods) relative mole fractions of each molecular species giving rise to the light scattering signals are shown in . Virtually all of the SKp and CaM molecules are bound in a 1/1 complex at equimolar concentrations (). The 1SKp/1CaM complex accounts for 80% of the protein in the solution, whereas the unbound or “incompetent” SKp and CaM proteins account for the other 20%. The estimated for the 1SKp/1CaM complex is <1 nM in zero Ca (). The affinity is strong enough that the protein concentrations of SKp and CaM in this experiment have little effect on the amplitude of the mole fraction of 1SKp/1CaM. In , 80% of protein is complexed at a 1SKp:1CaM molar ratio regardless of concentration used. Although the 1SKp/1CaM complex is the dominant species at equimolar ratios, there is another stoichiometry that must be included during fitting at zero Ca. When a 2SKp/1CaM complex is included in the model for fitting, the fits converge to the data (). There is a noticeable contribution from the 2SKp/1CaM complex to light scattering at zero Ca, but the contribution is much smaller than that of the 1SKp/1CaM complex (). Its contribution is noticeably greater when the concentrations of SKp and CaM are higher (). Only a small fraction of 2SKp/1CaM in the solution relative to the 1SKp/1CaM is needed to affect the light scattering signal (). The apparent affinity for formation of the 2SKp/1CaM complex is weaker than 8 µM (). The data do not support the addition of other stoichiometries to improve fitting to the data. At saturating Ca, only multiple stoichiometries with strong affinities are compatible with the data. Similar to what is observed at zero Ca, when the concentrations of SKp and CaM are nearly equal, nearly 80% of the protein forms a 1SKp/1CaM complex with a < 100 pM (; and ). However, if the molar ratio is not 1:1, other complexes form. When [SKp] > [CaM], a 2SKp/1CaM complex dominates the light scattering signal (). There is also more 2SKp/1CaM than 1SKp/1CaM in the solution when the molarity of SKp > CaM (). Because the is ∼200 nM for the 2SKp/1CaM complex, it is >10-fold stronger at saturating Ca than at zero Ca ( and ). In addition, a 1SKp/2CaM molecule must be included to account for the second hump at higher CaM ratios (). An attempt to include the 1SKp/2CaM complex in a fit at zero Ca produces a zero contribution, so this is strictly a phenomenon observed at saturating Ca. Although the 1SKp/2CaM complex is larger than the 2SKp/1CaM complex (see for predicted masses), the signal is considerably weaker (). This implies a lower affinity to add a second CaM to the 1SKp/1CaM complex. The modeled affinity to bind the second CaM is ∼2 µM (). In , 2SKp/1CaM accounts for 26% of the protein when SKp is in excess, but 1SKp/2CaM only accounts for 6% of the protein when CaM is in excess. When proteins are injected at higher concentrations (), 2SKp/1CaM can reach nearly 40% of the total protein when SKp is in excess, whereas 1SKp/2CaM approaches 15% when CaM is in excess. We converted the data in to apparent molar mass (). We plotted it as a function of the protein composition. Simulated data are compared with the measurements. The black symbols represent the measured apparent molar mass of complexes that may form from SKp and CaM. At the maximum concentration of SKp, there is no CaM in the solution. The measured mass is equal to the mass of free SKp (). Assuming no interaction, as the CaM concentration increases and the SKp concentration decreases the apparent molar mass would follow a straight line representing the weight-average of the two monomers (). At maximal CaM concentration, the molar mass is equal to the molar mass of free CaM (). The simulations represent what data would look like if 100% of the molecules formed a particular complex. The simulations were generated with the following numerical guides: the maximum concentration of each protein in the simulation is equal to the maximum concentration of each protein during the experiment, and each simulation uses subnanomolar affinities for the formation of each complex to get the maximum height possible if a complex were to form. Lower affinity interactions produce peaks that do not approach the actual molar mass of the presumed complex. The blue traces in the simulations show that if 100% of the protein forms a 1SKp/1CaM complex, the maximum occurs where the molar ratio of SKp and CaM is equal to one. The height of this peak is equal to the 1SKp/1CaM molar mass for a high-affinity site. In , the maximum apparent molar mass at zero Ca is very close to the actual mass of a 1SKp/1CaM complex. Thus any larger stoichiometries, such as a 2SKp/2CaM complex, must be substantially lower in concentration. In , at saturating Ca a dip in the amplitude of the data points is observed at 1:1 molar protein concentrations. The apparent molar mass is equal to the theoretical molar mass of the 1SKp/1CaM complex. This essentially excludes any possibility of a 2SKp/2CaM complex forming in solution, or else the apparent molar mass would not be smaller at 1:1 molar ratios than at other molar ratios. This is especially true because both 2SKp/1CaM and 1SKp/2CaM exist. Without a high-affinity 1SKp/1CaM complex, the apparent molar mass would be influenced by both larger stoichiometries and would “flatten out” the apparent molar mass as the composition changes from excess SKp to excess CaM. The green simulated curves () also show that if 100% of the proteins form a 2SKp/1CaM complex, the maximum apparent molar mass is expected to be smaller than if 100% of the proteins form a high-affinity 1SKp/2CaM complex shown in red, because a 1SKp/2CaM complex has a larger molar mass. The maximum height in the measured data approaches the molar mass of the 2SKp/1CaM complex when [SKp] > [CaM]. The smaller bump in the titration series occurs where [SKp] < [CaM]; however, the apparent molar mass is smaller than at high molar ratios of SKp. This is because there is less 1SKp/2CaM in solution than 1SKp/1CaM () so the weight-averaged molar mass is smaller, which is also indicative of a weaker affinity to form 1SKp/2CaM. A model for apparent molar mass was simulated using the fitted parameters in and . The simulation is shown as solid black lines in . The simulated curves overlap the measured data. Our data support the formation of two stoichiometries at zero Ca, but three stoichiometries at saturating Ca. In the presence or absence of Ca, we find no evidence of the 2SKp/2CaM complex suggested by prior crystal structures. The most striking difference between the data at zero Ca and at saturating Ca is the absence of the 1SKp/2CaM complex. This is the first clear evidence that Ca drives changes in stoichiometry between SKp and CaM. Concentration gradients of a protein without a partner are performed at the beginning and end of self+hetero-association experiments (). These data provide more detail into how the complexes form. For example, one CaM molecule may bind to a dimer of SKp molecules, or a 1SKp/1CaM complex may need to form before a second SKp molecule can bind. If proteins self-oligomerize, measuring the concentration dependence of apparent molar mass of a protein without partner will provide evidence for self-association. shows that across the concentrations used for CG-MALS, neither SKp nor CaM form homo-oligomers. Furthermore, and show that the fitted molar masses for SKp and CaM are very close to their predicted values in . The lack of homo-oligomerization, and the very strong affinity of SKp for CaM, suggests that the different stoichiometries are not formed by a partner binding to a homodimer. One disadvantage of CG-MALS is that it is performed in batch mode. Size-exclusion chromatography combined with light scattering (SEC-MALS) resolves species of molecules before entering the light scattering flow cell (). In addition, because CG-MALS was performed at low protein concentrations (<5 µM) that could have missed evidence of other weaker oligomers, we measured the mass of the complex that SKp forms with CaM at up to 100-µM protein concentrations by SEC-MALS. First, CaM was applied to a molecular-sieve column in the MALS system to measure light scattering and differential refractive index. In , the solid curves represent the protein signal from the refractometer. The circles are the real-time measurements of the molar masses of proteins after size fractionation. CaM elutes from the column at 6.85 ml. Data from the entire peak were analyzed to obtain the weight-average molecular weight (; ). These papers show that band broadening is best used for single component samples. We don’t use band broadening because it conceals components that elute close together. The entire peak is required because of the spreading of the peak between the light-scattering photometer and the refractometer (; ). CaM (34 µg) was injected, and the in-line refractometer measured 29 µg of protein, with the differential index of refraction (dn/dc) assumed to be 0.185 ml/g (). After averaging = 4 trials, the calculated weight-average mass of CaM was 17.7 ± 1.3 kD (), close to the predicted value of 16.9 kD based on the amino acid sequence (see ). We next incubated an equimolar (100 µM) mixture of CaM and SKp in 2 mM Ca for 24 h before the complex was injected onto the SEC. Protein elution is detected at 6.85 ml, the same volume at which free CaM elutes. However, the measured weight-average mass of this mixture is 27.7 ± 2.9 kD with = 3 trials (); this is consistent with a 1:1 stoichiometry, which has a predicted value of 28.0 kD (see ). These data caution against using size-exclusion chromatography exclusively as a method for analyzing the formation of protein complexes because the protein complex elutes at the same volume as one of its parts, depending on the confirmations involved. This is not surprising because size-exclusion chromatography separates molecules based on multiple physical parameters that include hydrodynamic volume, intrinsic viscosity, and interactions with the column matrix, in addition to solute mass (). Thus a dumbbell-shaped molecule like free CaM has the same retention volume on the G3000 SEC as the SKp/CaM complex, which may be more globular but occupies a similar hydrodynamic volume. SKp by itself with 0.1 M NaCl in the buffer sticks to the column matrix and fails to enter the light scattering chamber. No SKp eluted when NaCl was increased to a concentration of 0.5 M in the buffer. To remove SKp from the column, 100 µM CaM in the 0.5 M NaCl and 2 mM CaCl buffer was injected into the column. also shows that CaM is capable of displacing SKp from the column by forming a SKp/CaM complex, as demonstrated by the two peaks in the red curve. In 100 mM NaCl, CaM elutes at the same retention volume, 7.0 ml, as SKp/CaM (). In 500 mM NaCl, the CaM-SKp complex elutes at 7.05 ml, before the free CaM peak at 7.65 ml (). The molecule that elutes in the first peak in has a mass comparable to a 1SKp/1CaM complex with a mass of 28.7 kD. The second peak has a mass very close to that of unbound CaM, 18 kD. A change from 0.1 to 0.5 M NaCl has a greater effect on the elution volume (i.e., retention volume) of free CaM, which indicates that the hydrodynamic volume of CaM in solution is sensitive to salt. The main finding using MALS is that regardless of how the SKp/CaM complex journeys through the column matrix, the mass of the complex eluting from a column is consistent only with a 1/1 stoichiometry, at a 100-µM protein concentration. An independent approach to detecting protein complex formation is measuring sedimentation velocity in an analytical centrifuge (SV-AUC). The partial specific volumes of SKp, CaM, and 1SKp/1CaM were measured and recorded in . SV-AUC measures how a solution of particles sediments in a time-dependent manner in a centrifugal field, and the sedimentation progress is periodically observed by scanning the cell with UV light to monitor the distribution of protein concentration by its absorbance. As molecules sediment radially outward, the concentration of protein decreases at smaller radial distances, and the half-maximal absorbance moves radially outward with each successive scan. Example datasets that have been corrected for time-invariant and radial-invariant noise are shown in . After noise subtraction, the stochastic noise that remains is represented by the residual traces below each panel. The sedimentation of SKp and CaM at equimolar concentrations is shown in . Each scan is taken minutes apart. One experiment has up to 150 scans. Each scan has a characteristic sigmoidal shape. In contrast, there is a double sigmoid curve observed with fivefold molar excess SKp over CaM as shown in . This could be explained by protein absorbance having contributions from a faster sedimenting complex and from slower sedimenting free SKp. We used two methods to measure the sedimentation coefficient () values in our experiments. As a parameter, is quantitatively related to how fast particles can sediment normalized by acceleration. The coefficient has a unit of time, e.g., seconds. Particles that sediment faster will have a greater value for . One way to quantify the sedimentation is by taking the time derivative of the concentration profile, dC/dt (). This type of analysis has the advantage in that measurements are not influenced by small to moderate amounts of time invariant noise. Also, because the measurement is largely a derivative of a set of scans, the calculated parameters are not heavily model dependent. One major weakness is that diffusion is not accounted for, but diffusion can affect the measurement of , especially for slowly sedimenting molecules (). In brief, dC/dt is performed on a small subset of sequential absorbance scans, which are selected to have no interference from the meniscus and have a stable protein absorbance plateau at larger radii. shows distributions of apparent sedimentation coefficients calculated from dC/dt. Normalizing the distributions to the peak value allows one to compare experimental trials or conditions. vHW () is capable of determining the sedimentation coefficient without the need for intensive modeling or correcting for the contribution from diffusion. A subset of the scans from data, such as those in , are chosen to include only the data that have a stable absorbance plateau and are far removed from the meniscus. The absorbance axis of each dataset is divided into evenly distributed boundaries, and each boundary line crosses all of the scans, but intersects each scan only at one radial position. The distance between scans along a boundary shows net molecule displacement over time. The time displacements along a boundary are extrapolated to infinite time to provide a measurement for the sedimentation coefficient, , for that boundary. The extrapolated for each of 110 boundaries is plotted in , and plots are shown for each experiment. This type of plot can provide information on whether there is sample heterogeneity or nonideality. Heterogeneity of samples or samples with nonideality would produce a broad distribution of values. Alternatively, of homogeneous samples are confined to a nearly vertical line (). In , all samples appear mostly homogeneous, i.e., the distribution of values is almost vertical. There are some scans where increases toward the upper boundaries, which suggests that small amounts of oligomers or aggregates form. Drawing a vertical line through the data gives a good estimation of . Both dC/dt and vHW provide consistent measures for the molecular sedimentation. The mean sedimentation coefficients, , are listed in and shown in . The measured for SKp is ∼1.0 × 10 s, and this is easily resolved by eye in and . Even with different centrifuge speeds and different total protein concentrations, there was little difference in the measured mean, although there was some variability in the width of the dC/dt distributions (). We note that one noisy outlier was removed from these analyses because of an insufficient number of scans. CaM, a larger molecule, migrates faster with a coefficient near 2.0 × 10 s ( and ). Of the conditions we studied, the sedimentation profile of CaM seems the most sensitive to rotor speed. In , higher rotor speeds accounted for the narrower distributions, but this is partly expected as diffusion has a smaller effect at faster speeds. When SKp is combined with equimolar CaM to form a complex, a molecule with = 2.5 × 10 s is formed ( and ). The change in sedimentation of CaM in the presence of SKp strongly suggests the formation of a complex, which is consistent with light scattering data. For a more complete analysis using vHW, several datasets with the same protein composition were combined and the values were binned so that the integrated frequency of measurements could be plotted in . The data are normalized to show percent of total. In , the integrated values show a steep rise with a half-maximum, i.e., , at 2.51 × 10 s. We also combined dC/dt Gaussian fits into one graph for comparisons in . The means from all analyses are plotted in . In addition to 1:1 molar ratios of SKp to CaM, other molar ratios were analyzed by sedimentation velocity. These include the molar ratios of 2SKp:1CaM and 1SKp:2CaM shown as green and red traces, respectively, in and in . When molarities are different by twofold, the observed sedimentation coefficient is always greater than when SKp and CaM are at a 1:1 molar ratio, and these differences are significant (P < 0.01) as determined by a two-sample test or a nonparametric permutation test (see Materials and methods; and ). The species that forms at 1SKp:2CaM sediments faster than the one that forms at 2SKp:1CaM. Because the molar mass of the 1SKp/2CaM complex is larger than 2SKp/1CaM (), its faster sedimentation can best be explained by having the larger mass, as opposed to conformation. This is explored in more depth in the next paragraph. Some data are not completely vertical, which may indicate some heterogeneity or influence by nonideality, but by and large the data appear homogeneous. The behavior is not strongly concentration dependent, as demonstrated by overlapping data from different experiments (). A faster sedimentation indicates a larger mass, a change in the hydrodynamic volume, or a combination of these two factors. A different stoichiometry will have a different mass. Thus further analyses of the sedimentation velocity data are needed to determine whether excess of either SKp or CaM alters either the stoichiometry or the hydrodynamic volume of the complex. Because vHW is better suited for heterogeneous experiments, dC/dt is not used for molar ratios with more than twofold differences in concentrations of SKp to CaM. When SKp is at least fivefold molar excess over CaM, the absorbance profile of the scans show two different plateau phases (). Applying vHW analysis clearly resolves the sedimentation coefficients of both species. Instead of vertical, the profile has one phase that aligns with the 2SKp:1CaM data and another phase that aligns with the free SKp data (). The simplest interpretation is that these phases align with a 2SKp/1CaM complex and free SKp. Similarly, when CaM is in 10-fold molar excess over SKp, vHW analysis resolves a phase that aligns with the 1SKp:2CaM data and a second phase that aligns with free CaM (). In this case, we resolve the 1SKp/2CaM and free CaM. Together, these data support our interpretations of the CG-MALS data where both 2SKp/1CaM and 1SKp/2CaM stoichiometries were observed at saturating Ca. It may be possible to distinguish changes in mass (i.e., stoichiometry) or in hydrodynamic volume with sedimentation velocity data if simulated models solve for both parameters simultaneously. Although the measured values are model dependent, they can be used to distinguish qualitative changes that have occurred due to mass or shape change or both. Most of the samples that were analyzed by the vHW method were also analyzed by two-dimensional spectral analysis algorithms combined with Monte Carlo statistics, 2DSA-MC (). The main idea of 2DSA-MC is that the concentration of each protein solute in a solution has a different sedimentation behavior that can be uniquely measured by absorbance. The corresponding mass and hydrodynamic parameters for each solute are solved simultaneously with the partial protein concentration of each solute. This is possible because the absorbance profile depends both on the radial distribution of protein and the time at which absorbance scans across the window are collected. UltraScan III employs the use of supercomputers to search exhaustively over a large parameter space for unique solutions to the Lamm equation (), C is solute concentration at time t and at radius x, is the sedimentation coefficient, ω is the angular speed of the rotor, and k is the diffusion coefficient. Plots are created that show solutions to the sedimentation velocity datasets. One useful parameter that can be fitted is the frictional ratio (f/f), which is proportional to the volumetric radius times the solution viscosity (). UltraScan III uses the Stokes-Einstein relation to relate the diffusion and sedimentation coefficients to the frictional coefficient ratio (): Here, and again refer to diffusion and sedimentation, R and T are the gas constant and temperature, N is Avogadro’s number, η is the solvent viscosity, ρ is the solvent density, and is the partial specific volume. The f/f ratio is influenced by the compactness and the hydration state of a molecule. The purpose of using f/f instead of the diffusion coefficient in solving the Lamm equation is that it provides the user with an intuitive guess over which range to fit the data. For example, virtually all protein shapes, except filaments, will be in the range 1 < f/f < 4. A dehydrated sphere has a ratio f/f = 1 (). Any other value for f/f cannot distinguish a shape change from a change in hydration state, yet an f/f > 3 would strongly suggest a thin rod. Data in are presented in terms of f/f versus either or . After first correcting for noise to get an initial model that fit a single experiment, the solutions to each trial experiment are analyzed statistically using Monte Carlo simulations to determine how broad a range of parameters can be fitted and still be within tolerance of a good fit. A contour plot helps to show these parameter ranges. Multiple experiments can then be combined and a contour plot is redrawn to represent the combined data. The partial protein concentrations at saturating Ca represented at a particular f/f and are represented by contour levels in . An example of how the data are assembled is described in . Contour plots are generated for each trial, as shown in (inset) for equimolar SKp and CaM. All the trials for a sample are combined and rebinned to construct a contour plot of the combined data as in . Comparing different sample compositions is now possible, as shown in . As in and , is clearly resolved for SKp, CaM, and 1SKp:1CaM solutions. When equimolar SKp and CaM form a complex, the dominant density is clustered around = 2.5 (). Because the distribution is narrow and all other parameters are assumed, the uncertainty in f/f appears to be entirely caused by the implicit parameter estimation of diffusion (). Diffusion is not as easily measured as sedimentation at a single rotor speed. If is accepted with its uncertainty, the molar mass can be determined (). The greatest density of observation points for 1SKp:1CaM lies at ∼30 kD (). This is most consistent with a single SKp molecule binding a single CaM molecule, and there is very little experimental support for a 2SKp/2CaM complex near 55 kD. Also in and , it appears that the species that sediments with a coefficient of 2.75 × 10 s in 2SKp:1CaM molar ratio (green contours in ) is more massive (∼36 kD) than the species that sediments when the molar ratio is 1SKp:1CaM (blue contours at ∼28 kD in ). This suggests that an additional SKp binds to the 1SKp/1CaM complex to form a 2/1 complex when SKp is in excess. When CaM is in excess, it is less clear how to interpret the data. A peak appears within the same molecular weight range as the 1SKp/1CaM complex; however, moderate peaks are found in the ∼45 kD range consistent with one additional CaM binding to the complex. Because the CG-MALS data show that the affinity for a second CaM binding to the 1SKp/1CaM complex is weaker ( and ), perhaps this is a representation of both populations being present. The frictional ratio of the contours for 1SKp:2CaM that matches the molar mass of the 1SKp:1CaM data is slightly smaller. A conservative conclusion is that excess CaM influences either the conformation of the 1SKp/1CaM complex or the number of bound waters, perhaps through rapid associations. Sedimentation velocity experiments were also performed at zero Ca and analyzed by vHW. At protein concentrations up to 68 µM, varying the molar ratio of SKp:CaM has little effect on the sedimentation coefficient. In zero Ca, the sedimentation coefficients at <68 µM protein are 2.3 × 10 s for 1SKp:1CaM, 2 × 10 s for 2SKp:1CaM, and 2.3 × 10 s for 1SKp:2CaM (). These values suggest that sedimentation for the 1Skp/1CaM complex is a little slower at zero Ca than in saturating Ca. At these protein concentrations, we were not able to resolve by SV-AUC the 2SKp/1CaM complex that was hinted at in the CG-MALS data. Because the data in indicate that there is a weaker affinity for 2SKp/1CaM to form at zero Ca, we attempted sedimentation velocity at protein concentrations >100 µM. Although varying the total protein concentration at saturating Ca had little effect on the sedimentation of any stoichiometric complex ( and ), 100 µM protein uncovered a tendency toward aggregation at zero Ca when [SKp] > [CaM] (). In Ca, proteins in the 1SKp:2CaM sample sediment the fastest, but at zero Ca, proteins in the 2SKp:1CaM solution sediment at a higher, but variable, coefficient of 4.2 × 10 s (). Integrated 1SKp:1CaM data are not vertical, which shows heterogeneity in different samples (). We may interpret this to mean that minor aggregation is still present with a 1SKp:1CaM sample, which sediments with a mean coefficient of 2.9 × 10 s. In contrast, for a molar ratio 1SKp:2CaM, the results are homogeneous within a sample, which shows a nearly vertical vHW plot, and across experiments as shown as a vertical integrated plot (). We measured to be 2.5 × 10 s, which is approximately what we have interpreted to be the value of for a 1SKp/1CaM complex at saturating Ca. Because is smaller at zero Ca for 1SKp:2CaM solutions, this either means that the 1SKp/2CaM complex is more compact at zero Ca, or it means that only a 1SKp/1CaM complex forms and that a second CaM molecule does not bind to the complex. Second, because there is no heterogeneity in the sample, we can conclude that aggregation is substantially reduced when CaM is in molar excess. We measured the sedimentation over a broader range of SKp:CaM at zero Ca. The gray curves in show data with excess SKp. With a 10-fold excess of SKp over CaM, the curve shows one phase that covers a broad range of , and a stable phase where is ∼1 × 10 s, which is similar to free SKp (). This shows that not all of the SKp forms an aggregate. The aggregate only forms when CaM is present in smaller molar quantities. It appears that aggregates require multiple SKp and at least one CaM molecule. When SKp is fourfold higher than CaM, there is a mixture of dissociated SKp and aggregate (). The maximum at fourfold excess SKp, ∼4 × 10 s, is no greater than observed at twofold molar excess SKp, ∼4.5 × 10 s. Because the observation is weighted by all particles in solution, the heterogeneity may indicate that there are multiple oligomers that form in solution. We cannot resolve any oligomer or aggregate with certainty. When the molar ratio of CaM is 10-fold higher, the maximum is <3 × 10 s as shown in the cyan curve in . The of free CaM is ∼2 × 10 s, but there is no phase of the plot with excess CaM that clearly shows the sedimentation of free CaM. The vHW plot is not vertical, but slanted toward smaller values. This finding indicates that some heterogeneity or nonideality exists during sedimentation (i.e., charge or protein concentration effect). In either case, the 1SKp:10CaM data suggests a bias toward smaller sedimentation coefficient values that are most consistent with a 1SKp/1CaM complex. We noted that freshly thawed SKp is stable during the time course of our experiments, but after days at 4°C irreversible precipitation is observed by the eye. Only freshly thawed SKp is used in our experiments. The CG-MALS data does not show aggregation of SKp at zero or at saturating Ca (). At >100 µM protein concentrations, CaM can induce precipitation of SKp that is partially reversible if [SKp] ≥ [CaM] at zero Ca, but precipitation is not observed when [SKp] < [CaM]. This qualitative assessment suggests that excess CaM at zero Ca prevents runaway aggregation that forms a precipitate. In contrast, at saturating Ca no precipitation at any molar ratio of SKp:CaM is observed. Ca appears to have a negative effect on aggregation. shows an experiment that starts at zero Ca. Sedimentation velocity data were collected and are shown as the open symbols. After the SV-AUC run finished, Ca and a pH buffer was added to elevate the free Ca to 2 mM. As a cautionary note, proteins that have sedimented at zero Ca experience very high concentrations at the end of the centerpiece well and can precipitate out of solution. A 20–35% loss of protein was observed when comparing the runs at the maximum absorbance at time zero of each condition (unpublished data). The open symbols in show data at zero Ca, and the closed circles are data after Ca was added. When comparing the curves before and after Ca was added, proteins at 2:1 and 1:1 molar ratios of SKp:CaM have a smaller sedimentation coefficient, , at saturating Ca than at zero Ca. Notably, the 2SKp:1CaM sample sediments much more slowly with Ca present (). We hypothesized that the aggregates or oligomers at zero Ca had more molecules of SKp than CaM. This would indicate a greater loss of SKp if the aggregate precipitated over the time course of the centrifugation. In fact, we did notice a reduction in the absorbance signal indicating a small loss of material. Of the material that remained, the apparent sedimentation of 2SKp:1CaM in added Ca is comparable to or less than the observation with 1SKp:1CaM. We can conclude that Ca reversed the oligomerization state of the proteins in the 2SKp:1CaM solution so that only a 2SKp/1CaM complex is compatible with the SV-AUC data in added Ca. The opposite sedimentation behavior is observed when the ratio is 1SKp:2CaM. At this ratio, is greater when Ca is added (). This shows that just adding Ca to a 1SKp:2CaM molar solution can result in a molecular species that sediments faster. The following conclusions can be made from : at zero Ca, aggregation that is present in both the 1SKp:1CaM and 2SKp:1CaM is reversed with saturating Ca, and the 1SKp:2CaM condition is the only sample that shows an increased sedimentation coefficient when Ca is added, which is consistent with CG-MALS showing that the 1SKp/2CaM complex only forms in Ca. Our studies of CaM binding to SKp in solution are summarized as a scheme shown in . Multiple stoichiometries can form at zero and at saturating Ca. At zero Ca (<5 nM), SKp and CaM bind in a 1/1 complex (). In excess SKp, a second SKp may combine to form a 2SKp/1CaM complex. We presented some evidence that additional oligomeric or aggregated states can form. When Ca saturates the complex, SKp and CaM can combine in a 1SKp/1CaM, a 2SKp/1CaM, or 1SKp/2CaM complex (). The 1SKp/2CaM complex was not observed in any experiment at zero Ca. A 2SKp/2CaM complex observed in a crystal structure is not present in our solution studies in zero or saturating Ca concentrations. These results are not consistent with the 2/2 gating model, where Ca binding drives a transition from 1/1 to 2/2 complexes. The 2/2 gating model proposes a modular role of CaM in SK activation. At low Ca, the complex that forms is 1SK/1CaM, with only the C-lobe of CaM engaged. An increase in intracellular Ca drives Ca binding only to the N-lobe of CaM, and a 2SK/2CaM interaction forms. Although the 2/2 gating model is consistent with the crystallographic results for a single conformational state, there has been some physiological evidence that it is insufficient to describe channel gating. For instance, the part of CaM responsible for Ca dependent binding, the N-terminal lobe, also has an effect on Ca-free interactions with SK (). It is clear that our understanding of the role of CaM in SK gating is inadequate. Although our results find little support for 2SKp/2CaM, prior reports (; ,, ) mandate that we explore the possibility thoroughly. Our interpretations of CG-MALS data rely on fitting data that have multiple parameters, but the parameters may be correlated, thereby confounding the determination from the fits (unpublished data). We addressed this issue in several ways to increase our confidence in our conclusions. The molar masses of SKp and CaM can be measured independently during the CG-MALS experiment, and the measurements for molar mass are almost identical to what is predicted for monomers. The fact of whether or not these parameters are fixed or allowed to float during fitting of the hetero-association data does not affect the measured stoichiometries and apparent ( and ). We note that the Calypso software from Wyatt Technologies does not report an uncertainty for the log(K) values from a reversible association fit. We therefore presented ranges that were determined by fixing or floating certain parameters ( and ). In most modern methods the incompetent fraction is immeasurable and it is usually ignored, but we argue that protein chemists need to consider the amount of incompetent fraction for concentration-sensitive binding studies, especially because MALS appears capable of quantifying it. There is a precedent for including an incompetent fraction during fitting (; ). The inclusion of this term did not affect the measured stoichiometry. It also changed the measured by less than twofold for parameters within instrumental capability ( and ). There is no rationale for including additional stoichiometries in our models. We attempted to fit other stoichiometries, but the modeled light scattering signal approaches zero for each extra included stoichiometry. However, this obviously doesn’t rule out the potential formation of other higher-order associations at high concentrations (outside the range that we studied). A very small proportion of a 2/2 complex at zero Ca cannot be definitively ruled out, but there is not enough scattering to be conclusive. Concentrations >15 µM would be required to confirm or rule out this complex, which was outside the scope of our measurement limitations or secondary in importance to the loss of the 1SKp/2CaM complex and a decrease in binding affinity for the second SKp binding site. We also tried to force a 2SKp/2CaM interaction in the presence of calcium during fitting. Unlike the calcium-free case, this term was always thrown out of the “with calcium” fit, no matter what the initial guess was set to be. Using simulation, we can get the same M-shape only if we leave the fitting parameters as they are for the 1SKp/1CaM, 2SKp/1CaM, and 1SKp/2CaM terms and add a 2SKp/2CaM with log(K) ≤ 24. If we take this log(K) to represent the dimerization of two 1/1 complexes, that leaves a dimerization ≥ 250 µM. Because we were >100-fold below this concentration, this basically means that under the conditions tested, no 2/2 complex formed, and we would have to increase the concentration by at least 10-fold to see the effect if it exists. To extend this argument, our AUC data went up to at least 250 µM with still no evidence of a 2/2 complex (see point No. 5). Qualitative features of the data and the fits are consistent with our conclusions. We know the expected masses of all the complexes that form between SKp and CaM, which would most likely form given a molar ratio of SKp:CaM, and how close the calculated mass of a complex comes to reaching the expected mass. In saturating Ca, the peak is higher when SKp is in excess than when CaM is in excess, and there is a dip between the two peaks. Physically, this is consistent with the light scattering data only if the for the 2SKp/1CaM is stronger than it is for 1SKp/2CaM and if 1SKp/1CaM forms as the dominant species (>95% of active or “competent” protein) at 1:1 molar ratios with a very strong affinity. The stoichiometries determined from CG-MALS are fully supported by the AUC data. Because CG-MALS measurements are not affected by molecular shape, we can rule out virtually all exotic arguments that would explain the observations that the complex that forms at 1:1 molar ratios sediments with a smaller coefficient than when molar ratios are unequal; i.e., the complex at 1SKp:1CaM molar ratio sediments with a smaller coefficient because it is in fact smaller than the complexes that form at 1SKp:2CaM or 2SKp:1CaM. Any single approach for measuring the molar mass has experimental limitations and could lead to misinterpretations. Therefore we chose a multifaceted approach to gain confidence in our conclusions. Our results from different techniques are consistent enough to generalize our conclusions. Why is 2SKp/2CaM not resolved in our data? One explanation would be that the crystal packing environment favors intermolecular contacts that are not favored in a more physiological solution. Alternatively, perhaps the histidine tag altered the formation of the complex in earlier studies (). Tags in other proteins have led to some controversy. ERK2 is an extracellular regulated kinase that was believed to be capable of forming dimers, yet it was shown in solution that a histidine tag present in many early studies increases the presence of dimeric complexes of ERK2 (). Our approach is to avoid using tags in binding studies whenever possible. It is important to note that both crystal and solution studies may allow geometries of SKp/CaM that are not encountered when the full SK protein is embedded in lipid. We conclude from solution studies that the 1 to 1 stoichiometry for CaM to each SK binding site is more consistent with the traditional view of fourfold rotational symmetry of the homotetrameric channel and not the dimeric, twofold symmetry implied with the model derived from a crystal. Thus the basis for using a dimer of dimers model to describe SK gating may not be as relevant as a homotetramer model with fourfold symmetry (), which has been predominantly observed in other homotetrameric potassium channels (). An unexpected finding is that CaM may have chaperone-like properties with SKp. Although CaM causes SKp to aggregate if the final concentration of CaM is lower than SKp, when the concentration of CaM is greater than that of SKp, aggregation is prevented. When Ca is present, no aggregation occurs. CaM has been shown to prevent aggregation and degradation of some small proteins in a Ca-dependent manner (). CaM binding to ion channels senses both local and global changes to Ca (). CaM has been implicated in multiple functions on the same protein as exemplified in VDCCs (). Several structures of CaM bound to an IQ motif have been presented among various VDCC that led to various functional interpretations (; ; ; ). The structures show which residues of the peptides contact CaM in the presence of Ca in a crystal environment. However, we know that CaM must have other conformations or move to other sites in order for the channel to sample different functional states (). A static structure may miss dynamic motions or long-range allosteric effects on binding. Biochemical studies have also shown that residues from VDCC that do not contact CaM in the crystal can still affect Ca binding to CaM (). It is not unreasonable to infer that the level of complexity for functional CaM interactions with SK is equally as complex as CaM interactions with VDCC, especially because there are potentially more CaMs involved with SK. Our data show complexity in the interaction between SKp and CaM. If what we observe with CaM binding to SKp is representative of CaM binding to the full channel, this would suggest that an SK channel may present itself with a range of SK-to-CaM stoichiometry from perhaps 2 to 8 CaM molecules per channel tetramer. Our results can be used to facilitate the interpretation of more complicated findings from future investigations of full-length SK.

The voltage-insensitive calcium-activated potassium channel of intermediate conductance, KCa3.1, has been documented to play a prominent role in a large variety of physiological processes including immune reactions involving memory B and T cells (), transepithelial ion transport in Cl-secreting epithelial cells (), control of vascular tone (), and proliferation/migration in various cell types (; ; ). It follows that KCa3.1 is now recognized as a promising therapeutic target to treat life-threatening health disorders (; ; ; ). In line with this proposal are several reports demonstrating that a pharmacological activation of KCa3.1 and KCa2.3 channels constitutes a potential unique endothelium-specific antihypertensive therapy (; ). Of interest is also the observation that KCa3.1 activation could improve cAMP-induced Cl secretion in tissues coming from cystic fibrosis patients with partial CFTR function, thus identifying KCa3.1 activation as a complementary strategy to correct the basic ion transport defect in cystic fibrosis (). KCa3.1 channels are tetrameric membrane proteins with each subunit organized in six transmembrane segments, S1–S6, with a pore motif between segments 5 (S5) and 6 (S6). The channel assembly and trafficking are regulated by the constitutively bound calmodulin (CaM) molecule, which also confers Ca sensitivity (). The crystal structure of the KCa3.1 channel is not known. On the basis of homology modeling data and substituted cysteine accessibility method (SCAM) experiments, we proposed that residues V275, T278, A279, V282, and A286 were lining the channel pore with residues C276, C277, L280, and L281, oriented opposite to the pore lumen (; ). Additional experiments subsequently demonstrated that the bundle-crossing region located at the level of residues V282–A286 could not form a tight seal impermeable to K ions for the closed KCa3.1 channel. This conclusion came from experiments in which we showed that the accessibility of the thiol-modifying agent Ag to cysteines engineered in the channel cavity was independent of the channel-conducting state (). In agreement with these results, Ba-based experiments on the Ca-activated KCa2.2 channel demonstrated that the channel gate was most likely located deep in the channel central cavity (). In addition, data were presented indicating that negative gating of the KCa2.1 and KCa2.3 channels by NS8593 occurs through interactions with gating structures at a position close to the selectivity filter, deep in the pore inner vestibule (). It was also suggested that the selectivity region of KCa2.x channels represents a binding site for potentiators such as GW542573X and CM-TPMF, with the residue equivalent to L215 in the KCa3.1 S5 transmembrane helix playing a determinant role (, ). Therefore, these observations tend to support a model where the selectivity filter region constitutes a key determinant to the action of KCa3.1 regulators while demonstrating that residues distinct from the CaM-binding domain in the C terminus play a pivotal role in determining apparent Ca affinity and gating properties. However, channel gating is a dynamic process that involves concerted conformational changes in many parts of the channel protein. As Ca-dependent gating of KCa3.1 originates from the binding of Ca to CaM in the C terminus, the hypothesis of a gate located at the level of the selectivity filter requires that the conformational change initiated in the C terminus be transmitted to the S6 transmembrane helix and to the channel pore helix directly connected to the selectivity filter. The exact molecular mechanism underlying KCa3.1 opening in response to Ca binding to the CaM–KCa3.1 complex in the C terminus remains to be elucidated. Structural information pertinent to channel gating was, however, obtained through the crystallization of CaM bound to the rat KCa2.2–CaM-binding domain in the presence of Ca (). On the basis of this structure, it was proposed that a large-scale conformational rearrangement is taking place in the presence of Ca, where the N-lobe of CaM binds to a C-terminal segment of an adjacent channel monomer resulting in a dimerization of contiguous subunits. This rearrangement would in turn lead to a rotation/translation of the associated S6 transmembrane domains and to the opening of the ion-conducting pore (, ; ; ). As the CaM-binding domain of KCa3.1 is directly connected to the S6 transmembrane helix, activation of the channel gate at the level of the selectivity filter could depend upon the coupling between each of the channel pore helices and the associated S6 transmembrane segment. SCAM experiments have already provided evidence for widespread conformational changes in the pore helix of the cyclic nucleotide–gated channel during gating (). These observations were interpreted as evidence for a gate localized at the selectivity filter of the channel (). Mutations that perturb the pore helix of the Kir6.2 channel were similarly reported to affect the fast gating kinetics of the channel (). The importance to gating of the interactions between the selectivity filter and the pore helix in channels was also confirmed in a study on KirBac3.1, which showed that changes at the level of the cytoplasmic domain were relayed to the selectivity filter (). Similarly, changes in the conductivity of the KirBac selectivity filter were found to be correlated with remote conformational changes of the cytoplasmic domains, in agreement with the selectivity filter playing a role in gating by directly responding to global conformational changes of the channel (). Finally, work on the KcsA channel has led to the conclusion that C-type inactivation could arise from mechanical deformations propagating through a network of steric contacts between the channel inner helical bundle (equivalent to T278 KCa3.1) and the C-terminal end of the pore helix (equivalent to L249–T250 KCa3.1) (). Collectively, these observations suggest that residues located deep in the KCa3.1 channel pore, close to the selectivity filter, could not only contribute to the CaM–KCa3.1 complex to gating but also represent target sites for a pharmacological control of KCa3.1 activity. How the selectivity filter region in KCa3.1 is involved in gating remains unknown. However, one of the distinguishing features of the KCa3.1 Ca dependence is that Pomax, the channel open probability at saturating Ca concentrations (>20 µM), remains low, typically 0.1–0.2 for the wild-type channel, in contrast to a Pomax of 0.8 reported for the KCa2.2 channel in the high activity mode (). The observation of a low Pomax at saturating Ca conditions argues for the binding of Ca to the CaM–KCa3.1 complex, promoting the formation of a closed-state configuration from which the channel transits to an open configuration. Such behavior has been documented in numerous ligand-gated channels where the agonist does not alter the open state but brings the channel to a preopen configuration. It is likely that the setting of Pomax depends on the energetics governing the interactions taking place at the selectivity filter level. In this study, we tested the hypothesis that interactions of the KCa3.1 pore helix with the S5 and S6 transmembrane segments contribute in setting Pomax. More specifically, we intended to identify which residues of the pore helix are involved in functionally coupling the pore helix to the S5 and S6 transmembrane segments during gating. Our results demonstrate that important changes in Pomax can be obtained by modulating aromatic–aromatic interactions between F248 of the pore helix and W216 of the S5 transmembrane helical segment, and/or by perturbing the interactions between F248 and the S6 transmembrane segment at the level of the G274 hinge residue. A first class of 3-D structures for the KCa3.1 pore region (F190–R287) was generated through homology modeling () using as template the crystal structure of the Kv1.2 channel (Protein Data Bank accession no. ). The choice of the Protein Data Bank structure was essentially based on the analysis provided by the I-TASSER server, which identified Protein Data Bank as the template with the best TM score of 0.869, knowing that a TM score >0.5 corresponds approximately to two structures of similar topology (). The model includes part of the S4–S5 cytosolic linker plus the segment extending from the N terminus of the S5 transmembrane helix to the C terminus of the S6 transmembrane helix. A second class of models was obtained from the SAM-T08 server, which identified the Protein Data Bank accession number (bacterial cyclic nucleotide–regulated channel) structure as the top-ranking template (). Both templates were thus used to generate models of the KCa3.1 channel in the open (Protein Data Bank ) and closed (Protein Data Bank ) configuration. MD simulations were performed using the CHARMM-CGENFF force field with the channel pore region incorporated into a DPPC lipid bilayer in contact at top and bottom with an identical explicit water medium. Channel incorporation was accomplished according to the procedure implemented in the Charmm-Gui package (). Overall, the system contains 55,732 atoms including 161 DPPC lipid molecules, 9,576 TIP3P model water molecules, plus 19 K and 33 Cl ions to ensure electroneutrality at near physiological concentration. Cut-on and cutoff parameters needed to define nonbonded interactions were set to 10 and 12 Å, respectively, and SHAKE constraints were used to fix lengths of bonds involving hydrogen atoms. The system was equilibrated according to the six-cycle scheme implemented in the Charmm-Gui output for a total equilibration period of 500 ps. Trajectories were generated for 32–64 ns using a time step of 2 fs, and electrostatic and van der Waals interaction energies were computed from trajectories sampled at 0.2 ns. MD simulations were performed for a system at constant pressure (1 atm) and constant temperature (300°K). Solvent accessibility surface areas (SASAs) were averaged over a 64-ns trajectory using a standard probe of 1.6 Å. Wild-type and mutant KCa3.1 channels were expressed in oocytes by injection of their RNAs transcribed in vitro from pT7TS vector, which contains 5′ and 3′ untranslated regions of β-globin mRNA. Approximately 1–10 ng KCa3.1 RNA was injected into each oocyte. Recordings were performed 1–7 d after injection. All the mutants were generated using the Q5 Site-Directed Mutagenesis kit (New England Biolabs, Inc.) and verified by sequencing at the Université de Montréal genomics platform. Before patch clamping, the defolliculated oocyte was briefly incubated in a hyperosmotic solution containing (mM) 250 KCl, 1 MgSO, 1 EGTA, 50 sucrose, and 10 HEPES, buffered at pH 7.4 with KOH and the vitelline membrane peeled off using fine forceps. The oocyte was then transferred to a superfusion chamber for patch-clamp measurements. The bath and patch pipette solutions contained (mM) 200 KSO, 1.8 MgCl, 0.025 CaCl, and 25 HEPES, buffered at pH 7.4 with KOH (referred to 200 mM KSO). Sulfate salts were used to minimize the contamination from endogenous Ca-dependent chloride channels while enabling the chelation of contaminant divalent cations such as Ba. Calcium-free solutions were prepared by adding 1 mM EGTA to 200 mM KSO solutions without CaCl. Bath solution changes were performed as described previously using a rapid solution changer system (RSC-160; Biological) (). The solution exchange time was <10 ms. bis(2-mercaptoethyl)sulfone (BMS; EMD Millipore) was directly dissolved in the 200-mM KSO solution at a concentration of 8 mM. Multiple- and single-channel inside-out recordings were performed using an amplifier (Axopatch 200A; Molecular Devices). Patch pipettes were pulled from borosilicate capillaries using a pipette puller (model PP-83; Narishige), which was used uncoated. The resistance of the patch electrodes ranged from 2 to 5 MΩ. Data acquisition was performed using a Digidata 1320A acquisition system (Molecular Devices) at a sampling rate of 1.0 kHz unless specified otherwise. Single-channel analysis was carried out using the QUB package (, ). Dwell-time measurements were performed on data that were idealized according to the segmental-k means method based on a hidden Markov model–type analysis. Incorporation of unnatural amino acids was performed according to the procedure described by . In brief, we used a pU6-pMpa plasmid (provided by P.G. Schultz, Scripps Research Institute, La Jolla, CA), which encodes two genes: Tyr-tRNA (with anti-codon mutated to CUA, complement to the TAG stop codon) and Tyrosyl-tRNA-synthetase, which was mutated in Y37V/D182S/F183M to make the plasmid available to specifically recognize the unnatural amino acid -methoxy--phenylalanine (pMpa). The plasmid was coexpressed with the KCa3.1 channel mutated at position F248 into an amber stop codon (TAG) in oocytes, in which 5 mM pMpa (Sigma-Aldrich) had been injected. Statistical significance was analyzed using unpaired Student’s test. P < 0.05 was considered statistically significant. Data are expressed as mean ± SD. Fig. S1 shows a Western blot of full-length expression of a KCa3.1 channel tagged with 6-His at position 132 in the external linker S1–S2, where residue F248 has been replaced by the unnatural amino acid pMpa. Fig. S2 presents a Western blot obtained for the F248W–W216F and W216F mutants, confirming channel expression. Fig. S3 presents the open and closed dwell-time distributions computed for the wild-type, F248A, and F248W KCa3.1 channels. The online supplemental material is available at . The proposed structures for the 3BEH-based (bacterial cyclic nucleotide–regulated channel) and 2A79-based (Kv1.2 channel) KCa3.1 channel pore region are presented in . The selectivity filter is seen to be extending from T250 to T260 and is connected to the pore helix comprising amino acids G235 to L249. In both models, the residues V275, T278, V282, and A286 (blue) are predicted to line the channel pore, in agreement with our previous SCAM studies (). An analysis of the distance between the Cα of corresponding residues along S6 on opposite subunits indicates that the narrowest part of the conduction pathway should be located at the level of V282, with a van der Waals pore diameter of ∼4.5 Å for the 3BEH-based model compared with >12 Å for the model derived from the 2A79 template. It follows that the structure obtained from the 3BEH template is likely to be more representative of the KCa3.1 channel in the closed conformation relative to the 2A79-based model, which better accounts for the KCa3.1 open configuration. Both models predict that residues I244 and F248 of the pore helix should be projecting between the S5 and S6 transmembrane helices, whereas L243 should be oriented toward the selectivity filter in proximity of V256 at the C-terminal end of the selectivity filter region. The proposed configuration for the pore helix was investigated in experiments where the spatial proximity of the L243 and V256 residues was tested through disulfide bond formation between Cys engineered at L243 and V256, respectively. The inside-out patch-clamp recordings presented in confirmed in this regard that channel activity of the L243C–V256C double mutant was sensitive to the addition of the small reducing agent BMS to the bathing solution. These observations confirmed that disulfide bond formation between L243C and V256C tends to stabilize the channel closed state configuration by constricting the selectivity filter. This effect was specific to the double mutation L243C–V256C system, as demonstrated in the multi-channel (L243C) and single-channel (V256C) recordings presented in where channel activity appeared insensitive to BMS application for single mutation channels. A computer analysis was next undertaken to determine which residues of the KCa3.1 pore helix form the interface with the S5 and S6 transmembrane segments. Our approach consisted of computing for each residue of the pore helix the difference in SASA (ΔSASA) with and without the S5 and S6 helices. It is expected that KCa3.1 residues not located at the interface formed by the pore helix with the S5 and S6 transmembrane segments will have SASA values unaffected by the presence of the S5 or S6 helix (ΔSASA = 0). In contrast, residues of the pore helix contributing to the interface with the S5 and S6 helices should be characterized by a reduction of their SASA values when calculations include either the S5 or S6 transmembrane segment. summarizes the predicted contribution of each pore helix residue to the interface, with the S5 and S6 transmembrane helices averaged over a 64-ns MD simulation. As seen, the largest contact areas with both the S5 and S6 helices involve residue F248 at the C-terminal end of the pore helix regardless of the template used. In contrast, residues T240, L241, and I244 of the pore helix appeared to contribute more to the S5 than S6 transmembrane segment interface, with ΔSASA values remaining overall smaller relative to F248. Notably, calculations for residues L249 and W242 revealed contact areas almost entirely limited to the S6 transmembrane helix, an effect coming from W242 and L249 contributing to intersubunit contacts. As expected, ΔSASA values for L243 and I246 predicted to be oriented facing the selectivity filter were not affected by the presence of the S5 and S6 helices, indicating small if any interface contact surfaces. Collectively, this analysis showed that the largest contribution to the pore helix interface with both the S5 and S6 transmembrane segments is coming from residue F248. Because the CaM–KCa3.1 complex is directly connected to the S6 transmembrane helix, a similar procedure was applied to identify residues of the S6 segment contributing to the interface with the pore helix. The calculations presented in show that residues T278, C277, V275, G274, V272, G271, T270, and L268 in S6 are susceptible to interact with the channel pore helix in both models. Interactions involving V275 and T268 appeared unique, however, as they represent intersubunit contacts exclusively. Of interest, we noted that residue V275 can interface with L249 at the C-terminal end of the pore helix on an adjacent subunit. A close proximity between V275 and an adjacent pore helix would be in agreement with the observed strong channel inhibition by MTS reagents of the V275C mutant channel (), and with the contribution of V275 to the TRAM-34–blocking action (). Detailed representations of the interaction network between F248 and residues in S5 and S6 derived either from the 3BEH (bacterial nucleotide–activated potassium channel; A and C) or 2A79 (Kv1.2; B and D) templates are presented in . In both models, the side chains of T270, G274, and C277 of the S6 transmembrane helix are seen to be oriented toward the pore helix at the level of F248 (). However, the interaction pattern of the pore helix–S5 residues was found to be model dependent, with T212 and W216 facing F248 in the 3BEH-based model (), in contrast to the structure derived from the Kv1.2 (2A79) template where F248 essentially makes contact with L211 of S5 (). A detailed analysis of the nonbonded van der Waals plus electrostatic energy between F248 of the pore helix and residues extending from L268 to L280 along the S6 transmembrane segment is presented in . The most important interactions were seen between F248 and the glycine hinge G274, arguing for F248 being within close proximity of the G274 backbone. Strong interactions were similarly estimated between F248 and T270, G271, M273, C277, and T278. These predictions appeared template independent and could be obtained for the 3BEH- and 2A79-based models. Interaction energy between residue F248 and residues of the S5 transmembrane helix is presented in . The resulting interaction pattern appeared in this case model dependent, with F248 favorably interacting with W216, T212, and L215 in the 3BEH-derived structure, and with L211 for the Kv1.2-based open configuration. The previous analysis has led to the identification of residues predicted to line the interface formed by the pore helix with the S5 and S6 transmembrane segments. An alanine scan was thus performed to determine how important these residues are in setting Pomax. Typical single-channel recordings are presented in . We note that the mutation F248A resulted in a Pomax of 0.75 ± 0.10 ( = 3) compared with 0.22 ± 0.07 ( = 8) for the wild-type channel (, WT). The mutation T240A led to a less drastic increase in Pomax with a value of 0.42 ± 0.08 ( = 4) compared with I244A, which resulted in a Pomax decrease to 0.10 ± 0.04 ( = 4). The mutation T250A had a minor impact on the channel Pomax but resulted in an important decrease in the channel unitary conductance, from 30 pS for KCa3.1 wild type to 9 pS for the T250A mutant (not depicted). Collectively, these results show that the mutation F248A is unique, as it caused a 3.2-fold increase in Pomax. To determine if hydrophobicity and/or volume of the residue at position 248 control Pomax, experiments were performed in which F248 was substituted by residues differing in size and/or hydrophobicity. Typical unitary current traces are presented in . These recordings demonstrate that among all the mutants tested, F248T only yielded Pomax values (0.19 ± 0.05; = 3) not superior to wild type (F248, 0.22 ± 0.07; = 8). Clearly, there was no direct correlation between Pomax and the volume of the substituting residue, as replacing F248 by a smaller (F248A) or larger amino acid (F248W) resulted in both cases in an increased Pomax relative to wild type. In addition, we noted that although Val (V) and Thr (T) are of roughly the same shape and volume, there was a significant difference in channel activity, with Pomax of 0.77 ± 0.05 ( = 3) and 0.19 ± 0.05 ( = 3) for the F248V and F248T mutants, respectively. We concluded that the substitution of a methyl (V) by a hydroxyl group (T) modified the interaction energy pattern involving residue 248 as to cause a decrease in Pomax value. The results of this analysis are summarized in , which shows that the overall Pomax ranking reads T ≈ F < H < Y < S < A < V < L ≈ W. A scatter plot of Pomax as a function of residue 248 volume size is presented in . This analysis reveals two distinct sets of data. Whereas there was a modest but significant increase in Pomax as a function of volume size for Ala, Val, and Leu, drastic changes were observed with the aromatic (Phe, Tyr, His, and Trp) residues. Clearly, increasing the size of the aromatic side chain engineered at position 248 resulted in higher Pomax values. Importantly, aromatic residues such as Phe and Tyr showed significant lower Pomax values compared with nonaromatic amino acids of similar sizes such as Val or Leu, suggesting specific effects related to the presence of aromatic rings in these cases. To confirm that the size of aromatic residues at 248 is a key determinant to Pomax, experiments were undertaken in which we used the unnatural amino acid pMpa, a tyrosine analogue obtained by substituting the hydroxyl moiety of Tyr by a O-CH group. The resulting aromatic amino acid has an estimated volume of 227 Å compared with 192 Å for Tyr and 203 Å for Trp. Single-channel recordings confirmed that the substitution of F248pMpa caused an important increase in Pomax, with a mean value of 0.94 ± 0.03 ( = 4), in agreement with the proposal of a strong correlation between Pomax and the volume of the aromatic residue engineered at 248 (see ). The observation of a lower Pomax value for aromatic compared with nonaromatic residues engineered at 248 could be indicative of specific interactions involving aromatic–aromatic interactions potentially contributing to the energy barrier controlling channel opening. Aromatic amino acids are known to play a prominent role in determining protein structure through π–π or hydrogen–π interactions. For polar hydrogen–π interactions, the donors are hydrogen atoms connected to electronegative atoms (for instance, R-OH, R-NH), and the acceptors may be represented by various aromatic molecules and conjugate π groups (). More importantly, we note that the 3BEH-based model of KCa3.1 illustrated in shows possible face-edge interactions between F248 and W216 of the S5 transmembrane helix. As Tyr, Trp, His, and Phe constitute polar hydrogen acceptors, which can potentially interact with a polar hydrogen donor such as Trp (NH) (), the effect on Pomax seen with the substitutions F248Y/H/W could be indicative of polar hydrogen–π interactions and/or π–π interactions between the residue 248 in the pore helix and W216 in S5. Such interactions would be absent with the F248A/V/L mutations, leading to higher Pomax values as seen experimentally. As Trp can also behave as a hydrogen acceptor in polar hydrogen–π interactions, a strong interaction between F248T (hydrogen donor) and W216 could explain the unexpected observation of F248T leading to a smaller Pomax compared with F248A despite a larger side-chain volume. However, this interaction would be size dependent, as the Pomax for the F248S mutant is similar to the Pomax of F248A. It follows that substituting W216 (hydrogen donor) by nonaromatic residues such as Ala or Leu should impair aromatic–aromatic interactions involving F248 and reproduce the effects observed when F248 was substituted by a nonaromatic amino acid. Examples of single-channel recordings obtained with the W216A and W216L mutants are illustrated in . Clearly, replacing W216 by a nonaromatic residue caused a drastic increase in Pomax, with values of 0.68 ± 0.04 ( = 3) and 0.95 ± 0.01 ( = 3) for W216A and W216L, respectively (see ). The effect observed with W216A cannot be attributed to a decrease of the side-chain volume, as the substitution W216L led to a higher Pomax value relative to W216A. Collectively, these results support a model in which polar hydrogen–π and/or π–π interactions involving residue 248 of the pore helix and W216 on S5 () contribute to stabilize the gate closed configuration, while confirming the prominent role of the selectivity filter region to the KCa3.1 gating process. In line with this proposal is the observation that substituting W216 by the smaller Phe aromatic residue (W216F) resulted either in the absence of detectable channel activity (50% of the 25 oocytes tested) despite Western blot evidence for channel expression (see ), or in single-channel records where the channel gate switched within <3 min in 50% of the recordings considered, from a high Pomax state (0.7 ± 0.2; = 3) to a permanent closed configuration (). Such behavior was not seen when W216 was mutated to a nonaromatic residue such as Ala or Leu (). These observations are compatible with aromatic–aromatic interactions involving F248 and W216F significantly increasing the likelihood of forming a stable closed state. Notably, experiments in which the Phe at 248 and Trp at 216 were interchanged (F248W–W216F) failed to yield detectable channel activity despite biochemical evidence of channel expression (see Fig. S2). The structural model and the interaction energy diagram illustrated in and , respectively, point toward T212 and L215 in S5 as potential residues susceptible to be within the interaction range of F248. The contribution of these two residues in setting Pomax was investigated in single-channel experiments using the T212A and L215A mutant channels. In contrast to the mutation W216A, the substitution T212A or L215A failed to significantly affect channel activity, with Pomax values of 0.28 ± 0.15 ( = 3) and 0.19 ± 0.04 ( = 3), respectively (). These observations do not support a model whereby OH–π interactions involving T212 and F248 would participate in the control of Pomax. In contrast to the model derived from the 3BEH template, the model based on the Kv1.2 (2A79) structure predicts strong interaction energy between F248 and L211, and to a lesser extent with L215. The importance of these interactions in setting Pomax was investigated in inside-out patch-clamp experiments where L211 was substituted by residues of variable volume size and hydrophobicity. The results of these experiments are summarized in . On the basis of the model derived from the Kv1.2 template, one would expect a potential stabilization of the channel open state by favoring π–π or OH–π interactions between F248 and Phe (F), Trp (W), or Thr (T) residues engineered at 211. Our results indicate a small increase in Pomax with the L211F mutant relative to wild type, whereas the mutations L211T and L211W appeared ineffective in modifying Pomax. At best, we found a correlation between the volume size of the residue engineered at 211 and Pomax, with a Pearson correlation coefficient of 0.75 ± 0.15 (). A stronger correlation was obtained between Pomax and the solvation energy of the residue engineered at 211 with a Pearson coefficient of 0.84 ± 0.07 () (). Collectively, these observations suggest that the interactions between F248 and the S5 transmembrane helix are essentially governed by W216 for the channel in the closed configuration (3BEH-based model), with no significant contribution from L211, T212, and L215. The results presented in indicate that F248 can interact preferentially with the Gly hinge at 274 and with residues T270, M273, C277, and T278 of the S6 transmembrane helix. MD simulations further suggest that the strong interactions with G274 arise in part from van der Waals contacts indicating a very close proximity between the G274 backbone and the F248 side chain. As the CaM–KCa3.1 complex is directly connected to the S6 transmembrane segment, a double mutant cycle analysis was performed to determine how these interactions could contribute to Pomax. In this analysis, the coupling energy ΔΔG was computed as (ΔG + ΔG) − (ΔG + ΔG), with ΔG given by kT ln(Pomax), with k the Boltzmann constant and T the temperature (; ). Examples of cycle analyses are shown in . We note first the absence of coupling with either F248 or F248W for the C277A, M273A, and T270A mutants with ΔΔG of 0.02 ± 0.28 kcal/mol, −0.27 ± 0.30 kcal/mol, and −0.01 ± 0.10 kcal/mol for the F248W–C277A, F248W–M273A, and F248W–T270A double mutant cycle systems, respectively (not depicted). In addition, the observation that the mutations T270A and T270S were ineffective in modifying Pomax when added to the F248 or F248W template (not depicted) was seen as evidence for T270 not being involved in OH–π-type interactions with F248 or F248W. Strong ΔΔG of −1.04 ± 0.26 kcal/mol and −1.15 ± 0.56 kcal/mol was obtained, however, with the F248W–M273W and F248W–C277W double mutants (). No current could be detected with the T270W mutant (not depicted). Of interest, the mutation F248W failed to cause an increase in Pomax when added to the C277W template, an effect not seen with C277A (not depicted). These observations suggest potential aromatic–aromatic interactions between F248W and C277W, as seen with W216 (see ). We also note that the mutation M273W () led to a Pomax of 0.95 ± 0.005 ( = 3) with F248 and to 0.63 ± 0.05 ( = 4) when F248 was substituted by a Trp (F248W–M273W). With a Pomax of 0.86 ± 0.07 ( = 5) for F248W–M273, 0.80 ± 0.10 ( = 3) for F248W–M273A (not depicted), and 0.63 ± 0.05 ( = 4) for F248W–M273W (), there is no apparent correlation between the volume size of residue 273 and Pomax in conditions where F248 has been mutated into a Trp. Such behavior was not found when mutations of M273 were performed on the wild-type KCa3.1 template (F248) with a Pomax of 0.22 ± 0.02 ( = 3) and 0.95 ± 0.005 ( = 3) for the F248–M273A (not depicted) and F248–M273W () channels, respectively. Clearly, the presence at 248 of a Phe with a smaller side-chain volume compared with Trp caused the channel Pomax to become more sensitive to the volume of the residue substituting for M273. These results confirm the close proximity between F248 and M273, as predicted from the homology-based representations of the KCa3.1 open and closed configurations (see ). e r e s u l t s p r e s e n t e d h e r e d e m o n s t r a t e t h a t a r o m a t i c – a r o m a t i c i n t e r a c t i o n s b e t w e e n F 2 4 8 o f t h e p o r e h e l i x a n d W 2 1 6 o f t h e S 5 t r a n s m e m b r a n e s e g m e n t p l a y a c r u c i a l r o l e i n s e t t i n g t h e c h a n n e l m a x i m u m o p e n p r o b a b i l i t y , w h i l e s u p p o r t i n g t h e p r o p o s a l o f a n a c t i v e g a t e l o c a t e d a t t h e l e v e l o f t h e s e l e c t i v i t y f i l t e r . I t i s a l s o s h o w n t h a t F 2 4 8 i s w i t h i n t h e i n t e r a c t i o n r a n g e o f C 2 7 7 a n d M 2 7 3 a d j a c e n t t o t h e G l y h i n g e o f t h e S 6 t r a n s m e m b r a n e s e g m e n t a n d c o u l d t h u s c o n t r i b u t e t o t h e f u n c t i o n a l l i n k b e t w e e n t h e C a M – K C a 3 . 1 c o m p l e x a n d t h e p o r e h e l i x . T h e s e r e s u l t s s u g g e s t t h a t t h e c h a n n e l c l o s e d c o n f i g u r a t i o n i s s t a b i l i z e d b y a r o m a t i c – a r o m a t i c i n t e r a c t i o n s b e t w e e n F 2 4 8 a t t h e C - t e r m i n a l e n d o f t h e p o r e h e l i x a n d t h e S 5 t r a n s m e m b r a n e s e g m e n t , a n d s u p p o r t a m o d e l w h e r e t h e i n t e r f a c e b e t w e e n t h e p o r e h e l i x a n d t h e S 5 t r a n s m e m b r a n e s e g m e n t c o n s t i t u t e s a p o t e n t i a l t a r g e t s i t e f o r t h e d e s i g n o f K C a 3 . 1 p o t e n t i a t o r s .

Sunlight is crucial for life and has many beneficial effects, but, at the same time, the UV radiation (UVR) contained by sunlight is the most common environmental carcinogen (; ). Unlike other mammals that have fur to protect their skin, human skin is constantly exposed to solar UVR (280–400 nm) and is susceptible to its damaging effects, primarily skin cancers and photoaging. Human skin also has a unique protection mechanism against UVR: the presence of melanocytes in the epidermis allows skin to respond to UVR by increasing its pigmentation. Because UVR is omnipresent and is able to interact with human skin, identifying the molecular pathways that allow human skin to detect and elicit an immediate response to UVR is critical for developing new photoprotective methods. How does human skin detect UVR? UVR consists of photons; photons can activate G protein–coupled opsin receptors (GPCRs) in the eye that elicit cellular responses through the activation of different G proteins and downstream effectors. Gα is used by vertebrate photoreceptors (), whereas Gα mediates phototransduction () and non-image forming vision in the mammalian retina (; ; ). Activation of Gα pathways leads to stimulation of phospholipase C β (PLCβ), which induces hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP) into diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP). Changes in the levels of PIP, DAG, and IP modulate the activity of many proteins, including transient receptor potential (TRP) ion channels. We recently characterized a retinal-dependent UVR-sensitive phototransduction pathway in human epidermal melanocytes (HEMs) that is G protein and PLCβ dependent and results in the activation of TRP subfamily A1 (TRPA1) channels; activation of this pathway results in a rapid increase in intracellular Ca ([Ca]) and increased cellular melanin content (; ). In this study we investigated the G protein that mediates this pathway and the downstream molecular events. We found that UVR phototransduction in HEMs is mediated by Gα signaling, and provide evidence for a phosphoinositide cascade involving IP-mediated intracellular Ca release via IP receptors (IPR) and PIP regulation of Ca-permeable TRPA1 ion channels. The two sources of Ca have different dynamics and, combined, result in a Ca response with a fast rising phase and a slow decay. Our results demonstrate that UVR phototransduction in HEMs activates a Gα-dependent signaling pathway similar to well-characterized phototransduction pathways in the eye. Cholera toxin (CTX), pertussis toxin (PTX), HC-030031, 1-oleoyl-2-acetyl--glycerol (OAG), phosphatidylcholine phospholipase C (PC-PLC; from ), polylysine (PolyK, 70–150 kD), heparin, and ionomycin were purchased from Sigma-Aldrich. Endothelin, GPAnt-2, GPAnt-2a, and Xestospongin C (XeC) were from Tocris Bioscience. mSIRK and L9A were from EMD Millipore. DiC8-PIP2 was from Echelon Biosciences. Stocks of all reagents in water, DMSO, or ethanol were stored at −4°C or −20°C until use and diluted to the final concentration to contain <1% solvent. For Ca imaging experiments, HEMs were preincubated with pharmacological reagents for 3–15 min, with the exception of PTX and CTX, which used 24 h incubations. Primary HEMs isolated from neonatal foreskin were cultured in Medium 254 containing Human Melanocyte Growth Supplement (HMGS2; Cascade Biologics/Invitrogen) and 1% penicillin-streptomycin (Invitrogen), and propagated for a limited number of cell divisions (≤15). Vitamin A or retinoid derivatives are not components of either Medium 254 or HMGS2. Human embryonic kidney (HEK293) cells were cultured in Dulbecco’s Modified Eagle Medium and F12 nutrient mixture (DMEM-F12; Gibco/Invitrogen) containing 10% fetal bovine serum (Atlanta Biologicals) and 1% penicillin-streptomycin (Invitrogen). MicroRNAs (miRNAs) were designed and expressed in HEMs using a lentiviral system, as described previously (). BLOCK-iT miRNA oligos (Invitrogen) were cloned into the pcDNA6.2-GW/EmGFP-miR expression vector modified to contain mCherry instead of EmGFP to allow for simultaneous fluorescence detection and Fluo-4–based Ca imaging. miRNAs were recombined from pcDNA6.2-GW into pDONR221 and pLENTI6/V5-DEST vectors (Invitrogen) for lentiviral production. Lentiviral particles containing miRNA were obtained as described previously (). HA-tagged RGS2 (Missouri S&T cDNA Resource Center) was recombined into pDONR221 and pLENTI6/V5-DEST vectors (Invitrogen) for lentiviral production as described previously (). The mRNA expression level of Gα, Gα, or Gα in control or targeted miRNA-treated cells was determined ≥7 d after infection using comparative C quantitative PCR (qPCR). Total RNA was extracted from infected HEMs using the RNeasy Plus kit (QIAGEN) and converted to cDNA using RT-PCR (SuperScript III; Invitrogen). qPCR reactions were prepared according to manufacturer instructions using Power SYBR green. All reactions were done in triplicate and actin was used for normalization. Expression of HA-tagged RGS2 in HEMs was confirmed via Western blotting. Cells were homogenized ≥10 d after infection in ice-cold RIPA buffer (Thermo Fisher Scientific) containing protease inhibitor cocktail (Roche). Samples were agitated at 4°C for 30 min and then centrifuged at 16,000 for 30 min at 4°C. Protein content was determined using the Pierce BCA Protein Assay kit (Thermo Fisher Scientific). Equal amounts of protein were loaded onto each lane, separated by electrophoresis on NuPAGE Bis-Tris gels (Invitrogen), and transferred to PVDF membranes (Roche). Membranes were blocked at room temperature for 1 h and incubated overnight at 4°C with rat monoclonal anti-HA antibody clone 3F10 (1:500; Roche), followed by 1 h at room temperature with HRP-conjugated goat anti–rat IgG affinity-purified antibody (1:5,000; EMD Millipore). Antibodies were detected using the SuperSignal West Femto enhanced chemiluminescence system (Thermo Fisher Scientific) and imaged using autoradiography film (Thermo Fisher Scientific). Ultraviolet light stimulation of cultured HEMs was conducted using a 200 W Hg-Xe arc lamp with converging optics and appropriate filters (). A dichroic mirror (260–400 nm) was used in combination with 280-nm long pass and a 400-nm short pass filters (Newport). The levels of light lost due to scattering by imaging buffer were negligible. Physiological doses of UVR were applied by varying the duration and/or power of the pulse. A hand-held silicon detector was used to measure power (Newport). Ca imaging was performed as described previously (; ). Cultured HEMs plated on glass coverslips were incubated for 15 min at room temperature in Ringer’s solution with 2 µM Fluo-4 (Molecular Probes/Invitrogen) and 250 µM sulfinpyrazone (Sigma-Aldrich), followed by dark incubation for 15 min with 12 µM of 9-cis or all-trans retinal (Sigma-Aldrich). Imaging was performed in modified Ringer’s extracellular solution containing (in mM): 150 NaCl, 1.8 CaCl, 1.2 MgCl, 10 D glucose, 25 HEPES, pH 7.4, and 310 mOsm/liter. Fluorescence images were acquired every 2 s, and 2 µM ionomycin (Sigma-Aldrich) was added at the end of some experiments to elicit a maximal Ca response used for normalization. The fluorescence intensity of individual cells (measured using ≥25% of the cell area, F) was quantified using MetaMorph (Molecular Devices) and MATLAB (MathWorks) and plotted in Prism 6 (GraphPad Software). F values were normalized as F = (F − F)/(F − F), where F is maximal fluorescence with ionomycin, and F is baseline fluorescence averaged from ≥15 data points acquired before light stimulation. Final data values for each dish were obtained by averaging F values from individual cells. In experiments where ionomycin was not used for normalization, Fluo-4 fluorescence intensities were quantified as ΔF/F(t) = [F(t) − F]/F and averaged as described previously. ΔF/F values were plotted as a function of time and fitted with a single-exponential function in Prism 6 (GraphPad Software) to calculate decay time constants. Ca response initial slopes were calculated over the first 30 s after UVR stimulation. Paired Ca imaging experiments were used when the dish-to-dish variability was significant. Cells for each of the paired experiments were plated on glass coverslips and treated identically by incubation in the same Fluo-4 and retinal solution, then imaged sequentially in alternating order: control followed by experimental condition or experimental condition followed by control. The averaged fluorescence intensities of cells from one coverslip measured in each condition were plotted as the two experimental values connected by the dotted line. Statistical significance of paired experiments was evaluated using a paired Student’s test. When pairing is not mentioned, the cells were cultured and treated identically, but we did not strictly alternate control and experimental condition measurements. In such cases, we averaged the values from all the control and all the experimental conditions and represented them as bar graphs. Electrophysiology experiments were performed as described previously (). All-trans retinal was stored, solubilized, and applied as described previously (). Experiments were performed under dim-red or infrared illumination. Whole-cell patch clamp recordings were carried out using micropipettes with 3–6 MΩ resistance at room temperature using an EPC 10 amplifier (HEKA) with PatchMaster software (HEKA), filtered at 2.9 kHz and digitized at 20 kHz. Experiments were performed using modified Ringer’s solution (see “Calcium imaging”). Unless stated otherwise, internal pipette solution contained (in mM): 140 CsCl, 1 MgCl, 4 MgATP, 10 EGTA, 10 HEPES, pH 7.2, and 290 mOsm/liter. The low EGTA solution used in contained 20 µM EGTA. UVR-induced currents were measured using a step protocol consisting of a step from a holding potential of −60 mV to +80 mV immediately before UVR exposure. Current values were calculated by subtracting initial current at +80 mV (I) from maximal current after UVR exposure: I = I − I. All recordings were inspected for baseline drift before analysis. In most recordings, the baseline did not drift significantly. If the baseline drift was >20% of the UVR response, cells were excluded. Cell membrane capacitance values were used to calculate current densities. Current–voltage (I-V) relations were established using voltage step protocol from the −60-mV holding potential to voltages between −80 mV and +80 mV in 20-mV increments. HEMs are very difficult to patch clamp. Our experiments are further complicated by the fact that UV exposure often causes us to lose patches. Because UV photocurrents are small and exhibit strong outward rectification (), most of our voltage-clamp experiments use positive membrane voltages that are not physiological (+80 mV) in order to increase current amplitude. Although current amplitude is higher at positive membrane voltages, current kinetics are significantly altered compared with recordings carried out at more physiological membrane voltages (). In addition, positive voltages are not well tolerated by these cells, preventing us from recording and plotting the current over a long period of time for every single trace. We have previously recorded currents over longer periods using voltage pulses () and at more negative voltages (), and showed that the current returns to baseline after the UV pulse. Our success rate, even for the short periods of UV irradiation, is well below 10%. Consistent with the difficultly of these experiments, the large noise is likely caused by the relatively low seal resistance (∼1 GΩ) and patch stability. Experimental data are presented as mean ± SEM, where refers to the number of dishes for imaging data or the number of cells for electrophysiology. We calculated p-values by unpaired or paired Student’s test and considered results significant when P ≤ 0.05. Fig. S1 shows that RGS2 expression in HEMs reduces Ca responses to endothelin. Fig. S2 shows that HEMs expressing Gα-, Gα-, or Gα-targeted miRNA have reduced Ca responses to endothelin. Fig. S3 shows that neither IP-mediated Ca release nor DAG activate the UVR photocurrent. Fig. S4 shows selective inhibition of UVR phototransduction signaling components versus ion channels by cellular dialysis. Online supplemental material is available at . We have recently shown that exposure to physiological doses of UVR activates a retinal-dependent phototransduction pathway that is mediated by G protein activation (; ). However, we have yet to identify the type of G protein responsible for initiating the retinal-dependent signaling pathway activated by UVR. HEMs express a variety of GPCRs and their related G proteins, including the melanocortin-1 receptor coupled to Gα () and the Gα-coupled endothelin-1B (ET-1) receptor (; ). Because we previously showed that UVR exposure leads to retinal-dependent Ca responses (), we monitored intracellular Ca levels using the fluorometric Ca indicator Fluo-4 in response to treatments that modulate the function of different G protein subunits. We first tested if UVR-induced Ca responses in HEMs were mediated by members of the Gα or Gα family. CTX treatment inhibits Gα-mediated signaling and PTX inhibits Gα signaling (; ). HEMs incubated for 24 h with 500 ng/ml CTX, 500 ng/ml PTX, or vehicle, and stimulated with 150 mJ/cm UVR showed no difference in the amplitude of retinal-dependent Ca responses (; amplitude of the response: F = 0.55 ± 0.04 for vehicle, 0.56 ± 0.05 for CTX, 0.52 ± 0.04 for PTX). This result suggests that Ca responses elicited by physiological doses of UVR (150 mJ/cm; ) are not mediated by Gα or Gα signaling. The Gβγ subunit of heterotrimeric G proteins can lead to increase in intracellular Ca levels ([Ca]) by directly activating PLCβ (; ; ). To test if UVR-mediated elevation in [Ca] requires activation of Gβγ, we incubated HEMs with the Gβγ-inhibiting peptide mSIRK (10 µM) or its inactive analogue L9A (10 µM; ; ; ) before stimulation with 150 mJ/cm UVR. No significant difference was measured between UVR-induced Ca responses in HEMs treated with mSIRK compared with the inactive analogue L9A (; F = 0.43 ± 0.05 for mSIRK vs. 0.43 ± 0.04 for L9A), which suggests Gβγ does not mediate UVR-induced Ca responses in HEMs. We next investigated whether members of the Gα family, known to promote Ca mobilization through PLCβ activation (; ), mediate the UVR-induced Ca response. To alter Gα signaling, we expressed RGS2 in HEMs (; ; ), a member of the regulators of G protein signaling (RGS) family, which inactivates Gα by promoting the hydrolysis of GTP to GDP (). Lentiviral transduction of HA-RGS2 in HEMs resulted in the expression of a protein of the expected molecular size () that reduced Ca responses to activation of the Gα-coupled endothelin receptors endogenously present in HEMs (; ; Fig. S1, B and C). Because HEMs expressing HA-RGS2 exhibited reduced Gα-mediated signaling via endothelin receptors, we tested if UVR-induced Ca responses were also reduced. When compared with control transduced HEMs, cells expressing HA-RGS2 stimulated with 150 mJ/cm UVR had significantly reduced retinal-dependent Ca responses (; F = 0.62 ± 0.03 for control vs. 0.37 ± 0.06 for RGS2), which suggests that members of the Gα family mediate UVR signaling in HEMs. To determine which members of the Gα family are expressed in HEMs and might be activated by UVR, we examined the mRNA expression levels of Gα family members Gα, Gα, Gα, and Gα by qPCR. We found that in HEMs Gα and Gα are expressed at similarly high levels, whereas Gα and Gα are expressed at lower levels (). We investigated the contribution of Gα versus Gα to UVR signaling using RNA interference. Because Gα and Gα share a high degree of homology, we designed miRNA targeting either Gα or Gα individually, or both Gα and Gα (Gα). Expression of Gα-targeted miRNA in HEMs resulted in a significant reduction in the mRNA transcript levels of Gα, but not of Gα, relative to control (scrambled) miRNA-expressing cells (; Gα mRNA relative to control = 0.14 ± 0.01, Gα mRNA relative to control = 0.83 ± 0.16). Cells expressing Gα-targeted miRNA had a reduced Ca response to 6 nM endothelin when compared with control miRNA-expressing cells, which suggests that Gα-targeted miRNA significantly reduced Gα signaling (Fig. S2, A and B). We then measured retinal-dependent Ca responses elicited by 150 mJ/cm UVR and found that HEMs expressing Gα-targeted miRNA had significantly reduced Ca responses compared with HEMs expressing control miRNA (; F = 0.57 ± 0.04 for control miRNA, 0.35 ± 0.03 for Gα miRNA). Expression of Gα-targeted miRNA resulted in a significant reduction in the mRNA levels of Gα (Gα mRNA relative to control = 0.12 ± 0.03) and a smaller decrease in the mRNA transcript levels of Gα (; Gα mRNA relative to control = 0.75 ± 0.07). HEMs expressing Gα-targeted miRNA had reduced Ca responses to 6 nM endothelin (), which suggests that Gα signaling is also reduced. In response to stimulation with 150 mJ/cm UVR, HEMs expressing Gα-targeted miRNA had a reduced retinal-dependent Ca response compared with HEMs expressing control miRNA (F = 0.61 ± 0.04 for control miRNA, 0.40 ± 0.05 for Gα miRNA; ). Collectively, our data from HEMs expressing Gα or Gα-targeted miRNA suggest that both Gα and Gα contribute to the UVR-induced retinal-dependent phototransduction pathway in HEMs. We next examined if decreasing Gα and Gα levels simultaneously had a larger effect on UVR-induced Ca responses in HEMs. Expression of Gα miRNA that targets both Gα and Gα significantly reduced the mRNA levels of Gα and Gα relative to control miRNA-expressing cells (; Gα mRNA relative to control = 0.24 ± 0.08, Gα mRNA relative to control = 0.32 ± 0.08). Stimulation with 6 nM endothelin resulted in reduced Ca responses in HEMs expressing Gα-targeted compared with control miRNA (Fig. S2 E and ; F = 0.72 ± 0.04 for control miRNA, 0.44 ± 0.07 for Gα miRNA). UVR stimulation of HEMs expressing Gα-targeted miRNA led to smaller increases in intracellular Ca levels compared with control miRNA-treated cells (). Quantification of these responses showed that the UVR-induced Ca responses were significantly reduced in the presence of Gα-targeted miRNA compared with control miRNA (), which suggests that both subunits Gα and Gα contribute to retinal-dependent UVR-induced Ca responses in HEMs. UVR stimulation of HEMs leads to activation of a whole-cell current mediated by TRPA1 ion channels in a retinal and G protein–dependent manner (). We thus investigated if Gα is also required for UVR-induced whole-cell currents. We first tested the effect of Gα-targeted miRNA on whole-cell currents measured at +80 mV in response to 240 mJ/cm UVR and found that expression of control miRNA had no effect on the photocurrents, whereas Gα-targeted miRNA nearly abolished them (; I = 4.06 ± 0.29 pA/pF, I = 0.28 ± 0.17 pA/pF) at all voltages (, inset). The whole-cell patch clamp technique used to measure the photocurrents allowed us to use an alternative method to block Gα signaling by dialyzing GPAnt-2a, a Gα inhibitory peptide, into HEMs (). The time-dependent effect of peptide inhibitor dialysis allowed us to measure the UVR-induced current at 2 min after break-in, when the peptide was not effective, and thus the measurement was used as control, and at 10 min of dialysis, when the peptide became effective. This experimental protocol allowed us to compare the effects of the peptide inhibitors in the same cell. When GPAnt-2a was included in the pipette solution, the UVR photocurrents were similar to control cells 2 min after break-in, but decreased significantly after 10 min of dialysis (; I = 3.98 ± 0.50 pA/pF at 2 min, 0.69 ± 0.08 pA/pF after 10 min of dialysis with GPAnt-2a). As a control we performed a similar experiment using the GPAnt-2 peptide, which inhibits Gα signaling (), and detected no change in the amplitude of the UVR current after 2 or 10 min of dialysis (; I = 3.68 ± 0.49 pA/pF at 2 min, 3.64 ± 0.26 pA/pF after 10 min). These results suggest that UVR-induced photocurrents are dependent on Gα signaling. Our results so far indicate that UVR phototransduction leads to activation of Gα, which in turn activates PLCβ, required both for intracellular Ca release () and TRPA1 activation (). Nonetheless, the mechanism by which UVR leads to TRPA1 activation downstream of PLCβ remains unknown. Because PLCβ hydrolyzes plasma membrane PIP, generating DAG and the soluble messenger IP, we reasoned that PLCβ-dependent signaling could modulate TRPA1 channel activity in HEMs via IP, DAG, or PIP. We first tested whether IP or IP-mediated Ca release was sufficient to activate TRPA1 by using a control internal solution, allowing for an increase in [Ca] (see Materials and methods), internal solutions containing 100 µM IP to stimulate IPRs, or 1 mg/ml heparin to block IPRs, with both treatments occluding subsequent UVR-induced IPR-mediated Ca release. UVR (240 mJ/cm) exposure after dialysis for 5 min with each of the three solutions elicited retinal-dependent photocurrents with similar amplitudes (Fig. S3, A and B; I = 4.01 ± 0.33 pA/pF, I = 4.04 ± 0.47 pA/pF, I = 5.05 ± 0.56 pA/pF), which suggests that UVR-induced activation of whole-cell currents is not mediated by IP or IPR-mediated Ca release. We next examined the contribution of DAG to retinal-dependent UVR photocurrents. Bath application of the PC-PLC (10 U/ml), which generates DAG in the plasma membrane (), or of the DAG analogue OAG (100 µM), did not elicit a significant current in HEMs, and 5 min of incubation with the respective treatments did not affect the retinal-dependent photocurrents elicited by UVR (Fig. S3, C and D; I = 0.10 ± 0.10 pA/pF, I= 0.20 ± 0.09 pA/pF, I = 4.31 ± 0.70 pA/pF, I = 4.13 ± 0.48 pA/pF). Because increasing DAG levels failed to elicit whole-cell currents and had no effect on UVR photocurrents, we concluded that DAG does not modulate TRPA1 downstream of UVR. PIP hydrolysis is a key regulator for many ion channels (), including TRPA1 (; ; ). To test if the presumed decrease in PIP levels caused by PLCβ-mediated hydrolysis affects UVR photocurrents, we attempted to maintain elevated PIP levels by dialyzing HEMs with the PIP analogue diC8-PIP (20 µM; ; ). We found that after 5 min of patch pipette dialysis to allow diC8-PIP to diffuse into cells, UVR stimulation (240 mJ/cm) elicited significantly smaller photocurrents at all voltages compared with UVR photocurrents measured immediately after break-in (). This finding suggests that increased PIP prevents UVR-induced TRPA1 activation and raises the question of whether PIP hydrolysis is required for the UVR photocurrent. To test if the dialysis with diC8-PIP inhibited TRPA1, or another component of the UVR phototransduction cascade, we compared the whole-cell currents elicited by UVR and the TRPA1 agonist cinnamaldehyde (CA) in HEMs dialyzed with diC8-PIP or GPAnt-2a (). We found that dialysis with both diC8-PIP and GPant-2a inhibited UVR photocurrents, whereas only diC8-PIP inhibited currents elicited by CA (Fig. S4). These data suggest that GPAnt-2a inhibits an important component of the UVR signaling cascade, Gα, while diC8-PIP directly inhibits TRPA1 activity, which is required for both the UVR- and CA-elicited increase in whole-cell current. To test if a decrease in PIP levels is also sufficient to cause TRPA1 activation in HEMs, we dialyzed cells with polylysine (polyK; 50 mg/ml), which binds and sequesters PIP (; ; ), preventing it from acting on the TRPA1 channels. A subsaturating UVR dose (160 mJ/cm; ) evoked a submaximal UVR photocurrent in HEMs dialyzed with control internal solution. PolyK alone, when included in the pipette solution and allowed to diffuse into the cell and sequester PIP, did not lead to an increase in whole-cell current; however, it significantly enhanced UVR photocurrents elicited by the same subsaturating UVR dose (; I = 0.07 ± 0.03 pA/pF, I = 2.51 ± 0.07 pA/pF, I = 4.90 ± 0.44 pA/pF). The polyK-modulated UVR photocurrent was inhibited by the TRPA1 antagonist HC-030031 (HC; 100 µM; , right; and ; I = 0.27 ± 0.12 pA/pF), which suggests that the enhanced photocurrent was mediated by TRPA1. To address the specificity of the phospholipids that modulate the UVR photocurrents, we used recombinant pleckstrin homology (PH) domains from phospholipase C δ1 (PLCδ1-PH) that selectively bind PIP and from the general receptor for phosphoinositides type 1 (GRP1-PH) that selectively binds PIP (PIP; ). Dialysis of recombinant PH domains in HEMs will bind and sequester the phosphoinotides, resulting in decreased cellular levels. When PLCδ1-PH was included in the pipette, a significantly enhanced photocurrent was elicited by a subsaturating UVR dose (160 mJ/cm), compared with boiled PLCδ1-PH or GRP1-PH (; I = 3.01 ± 0.36 pA/pF, I = 1.32 ± 0.26 pA/pF, I = 1.34 ± 0.24 pA/pF). These results suggest that sequestering PIP, but not PIP, levels in HEMs leads to increased TRPA1 activity in response to UVR, leading us to hypothesize that UVR-mediated PIP hydrolysis releases the PIP-mediated inhibition of TRPA1. Retinal-dependent UVR-induced Ca responses triggered downstream of Gα signaling are mediated by two sources of Ca: efflux from intracellular thapsigargin-sensitive stores and influx via TRPA1 at the plasma membrane (). Because Gα/PLCβ signaling generates the second messenger IP, we hypothesized that IPRs mediate Ca release from intracellular stores. To test this hypothesis, we treated HEMs with the IPR antagonist XeC (25 µM; ; ) and found that UVR-induced Ca responses were significantly reduced (). To study the contribution of each Ca source (IPR vs. TRPA1) to the overall response, we measured Ca responses in HEMs preincubated with XeC in order to inhibit IPR, or HC-030031 (HC; 100 µM) in order to inhibit TRPA1 (). Ca responses elicited by 240 mJ/cm UVR were reduced by 71% in the presence of XeC, by ∼45% in the presence of HC, and by ∼90% in the presence of both antagonists, as compared with vehicle-treated cells (; fluorescence increase over baseline: ΔF/F = 3.21 ± 0.27 for vehicle, 0.94 ± 0.16 for XeC, 1.77 ± 0.29 for HC, and 0.34 ± 0.07 for XeC + HC). These results suggest that UVR phototransduction evokes a rise in [Ca] via intracellular Ca release from IPR and Ca influx through TRPA1 ion channels. We next sought to distinguish the contribution of each Ca source (IPR and TRPA1) to the biphasic nature of the transient UVR-induced Ca response. To do that, we measured the initial slope and the decay time constant of Ca responses in HEMs treated with XeC or HC, compared with vehicle. The mean initial slope of UVR-induced Ca responses (measured during the first 30 s after the beginning of UVR stimulation) was reduced by ∼76% with XeC treatment and by ∼35% with HC treatment when compared with vehicle-treated HEMs (; ΔF/s = 0.100 ± 0.009 for vehicle, 0.024 ± 0.003 for XeC, and 0.064 ± 0.01 for HC). These results suggest that both sources contribute to the initial rising phase of the Ca response, but IP-mediated Ca release has a significantly greater contribution. The time constant for the decay phase (τ) of the UVR-induced Ca response was measured by monitoring cellular Ca levels for ≥200 s after the peak of the response in cells treated with XeC, HC, or vehicle. HEMs treated with XeC had Ca responses with mean τ values ∼79% higher than vehicle-treated cells, which suggests that the IP-mediated response decays on a fast time scale. Interestingly, treatment with HC resulted in UVR-evoked Ca responses that decayed ∼38% faster than vehicle-treated cells (; τ = 225.90 ± 37.44 s for XeC, 78.42 ± 12.68 s for HC, and 126.20 ± 10.61 s for vehicle). Hence, TRPA1 activation extends the duration of the Ca responses, whereas IPR activation reduces the duration. Collectively, our results suggest that IPR-mediated Ca release is important to ensure a fast rising phase of the response. In contrast, consistent with our previous findings (; ), TRPA1 activation is slow but persistent, contributing to a prolonged Ca response (). We have recently discovered that primary human melanocytes are capable of rapidly detecting UVR by first increasing intracellular Ca and later producing more melanin (). The Ca response is retinal dependent and is in part due to calcium release from intracellular stores and in part to calcium influx through TRPA1 ion channels (). Both components of the response require heterotrimeric G proteins and PLCβ activation (; ). Here we investigated the identity of the G protein subunit that mediates the response and found that reducing the expression of both Gα and Gα, as well as using a peptide that inhibits Gα signaling, significantly decreased both components of the Ca response ( and ). These results suggest that the UVR-activated pathway is mediated by Gα, which, in turn, activates PLCβ. Signal transduction pathways associated with G protein activation often result in modulation of TRP ion channels. For example, in , light stimulation of rhodopsin results in activation of Gα and PLCβ, which hydrolyzes PIP to produce DAG and IP, to regulate TRP ion channels (). We reasoned that one of these messengers (IP, DAG, or PIP) might regulate the UVR-activated TRPA1 photocurrent and tested our hypothesis by exogenously altering the levels of these messengers ( and S3). We found that reagents used to manipulate the levels of IP and DAG had no effect on the photocurrent, whereas reagents that affect PIP levels did. Allowing diC8-PIP to diffuse into the cell blocked the current in response to a UVR dose that evokes a maximal response, which suggests that decreasing PIP levels is a necessary step in TRPA1 activation. But is the decrease in PIP sufficient to activate TRPA1? Sequestration of PIP using polyK did not elicit a significant current in the absence of UVR, but did potentiate TRPA1 currents evoked by UVR. Furthermore, using PLCδ1-PH to specifically sequester PIP enhanced UVR photocurrents, whereas sequestration of PIP by GRP1-PH had no effect. These results suggest that a decrease in PIP can modulate the photocurrent, but may not be sufficient for TRPA1 activation. Therefore, other messengers or proteins that contribute to the phototransduction pathway could be involved. Our results add to the already controversial role of PIP in modulating TRPA1. PIP was found to potentiate agonist-activated TRPA1 currents (), but also to inhibit TRPA1 activity (; ). GPCRs that activate Gα (bradykinin, PAR2, and Mrgprs) can regulate TRPA1, but the mechanism is unclear (; ; ; ). It remains to be determined in future experiments whether PIP directly interacts with TRPA1 to modulate its activity and what other cellular messengers contribute to TRPA1 activation in response to UVR and to Gα-coupled receptors. Examining the relative contribution of UVR-induced IPR- and TRPA1-mediated Ca responses revealed that IP-mediated Ca release is rapid, but also declines fast. In contrast, TRPA1-mediated influx is slower to increase intracellular Ca, as suggested by the slow time course of the photocurrent activation, and slow to decay, allowing intracellular Ca to remain elevated after UVR exposure, a necessary step for melanin production (). However, there is a discrepancy between the time course of photocurrents and that of Ca responses. UVR photocurrents recorded under voltage-clamp conditions peak during or shortly after light stimulation, whereas Ca responses measured in intact cells, in which the membrane voltage could change, peak seconds after irradiation. We recently found that UVR phototransduction depolarizes the plasma membrane of melanocytes to delay TRPA1 inactivation and prolong Ca responses (), a finding consistent with the TRPA1-mediated influx kinetics found in this study. The discrepancy between the time course of the response measured by the two methods is likely to be caused by the significant effects of membrane depolarization on the TRPA1 channel and consequent Ca signaling dynamics. Our analyses of Ca response dynamics also revealed the possibility of cross-talk between the Ca release and influx pathways. Inhibition of IPR resulted in a significantly slower decline of the TRPA1-mediated Ca response, which suggests that the IP-mediated response decays considerably faster than the overall response. However, when TRPA1 was inhibited, the decay of the IP-mediated Ca responses was only slightly (although significantly) faster, which suggests that Ca release may accelerate TRPA1 inactivation. Based on our data, we propose that UVR exposure of HEMs stimulates a retinal-dependent Gα-coupled receptor, which activates PLCβ. Active PLCβ hydrolyzes plasma membrane PIP into DAG and IP. Soluble IP binds IPRs in the ER, resulting in Ca release, whereas the decrease in PIP levels modulates TRPA1 activation and Ca influx (). Our model for UVR signal transduction in human melanocytes resembles visual phototransduction in photoreceptors () and nonvisual phototransduction in the mammalian retina (; ). The melanocyte UVR pathway also shares many similarities with a recently described UVR-activated signaling mechanism in larvae, which is mediated by Ca signals resulting from Gα and TRPA1 activation (). The receptor for the larvae UVR pathway appears to be a gustatory GPCR (Gr28b); how light can activate such a receptor remains unknown. One of the remaining questions for the UVR phototransduction cascade in melanocytes is the identity of the receptor. The retinal dependence and G protein involvement suggests the involvement of an opsin GPCR. We previously found that rhodopsin expression contributes to UVR-induced Ca responses (). However, the differences in spectral sensitivity and G protein coupling of rhodopsin and of the UVR pathway suggest that rhodopsin might work in conjunction with a different, possibly unidentified, UVR-sensitive receptor to mediate UVR phototransduction in melanocytes. Future experiments will identify the UVR receptor and determine the molecular mechanism of TRPA1 activation. Understanding this pathway might uncover new photoprotection strategies for human skin, thus lowering the incidence of skin cancer.

CFTR is a chloride ion channel () crucial for the salt water balance of several polarized epithelia. Mutations in CFTR are the cause of cystic fibrosis (CF) (), the most common lethal genetic disease among Caucasians, an incurable, devastating multi-organ disorder (). The most common CF-causing mutation (carried by >90% of patients), deletion of phenylalanine 508 (ΔF508), severely impairs both protein folding/stability (; ; ) and channel open probability (P) (; ). Thus, much effort is focused on the development of “correctors,” chemical chaperones that promote folding/stability of the ΔF508 CFTR protein, and of “potentiators,” which stimulate P of ΔF508 (or other mutant) CFTR channels. CFTR is an ATP-binding cassette (ABC) protein, with a typical ABC architecture consisting of two transmembrane domains (TMDs) and two cytosolic nucleotide-binding domains (NBDs). The additional cytosolic regulatory (R) domain, unique to CFTR (), is a substrate for cyclic AMP-dependent PKA, and its phosphorylation is required for CFTR channel activity (). CFTR’s chloride ion pore is formed by the TMDs and gated by a cycle of ATP binding and hydrolysis at the two NBDs (). Despite the functional divergence of the ion channel CFTR from other ABC proteins, which are mainly active transporters, a conserved molecular mechanism couples cycles of conformational changes in NBDs and in TMDs of all ABC proteins (). In the presence of ATP, NBDs of ABC proteins form tight head-to-tail dimers, occluding two molecules of ATP; each nucleotide is sandwiched between the conserved Walker A and B motifs of one NBD and the conserved signature sequence of the other, which together form a composite catalytic site with ATPase activity (e.g., ; ). NBD dimers glued together by two ATP molecules are extremely stable, but they dissociate after ATP hydrolysis (; ). In ABC-C family members, including CFTR (ABCC7), the composite binding site formed by Walker motifs of NBD1 and signature sequence of NBD2 (“site 1”) is degenerate and catalytically inactive (; ); ATP hydrolysis occurs only in “site 2” (formed by Walker motifs of NBD2 plus signature sequence of NBD1). In full-length ABC exporter structures, tightly associated NBD dimers are linked with outward-facing (; ) and fully or partially separated NBDs with inward-facing (; ; ) TMD conformations, suggesting that formation of ATP-bound NBD dimers flips the TMDs from inward to outward facing, whereas ATP hydrolysis drives dimer dissociation to reset the TMDs to inward facing (). In the ion channel CFTR, the NBD dimerization–dissociation cycle is coupled to the opening and closure of the transmembrane permeation pathway: the pore opens to a burst when the tight NBD dimer forms, and it closes from a burst when the dimer interface separates around site 2, after hydrolysis of the ATP bound there (). (CFTR current records display a bursting pattern: open events, which last for ∼2–300 ms, are interrupted by brief [<10-ms] ATP-independent “flickery” closures, distinct from the long closed events [∼1 s] that reflect NBD dimer separation [“interburst closures”; ]; in this study, the terms “pore opening” and “pore closure” are used synonymously with entering and exiting a burst.) In the absence of hydrolysis, pore closure is extremely slow: site-2 mutations that disrupt ATP hydrolysis slow the closing rate by ∼100-fold (). Thus, gating of WT channels is an essentially unidirectional cycle; channels that open to a prehydrolytic open state (O) preferentially progress to a post-hydrolytic open state (O), to then close through a pathway (O→C) distinct from pore opening (C→O) (shown in insets throughout –; ). This far-from-equilibrium operation is rare among ion channels but is an essential property of ABC transporters that mediate unidirectional uphill transport. At saturating [ATP], the CFTR functional cycle contains two rate-limiting steps with relatively high free energy barriers: step C→O (rate of ∼1 s) determines opening rate, whereas step O→O (rate of ∼4 s) rate limits closure (). Because the TMDs are the most divergent parts of ABC proteins, they are promising targets for the development of highly selective drugs. The arylaminobenzoate 5-nitro-2-(3-phenylpropylamino)benzoate (NPPB) is an anionic compound that binds to CFTR’s TMDs, and it has been studied extensively for its pore blocker properties (; ). Unexpectedly, discovered that NPPB, in addition to blocking the pore, dramatically increases P (i.e., potentiates gating) of both WT and ΔF508 CFTR. Given the potential therapeutic impact of CFTR potentiators, the present study aimed to identify the exact molecular mechanism by which NPPB stimulates P. As a control, we compared the effects of NPPB to those of MOPS, another well-characterized CFTR pore blocker (), but without potentiating effects. Our results offer a conceptually novel approach to robustly stimulate CFTR that exploits its unique nonequilibrium gating cycle, and that differs fundamentally from strategies applied so far to modulate the activity of any ion channel. WT and mutant CFTR cDNA in pGEMHE was transcribed in vitro using T7 polymerase, and 0.1–10 ng cRNA was injected into oocytes as described previously (). Current recordings were done at 25°C in inside-out patches excised from oocytes 2–3 d after RNA injection. Pipette solution contained (mM): 136 NMDG-Cl, 2 MgCl, and 5 HEPES, pH 7.4 with NMDG. Bath solution (mM: 134 NMDG-Cl, 2 MgCl, 5 HEPES, and 0.5 EGTA, pH 7.1 with NMDG) was continuously flowing and could be exchanged with a time constant of 20–30 ms. CFTR channels were prephosphorylated by 1–2-min exposure to 300 nM PKA catalytic subunit (Sigma-Aldrich). Recordings were done in the presence of saturating (2 mM) MgATP; for the K1250A mutant 10 mM MgATP was used to compensate for its greatly impaired apparent ATP affinity (). MgATP (Sigma-Aldrich) was added from a 400-mM aqueous stock solution (pH 7.1 with NMDG), and Na-pyrophosphate was added from a 500-mM aqueous stock solution, together with equimolar MgCl. NPPB (Sigma-Aldrich) was added from a 250-mM stock solution in DMSO; by spectrophotometry, NPPB solubility in water was >400 µM. Solutions containing MOPS were adjusted to pH 7.2; thus, [MOPS] was ∼50% of total [MOPS] (pK = 7.2). Currents recorded (Axopatch 200B; Molecular Devices) at a bandwidth of 2 kHz were digitized at 10 kHz (Digidata 1322A; Pclamp9; Molecular Devices). For gating analysis, current records were refiltered at 100 Hz. Steady-state mean burst and interburst durations in patches containing fewer or equal to two channels () were obtained by maximum likelihood fits of the closed-open-blocked scheme to the ensembles of all dwell-time histograms (); this approach efficiently separates brief ATP-independent “flickery” closures from long “interburst” closed events (). For burst analysis of one-channel records (), brief closures were suppressed using the method of . Distributions of burst durations longer than = 12 ms were fitted to gating models by maximum likelihood (); alternative models were compared using the log-likelihood ratio test (). Average unitary current amplitudes were estimated from Gaussian fits to amplitude histograms of heavily (10 Hz) filtered records (see ). Because titration of NPPB () had to be done in a dilute (100-µM) solution, the titration curve was fitted by accounting for the presence of ∼10 µM of aqueous CO. Thus, the data were fitted to a curve expected for a mixture of two weak acids: for one component (CO) a fixed concentration of 10 µM and pK* = 6.47 was used, whereas concentration and pK of the second component (NPPB) was left free during the fit. The slight deviation of the data from the fitted curve at pH >9 () reflects inevitable progressive accumulation of dissolved CO at very basic pH. All symbols and bars represent the mean ± SEM of at least five measurements, obtained from several patches. For parameters extracted from macroscopic currents, the average number of measurements was greater than nine, and the average number of patches was greater than five, for each data point (bar or symbol) shown. For the single-channel data shown in and , the numbers of patches analyzed for each condition are provided in the figure legends. Statistical significance was evaluated using Student’s two-tailed test (*, P < 0.05; **, P < 0.01). Fig. S1 illustrates the kinetics of pore block by NPPB and MOPS. Fig. S2 demonstrates the effect of NPPB on the unlocking rate of WT CFTR channels locked open by ATP plus pyrophosphate. The online supplemental material is available at . Cytosolic anionic pore blockers are driven into the CFTR pore by hyperpolarizing voltages and disrupt the flow of chloride ions, causing flickery block. To deconvolve effects on gating from those on permeation (pore block), we first quantitated the latter in inside-out patches of single WT CFTR channels gating in 2 mM of bath ATP (, left). Bath application of 210 µM NPPB elicited flickery closures whose frequency increased at more negative potentials to such an extent that distinct unitary gating events disappeared, leaving “fuzzy” noise (, middle). In contrast, bath application of 20 mM MOPS appeared to reduce unitary current amplitudes at negative voltages (, right). This seemingly different behavior reflects the lower affinity of MOPS for the CFTR pore, resulting in briefer interactions and hence blocked events too fleeting to be resolved at our recording bandwidth of 100 Hz (; ; ). On the other hand, after filtering the data at 10 Hz (), the effect of NPPB on unitary currents appears similar to that of MOPS; both appear to reduce unitary amplitudes at negative voltages. For the purposes of this study, it is useful to describe pore block in terms of a reduction of average unitary chloride current () flowing through the open pore (; compare blue and green to red symbols), as observed in heavily filtered (10-Hz) records. Plotting fractional unitary currents in the presence of NPPB or MOPS against membrane voltage () revealed, despite different affinities, similar voltage dependence of pore block: effective valences (from Boltzmann fits; , solid lines) were −0.51 ± 0.03 for NPPB and −0.45 ± 0.03 for MOPS (compare to ), suggesting that both blocking sites sense ∼50% of the transmembrane electrical field. A convenient macroscopic pore block assay is afforded by the nonhydrolytic E1371S CFTR mutant, which lacks the catalytic glutamate in site 2. These channels open rapidly in response to ATP but stay open for tens of seconds (), consistent with defective ATP hydrolysis. Because they are active in resting cells (compare to ), excision of E1371S multichannel patches into an ATP-free solution typically uncovers large, slowly decaying currents devoid of gating fluctuations (). As they flow through virtually locked open channels (P of ∼1), responses of such currents to brief (2–5-s) applications of blockers reflect pure pore block. Indeed, brief applications of 210 µM NPPB () or 20 mM MOPS () at various voltages yielded voltage-dependent current block (, closed symbols and lines), which essentially replicated the effects on unitary current amplitudes (, open symbols and dotted lines; replotted from ). Using this convenient methodology, we measured pore block at varying [NPPB] () and [MOPS] () at a fixed voltage of −120 mV, constructed dose–response curves (), and obtained apparent K values of ∼20 µM for NPPB (compare to ) and ∼8 mM for MOPS (compare to ). We next examined the effects of NPPB and MOPS on macroscopic currents of WT CFTR channels opening and closing at steady state in saturating (2 mM) ATP. At −120 mV, the application of increasing [NPPB] () or [MOPS] () inhibited these currents in a dose-dependent manner, and fractional inhibition by MOPS (, closed symbols; K of ∼15 mM) was reasonably explained by its effect on unitary currents (, open symbols; replotted from ). In contrast, for NPPB (), at any given concentration, the fractional reduction of macroscopic WT steady-state currents (closed symbols) was milder than the fractional reduction of unitary currents (open symbols; replotted from ), yielding an apparent K of ∼85 µM (compare to ). The macroscopic current () is given by = · · P, where is the number of channels in the patch, is understood as average unitary current during a burst, and P is the fraction of time a channel spends in the bursting state (bursting probability). Therefore, a discrepancy between fractional effects of a drug on and (, red arrow) indicates that the drug must also simultaneously affect P. The fractional effect on P () is given by the ratio of normalized macroscopic and unitary currents. A marginal increase in P (∼50%) was observed in very high (≥20 mM) MOPS (, green symbols). But a much more potent gating stimulation was seen with NPPB, at submillimolar concentrations, with P increased up to fourfold (, blue symbols): a close-to-maximal stimulation, considering that P is ∼0.2 under control conditions (see ). To understand the mechanism by which NPPB increases P, we systematically studied its effects on the rates of individual steps in the gating cycle (, cartoon). We first examined effects on the rate of normal, hydrolytic closure of WT CFTR (, cartoon, purple arrow; r) by comparing macroscopic current relaxation rates (at −120 mV) upon sudden ATP removal in the absence or presence of 210 µM NPPB (). Single-exponential fits (, solid lines, time constants indicated) revealed approximately fourfold slower closing rates in the presence of 210 µM NPPB relative to control (, blue vs. red bar). In contrast, in similar experiments, 20 mM MOPS () failed to alter channel closing rate (, green bar). To evaluate a potential effect of NPPB on opening rate, we exposed WT CFTR channels gating at steady state in 2 mM ATP to 210 µM NPPB (). Although in similar experiments MOPS application and removal caused simple monophasic current relaxations (), both the addition and removal of NPPB elicited biphasic responses (), attesting to the dual effects of this compound. Upon NPPB application, instantaneous (τ of ∼20 ms, reflecting solution exchange time) pore block was followed by a partial current recovery, as a larger pool of channels was drawn into the open burst state. Upon NPPB withdrawal, instantaneous relief from pore block revealed this larger pool of open channels in the form of a large current overshoot, which then relaxed back to the pre-application steady-state current level. The rate of macroscopic current relaxation after a sudden change in gating parameters reflects the sum of the average microscopic opening and closing rates in the new conditions. Interestingly, single-exponential fits to the current time courses after block (, blue fit line) and unblock (red fit line) yielded similar time constants, suggesting that the sum of opening and closing rates in the presence (, blue bar) and absence (, red bar) of 210 µM NPPB is not very different. However, in the absence of NPPB, this sum is largely dominated by the closing rate (compare , red bar, to , red bar). In contrast, in the presence of NPPB, closing rate is slowed to ∼1 s (, blue bar); thus, the sum (, blue bar) must be dominated by an increased opening rate (see also ). These data suggest that NPPB increases channel opening rate, which, under our conditions of saturating ATP, reflects the rate of step C→O (, cartoon, purple arrow; r). For most ion channels, potentiators stabilize open states. If NPPB acted by stabilizing state O relative to C, then, in addition to speeding opening (see ), it might be expected to slow the reverse of the opening step, i.e., nonhydrolytic closure (, cartoon, purple arrow; ). To test this, we studied the closing rate of K1250A CFTR channels, in which mutation of the NBD2 Walker A lysine abrogates ATP hydrolysis at site 2 () and reduces gating to reversible C↔O transitions (, cartoon). Consistent with previous reports (; ; ), macroscopic closing rate of K1250A upon removal of bath ATP () was ∼100 times slower than for WT channels (, red fit line, and B, red bar; compare to ). Whereas 20 mM MOPS did not affect nonhydrolytic closure (, green fit line, and B, green bar), 210 µM NPPB unexpectedly accelerated it by two- to threefold (, blue fit line, and B, blue bar). As an alternative approach for studying effects on nonhydrolytic closure, we also compared unlocking rates in the absence and presence of NPPB for WT CFTR channels locked in the open state by exposure to a mixture of ATP plus pyrophosphate (; ). Consistent with its effect on K1250A closing rate (), 210 µM NPPB also accelerated the unlocking rate of pyrophosphate-locked WT CFTR channels by two- to threefold (). Because NPPB accelerated both forward () and backward () transitions of the C↔O step, we examined whether it affects the equilibrium constant between those two states (, cartoon, purple double arrow; K), i.e., P for K1250A channels. Even prolonged exposure to 210 µM NPPB of K1250A channels gating at steady state in 10 mM ATP (; note V = −40 mV) failed to elicit biphasic responses like those seen for WT channels (see ). Moreover, fractional current inhibition at steady state was identical to that instantaneously observed upon brief application of NPPB long after ATP removal (, yellow box, expanded in inset). The latter maneuver measures pure pore block of surviving open channels, i.e., fractional reduction of (see ). The very similar fractional NPPB effects under those two conditions (, red vs. yellow bar; compare to ) revealed no fractional change in P (, blue bar). Thus, in contrast to its effect on hydrolytic gating (), NPPB does not affect P for steady-state nonhydrolytic gating. Given the observed increase in nonhydrolytic closing rate (), this result indicates that NPPB must also speed the opening rate of K1250A CFTR, just as it does for WT (). The magnitude of the NPPB effects on opening and closing rates must be similar, such that the C↔O equilibrium constant is not affected; i.e., by reducing the height of the energetic barrier separating the two states, NPPB acts as a catalyst for this step. NPPB slowed the hydrolytic closing rate of WT CFTR (), which reflects sequential transition through the prehydrolytic O and post-hydrolytic O states (, cartoon, purple arrow; r). Dissection of the rates of the two sequential steps requires maximum likelihood fitting of the distributions of open burst durations (). To achieve this, we analyzed steady-state gating of single WT CFTR channels under control conditions, in 20 µM NPPB (unitary currents in 210 µM NPPB could not be resolved at −120 mV [], whereas 20 µM NPPB did not compromise reliable event detection), or in 20 mM MOPS (). 20 mM MOPS did not significantly affect either mean burst (308 ± 60 ms; = 9) or mean interburst durations (, green bars). However NPPB, even at this low concentration, significantly prolonged mean burst duration (from 277 ± 24 ms [ = 37] to 412 ± 56 ms [ = 15]; P = 0.013), i.e., slowed closure; reduction of mean interburst duration was not significant (, blue vs. red bars). To determine which of the two sequential closing steps is slowed by NPPB, we reconstructed the distributions of burst durations for single WT CFTR channels gating at −120 mV in the absence () or presence of 20 µM NPPB (). Both distributions were peaked and fit (, solid lines) significantly better (P = 4 × 10 and 9 × 10, respectively, by the log-likelihood ratio test) by the two-parameter (, ) irreversible sequential closing mechanism (, cartoon, purple arrows) than by a single exponential. The fit parameters (, bars) attested to a significant (P < 10) reduction by NPPB of the slow rate , which represents the rate of step O→O associated with ATP hydrolysis in site 2 (). NPPB was first identified as a potentiator at positive voltages () because at negative voltages gating stimulation is masked by the pore block, which is voltage dependent. To assess whether any of its effects on gating are also voltage dependent, we systematically studied gating in NPPB at +60 mV and compared NPPB effects at +60 mV with those at −120 mV (–). Because at positive voltages unitary current in 210 µM NPPB remains ∼67% of control (as opposed to ∼9% at −120 mV; ), gating stimulation prevails over pore block, causing dose-dependent overall stimulation of macroscopic WT CFTR current (). Indeed, dividing fractional macroscopic currents (, open squares) by fractional unitary currents (, open diamonds) revealed a powerful increase in P (, closed circles); the approximately threefold stimulation by 210 µM NPPB approached that observed at −120 mV (, blue circles). Furthermore, 210 µM NPPB slowed closing rate by approximately threefold (; compare to ), and similar macroscopic relaxation rates upon the addition and removal of NPPB attested to a simultaneous increase in opening rate in the presence of NPPB (; compare to ). Finally, at +60 mV, 210 µM NPPB increased the nonhydrolytic closing rate of K1250A CFTR by approximately threefold (), just as it did at −120 mV (). Because at +60 mV unitary currents remained resolvable even in the presence of 210 µM NPPB (), the conclusions drawn from nonstationary macroscopic analysis regarding effects on WT CFTR gating () could be corroborated in steady-state single-channel recordings: as expected, 210 µM NPPB prolonged mean burst duration but shortened mean interburst duration, both by approximately threefold (, blue bars). In contrast to the three- to fourfold stimulation by NPPB of P for WT CFTR observed here, reported 10–15-fold stimulation of ΔF508 CFTR by NPPB. Indeed, we found that gating stimulation of low temperature–rescued ΔF508 CFTR by 210 µM NPPB is so robust that it overrides the pore block effect at all voltages, yielding overall current stimulation even at −120 mV (, yellow bar), and greater than fivefold stimulation at +60 mV (, cyan bar). Accounting for the pore block effect, these results suggest ∼20- and ∼8-fold stimulation of P at −120 and +60 mV, respectively (). Interestingly, also reported a dependence of NPPB stimulation on the phosphorylation state of WT CFTR. Whereas all the data in – were obtained in the relatively stable, partially dephosphorylated state of CFTR, after removal of PKA (), we indeed observed an ∼10-fold stimulation of the small current, carried by low P activity of unphosphorylated WT CFTR channels, when 210 µM NPPB was applied at +60 mV before exposure of the patch to PKA (, first application of NPPB [expanded in inset], and B, left). In contrast, in the same patch, 210 µM NPPB had little effect on overall current at +60 mV when applied in the presence of PKA, i.e., to fully phosphorylated channels that are gating at high P (, second application of NPPB, and B, right), whereas soon after PKA removal, it increased by approximately twofold the current carried by partially dephosphorylated channels, which are gating at intermediate P (, last NPPB application, and B, middle). Accounting for the pore block, these effects of NPPB on overall WT CFTR current at +60 mV correspond to ∼15-, ∼3-, and <2-fold stimulation of P of poorly, partially, and fully phosphorylated CFTR (). These data confirm the findings of and suggest that NPPB stimulates P much more efficiently when the starting P value is low. At our bath pH of 7.1, the anionic, deprotonated form of NPPB dominates, and only a small fraction of the compound is protonated, uncharged. Although the pore block is no doubt attributable to the anionic form of NPPB (NPPB), we wanted to identify whether it is this anionic form or perhaps the less prevalent protonated form (NPPB-H) that is responsible for gating stimulation. Because we found no reliable data in the literature for the pK value of NPPB (most studies assumed a pK of ∼4.5 by referring either to the study of , in which the pK was not measured, or to that of , in which no data were shown), we determined the latter by electrometric titration. Because of the limited solubility of NPPB in water (especially at low pH), titration by NaOH (, closed circles) and HCl (, open circles) could be performed only for a dilute (100-µM) aqueous solution of the compound, restricting reliable determination of potential pK values to the range of 4–10. (At this low concentration, a weak acid with a pK of <4 is largely dissociated, whereas a site with a pK of >10 remains largely protonated even after the addition of 1 mol/mol NaOH.) The data were well fit (, solid line) by a titration curve expected for a monoprotic weak acid and revealed a single pK value of ∼5.6 in the range of 4–10. To deconvolve gating and permeation effects of NPPB at various pH, we first characterized pore block at +60 mV by 210 µM of total [NPPB] applied at bath pH values of 8.1, 7.1, and 6.1 (). Whereas average unitary currents, determined from heavily filtered current traces, decreased to ∼67% of control when 210 µM NPPB was applied at pH 7.1 or 8.1 (, left; compare to ), at pH 6.1 the pore block was significantly (P = 8 × 10) milder, yielding a fractional decrease in unitary current of only ∼15% (, third bar). This is consistent with a significantly decreased concentration of NPPB at this low pH, as predicted by the measured pK value of 5.6 (see table in ). We next assayed macroscopic current stimulation at +60 mV by 210 µM of total NPPB at various pH (). Fractional current stimulation at pH 6.1 was not enhanced relative to that measured at our control pH of 7.1 ( and middle blue bars in G), despite an ∼10-fold higher concentration of uncharged NPPB-H at the lower pH (, table); in fact, the calculated effect on P was significantly blunted at pH 6.1 (, middle dark blue bars). Furthermore, no significant stimulation of P was observed when applying 21 µM of total NPPB at pH 6.1 ( and right side of G), a condition in which [NPPB] is selectively lowered, whereas [NPPB-H] remains comparable to that calculated for 210 µM of total NPPB at pH 7.1 (, table). These results rule out uncharged NPPB-H as the active form responsible for gating stimulation and suggest that the latter is largely caused by anionic NPPB. In principle, the reduced fractional stimulation of P by 210 µM of total NPPB at pH 6.1 could be taken as an indication that uncharged NPPB-H is entirely inactive; however, we refrain ourselves from such a conclusion, as fractional current stimulation was also blunted at pH 8.1 ( and left blue bar in G) relative to pH 7.1, despite slightly higher NPPB at the higher pH (, table). The similarly reduced fractional effect on P at both pH 6.1 and 8.1 (, compare first and third dark blue bars) suggests that changes in surface properties and/or gating kinetics of the CFTR protein at both acidic and basic pH should also be considered (compare to ). Until very recently, treatment of CF has been exclusively symptomatic. This has changed with the approval of VX-770 (ivacaftor; Vertex Pharmaceuticals; ; ), the first drug shown to act by binding specifically to the CFTR protein. However, to date it has been shown to benefit only patients carrying at least one G551D allele, which constitute <5% of the CF population. Effective drug therapy to treat patients carrying ΔF508 alleles is still lacking. At present, there is a wide gap between industrial efforts to obtain drugs targeting CFTR and academic research aimed at understanding CFTR structure and mechanism at a molecular level. Although industrial high-throughput screening has clearly met some success (; ), the wealth of information that has emerged from basic research has yet to impact drug discovery. In this study, we have identified two strategic points in CFTR’s unique functional cycle that are eminently suited as intervention points for altering CFTR activity: the channel opening step, and the step that rate limits channel closure (hydrolytic O→O step). Our dissection of the mechanism of action of NPPB, presented here, shows how affecting these transitions can powerfully influence P of WT and mutant (including ΔF508) CFTR channels, and offers a route to rational design of drugs targeting CFTR. Our interpretation builds on the nonequilibrium nature of the CFTR gating cycle (see the simplified cyclic gating model depicted throughout the insets of –), which is rooted in its evolutionary descent from an ancestral ABC transporter. First proposed by , this model was later tested and refined (; , ; ; ) before becoming generally accepted (; ; ; but compare to ). Recently, various extensions of this scheme have been suggested to explain the rare openings (P of ∼0.004), in the absence of ATP, of unliganded WT () or G551D CFTR (), enhancement of such activity by cytosolic loop mutations (), and [ATP]-dependent prolongation of open burst durations in mutants () or by drugs (). Although these extended models differ in detail (; ; ), they all contain, as a common core, the basic cycle depicted in our figure insets, which describes the majority of gating events for both WT CFTR and many CFTR mutants, including ΔF508 (; but compare to , for ATP-independent G551D). We will therefore use the core cyclic model as a framework for discussing our results.

Polyunsaturated fatty acids (PUFAs) are naturally occurring substances with important functions in normal physiology. As a component of the cell membrane, PUFAs and other fatty acids can directly affect the activity of membrane proteins like voltage-gated ion channels (; ; ; ). In addition, free PUFAs can affect different ion channels (; ), and their beneficial effects on heart arrhythmias and epilepsy have been known for a while (; ; ; ; ; ; ). PUFAs have been suggested to regulate neuronal excitability by closing sodium or calcium channels (; ) and/or by opening potassium (K) channels (, ; ; ; ). A voltage-gated ion channel consists of a pore-forming unit surrounded by four voltage sensor domains (VSDs; ; ). Each VSD is composed of four transmembrane helices (S1–S4), where S4 contains several regularly spaced positively charged amino acid residues (). These positive charges respond to alterations in the membrane voltage by sliding along negative countercharges in S1–S3, and this movement regulates whether the channel is open or closed (; ; ; , ; ; ; ). The VSD of voltage-gated K channels can be in at least four closed/resting configurations (C1 to C4) and one open/activated configuration (O; ; ). PUFAs open the voltage-gated Shaker K channel by shifting the voltage dependence of the opening toward more negative voltages (, ; ; ). The PUFA molecule is suggested to be inserted into the lipid membrane, close to or in direct contact with the ion channel (). The negative charge of the fatty acid attracts the positively charged voltage sensor S4, primarily the last molecular conformational step (C1→O) that swings out the top charge (arginine R362 called R1) against the lipid bilayer (; ; ). Thus, R1 is suggested to be a key player in determining the sensitivity to PUFA and PUFA-like molecules (). From an earlier study, we know that the Shaker ILT mutant is more sensitive to docosahexaenoic acid (DHA) compared with the WT Shaker channel, supporting an effect by DHA on the last opening step (). In the present investigation, we aimed to construct a channel with increased sensitivity to PUFAs. Such a channel would (a) gain us more insight into the biophysical mechanism of action of PUFA, (b) explain why some ion channels are more sensitive for PUFAs than others (), and (c) function as an important tool in the search for new substances with lipoelectric properties, acting as drugs against epilepsy, cardiac arrhythmias, and pain. In addition, we also report on electrostatic channel-opening effects of a small-molecule compound. Experiments were performed on the Shaker H4 channel (), made incapable of fast inactivation by the Δ(6–46) deletion (). Mutagenesis, cRNA synthesis, oocyte preparation, cRNA injection, and oocyte storage were performed according to the procedures described previously (; ). Animal experiments were approved by the local Animal Care and Use Committee at Linköping University. Currents were measured with the two-electrode voltage-clamp technique (GeneClamp 500B amplifier, Digidata 1440A digitizer, and pClamp 10 software; Molecular Devices) 1–6 d after injection of RNA. The amplifier’s leak and capacitance compensation were used, and currents were low-pass filtered at 5 kHz. All experiments were performed at room temperature (20–23°C). The holding voltage was set to −80 mV (−120 mV for the L361R/R362Q mutant), and steady-state currents were achieved by stepping to voltages between −80 and 50 mV (adjusted for some of the mutants) for 80 ms in 5-mV increments. The control solution contained (mM): 88 NaCl, 1 KCl, 15 HEPES, 0.4 CaCl, and 0.8 MgCl. pH was adjusted to 7.4 with NaOH, yielding a final sodium concentration of ∼100 mM. Pure control solution was added using a gravity-driven perfusion system. DHA was prepared, stored, and applied as previously described (). Arachidonyl amine (AA) was provided by T. Parkkari (University of Eastern Finland, Kuopio, Finland) and prepared, stored, and applied as previously described (). For AA measurements, cells were preincubated in 1 µM indomethacin, and all recording solutions were supplemented with 1 µM indomethacin, as previously described (), to prevent COX-induced metabolization of AA. Pimaric acid (PiMA) was obtained from Alomone Labs and treated as DHA (however, the stock concentration of PiMA was 50 mM). The effective DHA, AA, and PiMA concentrations were assumed to be 70% of the nominal concentration because of binding to the chamber walls (). All concentrations given in the main article are the effective concentrations. To improve the washout of DHA and PiMA, albumin-supplemented (100 mg/liter) control solution was added manually to the bath, followed by continuous wash by control solution. For low concentrations of DHA, the recovery was almost complete, but for higher concentrations less complete. For 70 µM DHA, the recovery ranged from 40 to 85% for the different mutants. All chemicals were obtained from Sigma-Aldrich, if not stated otherwise. The K conductance () was calculated aswhere is the steady-state current at the end of an 80-ms pulse, is the absolute membrane voltage, and is the reversal potential for the K channel, set to −80 mV. and for Shaker mutants was determined by fitting a simple Boltzmann ( = 1) curve to the conductance data:where is the amplitude of the curve and and are the midpoint and the slope, respectively. The DHA-induced shift of the () curve was quantified at the 10% level as previously described (). For illustrative reasons, the figures presented in this manuscript were generated by fitting a Boltzmann curve raised to the power (i.e., no restriction for ) to the conductance data. When several concentrations (in increasing order) of DHA were applied on the same oocyte, all shifts were calculated compared with the first control curve. Average values are expressed as mean ± SEM. When comparing DHA-induced shifts of mutants with control (R362Q), one-way ANOVA together with Dunnett’s multiple comparison test was used. When comparing groups, one-way ANOVA together with Bonferroni’s multiple comparison tests was used. Correlation analysis was performed by Pearson’s correlation test and linear regression. P < 0.05 is considered significant for all tests. The crystal structure of the Shaker K channel is not determined. Therefore, we used the structure of the Kv1.2/2.1 chimera channel () with Shaker side chains () for the structural evaluations. The Kv1.2/2.1 chimera shares high sequence identity with the Shaker K channel and has previously been shown to serve as an accurate Shaker model (). This chimera was also used for generation of models for the closed states as previously described (). Fig. S1 shows representative () curves and DHA-induced () shifts for arginine mutants. Fig. S2 shows the correlation between values of () curves and DHA-induced () shifts for arginine mutants. Fig. S3 shows representative data for the AA effect on A359E/R362Q. Online supplemental material is available at . To explore the place dependence of the top arginine (R1) in S4 for DHA effects, we expressed the Shaker K channel in oocytes and measured ion currents with a two-electrode voltage-clamp technique. We studied eight channels in which either the region 356–362 was neutral or in which one charge at a time was introduced in each of the seven positions 356–362 (). All mutants expressed well, and the opening kinetics was essentially unaffected by the mutations (opening time constants were 0.5–2.0 times the opening time constant for the WT channel). By shifting the positive charge along S4, the voltage dependence for the conductance versus voltage, (), curve also shifted along the voltage axis (); some mutated channels were opened at more positive voltages (green symbols) than the channel with a neutral 356–362 segment (open symbol), and some were opened at more negative voltages (red symbols; for () curves, see ). The two groups of residues are positioned on opposite sides of the S4 helix (). The simplest explanation for this orientation dependence is that a charged residue dislikes a hydrophobic environment and therefore destabilizes either the open state or closed state depending on in which state a particular residue faces the lipid bilayer (). Side chain interactions within the VSD may also contribute to the voltage dependence of the arginine mutants: The right-shifted () curve for M356R and A359R is probably supported by charge interactions with negative countercharges in S2 and S3 normally interacting with the gating charges in S4 (; ), whereas the open conformation for S357R, L358R, and L361R might be stabilized by interactions with negatively charged clusters in the S1–S2 loop or the S3–S4 loop (). Shifting the top charge along S4 also altered the Shaker K channel sensitivity to DHA. Moving the arginine from position 362 to 361 abolished the effect of DHA (), whereas moving it two steps further clearly potentiated the effect (). A summary of DHA-induced shifts of the () curves for all mutations () shows that three (yellow bars) are not significantly different from the neutral mutant (open bar), two are significantly less affected (red bars), and two are significantly more affected (green bars), with A359R/R362Q as the most potent mutation, nearly doubling the shift from −6.1 ± 0.6 ( = 9) to −11.8 ± 0.8 mV ( = 9; Fig. S1). When coloring the mutated residues in two suggested Kv channel structures (), C3 and O, based on their impact on DHA sensitivity, a clear pattern emerges (). The red residues point toward the lipid bilayer and the expected position of DHA () in the closed C3 state, supporting the lack of channel-opening effects of DHA on these mutations. The green residues point toward the lipid bilayer and the expected DHA position in the open state, thus supporting DHA-induced promotion of channel opening. The yellow residues are positioned toward the lipid bilayer in the open state, but at larger distances (vertically or horizontally) from the expected position of DHA in the open state, supporting a lack of electrostatic interaction. There is a significant correlation between the DHA-induced shifts of voltage for activation and the voltage required to reach 50% of the maximum conductance (; ). This correlation suggests that, at least to some extent, the mechanisms for the induced shifts and the opening of the channels are common. A similar correlation has been found for metal ion effects on different WT as well as mutated K channels and suggested to depend on alterations in fixed surface charges (; ; ). In contrast, there is no correlation between the slope of the fitted Boltzmann curves of the arginine mutants and the DHA-induced shifts, suggesting that the altered sensitivity to DHA does not depend on altered energy barriers between the states with altered state occupancy as consequence (). To further test the hypothesis that there is an electrostatic interaction between charges in S4 and the DHA molecule, and not simply that additional charges in S4 dislike the lipid bilayer or hydrophobic pockets of the channel protein, we tested whether negatively charged glutamates at three consecutive positions, 359 to 361, had opposite effects to the arginines. A glutamate in position 359 significantly reduced the DHA-induced () shift from −6.1 ± 0.6 to −3.8 ± 0.4 mV ( = 7; ). This alteration is in the opposite direction of the positively charged arginine. The 359E mutant was further investigated by testing whether AA, a positively charged PUFA analogue, shows the opposite effect compared with DHA. Although AA previously was shown to close the Shaker WT channel by shifting the () curve in a positive direction along the voltage axis (), AA opened the 359E channel by shifting the () curve by −8.9 ± 1.7 mV ( = 4; ), which is significantly more than the shift induced by the negatively charged DHA on the same channel (−3.8 ± 0.4 mV). These experiments clearly support an electrostatic mechanism. In contrast to the experiments for 359 above, swopping the charge at position 361 from positive to negative had the opposite effect; the DHA-induced shift increased from −0.8 to −7.0 mV (), slightly more than for R362Q. The AA effect on 361E was not possible to evaluate because AA induced severe cellular toxicity for this mutant (see also for toxicity description) and caused substantial inactivation of 361E. For position 360, there was no difference between the glutamate and the arginine (). Calculations of the electrostatic effects are complicated because of several unknown parameters, like the exact positions of the amino acid charges, the dielectric constant, and uncertainty regarding the number of bound DHA molecules. However, the data are consistent with a simple model, locating one DHA molecule in the lipid bilayer just outside of S4 as indicated in . The last step in the channel opening, the transition from C1 to O (), is the step most sensitive to DHA (). During this step (, arrows), 359 approaches the negatively charged DHA molecule, 361 departs from the DHA molecule, and 360 keeps a relatively constant distance from the DHA molecule (). A positive charge at 359 or a negative charge at 361 promotes opening. A negative charge at 359 or a positive charge at 361 prevents opening. Notably, the negative glutamates have less of an effect than the positive arginines, compared with R362Q, probably because the negative charge of the glutamate can partly push away the negative DHA molecule from the channel or deprotonate the DHA molecule. These data strongly suggest that S4 is required to rotate in the last step to open the channel and that DHA electrostatically affects this rotation. To search for a channel with even higher sensitivity to DHA than the single mutants, we explored the effect of DHA (70 µM at pH 7.4) on several combinations of positively charged residues in the positions 356, 358, 359, 360, and 362 (). The potentiating effect of A359R was independent (i.e., additive) of single-positive charges in positions 356, 358, or 362, and thus, A359R could easily be combined with these arginines to gain a larger sensitivity to DHA. In contrast, the potentiating effect of I360R was abolished in various combinations with A359R. In combination with A359R alone, I360R even abolished the potentiating effect of an arginine at position 359. Thus, I360R was not a good candidate for the construction of a highly DHA-sensitive channel. The L358R/A359R/R362Q mutant was generated to mimic the charge profile of the Kv2.1 channel because this channel is known to be sensitive to DHA (). Also in this system, the Kv2.1 look-a-like channel was sensitive to DHA, which displayed a relatively large DHA-induced shift of −10.7 ± 1.2 mV ( = 8), but this effect was gained from 359R rather than 358R. The triple-R mutant M356R/A359R/R362R (hereafter referred to as 3R) turned out to be the most sensitive channel and was thus further explored. There are no major differences between the activation kinetics of the 3R mutant and the control channel (i.e., a difference less than a factor of 2; ). For the 3R mutant, 70 µM DHA at pH 7.4 increased the current 11-fold at 10 mV () and shifted the voltage dependence by −19.6 ± 1.0 mV ( = 15; ), a shift more than three times larger than that for the R362Q and WT channels. In a previous study, we showed that the DHA-induced () shift of the Shaker WT channel was pH dependent (). The pH dependence is explained by incomplete deprotonation of DHA at pH 7.4, as the apparent pKa value of PUFAs in a lipid membrane is ∼7.5 (). Thus, the effect of DHA on the 3R channel was further investigated by altering the pH. At −15 mV and pH 9.0, a DHA concentration ≥ 3 µM clearly increased the current; 7 µM caused a 10-fold increase (orange trace), and 70 µM caused a 40-fold increase (green trace; ). Increasing concentrations of DHA at pH 9.0 gradually shifted the voltage dependence to more negative voltages (up to −60 mV), with minor effects on the maximal conductance (). However, at high concentrations, DHA also induced some inactivation, as previously reported for the WT Shaker channel (). The DHA-induced shift caused by the highest DHA concentrations may therefore be the result of two separate mechanisms: lipoelectric opening of the channel and channel inactivation (possibly by DHA-induced conformational changes of the selectivity filter or block of the gate). As inactivation mainly occurs at the more positive voltages, we expect only minor effects of DHA-induced inactivation on the () shifts measured at the 10% level. DHA concentrations >210 µM were not tested because free fatty acids form micelles around 100 µM in physiological solutions (; ). The effect of DHA on the control channel versus the 3R channel was measured at different concentrations at pH 7.4 and 9.0 (). From these data, it is clear that the DHA sensitivity of the WT channel at pH 9.0 is similar to the DHA sensitivity of the 3R channel in pH 7.4. The increased DHA sensitivity of the 3R channel is likely explained by the additional three positive charges in the 3R channel that all point toward the lipid bilayer and therefore are able to electrostatically interact with DHA (). Thus, although 70 µM DHA at pH 9.0 shifted the () of the WT channel by −18.0 ± 1.4 mV ( = 9; ), it shifted the () of the 3R channel by −47.9 ± 4.2 mV ( = 4). At low voltages these shifts can be converted to equivalent increases in current magnitude, = exp(−Δ/4.7; ). The amplitude increase from WT to 3R is thus / = exp(−ΔΔ/4.7). Thus, introducing two extra positive charges at positions 356 and 359 increases the open probability of the Shaker K channel by a factor of 580 at 70 µM DHA at pH 9.0. This large potentiation simplifies the search for other compounds with similar properties. DHA and other PUFAs display promising anti-excitable effects on the Shaker K channel (, ; ; ; ). However, they are very promiscuous, a characteristic not appreciated in drug design. Therefore, in an attempt to find other candidate drugs, we have searched for small-molecule compounds with similar effects to DHA. PiMA, an amphipathic resin acid known to open Ca-activated BK channels () has, in conformity with the fatty acids, a lipophilic domain and a carboxyl group supposed to be negatively charged at high pH (). Therefore, PiMA is also a possible candidate to open Kv channels. For the WT Shaker K channel, 70 µM PiMA at pH 7.4 shifted the () by −4.6 ± 0.7 mV ( = 15; ). For the 3R channel, the shift was almost doubled to −8.4 ± 1.2 mV ( = 4; ), indicating that PiMA also activates the Kv channel by electrostatically affecting the positive charges in the top of S4. Similar to the DHA effect, the PiMA effect was potentiated at increased pH (). Thus, we have found a small-molecule compound able to shift the () of the 3R channel at pH 9.0 by almost −30 mV. In this study, we have investigated introduced charged amino acid residues in S4, and combinations of them, and how they affect the DHA-induced alteration of the Shaker K channel’s voltage dependence. We found that the single mutations A359R and I360R and combinations including A359R, in particular 3R, significantly increased the sensitivity to DHA. We have also identified a small molecule compound, PiMA, with similar effects as DHA on the channel’s voltage dependence. Furthermore, we found that residues on the opposite side of S4, S357R and L361R, significantly reduced channel sensitivity to DHA. The place dependence of arginines and glutamates for the DHA sensitivity supports a rotational S4 movement in the last opening step that is promoted by negatively charged lipophilic compounds, like PUFAs and PiMA. This model is supported by data for the positively charged PUFA analogue AA, which shows opposite the effects to DHA on 359E. An understanding of the molecular details for the PUFA–channel interactions is important for explaining and predicting differences in PUFA sensitivity between channels. One of the channels constructed in this study mimics the S4 arginine profile of Kv2.1 (L358R/A359R/R362Q). The Kv2.1-mimicking mutant demonstrated increased DHA sensitivity, in line with experimental findings reporting clear shifts in Kv2.1 channel voltage dependence from low micromolar PUFA concentrations (). BK channels also display a favorable S4 charge profile for possible PUFA effects by having a positive charge at the position equivalent to 359 in the Shaker K channel. BK channels are reported to be highly sensitive to PUFAs (; ; ,). However, these large effects on the BK channel could, at least partly, be explained by another mechanism because residues on the intracellular side of the channel are involved in the effect (). Other DHA-promoting or -preventing charge profiles are found in different pseudotetrameric sodium and calcium channels. However, the physiological relevance is difficult to evaluate because only one out of four VSDs is expected to show increased or decreased PUFA sensitivity. We have previously reported that part of S4 adapts a 3-helical structure and that the segment forming a 3 helix slides along the S4 helix during gating (). In our previous study, we proposed that the 3-helical structure is downstream of R3 in the last opening step, the step most sensitive to DHA. Because the electric force varies inversely with the square of the distance between two charges, the electrostatic interaction between DHA and the gating charges in the 3-helical segment during the opening step will be weak. Hence, the 3 helix structure and the sliding of that structure will likely be preserved in the presence of DHA. The main finding in the present work is the 3R mutant, a constructed channel with an increased sensitivity to PUFAs with respect to the channel’s voltage sensitivity. 70 µM DHA increased the current of the 3R channel at pH 9.0 and negative voltages >500 times the WT channel. Thus, the 3R channel is a promising tool in the search for pharmacological compounds with beneficial affects against cellular hyperexcitability in diseases such as cardiac arrhythmia, epilepsy, and pain. As a first example of this, we here demonstrate 3R channel sensitivity to PiMA, a novel Shaker-channel opener.

In 2008, two members of the anoctamin superfamily, Ano1 and Ano2, were found to encode Ca-activated Cl channels (CaCCs; ; ; ). Since then, it has been shown that Ano1 (also known as Tmem16A) plays key roles in diverse physiological processes. Ano1 mediates Ca-dependent fluid transport by a variety of epithelia (), including salivary gland (), airway (), and bile duct (). Furthermore, Ano1 modulates mucin secretion by airway epithelium (), regulates slow wave motility of the gut (; ; ; ), participates in nociception by dorsal root ganglion neurons (; ), regulates vascular and airway smooth muscle contraction (; ; ; ; ; ; ), and may participate in the sperm acrosome reaction (). Additionally, it has been suggested that Ano1 may impact cell proliferation and metastasis (; ; ; ). Ano1 is activated by increases in cytosolic Ca concentration with an EC in the low micromolar range (; ), but the gating mechanisms remain unresolved. Two possible mechanisms have been considered: binding of Ca directly to the channel or binding of Ca to a separate Ca sensor protein such as calmodulin (CaM). We have proposed that Ano1 is regulated directly by Ca binding to the channel because mutagenesis of two amino acids, E702 and E705, alters the Ca sensitivity of the channel by several orders of magnitude (). Mutation of homologous residues in the Ano1 paralogue Ano6 also dramatically decreases its Ca sensitivity (). Despite these dramatic results, their interpretation is ambiguous. The simplest interpretation is that these mutations alter a Ca-binding site. However, allosteric consequences of mutations are difficult to exclude: for example, the mutation might alter the association of an accessory Ca sensor. Furthermore, gating of the channel by direct Ca binding has been questioned because the Ano1 sequence does not contain canonical Ca-binding motifs and a sequence in the first intracellular loop resembling the “Ca bowl” of the large conductance Ca-activated K channel does not appear to be a principal Ca-binding site (; ). Several studies have implicated a role of CaM in regulating Ano1 currents. have reported that trifluoperazine (TFP) or J-8, classical inhibitors of CaM, decreases activation of Ano1() by Ca. They identified an alternatively spliced segment (termed segment ) in Ano1 as a CaM-binding site. However, the conclusion that CaM is required for Ano1 activation is contradicted by the robust Ca-dependent activation of the Ano1() splice variant that lacks the segment (). Recently, reported that CaM binds in a Ca-dependent manner to two different sites in Ano1(): one site (CBM1) is immediately N-terminal to the putative first transmembrane segment, and the other is immediately N-terminal to the putative seventh transmembrane segment (CBM2). show that Ca-CaM binding increases the relative Cl:HCO permeability of the channel. While our paper was in review, two other papers appeared on the role of CaM in Ano1 regulation. report that TFP and J-8 do not inhibit Ano1 currents (in contrast to ), but they show that purified peptides corresponding to CBM1 bind to CaM and that overexpression of certain CaM mutants reduces Ano1 current density. Finally, have purified recombinant Ano1 expressed in Sf9 cells and find that the purified Ano1 protein reconstitutes Ca-activated Cl channels in liposomes without the presence of CaM. Furthermore, they were unable to find evidence that CaM bound to purified Ano1 by a variety of assays with the purified proteins. In light of these conflicting results, we wanted to clarify whether activation of Ano1 by Ca involves CaM. There are at least six ways by which CaM can regulate its effectors (). Class A effectors bind essentially irreversibly to CaM irrespective of Ca concentration. In these effectors, CaM is a “constitutive” subunit of the protein complex. The effector is activated allosterically by Ca binding to the tethered CaM. This mode of CaM signaling has been demonstrated to be responsible for activation of the small-conductance K (SK) channel by Ca () and for Ca-dependent facilitation and inactivation of voltage-gated Ca1 and Ca2 channels (). Regulation of class A effectors by CaM is relatively insensitive to pharmacological inhibitors like TFP. Class B effectors also bind Ca-free CaM (apo-CaM), but Ca binding stimulates CaM dissociation from the effector. Class C effectors bind to CaM with low affinity at low Ca concentrations (<2 mol Ca/mol CaM), whereas at high Ca concentrations, class C effectors form a high-affinity complex with Ca-CaM. Class D and E effectors do not bind apo-CaM, but bind Ca-CaM reversibly and are inhibited or activated, respectively, by Ca-CaM binding. Class F effectors are activated by phosphorylation, but phosphorylation is promoted by CaM binding to the effector. The results of and are consistent with Ano1 being a class E effector because CaM binding is only observed in the presence of Ca and the activation of the channel by Ca is inhibited by TFP. However, the data we present here are more consistent with the conclusion that Ano1 activation by Ca occurs by direct binding of Ca to Ano1 and that the activation of Ano1 by Ca is not mediated by CaM. mAno1() tagged with EGFP at the C terminus was provided by U. Oh (Seoul National University, Seoul, South Korea). Unless indicated otherwise, experiments used mouse Ano1 (mAno1). Both the (UniProt accession no. ) and (QGEGRRKDSALLSKRRKCGKYG inserted after position 266 in Ano1) splice variants were used. SK2 cDNA was provided by J. Adelman (Vollum Institute, Portland, OR). cDNA was transfected into HEK293 cells (0.1–1 µg total DNA per 3.5-cm plate) using Fugene-6 (Roche). Single cells identified by EGFP fluorescence were used within 72 h. Transfected HEK293 cells were recorded using conventional whole-cell or excised inside-out patch-clamp techniques with an EPC-7 amplifier (HEKA). Fire-polished borosilicate glass patch pipettes were 3–5 MΩ. Experiments were conducted at room temperature (20–24°C). Because liquid junction potentials were small (<2 mV), no correction was made. The zero Ca intracellular solution contained (mM): 146 CsCl, 2 MgCl, 5 EGTA, 10 sucrose, and 10 HEPES, pH 7.3, adjusted with NMDG. The high Ca pipette solution contained 5 mM Ca-EGTA (free Ca ∼20 µM) instead of EGTA. Solutions with various free Ca concentrations were made by mixing CaCl with EGTA or hydroxyethyl-EDTA (HEDTA), as calculated by the MaxChelator Program (), and the free Ca concentration was verified using a Ca ion–selective electrode. In experiments with SK2 channels, 146 mM KCl replaced equimolar CsCl in the high-Ca pipette solution. Ba-containing solutions were made by adding BaCl and 1 mM EGTA to Chelex-100–treated zero-Ca solution to generate a free calculated Ba concentration using the MaxChelator program. EGTA affinity is much lower for Ba than for Ca. The standard extracellular solution contained (mM): 140 NaCl, 5 KCl, 2 CaCl, 1 MgCl, 15 glucose, and 10 HEPES, pH 7.4 with NaOH. For recordings of SK2 currents, 140 mM KCl replaced equimolar NaCl in the extracellular solution. Purified CaM (EMD Millipore) was added to recording solutions from a stock solution of Ca-CaM. Osmolarity was adjusted with sucrose to 303 mOsm for all solutions. Cells were placed on the stage of an Axiovert inverted microscope (Carl Zeiss). The microscope condenser assembly was replaced with a 100-W xenon model JML flash lamp (Rapp Optoelektronik GmbH) filtered by a UG11 filter (∼300–400 nm band pass) focused onto the recording chamber with an 18-mm-focal-length lens, which produced a 4-mm-diameter spot of illumination (; ). Flash intensity was adjusted by changing the condenser charging voltage. We typically used 4-mJ/mm flashes. The duration of the flash was <1 ms. Cells were loaded with caged Ca, -nitrophenyl–EGTA (NP-EGTA; Invitrogen; ), from the patch pipette. The pipette solution contained (mM) 2 NP-EGTA, 136 CsCl, 1.5 CaCl, 1 MgCl, and 25 HEPES-NMDG, pH 7.5. Before photolysis, the free Ca concentration was calculated to be 80 nM and the Ano1 current was typically l ≤ 100 pA in amplitude. The fast application of Ca to excised inside-out patches was performed using double-barreled theta tubing (1.5 mm o.d.; Sutter Instrument) with a tip diameter of ∼50 µm attached to a Piezo bimorph on a micromanipulator (; ). One barrel was filled with standard zero-[Ca] solution, and the other barrel was filled with intracellular solution containing Ca (or Ba for experiments in ). Excised patches were positioned at the opening of one barrel of the theta tubing, and the barrel flooding the patch was switched by applying ∼100 V to the Piezo bimorph. The time course of solution exchange across the laminar flow interface was estimated by liquid junction potential measurements to be <5 ms. For immunoprecipitation experiments in , HEK293 cells were transfected with Ano1-EGFP or SK2 plus CaM tagged on the N terminus with FLAG and myc (provided by R. Dolmetsch and C.Y. Park, Stanford University, Stanford, CA; ). 2–3 d after transfection, one set of dishes was treated with 10 µM ionomycin for 4 min in Hanks BSS to elevate intracellular Ca, and the other set was untreated. The untreated cells were lysed in Ca-free Buffer A (mM): 150 NaCl, 10 HEPES, 1 EGTA, and 0.1 MgCl, pH 7.4, containing 0.5% Triton X-100 and Complete Anti-protease (Roche). Ionomycin-treated cells were lysed in Ca-containing Buffer A that lacked EGTA and had 1 mM of added CaCl. For ionomycin-treated cells, Ca was present in all subsequent steps. The clarified supernates were applied to 20 µl anti-FLAG antibody–coated Dynal magnetic beads and incubated for 2 h. The beads were washed three times with 300 µl Buffer A (with EGTA or Ca, as appropriate). Proteins were solubilized from the beads with 50 µl Laemmli SDS sample buffer, electrophoresed on 4–20% acrylamide SDS-PAGE gels, and subjected to Western blotting using anti-Ano1 (Ano1-EGFP = 150 kD) and anti-myc (CaM = 17 kD). For pull-down experiments in , HEK293 cells were transfected with human Ano1 (hAno1; provided by M.G. Lee, Seoul National University; ), mAno1-FLAG(3X), or CaMKIIα-Venus (plasmid #29427; Addgene; ). HEK cell lysate (300 µl containing 750 µg protein) was mixed with 600 µl of binding buffer (with EGTA or Ca) and 120 µl CaM-GST–Sepharose beads containing 450 µg CaM as described by . After an overnight incubation at 4°C, bead complexes were washed five times. Bound protein was eluted with 60 µl SDS sample buffer and immunoblotted with antibodies against hAno1 (Dog1.1; gift of R. West, Stanford University), anti-flag (M2 monoclonal; Sigma-Aldrich), or anti-CaMKII (Sigma-Aldrich). Electrophysiological traces were analyzed with Clampfit 9 (Molecular Devices). Data are presented as mean ± SEM. Statistical difference between means was evaluated by two-tailed test. Statistical significance was assumed at P < 0.05. Immunoblots were scanned using an Epson Perfection V700 desktop scanner and analyzed using myImageAnalysis version 1.1 software (Thermo Fisher Scientific). As a first step to exploring the mechanism of Ano1 activation by Ca, we first wanted to know the speed of Ano1 activation by Ca. To determine the rate of Ano1 activation by Ca, we recorded Ano1 currents in response to photolysis of caged Ca (NP-EGTA) with whole-cell patch-clamp using solutions that lacked ATP to minimize phosphorylation. With a 4-mJ/mm intensity UV flash, Ano1 currents activated very rapidly. Fitting the rising phase of the current to the sum of two exponentials revealed that the major component of the current increased with a time constant τ = 1.2 ± 0.1 ms (see legend for details). The precise instant that the current began to increase was obscured by the flash artifact, but invariably the current had increased significantly above baseline by the end of the 600-µs flash (). Some of the latency may be explained by the rate of Ca release from NP-EGTA, which is 6.8 × 10/s as measured with a 180-mJ/mm laser flash at the optimal excitation wavelength of 357 nm (; ). Because the energy density of our flash was >40-fold smaller, the rate of Ca release is likely slower. This rapid activation rate is comparable with rates reported for the class A CaM effector SK channel () and ligand-gated ion channels such as the nicotinic ACh receptor (). The activation rate is slower than would be expected if Ano1 was gated open by Ca-stimulated phosphorylation. Although protein kinases can operate at rates as high as 20/s (), Ca activates the Ano1 current at a rate of >800/s (τ = 1.2 ± 0.1 ms) without added ATP. The observation that the Ano1 current is activated in the absence of added ATP () suggests that Ca does not activate the channel by stimulating protein phosphorylation. However, in whole-cell recording, phosphorylation might occur because the cell may produce ATP endogenously by glycolysis or oxidative phosphorylation. Furthermore, activation of Ano1 could hypothetically be explained by Ca-dependent dephosphorylation of a previously phosphorylated channel. To test these possibilities, we examined activation of the Ano1 current in excised inside-out patches where cytosolic ATP could be rigorously controlled. Ano1 current was activated by Ca in excised patches even when the cytosolic face of the excised patch was perfused with intracellular solution lacking any high-energy substrates (ATP, GTP, and glucose). Furthermore, the current could be turned on and off repeatedly by switching between Ca-free and Ca-containing solutions (). This result eliminates reversible dephosphorylation/phosphorylation as the triggering event in Ca-dependent opening and closing of the Ano1 channel. The result in places certain constraints on an activation mechanism that involves gating by Ca-CaM. Because the current can be activated many times by repeatedly switching between Ca-containing and Ca-free solutions, the Ca sensor must be effectively anchored to the excised patch so that it does not freely diffuse away between Ca applications. In other words, it is unlikely that Ano1 is a class E CaM effector where Ca-CaM reversibly binds to the channel to open it (see Discussion for a more thorough discussion of CaM dissociation and diffusion rates). These experiments provide evidence that the activation of Ano1 is membrane delimited: freely diffusible, reversibly bound soluble proteins are unlikely to be necessary for activation of the current by Ca. Although the Ano1 current can be repeatedly activated in excised patches, the amplitude of the current runs down with time. In the example of , the current decreased exponentially with τ = 2.45 min (). The rundown could be explained by various mechanisms, but one possibility would be a slow loss of CaM from the patch. Rundown was quantified by exposing excised patches to cytosolic solution containing different concentrations of free Ca. With 1.4 or 3.4 µM free Ca in the solution bathing the cytosolic face of the patch, current rundown was relatively slow and incomplete during the recording (). With these low micromolar concentrations of Ca, Ano1 current decreased ∼25% over a period of 5 min and then stabilized (, open symbols). Higher concentrations of Ca (>10 µM) accelerated current rundown. With 20 µM Ca, the current ran down ∼55% in 5 min, whereas with 74 µM Ca, the current ran down >95% in <5 min (). Current rundown could not be reversed by washing the patch with zero-Ca solution for several minutes before switching back to Ca-containing solution. The finding that current rundown depends on the concentration of cytosolic Ca and that rundown is irreversible is consistent with rundown being caused by CaM loss from the patch. In the next section this hypothesis is tested experimentally. If the rundown of current is caused by loss of CaM, one would expect that addition of exogenous CaM to the cytosolic solution would restore the current. A large number of experiments were performed applying purified brain CaM at concentrations ranging from 1 to 25 µM to the cytosolic side of the patch. However, CaM application never reversed or obviously slowed rundown. Several examples are shown in . Addition of 10 µM to the cytosolic solution had no effect on currents carried by Ano1() () or Ano1() () in inside-out excised patches. The effect of CaM was quantified by measuring rundown for 1 min before CaM application and 1 min during CaM application (). With both isoforms, rundown occurred at statistically the same rates whether CaM was present or not. One possible explanation for the absence of effect of CaM is that its association with Ano1 is very slow and that a 1-min exposure is insufficient time for CaM to bind. However, applications of 10 µM CaM for >8 min had no effect on the rate or extent of rundown (, inset). Exponential fits of the rundown from the time of switch to CaM-containing or control solution were control: τ = 2.9 ± 0.4 min, y = 0.51 ± 0.02; and CaM: τ = 3.6 ± 0.7 min, y = 0.49 ± 0.04, where y is the level of the current remaining after rundown has completed compared with the initial current. After the current had run down completely in solutions containing high Ca concentrations, addition of CaM did not restore the current (). The finding that exogenous Ca-CaM had no effect raised the question of whether exogenous apo-CaM would have any effect. To test the effect of apo-CaM, the patch was first activated with 20 µM Ca and then the solution was replaced with a zero-Ca solution containing 25 µM apo-CaM that had been dialyzed against the zero-Ca EGTA solution (). Current amplitudes in zero-Ca solution and in zero-Ca plus apo-CaM were identical ( = 4). Subsequent exposure of the patch to 20 µM Ca evoked a current similar in amplitude to that expected if rundown had continued at a constant rate during apo-CaM exposure. Subsequent addition of 25 µM Ca-CaM had no effect (). The rate of current rundown was virtually identical in 20 µM Ca in the presence and absence of 25 µM CaM and before or after exposure to apo-CaM (). The data indicate that rundown is not caused by the loss of CaM from the patch and that Ano1 is not a class E CaM effector. Several caveats should be mentioned. Rundown may be a process unrelated to Ca-dependent gating, for example, a process involving loss or modification of requisite components including protein subunits or lipids. If that is the case, one would not expect CaM to reverse the rundown process even if CaM is required for activation of the current by Ca. The present experiments also would not exclude a mechanism in which rundown is caused by a loss of CaM from excised patches but the conditions required for rebinding of exogenous CaM are suboptimal or very slow. Class A CaM effectors, such as the small conductance K channel SK2 (; ), have CaM tethered to the pore-forming polypeptide as an integral subunit: both apo-CaM and Ca-CaM are tightly bound. It has been shown that overexpression of CaM mutants defective in Ca binding () can compete for native CaM and eliminate effects of Ca on class A effectors, including SK2 () and Ca channels (). Ca-insensitive CaM generated by mutagenesis of aspartate to alanine in the second, third, and fourth EF hands (CaM) or all four EF hands (CaM) greatly reduces SK2 currents (). shows that currents recorded from Ano1 coexpressed with wild-type CaM or CaM had the same mean amplitudes and exhibited similar voltage and time dependence and current-voltage relationships (). CaM had no effect on the outward current but produced a small increase in the inward current at potentials negative to −60 mV. Control experiments performed on the same day with the SK2 channel showed that both of the mutant CaM constructs completely eliminated SK currents (). From these experiments, we conclude that CaM is not a constitutively bound subunit of the Ano1 channel necessary for Ano1 activation by Ca. However, we cannot exclude the possibility that the mutant CaMs are incapable of binding to Ano1 even though they are capable of binding to SK2. Analysis of the mouse Ano1 sequence by the Calmodulin Target Database identifies two potential CaM-binding sites with scores >5 (out of a possible 10). Neither of these sites conform exactly to established CaM-binding sequences, but they resemble CaM-binding “1-8-14 motifs.” 1-8-14 motifs are characterized by bulky hydrophobic amino acids at positions 1, 8, and 14 and are typically found in class E effectors that bind reversibly to Ca-CaM (). The first site (LLSKRRKCGKYG, numbering based on the isoform) overlaps with the splice segment and is not present in the Ano1 isoform (; ). The second site (DLVRKYGEKVG; CBM1 []) is located immediately N-terminal to the putative first transmembrane segment and is present in both and isoforms. To determine whether CaM binds to either isoform of Ano1, we performed immunoprecipitation experiments on HEK293 cells cotransfected with CaM (FLAG and myc tagged) plus Ano1(), Ano1(), or SK2. To test whether the association of CaM with Ano1 is Ca dependent, immunoprecipitation was performed on cells with basal cytosolic Ca (presumed to be ∼100 nM) and on cells in which Ca was elevated with ionomycin before lysis. After lysis in Ca-free or Ca-containing buffer, respectively, CaM was immunoprecipitated using magnetic beads coated with anti-Flag antibody, and the immunoprecipitate was immunoblotted for the presence of Ano1, SK2, and CaM (). The immunoprecipitate from cells expressing CaM plus SK2 contained large amounts of SK2 (which appeared as monomers, dimers, and higher-order oligomers on the gel). In contrast, in cells expressing CaM plus either mAno1() or mAno1(), there was no Ano1 protein in the immunoprecipitate regardless of whether the lysate was obtained from cells with resting Ca or elevated Ca. The positive control with the SK2 channel validates the method. The absence of Ano1 coimmunoprecipitation with CaM supports our physiological data that CaM binding is not required for activation of the current by Ca. The conclusion from is different from that recently reported by . These authors demonstrate a physical interaction between human Ano1() and CaM-GST in a pull-down assay. Therefore, we tested the ability of CaM-GST beads to pull down mouse Ano1() human Ano1() from lysates of transfected HEK cells (). First, we performed a positive control to evaluate this method. We prepared lysates from HEK cells overexpressing CaMKIIα and examined the ability of CaM-GST to capture CaMKIIα from the lysate (). As estimated from densitometric scans of band intensity, ∼10 times more CaMKIIα bound to CaM-GST beads in the presence than in the absence of Ca (). To estimate the fraction of CaMKIIα in the lysate that was captured by the beads, we compared the intensity of the CaMKIIα band bound to the beads to various dilutions of the lysate (lanes 1–4). Densitometric scans show that virtually all of the CaMKIIα in the lysate was captured by the beads (). Consistent with the results of , we found that two times more hAno1() bound to CaM-GST in the presence of Ca than in the absence of Ca (). A similar result was obtained with mAno1() (). To evaluate the fraction of Ano1 that bound to the CaM-GST beads, we compared the amount of Ano1 that bound to the CaM-GST beads relative to the amount of Ano1 present in the lysate. Different amounts of lysate were loaded onto the gel, and the densitometric area of the bands was measured. Less than 1% of either hAno1() or mAno1() in the lysate bound to the CaM-GST beads (). As shown above, this compares with ∼100% of CaMKIIα that binds. These results show that although CaM may bind to Ano1(), the binding is less robust than the binding to CaMKIIα. This could be explained if CaM has a much lower affinity for Ano1 than for CaMKIIα or if the stoichiometry of CaM binding to Ano1 is low, for example if CaM binds only to Ano1 in a certain state (conformational, posttranslational modification, etc.) that is poorly represented in the Ano1 preparation we have used. Alternative explanations are considered in the Discussion. have shown that Ba has an extremely low affinity for CaM. The affinity of divalent cation binding to CaM exhibits the sequence Ca > Sr >> Mg > Be > Ba. Furthermore, the maximal effect of Ba on CaM conformation as measured by tyrosine fluorescence was <2% as large as that produced by Ca. As an additional test of the involvement of CaM, we tested the effect of Ba on Ano1 currents. In both whole-cell recording and in inside-out excised patches, >10 µM Ba activated large Ano1 currents (). In whole-cell recording, the EC was estimated to be 25 µM Ba (). To estimate Ba affinity in inside-out excised patches, we corrected for variations in the size of the patch and number of channels contained in different patches by normalizing the currents to the maximum Ano1 current activated by 1 mM Ba in each patch. Furthermore, to correct for rundown, patches were exposed to 1 mM Ba, then the test Ba concentration, and then again 1 mM Ba, each for ∼5 s. The response to the test concentration was then expressed as a fraction of the mean of the responses to the two 1 mM Ba applications. These experiments show that the EC for Ba in excised patches is very similar to whole-cell recording, 14 µM. The differences may reflect rundown that occurs in whole-cell recording before complete equilibration of the pipette solution with the cytoplasm. The EC for Ba is ∼50-fold larger than the EC we have estimated for Ca (). Additional support for the conclusion that Ca does not activate Ano1 by binding to CaM is the finding that Mg does not activate Ano1, despite the fact that the affinity of CaM for Mg is higher than for Ba (). 2 mM Mg is present in our zero-Ca intracellular solution, but it does not activate Ano1. Another piece of evidence that has been offered to support the hypothesis that Ca activates Ano1 by binding to CaM is the finding that CaM antagonists such as TFP suppress the Ano1 current (), although found that TFP did not inhibit Ano1. TFP is notoriously nonspecific. For example, TFP inhibits the gating of the large-conductance Ca-activated K channel (BK channel) with an apparent of 1.4 µM by a mechanism independent of CaM (; ), and TFP is known to disrupt the physical properties of phospholipid monolayers at a low molar ratio of TFP to lipid (). Therefore, we wanted to examine the mechanisms of TFP action on Ano1. confirms another report () that extracellular 10 µM TFP suppresses Ano1 current. If TFP is blocking Ano1 by binding to CaM, one would expect that the site of action of TFP would be intracellular. Because TFP is hydrophobic, it is possible that extracellularly applied TFP crosses the membrane and acts intracellularly. To test this prediction, we applied concentrations of TFP as high as 100 µM to the cytosolic face of inside-out excised patches. No effect of TFP was observed, suggesting that TFP acts at a site that is not accessible to intracellularly applied drug (). As another test for the involvement of CaM, we tested the effect of a peptide inhibitor of CaM. The peptide corresponds to the high-affinity CaM-binding site on CaMKII (281–309) and would be expected to compete for CaM binding to its targets (). This peptide does not cause any effect when applied to the cytoplasmic side of excised patches (). xref #text

First discovered in voltage-gated cation channels and later identified in voltage-sensitive phosphatases and proton channels, the voltage-sensor domain (VSD) is a biological molecular device that responds to changes in transmembrane (TM) electrical potential (; ; ). Moreover, VSDs are demonstrably modular and widely distributed in both prokaryotic and eukaryotic cells, indicating an early origin and consequently vast evolutionary exploration of sequence space (; ). Because of their ubiquity in cellular electrical signaling, mutations in VSDs give rise to various heritable diseases (). The VSD consists of a four-helix bundle (denoted as S1–S4) embedded in a membrane (; ; ). The S4 helix contains a highly conserved motif of three to eight positively charged residues, referred to as “gating charges,” which occur at three-residue intervals and occupy a common helical face (; ; ). These positive charges are stabilized in the TM position through salt-bridging interactions with the negative “countercharge” residues of S1–S3 (; ; ). These charges exist in an aqueous-like environment as water molecules protrude into the VSD lumen (; ; ). A cluster of bulky hydrophobic residues partitions the lumen into two disconnected intracellular and extracellular regions (; ; ). In response to changes in membrane potential, S4 translates so as to sequentially transfer the gating charges across the hydrophobic region (; ). Because of the partial solvation of the lumen, the TM electric field is focused on this hydrophobic region (; ). A rather small displacement of S4 is therefore sufficient to achieve a complete transfer of positive charge across the field (). This conformational change in S4 allows the VSD to act as a transducer of electrical signals. Major mechanistic features of VSDs appear well identified (; ). Even so, the protein sequences encoding VSDs in various families can be very different (). Here, we aim to show how the functional requirements of voltage sensing have shaped the sequence distribution of VSDs during evolution. After addressing the specific problem of aligning the register of S4 helices in a statistically rigorous way, we construct a large multiple sequence alignment (MSA) of VSD sequences and search for patterns and regularities that reflect the evolutionary “design principles” underlying voltage-sensor function. Success in this endeavor implies that from the design principles we will be able to recapitulate experimental findings and, importantly, formulate novel, testable hypotheses concerning the structural and functional properties shared among VSDs. Site-specific residue frequencies are arguably the most intuitive feature that can be extracted from MSAs. Because the analysis may contain thousands of sequences from long-diverged organisms, a frequency distribution of amino acids can be extracted for each column/position of the MSA. The peculiar nature of this distribution is, in general, informative about the evolutionary pressure acting on that particular position (). For instance, a distribution peaked around tyrosine, phenylalanine, and tryptophan suggests a functional requirement for aromaticity at that position. Here, we use this sort of single-site analysis to detect “evolutionarily important” functional sites. Much information can also be extracted from joint frequency distributions involving pairs of positions. In the native folded state, each residue of the polypeptide chain engages in residue–residue interactions. The specific mutations allowed at a given position are then highly dependent on the chemical identity of neighboring residues. As a result, frequency distributions of amino acids at different positions are expected to be statistically dependent on each other. Detection of these correlations can potentially unveil the physical interactions defining the native structure. However, residue–residue correlations per se are not able to convey this information; in a highly connected network, such as the contacting residues in a folded protein, an extensive set of pairwise interactions can give rise to long-range indirect statistical correlations (). As a result, pairs of correlated positions are not necessarily in contact. Reconstructing the network of direct interactions entails the logical process of inference: within a stochastic framework (a probabilistic model), it is assumed that residues interact in pairs (are directly coupled) in such a way as to best reproduce the distributions of sequences observed in the MSA (). Analysis of these direct couplings (direct-coupling analysis [DCA]) is able to highlight functionally crucial residues as those involved in enzymatic active sites and regions of conformational change (; ; ). It is important to note that these methodologies rely on a large, confidently aligned MSA. Anticipating our findings, we show how analysis of site-specific frequencies and direct evolutionary couplings (ECs) unveils the major sequence determinants of voltage sensing. The requirement for a translation of S4 through the electric field results in strong conservation of the gating charges, the countercharges, and the S2 phenylalanine. With DCA, we identify unexpected networks of strongly evolutionarily coupled residues surrounding these conserved positions. We also identify sites that have been under significant evolutionary pressure and have not yet been tested for functional relevance. We suggest these sites as potential targets for new mutagenesis experiments. Additionally, we show that the helical interfaces between S1–S4 and S3–S4 do not feature strong ECs, suggesting that no specific residue–residue contacts were conserved along evolution apart from the canonical salt bridges. This lack of expected coevolution is shown to characterize the entire paddle motif, rationalizing chimeragenesis and deletion analysis experiments highlighting the modular, mobile nature of this region. Lastly, observation of specific evolutionarily coupled residue pairs involving positions 96, namely 25–96 and 49–96 (NavAb numbering), provides independent support for conformational models of the VSD along the activation pathway and suggests that positions 25, 49, and 96 have important roles in tuning the activation properties of the VSD. Profile HMMs are among the most important approaches to analyze ensembles of sequences. Conceptually, HMMs characterize sequences as stochastic processes: discrete time random walks across a set of states with defined state and transition probabilities (). In practice, HMMs are initialized with a protein sequence of interest and iteratively refined by scanning many millions of known protein sequences. Homologous sequences are then collected in an MSA, an array of statistically significant protein sequence alignments. From an abstract perspective, an MSA can be thought of as the physical record of evolution’s exploratory search for structurally and functionally analogous proteins. Subsequent sequence analysis is highly dependent on the quality and quantity of alignments in the MSA, and for this reason, we give special care to its construction. To train an HMM to identify and align VSD sequences from a large sequence database, we construct a small, confidently aligned “seed” MSA. In the seed MSA, we included homologues representing phylogenetically diverse VSD-containing families (). Initial multiple alignments were generated using ClustalW2 and manually refined such that the alignment was consistent with structural alignments (in the cases where high resolution structures were available) and TM-helix topology predictions (; ; ). See for full seed MSA. VSDs contain several conserved positive gating charges, which come at three-residue intervals on the S4 helix. However, not all VSDs contain the same number of gating charges, and this may lead to uncertainty in the alignment of S4, as three-residue shifts in the alignment register will still match positive charges (; ). We were interested in quantifying how uncertain such alignments actually were. Position-wise reliability of a sequence alignment can be calculated as the posterior probability of each symbol in the alignment being emitted by a pair HMM based on the Blosum62 substitution matrix (). As an example, for the pairwise alignment of human Hv1 S4 to that of human Kv1.2 S4, we calculated the positional posterior probability of (a) an alignment generated by the Needleman–Wunsch algorithm, and (b) alignments “sampled” from a pair HMM with Hv1 S4 in alternate registers using ppAlign (; ). The posterior probability over the S4 helix in (a) is much higher than any of the alignments in (b), suggesting that, given the widely used Blosum62 substitution matrix, the alignment of S4 is unambiguous. An HMM was trained based on the seed MSA using HMMER3.0 (). Scanning this HMM against the NCBI protein sequence database, we collected and aligned 6,652 unique VSD sequences containing all four TM helices (). Only alignments with an E-value of <0.01 were included. Sequence logos were generated with Weblogo3 (). To describe the constitution of the VSD MSA, hierarchical clustering was performed to partition the MSA into protein families. Specifically, we clustered with the neighbor-joining algorithm in the phylogeny inference package (PHYLIP; version 3.695). Neighbor-joining is a fast tree–generating method appropriate for datasets of the size considered here. By identifying sequences in this tree with well-curated annotations from the NCBI protein database, branches can be assigned to known VSD-containing protein families (). This tree is shown in . The seven largest branches corresponded to the first three VSDs in voltage-gated calcium and sodium channel pseudotetramers (Navs + Cavs I-III), the fourth VSD in voltage-gated calcium and sodium channel pseudotetramers (Navs + Cavs IV), eukaryotic voltage-gated potassium channels (Euk. Kvs), prokaryotic voltage-gated potassium channels (Prok. Kvs), hyperpolarization-activated cyclic nucleotide–gated channels (HCNs), and voltage-gated proton channels/voltage-sensitive enzymes (Hvs + VSEs). An additional set of VSD sequences did not cluster with VSD sequences with well-curated annotations, and we labeled them “unclassified.” A piechart of the VSD MSA illustrating these families can be found in . To quantify the diversity of sequences in the VSD MSA, we calculated the histogram of pairwise sequence identities for the aligned regions of all sequences (). This histogram shows an approximately unimodal distribution peaked around 20% sequence identity. Additionally, we calculated histograms of pairwise sequence identities for MSAs of sequences in each of the seven protein families described above (see Fig. S1). Several histograms showed polymodal distributions, suggestive of VSD subtypes within the identified protein families. With such a large quantity of homologous protein sequences in the VSD MSA, site-specific frequency distributions can be determined quantitatively. We can think of the difference between such a distribution and a reference as a measure of the “evolutionary pressure” exerted on that site. The Kullback–Leibler divergence, D(i), measures the information that differentiates an observed empirical distribution of amino acids, P, from a suitable reference distribution of amino acids, Q: Single-site amino acid frequency distributions, P, were calculated from the MSA with python scripts. Biases caused by phylogenetic relationships and incomplete sampling were addressed by reweighing sequences in MSA; aligned sequences with high sequence identity (>0.9) were weighted together as a single sequence as described by . By these criteria, the effective number of sequences in the MSA was calculated to be 3,821. Reference amino acid distributions, Q, were calculated for three topologically distinct regions of TM proteins: the inner membrane–water interface, the outer membrane–water interface, and the lipid-facing TM region. The Orientations of Proteins in Membrane (OPM) database consists of membrane protein structures with predicted outer and inner membrane interfaces. From the OPM, 533 polytopic α-helical TM protein structures were downloaded (). A python script was used to measure the amino acid frequencies within an 11-Å window of each predicted interface (representing the inner and outer membrane–water interfaces) and also between these windows (representing the TM region). These reference distributions and the assignment of topological region to individual positions in the NavAb VSD are available in Tables S1 and . To accurately reconstruct the network of evolutionarily important interactions from statistical correlations in the MSA, we infer the probabilistic model that makes the least possible number of assumptions about the underlying structure of the data. This criterion is satisfied by the model structure that maximizes the Shannon entropy, namely a Potts model. Additionally, determination of the interactions (the parameter-learning step) must rely on a computationally tractable algorithm. Several solutions to this second problem have been proposed recently (; ). These modeling approaches have been implemented in a family of algorithms known as DCA. DCA has demonstrated unprecedented success in dissecting the correlations present in MSAs into scored pairwise interactions. The set of all possible scored pairs generates an EC score matrix. This matrix appears highly similar to contact maps derived from the crystal structures of members of the protein family and contains sufficient information to allow ab initio structure prediction, identification of interfaces in homo-oligomers, and prediction of conformational changes (; ). Using a large MSA as input, DCA infers the parameters of a probabilistic model that reproduces single- and two-site marginals of empirical distribution of sequences contained in the MSA. The least constrained probabilistic model, known as a Potts model, has the following functional form: P(,…,) is the probability of observing the sequence {,…, }, where takes values in the alphabet containing the 20 amino acids plus the gap (“-”) symbol. Local fields, (), give rise to the propensity for an amino acid, , to be observed at sequence position , whereas the coupling constants, (, ), similarly encode the joint propensity of observing amino acids and at positions and , respectively. For a given pair, the set of coupling constants is arranged in a matrix; the Frobenius norm of this matrix is defined as the EC score for the pair ,. The full set of EC scores composes the EC score matrix. EC score matrices have been shown to be significantly correlated with contact maps. Although a rational interpretation of this correlation would posit that residues engaged in energetically favorable interactions are likely to coevolve, it has yet to be shown to which extent a statistical coupling can be identified with an energetic one. Thus, a strict correspondence between data from double mutant cycle analysis and DCA is not expected a priori. We obtained the EC scores of the VSD using Matlab scripts for pseudolikelihood maximization DCA described in . Molecular visualizations were created with visual molecular dynamics (). Statistical inference to determine direct couplings has been performed using a model structure (Potts model) that assumes a large set of parameters; even considering only 115 positions from the VSD sequences, there are ∼2.5 million parameters corresponding to pairwise coupling constants and local fields. The optimal values of these parameters are found, in general, by maximizing the likelihood of the data, here approximated by a computationally tractable proxy, the “pseudolikelihood” (). The quality of the results is therefore strongly dependent on the size of the dataset and on a correctly stratified sampling. To test whether our MSA of 6,652 unique VSD sequences satisfies these requirements, we calculated the EC score matrices for two disjoint subpopulations in our sample. Specifically, we extracted MSAs from the original dataset corresponding to two distinct phylogenetic lineages, the pseudotetrameric Nav/Cav family and the Kv family. These subpopulations were identified based on the partitioning provided by the hierarchical clustering described above, resulting into 3,391 and 1,832 sequences for Nav/Cav and Kv VSDs, respectively. EC score matrices for both of these families are presented in the supplemental material for comparison with the EC score matrix of the full dataset (, and ). The supplemental material contains the seed MSA used to infer the HMM (Table S1). Fig. S1 (A–G) shows pairwise sequence identity histograms for each of the seven protein families identified in our large MSA by hierarchical clustering. Fig. S1 H shows the dendrogram produced by hierarchical clustering, with assigned branches highlighted and labeled. Fig. S2 (A–D) shows the contact maps of four experimentally determined VSD structures and a structural superposition with the NavAb VSD. Fig. S3 (A and B) shows EC score matrices for partitioned MSAs containing either Nav and Cav or Kv VSD sequences. Tables S2 and S3 provide Shaker numbering for positions discussed in this work. Table S4 contains the membrane protein reference distribution. To help the readers map our results on each specific VSD, the MSA is available for download in the supplemental material, as well as a file with the complete set of parameters of the HMM used in this work The file format of the latter (.hmm) is compatible with the HMMER webserver () and can be used to generate an MSA using a different (possibly the most up to date) sequence database. The online supplemental material is available at . To generate the large MSA on which our subsequent analysis is based, we first train a profile HMM to recognize and align VSDs based on a small “seed” alignment (Table S1). This alignment contains representative VSDs from phylogenetically diverse protein families such as the voltage-gated potassium, sodium, and calcium channels, the voltage-gated proton channels, and the voltage-sensitive phosphatases (). Using structural alignment and TM helix topology predictions, we manually refine automatic MSAs for all of the representative sequences included and generate a confident seed alignment, available in the supplemental material and described more fully in Materials and methods. However, recently published homology models of the human voltage-gated proton channel (hHv1) suggest that alignment of the S4 helix among VSDs may be ambiguous (; ). Because hHv1 contains three gating charges and other VSDs contain four to six, alignments that only consider the matching of gating charges produce several different possible registers. Without structural alignments for all of the VSD representatives included in the seed MSA, we are forced to rely on score-based sequence alignment of S4. A statistically well-founded metric to judge the reliability of score-based alignments is the positional posterior probability (). The posterior probability for a certain position in a pairwise alignment is calculated by marginalizing a posterior distribution of scored possible alignments. Simply, the posterior probability that two particular positions in a pair of sequences should be aligned is a measure of how probable it is to see those particular residues aligned if the full distribution of possible alignments is considered. The profile of posterior probabilities for a given pairwise alignment therefore shows where the alignment is probable and where it is uncertain. This is exactly what we would like to see for alternate alignments of S4. In we show the optimal Needleman–Wunsch pairwise alignment of hHv1 to human Kv1.2, that is, the pairwise alignment with the best possible score. Incidentally, the register of this alignment is consistent with that of . Over the region of the gating charges, the posterior probability is high. Taking advantage of the probabilistic pair HMMs underlying the posterior probability calculation, alternate alignments were sampled with hHv1 in different registers. shows that in three alternate registers of S4, the posterior probability of the region covering the gating charges is decimated. This indicates that given the fundamental assumptions of score-based pairwise sequence alignment, the proper register of hHv1 S4 is not ambiguous. This method was used to check the reliability of the S4 register for other pairs of sequences in the seed MSA. With the technical issue of aligning S4 addressed, we train an HMM based on the seed MSA and collect a larger MSA of ∼6,600 VSD sequences as described in Materials and methods. We note that by using an HMM trained on a small but diverse seed MSA, we are able to automatically construct a much larger MSA than those reported previously for the VSD (; ; ). As observed in , the hierarchical clustering naturally breaks down the MSA into previously described protein families including voltage-gated calcium and sodium channels, eukaryotic and prokaryotic voltage-gated potassium channels, prokaryotic sodium channels, hyperpolarization activated cyclic nucleotide–gated channels, voltage-gated proton channels, and voltage-sensitive enzymes. Interestingly, the first three VSDs of the pseudotetrameric voltage-gated sodium and calcium channels clustered together, and the fourth clustered separately. This may be expected from the findings of , in which the functionally distinct kinetics of the first three “fast” domains of Nav1.4 compared with the fourth “slow” domain were shown to arise from specific sequence differences in their respective VSDs. shows the distribution of pairwise sequence identity, where we observe a single large peak around 20% identity. If the dataset consisted of large clusters of mostly identical sequences, we would expect to see additional large peaks at high identity values. Our MSA therefore conforms to our requirements for extensive phylogenetic coverage and sufficient intra-family variability. The large VSD MSA is represented as a sequence logo in (), with positions numbered according to the voltage-gated sodium channel (NavAb) sequence (). The height of the columns in suggests that conserved residues on the VSD occur with periodicity. This trend was noticed in an early study done with a smaller MSA of VSD sequences (). To quantify this observation, we calculate the Kullback–Leibler divergence (D) for each position in the MSA, as described in Materials and methods. In general, D values quantify the additional amount of information (in nats) differentiating the empirical distribution of amino acids at a certain position from a suitable reference distribution (). For instance, residues located in the TM region are expected to be mostly hydrophobic; a TM position with a distribution significantly enriched in hydrophilic residues arguably suggests evolutionary pressure and results in a large value of D. Therefore, we use D as an empirical site-specific measure of evolutionary pressure. In , we map high D residues above a threshold (1.1 nats) onto the NavAb VSD and notice that the internal core of the bundle has the highest D. The outer surface of the VSD has much lower D, consistent with the idea that these positions make nonspecific interactions with membrane lipid tails and are thus subject to less strict evolutionary constraints (; ). Interestingly, the top of S3 (termed “S3b,” consisting of positions 86–91) does not exhibit this periodicity (), suggesting a lack of evolutionary pressure on this region. shows the relative variation of D in the VSD. Unsurprisingly, the S4 gating charges (, positions 99, 102, 105, and 108) and the S1–S3 countercharges (, positions 32, 59, and 80) exhibit high D with respect to the rest of the VSD. This is consistent with their conspicuous placement as charged particles in the TM region and with their conserved mechanistic importance among VSDs (; ; ). Extensive mutagenesis studies have identified an S1 isoleucine (, position 22) and an S2 phenylalanine (, position 33) as the primary constituents of a conserved hydrophobic region present in VSDs (; , ). Mutation of residues in this region modulates both the kinetics and voltage sensitivity of S4 translation. Both of these positions have high D (). Interestingly, voltage-gated proton channels are not significantly represented in the dataset; the position identified as the selectivity filter in hHv1 (, position 25 and hHv1 D112) shows high D, suggesting a functional significance for this position, even in VSDs that do not conduct protons. Given that D measures evolutionary pressure and that identified functional residues exhibit high D, it may be valuable to investigate other high D positions throughout the VSD. To the best of our knowledge, residues at the lower membrane interface (, positions 11, 14, 63, 71, 74, 76, and 77) have not been experimentally characterized. From we see that these positions take polar, aromatic, and positively charged amino acids. These amino acids are commonly observed at the membrane interface, specifically at the inner leaflet, as they are involved in anchoring the protein to membrane headgroups (; ). However, in VSDs their relative abundance cannot be explained only by their localization. Indeed, our formulation of D explicitly accounts for the propensity for observing these residues at the lower membrane interface. That the presence of these amino acids has been maintained throughout evolution is suggestive of distinct, conserved functional roles. We performed DCA on the VSD and ranked pairs according to their EC scores (see Materials and methods for the definition of EC scores). Simply, the EC score for a pair of residues is a heuristic measure of the total coupling as inferred from the probabilistic modeling. It is important to note that the distribution of EC scores does not show an obvious threshold of significance. shows a histogram of EC scores for non-neighbor pairs (more than five residues apart in sequence). The distribution appears to be approximately Gaussian with a fat right tail. Because we have no prior expectation about the distribution of EC scores, we set an arbitrary threshold of 0.10 (approximately corresponding to a standard deviation from the mean), as depicted by the yellow region in . only considers those couplings with EC scores >0.10. Considering only positions separated by at least five residues in the primary structure, we find that 20 of the top 23 pairs are in direct physical contact in the crystal structure of the activated NavAb VSD (). This gives a true-positive rate of 87% contact prediction for this first small subset of pairs. This is consistent with reported true-positive rates for contact prediction with pseudolikelihood maximization DCA (). These 20 contacting pairs define the interfaces between S1–S2 and S2–S3. Curiously, only two pairs in the top 23 involve S4 coupled to the S1–S3 bundle, and these pairs are not in physical contact in the NavAb VSD structure (, positions 25–96 and 49–96). For comparison with , contact maps of four other experimentally determined VSD structures are available in , accompanied by structural superpositions with the NavAb VSD. To further investigate the spatial distribution of pairs with large EC scores on the VSD structure, we compare the EC score matrix to the contact map of the NavAb VSD (). The trend is similar to that observed in the top pairs: the contacting interfaces of S1–S2 and S2–S3 are well-defined in , whereas the contacting interfaces of S1–S4 and S3–S4 observed in the contact map are completely absent in the EC score matrix (shown schematically in ). EC score matrices calculated from partitioned MSAs of only Cav/Nav or Kv VSDs show similar features (Fig. S3, A and B). In the next section we consider how design principles of VSD function may have constrained evolution to generate these observations. xref #text

The classic experiments of Hille defined selectivity profiles for voltage-gated sodium (Nav) and potassium channels (, , ), and later for nAChR channels at the neuromuscular junction (). These determinations were based on relative permeability values calculated from shifts in reversal potential determined under voltage clamp. The approach offered a well-defined protocol for comparison of the ease with which different ions entered a particular channel. Such measurements, systematically performed for alkali cations, and a series of variously sized organic cations roughly defined the limits for the size, and to some degree the cross-sectional shape, of the narrowest part of each channel, henceforth known as the “selectivity filter.” A similar analysis was performed for voltage-gated Ca channels from skeletal muscle (). Later, single-channel recording allowed direct comparisons of unitary conductance, measured in the presence of different permeating ions, which provides an alternate indication of channel selectivity (; ). Although, depending on the molecular details of conduction, relative permeabilities and conductances obtained from these two approaches may differ, the analysis of reversal potential shifts resulting from external cation replacement endures as one convenient and widely used approach to measure ion channel selectivity. Studies on Nav channels from frog node of Ranvier suggested a selectivity filter of asymmetric cross section, measuring a minimum of ∼3 × 5 Å (). Subsequently, the selectivity filter was later identified as the highly conserved signature motif, DEKA (see rSkM1 pore sequence fragments in ), to which one amino acid residue was contributed by each of the homologous repeat domains of the Nav channel α subunit (). In contrast, Hille’s analysis for delayed rectifier K channels suggested a symmetric selectivity filter of ∼3 Å in diameter (). The first crystal structure determined for a K channel revealed a size closely matching this estimate, and showed that the lumen of the selectivity filter was lined by backbone carbonyls () of residues from the K channel signature sequence (). In contrast, functional data for Nav and voltage-gated calcium (Cav) channels suggested a filter lumen lined by amino acid side chains providing a net negative charge (; ; ). Functional characterization by the Clapham laboratory of a homotetrameric, prokaryotic Nav channel, NaChBac (), prompted hopes for an Nav channel crystal structure. After 10 years, these hopes were realized by Catterall and collaborators (, ), using the NaChBac relative NavAb, closely followed by and . Studies on these prokaryotic channels shattered the maxim that an asymmetric selectivity motif was required to achieve selectivity for Na over K and common divalent metal ions (; ). In the prokaryotic Nav channels, sequence alignment and structure are consistent with a selectivity ring made up of one glutamate residue from each of the four monomers, reminiscent of the EEEE selectivity filter of four-domain eukaryotic Cav channels (). In eukaryotic Nav channels, there is a conserved outer ring of acidic residues, three to four positions C terminal from the DEKA motif (), but this outer ring appears to influence conductance and proton block, rather than determine selectivity (). Although follow-up studies of NaChBac showed that it could easily be converted into a Ca-selective channel by the addition of one or more additional acidic residues into the outer vestibule-lining P loop (), the detailed mechanism by which the NaChBac family of prokaryotic channels achieves Na selectivity with an EEEE selectivity filter has yet to be clarified. Similar to voltage-gated potassium channels, the NavAb channel is a homotetramer. Each monomer is composed of six transmembrane helices, S1–S6. The first four helices, S1–S4, form the voltage-sensing domain, whereas helices S5 and S6, plus their connecting loop, form the pore domain, through which ions permeate. Notably, the presumed selectivity filter of the NavAb channel is wider than those of K channels. The resolution of the crystal structures renders unambiguous positioning of permeant ions difficult. Nevertheless, up to three binding sites were proposed based on a combination of functional and structural studies (). T175 of each monomer forms the inner (cytoplasmic) site, S, whereas main-chain oxygens of L176 aided by the water molecules comprise the central site, S, and the acidic side chain of E177 ligates permeant ion in the outermost, high field–strength site, S (). The question of how this filter gives rise to selective permeation, including the functional role of the varying polarity among its different binding sites, has yet to be resolved. The general physical principles underlying selectivity in prokaryotic and eukaryotic Nav channels may be similar, but the molecular details show some striking differences. A broad structural similarity is seen in the asymmetry along the pore axis. Both prokaryotic and eukaryotic Nav channels possess a selective periplasmic, or extracellular, section of the pore, with an outer vestibule that quickly grades into the narrow selectivity filter, followed by an expansive aqueous inner cavity, roughly aligned with the hydrophobic center of the surrounding bilayer, and then, a nonselective cytoplasmic mouth. The points of difference are tantalizing and include, in prokaryotes, the more negative net charge of the EEEE selectivity ring and the radial quasi symmetry, which contrasts with the strongly asymmetric DEKA ring of eukaryotic Nav channels (). Among the eukaryotic members of the P-loop ion channel molecular family, Cav and Kv channels show functional characteristics that are most easily explained if each channel normally accommodates two to four conducting ions. These properties include tight binding of the preferred ion species (; ), strong coupling of unidirectional ion fluxes through the channel (; ), and a nonlinear dependence of conductance on the mole fraction for mixed-ion solutions (). In contrast, eukaryotic Nav channels show a saturating dependence of single-channel conductance on sodium concentration with half-maximal conductance in the millimolar range (; ; ), very weak flux coupling (), and other features consistent with only a single ion being bound to the channel most of the time (; ). Perhaps the most obvious connection is that, functionally, bacterial Nav channels tend to be leakier to calcium ions than their eukaryotic counterparts, and several point mutations allow them to preferentially conduct calcium (). A key question that we address here is how bacterial Nav channels can show selectivity and conductance closely approximating those of their eukaryotic counterparts, whereas their multi-ion occupancy, suggested by available structures, is more akin to that of eukaryotic Cav and Kv channels. We experimentally evaluate NaChBac’s alkali cation selectivity under a variety of conditions and explore possible mechanistic and molecular underpinnings of that selectivity using computer simulations based on the related NavAb crystal structure. Although these are two different channels, they display a considerable structural similarity; for example, both channels are homotetramers with high sequence identity, especially in the pore domain. Judging by multiple studies of 2TM and 6TM potassium channels, a wide variety of channels with apparent difference in sequence produces similar structures of the pore domain (within 1–2-Å resolution for backbone atoms). We identify possible determinants of conduction and alkali cation selectivity of bacterial sodium channels by analysis of molecular dynamics (MD) simulations and calculations of 1- and 2-D potentials of mean force (PMFs) for ions moving within the conducting pore. We find that the functions of selectivity and conduction resemble those of eukaryotic Navs, even though patterns of ion interaction and occupancy are more like those of Cav and, to a lesser extent, Kv channels. The original NaChBac channel construct in pTracer-CMV2 (Invitrogen) was provided by D. Clapham (Howard Hughes Medical Institute, Children’s Hospital, and Harvard University, Boston, MA). Single amino acid mutants were generated as described previously (), using overlapping PCR amplification with oligonucleotide containing the sequence for the desired amino acid substitutions, followed by subcloning into pTracer (Invitrogen). All clones were completely sequenced. Mammalian tSA 201 cells () were transfected with DNA encoding WT or mutant NaChBac, plus GFP using Polyfect (QIAGEN). Significant channel expression occurred within 24 h, after which whole-cell patch-clamp recordings were made at room temperature with an amplifier (Axopatch 200B; Molecular Devices). Patch pipettes were pulled from glass (8161; Corning; Potash-Rubium-Lead; softening temperature, 600°C; dielectric constant, 8.3) to a resistance of 1.5–2.5 MΩ. Recordings were made 24 h after transfection in control external solution that contained (mM): 142.5 NaCl, 2 CaCl, 2 MgCl, 10 glucose, and 10 HEPES, pH 7.4. External ion replacement solutions were made by substituting 142.5 mM Na by Li, K, Rb, or Cs. Control intracellular (pipette) solution (“Na”) contained (mM): 105 CsF, 35 NaCl, 10 EGTA, and 10 HEPES, pH 7.2. “K” was made by substituting 35 mM NaCl by 35 mM KCl. For bi-ionic experimental conditions, the intracellular solution contained either 140 mM Na, K, or Cs. External solution changes were achieved by local superfusion of the replacement solution over the cell, with appropriate corrections for changes in junction potential (see Data analysis below). Data were analyzed using Clampfit (Molecular Devices) and Igor (WaveMetrics) software. Peak I-V curves were fitted using I(V) = (V − V) * G/(1 + exp((V − V)/V)), where I is the macroscopic current, V is the command potential, V is the reversal potential, G is the maximal conductance, V is the half-activation potential, and V is the slope factor (mV/e-fold). For weakly permeant external ions, the reversal potential is expected to occur in a negative voltage range, at which very little conductance is activated (e.g., , left, internal sodium, external potassium). Thus, in the presence of weakly permeant ions, a prepulse to −10 mV was applied to induce maximal activation, and V was determined from the instantaneous I-V relation, measured from the initial point in the tail current decay after steps to a series of voltages encompassing V. For weakly permeant ions, this protocol provided a more precise and reproducible estimate of V. Relative permeabilities were calculated according to: P/P = exp((E − E)/(RT/F)), where P is permeability to ion X, P is Na permeability, and RT/F is 25.4 mV. Net junction potentials were balanced to reduce the pipette current to zero before seal formation. In experiments where solution replacement happened after the seal was established, the theoretical junction potential for each solution pair was determined using JPCalc from Clampfit, which was also used for V correction before the calculation of the permeability ratio. Series resistance compensation was applied conservatively to favor voltage-clamp stability; we note that the series resistance correction approaches zero near the reversal potential, the most critical measurement in this study. Net junction potential corrections applied to estimates of V fell in the range of 1.8–9.9 mV. All summary data are presented as mean ± SEM (), where is the number of determinations. Statistical significance was evaluated using the unpaired Student’s test; the criterion for a significant difference was taken to be P < 0.05, unless otherwise stated. The initial structure of NavAb was taken from the high resolution x-ray crystallographic structure with the Protein Data Bank accession number (). The structure corresponds to a closed-pore conformation. The molecular simulations reported in this study were focused on the WT NavAb () and the E177D mutant (). The tetrameric channel was embedded into a pre-equilibrated DMPC lipid bilayer and solvated in a 104 × 104 × 80–Å box filled with TIP3P water using CHARMM-GUI membrane builder protocol (). The whole assembly was bathed with 150 mM NaCl, ensuring electroneutrality. Although Na ions were not directly observed in the crystallographic structure, three Na ions were positioned in the filter as the starting conformation (as shown in ), based on continuum electrostatic computations. All MD simulations were performed by the program CHARMM (). The CHARMM-27 force field was used for protein and lipids (; ), and Na and K ions as reported previously (; ). The NpT ensemble was used for all simulations, with pressure set to 1 atm and temperature set to 315 K. Long-range electrostatic interactions were treated by the particle mesh Ewald algorithm (). Nonbonded interactions were switched off at 10–12 Å. The systems were simulated with periodic orthorhombic boundary conditions applied in all directions with the time step of 1 fs. After a staged equilibration with a gradual decrease in harmonic constraints that act on heavy protein atoms only, further nonconstrained equilibration was run for 23 ns. To unravel energetics of ion permeation, we used multidimensional umbrella sampling methods, a powerful computational technique used with considerable success in studies of K channels. Umbrella sampling simulations were performed with harmonic biasing potentials with a force constant of 10 kcal/(mol · Å) along the z axis. The zero position along the z axis is the center of mass of the selectivity filter backbone atoms of residues T175, L176, and E177. 1- and 2-D profiles for ions were computed. The lateral displacement of ion(s) was restrained to be within a cylinder with a radius of 10 Å and the central axis along the z axis. The reaction coordinate for each ion was the distance along the z axis between the ion and the center of mass of the selectivity filter of the protein, as just defined. The final snapshots of the conventional MD were used as the starting conformations for the umbrella sampling. The sampling windows were spaced every 0.5 Å from +14.5 to −10 Å, resulting in 50 windows for 1-D PMF computations and 1,045 windows for 2-D PMFs. The simulation time per window was set to 3 ns for 1-D PMF computations and 0.5 ns for 2-D PMF computations, respectively. 1-D PMFs (e.g., and ) were obtained by integration along a pathway for a single ion across a 2-D map (e.g., ). Such 1-D PMFs, in general, would be expected to trace out a likely low energy pathway of one ion across the 2-D PMF surface, as illustrated in , and would reflect the ion’s interactions with its environment, including other ions, protein, and solvent. Convergence of computed 1-D PMFs obtained from a 2-D PMF map obtained after simulation times of 100–500 ps is illustrated in . The energy surfaces were rebuilt with the weighted histogram analysis method (WHAM) (; ), and the tolerance for WHAM was set to 0.001. In some cases, 2-D PMFs were further analyzed by extracting 1-D “cross sections,” for which one ion is held constant, whereas the other is moved on a single reaction coordinate across the profile (e.g., , and see Results). An indication of the significance of differences between PMFs or profiles of other parameters is given by SEMs estimated from standard block-average calculations () using progressively increasing block sizes (see , and and associated text). The single equilibrium dissociation constant K(single) from 1-D PMF in the presence of a cylindrical constraint can be expressed as follows (; ): where R is the radius of the cylindrical restraint oriented normal to the z axis with = −10 Å and = 14.5 Å. The () was offset to zero for an ion in the bulk phase. The equilibrium dissociation constant for the double occupancy state of the channel, K(double), can be expressed in accordance with : with the following integration limits: −7 Å ≤ z ≤ 20 Å and −10 Å ≤ z ≤ 20 Å. The supplemental figures provide additional experimental data to complement that presented in the main text, as follows. Fig. S1 illustrates the time courses of wash-in and washout, as K replaces Na in the external solution. Combined with the results in the main text, , the data make two points: washout of K, and its replacement by Na, is significantly slower than the reverse, suggesting stronger binding of K under these conditions; and washout of K, in particular, depends significantly on the complement of intracellular ions, with the slowest being observed with intracellular solutions containing Cs or Na as the sole internal alkali cation species. Fig. S1 shows that no anomalous mole-fraction dependence is seen when external Na is replaced by K, with intracellular media containing 105 mM Cs with 35 mM K or Na. This contrasts sharply with the observations for intracellular solutions containing either only K or only Na, which reveal a dramatic anomalous mole-fraction dependence (). Fig. S2 provides voltage-clamp records from NaChBac E191D taken at extracellular pH values of 7.4 and 5.8. Even though the E191D mutant is much less selective for Na over K than is the WT channel, an additional significant loss of selectivity (i.e., an increase in P/P) was observed with a decrease of extracellular pH (). Fig. S3 illustrates convergence of a 1-D PMF computed from a 2-D map. The online supplemental material is available at . In this Results section, we stress the overriding goal of our study: to systematically integrate experimental and computational results that suggest the mechanisms underlying selectivity among alkali cations by bacterial Nav channels, exemplified by NaChBac and NavAb. To this end, we interleave results from experiments and simulations throughout. We begin with the classic approach of measuring relative permeabilities, based on reversal potential shifts observed when the external cation species is changed, and calculations of the PMF as a single ion is moved through the channel. Intuitively, one might expect that reversal potential shifts generated by external ion substitution are expected to be primarily influenced by the energetics of ion entry into the pore (; ), and hence by the outermost parts of the pore structure. It is this part of the pore structure that is likely to differ least between open and closed conformations in channels for which control of channel opening is presumed to lie at the S6 bundle crossing on the cytoplasmic side of the selectivity filter. Thus, we believe that the putative preopen structure of is an appropriate basis for a consideration of selectivity based on relative permeability measurements. The qualitative selectivity fingerprint (), defined by the sequence of relative permeability values, P, where P is defined as 1, matches both the sequences determined from similar experiments for eukaryotic Nav channels (e.g., ) and the sequence for equilibrium binding to high field–strength anionic sites (). Our experiments used 105 mM Cs as the major internal cation with 35 mM of internal Na to allow easy measurement of either positive or negative shifts in Erev, with a quasi-physiological inward electrochemical gradient for Na over most of the working experimental voltage range. The resulting selectivity sequence is P ∼ P > P ≈ P ≈ P. As a first step toward defining the energetics of ion permeation, we evaluated 1-D PMFs as single ions were moved through the NavAb model pore (). In the absence of other ions, a solitary Na would bind much too tightly to permit physiological rates of throughput (single-channel conductance, ∼12 pS or ∼10 ions/s, with a 100-mV driving force; ). From the one-ion PMF, dissociation rates of 10 ∼ 10 ions/s are predicted from a single-barrier calculation based on Eyring rate theory. This is reminiscent of tight binding of Na and Ca to binding sites in model channels (; ). To reconcile large ionic fluxes, argued for the existence of multiple binding sites and the importance of the local crowding effects. More realistic permeation pathways appear on the energy landscape when two monovalent cations enter into the pore (), with energy minima for four doubly occupied configurations in the range of −2 to −4 kcal/mole (). Then, low energy permeation pathways (blue tones) become clearly apparent; the potential profiles traversed by a single ion were obtained by integrating along the reaction coordinates ( and see Materials and methods, Computational strategies). Similar findings have been reported recently from computational studies by and by . They also emphasize the need for a second bound ion to achieve physiological conductance levels. The relatively deep energy well for Na–Na would contribute to higher stability of Na–Na-occupied filter, with the essentially flat entrance barrier for Na–Na also playing a significant role (Z = 10 ∼ 12 Å) in discrimination among different ions. It is important to stress that uncertainties in 2-D PMF computations are rather high, but the resolution in the computed maps is still sufficient to clearly discriminate between Na–Na and K–K occupancy in the filter. Interestingly, in the case of K–K binding to the filter, there is a small barrier separating two ion-binding energy minima. Extended analysis of the 2-D PMFs revealed multiple binding sites separated by small barriers (indicated in ). The integrated single-ion PMFs (e.g., ) and 1-D “cross-section” profiles (e.g., ), derived from the 2-D PMF surfaces, help us to understand the significance of this apparent complexity in the filter. We estimated the corresponding equilibrium dissociation constants in the presence of constraints for the singly and doubly occupied channel using the following theoretical developments of and . In general, the channel binds Na in preference to K, but when two ions are bound, the magnitude of Na/K selectivity for binding at the first site depends on the identity of the second ion. With symmetric occupancy of the filter, binding of an Na–Na pair is favored over a K–K pair (). Up to three well-defined potential binding sites can be distinguished, further supporting the idea that multiple ions can occupy the pore. The free energy cost to bring in a third ion would render this occupancy unstable, thus leading to a permeation event. The effective free energy cost caused by ion–ion repulsion for bringing a second ion as determined from the single- and double-occupancy dissociation constants, K and K, is ∼13–16 kcal/mol depending on the cation species. The mixed occupancies, Na–K and K–Na, yield intermediate depths, reflecting an effect on the stability of the binding around the middle of the selectivity filter. However, the PMF calculations are accurate only within ∼1 kcal/mol according to our block-average analysis placing restraints on the interpretation of obtained PMFs. Nevertheless, in the limits of method resolution, the barriers at the extracellular side of the channel display dependence on the combination of ions present, although the differences are subtle. In the case of the barriers, the mixed ion combinations provide the extremes, whereas double occupancy, by a single-ion species, Na–Na and K–K, yields the extremes. The number and the conformation of binding sites definitely differ considerably depending on the identity of the second ion. The free energy profile for Na, double occupancy (the solid black line in ), has a local minimum of around Z = 3 Å, which is not present in the case of K (the solid red line in ). Also, binding sites around Z = 4.5 Å for Na shifted to the right by ∼0.5 Å with respect to those for K. reported a striking asymmetry of reversal potentials for skeletal muscle Nav channels in planar lipid bilayers, measured when a bi-ionic Na/K gradient was reversed. Similar behavior has also been reported recently by a mutant version of nonselective NaK channel by . We observed an analogous result for NaChBac whole-cell currents, when K was exchanged for Na in the presence of either internal Na or K (). The magnitude of the reversal potential shift, after external ion replacement, was ∼100 mV with internal Na but only ∼40 mV with internal K, yielding a 10-fold difference in P/P calculated with oppositely directed gradients (P/P = 0.2 with internal K, and 0.02 with internal Na). In bi-ionic conditions, with isotonic Na and K solutions on opposite sides of the membrane, the reversal potentials were approximately +45 mV (Na inside) and −100 mV (Na out; instantaneous I-V relations; data points plotted as open triangles; see ). This functional asymmetry may reflect a common property among the various members of the P-loop family of channels, which possess a nonselective entrance to the intracellular cavity, and a selectivity filter located nearer to the extracellular mouth. In doing these experiments, we also noted a kinetic asymmetry. The approach to a new steady-state level of peak currents elicited by a succession of identical voltage-clamp pulses took significantly longer for washout of external K by Na than vice versa (), suggesting stronger binding of K within the pore. This difference was most obvious with Na as the only internal cation species, but it was also significant with K as the only intracellular cation. Slow washout of potassium was also seen with cesium as the only internal cation, but it was not detectable using mixtures of cesium or potassium with sodium (). Progressive substitution of external Na by K reveals a nonmonotonic (anomalous) dependence of conductance on mole fraction of sodium, (). Such behavior has frequently been associated with multi-ion occupancy of channels and repulsive interactions among ions in the channel, although alternate explanations have been proposed. The following features are of interest in our data. With a single monovalent cation species inside—either sodium or potassium—there is a general tendency for conductance to increase in a supra-linear fashion up to = 0.6. With potassium as the sole internal cation species, a conductance maximum occurs at ≈ 0.8, and the conductance then declines monotonically to the value characteristic of a full sodium external solution. Notably, the appearance of this maximum appears specific to interactions between Na and K, as we saw no maximum with internal solutions containing 105 mM Cs, plus 35 mM of either Na or K (see Fig. S1, E and F). If the sole internal cation species is sodium, a shallow local minimum also appears in the conductance at ≈ 0.95 (). This result provides further evidence that the last K ion to exit binds tightly enough to make it relatively difficult to completely wash potassium ions out of the channel (see also ), consistent with the tight binding predicted by 1-D PMFs. Two features of these data deserve attention. First, the maxima we see for intermediate mole fractions ( = 0.6–0.8) are unusual, and to our knowledge have not been reported previously for other systems. Thus, the conductance is higher for external mixtures of Na and K than with either Na or K as the sole external species. This would be consistent with an evolutionary optimization for conduction in physiological solutions containing mixtures of sodium and potassium. In other reported cases of an anomalous mole-fraction effect involving K or Cav channels, a minimum in mole-fraction dependence has been seen, suggesting that one ion species binds strongly enough to impede permeation by another (; ; ). A similar mechanism may underlie the local minimum that we show in , at ≈ 0.95, with Na as the internal monovalent cation species. To explore the role of a particular bound ion on the entrance barriers for the second ion, we used a cross-section methodology from applied to a bi-ionic solution, where a 1-D PMF (“cross section”) for a permeating ion, in the presence of a fixed copermeant ion, was obtained from a 2-D map (). The bound ion was fixed at Z = 1.0 Å. As expected, the barriers presented near the entrance (Z > 12 Å) display relatively little dependence on a particular combination of cations. Within the selectivity filter (Z = 4–8 Å), the situation is drastically different among the Na–Na, Na–K K–Na, and K–K combinations. The interaction between cations in the filter can significantly affect the relative height of the barrier and the depth of the well experienced by a second cation. The ion bound to the lowest affinity site (Z = 1.0 Å) also affects the location of the binding site for the entering cation. Notably, K–Na mixtures tend to display flatter energy profiles, suggesting that the conductance for a mixture of cations could be as high as for pure Na solution based on reduction of the entrance barrier for the second cation. Discrimination among alkali cations by NaChBac is qualitatively similar to that shown by eukaryotic Nav channels (). For NaChBac, under the conditions of our experiments (single monovalent cation species, X, in the extracellular solution, with an intracellular solution containing 35 mM Na and 105 mM Cs), the relative permeabilities, P/P, fall into the following sequence: Li ≈ Na > K ≈ Rb ≈ Cs. Given the difference in net charge on the selectivity rings between bacterial (EEEE) and eukaryotic sodium channels (DEKA), an obvious question arises: how do changes in the net charge on the selectivity filter affect selectivity? Replacement of the selectivity filter glutamate with aspartate (E191D in NaChBac and E177D in NavAb) would yield a DDDD selectivity ring. With minimal perturbation in the backbone structure, this substitution might have been expected to lead to a small increase in diameter, with a consequent reduction in negative charge density at the selectivity filter, and because of the shorter aspartate side chains. Actually, the energy-minimized structure for E177D shows a slightly reduced diameter because of the more direct projection of the side chain toward the pore axis than for the longer, more flexible glutamate ( and ). The resultant decrease in radius is associated with the removal of the multiple energy minima seen in 2-D PMFs, and it yields almost indistinguishable single-ion PMFs for Na and K (). Experimentally, this mutant showed a decrease in selectivity for Na, with significant increases in relative permeability to both potassium and rubidium compared with the WT channel (). In a complementary set of experiments, lowering the extracellular pH from 7.4 to 5.8, which is expected to titrate some of the charge on the EEEE selectivity ring, reduced the discrimination among alkali cations ( and S3). For both WT NaChBac and the mutant E191D, the lowered external pH yielded an increased relative permeability to potassium. The permeability increase was slightly larger for the E191D mutant, such that this construct did not distinguish between extracellular Na and K at pH 5.8. Collectively, the experiments in this section suggest that charge density within the selectivity ring is an important contributor to overall selectivity among the alkali cations. To test whether the pore cross section is affected by E191D replacement, we ran a complementary set of all-atom MD simulations on a mutated pore. The calculated cross-sectional profiles for NavAb E177D are shown in . The pore radius at the constriction is still considerably wider than that for K channels. This finding is in keeping with previous data on the ion hydration in the pore (; ), which suggest that the permeant cation (either Na or K) is able to maintain partial hydration and could possibly be accommodated by either WT or E/D mutant. The pore radius decreased slightly with the E177D mutation despite the shorter side chain introduced. To further investigate the mechanisms by which the E/D mutation might reduce selectivity, we again used a 2-D PMF-based strategy similar to that in the WT sections. The resulting 2-D PMFs for Na/Na and K/K are shown in , and the corresponding single-ion PMFs are shown in . The PMF calculations show considerable changes in the energy surface, which controls ion transport across NavAb E177D. The computations suggest that the degree of ion stabilization in two of the three proposed binding sites present in the WT filter () is significantly affected by the mutation. The first binding site (S) is destabilized for both Na and K (), whereas the second binding site (S) remains virtually unaffected. It is tempting to think that E/D mutation not only affects the cross section of the pore but also ion coordination environment and number of sites available to an ion along the permeation pathway. The significant destabilization of one of the ion-binding sites caused by mutation would reduce the likelihood of two ions binding simultaneously to the filter. The computed PMFs suggest that the differences in Na/K profiles, present in WT, are largely gone when this mutation is present, and it is also evident that there are several significant rearrangements in the ion coordination in the filter region. The entrance binding site, , presented in the WT () is eliminated in the mutant for both Na and K binding to E/D–NavAb, whereas central binding sites and ) display similar stability for Na and K. This finding is in agreement with our experimental data on ion selectivity (). Thus, the reduced selectivity of the E-D mutant may be primarily attributed to the removal of one of the binding sites from the selectivity filter coupled with a nondiscriminatory binding of Na and K to the remaining site (). These results highlight an importance of the multiple binding sites present in the selectivity filter, which appears to modulate selective conductance. A definition of the “selectivity” of an ion channel could be based on measurements of ion-binding affinity, conductance, or reversal potential for currents measured under particular conditions. Under fairly general conditions (), “relative permeabilities” thus calculated (see Materials and methods) directly reflect the rate of unidirectional flux across the membrane (; ), or the rate of entry into the channel (). Although this simple physical interpretation likely breaks down with complex barrier profiles or ion–ion interactions (), relative permeability provides a widely used method to establish selectivity sequences among the various ions that can pass through a channel. Homotetrameric NaChBac, despite its more symmetric structure, qualitatively resembles eukaryotic Nav channels in that it selectively conducts Na in preference to most other monovalent ions and common physiological divalents. The selectivity “fingerprint,” defined as the order of relative permeabilities, P/P, for the alkali cations is indistinguishable from that seen for eukaryotic channels: Li ≈ Na > K ≈ Rb ≈ Cs (). As suggested by earlier work (), the critical physiological selection for Na over K appears to be somewhat more stringent for NaChBac than for eukaryotic channels (NaChBac, P/P = 0.03 ± 0.007 compared with 0.048–0.086 for four eukaryotic Nav channel variants; see Table 14.2 in ). A functional asymmetry, common to NaChBac and eukaryotic channels, is seen in the inequality of magnitude of reversal potentials observed when a bi-ionic Na/K gradient is reversed. Reversal potentials of approximately −100 mV (Na/K) and approximately +40 mV (K/Na) indicate a 10-fold stronger discrimination between K and Na when the internal cation species is Na than for the reverse case (). Alkaloid-modified rat muscle Nav channels (bound by a large alkaloid, batrachotoxin or veratridine, in the cytoplasmic cavity) show many of the qualitative properties of unmodified channels, including an asymmetry of bi-ionic reversal potentials (). This suggests that important determinants of selectivity reside near the outer end of the pore. The results also indicate the usefulness of a “pre-open” structure as a relevant basis for simulations of selectivity, particularly discrimination among extracellular ions, even though the intracellular bundle-crossing gate may be closed or modified. The dependence on external mole fraction in bi-ionic Na–K mixtures ( = [Na]/([Na] + [K])) of the normalized maximal conductance is highly nonmonotonic (). A clear maximum appears at intermediate values of (0.6–0.8), suggesting that mixed occupancy by Na and K, as would be expected under physiological conditions, enhances conductance. In addition, a less conspicuous but significant local minimum, which occurs at low levels of K ( ≈ 0.95–0.97), hints that a single potassium ion binding at the outer end of the pore may inhibit sodium permeation. Anomalous (nonlinear and nonmonotonic) mole-fraction dependence has generally been associated with multi-occupancy with ion–ion interactions in the pore. This is consistent with our own calculations () and those of others (), but alternate explanations have been proposed (; ). NaChBac’s strong anomalous mole-fraction dependence, with its distinctive enhancement of conductance at intermediate mole fractions, was observed only with intracellular solutions containing solely Na, or solely K, as cations. Two maneuvers, which might be expected to change the charge density on the selectivity filter (E191D mutation and lowering of pH), each reduce NaChBac’s ability to discriminate between Na and K (), as predicted by . Accordingly, our data show that lowering pH from 7.4 to 5.8 increases P/P significantly from 0.03 ± 0.006 to 0.15 ± 0.03 (P = 0.018; see ). The E191D substitution also reduces selectivity, with P/P = 0.57 ± 0.02 at pH 7.4 and a further increase to 0.71 ± 0.01 at pH 5.8. It is not intuitively obvious why the decrease in pore diameter () should be associated with a decrease in selectivity, but at least it does suggest that energetic factors beyond simple steric filtering, or simple binding interactions dependent on the field strength near a coordinating site, must be considered to account for this change, together with the other complexities of the dataset. Insight from atomistic simulations is essential to study the mechanism of ion permeation through NavAb channels. Without such a quantitative approach, coordinated with experimental studies, the binding sites proposed based on the crystal structure would remain speculative. The deep energy wells shown in 1-D PMF profiles (e.g., ) indicate that single Na or K ions are extremely unlikely to permeate though the channel. Calculations based on 2-D PMFs provide more realistic permeation pathways, with energy minima in the range of −2 to −4 kcal/mol (). The blue tones in the energy landscapes show many stable conformations and multiple favorable ion-binding sites. All three binding sites, which were previously hypothesized based on the structure (), are consistent with our simulations. In the 2-D PMF maps, S appears in the stable conformations but not necessarily S. The outer binding site S is not stable because of the flexibility of the Glu177 side chains, but Glu 177 can form a binding site with the help of the side chain oxygen atom of Ser178. It can also contribute to the coordination shell of a Na ion, together with the Leu176 side chain of the neighboring monomer in S. Finally, it can even form a binding site in partnership with Leu176 of the same chain. The computational results also indicate that the ion does not move along the central pore axis () because of the relatively large radius of the channel (). Instead, an ion is coordinated by the main chains and charged side chains from one or two of the monomers, and by water molecules (, and ). In addition, MD simulations with different ion configurations inside the channel pore suggest that movement of the ions is weakly coupled compared with the strong coupling in K channels, which provide a very snug fit for their preferred ion species. The most important biological function of Na channels is to catalyze Na permeation into the cell, while simultaneously preventing permeation by K ions. Within the integrated 2-D PMFs (), the barriers and wells depend on the occupancy in the filter. Na–Na displays the deepest energy wells, whereas K–K has the shallowest ones in the filter, indicating that the channel selectively binds Na. Calculation of the free energy of binding for two ions indicates that binding of the ion pair, Na–Na, is favored over K–K (). Furthermore, an entering Na ion needs to overcome a relatively high energy barrier to reach the energy minimum in the selectivity filter, when a K is already present (panel Na–K). This is further illustrated in the 1-D cross sections (), where one ion, K, is fixed in the filter at the position corresponding to the stable minimum. This example shows that it is energetically feasible for Na to pass K in the narrowest part of the filter, even though this may not represent the most probable reaction pathway for permeation. Comparison of the 2-D PMFs for E177D () shows that this mutation decreased the number of potential binding sites (). Furthermore, this conservative amino-acid substitution modulates the relative numbers of different coordinating ligands, thus altering local chemical moieties involved in ion coordination leading nearly the same relative binding free energies for Na and K to the filter. To illustrate how ion coordination along the permeation pathway is affected by the mutation, we performed analysis of the position-dependent ion coordination sphere for stable binding sites found in 2-D PMF maps (). The average coordination number of Na and K does not change substantially with the E/D mutation. However, the chemical composition of the coordination sphere does. There is a small decrease in the average number of carboxylates coordinating each Na ion from the extracellular side to the position of Z = 2.5 Å (free energy minimum for Na), as shown in . Meanwhile, the probability of finding carboxylates in the potassium coordination sphere increases by ∼60% in both stable sites identified by the PMF computations. Therefore, in the mutant, K can gain additional stabilization from carboxylates, whereas this mode of K coordination is largely missing in the WT. At the same time, K coordination by water decreases as compared with the WT protein. The water ligating K in the WT is replaced by carboxylates in the mutated pore. For example, in NaVAb E177D, the coordination with carboxyl oxygens was substituted by coordination with water molecules for Na but not for K (). Therefore, hydration and coordination in the pore combine to play an important role, which appears sufficient to define the properties of an important ion-binding site and the entry barrier of bacterial Nav channels. The results of PMF computations together with coordination analysis are consistent with the experimentally observed increase in relative potassium permeability. Thus, the E/D mutation allows for more favorable K binding to the filter as compared with WT. There are, of course, limitations imposed on our analysis by the use of a closed or “pre-open” structure to explore the underpinnings of selectivity. Certainly, a prediction of absolute ion throughput and conductance will require detailed analysis of an open pore structure, preferably for a channel from which single-channel data can be obtained. However, by focusing our analysis on experiments evaluating selectivity based on ion interactions near the extracellular entrance, we believe that we have obtained significant insight into the physiologically crucial discrimination between Na and K by these channels. The key strategy adopted by prokaryotic Nav channels to discriminate against K is as follows. First, the pore is large enough to allow for permeation of variably hydrated cations. Second, within the selectivity filter, ion coordination is highly dynamic, and a fine balance among ion ligation from carboxylates, carbonyls, and water molecules is at the heart of the selective entry and occupancy by external ions. Third, although a single binding site inside the filter displays a preference for Na over K, two-ion states are thermodynamically stable and are essential for rapid permeation. Finally, two procedures likely to modify the charge density near the narrowest point in the selectivity filter (E-D substitution and reduction of pH) reduce the discrimination between Na and K, likely because of small changes of both energy minima and maxima in the PMFs for the two ions.

Titin, the largest known protein (3–4 MD), resides in the sarcomere of striated muscle, where it extends from the Z disk to the M band and is responsible for the intracellular passive stress that develops when muscle is stretched (; ; ). Titin-based passive stress maintains the central position of the A band in the sarcomere (), which is important for efficient contraction, controls the physiological sarcomere length (SL) range (), and plays a role in various stress-dependent signaling pathways through its interaction with multiple binding partners (; ; ; ). Titin is encoded by a single gene, and differential splicing of the extensible spring-like segment of the titin molecule can vary the passive stress of muscles (; ). The spring segment in skeletal muscle consists of the PEVK region (so named because it is rich in proline, glutamic acid, valine, and lysine residues) and the proximal and distal tandem Ig-like–containing segments, located near the Z disk and the A band, respectively (; ; ). Titin’s serially linked Ig-like domains have a β-barrel fold characteristic of the intermediate I-set of the Ig superfamily (); their linker sequences unbend upon sarcomere stretch, giving rise to an increased end to end length of the tandem Ig segments. Differential splicing of the PEVK and Ig-like regions leads to the production of a titin protein that varies in size depending on the species, muscle type, and developmental stage, with a decrease in size that is associated with an increase in passive stress in the sarcomere (; ). Advancement in our understanding of the contribution of these spring-like elements of titin to the generation of passive stress have been made through the generation of mouse models in which various portions of the spring have been deleted. These include the deletion of exon 49, which encodes the cardiac-specific N2B element (), exons 219–225, which encode part of the PEVK region (), and exons 30–38, which encode 9 of the 15 constitutively expressed proximal tandem Ig domains (). Such models have been used to study the effects of spring element deletion on the physiology of cardiac muscle, but an in-depth investigation of the role of titin in skeletal muscle has not been performed. This is an important area of research considering that passive stress is altered in various skeletal muscle myopathies (, ; ; , ); it is unknown to what extent changes in titin are an effect of the myopathy or can cause myopathic changes. Here, we studied how skeletal muscle responds to deleting the constitutively expressed proximal Ig domains 3–11 (titin exons 30–38; mouse model referred to as IG KO) using a multidisciplinary approach that includes gene expression analysis, titin protein analysis, muscle mechanics, and mouse genetics. We found that in the IG KO mouse, skeletal muscle undergoes additional differential splicing to yield smaller titin isoforms, which results in greatly increased passive stress. Surprisingly, there are also changes in muscle trophicity and contractility. Exon expression analysis revealed that the small titin isoforms expressed in IG KO mice are full-length titins that arise through removal of PEVK sequences. We analyzed titin-binding proteins and found (also called CARP [cardiac ankyrin repeat protein] or MARP1 [muscle ankyrin repeat protein 1]) to be highly up-regulated, but by crossing the IG KO mouse with a mouse deficient in , we obtained conclusive evidence that is not necessary for additional titin splicing to take place. We also determined the expression level of the recently discovered titin splicing factor RBM20 (RNA-binding motif protein 20; ) and found it to be up-regulated in IG KO soleus muscle. Reducing RBM20 function in the IG KO mouse by breeding mice with a mouse model that is heterozygous for a deletion in RBM20 (IG KO, RBM20 HET) normalized titin isoform expression. Thus, the reduced size of titin’s tandem Ig segment in IG KO mice triggers elevated RBM20 protein levels, which further reduces the size of titin and increases passive stress. The implications of these findings for understanding and treating skeletal myopathies with altered passive stress are discussed. IG KO mice were created by deletion of titin exons 30–38, which correspond to nine proximal Ig domains (Ig 3–11). For details see . Only male mice were used for experiments and studied at 3 mo of age unless otherwise indicated. mice were generated in the University of Arizona (UA) BIO5 GEMMCore from embryonic stem cells obtained from the KOMP consortium (); the gene (encoding the protein) was targeted using the VelociGene technique with the ZEN-UB1 cassette. RBM20 mice were made at the UA BIO5 GEMMCore using homologous recombination. Exons 6 and 7 from the RBM20 mouse gene were deleted causing an in-frame deletion of the RNA recognition motif (RRM). All mouse strains were backcrossed onto a C57BL/6J genetic background. All animal experiments were approved by the UA Institutional Animal Care and Use Committee and followed the US National Institutes of Health “Using Animals in Intramural Research” guidelines for animal use. All mice were genotyped using GoTaq Green Master Mix (Promega). For IG KO mice the following primers were used: Common, 5′-GCAGCTACCCATATCATAGC-3′; KO specific, 5′-CACTAGCAGGAACATGTGTC-3′; and WT specific, 5′-GAACGGTGTGGAGATCAAGT-3′; expected product sizes: 319 (WT) and 268 bp (KO). For mice the following primers were used: Common, 5′-TCACTAGAGGATATTTTAACACC-3′; KO specific, 5′-TCATTCTCAGTATTGTTTTGCC-3′; and WT specific, 5′-CAGTCACCCGGAAGTCAAA-3′; expected product sizes: 318 (WT) and 286 bp (KO). For RBM20 the following primers were used: Common, 5′-ATATCTGCACCCATGTTTAGTTTCC-3′; KO specific, 5′-GAAGCCAGTGTGTTGGTATGG-3′; and WT specific, 5′-GTGGCCAGCCACGATAGC-3′; expected sizes: 498 (WT) and 817 bp (KO). To quantify spinal shape, animals were anesthetized with Avertin via i.p. injection. Animals were imaged on a GE Healthcare Lunar PIXImus and scanned. Animals were placed on their side in a right lateral recumbent position, and their ventral side was aligned using a straight template. Analysis of the high-energy image of a DXA scan was captured using PIXImus software analyzed offline by manual tracing of images. The lower-energy image showing body composition was not used for this study. Kyphosis index was used as described by others () and measured as AB/CD, where AB is the distance between the posterior edge of C7 and posterior edge of L6 and CD is the distance from line AB to the distal border of the vertebral body farthest from the line (). The soleus muscle was dissected and pinned onto cork at slack length. The tissue was then covered with OCT (Tissue-Tek) and frozen with liquid nitrogen–cooled isopentane and stored at −80°C. The belly of the muscle was cut crosswise, and 8-µm sections were collected on VWR glass microscope slides and stored at −20°C overnight. The microscope slides were then taken out and left to dry for 5 min. An ImmEdgePen (Vector Laboratories) was used to mark a circle around each section collected. Sections were skinned with 0.2% Triton in PBS for 20 min, followed by treatment with blocking solution (2% BSA and 1% donkey serum) in PBS at 4°C for 1 h. Primary antibodies were then applied to the sections () overnight at 4°C. The monoclonal anti-Myosin (skeletal, slow, myosin heavy chain I [MHC-I]; M8421; Sigma-Aldrich) recognizes an epitope located on the heavy meromyosin portion of human adult skeletal muscle slow myosin. The MHC-II antibody (M1570; Sigma Aldrich) stains the fast (type II) and neonatal isomyosin molecules found in skeletal muscle. After primary incubation, sections were washed with PBS twice for 30 min, secondary antibody was applied (2–4 h at room temperature), and they were washed with PBS for 30 min and then washed twice with water. Mounting media (K0424; Vector Laboratories) was then added to each section and sealed with a coverglass. Images were collected on an Axio Imager M.1 microscope (Carl Zeiss) using an Axio Cam MRC (Carl Zeiss). CSAs of the slow and fast myosin fibers were measured using the ImageJ program (National Institutes of Health). Stained cells were traced manually, and the CSAs were determined. The CSAs of ∼150 fibers of each fiber type from the soleus muscle were collected from two to four stained sections, and the total number of fibers in soleus muscle was counted. Titin was visualized on 1% agarose gels (16 × 18 cm) stained with Coomassie blue as described previously (). Total titin/MHC ratios and molecular weight estimates have been described previously (). MHC isoform composition was visualized using 8% acrylamide gels (16 × 18 cm; ). For Western blotting of titin, 0.8% agarose gels (16 × 18 cm) were ran for 3 h at 15 mA/gel and transferred to polyvinylidene fluoride membrane using a semi-dry transfer unit (Bio-Rad Laboratories) for 2.5 h at 150 mA. Western blots of titin-binding proteins were performed with 12% acrylamide gels, except RBM20, calpains, and obscurin, in which 4–20% acrylamide gels were used (13.3 × 8.7 cm; Criterion Cassettes; Bio-Rad Laboratories). All transferred blots were stained with Ponceau S to visualize total transferred protein. The blots were then probed with primary antibodies (Table S1) at 4°C overnight. To normalize for loading differences, actin from the Ponceau S–stained membrane was used unless otherwise indicated. Secondary antibodies conjugated with fluorescent dyes with infrared excitation spectra were used for detection. One-color IR Western blots were scanned (Odyssey Infrared Imaging System; LI-COR Biosciences), and the images were analyzed with One-D scan EX (Scanalytics Inc.). Soleus muscles were collected from WT and IG KO mice ( = 3) and stored in RNAlater (Ambion). RNA was isolated using the RNeasy Fibrous Tissue Mini kit (QIAGEN). A custom microarray with all titin exons has been described previously (; ). RNA was reverse transcribed and amplified using SenseAmp (Genisphere) and SuperScript III (Invitrogen) and dye-coupled with Alexa Fluor 555 or Alexa Fluor 647 (Invitrogen). Sample pairs were hybridized with 70 SlideHyb Glass Array Hybridization Buffer #1 (Ambion) in a GeneTAC Hybridization Station (Genomic Solutions); slides were scanned with an Axon GenePix scanner, and the results were analyzed using the R package CARMA (). For Affymetrix microarray analysis, soleus muscle tissue was dissected ( = 3) and stored in Ambion RNAlater (Invitrogen). RNA quality was assessed by NanoDrop 1000 Spectrophotometer and 2100 Bioanalyzer (Agilent Technologies); all samples had RIN 9.0–9.3. Samples were hybridized with the GeneChip Mouse Gene 1.0 ST Array (Affymetrix); processing (labeling through scanning) was performed by the Genomics Core, UA, according to Affymetrix protocols and using Affymetrix supplies and equipment. Data analysis was conducted using three different packages: Expression Console (Affymetrix) with RMA, BRB-ArrayTools (Biometric Research Branch, National Cancer Institute) with RMA, and Gene Array Analyzer () with either PLIER or RMA. The criteria for a significant difference was that the gene was significantly changed in analysis with all three packages (P < 0.001) and had a fold change greater than two between IG KO and WT samples. Intact muscle mechanics analysis was performed using the Aurora 1200A in vitro test system and has been described previously (; ). In brief, soleus muscle was attached between a combination servomotor-force transducer and fixed hook via silk suture in a bath containing oxygenated Ringer solution (145 mM NaCl, 2.5 mM KCl, 1.0 mM MgSO, 1.0 mM CaCl × 2HO, 1.0 mM HEPES, and 10 mM glucose, pH 7.4, 30°C). For passive force, the muscle was stretched from slack length to 10, 20, and 30% of the muscle length at 10%/s. The muscle was held for 60 s and then returned to slack length, waiting 7 min between each stretch. Measured force in millinewtons was normalized to CSA (muscle mass [mg]/(L0 [mm] × 1.056) to obtain stress (mN/mm). The optimal length (L0) was determined by adjusting muscle preload force until optimal fiber length for maximal twitch force was achieved (pulse duration of 200 µs with biphasic polarity). The procedures for skinned muscle contractility were as described previously (; ), with minor modifications. Soleus muscles were skinned overnight at ∼4°C in relaxing solution (mM: 20 BES, 10 EGTA, 6.56 MgCl, 5.88 NaATP, 1 DTT, 46.35 K-propionate, and 15 creatine phosphate, pH 7.0 at 20°C) containing 1% (wt/vol) Triton X-100. Preparations were washed thoroughly with relaxing solution and stored in 50% glycerol/relaxing solution at −20°C. All solutions contained protease inhibitors (in mmol/liter: 0.01 E64, 0.04 Leupeptin, and 0.5 PMSF). Small muscle bundles (diameter ∼0.06 mm) were dissected from the skinned muscles. Muscle bundles were attached to a strain gauge and a high-speed motor using aluminum foil clips. Experiments were performed at 20°C. SL was measured online by laser-diffraction using a He-Ne laser beam. The width and depth (using a prism) bundle diameters were measured with a 40× objective. The muscle bundle CSA was calculated from the mean of three width and depth measurements made along the length of the muscle bundle, and passive stress was determined by dividing the passive force by CSA. Muscle bundle length was activated in pCa4.5 activating solution (mM: 40 BES, 10 Ca-EGTA, 6.29 MgCl, 6.12 Na-ATP, 1 DTT, 45.3 potassium-propionate, and 15 creatine phosphate) and protease inhibitors (mM: 0.01 E64, 0.04 Leupeptin, and 0.5 PMSF) at SL 2.4 µm to record maximal active stress. The bundles were then set at slack in relaxing solution, and passive force was recorded while SL was increased to 3.0 µm (velocity, 0.1 muscle length/s), after which length was held constant for 60 s to observe stress relaxation, followed by a release to slack SL. Subsequently, thick and thin filaments were extracted by immersing the preparation in relaxing solution containing 0.6 M KCl (45 min at 20°C), followed by relaxing solution containing 1.0 M KI (45 min at 20°C). After the extraction procedure, the muscle bundles were stretched again at the same velocity, and the passive force remaining after KCl/KI treatment was assumed to be collagen based, and titin-based passive force was determined as total passive force minus collagen-based passive force. All data are represented as mean ± SEM. For data with numbers of mice () greater than or equal to eight, an unpaired test with a p-value <0.05 was considered significant. For experiments performed with data less than eight, a Mann–Whitney test, which does not assume Gaussian distributions, was performed (P < 0.05 significant). For MHC isoform analysis, active specific tension, and passive stress analysis, two-way ANOVA with a Bonferroni post-test was performed (P < 0.05 significant). For all figures: *, P < 0.05; **, P < 0.01; ***, P < 0.001. Figures showing the kyphosis analysis (Fig. S1), titin size analysis during development (Fig. S2), examples of Western blot results for titin-binding antibodies (Fig. S3), and single fiber protein analysis (Fig. S4) are included. A list of the antibodies used can be found in Table S1. Online supplemental material is available at . Mice deficient in Ig domains 3–11 of titin’s proximal tandem Ig segment (IG KO) have recently been shown to have a mild cardiac diastolic phenotype (), but the role of these proximal Ig domains in skeletal muscle had not been studied. Upon more detailed characterization, it was found that IG KO mice displayed a slight but significant spine curvature (kyphosis) as quantified by the kyphosis index (Fig. S1). Kyphosis is a commonly associated phenotype in mouse models of disease in which skeletal muscles are affected (; ). Compared with the kyphosis of other muscle disease models, the kyphosis in the IG KO was less prominent but nevertheless suggested that the IG KO mouse model outwardly displayed signs that could be associated with skeletal muscle myopathy. This prompted us to look at the effect of this deletion in skeletal muscle. Analysis of skeletal muscle mass revealed that the soleus and diaphragm muscles from IG KO mice were significantly smaller (raw weight and normalized to tibia length; ), whereas all other muscles exhibited no difference. In the mouse, most muscle types express mainly fast MHC. The two well-known exceptions that express a greater abundance of slow MHC are the soleus and diaphragm muscle (), and this suggested that there may be fiber type specificity to the atrophy phenotype whereby slow myosin-expressing fibers (type I) were more affected than fast myosin-expressing fibers (type II). We therefore measured the fiber CSA of type I and type II fibers from soleus muscle using immunofluorescence (). Interestingly, both fast and slow muscles underwent similar fiber atrophy (37.9% and 49.7%, respectively), indicating that the atrophy observed in the soleus muscle was not caused by effects specific to slow myosin-expressing muscle. Additionally, we found that the total number of muscle fibers was not different (not depicted), supporting the idea that atrophy of the muscle was caused by a decrease in the size of fibers. Gel electrophoresis of MHC isoforms revealed an increase in slow MHC in the IG KO soleus muscle ( and ). This shift in MHC isoform composition is likely part of a compensatory mechanism that develops postnatally as no shift in MHC isoform composition was observed in neonatal day five soleus muscle (). This isoform shift to a slow myosin isoform composition appeared to be soleus muscle specific as a survey of various skeletal muscles showed either no significant difference (tibialis cranialis [TC] and extensor digitorum longus [EDL]) or a significant decrease in slow MHC (diaphragm; ; and ). The changes in MHC isoform composition in soleus muscle of IG KO mice suggest that there may be changes in contractility of the soleus muscle. Slow myosin fibers differ from fast fibers in that they are more oxidative, have a lower power output, and resist fatigue (). Results from intact mechanics of soleus muscle showed that IG KO mice had a significant decrease in twitch force ( and ) and were more resistant to fatigue (). These findings are likely to reflect the change in MHC isoform composition from fast to slow. Passive properties of intact muscle were also studied. The soleus muscle contains one of the largest titin isoforms of adult striated muscle (), and the deletion of nine proximal Ig domains is expected to have a minimal effect on passive stress (the effect is a ∼2% stress reduction in the much smaller cardiac N2B titin []). However, passive stress in the intact soleus muscle was significantly increased in the KO, when stretched from its slack length, to a level much greater than in WT muscle (mean of 30.9 ± 0.2% change; ). A similar increase in passive stress also occurred in the EDL muscle (not depicted). Thus, passive stress of skeletal muscle is increased in IG KO mice but to an extent greater than expected by the deletion of only nine proximal Ig domains. At the level of the intact muscle, the extracellular matrix (ECM) contributes to passive stress in addition to titin and could have been the cause of the greater than expected increase in passive stress. To distinguish between these two contributors, skinned fibers were studied. Demembranized fiber bundles were passively stretched to determine the total stress, followed by an extraction of thin and thick filament proteins with KCl/KI to remove titin’s anchoring points in the sarcomere (). The fiber bundle was passively stretched again to determine the nontitin, ECM passive stress contribution. Titin-based stress is defined as the difference between total stress and ECM-based stress. No difference in maximal activated stress was observed between genotypes, consistent with the tetanic stress observed in intact soleus muscle (153 ± 7.5 mN/mm WT vs. 151 ± 13 mN/mm KO). The soleus muscle from IG KO mice was found to have a significant increase in titin-based passive stress above 2.5 µm SL, whereas the ECM was not significantly increased (). The mean percent change (defined as ((KO-WT)/WT) × 100) in titin-based passive stress is 30.2 ± 2%, similar to that observed in intact muscle. These results indicate that the large increase in passive stress observed in intact muscle reflects an increase in titin-based passive stress. Titin-based passive stress can vary through alterations in titin size, phosphorylation status, and titin/MHC stoichiometry (; ). Because a large increase in passive stress was observed in IG KO mice, we sought to determine the contribution of these various factors to the mechanical properties of muscle. Large pore agarose gels were used to visualize titin isoform size. The genomic region deleted to produce the IG KO was expected to yield a single mutant titin protein 88 kD smaller than WT titin. IG KO mice instead contained two faster-migrating bands (N2A and N2A) in the soleus, EDL, and diaphragm muscles (). The second faster-migrating band was most prominent in the soleus muscle. To test whether this faster-migrating band (N2A) was either a titin degradation product or a full-length titin isoform, Western blots were performed with antibodies raised against titin’s N terminus (Z1Z2 antibody) and C terminus (M8M9 antibody). The smaller band contained both of titin’s ends in both soleus () and EDL muscle (), and thus it is a full-length titin isoform. The molecular weight of the titin isoforms N2A and N2A in the soleus muscle was estimated as described previously (). The N2A and N2A isoforms were reduced in size by more than the expected 88-kD deletion (), indicating that adaptations in splicing had taken place in the IG KO. Interestingly, these mutant titin isoforms were developmentally regulated; they were not present at birth, and the N2A over N2A ratio increased with age (). Thus, changes in splicing that occur postnatally in the IG KO are both progressive and adaptive. To more directly observe changes in titin splicing, a titin exon microarray analysis was performed on 3-mo-old soleus muscle. As expected, the nine Ig domains removed in the IG KO (exons 30–38) were absent (). In the IG KO soleus muscle, there was also a down-regulation in exons that encode PEVK sequences. This down-regulation was confirmed at the protein level via Western blots performed using the 9D10 antibody (). This antibody recognizes repetitive sequences throughout the entire PEVK region (; ). There was a decrease in labeling of the two mutant titin isoforms with the 9D10 antibody in the IG KO as normalized to total titin (Z1Z2 antibody), supporting the idea that the difference in isoform size was, at least in part, caused by additional differential splicing in the PEVK region of titin. In addition to alternative splicing, titin-based passive stress can also be modulated by phosphorylation. Skeletal muscle titin is a target of phosphorylation by PKC-α in the PEVK domain. Phosphorylation of two serines (PS11878 and PS12022) has previously been shown to significantly increase passive stress (). Phosphorylation of titin at serine 11878 was significantly reduced in IG KO soleus muscle, whereas phosphorylation at serine 12022 was unchanged (); the expected net effect of this reduction in phosphorylation is a lower passive stress. This expected effect will partially offset the passive stress increase produced by the reduction in titin size caused by differential splicing in the IG KO (see also Discussion). The amount of titin relative to the sarcomeric protein myosin reflects the stoichiometry of titin to thick filaments. Total titin/MHC ratios were not significantly different between genotypes in all muscles observed (). This indicates that the deletion of Ig domains in the IG KO did not affect the number of titin filaments per thick filament. The titin degradation product, T2, was also not significantly altered by Ig domain deletion (). In summary, soleus muscle titin from IG KO mice undergo additional alternative splicing and alterations in phosphorylation. It has been proposed previously that titin-binding proteins can function in signalosomes to sense strain and modulate protein expression (; ), and it has been hypothesized that changes in these signalosomes could be responsible for changes in gene expression. Titin-binding protein abundance was assayed by Western blot analysis in the soleus muscle, and it was found that proteins that bind in the I-band region were differentially regulated, whereas those in the M line and A band were less affected ( and ). The calpain (CAPN) family are a group of calcium-dependent intracellular proteases, some of which bind titin in the N2A region (calpain-3; ) or the proximal Ig domains (calpain-1, binding site deleted in IG KO; ; ). CAPN1 (catalytic subunit of calpain-1) amount was quantified by antibodies against the full-length inactive CAPN1 (pre-calpain) and self-autolysis–activated CAPN1 (post-calpain; ). There was no significant difference between active and inactive forms of calpain-1 between WT and IG KO mice in soleus muscle. There was also no significant difference in the amount of calpain-3 present between genotypes (). Of the proteins that bind titin’s I band, FHL-1 and FHL-2 were modestly up-regulated, whereas (also called MARP2) and (DARP/MARP3) were modestly down-regulated in 3-mo-old IG KO soleus muscle as compared with WT (). Most dramatic was the induction of (CARP [cardiac ankyrin repeat protein]), which was up-regulated in the IG KO soleus muscle >60-fold (). was considered a candidate for inducing the additional differential splicing of titin in IG KO mice for several reasons. First, is normally expressed at high levels in cardiac muscle where titin size is greatly reduced. Second, mice deficient in the three muscle ankyrin repeat family members, , , and (MKO), displayed an increase in titin size (). Finally, can act as a transcription factor in the nucleus to alter gene expression through inhibition of the NF-κB pathway (). If up-regulation were necessary for additional differential splicing to occur, then titin would not undergo additional differential splicing in -deficient IG KO mice. To test this, KO mice were bred to IG KO mice to produce double KOs. Mice were studied at 1 mo of age, a time when the N2A isoform is present. As highlighted in , deletion of did not prevent the additional differential splicing, and both N2A and N2A were expressed in the double KO mice. Recently, it has been uncovered that the splicing factor RBM20 plays a role in regulating alternative splicing of titin, with an increase in RBM20 expression leading to exclusion of PEVK exons in a splice reporter assay (). A 2.1-fold increase in RBM20 was found in IG KO soleus muscle relative to WT (). RBM20 expression was normalized to both MHC and another nuclear splicing factor U2AF65, as performed by others (), but the same result was found. These findings support the idea that RBM20 up-regulation leads to exclusion of PEVK exons in titin. To identify potential pathways in which increased titin-based stress leads to up-regulation of RBM20 and additional differential splicing, an Affymetrix GeneChip was used. The microarray analysis found 28 mRNA transcripts significantly altered in IG KO mice. Many of the genes identified in the Affymetrix microarray reflect the transition to a more slow fiber phenotype. These include up-regulation of mitochondrial gene UCP2 and down-regulation of genes expressed in fast fiber types, including Myoz1, Kcng4, Peg3, Myom2, and Nos1 (; ). Among the 28 differentially expressed transcripts, only two transcription factors were found: and Runx1 (). Runx1, a transcription factor which regulates blood cell differentiation (), is unlikely to be responsible for changes in skeletal muscle titin. Furthermore, we found no Runx1-binding sites within the RBM20 promoter region (not depicted). These data suggest that the transcription factor that is responsible for up-regulation of RBM20 might not be regulated at the level of mRNA expression. To test whether RBM20 activity is necessary for additional differential splicing, IG KO mice were crossed with mice deficient in the RRM domain of RBM20 (RBM20). Mice heterozygous for RBM20 displayed a single titin band intermediate of the titin band in mice that were either WT or homozygous for the RBM20 (). This is consistent with what has been shown previously in the rat model with a spontaneous mutation in RBM20 (). These data indicate that RBM20 functions in a dose-dependent manner to regulate differential splicing of titin. In 3-mo-old mice that are deficient in Ig 3–11 and that are heterozygous for RBM20 (IG KO, RBM20 HET), only a single titin bands appears. This data support the interpretation that RBM20 up-regulation is required to cause additional differential splicing of titin in IG KO soleus muscle. sup #text

Transient receptor potential classical/canonical (TRPC) 3, C6, and C7 channels are the closest mammalian homologues of the TRP channel and are expressed in various cell types, including smooth muscle and neurons (; ; ). (Na, Ca) in response to stimulation of receptors coupled to phospholipase C (PLC), namely G protein–coupled receptors and certain tyrosine kinase receptors (). currents mediated by these channels are often called receptor-operated cation currents (; ). also activated by synthetic membrane-permeable diacylglycerol (DAG) analogues and are thus considered to be DAG-sensitive or activated channels (; ). hydrolytic activity of PLC, which is located downstream of the receptors for neurotransmitters and hormones. TRPC3/6/7 channels through the production of DAG to generate the receptor-operated TRPC currents (; ; ). the major substrate of PLC, and its hydrolysis produces DAG. known to regulate numerous ion channels, modulating electrical signal outputs from metabotropic receptors in diverse physiological contexts (; ; ). of PIP inhibits the activity of DAG-sensitive TRPC3/6/7 channels, both in an exogenous expression system and in smooth muscle–derived cells (A7r5) (; ). inhibition of TRPC3/6/7 currents was detected even in currents evoked by a membrane-permeable DAG analogue (OAG), which suggests that reduction in PIP can inhibit TRPC3/6/7 channel opening regardless of the presence of DAG. PIP hydrolysis (breakdown) by the receptor-activated PLC largely contributes to the production of DAG. of PIP and DAG suggests that TRPC3/6/7 channel activity may be regulated in a self-limiting manner. hydrolysis of PIP are not known. simultaneously measured PIP or DAG dynamics and receptor-operated TRPC currents evoked by carbachol (CCh; a muscarinic receptor agonist) or vasopressin (a vasoconstrictor). PIP and DAG using a quantitative Förster resonance energy transfer (FRET)-based sensor alongside detection of TRPC6 and TRPC7 currents, which are more sensitive than TRPC3 currents to reduction in PIP in human embryonic kidney (HEK)293 cells and smooth muscle–derived cells. correlate with the time course of activation and inactivation of receptor-operated TRPC6/7 channel currents. PIP–DAG signaling. DrVSP-derived functional dissociation constants for PIP binding to TRPC3/6/7 channels, closely resembled our experimental results. experimental and computer simulation data, revealed the crucial role of receptor-stimulated PIP hydrolysis in TRPC6/7 currents. work presented here has appeared in abstract form (). The pcDNA3 expression vector encoding human TRPC6 (GenBank accession no. ) was provided by T. Pharmakologie und Toxikologie, Zürich, Switzerland); pCI-neo expression vectors encoding mouse TRPC3 (GenBank accession no. ) and TRPC7 (GenBank accession no. ) were provided by Y. Japan). QuikChange Site-Directed Mutagenesis kit (Agilent Technologies) according to the manufacturer’s instructions. super-enhanced YFP or CFP isolated from a RhoA FRET sensor (provided by M. a monomeric form (CFP or YFP) (). fused to the N-terminal side of the PLCδ Pleckstrin homology domain (PHd; provided by K. CFP-PHd or YFP-PHd. CFP was fused to the C-terminal side of PKCε (provided by M. PKCε-CFP. PKCE-CFP, YFP was attached to the C-terminal side of the GAP-43 myristoyl domain (Invitrogen) through an octaglycine (G8) linker (Myr-YFP). each incorporated into an IRES-reporter region–excluded pIRES2 expression vector (Invitrogen). 1 receptor (MR) was provided by T. Tokyo, Japan). Human phosphatidylinositol-4-phosphate-5-kinase (PIP5K; β isoform) in pcDNA3.1 vector (Invitrogen) was provided by S. Kita and T. (Fukuoka University, Fukuoka, Japan). entirely. modified Eagle’s medium (Invitrogen) supplemented with 10% FBS (Gibco) and antibiotics (penicillin and streptomycin; Gibco) at 37°C (5% CO). poly--lysine–coated glass coverslips (Matsunami) in 35-mm culture dishes and transfected with a mixture of plasmid vector–incorporated DNAs using the SuperFect transfection reagent (QIAGEN). cotransfected with 1 µg each of plasmids encoding TRPC3, 6, or 7 together with MR (or without it, in endogenous muscarinic receptor stimulation) and 0.3 µg each of plasmids encoding CFP-PHd and YFP-PHd. For DAG detection, 0.3 µg each of plasmids encoding Myr-YFP and PKCε-CFP were cotransfected instead of the PIP sensor plasmids. of local PIP around the TRPC7 channel, the sequence encoding the donor protein (CFP) was inserted before the stop codon of TRPC7. (YFP-PHd), and MR plasmids were used for transfection. after transfection. the ATCC, maintained in medium identical to that used for HEK293 cells, and passaged every 5–7 d. as the one used with HEK293 cells. CFP-PHd and YFP-PHd were reseeded on poly--lysine–coated glass coverslips and incubated at 37°C (5% CO) for at least 15 min before use. always used within 2 h of reseeding. The standard external solution contained (mM): 140 NaCl, 5 KCl, 1 CaCl, 1.2 MgCl, 10 HEPES, and 10 glucose (pH 7.4, adjusted with Tris base; 300 mOsm, adjusted with glucose). contained (mM): 120 CsOH, 120 aspartate, 20 CsCl, 2 MgCl, 5 EGTA, 1.5 CaCl, 10 HEPES, 2 ATP-Na, 0.1 GTP, and 10 glucose (pH 7.2, adjusted with Tris base; 290–295 mOsm, adjusted with glucose). concentration (100 mM in HO). currents, NMDG solution (150 mM -methyl--glucamine chloride, 10 mM HEPES, and 1 mM CaCl, with pH 7.4 adjusted with HCl) was applied at the end of each stimulus. in DMSO (Wako Chemicals USA). vasopressin (AVP; 100 µM; MP Biomedicals) and nifedipine (10 mM; EMD Millipore) were dissolved in HO and DMSO, respectively. nifedipine were freshly prepared in the standard external solution to final concentrations of 1 and 5 µM, respectively, before applying to A7r5 cells. external solution and gravity-fed at a flow rate of 0.25 ml/min. was turned on and off using electromagnetic solenoid microvalves (The Lee Co.). #text F l u o r e s c e n c e f r o m v o l t a g e - c l a m p e d c e l l s w a s d e t e c t e d u s i n g a m i c r o s c o p e ( 6 0 × 0 . 9 N . A . o b j e c t i v e ; T E 3 0 0 E c l i p s e ; N i k o n ) e q u i p p e d w i t h a t w o - c h a n n e l s i m u l t a n e o u s b e a m - s p l i t t e r ( D u a l - V i e w 2 ; P h o t o m e t r i c s ) a n d a h i g h s e n s i t i v i t y E M C C D c a m e r a ( E v o l v e 5 1 2 ; P h o t o m e t r i c s ) . E x c i t a t i o n l i g h t f i l t e r e d a t 4 2 7 / 1 0 a n d 5 0 4 / 1 2 n m w a s a l t e r n a t e l y i n t r o d u c e d v i a a n o p t i c a l f i b e r f r o m a l a m p h o u s e e q u i p p e d w i t h a h i g h s p e e d e x c i t a t i o n w a v e l e n g t h s e l e c t o r ( 7 5 W x e n o n l a m p ; O S P - E X A ; O l y m p u s ) . E p i f l u o r e s c e n c e f r o m t h e c e l l s w a s p r e f i l t e r e d u s i n g a m u l t i b a n d d i c h r o i c m i r r o r ( 4 4 9 – 4 8 3 a n d 5 3 0 – 5 6 9 n m ) c o n t a i n e d i n t h e m i c r o s c o p e , a n d t h e n f u r t h e r s e p a r a t e d i n t h e b e a m - s p l i t t e r ( a t 5 0 5 n m ) a n d f i l t e r e d a t 4 6 4 / 2 3 n m ( d e t e c t i o n o f t h e d o n o r f l u o r e s c e n c e ) o r 5 4 2 / 2 7 n m ( d e t e c t i o n o f t h e a c c e p t o r f l u o r e s c e n c e ) . O p t i c a l f i l t e r s w e r e o b t a i n e d f r o m S e m r o c k , e x c e p t t h e s p l i t t e r ( C h r o m a T e c h n o l o g y C o r p . ) . T h e d u r a t i o n o f c a m e r a e x p o s u r e w a s 1 0 0 m s a n d o c c u r r e d w i t h i n 1 5 0 - m s p e r i o d s o f i l l u m i n a t i o n a t e a c h e x c i t a t i o n w a v e l e n g t h . I m a g e s w e r e c a p t u r e d w i t h a n E M g a i n o f 3 0 0 a n d t h e n d i g i t i z e d a s 5 1 2 × 5 1 2 p i x e l s b y 1 6 - b i t a r r a y s i n t h e m i c r o s c o p e s o f t w a r e ( M i c r o - m a n a g e r v . 1 . 4 ) . T h e i m a g e p i x e l r e s o l u t i o n w a s ∼ 0 . 2 6 µ m . A v e r a g e d i n t e n s i t i e s f r o m t h e w h o l e - c e l l r e g i o n ( t y p i c a l l y 2 0 × 2 0 t o 4 0 × 4 0 s q u a r e p i x e l s ) w e r e a n a l y z e d t o c a l c u l a t e F R E T u s i n g a c u s t o m - w r i t t e n M A T L A B p r o g r a m . T h e e l e c t r o p h y s i o l o g y a n d F R E T m e a s u r e m e n t s w e r e s y n c h r o n i z e d u s i n g b r i e f t r i g g e r s f r o m t h e A / D D / A c o n v e r t e r l i n k e d t o t h e e x c i t a t i o n l i g h t s h u t t e r . A l l o f t h e d a t a i n t h i s p a p e r w e r e r e c o r d e d f r o m t h e f i r s t a p p l i c a t i o n o f a n y o f t h e a g o n i s t s . sub xref disp-formula #text Here, we calculated the relationship between FRET and PIP. YFP-PHd can transit between a soluble cytoplasmic state and a PIP-bound membrane state; and FRET between CFP-PHd and YFP-PHd, where both probes transition between a soluble and membrane-bound state. PIP, is the dissociation constant of PHd bound to PIP (reported as 2.0 µM; ; ), and is the maximum at an infinitely high concentration of PIP that induces all the fluorophore-fused PHd probes to bind to the plasma membrane. is a purely theoretical value, because physiological PIP levels (5–40 µM) are insufficient to localize all PHd proteins to the plasma membrane (; ). is impossible to demonstrate, overexpression of PIP5K can greatly increase the cellular levels of PIP by approximately two- to threefold (). cells overexpressing PIP5K demonstrated ∼1.2 times higher than control cells with resting PIP levels (). value by multiplying the resting by this correction factor of 1.2. was used to simulate dynamics in cells expressing TRPC7-CFP and YFP-PHd, as shown in Eq. 25 () and . to PIP concentration (i.e., case 2) is described in the Results. cells was 1.6 µM. fluorescein (Sigma-Aldrich) as follows. Fluorescein was dissolved in 50 mM borate buffer (pH 9.1). fluorescein solution were measured, under the same conditions as the cells overexpressing YFP-PHd. standardized by the quantum yields for fluorescein (0.92) and YFP (0.57), and then the average [YFP-PHd] in living cell was determined to be 0.8 µM. YFP-PHd plasmids were transfected, so total amounts of CFP-PHd and YFP-PHd proteins could be extrapolated from this value. as ordinary differential equations, and the concentrations of PIP and DAG derived from these were incorporated into the channel-operation models. the forward Euler method with a time step of 0.05 s. translated into differential equations based on the proposed kinetic scheme. a Generalized Reduced Gradient algorithm of the Solver function in Excel. described in the Results. by model fitting to the currents ( and ) were evaluated by the standard deviations of residual as follows:where Norm and Norm. and back-calculated at the points. indicates the total number of time points (20/s). / relationship (Fig. S1). (Fig. S2). regulation by PIP–DAG signaling (SPD) model-based simulation (Fig. S3). (Fig. S4). under the AVP stimulation (Fig. S5). demonstrated less matching to dynamics and vice versa (Fig. S6). receptor-operated TRPC7 currents (Fig. S7). available at . 1,4,5-trisphosphate (IP) (). fluorophore-tagged PHds are located at the plasma membrane, making it possible to measure PIP at the plasma membrane (; ; ) (). measurements of PIP and receptor-operated TRPC6 current. FRET pairs, which consisted of donor (CFP) or acceptor (YFP) fused to the PHd, were coexpressed with TRPC6 channel and MR in mammalian HEK293 cells (). shows a typical example of the simultaneous measurement of a TRPC6 current and PIP levels in a HEK293 cell after stimulation with 10 µM CCh. Soon after CCh application, an inward TRPC6 current (, top) and FRET reduction (middle) were observed concurrently. the time required for receptor-operated TRPC6 current to increase from 10 to 90% of its peak amplitude and then to decay from 90 to 50% (, top). alteration in PIP level using the 3-cube FRET measurement (described in Materials and methods). CFP-PHd and YFP-PHd (, middle) were calculated from the respective fluorescence intensities (, bottom). fluorescence upon donor excitation (black circles; ) displayed inverted changes, whereas acceptor fluorescence (filled triangles; ) stayed largely constant (, bottom). quenching of sensor proteins during the recordings. reduction (τ) and the minimum amount of () under the receptor stimulation were obtained by fitting to the FRET data with a single-exponential decay function (, middle). We then explored the effect of PIP reduction on TRPC6 or TRPC7 channel currents at varying levels of receptor stimulation. endogenous muscarinic receptors in HEK293 cells with 100 µM CCh evoked prolonged TRPC6 current and a small amount of reduction in (). (Δ90–50%) were 32 ± 7 s and 39.2 ± 6.7 s ( = 6), respectively. current time course, τ was lengthened to 55.3 ± 4.5 s, and stayed near the resting level. Reduction in was only 18 ± 6%. shown that depletion of PIP can be induced by overexpression of MR (; ). with these studies, overexpressing MR and treating the cells with a high concentration of CCh (100 µM) greatly accelerated the TRPC6 current and FRET reduction () (Δ10–90% = 1.6 ± 0.3 s; Δ90–50% = 6.2 ± 0.9 s; τ = 4.5 ± 0.9 s), and enhanced the reduction with a near zero FRET efficiency ( = 1.15 ± 0.1 and = 0.016 ± 0.01; = 11). current was remarkably shortened compared with endogenous muscarinic receptor stimulation at the same concentration of CCh (100 µM; +MR = 2.8 ± 0.7 s; endo = 63 ± 8 s). peak time were observed when TRPC7 was expressed instead of TRPC6, except for the shorter activation and inactivation time for TRPC7 current (). TRPC7’s τ was slightly delayed and was also slightly attenuated compared with that of TRPC6. (Data from the different strength of receptor stimulation was summarized in .) To elucidate the functionality of PIP hydrolysis, we focused on the kinetic relationships between TRPC6/7 currents and FRET reduction. purpose, the log–log plots for the activation or the inactivation of TRPC6/7 currents and the values of τ were made using data obtained at varying levels of agonist stimulation, with and without MR overexpression (). τ and the current time courses (Δ10–90% [left panels] and Δ90–50% [right panels]) for both TRPC6 and TRPC7 currents. plot of the relation of Δ90–50% to τ in TRPC7-expressing cells showed significantly steeper slopes (slope = 1.33) than that in TRPC6-expressing cells (slope = 0.76) (). This steepness may reflect higher TRPC7 sensitivity to reduction in PIP than TRPC3 or TRPC6 (). FRET decay, the extent of the reduction or depletion of PIP levels () also showed a similar tendency to the time course of current activation or inactivation of each channel (). those of Such a scattering was probably caused by cell-to-cell variability in the released IP in response to the hydrolysis of PIP (). TRPC currents demonstrated that the activation and inactivation time courses related to both the kinetics and the extent of PIP reduction. unknown. with receptor-operated TRPC6/7 channel currents. relies on membrane translocation of DAG-activated PKC in response to increasing DAG levels at the plasma membrane () (). Ca-insensitive PKCε as a fluorescence donor molecule to exclude Ca-dependent translocation of PKC (). stimulation with CCh through the endogenous muscarinic receptors, TRPC6 current and DAG production were initiated almost simultaneously (). This parallel response is consistent with TRPC6 being a DAG-sensitive channel. Furthermore, during the inactivation of the TRPC6 channel current, DAG level also declined. the strength of receptor stimulation was weak, the production of DAG levels seemed to be a critical factor to the current appearances. receptor stimulation, when the cells overexpressed MR, the simultaneous measurement was revealed to be inconsistent. TRPC6 channel current paralleled DAG production, whereas the inactivation of the channel did not; there was no decline in DAG production (, left, purple zone). consistency between the current decay and DAG levels was even more prominent in the TRPC7 channel, which exhibited inactivation while DAG levels were still increasing (, right, and summarized in D). PKC-mediated phosphorylation inhibits channel opening, which has been proposed for TRPC3 (). inactivation (Δ90–50%) between cells overexpressing PKCε (TRPC6, MR, and DAG sensor–expressing, 7.5 ± 2.5 s; = 5) and control (TRPC6 and MR, 6.2 ± 0.9 s; = 11). has shown that PKCδ exerts a negative feedback effect via phosphorylation of Ser448 of TRPC6 (). TRPC6 and its corresponding mutant TRPC7, we tested whether the PKC-mediated phosphorylation of these residues is involved in the current decay. not detect any clear difference in the Δ90–50% time (100 µM CCh: TRPC6, 11.4 ± 2.4 s; TRPC6, 12.7 ± 2.7 s; TRPC7, 1.8 ± 1.1 s; TRPC7, 1.9 ± 0.9 s; = 8). phosphorylation site of TRPC6 has been identified at Ser (), which may play a role in the channel function. Δ90–50% (10.1 ± 3.5 s; = 7), which is consistent with the previous report (). PKC-mediated phosphorylation inhibits channel opening, because of the variety of cell and measurement conditions. DAG production level and PKC phosphorylation–mediated channel modulation may contribute less to the inactivation of TRPC6/7 channels at the robust stimulation. PIP and DAG, as described in the latter section. purpose, we presented the relation between the current increase and the production of DAG. (Δ10–30%) versus DAG production kinetics (τ) exhibited a clear correlation, with a smaller slope for the TRPC7 compared with the TRPC6-expressing cells (). the TRPC7 channel is highly sensitive to the increment of DAG. DAG sensitiveness in the initial parameter as TRPC7 > TRPC6 in the model simulation (, rows 21 and 22). PIP to the inactivation of TRPC6/7 channel currents, but the affinity of PIP to these channels is not yet known. voltage-controllable phosphoinositides phosphatase, was used to reduce intrinsic PIP (). PIP by activation of DrVSP led to concomitant inhibition of TRPC3/6/7 currents (). inhibition. To evoke the currents, the DAG lipase inhibitor, RHC80267, was used. by elevating the resting level of DAG (). duration of 500 ms, enables the current inhibition and FRET reduction to be observed simultaneously (). “()” and FRET reduction “()” against the depolarizing pulses that activate DrVSP, the various sensitivities of the TRPC3, TRPC6, and TRPC7 channels were quantified (). compared with the reduction in FRET (triangles). TRPC7 was highly sensitive to the reduction in FRET (, right). gradual changes in () and () (, middle). showing differential sensitivities of TRPC3/6/7 channels to PIP reduction, with an order of TRPC7 > C6 > C3 (). FRET reduction by DrVSP activation was also insufficient to reach a zero FRET level ( = 1.46 ± 0.07; = 14 at 120 mV). may simply be because of the incomplete depletion of PIP. idea raises a challenging question: how is TRPC3/6/7 inhibition by DrVSP activation observed at the single-channel level? To answer this question, DrVSP was activated during receptor stimulation in the cell-attached patch mode. TRPC6 channel during the bursting activity (). isotherm for PIP–TRPC3/6/7 channel binding. calculation of PIP concentration from the reduction in was done by a boundary function (described in Materials and methods). YFP-PHd proteins is cancelled by detachment of either the donor or acceptor fluorophore-fused PHd proteins from the membrane, reduction could be approximated as a cooperative square law of the membrane-bound fraction of PHd () as follows:where is the dissociation constant of PIP binding to the PHd. of is always positive, and solving PIP yields:where is the maximum value at an infinite concentration of PIP. PIP5K-overexpressing cells, that value was estimated as 1.2-fold higher than the resting of the control cells (). PIP concentration (, PIP) plots were fitted using the Hill equation () (). PIP − PHd = 2 µM (), the functional dissociation constants of PIP binding to TRPC3, TRPC6, and TRPC7 channels were estimated at 1, 2, and 5 µM, respectively. factors were incorporated into as the initial parameter for simulation in channel regulation by PIP (see below). TRPC channels, we attempted to simulate channel activity and compare the results with the experimental data. Our model consists of three components. of PIP recovery (, Eqs. 6–17). probability () and the resultant current based on the dynamics of DAG and PIP concentrations (, Eqs. 18–20). back-calculation of normalized of PIP sensor according to the concentrations of PIP and IP (; Eqs. 21–26). below. measurements. sum of the squared errors with 19 free parameters, the initial values of which are listed in . the model was assessed by a similarity of between the experimentally measured and simulated , which was obtained from the back-calculation of the resultant PIP concentrations by fitting to the currents, according to the equations described in the section on model Part 3 and . endogenous muscarinic receptors, the experimental was substantially compatible with the calculated from the simulated PIP changes (Fig. S4 A; SD = 0.12). In contrast, fitting the currents in MR-overexpressing cells to the DG model highly deviated from the experimental changes (Fig. S4 B; SD = 0.63). than that observed in the PKC-based FRET dynamics (Fig. dashed line). be useful for mimicking the delayed receptor-operated currents, but it is not totally suitable for the rapid case. PLC activity () was set to an accelerated kinetics ( = 0.7), a marked TRPC6 current inactivation emerged only in the SPD model but not in the DG model (, top). The SPD model was then tested. was used. current data, the computed PIP data were compared with the experimental dynamics. incorporated at (, row 23). whole-cell TRPC6 currents did indeed show overlapping of the dynamics of the experimental Norm. Norm., with a smaller SD value than in the DG model (SD = 0.04; = 4). was clearly evident in TRPC6 currents in MR-overexpressing cells (SD = 0.05; = 4; , middle). DAG, persisting well beyond the current decay (, bottom, solid dashed line). DAG was seen experimentally after robust receptor stimulation () and in recent work by . correlation between the experimental and modeled data supports the description of TRPC6 currents and PIP dynamics in the SPD model (fitted parameters are summarized in ). muscle–derived A7r5 cells (; ). (CFP-PHd and YFP-PHd) were coexpressed in A7r5 cells without exogenous expression of channels or receptors (, inset). and FRET reduction were observed similarly to the observations after stimulation of HEK293 cells with CCh (, top and middle). quickly (5–20 s after the application of AVP). is expected with the rapid desensitization of vasopressin receptors, and was reproduced by coexpressing vasopressin type 1A receptors with TRPC6 or TRPC7 channel in HEK293 cells (). simulation parameters for rapid desensitization (, rows 7 and 8) also overlapped the Norm. dynamics (, middle, blue line; SD = 0.12). the SPD model is useful even in the context of physiological cells. similarity to the experimental FRET achieved by fitting to TRPC7 currents observed in HEK293 cells and vice versa (; SD = 0.18; = 4). MR was overexpressed, but the FRET did not clearly show such irregular dynamics. was too slow to respond to the PIP dynamics compared with the time course of TRPC7 current, recorded by the electrophysiological method. To improve this issue, we redesigned the FRET pairs to detect PIP changes in the local vicinity of the TRPC7 channels. (CFP) directly fused to the channel was coexpressed with YFP-PHd (, inset). the plateau or biphasic current (, red arrow). “down-up-down” PIP dynamics (, bottom), we achieved an improved matching of Norm. (SD = 0.14) as well. up-regulation of PIP reflects a quick replenishment by diffusible PIP in the SPD model (). Norm. strongly support the fidelity of the SPD model for reproducing simultaneous events of receptor-operated TRPC currents and changes. #text

All animals were handled and experiments were conducted in accordance with procedures approved by the Institutional Animal Care and Use Committee at the University of California, Irvine. The ages and numbers of animals used for different types of experiments are detailed in Table. In the Calb2-Cre:tdTomato double transgenic mouse obtained by crossing the Calb2-Cre mouse line (Taniguchi et al., ) to a Rosa-CAG-LSL-tdTomato Cre reporter line (Madisen et al., ), dentate granule cells express strong red fluorescent (tdTomato) proteins so that their axon bundles (i.e., the mossy fiber tract) are fluorescently visible in hippocampal sections. All primary antibodies (Table) used in our immunostaining experiments are commercially available from major companies. The PCP4 gene encodes the PCP4 protein (also known as Pep19), which modulates calmodulin activity by activating the interaction of calcium with calmodulin (Kleerekoper and Putkey, ; Putkey et al., ). Expression of PCP4 is detected in the brain with the highest level in Purkinje cells of the cerebellum. In the mouse hippocampal formation, gene expression profiling studies via in situ hybridization show that the PCP4 gene has its strong expression around the region roughly corresponding to CA2, in addition to the granule cell layer of the dentate gyrus (Lein et al., ,). In the present study, we chose to immunochemically label mouse hippocampal sections with a PCP4 antibody (Santa Cruz Biotechnology, Santa Cruz, CA). Based on technical information from Santa Cruz Biotechnology, the PCP4 antibody is an affinity-purified antibody raised against a peptide mapped at the C-terminus of PCP4 of human origin; its specificity has been confirmed by western blot analysis of PCP4 expression in nontransfected and human PCP4-transfected 293T whole-cell lysates. As shown in , we independently confirmed the specificity of the PCP4 antibody in both mouse hippocampal and cerebellar sections by immunoabsorption tests using 10 μg/ml of the PCP4 peptide (sc-74816 P, Santa Cruz Biotechnology). In different sets of immunostaining experiments, we stained mouse hippocampal sections against calcium binding proteins such as calretinin (CR), calbindin-D28 K (CB), and parvalbumin (PV) (Swant, Bellinzona, Switzerland) to compare with PCP4 staining. In addition, in a few sections we used the antibody of synapsin 1 (Sigma-Aldrich, St. Louis, MO) to restain the same PCP4-stained sections to confirm the PCP4 immunoreactivity of the mossy fiber tract, which is known to course in the suprapyramidal layer (also known as the stratum lucidum) of CA3. These antibodies (included in the Antibody Database) have been widely used, and their specificity and effectiveness are confirmed in our previous work or in the published studies of others (Burkhalter, ; Dailey et al., ; Fletcher et al., ; Kawaguchi and Kondo, ; Xu et al., ,). Antibodies against fibroblast growth factor-2 (FGF2) and α-actinin were initially used in examining CA2 immunoreactivity, as some previous studies had used them as CA2 markers (Chevaleyre and Siegelbaum, ; Mercer et al., ). However, we found that in mouse hippocampal sections, their immunostaining was not clear or was diffuse in the CA2 region. We did not further examine their distribution or compare them with other markers. Based on technical information from EMD Millipore, the FGF2 antibody is routinely evaluated by western blot on Huvec lysates; it reacts strongly with basic fibroblast growth factor (FGF-2), but no cross-reactivity is seen with acidic FGF (FGF-1). Based on technical information from Sigma-Aldrich, α-actinin is an actin-binding protein present in both muscle and nonmuscle cells. The actinin monoclonal antibody shows wide reactivity with α-actinin in many species with various immunochemical techniques. The secondary antibodies, Cy3-conjugated or Alexa Fluor 488–conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch, West Grove, PA; 711-165-152 or 711-225-152,1:200 dilution) were used for fluorescent visualization of the immunostaining. None of the observed labeling was due to nonspecific binding of secondary antibodies or autofluorescence in the fixed tissue because sections labeled with secondary antibodies alone showed no detectable labeling. To stain tissue sections with antibodies, conventional fluorescent immunohistochemistry was performed as follows. Free-floating sections were rinsed 3–5 times with PBS with 0.1% Triton X, and incubated in a blocker solution for 2 hours at room temperature. The blocker solution contains 10% normal donkey serum, 2% bovine serum albumin, and 0.25% Triton X in PBS. Sections then were incubated with the primary antibody in the blocker solution at the appropriate dilution for 24–36 hours at 4°C. After the primary antibody incubation, sections were rinsed thoroughly with PBS (or working buffer: 10% blocker and 90% PBS), and then incubated with an appropriate secondary antibody in the blocker solutions for 2 hours at room temperature. After the secondary antibody solution was rinsed off, sections were counterstained with 10 μM 4′-6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich) for 10 minutes to help distinguish hippocampal subfields. Finally, sections were rinsed and wet-mounted, and were either directly coverslipped with the mounting medium Vectashield (H-1000, Vector, Burlingame, CA) or air-dried overnight, dehydrated, defatted, and then coverslipped with the mounting medium Krystalon (EM Science, Fort Washington, PA; 64969-95). The sections were examined, and low- and high-power images were acquired with an Olympus BX61 microscope equipped with a CCD camera (Hamamatsu Photonics, Tokyo, Japan) or a confocal microscope (LSM 700, Carl Zeiss Microscopy, Nussloch, Germany). For immunoabsorption control tests, the PCP4 antibody was first preincubated with the PCP4 peptide overnight in blocker solution and this solution was then applied to the sections for 24 hours and the remaining staining procedures were completed as described above. We did not see any specific immunolabeling in the sections used for these control tests (). To identify the CA2/CA1 border in a less arbitrary fashion in practice (see A,B), we used Adobe (San Jose, CA) Photoshop tools (blurring and thresholding) to perform Gaussian blurring with an empirically determined radius of 20 μm (about the size of two PCP4+ cell bodies), which resulted in exclusion of sparsely labeled PCP4-positive cells across hippocampal subregions. Furthermore, applying a threshold of 50% local maximal intensity (measured from distal CA3 and CA2) to the smoothed image facilitated identification of the CA2/CA1 border. This CA2/CA1 border identification was consistent with the cytoarchitectural border determined from the transition of DAPI nuclear staining in the pyramidal cell layer. In the processed images, the extent of PCP4-expressing CA3 was similarly determined. A custom analysis program written in Matlab (MathWorks, Natick, MA) was used to measure the extent of the PCP4 densely stained region versus the delineated CA2 portion, and CA3 length of the hippocampus. The measurements were obtained for the various ages of mouse brains (Table). We only took measurements from the sections in which we had good confidence in properly delineating distal CA3, CA2, and CA1 subregions. Unless specified, the data were measured from pyramidal cell layer of distal CA3 (within about 200 μm from the CA2/CA3 border), CA2, and proximal CA1 (200 μm from the CA2/CA1 border). In addition, the manual measurement tools of Adobe Photoshop were used to measure somatic sizes of clearly distinguishable individual PCP4+ cells in distal CA3 and CA2 using confocal image stacks. The Adobe Photoshop tool was also used to measure optical densities associated with DAPI nuclear staining in the pyramidal cell layer of distal CA3, CA2, and proximal CA1. As for morphological characterization of electrophysiologically recorded cells, the cells were first revealed with biocytin staining (see below), and reconstructions were conducted based on stacks of optical sections acquired by the confocal microscope. Then we examined the number of primary apical dendritic branches, and measured the distance between a primary branching point and the base of the apical dendrite (i.e., primary branching distance) using the Adobe Photoshop measurement tool. Horizontal hippocampal slices 400 μm thick were cut at the angle optimized to conserve the intrahippocampal axonal projections (Bischofberger et al., ) in well-oxygenated, ice-cold sucrose-containing cutting solution (in mM: 85 NaCl, 75 sucrose, 2.5 KCl, 25 glucose, 1.25 NaHPO, 4 MgCl, 0.5 CaCl, and 24 NaHCO). Two morphologically intact slices intermediate between dorsal and ventral hippocampus from each animal were used for experiments. Slices were first incubated in sucrose-containing ACSF for 30 minutes to 1 hour at 32°C, and then transferred to recording ACSF (in mM: 126 NaCl, 2.5 KCl, 26 NaHCO, 2 CaCl, 2 MgCl, 1.25 NaHPO, and 10 glucose). Throughout the cutting, incubation, and recording, the solutions were continuously supplied with 95% O−5% CO. Our overall system of electrophysiological recording, photostimulation, and imaging was described previously (Xu et al., ; Xu, ). Electrophysiological experiments were conducted at room temperature. To perform whole-cell recordings, neurons were visualized at high magnification (60× objective) with an upright microscope (BX51WI, Olympus, Tokyo, Japan). Cell bodies of recorded neurons were at least 50 μm below the slice-cutting surface and were initially targeted based on the pyramidal appearance of the cell soma and thick apical dendrite when possible. Patch pipettes (4–6 MΩ resistance) made of borosilicate glass were filled with an internal solution containing (in mM) 126 K-gluconate, 4 KCl, 10 HEPES, 4 ATP-Mg, 0.3 GTP-Na, and 10 phosphocreatine (pH 7.2, 300–305 mOsm). The internal solution also contained 0.1% biocytin for cell labeling and morphological identification. Once stable whole-cell recordings were achieved with good access resistance (usually <20 MΩ), basic electrophysiological properties were examined through hyperpolarizing and depolarizing current injections (ranging from −100 pA to 400 pA with a duration of 1,000 ms). We focused on analyzing intrinsic properties such as neuronal resting membrane potentials, spike rates, and spiking interval adaptation. We also performed spike shape analysis similar to that of Murphy and du Lac . The data are summarized in Table. The LSPS method has been previously used in the neocortex and hippocampus (Brivanlou et al., ; Dantzker and Callaway, ; Shepherd et al., ; Weiler et al., ; Xu and Callaway, ); it is a precise and useful approach for detailed local circuit mapping. A laser unit (DPSS Lasers, Santa Clara, CA) was used to generate 355-nm UV laser pulses for glutamate uncaging. Short pulses of laser flashes (1 ms, 20 mW) were controlled by using an electro-optical modulator and a mechanical shutter. The laser beam formed uncaging spots, each approximating a Gaussian profile with a width of 100 μm laterally at the 4× objective focal plane (Xu et al., ). Physiologically, under our experimental conditions, LSPS evoked action potentials only from stimulation locations within ∼100 μm of targeted somata (see ). In these calibration experiments, we did not find evidence of synaptically driven spiking evoked by distant photostimulation, because such trans-synaptic driving would appear in excitation profiles as spike-evoking sites far away from the perisomatic area of the recorded neurons. Additional data against synaptically driven spiking came from mapping experiments, in which synaptic input maps of the same recorded cells were acquired multiple times with different photostimulation powers. Even at the double uncaging power, the topography of input sources was unchanged with only increased input strength at the input locations. The stable map topography cannot exist if LSPS evoked inputs are derived from disynaptic or polysynaptic activation. Based on these control experiments, we conclude that synaptic inputs mapped in LSPS experiments are monosynaptic inputs from directly photostimulated sites. During mapping experiments, photostimulation was usually applied to 16 × 16 patterned sites (with an intersite space of 100 μm) covering the whole hippocampal slice. Glutamate uncaging was delivered sequentially in a nonraster, nonrandom sequence, following a “shifting-X” pattern designed to avoid revisiting the vicinity of recently stimulated sites (Shepherd and Svoboda, ). Whole-cell voltage-clamp recordings were made from the recorded neurons to measure photostimulation-evoked excitatory postsynaptic current (EPSC) responses at the holding potential at −65 mV (P6–P7) or −70 mV (P15–P18), which was based on the empirically determined γ-aminobutyric acid (GABA)ergic reversal potentials at the recorded mouse ages. Photostimulation data analysis has been described in detail (Shi et al., ). Photostimulation can induce two major forms of excitatory responses (Shi et al., ; Xu and Callaway, ): 1) direct glutamate uncaging responses (direct activation of the recorded neuron’s glutamate receptors); and 2) synaptically mediated responses (EPSCs) resulting from the suprathreshold activation of presynaptic excitatory neurons. Responses occurring within the 10-ms window from laser onset are considered direct. Synaptic currents with such short latencies are not possible because they occur before the generation of action potentials in photostimulated neurons. Therefore, direct responses need to be excluded from synaptic input analysis. However, at some locations, synaptic responses were over-riding on the relatively small direct responses and were identified and included in synaptic input analysis (see ). As for individual map construction, input measurements from different stimulation sites were assigned to their corresponding anatomical locations in the hippocampus; color-coded maps of average input amplitude and the number of events per site were plotted to illustrate overall input patterns to the recorded cell. The total input from each stimulation site was the measurement of the sum of individual EPSCs within the analysis window (>10 ms to 160 ms post photostimulation), with the baseline spontaneous response subtracted from the photostimulation response of the same site; the value was normalized with the duration of the analysis window (i.e., 150 ms) and expressed as average integrated amplitudes in picoamperes (pA). The analysis window was chosen because photostimulated neurons fire most of their action potentials during this time. To quantitatively compare input strength across cell groups, we measured the ESPC input amplitudes and the numbers of EPSCs across specific hippocampal subfields for individual cells. Note that as stratum lacunosum-moleculare (S-LM) only has sparse inhibitory neurons, but that pyramidal neurons located in the pyramidal cell layer could fire action potentials when their distal apical dendrites were stimulated in the S-LM layer (e.g., see ), EPSCs detected after photostimulation in the S-LM layer were not included for analysis to avoid repeated sampling. When two independent groups were compared, normally distributed data were analyzed by using a Student’s -test; when data were not normally distributed, a Mann–Whitney U-test was used. For statistical comparisons across three groups, we used the Kruskal–Wallis test (nonparametric one-way analysis of variance [ANOVA]) to identify the overall significance (e.g., CA3a, CA2, and CA1 cells) and Mann–Whitney U-tests for group comparisons (e.g., CA3a vs. CA2, CA3a vs. CA1, and CA2 vs. CA1). Alpha levels of  ≤ 0.05 were considered significant. All the values were presented as mean ± SD. For electrophysiological experiments, sample size was defined as cell number. For morphological measurements, unless specified, individual data points were measured and calculated for each section/slice and averaged for each animal, in which stood for animal number. In 6–8-week-old mice, hippocampal subfields CA1, CA2, CA3, and the dentate gyrus (DG) were differentiated on the basis of highly specific and strong anti-PCP4 immunostaining (). We follow the basic nomenclature of Lorente de Nó and Ishizuka et al. and use the terms (nearer the dentate gyrus) and (further away from the dentate gyrus) to designate positions along the transverse axis of the CA1 and CA3. In the mouse hippocampal proper, consistent with previous PCP4 in situ hybridization (Lein et al., ,; Zhao et al., ), strong PCP4 immunoreactivity was localized in both CA2 and the distal part of CA3 (CA3a), which is distinct from proximal CA1 (CA1c) (B,F,J). However, different from in situ hybridization, PCP4 immunostaining resulted in concurrent detection of the mossy fiber tract (A–D,I,J), which allows for distinguishing distal CA3 from the CA2 region (see below). Little PCP4 immunoreactivity was seen in the proximal portion of CA3 pyramidal cells, and CA1 pyramidal cells were sparsely labeled. As shown in A and B, an unbiased smoothing and thresholding protocol was applied for every image of interest to facilitate identification of the CA2/CA1 border in a less arbitrary manner (see Materials and Methods). Dentate granule cells were also strongly immunopositive for PCP4 (B,J), so that the bundles of their axons (mossy fibers) were detected in the stained sections. Double labeling using the synapsin-1 antibody in wild-type mice and genetic labeling of granule cell axons in Calb2-Cre:tdTomato double transgenic mice, confirmed the mossy fiber tract revealed by PCP4 immunoreactivity (C,G,K). Given that the extent of mossy fiber innervation remains a functionally useful criterion, the presence versus absence of mossy fiber tract distinguishes the CA3a/CA2 border in the PCP4-stained hippocampal sections. That is, as the termination of the mossy fiber tract traveling in the suprapyramidal layer of CA3 (Dailey et al., ; Romer et al., ; Zimmer and Haug, ) allows for precise delineation of PCP4-immunopositive distal CA3a, the other portion (∼200 μm in length) of the dense PCP4-stained region is unambiguously localized as CA2 that lacks the presence of the mossy fiber tract. Because the mossy fiber tract ends relatively bluntly at the CA3/CA2 border at horizontal sections, but gradually tapers off toward the border in coronal sections, horizontal sections should be preferred for CA2 examination when possible, even though we used both horizontal and coronal sections for our analysis. Differential cytoarchitecture supports the finding that the PCP4-delineated CA2 region differs from distal CA3 or proximal CA1 (E–H, C,D). In the DAPI-stained sections, distal CA3 had loosely packed, large cell nuclei located in the pyramidal cell layer, whereas CA1 had densely packed, compact nuclei in the pyramidal cell layer. In comparison, the CA2 pyramidal cell layer had cell density that was intermediate between CA1 and CA3. Consistent with qualitative observations, the mean optical densities of the pyramidal cell layer in distal CA3 (within about 200 μm from the CA3/CA2 border), CA2 and proximal CA1 (200 μm from the CA2/CA1 border) showed differences, with their values being 11.2 ± 4.5 (mean ± SD; unadjusted raw gray scale unit, 0 for black, 255 for white), 14.8 ± 5.8, and 16.3 ± 5.8, respectively ( = 10 slices from five mice, Kruskal–Wallis test:  = 0.04, Mann–Whitney U-tests for group comparisons: distal CA3 versus CA1  = 0.03, distal CA3 versus CA2  = 0.08; CA2 versus CA1  = 0.3). Furthermore, the overall somatic sizes measured from PCP4+ cells in distal CA3a were larger than those of CA2 (C,D), with the average values being 363.3 μm ± 58.7 μm and 264.4 μm ± 68.4 μm ( = 20 and 16 cells for CA3a and CA2, pooled from five sections from four mice), respectively ( = 0.0002). In addition, the PCP4-defined CA2 was verified by its neuronal morphology, electrophysiology, and local circuit connectivity, which differed from distal CA3 and proximal CA1 (see below). Together, these data establish the PCP4-based method for proper identification of the CA2 region. Considering the dorsal and ventral variations of the hippocampus in its anatomical and neurochemical organization (Jung et al., ; Sahay and Hen, ), we further examined the variations of PCP4 immunostaining and the CA2 extent at different dorsal–ventral levels across horizontal sections (), and at different anterior–posterior positions across coronal sections () of mouse hippocampus (6–8 weeks old). In the dorsal sections, the average lengths of delineated CA2 region and total PCP4 stained region were 224.5 ± 30.4 μm and 317.5 ± 33.3 μm (measured from three to six sections of each animal and averaged across four mice), respectively, whereas in the ventral sections, the lengths of delineated CA2 region and total PCP4 densely stained region were 197.9 ± 26.4 μm and 285.6 ± 54.4 μm, respectively. In the anterior sections (AP ≥ −1.5 mm), the average lengths of delineated CA2 region and total PCP4 densely stained region were 116 ± 16.5 μm and 375 ± 23.8 μm (measured from three sections of each animal and averaged across two mice), respectively; in the middle sections (AP −1.7–2 mm), the average lengths of delineated CA2 region and total PCP4 densely stained region were 95.8 ± 14.8 μm and 347.3 ± 14.7 μm, respectively; in the posterior sections (AP −2.2–2.8 mm), the average lengths of delineated CA2 region and total PCP4-stained region were 142.8 ± 23.2 μm and 446 ± 38.7 μm, respectively. It should be noted that our immunostaining data indicate that PCP4 gene expression encompasses a large proportion of CA3a particularly at dorsal or ventral sections, similar to what was previously acknowledged with in situ hybridization (Lein et al., ,). We compared the use of PCP4 with other molecules that can be potentially used as CA2 markers. Although FGF-2 and α-actinin have been used for CA2 studies (Chevaleyre and Siegelbaum, ; Mercer et al., ), we found that their immunostaining was not clear in the CA2 region. Among the calcium binding proteins tested (calretinin, calbindin-D28K, and parvalbumin), we found that calbindin-D28K (CB) immunostaining can be used for hippocampal subfield delineation (), which is supported by previous work (Forster et al., ; Sloviter, ). Strong CB immunoreactivity was present in the mossy fiber tract and CA1 pyramidal cell layer, and was relatively weak in the transitional CA2 region, but not in the CA3 pyramidal cell layer (B,F). Given that there is some uncertainty in the use of PCP4+ mossy fibers to define the border between distal CA3 and CA2, particularly at anterior coronal or dorsal horizontal sections where the ending of mossy fibers is not well defined, colabeling of mossy fibers with CB staining (B,F) together with PCP4 staining (A,E) can help with precise determination of the distal CA3/CA2 border (C,D, G,H). Nevertheless, it appears that single PCP4 immunostaining is more effective and definitive than CB staining in determining the CA2 region. We continued to determine whether the PCP4-delineated CA2 region can be distinguished from distal CA3 and proximal CA1. We examined the detailed morphology, intrinsic physiology, and local circuit connections of mouse hippocampal CA2 cells, using whole-cell recordings and LSPS in living brain slices of P15–P18 mice. Under differential interference contrast (DIC) microscopy, the CA2 region in the slice can be visually identified as a narrow region intermediate between CA3, determined by its suprapyramidal layer (i.e., the stratum lucidum) through which the mossy fiber tract courses, and CA1, characterized by its compact cell body layer. The PCP4 staining worked sufficiently well in 400-μm-thick slices postfixed after physiological recording, and confirmed the CA2 region identified in the living slice image (A–H). Differential cell morphology was revealed by cell fills followed by intracellular biocytin staining across CA3 (including CA3a, CA3b), CA2, and CA1 (, ). These results are in accordance with CA2 identification by PCP4 immunostaining. Wherease excitatory pyramidal cells in CA3 had complex postsynaptic spines termed thorny excrescences in their proximal dendrites (J,M,N), CA2 excitatory cells, like CA1 cells, had dense dendritic spines, but did not have such thorny excrescences (K,L,O). This morphological feature, as originally identified by Lorente de Nó using Golgi staining, suggests that CA2 excitatory cells do not receive direct synaptic inputs from dentate granule cells as do CA3 excitatory cells. As CA3 and CA1 have large and small pyramidal cells, respectively, and CA2 have CA3-like pyramidal cells shown in Golgi staining (Lorente de Nó, 1934) (also see C,D), our quantification indicates that the average somatic sizes of intracellularly labeled pyramidal cells differ across these regions; the least to greatest average somatic size is CA1<CA2<CA3 (D; Kruskal–Wallis test:  = 0.002, Mann–Whitney U-tests for group comparisons: CA3 versus CA2  = 0.02, CA3 versus CA1  = 0.002, CA2 versus CA1  = 0.045). In comparison with CA3 cells, both CA2 and CA1 excitatory cells tended to have fewer primary apical branches and more distant branching points (E,F). Whereas CA2 neurons show some morphological differences with CA1 neurons, CA2 excitatory cells are generally similar to CA1 excitatory cells electrophysiologically (Table). In contrast, CA2 neurons differed significantly from distal CA3 cells for many electrophysiological parameters including membrane capacitance, spike width, firing rate, and degree of spike interval adaptation (Table, C,G–I). To address whether PCP4-positive CA3a neurons electrophysiologically differ from other CA3 neurons not expressing PCP4, we compared the parameters of distal CA3 cells closely adjacent to CA2 ( = 9 cells) with those of CA3c/b excitatory cells ( = 15 cells). The two groups of cells (CA3c/b vs. PCP4+ CA3a) differed significantly overall in terms of such parameters as average spiking rate at the current injection of 200 pA (7.53 ± 4.52 Hz vs. 16.7 ± 8.52 Hz,  = 0.005), the average of the first two interspike intervals (82.9 ± 36.7 ms vs. 44.1 ± 38 ms,  = 0.009), and spike interval adaptation index (0.54 ± 0.2 vs. 0.33 ± 0.24,  = 0.02). Thus, the results indicate that PCP4-positive CA3a neurons are different from other CA3 neurons in CA3c/b subdivisions not expressing PCP4. Compared with CA3c/b cells, these PCP4-positive distal CA3a cells are more similar to CA2 cells. We conducted photostimulation mapping experiments to examine intrahippocampal circuit connections to CA2, CA3, and CA1 excitatory cells. LSPS combined with whole-cell recordings allows for mapping inputs from many stimulation sites over a large region to single recorded neurons, which has a great technical advantage over paired intracellular recordings limited in highly localized circuits. Specifically, the LSPS approach involves recording from single neurons, and then sequentially stimulating at other sites in order to generate action potentials from neurons in those sites to establish the map of input sources for the recorded neuron based on activation of presynaptic inputs. As LSPS evoked action potentials from the recorded cells in DG, CA3, and CA1 in a spatially restricted manner (), combined with other control experiments (see Materials and Methods), it is inferred that this approach has a sufficient spatial resolution to map direct synaptic inputs from specific hippocampal subfields and DG to individual excitatory neurons. Our results indicate that CA2 excitatory pyramidal cells receive photostimulation-evoked input from CA3, but do not have photostimulation-evoked input from DG (). The CA2 cells received most excitatory inputs from CA3a and CA3b, and the average strengths of summed excitatory synaptic input (excitatory postsynaptic currents, EPSCs) measured from stimulation sites of CA3a and CA3b were 50.1 ± 46.8 pA and 88.9 ± 154.6 pA ( = 11 cells), respectively. Whereas CA2 cells received input from within CA2 (19.5 ± 27.7 pA), they did not receive photostimulation-evoked input from DG or CA1, compared with baseline spontaneous activity. That is, statistical analysis showed that the EPSC input amplitudes from DG or CA1 during photostimulation did not differ significantly from those of spontaneously occurring EPSCs. To further distinguish spontaneous events from photostimulation-evoked inputs from DG sites, multiple photostimulation maps were collected from the same CA2 cells and maps were merged into an average map by a “two successes” filtering criterion described in a published study (Zhang et al., ). Only sites associated with EPSCs in the analysis windows during at least two out of three trials were considered to be actual input sites. This analysis confirmed the absence of DG input to CA2 cells (D–F). In contrast, in the same slices in which CA2 cells ( = 6) were recorded, CA3 cells received clear DG input (G–I), with the average input strength from DG being 49.7 ± 48.3 pA (measured from five CA3a and three CA3b cells). Overall, CA2 cells had weaker intrahippocampal excitatory inputs than CA3 cells, as the average strengths of summed EPSC inputs measured from CA2 ( =11), CA3a ( = 8), and CA3b ( = 7) excitatory pyramidal cells were 149 ± 127.4 pA, 262.9 ± 235 pA, and 386.9 ± 301 pA (CA2 vs. CA3a  = 0.045, CA2 vs. CA3b  = 0.02), respectively. Whereas CA3a cells received excitatory inputs from CA2 region with its average strength being 52.1 ± 65.9 pA, CA3b did not receive CA2 input. In addition, we found that CA1 excitatory pyramidal cells ( = 10 cells) received the majority of excitatory synaptic input from the CA3 region and weaker inputs from the CA2 and CA1 regions (), with the average input strengths measured from proximal CA3, distal CA3, CA2, and CA1 being 94.2 ± 63.9 pA, 51.1 ± 42.1 pA, 8.9 ± 18.7 pA, and 16 ± 13 pA, respectively. With the PCP4-based method for CA2 identification in hand, we used immunostaining of the surrogate marker PCP4 in conjunction with gross slice morphology and circuit connection mapping to examine the developmental property of CA2. We assessed PCP4 immunostaining in horizontal hippocampal slices of postnatal animals at P1–P2, P4–P5, P6–P7, P10–P15, and P21–P25 (; see Table for the number of animals for each age group). PCP4 immunoreactivity appeared weak and diffuse around the presumptive CA2 and CA1 at P4–5 (C,D). The intensity of staining gradually increased by the end of first postnatal week (P7) (E–H). The PCP4-positive region became successively more refined by the middle of the second week (i.e., around P12); CA2 could be delineated from the adjacent distal CA3 region (I–L). By P21 and beyond, the CA2 region revealed by PCP4 immunostaining became spatially restricted and reached adult-like form. Note that the appearance and strengthening of PCP4 immunoreactivity at CA2 concurs with the development of mossy fiber tract revealed by PCP4 or CB staining. Thus, assuming that the strength of immunolabeling corresponds directly to the strength of PCP4 gene expression, gene driven labeling would be expected to be more useful for later ages due to the gradual emergence of PCP4 immunostaining. The CA2 region does not appear to grow much in size during development (G,K,O), as the widths of PCP4-delineated CA2 regions are 176.1 ± 7.6 μm, 183.4 ± 34.8 μm, and 204 ± 25.5 μm (measured from three to five sections of each animal and averaged across three to four mice each age group) for the age groups of P6–P7, P10–P14, and P21–P25 mice, respectively. In addition, we measured the CA3 length (from proximal to distal end) of the same age groups, with the average lengths being 1,057.6 ± 55 μm, 1,198 ± 52.8 μm, and 1,307.1 ± 39.8 μm (measured from two to three sections of each animal across two to three mice of each age group) for P7, P14–P15, and P25 mice, respectively. By comparison, the degree of CA3 developmental growth was larger than that of CA2. We also took advantage of brightfield images of living slice morphology to examine the CA2 development, and confirmed the CA2 regional differentiation during the postnatal development as seen in the PCP4 immunostaining. In acutely cut living slices, by P4 there was no differentiated region intermediate between CA3 and CA1 in gross morphology (A,B). Starting at P4–P5, a presumptive CA2 region appeared (C,D); by P6–P7 it was more distinguishable with differential gross cytoarchitecture of CA3 and CA1, and the detection of the mossy fiber tract (E,F). The CA2 region could be confidently identified by P14 and beyond (H–I). The overall CA subfield development and lamination patterns observed in the living slices of mouse developing hippocampus are consistent with previous studies using in situ hybridization of field-specific gene expression (Tole et al., ) and Timm’s silver staining (Zimmer and Haug, ). Electrophysiological recordings from the developing slices indicated that P1–P4 neurons exhibited immature spiking patterns (absent or broad poorly defined spikes) in response to depolarizing current injections, whereas cells beyond P4 exhibited more mature sharply defined spiking patterns. In addition, there was rapid morphological development of excitatory cells in distal CA3 and CA2 regions during the first postnatal week; excitatory pyramidal cells revealed by intracellular biocytin labeling had close to adult-like morphology with wide dendritic fields and extensive branches at or beyond P7 (A,B). Recordings from single cells in the emerging CA2 region revealed local excitatory circuit connections to CA2 excitatory cells at P7 (C–F). Although the total EPSC input strength for the cells of two age groups (P7 vs. P15–P18) were similar, the P7 cells had more input from distal CA3 than proximal CA3 compared with the CA2 cells at later developmental ages (c.f., and ), which suggests age-related developmental connections. The average strengths of summed EPSC inputs to P7 CA2 excitatory cells ( = 5 cells) were 28.8 ± 49.2 pA, 146.8 ± 157.9 pA, and 8.9 ± 15 pA for CA2, CA3a, and CA3b, respectively. u s i n g a s u i t e o f i m m u n o c h e m i c a l s t a i n i n g , a n a t o m i c a l , a n d p h y s i o l o g i c a l t e c h n i q u e s , w e s h o w t h a t i m m u n o s t a i n i n g f o r P C P 4 i s t h e b e s t , m o s t f u n c t i o n a l l y u s e f u l w a y t o l o c a l i z e t h e “ c l a s s i c a l l y d e f i n e d ” h i p p o c a m p a l C A 2 . O u r a n a l y s i s b r i d g e s t h e d i s c r e p a n c y o b s e r v e d b e t w e e n “ c l a s s i c a l l y d e f i n e d ” C A 2 a n d “ m o l e c u l a r l y d e f i n e d ” C A 2 . T h e C A 2 r e g i o n d e l i n e a t e d b y P C P 4 i m m u n o s t a i n i n g h a v e c e l l m o r p h o l o g y , p h y s i o l o g y , a n d s y n a p t i c c i r c u i t c o n n e c t i o n s t h a t a r e d i s t i n g u i s h a b l e f r o m d i s t a l C A 3 a n d p r o x i m a l C A 1 r e g i o n s . U n l i k e d i s t a l C A 3 c e l l s , o u r p h o t o s t i m u l a t i o n f u n c t i o n a l c i r c u i t m a p p i n g s u p p o r t s t h e i d e a t h a t C A 2 e x c i t a t o r y p y r a m i d a l c e l l s d o n o t r e c e i v e e x c i t a t o r y s y n a p t i c i n p u t f r o m d e n t a t e g r a n u l e c e l l s . O u r d a t a a l s o s u g g e s t p o s t n a t a l e m e r g e n c e o f a d i s t i n c t C A 2 r e g i o n , a s P C P 4 s t a i n i n g g r a d u a l l y e m e r g e s w i t h i n c r e a s i n g a g e , r e a c h i n g a d u l t - l i k e f o r m a t a g e P 2 1 o n w a r d .

Alzheimer's disease (AD), the most common dementia associated with an accumulation of amyloid-β plaques and tau tangles, affects over 35 million people worldwide and there currently is no cure . A major problem with treating AD is that by the time clinical symptoms (e.g., memory loss) appear, the disease is so advanced that reversing or slowing the process is largely ineffective . Medical imaging research suggests hippocampal atrophy and reduced brain perfusion , coupled with low brain metabolism , could be early biomarkers of AD. Amyloid-β plaques can be directly visualized with molecular imaging compounds such as Pittsburgh compound B (PIB) and 18F flutemetamol (flute) . However, these medical imaging techniques are not ideal for screening at-risk populations because they are expensive, radioactive, and limited mainly to research facilities. We have been actively investigating diffuse optical spectroscopy (DOS) as a non-ionizing, sensitive, and less expensive alternative to detecting AD biomarkers. DOS techniques probe thick living tissue with red and near-infrared light (600–1,000 nm) to measure the absorption (µ) and reduced scattering (µ′) coefficients of the tissue. The absorption spectra can be decomposed into functional parameters, such as oxy-hemoglobin (HbO), deoxy- hemoglobin (Hb), tissue oxygen saturation (O Sat), and water and lipid content . Reduced scattering coefficients are indicative of the tissue architecture, which may change due to cellular morphological changes or apoptosis . DOS instrumentation can range from fiber-based designs meant to interrogate human brain cortex through skin and skull , to non-contact, camera-based measurements that allow for a wider field-of-view with reduced penetration depth . Our strategy is to use mouse brain optical properties measured with a camera-based instrument to inform human brain measurements with the fiber-based DOS instruments. This study employs a camera-based DOS technique, spatial frequency domain imaging (SFDI), to characterize the , spatially resolved optical properties in mouse models of AD. In a previous study , we demonstrated extensive optical property differences in the brain between a triple transgenic mouse model of AD (3×Tg-AD) and age-matched controls. Average µ′ values measured over the 650–970 nm wavelength range were 13–26% higher than those in controls, and the heavy amyloid-β and tau build-up in the 3×Tg-AD mice was associated with a 27% reduction in total hemoglobin (Total Hb), as compared with controls . While the 3×Tg-AD mice suffer from microgliosis and astrocytosis , key components of AD that could contribute to changes in measured µ′, these mice do not develop the significant neuronal loss seen in human AD. Therefore, in this work we seek to better characterize the AD scattering biomarker in the CaM/Tet-DT mouse model , which allows controlled lesioning of forebrain neurons by removing doxycycline from the diet. All mice were bred and raised in the UC Irvine Institute for Memory Impairments and Neurological Disorders (UCI MIND). We imaged 3-month-old male CaM/Tet-DT mice , which harbor transgenes for the doxycyline-regulated neuronal expression of diphtheria toxin. When doxycycline is removed from the diet, CaM/Tet-DT mice develop progressive loss in forebrain neurons. CaM/Tet-DT mice ( = 5) were imaged longitudinally immediately prior to and after 23 days of lesion induction, and compared to images from Tet-DT controls ( = 5) for scattering and absorption (30 wavelengths, 650–970 nm). An additional five CaM/Tet-DT and nine Tet-DT controls were imaged only after 23 days of lesion induction to test the reliability of longitudinal imaging. Anesthesia was induced with isoflurane anesthesia (2% maintenance in 21% oxygen balanced by nitrogen, induction chamber). The head was secured in a stereotactic frame (Stoelting Co., Wood Dale, IL) equipped with a gas mask to enable repeatable positioning of the animal during imaging. The scalp was shaved and imaging windows were created by making a midline incision in the scalp and retracting the skin bilaterally to the temporalis muscle attachments. The skull was cleaned with a sterile cotton swab. Petroleum jelly (Vaseline, Unilever, London, UK) was used to create a well on the skull surface, and the well was filled with saline and covered with a glass coverslip. This approach prevents scattering artifacts from drying of the skull. During the surgery and the subsequent imaging sessions, the animal was kept at 37°C by a thermister-controlled heating pad (CWE, Inc., Ardmore, PA). Following surgery, isoflurane was turned down to 1% concentration and maintained at 1 L/minute. Excess gases were scavenged via a adsorber unit (A.M. Bickford, Inc., Wales Center, NY). After imaging, the scalps of animals undergoing longitudinal imaging were closed via suture and allowed to recover. An antibiotic, Baytril (10 mg/ml), was administered in the drinking water to prevent infection. All skin wounds healed within 2 days. For imaging on Day 23, the same procedures listed above were performed. All procedures were performed in accordance with the regulations of the Institutional Animal Care and Use Committee (IACUC) at the University of California, Irvine (protocol no. 2010–2934). A schematic of the experimental arrangement is illustrated in A. A complete description of SFDI instrumentation and data analysis has been previously presented in detail ,,. Briefly, a 250 W broadband near-infrared light source (Newport Corp., Irvine, CA) was directed onto a spatial light modulator (Texas Instruments, Dallas, TX) in order to create twodimensional sinusoidal patterns at two spatial frequencies (0, 0.125 mm) and phase-shifted 120° apart, resulting in six total projections (three at each frequency). These light intensity patterns were serially projected onto the imaging window of the animal and the remitted reflectance was imaged with the Nuance Multispectral Imaging System (CRI, Inc., Woburn, MA). A liquid-crystal tunable filter was used to image specific wavelengths in a serial fashion, resulting in images at 30 equally-spaced center wavelengths from 650 to 970 nm (10 nm bandwidth). After demodulating the three phase-shifted images at each spatial frequency, µ and µ′ maps were created at each wavelength by performing a linear least-squares fit of the diffuse reflectance from the two spatial frequencies to a Monte Carlo model of light transport in tissue . All image processing and analysis of SFDI data was done with MATLAB software (MathWorks, 2007b, Natick, MA). In each SFDI image, a region of interest (ROI) was selected between the suture junctions, bregma and lambda, and bilaterally to the temporalis muscle attachments (B). The average of pixel intensities in the ROI for each animal was calculated and was used for all subsequent analysis. We applied a Levenberg–Marquardt least-squares fit with the wavelength-dependent Beer–Lambert law, to decompose the absorption spectra to estimates of HbO, Hb, total hemoglobin (Total Hb = HbO + Hb), and tissue oxygen saturation (O sat = HbO/Total Hb × 100). All averages, standard deviation bars, and -values shown were calculated from mean ROI and standard deviation values between animals in each group. A two-tailed student's -test analysis was applied to the absorption and scattering data, to assess differences between CaM/Tet-DT and Tet-DT mice. To determine whether removal of doxycycline from the diet in the CaM/Tet-DT mice resulted in expected trends, immunohistochemical methods were used. We expected to observe a decrease in neuronal cells and an increase in inflammatory microglia and astrocytes. After the last imaging session for a given animal, the animal was euthanized, and the brain was preserved in 10% formalin, cryo-protected with 30% sucrose, and subsequently flash frozen. Coronal sections of 40 µm thickness were incubated overnight at 4°C with 1:1,000 dilutions of NeuN (EMD Millipore, Billerica, MA), a neuronal cell marker; Iba-1 (Wako Chemicals, Richmond, VA), a marker of microglia; and GFAP (EMD Millipore) primary antibodies, a marker of activated astrocytes. Washed sections were then incubated for 1 hour with the appropriate secondary fluorescent antibodies. An area of cortex corresponding to the ROI was imaged with a 10× objective on a Zeiss confocal microscope. After initial settings were determined for the best signal-to-noise imaging, all slices were imaged with the same settings at 1,024 × 1,024 pixel resolution. Further analysis included manual selection of a rectangular cortical ROI (average area = 130,000 square pixels) and calculation of the average optical density of staining, using ImageJ software . Four animals were used in each group for statistical analysis. Baseline reduced scattering and absorption coeffcients showed no significant difference between CaM/Tet-DT mice and Tet-DT controls (B,C, and A,B). Visual inspection of the reduced scattering maps showed a global scattering change in the lesioned mice (A). After 23 days of lesion induction, including both naïve and chronically imaged mice, brain cortical scattering was 11–16% higher in the CaM/Tet-DT mice as compared to controls ( < 0.005) across the 650–970 nm range (D). Longitudinal imaging, which involved exposing the skull and suturing the scalp together after the first day of imaging, was not associated with any significant difference in reduced scattering coefficient between the first and 23rd day of imaging in Tet-DT controls. Data in A suggests that removal of doxycycline from the diet causes a significant decrease in absorption for all wavelengths from 650–970 nm in the Tet-DT mice ( < 0.05). This lower absorption was associated with a difference in Total Hb (119 ± 21 µM vs. 91 ± 25 µM) ( < 0.05) in controls. This difference in Total Hb was not observed in CaM/Tet-DT mice (D). As expected, after 23 days of lesion induction, neuron loss, and infiltration of inflammatory cells were seen in the cortex of CaM/Tet-DT mice but not in Tet-DT controls (A). We quantified these structural changes in the cortex by comparing the average optical density of the staining in the cortex (B). Neuronal staining decreased 61%, microglial staining increased 97%, and activated astrocytes increased ∼100-fold in the lesioned animals. This is the first report, to our knowledge, of the increase in µ′ as a result of neuronal death and increased brain inflammation . Our previous work showed increased µ′ in the triple transgenic (3×Tg-AD) mouse model of AD . However, it was unclear if this increase was due to inflammation of the brain, misfolded protein build-up, or differences in skull thickness. In this study, we have controlled for differences and changes in skull thickness by comparing CaM/Tet-DT mice and Tet-DT controls at baseline and after lesion induction. . Interestingly, we observed a significant decrease in absorption due to reduced blood perfusion in the control mice after removing doxycycline from their diet for 23 days. This is a previously unknown phenotype of these mice, but it may be due to the role of doxycycline in vascular tone regulation . The reduction in absorption is not seen to the same extent in the CaM/Tet-DT mice, perhaps because of inflammation-induced vasodilation. These results underscore the importance of separating the effects of scattering from absorption. Unlike planar reflectance imaging, SFDI enables the mathematical separation of scattering from absorption. Without this additional information, the measured decrease in absorption combined with an increase in scattering would confound our interpretation of the underlying physiology. In conclusion, we have shown that µ′ increases in a model of brain injury that mimics the cellular and structural changes found in late human AD . Monitoring µ′ could be useful in evaluation of pre-clinical therapies in animals, as well as for tracking AD progression in humans. The ultimate goal, however, is to detect AD before neuronal loss occurs. AD is intimately associated with microglial activation, which has roles both in clearing amyloid-β and in secreting pro-inflammatory factors . In addition, inflammatory genes have been shown to be up-regulated early and chronically throughout the brain in AD . Further studies need to be done to determine if light scattering may be uniquely sensitive to the broad cellular changes seen in an inflammatory state before brain atrophy occurs. Previous studies have demonstrated the sensitivity of optical absorption to neurovascular dysfunction in AD . As a result, clinical DOS techniques that separate scattering from absorption could be useful as a point-of-care approach for detecting and characterizing AD.

Efficient cell delivery of bio-molecules, such as oligonucleotides and peptides, is a major hurdle in development of novel therapeutics. As a result, higher drug dosages are often required than would otherwise be needed, which increases costs and the possibility of off target effects. A promising method of enhancing cell delivery is by use of cell-penetrating peptides (CPPs) to facilitate cell-uptake. CPPs often contain positive charges and/or hydrophobic elements. Some are based on cell-permeable peptides obtained from larger proteins and are known as protein transduction domains (PTD), whereas others are synthetically derived. CPPs have been shown to deliver a range of different biologically active cargoes into cells and , including proteins, peptides, oligonucleotide analogues, siRNAs and small molecule drugs. In the case of oligonucleotide cargoes, CPPs can either be complexed with or conjugated covalently to the cargo. For example, CPP conjugates of charge neutral antisense phosphorodiamidate morpholino oligonucleotides (PMO) or peptide nucleic acids (PNA) have been shown to be very effective in induction of splicing redirection or exon skipping in cells and . However, there have been very many CPPs proposed, and individual research groups often utilise their own preferred peptide sequences. No single CPP has been found to be universally successful for the conjugate delivery of all cargoes into all cell types. Instead for a new drug target or application, success is often only achieved through painstaking conjugate synthesis on an individual basis and search for a suitable peptide sequence by trial and error using a cell or assay. Only a small number of CPPs (commonly 2 or 3 known CPPs) can usually be tested as bio-cargo conjugates because of the difficulty and the time-consuming nature of their syntheses, even though cell-screening reporter assays, luciferase-based, are frequently available and adaptable for high-throughput. Significant demand exists therefore for a method of peptide conjugate library synthesis that is quick and efficient and is suitable for subsequent use of rapid screening assays. Following cell assay of an initial library of conjugates using peptides of widely ranging sequence, fine-tuning of a peptide candidate can then be accomplished either by further narrower library synthesis and cell re-assay, or by more conventional synthesis if further screening is to be carried out by an assay. The most time-limiting step in synthesis of peptide-cargo conjugates is generally not the conjugation itself but the individual purifications required for peptide, cargo and conjugate. Recently, parallel multi-peptide synthesis machines have become available as well as new methods for rapid affinity-based purification of bio-molecules ( histidine tags or biotin-streptavidin) that provide an opportunity both for rapid synthesis and for reaction workup and purification. In addition, recent developments in bio-orthogonal “click”-type ligation reactions provide a chemical basis for efficient conjugation of bio-molecules in aqueous solution. These new methodologies have now inspired us to develop a new strategy for the SELection of PEPtide CONjugates (SELPEPCON) of bio-cargoes that can act as an initial parallel synthesis approach useful for therapeutic screening. SELPEPCON allows for convenient and rapid parallel conjugation reactions and workup that avoid the need for HPLC purification (). The method utilizes a functionalized cargo linked through a cleavable linker to an affinity tag. The functionalized cargo is conjugated to the peptide after which it is purified by immobilization and isolated by release of the tag. Affinity purification is very rapid and is readily automated, making the overall strategy very suitable for high-throughput synthesis of conjugates for screening. To demonstrate the utility of this SELPEPCON methodology, two case studies are shown. In the first example, a small conjugate library is synthesized to investigate the roles of individual or groups of amino acids in a pre-selected CPP attached to a cargo. The second shows a larger CPP screen of a variety of quite different peptide sequences to find CPP candidates suitable for delivery of a novel cargo into a cell. Both studies utilize CPP conjugates of a PNA705 cargo, a well-known 16-mer splice-redirecting oligonucleotide analogue cargo that has been used many times for CPP development in the past. A HeLa pLuc705 cell assay was employed to assess the construct for their ability to enter cells. In this assay, PNA705, upon entering the cell, causes splicing redirection of an aberrant β-globin intron, which leads to the up-regulation of firefly luciferase. The assay is convenient, has a high dynamic range and is very suitable for testing of a large number of conjugates in parallel. Conjugations are carried out using copper catalyzed Huisgen “click” reactions between alkyne-functionalized CPPs and an azide-functionalized PNA705 containing a disulphide-linked biotin-tag. The disulphide-linked biotin tag allows for solid-phase immobilization purification of the resultant conjugates after which the conjugates can be isolated by reduction of the disulphide releasing it from the solid support. The results demonstrate how SELPEPCON can be utilized to find active CPPs for a cargo such as PNA705 in a rapid synthesis, isolation and screening procedure. The SELPEPCON method of synthesis of a bifunctionalized PNA705 cargo suitable for attachment of peptides and subsequent immobilization is shown in . Because a known side product of the click reaction is alkyne homocoupling, we decided to locate the azide functionality on the PNA cargo and the alkyne functionality on the peptide component. In this setup any alkyne-homocoupling side reaction will lead to a product that is easily removed by washing of the streptavidin solid support. Thus, 5-azidopentanoic acid was used for N-terminal modification of the PNA by manual solid phase amide coupling following machine-aided assembly of the PNA by solid-phase synthesis. The PNA contains a C-terminal -trityl-protected Cys residue, which was introduced at the first step of solid-phase synthesis. After deprotection and cleavage from the solid support, a biotin group was introduced on to the N-PNA705 by disulfide bridge formation reaction of the C-terminal Cys residue with commercially available -[6-(biotinamido)hexyl]-3′-(2′-pyridyldithio)propionamide (EZ-link™ HPDP Biotin). This straight forward reaction provided the N-PNA-S-S-biotin after HPLC purification in 80% yield. In order to demonstrate the methodology, a small test library of 16 peptides was synthesized on Tentagel using an Intavis Parallel Peptide Synthesizer (). The test library was narrowly based around a known relatively short (17 amino acids) CPP, Pip5h, which we have previously evaluated in exon skipping as a conjugate with a PMO targeting the pre-mRNA of dystrophin, but which was not chosen for further development. Peptide variants included modifications in the central 5-amino acid core () or on one of the Arg-rich flanking regions () of the sequences. In each case an N-terminal alkyne was introduced using 4-pentynoic acid through standard amide bond formation as the final coupling step (). The peptides were deprotected and cleaved from the solid support by standard TFA treatment for 4 h with gentle mixing and rapidly isolated by solid phase extraction (SPE) using Oasis HLB cartridges. Good yields were achieved using this synthesis and purification method (). MALDI-TOF mass-spectrometry of the products showed in all cases a major signal corresponding to the expected peptide and only minor amounts of peptide impurities (ESI Table S1 and Fig. S1, S2 and S13, S14). None of the observed shorter peptide impurities should form conjugates in the subsequent conjugation step because the alkyne linker is introduced in the final step of the synthesis and truncated peptides are capped after each coupling step during solid phase synthesis. Thus, time-consuming HPLC purification of each synthesized peptide is avoided. The peptide purification method used is fast, such that cleavage, deprotection and purification of 16 peptides can be carried out within one day, thus allowing for easy parallelization. Peptide-PNA conjugates were synthesized, purified and isolated in parallel using the procedure shown in . Copper() sulphate in combination with sodium ascorbate proved to be the most promising catalyst system. Procedures involving copper() salts, such as copper() bromide or copper() iodide, were also tested but provided less reproducible conversions. Diisopropylethylamine was used as base to quench TFA salts of the cargo and peptides, and 2,6-lutidine and tris(benzyltriazolylmethyl)amine (TBTA) were utilized to stabilize the active copper() species. After 1 h, reactions were quenched by the addition of EDTA, diluted and immobilized on a streptavidin-functionalized Sepharose™-support held in SpinTrap™ centrifuge tubes. Following several washing steps, the conjugates were released from the support by addition of a TBS buffer solution containing 10 mM tris(2-carboxyethyl)phosphine (TCEP) followed by centrifugation, leaving the biotin tag on the support. The use of organic solvent ( 40% acetonitrile) for dilution of the reaction mixtures, washing and release proved to be essential for good recovery of the conjugates. The resultant conjugates still contained buffer and TCEP, which were removed by SPE using Oasis HLB cartridges. The conjugates were obtained in good isolated yields (50–70%, ). In all cases, analysis by MALDI-TOF mass-spectrometry showed the expected signal for the conjugation product (ESI Table S1 and Fig. S15–S17). In addition, a signal corresponding to the unconjugated PNA705 was observed. Quantitative analysis by RP-HPLC revealed <10% of the unconjugated cargo. As the cargo by itself is unable to enter cells efficiently and has been demonstrated to be inactive in the HeLa pLuc705 cell splicing redirection assay (data not shown), the conjugates containing small amounts of unconjugated PNA705 were used without further purification. The PNA705 conjugates of were assessed for their ability to enter HeLa pLuc705 cells and redirect splicing, resulting in up-regulation of luciferase expression. Attempts were made to use 96-well plates, instead of the commonly used 24-well plates, to allow for faster screening. However, the obtained results were hard to interpret because of large inconsistencies in luminescence values. More reliable and reproducible results were obtained with 48-well plates, which were used for all screening assays shown in this work. Several conjugates were purified by RP-HPLC and assayed, which provided similar results compared to the crude conjugates obtained by SELPEPCON (ESI Fig. S30). The full –PNA705 conjugate library was assayed at a single-concentration (5 μM) and the results analysed as the fold increase in luminescence compared to buffer only (). Several of the –PNA705 constructs were strongly active, as was expected for Arg-rich peptides. Substantial reduction in the number of Arg residues ( multiple replacements by His, Ser or Glu) led to significantly lower activity (. When the Arg residues were replaced by Lys, retaining the overall positive charge, some splicing redirection activity was retained ( and ). –PNA705 conjugates differ slightly from CPP–PNA constructs previously synthesized by us. Thus, –PNA conjugates contain an additional free Cys residue and the peptide to PNA conjugation is N- to N-terminus rather than C- to N-terminal conjugation used in previous work. However, containing the original Pip5h sequence showed similar splicing redirection activity to that of previously synthesised Pip5h-PNA705 (data not shown). Amongst the more active conjugates, variations in splicing redirection activity observed in the HeLa cells were not identical to those that might have been expected based on knowledge of exon skipping effects in mouse muscle cells for which the Pip5 and the later Pip6 series were designed. Scrambling of the hydrophobic core (), replacement by Ala residues (), substitution of Leu by Pro () or a negatively charged Glu () all appeared to be detrimental to the activity (). Good splicing redirection activity was retained by reversal of the core sequence () or by replacement of a Leu by a Trp (). The latter gives an IWFQ sequence also found in Penetratin, from which the hydrophobic core sequence of the Pip series was originally derived. Interestingly, the replacement of same Leu by a Lys residue significantly increased activity (). The results of CPP–PNA705 conjugate library based on demonstrate how SELPEPCON can be effective for study and optimization of the amino acid sequence of a CPP for delivery of a cargo into a particular cell system providing that a cellular assay is available. We now wished to apply the SELPEPCON methodology to a significantly larger library in order to demonstrate its potential for selection of a CPP from a large range of sequence-dissimilar peptides, which would be necessary in the case of a new cargo candidate where relatively little was known about likely requirements for intracellular activity. Thus we designed peptide library consisting of 78 peptides that included many well-known CPPs, some sequences obtained from searching scientific literature, and some newly designed (Table S2, ESI). The peptides in this library consisted of both hydrophobic and hydrophilic peptides, ranged in length from 6–28 amino acids and carried a net charge ranging from –1 to +13. The library consisted of standard -amino acids with the addition of β-alanine (B) and ε-aminocaproic acid (X), which are commonly found in CPPs. Since it was found that free thiol groups inhibited the copper-catalysed cycloaddition reaction, Cys residues could not be introduced using Cys(Trt), which is the standard protecting group used in peptide synthesis. Instead an alternative procedure was developed using -butylthio (StBu) protected -cysteine, which contains a disulfide-protecting group that is stable to TFA cleavage but which is released in the reducing step of the conjugate workup procedure. The library of CPPs () was synthesised in parallel using the procedure described above for . Only standard synthesis protocols were used for the parallel synthesizer and a single method for purification and isolation was employed without any peptide specific optimization. Using these standard conditions, all but 5 of the 78 peptides (>90%) were successfully obtained, where the expected masses were seen as main signals in mass spectrometry (ESI Tables S2, S3 and Fig. S3–S12). Because of the type of work-up procedure, small amounts of impurities were seen in most of the mass-spectra obtained for the successfully synthesised peptides. In 4 peptide syntheses a significant amount of shorter acetyl-capped peptide sequences was observed, which is the result of incomplete coupling reactions (Fig. S11 and S12). Overall yields for the 73 successful syntheses were good (±70% average) and provided plenty of peptide material in principle for many different conjugation reactions on nanomole scale for this CPP library. The peptides were conjugated to PNA705 using the procedure described previously for conjugations. Most conjugates were obtained in good yield (>60% average, ESI Tables S2 and S3). As in the case of , a small amount of unconjugated PNA was observed by MALDI-TOF mass-spectroscopy and HPLC for all conjugates obtained (ESI Fig. S18–S29). Out of the 73 peptides where there was a peptide product available for conjugation, 10 conjugates were either not obtained or showed only small amounts of conjugate product compared to unconjugated PNA following affinity purification and desalting. The failed conjugations were practically all for the longer peptides that are likely to be more prone to form secondary structures, which may sterically inhibit the conjugation reactions. In the cases of all 4 peptides () that showed significant amounts of acetyl-capped peptide impurities, good quality conjugate products were nevertheless obtained. This demonstrates the effectiveness of the introduction of the alkyne linker in the final step of assembly in combination with acetic anhydride capping after each synthesis cycle during peptide synthesis, which allows for the removal of these peptide-impurities during affinity purification of the conjugates. The four –PNA705 conjugates that contained Cys residues (see Table S3, ESI) showed by mass-spectrometry impurities that resulted from a statistical mixture of StBu modifications. Some modification was even observed of the Cys residue that remains attached to PNA705 after release from the affinity support, which was evidenced by the fact that conjugate masses were found corresponding to two StBu groups for conjugates that contained only one Cys residue in the peptide part. This may be because residual free StBu that is not effectively removed by the Oasis HLB cartridge workup of the conjugates can slowly allow StBu disulfides to form with available Cys residues in the conjugates. The StBu disulfide protecting groups are clearly cleaved off during the reduction step in SELPEPCON, since the alternative use of centrifugal filter units (Amicon Ultra 3 K 0.5 mL, Millipore Ireland Ltd) to remove salts, TCEP reducing agent and StBu did provide the main product conjugates without StBu modifications. However, this procedure resulted in low yields, especially for hydrophobic conjugates. The –PNA705 conjugate library was screened at 5 μM concentration for their ability to enter HeLa pLuc705 cells and redirect splicing, resulting in up-regulation of luciferase expression (). A significant number of active conjugates were identified from this initial screen, showing 20- to 200-fold increases in luminescence values. In some cases, high error-bars were observed for conjugates that showed over a 20-fold increase, because conjugates could not all be tested in a single 48-well plate experiment and samples with high luminescence readouts show large variability between different experiments. However, the main purpose of the luciferase readout is to distinguish active from inactive conjugates. As a check on the reliability of purifications of conjugates several conjugates were purified by HPLC and tested for their ability to up-regulate luciferase expression (). None of the tested low-activity crude conjugates showed significantly higher activity after purification, revealing good reliability of the crude conjugates in this assay. The mildly active PNA705 conjugate of does show improved activity after purification. This can be explained because the crude conjugates contain a small amount (±10%) of unconjugated PNA, lowering the effective concentration of the –PNA705. This highlights once again that the assay results using this crude library should only be used to find suitable active CPP-candidates for further investigation and active conjugates should not be directly compared. Some interesting observations can be inferred from the results of the luciferase expression readouts. First, strong splicing redirection for the PNA705 cargo in this HeLa-cell assay is clearly directly related to the numbers of positive charges, particularly Arg residues, in the peptide part of the conjugates ( and ). A charge of at least +8 and at least 7–8 arginines are revealed as good CPPs properties. Association between the number of Arg residues and cell penetrating ability for CPP conjugates of PNA and PMO cargoes is well known from previous literature, but never before has such a large number and variety of CPPs of varying Arg content been tested in parallel. Interestingly several standard CPP sequences previously proposed for conjugation to PNA/PMO cargoes were found to be ineffective in this assay (for example, two containing Tat sequences and , and two containing Penetratin sequences and ). These results show that effective cell penetration is cargo and cell dependent and demonstrates how SELPEPCON can be a valuable tool in the selection of a good CPP candidate for a particular application. An additional experiment was carried out in which all conjugates were evaluated in a single point assay to demonstrate how a first round of selection can be performed after a minimal screening effort (ESI Fig. S31). Although several active conjugates give very high activity spikes, the same trend for active conjugates is observed as for the multiple data-point screening shown in . A follow-up to screening can provide further information on selected candidate conjugates. For example, initial “hits” can be confirmed by evaluation of splicing redirection at the RNA level. This assay is more reliable compared to the luciferase readout but is less suitable in a high-throughput mode because of the need for gel electrophoresis to separate and quantify DNA strands. An example is shown in in which several –PNA705 conjugates were assessed by extraction of RNA from the treated HeLa pLuc705 cells followed by RT-PCR. Previously reported Pip-1-PNA, prepared by conventional conjugation through a C- to N-terminal disulfide linkage, was used as a positive control. As expected, the splice corrected RNA levels were low when cells were exposed to –PNA705, which also shows very low activity in the luciferase readout. contains the same Pip1 sequence as the positive control. The PNA705 conjugate was N- to N-terminally linked as obtained through SELPEPCON and showed high splice correction activity at both 2.5 and 5 μM. At 5 μM fully spliced RNA levels are high for the PNA705 conjugates of –PNA705, and , while showed significantly lower luciferase activity. The saturation of the splicing redirection observed at 5 μM can explain the large variation in the luciferase activity assay. At 2.5 μM these conjugates show the same activity trend by RT-PCR as seen in the luciferase readout obtained at 5 μM, highlighting the difference in sensitivity between the two methods. The SELPEPCON methodology described above has been exemplified by use of libraries of CPPs conjugated to a PNA705 cargo where a reliable cell assay is available to gauge the effectiveness of the peptides in delivering the cargo into the nucleus of HeLa cells. Clearly the methodology is in principle applicable to any bio-cargo capable of being functionalized by an azido group, such as PMO or other oligonucleotide types, siRNA, peptides or other biomolecules where there is a need to search for peptides that when conjugated to a cargo may enhance its delivery into cells or for cell or tissue targeting. For example, we are currently applying the SELPEPCON technology to cell delivery of a potential anti-cancer peptide known to poorly enter cells itself. The only limitation is that the constructs need to be compatible with the conjugation step. For example, a conjugation can fail for longer peptide sequences that contain strong secondary structures, as shown for several peptides in the library. This might be overcome by further optimizing the conjugation conditions, such as use of a denaturing agent in the conjugation step. We chose to conjugate the N-terminus of peptides to the N-terminus of the cargo, since N-terminal peptide functionalization is technically easier and in addition capping of unreacted chains during synthesis reduces the possibility of shorter chain peptide impurities becoming conjugated to the cargo. Thus, 4 less pure peptides could nevertheless be conjugated to the PNA cargo and still lead to an acceptable conjugate product. However, the method is not limited to N-terminal conjugation a N-terminal alkyne. To demonstrate this we synthesized two peptides based on containing a C-terminal alkyne ( and , Tables S4 and S5). The alkyne was placed on the C-terminus by using -Fmoc--bishomopropargylglycine-OH (Fmoc-Bpg-OH, Chiralix, Nijmegen, The Netherlands) during automated synthesis. Both peptides were obtained in good yield. The double glycine spacer of did not appear to be necessary to improve the conversion of the conjugation reaction, as both conjugates were readily obtained in good yield and purity. Note that it should also be possible to achieve conjugation at the C-terminus of PNA by use of an azido side-chain derivative of an Fmoc amino acid (such as Fmoc--azido-lysine) at the C-terminus and a Cys attached at the N-terminus of the PNA. Following initial selection and verification of hits by the SELPEPCON methodology using a wide CPP library, such as described above, lead optimization would likely involve further rounds of SELPEPCON screening using narrower peptide libraries, as exemplified in . Because the peptides in are relatively similar, conjugation efficiency and thus effective concentration and purity of the conjugates obtained by SELPEPCON will be less variable compared to a broad library such as . The resultant libraries are also suitable for other screening purposes such as comparisons of serum stabilities of the conjugates, if studies are to be contemplated. Once one or more candidate peptides have been chosen for the desired cargo, thereafter drug development might consist of investigation of alternative conjugation methods to vary the type of linkage between peptide and cargo, as well in some cases their respective orientations to each other. Commonly such studies are best carried out in an model of a disease state using conjugates prepared more conventionally on larger scale, since pharmacological properties of conjugates may vary substantially depending on cargo type as well as on peptide sequence. It is possible that SELPEPCON could be used for preparing conjugates for assay by intramuscular injection ( into a mouse model of Duchenne muscular dystrophy) where only μg quantities are needed. However, use of SELPEPCON should reduce considerably the time and effort required to obtain potential leads that are demonstrated to be capable of the required biological effect in a cell model, hitherto a major bottleneck. MALDI-TOF mass spectrometry was carried out using a Voyager DE Pro BioSpectrometry workstation. A stock solution of 10 mg mL of α-cyano-4-hydroxycinnamic acid in 60% acetonitrile in water was used as matrix. Error bars are ±0.1%. Reversed phase HPLC purifications and analyses were carried out using a Varian 940-LC. A Nanodrop 2000 UV analyser (Thermo Scientific) was used for the quantification of PNA concentrations using the absorption at 260 nm with the following values to calculate absorption coefficients; C 6.6 cm M, T 8.6 cm M, A 13.7 cm M, G 11.7 cm M. TBTA (tris-(benzyltriazolylmethyl)amine) was synthesised according to a literature procedure. The synthesis of the Pip-1-PNA705 conjugate had been reported previously. The PNA705 sequence containing a C-terminal Cys and two flanking Lys residues (Lys-CCTCTTACCTCAGTTACA-Lys-Cys) was synthesised on a 50 μmol scale using a modified Liberty Peptide Synthesizer (CEM) according to a published procedure using a Chem-Matrix solid support and Fmoc amino acid monomers (Novabiochem) or Fmoc (Bhoc) PNA monomers (PolyOrg Inc.). After the final deprotection, the support was removed from the synthesizer and 5-azidopentanoic acid was manually double coupled using a standard PyBop/NMM coupling reaction at room temperature for 2 × 15 min. The PNA was cleaved from the support and deprotected using TFA (1.5 mL) containing 10% triisopropylsilane/2.5% water and 1% phenol. The mixture was agitated for 2 h after which it was filtered and concentrated. The crude PNA was isolated by cold diethyl ether precipitation, dissolved in water, filtered and purified by HPLC. A Phenomenex Prep C18 Jupiter column (250 × 21.2 mm, 10 micron) was used with the following gradient (A: 0.1% TFA, B: 90% acetonitrile, 0.1% TFA) 0–2 min 7.5% B 2–20 min 7.5%–25% B 20–25 min 25%–90% B (retention time: 20.4 min). The fractions containing the desired PNA were combined and freeze-dried giving a fluffy white solid (yield 10–40% based on support loading). Mass, expected: / 5250.0 found: / 5256.9. To a solution of 4 μmol PNA705 in 3 mL water, was added 8 μmol EZ-link HPDP-Biotin (Thermo Scientific) from a 3.7 μmol mL stock in DMSO. To this solution 600 μl 2 M sodium acetate pH 7 was added and the resulting solution shaken for two hours. The reaction was quenched by the addition of 15 mL 0.1% TFA and 150 μl TFA in order to solubilise the precipitate formed. This solution was filtered and then purified by HPLC using a Phenomenex Prep C18 Jupiter reversed phase column (250 × 21.20 mm, 10 micron) with a the following gradient (A: 0.1% TFA, B: 90% MeCN 0.1% TFA) 0–2 min 12% B 2–25 min 12%–25% B 25–30 min 25%–90% (retention time: 25.0 min). The purified N-PNA-705-S-S-biotin was obtained as a fluffy white powder in a typical yield of 60–80%. Mass, expected: / 5678.4 found: / 5687.1. Peptide library synthesis was carried out on a 5 μmol scale using an Intavis Parallel Peptide Synthesizer, applying standard Fmoc chemistry and following manufacturer's recommendations. The solid support was as supplied by Intavis (Tentagel, 0.2 mmol g). Double coupling steps were used with a PyBop/NMM coupling mixture followed by acetic anhydride capping after each coupling step. Terminal alkyne functionalization involved a standard coupling procedure using 4-pentynoic acid in the final step of the synthesis. The peptides were cleaved from the support and deprotected by addition of TFA (1.5 mL) containing 5% triisopropylsilane/2.5% water and 1% phenol with shaking for 4 h. The support suspension was then concentrated to a volume of ±500 μl using a flow of nitrogen and diluted with 5 mL water. After mixing, the resulting mixture was loaded on a 20 cc Oasis HLB cartridge (Waters), which was previously washed with acetonitrile (10 mL) and equilibrated with 0.1% TFA (3 × 10 mL). After loading, the cartridge was washed with 0.1% TFA (2 × 10 mL) and 5% acetonitrile in 0.1% TFA (2 × 10 mL). The peptide was eluted with 40% acetonitrile (10 mL). The solution obtained was freeze-dried and the yield was calculated using the weight obtained (), corrected for the amount of TFA salts based on the number of positive charges present in the peptide. For mass spectroscopy data see Tables S1 and S2. A mixture was prepared containing 30 nmol PNA from a stock solution in water (±2 mM), 150 nmol peptide from a stock solution in NMP (±10 mM), 0.2 μl 2,6-lutidine (1.7 μmol) and 1 μl diisopropylethylamine (5.7 μmol). 7.5 μl of 20 mM CuSO–TBTA solution (150 nmol) premixed in a 1 : 1 mixture of DMSO/water and 10 μl of a 20 mM solution of sodium ascorbate (200 nmol) were added to this mixture. The mixture was left for one hour and quenched with 100 μl 0.2 M EDTA 40% acetonitrile and 800 μl TBS 40% acetonitrile. The solution obtained was vortexed and loaded in two batches of 500 μl on a streptavidin HP SpinTrap (GE Healthcare UK Limited) and each batch was incubated for 20 minutes while being mixed by inversion. The column was washed with 400 μl 0.1 M EDTA in TBS 40% acetonitrile and with 5 × 400 μl TBS 40% MeCN. The conjugate was released from the resin by 2 × 20 minutes reactions with 2 × 400 μl 10 mM TCEP in TBS 40% acetonitrile and the solid support washed with 200 μl 40% acetonitrile in TBS. The solutions collected were combined and freeze-dried. The solid was dissolved in 500 μl 20% acetonitrile and loaded on an equilibrated 1 cc Oasis HLB cartridge (Waters) together with 500 μl 0.1% TFA. The column was washed with 3 × 1 mL 0.1% TFA, 3 × 1 mL 5% acetonitrile 0.1% TFA and 1 × 1 mL 10% acetonitrile 0.1% TFA. These extensive washes are required to remove the TCEP from the cartridge. The conjugates are released using 500 μl 60% acetonitrile and diluted with 500 μl 0.1% TFA. The resulting solution was freeze-dried and dissolved in 500 μl water. The concentration was determined and the solution was stored in the freezer for use in cellular assays. The splicing redirection assay was carried out similarly to a previously reported procedure with minor changes. The assay was performed using 48-well plates with cells seeded the previous day (7.5 × 10 cells per well) and a total volume 100 μl per well for the conjugates’ solutions in OptiMEM. Note that we found that 96-well plates tended to give too high a scatter probably because of the smaller volumes in pipetting. RT PCR experiments were carried out as reported in previous publications. All experiments were performed in triplicates, unless specifically stated as designed one-point experiments. #text

Published data indicate that CD80 and CD35 can be expressed by the antigen specific B cells. In this study, we characterized the CD19 (a B cell marker) CD80 B cells and CD35 B cells in the intestine of a food allergy mouse model. The mice were sensitized to ovalbumin (OVA) and then treated with or without SIT. The mice were sacrificed one week after the last treatment. The lamina propria mononuclear cells (LPMC) were isolated and analyzed by flow cytometry. The frequency of CD19 B cells was markedly increased in sensitized mice than naïve control mice (). We then isolated the CD19 B cells from the intestine by magnetic cell sorting (MACS), labeled with carboxyfluorescein succinimidyl ester (CFSE) and cultured in the presence of the specific Ag (OVA) for 3 days. The CD19 B cells from the sensitized mice, treated with either saline or SIT, proliferated markedly while those from naïve mice did not (). Further analysis showed that the phenotypes of the proliferating CD19 B cells were CD80 CD35 (41.2%), CD80 CD35 (19.7%) and CD80 CD35 (21.0%) in sensitized mice. After treating with SIT, the phenotypes of B cells were altered dramatically to be CD80 CD35 (10.8%), CD80 CD35 (16.4%) and CD80 CD35 (49.2%) (). The results indicate that the Ag-specific B cells have developed in the intestine after the sensitization, which can be activated upon re-exposure to specific Ag. Treating with SIT can alter the phenotypes of antigen specific B cells. TSP1 is an immune regulatory molecule, which may contribute to the development of immune tolerance. Since one of the mechanisms by which SIT-induced immune tolerance contributes to suppression of the ongoing allergic inflammation, we hypothesize that SIT increases the production of TSP1 in the antigen specific B cells. We then assessed the expression of TSP1 in the B cells by flow cytometry. The results showed that TSP1 was detected in 4.27% CD35 CD80 B cells, 26.1% CD80 CD35 B cells and 89.7% CD80 CD35 B cells (), which was not only observed in the intestine, but in the mesentery lymph nodes and the spleen as well (). The expression of TSP1 by the B cell subtypes was further confirmed by the assessment of qRT-PCR (). To elucidate if the TSP1 can be released to the microenvironment, we measured the levels of TSP1 in the culture medium, in which CD19 CD35 B cells (isolated from the OVA-tolerant mouse intestine; the mice were fed with OVA at 5 mg/mouse daily for 7 days) were cultured in the presence of OVA or BSA for 3 days. As shown by ELISA data, the levels of TSP1 in the supernatant were detectable in the saline group, which were markedly increased when the cells stimulated by OVA, but not in those stimulated by BSA (). The results indicate that exposure to specific antigen can induce TSP1 release from CD35 B cells. In addition, treating mice with a specific antigen, OVA, daily for 7 days also increased the CD35 TSP1 B cells in the intestine (). Since TSP1 has the immune regulatory property, the CD35 TSP1 B cells may be a subset of tolerogenic B cells, which was further explored as presented below. The amount of costimulatory molecules on the DC surface is one of the indicators of the tolerogenicity of DCs. The fact that TSP1 can cleave cell surface proteins implies that the B cell-derived TSP1 may be a regulator of the amount of costimulatory molecules on the surface of DCs. To test the hypothesis, CD19 B cells were isolated from the intestine of OVA-sensitized mice with or without treating with SIT and isolated to be CD35 and CD35 B cells respectively (the TSP1 cells were 89.7% of the CD35 B cells as checked by flow cytometry). DCs were isolated from the naïve mouse spleen and pulsed by lipopolysaccharide (LPS; 100 ng/ml, overnight) in the culture to boost the expression of the costimulatory molecules, CD80 and CD86. CD35 B cells (or CD35 B cells) and DCs were co-cultured overnight in the presence or absence of the specific antigen (OVA). The results showed that the exposure to the activated CD35 B cells significantly reduced the frequency of CD80 DCs (). The exposure to BSA, an irrelevant antigen, did not alter the amount of costimulatory molecules from DCs (). To elucidate the role of TSP1 in the reduction of CD80 from DCs, the TSP1 inhibitor, peptide LSKL, was added to the culture, which blocked the reduction of CD80 in the DCs (Fig. A6–A7). The suppressor effect of CD35 B cells was also observed on CD86 in DCs (). In contrast to CD35 B cells, the CD35 B cells did not have any effect on the levels of CD80/CD86 in DCs (). The results were confirmed by Western blotting data () and by exposing the LPS-pulsed DCs to recombinant TSP1 in the culture (). To clarify if the effect of TSP1 on the suppression of CD80/CD86 in DCs was to suppress their gene expression, DCs were isolated by MACS and analyzed by qRT-PCR. However, the treatment did not disturb the gene expression of CD80/CD86 in DCs (). The data indicate that the CD35 B cell-derived TSP1 can reduce the amount of costimulatory molecules in DCs; but does not alter the gene transcription in the cells. The results implicate that CD35 B cells may contribute to maintaining the tolerogenic properties of DCs, in which TSP1 plays an important role. In addition, to understand if the cell-cell contact is required in the suppression of costimulatory molecules on DCs by the CD35 B cell-derived TSP1, in separate experiments, LPS-primed DCs and CD35 B cells were physically separated in Transwells. The results showed that the amount of costimulatory molecules was still suppressed (). The results indicate that the cell-cell contact is not required in the CD35 B cell-reducing CD80 amounts on DCs. On the other hand, the supernatant of CD35 B cells also showed a suppressive effect on the CD80 amounts on DCs (). Published data indicate that SIT can induce Tregs in sensitized subjects. In line with the reports, we also detected the expansion of Foxp3+ Tregs in the intestine of the sensitized mice treated with SIT but not in those treated with saline (). To clarify if TSP1 plays a role in the generation of Tregs, a group of mice was treated with TSP1 inhibitor during SIT, which abrogated the increase in Tregs in the intestine (). Furthermore, we treated mice with a neutralizing anti-CD35 antibody right before the commence of SIT to remove the CD35 cells from the mice (). Indeed, the increase in Tregs was abolished. To confirm the effect of TSP1 on the induction of Tregs, a group of sensitized mice was treated with recombinant TSP1 daily for 5 days. The frequency of Tregs was increased about 4.7 folds as compared to the controls (). We then isolated CD4 T cells from the intestine of mice sensitized to OVA. Abundant CD4 T cell proliferation () was observed in response to a specific antigen stimulation in the culture (), which was not apparent in the CD4 T cells isolated from mice treated with SIT (). The CD4 T cells from mice treated with TSP1 inhibitors () also showed marked proliferation in the culture after treated with the specific antigen, OVA. The removal of CD35 cells resulted in high proliferation of CD4 T cells. The results indicate that CD35 B cells play a critical role in the regulation of the properties of CD4 T cells in the allergic intestine by SIT. We also measured the levels of IL-4, IL-5 and IL-13 in the culture supernatants of . The results showed that, compared to the cells isolated from the sensitized mice (not treated by SIT), the CD4 T cells from SIT-treated mice produced much less Th2 cytokines (p < 0.01) in the culture in response to the stimulation of the specific antigen, OVA. The suppressor effect of SIT on Th2 cytokine production was abolished in the cells isolated from mice with CD35 cell depleted, or treated with TSP1 inhibitors (). Collectively, the results indicate that the TSP1-producing CD35 B cells can suppress the skewed Th2 response in the intestine of sensitized mice. We next observed if TSP1-producing CD35 B cells could induce Tregs directly. We isolated CD35 B cells from the OVA-tolerant mice, and Th0 cells from DO11.10 mice. The two cell populations were co-cultured for one week in the presence of the specific antigens. About 25.4% T cells were converted into the Foxp3 Tregs (; ). Because IL-10 or TGF-β from tolerogenic DCs plays the key role in the conversion of naïve T cells to Tregs, we added the neutralizing anti-IL-10 antibody or anti-TGF-β antibody to the above co-culture system. The results showed that the addition of anti-TGF-β, but not the anti-IL-10, abolished the generation of Tregs (; ). To elucidate if the B cell-derived TSP1 played any roles in the conversion of Tregs from Th0 cells, TSP1 inhibitor, the LSKL peptide, was added to the culture; the Treg generation was abolished (; ). The results indicate that the CD35 B cells produce TSP1 upon the exposure to the specific antigen; the TSP1 converts naïve Th0 cells to Tregs. We further co-cultured the CD35 B cells and Th0 cells in the presence of recombinant TSP1; the Th0 cells were converted to Tregs in a TSP1 dose-dependent manner (; ). To corroborate the finding, we isolated the Tregs from the co-cultured cells and analyzed the cell extracts by Western blotting. The results showed that the CD35 B cells or recombinant TSP1 markedly increased the TGF-β levels in the cells (). To further characterize the SIT-induced CD35 B cells, we developed an intestinal allergy mouse model. The mice were treated with SIT and with or without depletion of CD35 B cells are treated with TSP1 inhibitors. The effect of SIT was determined by assessing the response of the intestine to the challenge with the specific antigen, OVA, at a dose of 5 mg/mouse in gavage. The results showed that, after the challenge with OVA, the sensitized, saline-control mice showed an allergic attack manifesting high serum levels of OVA-specific IgE (), high levels of IL-4 () and IL-13 () in the tissue extracts of the intestine, a significant drop of the core temperature () and 100% mice had diarrhea (). The treatment with SIT suppressed the intestinal allergic responses, which was abolished by depletion of the CD35 cells, or using the TSP1 inhibitors during the SIT. Furthermore, the suppressor effect of SIT on the intestinal allergy phenomenon was mimicked by administration with recombinant TSP1, or adoptive transfer with the CD35 B cells (isolated from OVA-tolerant mice; ). The results emphasize the importance of CD35 B cells in the maintenance of the immune homeostasis in the intestine. On the other hand, we also observed the effect of SIT on the frequency of CD11b CD35 DCs, CD11c CD35 DCs and CD19 CD35 B cells in the intestine of mice sensitized to OVA. After treating with OVA or saline, LPMCs were isolated from the mouse small intestine and analyzed by flow cytometry. The results showed that much less CD35 cells were CD11b or CD11c as compared with the frequency of CD19 CD35 B cells in the mice treated with saline. After treating with SIT, markedly increase in the frequency of CD19 CD35 B cells was observed (p < 0.01), while the frequency of CD11b CD35 DCs and CD11c CD35 DCs was not changed much (p > 0.05) (). The present study revealed a novel subset of B cells, the antigen specific CD35 B cells, could be induced by SIT. The CD35 B cells expressed high levels of TSP1, the latter can down regulate surface domain of the costimulatory molecules, CD80 and CD86, of DCs in a cell co-culture model. Such an effect was reproduced by adding the recombinant TSP1 to the culture. The results also showed the tolerogenic function of CD35 B cells. This subset of B cells was capable of inducing Tregs and suppressing the ongoing allergic inflammation. The promotion of immune tolerance by SIT in the subjects with allergic disorders has been noted. Akdis et al. noted that SIT could up-regulate the expression of IL-10 and T cell anergy. The production of TGF-β was also observed after SIT; the resulted inhibitory effect on allergic reactions could be blocked by pretreatment with TGF-β antagonists. Thus, it seems that SIT can promote the expression of both IL-10 and TGF-β by immune cells under an allergic environment. Our data have expanded the knowledge in this area. By treating the sensitized mice with SIT, A novel subset of B cells, the CD35 B cells, was induced in mice with intestinal allergy. In line with the published data, such a subset of B cells could induce Treg development in the tissue with allergic inflammation and could suppress the ongoing allergic reactions. Further finding from this study pinpointed that TSP1 mediated the immune suppression of the CD35 B cells. Blocking TSP1 abolished the production of Foxp3 Tregs in the intestine of mice with intestinal allergy. The role of the costimulatory molecules in the T cell activation has been well described. CD80 and CD86 are the mostly investigated costimulatory molecules produced by DCs or B cells. An amplificatory role of CD80/CD86 in T cell activation is suggested based on the finding of blocking CD80/CD86 can dampen the interactions between DCs and T cells. The levels of CD80/CD86 on DCs are one of the major factors to dictate naïve T cells to differentiate into the inducible Tregs or other effector T cells. A number of factors, such as the stimulation of Toll-like receptors, are suggested in up-regulation of the expression of costimulatory molecules in DCs and so as to induce skewed immunity such as autoimmune diseases and other immune disorders. Thus, the control of the expression of costimulatory molecules in DCs may provide an important tool to modulate aberrant immune responses. The present data indicate that the antigen specific CD35 B cells are capable of modulating the levels of CD80/CD86 on the surface of DCs; the underlying mechanism is that upon the stimulation of the specific antigens, the CD35 B cells are activated to produce TSP1; the latter down regulate the levels of CD80/CD86 on surface domain of DCs. TSP1 does not affect the gene expression of CD80/CD86 manifesting the mRNA levels of CD80/CD86 were not altered in DCs after the treatment. The data showed that the TSP1 levels were increased in the supernatants of the co-culture of DCs and CD35 B cells indicating that the CD35 B cells can release TSP1 to the microenvironment and has the potential to interact with other immune cells. It is accepted that the major function of Tregs is to suppress other effector T cells' activities and to suppress or prevent immune inflammation. Various approaches have been tried to generate Tregs in experiments; SIT is one of them. The key point in the mechanism of Treg generation is the induction of Foxp3 expression. In line with previous studies, our data also show that SIT induces an increase in the Foxp3 Tregs in sensitized mice. Such an increase could be blocked by antagonizing the activities of TSP1, which also could be mimicked by the administration with the recombinant TSP1. The results indicate that the antigen specific CD35 B cells induce Tregs via producing TSP1 upon the exposure to specific antigens. The administration of the DC vaccines has been used in the studies of immune regulation. The immature DCs produce IL-10 or/and TGF-β, can induce Th0 cells to differentiate into Tregs. In line with the previous studies, the present study also revealed that the exposure to the specific antigens induced DCs to express TGF-β. Similar to previous reports, which described the critical role of TGF-β in the conversion of Treg, the present data also showed that DC-derived TGF-β is critical in the present experimental setting to convert the Th0 cells to Tregs. The novel aspect of the present study is that the CD35 B cell-derived TSP1 plays a critical role in the conversion of Tregs. This is supported by others' studies. Crawford found that TSP1 could convert the latent TGF-β to the active TGF-β; Chen et al demonstrated the role of TGF-β in the conversion of Tregs. Together with the results that the recombinant TSP1 also induces the conversion of Tregs from Th0 cells; it suggests a potential of the exogenous TSP1 in the treatment of immune disorders. The inference is supported by the animal model of intestinal allergy of the present study, which shows that the adoptive transfer with the TSP1-producing CD35 B cells can suppress the intestinal allergic inflammation. It is reported that several cell types can produce TSP1, such as endothelial cells, adipose cells, fiberocytes, macrophages, dendritic cells and B cells. The production of TSP1 by the cells needs proper stimulation. Our data demonstrate that SIT can drive antigen specific CD35 B cells to produce TSP1 in mice with intestinal allergy. The production of TSP1 by this subset of B cells is associated with the induction of Tregs in the intestine; the supportive evidence includes that blocking TSP1 could abolish the induction of Tregs in the intestine, which was also occurring in the depletion of the CD35 B cells. In separate experiments, we noted that none-specific activation of B cells by anti-IgM antibody also increased the expression of TSP1 (data not shown). Therefore, although there are a number of sources of TSP1 in the intestine, the proper stimulation is required for its production. In summary, the present data indicate that the treatment with SIT can induce the development of TSP1-producing CD35 B cells; this subset of B cells can down-regulate the costimulatory molecule levels in DCs, induce Tregs and inhibit food allergy associated inflammation in the intestine. Following the published procedures with minor modification, BALB/c mice were i.p. injected with OVA (1 mg/mouse) mixing with cholera toxin (2 μg/mouse) on day 0 and boosted on day 3. In addition, mice were gavage-fed with OVA (1 mg/mouse) or/and cholera toxin (CT; 10 μg/mouse; using as an adjuvant) weekly for 6 weeks. The negative control mice were gavage-fed with PBS only. The experimental procedures were approved by the Animal Ethic Committee at Shenzhen University. Following the published procedures with minor modification, the sensitized mice were treated with oral SIT. Briefly, ovalbumin (OVA) was gavage-fed with the doses of 1 mg (days 1 and 2), 5 mg (days 3 and 4), 10 mg (days 5–7), 25 mg (days 8 and 9), and 50 mg (days 10–14). Control mice were treated with bovine serum albumin (BSA). Peptides of LSKL, SLLK, recombinant TSP1, mutated TSP1 (mTSP1) were dissolved in saline (1.0 mg/ml), and i.p. injected at 30 mg/kg body weight. Mice were received neutralizing anti-CD35 antibody (0.25 mg/mouse; i.p.; one dose). The frequency of CD35 cells was not detected in the intestine after the treatment (). More experimental procedures were presented in . H . P . Z . , Y . W . , J . L . , J . J . , X . R . G . , G . Y . , L . M . , Z . Q . L . a n d Z . G . L . p e r f o r m e d e x p e r i m e n t s , a n a l y z e d d a t a a n d r e v i e w e d t h e m a n u s c r i p t . P . C . Y . d e s i g n e d t h e p r o j e c t , s u p e r v i s e d t h e e x p e r i m e n t s a n d w r o t e t h e m a n u s c r i p t .

After germination of a seed, the seedling enters the vegetative phase where rosette leaves are produced by the apical meristem in a spiral arrangement separated by short internodes. The plant then enters a transient phase, marked by a reprogramming of the apical meristem into an inflorescence meristem that produces spirally patterned cauline leaves, separated by long internodes. Entering the reproductive stage, the inflorescence meristem again reorganizes and starts producing floral meristems in the same spiral pattern, where each floral meristem develops into a single flower. Our current understanding of vegetative development and phase transitions is mainly based on different levels of transcriptionally based regulation, incorporating various roles of environmental triggers, chromatin and distribution of hormone gradients (for review see refs. –). An important level of protein regulation is, however, also mediated by UBIQUTIN-26S proteasome dependent pathways and an emerging perspective of both proteolytic and non-proteolytic roles of ubiquitination matches the importance of transcriptionally based regulation. We have previously demonstrated that the WD × R motif containing protein WDR55 plays a major role in reproductive development in . WDR55 is essential for gametogenesis and embryogenesis and is required to break radial symmetry and establish bilateral symmetry in the apical embryo domain. WDR55 belongs to the WD repeat (WDR) family, which is a large family of proteins with diverse functions but with a common conserved motif; the WD repeat. WDR55 also contains a WD × R motif, which is a submotif of the DWD box [DAMAGED DNA BINDING PROTEIN1 (DDB1)-binding WD-40 box]. The DWD box and the WD × R motif are signatures for potential substrate receptors of the CULLIN4 (CUL4) RING ubiquitin ligase complexes (CRL4s). Proteins with these motifs, which interact with CUL4-DDB1 based E3 ligases, are referred to as DCAF (DDB1-CUL4 ASSOCIATED FACTOR) proteins, and potentially target proteins for ubiquitination after they assemble with DDB1. WDR55 interacts with DDB1A, suggesting that a putative CUL4-DDB1 E3 complex regulates embryo or endosperm development. The CRL4s are associated with many processes such as cell cycle, DNA repair, photomorphogenesis and modulation of chromatin. Their mode of action appears to include both proteolytic and transcriptional mechanisms. In photomorphogenesis, CRL4 has been shown to act as a negative regulator of transcription in a CRL4 26S targeting complex. CRL4 has recently been shown to associate with the Polycomb Repressive Complex (PRC2-EMF) leading to PRC2 mediated H3K27me3 repression of FLC and MSI5 may also participate in this regulation. Moreover, CRL4 is required for maintaining PRC2-FIS mediated imprinting. We have generated a mutant that is homozygous for a weak allele of . Here we report that WDR55 is also required for proper vegetative growth and for organization of the adult plant body. seedlings display asymmetric and often multiple cotyledons and mature plants show phenotypes often associated with auxin misregulation, suggesting that WDR55 may play a role in hormonal control of plant development. Two mutant alleles of WDR55 have previously been identified to affect gametophyte development and function, as well as the establishment of apical symmetry in the embryo. The weaker allele, , initially displayed a close to mendelian ratio of mutant seeds (22.8%, n = 548), but no homozygous plants were observed among the germinating progeny. In order to search for a small fraction of homozygous plants (~2%) anticipated from the genetic data, a large number of seeds from heterozygous plants were germinated on MS-2 plates and observed over a longer period. Upon closer inspection, we identified a minor class of seedlings (3.6%, n = 1,035) that were late germinating, slow growing and considerably smaller than wild type seedlings. Using PCR genotyping and sequencing, we verified that this class was homozygous for . Using the same screening setup as above, we also inspected the stronger allele, but no small homozygous plants could be observed, much in agreement with the early embryo abort phenotype described previously. Transgenic plants with a full length genomic rescue construct restored wild type phenotype in both and alleles. Furthermore, in a complementation cross, the phenotype is partially rescued to phenocopy the embryo phenotype. The same partial rescue is also found in transgenic plants with rescue constructs containing the splice form controlled by its endogenous promoter. We therefore also inspected the progeny of crossed with on MS-2 media selecting for both mutant alleles, but were not able to identify any putative homozygous seedlings (n = 319). The identified homozygous mutant class appeared to be unique to the allele, which has a T-DNA insert in exon 13 (out of 14) and the protein thus may be able to create the proposed propeller structure characteristic for WDR proteins, but will lack the features of the C-terminal unstructured region. To investigate a putative role of WDR55 in the vegetative phase and whether the main requirement of WDR55 is restricted to gametophyte and seed development, we transferred the mutants to fresh S-2 agar plates to allow further growth. Generally, seedlings continued to develop, but compared with a wild type pattern their morphology and development largely diverged. Development of mature mutant plants was marked by a plethora of developmental defects in both roots and aerial structures, indicating that WDR55 is not only required in the reproductive phase, but represents a more general mechanism required throughout plant development. We therefore analyzed the vegetative phenotype. The (WiscDsLox430F06) line is a T-DNA insertional mutant previously characterized in Bjerkan et. al. Seeds were surface sterilized with ethanol, bleach and Tween20 before transfer to MS media supplemented with 2% sucrose (MS-2). Seeds were stratified on MS-2 plates at 4°C overnight before incubation at 18°C under long day conditions (16 h.) until germination. Seedlings were then transferred to soil and grown under the same conditions as above. Genomic DNA isolation was made using the SP Plant DNA kit (OMEGA bio-tek) according to the manufacturer’s instruction. Genotyping was performed using genomic and T-DNA insertion specific primers; ASP Wiscdslox 430F06 (5′-TTCCTCTTGCTGCCTGTAAAGC-3ʹ), SP Wiscdslox 430F06 (5ʹ-CAACAGAATCATACAGCCGATTGG-3ʹ) and p745 (5ʹ-AACGTCCGCAATGTGTTATTAAGTTGTC-3ʹ), as described previously. A Zeiss Axioplan Imaging2 microscope system equipped with Nomarski optics was used when examining tissue by light microscopy. Some sporophytic tissues were examined using a Leica WILD MZ8 stereo light microscope and images were captured using a Nikon Coolpix 995 mounted on the microscope. Histochemical GUS assays were performed by incubating sampled tissue in a substrate staining solution (0.05 M NaPO, 0.1% Triton X-100, 2 mM KFe(CN)3HO, 2 mM KFe(CN) and 2mM X-Gluc) overnight in the dark at 37°C. The substrate was then removed and the tissue was stored in 50% ethanol at 4°C. Slides were prepared by mounting the material directly in clearing solution (8:2:1, Chloral Hydrate:HO:Glycerol) and inspected after at least 1 h incubation at 4°C.

Plants can release organic compounds into the environment. These secondary metabolites may accumulate in the soil environment and influence the growth and development of neighboring plants, with positive and negative effects. This is called allelopathy, a natural ability of plants to protect themselves through natural allelochemicals. Allelochemicals typically inhibit seed germination and seedling growth. Moreover, they alter several physiological and biochemical processes including water utilization, mineral uptake, foliar expansion, photosynthesis, amino acid metabolism, protein synthesis, glycolysis, mitochondrial respiration and ATP synthesis, among others. Dopamine is one of these compounds; it is widespread in animals and has also been detected in many plant families. Velvetbean [ (L.) DC. var utilis] is widely used in tropical regions for intercropping with maize, sorghum and millet and for providing benefits, such as suppression of the nematode population, weed smothering, symbiotic nitrogen fixation, nutrient recycling and control of erosion. Many secondary compounds are produced by velvetbean. Using HPLC coupled with mass spectrometry, Wichers et al. identified dopamine in 2–3-week-old leaves of . The dopamine content of the leaves even exceeded the content of L-DOPA, the most abundant allelochemical in . However, in the roots, stems and seeds, no dopamine could be detected at any stage of development. Matsumoto reported that mucuna metabolizes L-DOPA to dopamine in leaves as a protective mechanism against the toxicity of L-DOPA. Dopamine has also been detected in many other plant families. Yue et al. demonstrated that noradrenaline and dopamine are major bioactive components of L., a traditional herbal medicine. Dopamine is found at high concentrations in potato () plants, the spathes of Araceae inflorescences, as well as the pulp of yellow banana (), red banana (), plantain () and fuerte avocado (). The role of dopamine in plants is poorly documented. It has been proposed to be a precursor for various alkaloids benzylisoquinolines like papaverine and morphine or of the hallucinogenic alkaloid mescaline. Some studies have addressed the effects of dopamine on plants and have revealed that it has attributes typical of an allelochemical. Dopamine is associated with defense against herbivores, processes such as nitrogen fixation, flowering and prevention of IAA oxidation, intercellular regulation of ion permeability and photophosphorylation of chloroplasts. Exogenous dopamine at concentrations of 5–100 μM stimulates ethylene biosynthesis in illuminated chloroplast lamellae from sugar beet leaves. Dopamine is formed in plants from tyrosine, either via tyramine or via L-DOPA. The physiological mechanism of action of dopamine in plants, however, is poorly understood. In animals, dopamine is a well-known neurotransmitter; its absence in nerve cells can cause Parkinson disease. One of the more important chemical changes is dopamine auto-oxidation or oxidation by enzymes, leading to melanins. During oxidation, both semiquinones and quinones are generated in a chain autoxidation process, which also results in the production of ROS like O, O, HO and HO. These ROS, as well as semiquinone and quinone products of catecholamine oxidation, can interact with proteins, lipids, nucleic acids and membrane components, thus causing cell damage. The phenylpropanoid pathway is one of the most important metabolic pathways since it synthesizes phenolic compounds and a wide range of secondary products in plants, including lignin. PAL is regarded as the primary enzyme of the phenylpropanoid biosynthetic pathway, wheareas POD within the cell wall, in either the free or bound state, has been shown to be associated with monolignol polymerization and, therefore, is important in lignin synthesis. To date, no reports on the effects and mode of action of exogenous dopamine on soybean roots are available. Due to the important role of lignification in root growth, the question addressed in the current research was whether dopamine affects PAL and cell wall-bound POD activities, as well as the lignin content of soybean roots. Root fresh and dry weights and root lengths decreased with increasing concentrations of dopamine in soybean seedlings grown during exposure (24 h) (). Mean total root lengths were 9.2, 29.5 and 48% less than the controls for the 0.25, 0.5 and 1.0 mM dopamine. Root fresh weight decreased by 8.8, 20, and 31.1% at the 0.25, 0.5 and 1.0 mM treatments, respectively, compared with the control. Similar behavior was observed in root dry weights, which were found to be 13.4, 19.8 and 37.6% less than those of the controls with 0.25, 0.5 and 1.0 mM dopamine. PAL activities were significantly distinct from the controls (). Dopamine decreased the activity of PAL by 12.8% to 34% with 0.25 to 1.0 mM dopamine. Cell wall-bound POD activities () decreased 16.3% indifferent of the dopamine concentration and no significant modify was detected in the lignin content (), compared with control. The data reveal that 0.25 to 1.0 mM dopamine activated SOD by 73.2% on average, in relation with the control (). Finally, it was found that the root cell was also affected significantly by exposure to dopamine. A loss of cell viability was increased by around 59.7% compared with the control (). The main outcome exposed in the present study is that root growth (), PAL (), cell wall-bound peroxidase activity () and cell viability () decreased after dopamine treatment, while SOD activity increased () and the lignin content () was not changed. The discovery that dopamine did not stimulate lignin production () is in agreement with the observed inhibition of PAL () and cell wall-bound peroxidase activities (). The reduction in seedling roots has been related with lignification of the cell walls and increases in PAL and POD activities and phenolic compounds. PAL is regarded as the access enzyme into the phenylpropanoid pathway and catalyzes the synthesis of phenolic compounds. In relation to cell wall-bound POD, there is long-standing evidence that this enzymatic form causes oxidative polymerization of monolignols from phenolic compounds, causing cell wall lignification and resulting in growth reduction. Although root growth of soybean was reduced by dopamine (), the enzymatic activities of PAL and cell wall-bound POD ( and ) also declined, with no increase in the production of lignin (). Thus, our results suggest that dopamine-induced inhibition in soybean roots does not appear to be related to the synthesis of lignin. Changes in the growth of roots by dopamine have been reported in a few plant species. Protacio et al. showed that catecholamines caused a stimulation of growth in root cultures of and cultures. According to the authors, dopamine affects plant development by acting with hormones, leading in elevated contents of auxin. It was shown that dopamine can inhibit IAA oxidation in vitro as well as in vivo via the inhibition of IAA oxidase. It is known that auxins promote the growth of stems and coleoptile and inhibit the growth of roots. It is likely that roots may require a minimum concentration of auxin to grow, but growth is strongly inhibited by concentrations of auxin required to promote elongation of stems and coleoptile. Thus, if dopamine actually inhibits IAA oxidase, thereby increasing the auxin content and high levels of this hormone in the roots inhibit growth, it is no exaggeration to suggest that this could be one of the modes of action of dopamine applied to the roots of soybean seedlings. Other studies have indicated that the dopamine toxicity may be related to its oxidation. This compound can be enzymatically or spontaneously metabolized by molecular oxygen in physiological solutions to form ROS, leading to the formation of melanins. Melanin biosynthesis is also linked to other dopamine oxidation substances, like quinones and semiquinones. Low concentrations of these ROS can stimulate several cellular processes and regulate several important physiological functions. Moreover, ROS, semiquinone and quinones can interact with proteins by denaturing them, thus causing injury to DNA and the cell membrane. To confirm whether melanin was generated from dopamine, soybean seedlings were developed with this compound (0.25–1.0 mM) in a nutrient solution. Our data (not shown) reveal that roots in contact to dopamine synthesized a significant quantity of melanin. As a method to indirectly assess possible increased levels of ROS like O, we determined the activity of SOD (). SOD enzymes are famous scavengers of O. Takano et al. observed the effects of dopamine on extracellular SOD expression in cultured rat cortical astrocytes. SOD was increased by 24 h of dopamine exposure in a dose-dependent manner. Dopamine has been shown to markedly reduce the neuroblastoma cells in the cerebrum of Parkinson disease patients and the production of dopamine semiquinone was observed in the brains of Parkinson disease patients by electron spin resonance spectrometry but semiquinone generation and toxicity were avoided by SOD, indicating the participation of superoxide radicals as the mechanism of toxicity. In our studies, SOD activities were significantly increased (), suggesting a possible increase in the levels of O in the roots of seedlings exposed to dopamine. Several studies have found that the cytotoxicity of dopamine, in animal cells, is dependent on ROS. Lai and Yu demonstrated a correlation between ROS that broke plasma membrane integrity and reduced growth or cell death. The data also demonstrated a considerable loss of cell viability, which was verified by superior Evans blue absorption. The data exposed could be the result of dopamine-induced oxidative stress, due to possible ROS produced during its decomposition and auto-oxidative transformation into melanin. We suggest that the decrease in cell viability () and reduced root growth () could also be related to increased levels of ROS in the roots, as a result of dopamine-induced oxidative stress due its conversion into melanin and due to its autoxidation. However, the possibility cannot be ruled out that there could be an increase in the level of auxin in roots treated with dopamine, contributing to the reduction in growth. Further studies are required to investigate dopamine metabolism in roots and to quantify auxin and ROS levels produced by this pathway. This is the purpose of a research currently in progress. Soybean [ (L.) Merr. BRS-232] seeds, surface-sterilized with 2% sodium hypochlorite for 5 min and rinsed extensively with deionized water, were dark-germinated (at 25°C) on three sheets of moistened filter paper. Twenty-five 3-d-old seedlings of uniform size were supported on an adjustable acrylic plate and transferred into a glass container (10 × 16 cm) filled with 200 ml of half-strength Hoagland solution (pH 6.0), without or with 0.25 to 1.0 mM dopamine. The container was kept in a growth chamber (25°C, 12L:12D photoperiod, irradiance of 280 μmol m s). Roots were measured at the start and at the end of experiments (24 h). Fresh root weight was determined immediately after incubation and dry weight estimated after oven-drying at 80°C, for 24 h. Dopamine was purchased from Sigma Chemical Co. and all other reagents used were of the purest grade available or chromatographic grade. After incubation, all treated or untreated seedling roots were detached and enzymes were extracted. PAL was extracted, as described by Ferrarese et al. Fresh roots (2 g) were ground at 4°C in 0.1 M sodium borate buffer (pH 8.8). Homogenates were centrifuged (2,200g, 15 min) and the supernatant was used as the enzyme preparation. The reaction mixture (100 μmoles sodium borate buffer, pH 8.7 and a suitable amount of enzyme extract in a final volume of 1.5 ml) was incubated at 40°C, for 5 min, for PAL activity assay. Fifteen μmoles of phenylalanine were added to start the reaction which was stopped after 1 h of incubation by the addition of 50 μl 5 M HCl. Samples were filtered through a 0.45 μm disposable syringe filter and analyzed (20 μl) with a Shimadzu Liquid Chromatograph equipped with a LC-10AD pump, a Rheodine injector, a SPD-10A UV detector, a CBM-101 Communications Bus Module and a class-CR10 workstation system. A reversed-phase Shimpack GLC-ODS (M) column (150 × 4.6 mm, 5 μm) was used at room temperature, with an equivalent pre-column (10 × 4.6 mm). The mobile phase was methanol:water (70%:30%) with a flow rate of 0.5 ml min. Absorption was measured at 275 nm. Data collection and integration were performed with class-CR10 software (Shimadzu). -Cinnamate, the product of PAL, was identified by comparing its retention time with that of standard’s. Parallel controls without -phenylalanine or with -cinnamate (added as internal standard in the reaction mixture) were made as described elsewhere. PAL activity was expressed as μmol -cinnamate h g of fresh weight. Cell wall-bound POD was extracted from fresh roots (0.5 g) with 67 mM phosphate buffer (5 ml, pH 7.0). Extract was centrifuged (2,200, 5 min, 4°C) and the pellet was washed with deionized water until no soluble POD activity was detected in the supernatant. Pellet was then incubated in 1 M NaCl (2 ml, 1 h, 4°C) and the homogenate was centrifuged (2,200, 5 min). The supernatant contained the cell wall (ionically) bound POD. Guaiacol-dependent activities of the cell wall-bound POD were determined according to Cakmak and Horst, with slight modifications. The reaction mixture (3 ml) contained 25 mM sodium phosphate buffer, pH 6.8, 2.58 mM guaiacol and 10 mM HO. Reaction started by adding the enzyme extract in phosphate buffer. Guaiacol oxidation was followed for 5 min, at 470 nm and enzyme activity was calculated from the extinction coefficient (25.5 mM cm) for tetraguaiacol. Blank consisted of a reaction mixture without enzyme extract whose absorbance was subtracted from the mixture with enzyme extract. POD activities were expressed as μmol tetraguaiacol min g fresh weight. SOD activity was assayed by using the photochemical nitroblue tetrazolium (NBT) method. Fresh roots (0.5 g) were ground in a mortar with 0.01 g of PVPP and 2 ml of 50 mM potassium phosphate buffer (pH 7.8) containing 1.0 mM EDTA. After centrifugation (2,200g, 20 min, 4°C), the supernatant was used for the determination of the enzyme activity. The reaction mixture (1.5 ml) contained 75 μM NBT, 4 μM riboflavin, 13 μM methionine, 50 mM phosphate buffer (pH 7.8) containing 1.0 mM EDTA and 20 μl of enzyme extract. To exclude eventual interference on SOD activity, parallel controls with dopamine added in the reaction mixture without enzyme preparation were undertaken under the same experimental conditions. The reaction was initiated by placing the reaction-mixture tubes under 15 W fluorescent lamps (56 μmol m s) for 10 min. The reaction was stopped by keeping the tubes in the dark for 10 min. The photoreduction of NBT (formation of purple formazan) was measured at 560 nm. One unit of SOD enzyme activity was defined as the amount of enzyme required to produce a 50% inhibition of the reduction of NBT. SOD activity is expressed as unit g fresh weight. About 20 mg of protein-free cell wall from different tissue was weight into a screw capped glass tube and added 0.5 ml of 25% acetyl bromide (v/v in glacial acetic acid) as described by Hatfield. The mixture in the tube was allowed to react for 30 min at 70°C. After cooling in ice bath, 0.9 ml of 2 M NaOH, 0.1 ml of 5 M hydroxylamine-HCl and 2 ml of iced acetic acid were added to the samples and centrifuged (1,400g, 5 min). Blank was run in conjunction with the samples and the UV absorption spectrum measured against the blank at 280 nm. Standard curve was generated by commercial alkaline lignin (Aldrich) to the same procedure. The extinction coefficient was ε = 16.4 cmmg mland lignin results were expressed as mg g cell wall. After incubation, the loss of cell viability in the seedlings was determined using the Evans blue staining spectrophotometric assay. All freshly harvested roots were incubated for 15 min with 30 ml of 0.25% Evans blue solution. Next, the roots were washed in distilled water for 30 min to remove excess and unbound dye and the excised root tips (3 cm) were soaked in 3 ml of N,N-dimethylformamide for 50 min at room temperature. The absorbance of released Evans blue solution was measured at 600 nm, using deionized water as a blank. The loss of cell viability is expressed as the absorbance at 600 nm of treated roots in relation to untreated roots (control). The experimental design was completely randomized and each plot was represented by one glass container with 25 seedlings. Data are expressed as mean of three to six independent experiments ± SE. Significance of differences was undertaken by one-way variance analysis with GraphPad Prism package (Version 2.0, GraphPad Software Inc.). Difference between parameters was evaluated by Dunnett’s multiple comparison test and p values < 0.05 were considered statistically significant.

The development of therapeutic cancer vaccines is based on the notion that the induction of immune responses against self antigens, particularly those either mutated or overexpressed in tumors vs. the corresponding normal tissues, may attenuate cancer growth and metastasis. In particular, since cytotoxic T lymphocytes (CTL) have been shown to play a crucial role in tumor surveillance, cancer vaccines are being designed to elicit strong and durable T-cell responses against selected tumor-associated antigens (TAAs). CTLs recognize short peptides derived from intracellularly synthesized proteins that are presented on the surface of target cells in association with Major Histocompatibility Complex class I (MHC-I) molecules. CTL epitopes have been identified in multiple clinically relevant TAAs defining the so called “cancer antigenome.” In order to break tolerance and potentiate immune responses against self-antigens, several efforts have been spent over the years to generate anchor-modified analog peptides. By virtue of their higher affinity for MHC class I, analogs are able to bind to the MHC complex with a longer half-life thus soliciting a more efficient priming of T cells which, once primed, are capable of subsequent recognition of the wild-type epitopes on the surface of target cells, including cancer cells. Electro-gene-transfer (EGT) of plasmid DNA in vivo is an efficient and safe methodology that results in greater DNA uptake per cell and enhanced protein expression. DNA-EGT results in long-term immune responses against target antigens in a variety of species and can be repeatedly administered to boost immune responses as required for the maintenance of antitumor immunity. In addition to increased gene expression, DNA-EGT is believed to enhance the immune response through stimulating local secretion of inflammatory chemokines and cytokines, the recruitment of antigen presenting cells (APCs) to the EGT site and by promoting the trafficking of APCs to the draining lymph nodes. Indeed, the addition of in vivo EGT has been associated consistently with an enhancement of cell-mediated and humoral immune responses in small and large animals, supporting its use in human clinical trials. Our group has shown that DNA-EGT is able to induce high levels of cell-mediated immunity (CMI) to a variety of TAAs in small and large animal species upon injection of plasmids expressing codon-usage optimized variants of full-length or truncated forms of TAAs, including those derived from carcinoembryonic antigen (CEA), the HER2/ oncogene, telomerase reverse transcriptase (hTERT) and matrix metalloproteinase 11 (MMP11). It has been surmised that immunizations with minigenes containing select, minimal T-cell epitopes may have several advantages as compared with full-length or even truncated proteins. A technical advantage of smaller minigenes is their compatibility with commonly used delivery agents, including plasmid DNA vectors. Also, full-length proteins may have unknown, non-desirable and potentially even toxic biological activity in contrast to minigenes that deliver only specific, immunologically relevant targeting information. Immunization with the entire gene may lead to the processing of immunodominant epitopes that may be highly competitive for binding to MHC, although some of these epitopes may be ineffective due to immune tolerance mechanisms and thymic ablation of the corresponding T-cell clones. In contrast, minigenes can be designed to contain only a select number of non-immunodominant epitopes with reduced frequency of negative thymic selection. Furthermore, polyepitope DNA vaccines can be constructed to contain epitope analogs to increase the chances of breaking immune tolerance and epitopes can be spaced by suitable linkers conducive to efficient processing. Numerous approaches have been previously described for the generation of minigenes targeting TAAs engineered for exploitation as cancer vaccines. However, despite intensive studies and various strategic approaches, the design of an optimal minigene that maximizes epitope-specific immune responses has so far remained elusive. Here, we have undertaken a systematic effort to identify an optimal scaffold for epitope-modified minigene (EMM) constructs. We have recently described an efficient T-cell epitope in silico prediction approach based on 3 criteria: 1) binding to 1 out of 5 common MHC class I alleles; 2) uniqueness to the antigen of interest; and 3) increased likelihood of natural processing. We characterized 225 candidate T-cell epitopes (wild-type and fixed-anchor analogs) selected within CEA, HER2/ and hTERT by high-throughput stable binding to MHC using the iTopia epitope discovery assay. On the basis of these results, we concluded that the combination of in silico prediction and a biochemical binding/stability assay represents an accurate prediction of novel TAA-derived epitopes. Indeed this was later confirmed by further validation of Human Leukocyte Antigen (HLA)-A*0201 restricted fragments in HLA-A*0201 (HHD) transgenic mice. In the present study, we generate and functionally characterize a series of EMMs targeting human CEA as a model antigen. Human CEA is one of the most well-studied TAAs over the past 20 y. Its aberrant expression has been long correlated with many cancer types. Furthermore several therapeutic strategies have been developed against CEA and brought into advanced clinical trials. Here, by comparatively assessing distinct approaches to EMM design both in vitro and by iterative immunization studies upon DNA-EGT in HHD/CEA double transgenic mice in vivo, we define an optimal minigene construct exhibiting strong immunogenicity and therapeutic efficacy. By means of an in silico prediction algorithm previously developed by our group, we selected 17 CEA epitopes restricted to HLA-A*0201 (). In addition to the predicted MHC-I allele HLA-A2 binding affinity, primary epitope selection was also based on increased susceptibility to proteolytic processing and the uniqueness of the target antigen in the human genome. A further criterion was the possibility of designing fixed-anchor modified epitope analogs. The algorithm selected nonamers (CEA.569, 589, 687, 691) or decamers (CEA.100, 307, 411, 589, 682, 690) predicted to bind to MHC-I pocket and identified by the position of the first residue occurring within the CEA protein primary sequence. CEA.589 was selected in 2 versions, both as nonamer and decamer (referred as 589), as both forms were predicted to exhibit MHC-I pocket binding capabilities. In 6 epitopes (CEA.100, 307, 411, 589, 589, and 682), wild-type residues at position 9 or 10 were replaced with valine. CEA.690 was modified by replacing isoleucine with leucine in position 2. CEA.691, 569 and 687 were left without modifications. To determine their biochemical properties, wild-type peptides and corresponding analogs were synthesized and characterized for MHC binding and complex stability using the iTopia Epitope Discovery System. Peptides were incubated in duplicate in MHC-coated wells (refer to Methods) and peptides exceeding the threshold of 30% of the assay positive control were classified as binders. As shown in , according to this standard, all peptides were able to bind HLA-A2 and the percentage of the positive control is reported. Interestingly, most of the wild-type and anchor-modified peptides showed high affinity to HLA-A*0201 exhibiting a comparable half-maximal effective concentration (EC) to the CAP-1 peptide (CEA.605) used as positive control (). Although the majority of the analogs displayed a lower EC, and thus had higher affinity for HLA-A2 than the wild-type epitopes from which they were derived, no improvement in binding affinity was observed for either CEA.690L2 or CEA.307V10, the latter of which actually displayed lower affinity than its wild-type counterpart. The binding stability of the peptide-MHC complex was also evaluated over time. Despite the relatively high affinity of candidate epitopes, MHC complex stability varied considerably with 10/17 peptides (59%) forming stable complexes, here arbitrarily defined as a half-life (T) > 4 h (). An improved stability of peptide analogs compared with native sequences was observed, in particular for CEA.411V10, 589V9, 589V10, 682V10 and 690L2 (3.1, 4.5, 5.3, 7.4 and 4.4-fold, respectively). These data were further confirmed in a cell-based T2 binding assay (data not shown). Due to poor in vitro binding features, CEA.100 and its analog were excluded from subsequent analyses. We also decided to exclude CEA.690 and CEA.690L2 considering its high sequence similarity with the already clinically validated, stable CEA.691 epitope. To verify the biologic relevance of the epitope modification suggested by our algorithm, we next set out to test the impact on their immunogenicity in vivo and further, assess their cross-reactivity with wild-type epitopes. To this end, HHD mice were immunized with a mixture containing the candidate CD8 T cell peptide immunogen along with hepatitis B virus (HBV)-core128 helper peptide and immunostimulatory synthetic oligonucleotides (CpG) in Incomplete Freund’s Adjuvant (IFA). Mice were given a second injection 15 d later with the same mix of components. Two weeks later, peripheral blood mononuclear cells (PBMCs) and splenocytes were recovered for immunological assays, including intracellular staining for IFNγ release upon stimulation with wild-type or analog peptides. As shown in , immunization with any of the designed analogs induced a cross-reactive response (), eliciting IFNγ release from PBMCs upon secondary stimulation with wild-type epitope. Furthermore, analogs were much more immunogenic than their wild-type native peptides in activating PBMCs () and splenocytes (). In particular, CEA.589 was highly potent, in sharp contrast to its native counterpart not modified at position 10 with valine, exhibiting a 310-fold increase in immunogenicity. Similar results were obtained for CEA.682, while lower immunogenicity was measured for CEA.411V10, albeit significantly higher than the corresponding wild-type epitope, and comparable to that of CEA.691 which was not modified by the algorithm. Overall, the enhancement of immune response for peptide analogs relative to the corresponding wild-type peptides ranged from 11.8- to 310-fold (). Specifically, the fold increase in the immunogenicity of peptide analogs were 11.8, 310, 12.2, 105, and 16-fold for CEA.411V10, 589V10, 589V9, 682V10, and 307V10, respectively, compared with wild-type counterparts. Importantly, these immunogenicity data correlate with binding data (), thus confirming that binding affinity and off-rate are important parameters to predict the immunogenic potential of a defined epitope. To verify whether elicited effectors are indeed capable of lysing human cancer cells, we used as targets different human HLA-A*0201 colon cancer cells, including SW480, Colo705, and Colo201 which express CEA, and Colo205 that is negative for CEA. These results demonstrate that cells from mice immunized with CEA.411V10, 589V9, 589V10, 682V10, and 691 peptides were able to efficiently lyse only CEA positive colon cancer cells, although with some variability (). In a different experimental setting (data not shown), CEA.569- and CEA.687-specific effectors were also able to recognize CEA cells whereas poor lytic activity was observed by CEA.307 and CEA.307V10 primed CTLs. Importantly, these data show that most of the epitopes predicted by the algorithm are immunogenic and effectively presented in complex with MHC-I on malignant cells. To define a universal scaffold for the construction of an immunogenic and therapeutically effective minigene, several EMM variants were designed and constructed for functional characterization. A first series was based on proteasome-dependent epitope processing. The ubiquitin degradation pathway is an efficient endogenous polyepitope processing mechanism and ubiquitin fusion has been shown to improve the induction of CD8 T-cell response to genetic epitope vaccines by targeting the polypeptide for rapid degradation by the proteasome. Thus, 2 EMMs were designed to covalently linking the same string of CEA epitopes to a mutant form of ubiquitin (G76V, 37) encoded in the vector and fused to the CEA peptide via the flexible linker peptide VGKGGSGG (). The design also included the use of 3 spacer sequences, AAY (), LRA or RLRA () designed to ensure efficient epitope processing by the proteasome. A second series () was based on the proteasome-independent mechanism of proprotein processing operated by furin in the -Golgi compartment of the secretory pathway. For this reason, the leader peptide from the secretory protein tissue plasminogen activator (TPA) was included in all these particular EMMs in order to ensure translocation of the nascent protein into the endoplasmic reticulum (ER). The leader peptide was followed by the string of the previously selected four epitopes CEA.411V10, CEA.691, CEA.589V10, and CEA.682V10 to comprise a polyepitope. Between each epitope, a furin-specific cleavage site (REKR) from the human immunodeficiency virus type-1 (HIV-1) glycoprotein gp120/gp41 was inserted. The first minigene (TPA-CEA-Furin-p30, ) bears the p30 helper epitope from toxin in addition to the above-described elements. In a second furin-specific minigene construct, the membrane translocating sequence (MTS) from the HIV-1 derived Tat gene at the C-terminus (TPA-CEA-Furin-p30-MTS, ) was included for its ability to deliver exogenous antigens into the intracellular compartments where processing into MHC-binding peptides occurs. To assess if fusion of the polypeptide to the heat-labile toxin B subunit of (LTB) is able to increase cell-mediated immune responses after DNA-EGT, as previously observed by our group for several antigens such as CEA, hTERT and MMP-11, 2 other LTB-containing minigenes were constructed, with or without the p30 helper epitope (TPA-CEA-Furin-p30-LTB, and TPA-CEA-Furin-LTB, and , respectively). Next, in order to assay our candidate constructs specifically designed to elicit targeted immunity, these EMMs were used to vaccinate HHD mice by DNA-EGT. Mice received 4 weekly injections and CMI was analyzed 2 wk after the last boost. PMBCs from the immunized animals were challenged in vitro using the pool of modified or native peptides, and CMI was measured by intracellular staining for IFNγ. As shown in , we found that all the minigenes tested elicited a strong CD8 T cell immune response among PBMCs from EMM-immunized HHD mice. Importantly, the observed CMI was cross-reactive with wild-type epitopes. Of note, the fusion with LTB resulted in significantly higher levels of CMI. To assess the contribution of each epitope to the overall immunogenicity of the polyepitope containing EMMs, immunized mice were euthanized and the immune response against individual wild-type epitopes was measured by intracellular staining of splenocytes for IFNγ. As shown in , epitope- and scaffold-specific differences in the percentage of CD8IFNγ lymphocytes were observed. In response to immunization with EMMs acting through the ubiquitin degradation pathway with the AAY spacer, the immune response against CEA.682 was consistently relatively high, followed by the epitopes CEA.691 and CEA.589. However, CEA.411V10 component of the EMM was non-immunogenic, as no response was observed upon stimulation with the wild-type CEA.4ll epitope (). Use of the LRA spacer gave rise to an effective immune response only against CEA.682 epitope () but not against the others, thus indicating that LRA spacer sequence is not an efficacious spacer element. Similar results were obtained with the RLRA spacer constructs (data not shown). Overall the proteasome-independent, furin cleavage-dependent EMMs gave a similar pattern of immunogenicity (), revealing a strong CD8IFNγ immune response against CEA.682 and CEA.691, followed by a good response to CEA.589, but again an ineffective response against CEA.411. Interestingly, the fusion with LTB alone (TPA-CEA-furin-LTB) in the absence of both the MTS leader peptide and the p30 helper peptide () appeared to stimulate the highest level of immune response in this assay. To address the first point, a new furin-LTB-based EMM (TPA-CEA-Furin-Inverted-LTB) was designed in which the CEA.411V10 and CEA.682V10 positions were switched (, upper panels). Upon immunization in HHD mice, the immunogenicity of the new construct was virtually identical to that of the original orientation (, compare the right and left lower panels). Although neither construct elicited a strong response against the CEA.411 epitope, CEA.682V10 remained highly immunogenic when placed in the first position, suggesting that the poor immune response to CEA.411V10 was not due to a positional effect. In order to test the second hypothesis, we performed peptide immunizations using either individual peptides or a mixture of the 4 and measured CMI by intracellular staining for IFNγ in response subsequent stimulation with each individual epitope. As shown in , mixed peptide vaccinations with CEA.682 all maintain comparable immunogenicity to single injection whereas CEA.691 and CEA.589 are slightly less immunogenic when injected as mixtures. In sharp contrast, CEA.411 is dramatically affected by co-injection. These data confirm that the intrinsic affinity to MHC in conjunction with epitope competition in the context of a polyepitope construct may affect the performance of a selected epitope candidate. In order to compare the immunogenicity of our optimized EMM with that of other platforms, we performed parallel anti-CEA vaccinations in tolerant HHD/CEA mice using three different immunogens, including the best performing EMM minigene (TPA-CEA-furin-LTB), the mixture of the same CEA peptides encoded by the minigene plus a helper epitope, and the full-length codon optimized cDNA coding for CEA fused to LTB. Mice receiving either full-length CEA or EMM were immunized by EGT and received 4 weekly DNA injections, whereas mice immunized by peptide mixture received two injections of peptides spaced by a 14-d interval. In either case, the CMI was measured 2 wk later against wild-type peptides as assessed by intracellular staining of splenocytes for IFNγ. Results () show that for each epitope, the highest level of CMI was achieved using the EMM construct, followed by peptide immunizations. In particular, CEA.682 immunogenicity was approximately 13-fold higher with EMM as compared with the full-length cDNA and 27-fold and 2-fold higher for CEA.691 and CEA.589, respectively. For this latter epitope, vaccination with EMM was the only approach capable to overcome the 0.1% response threshold, and was thus considered of sufficient strength to break immune tolerance. On the basis of these results we were encouraged to test the antitumor efficacy of the most immunogenic EMM in a prophylactic model. To this end, 2 groups of 5 HHD/CEA mice were first immunized in parallel by 4 weekly minigene vaccinations or with 2 bi-weekly injections of the peptide mixture. Two weeks after the end of the immunization schedule mice were challenged with B16-HHD/CEA cells injected Two weeks later, the mice were euthanized and the number of lung metastases was determined. As shown in , the full-length CEA cDNA was able to confer a significant antitumor effect, as previously observed in this model. However, TPA-CEA-furin-LTB vaccination exhibited a stronger anticancer activity in this aggressive model. Peptide mixture vaccination was also capable of exerting a potent anti-tumor effect, albeit to a lesser extent than TPA-CEA-furin-LTB. T lymphocytes are a critical cellular component of immunity and play a crucial role in the eradication of cancer cells in mammals. The activation of cytotoxic (CD8) and helper (CD4) subsets of T lymphocytes is integral to cellular immunity, including immune responses targeting neoplastic cells. CTLs and their T-cell receptors (TCR) recognize small peptides derived from intracellular antigens and presented by MHC-I molecules on the cell surface via the endogenous antigen processing and presentation pathway. Peptides for human CD8 T-cell epitopes range in length from 7 to 14 amino acids, although they are typically 9–10 amino acids long. TCR recognition of the peptide-MHC class I molecule complexes on the cell surface triggers the cytolytic activity of CTLs, resulting in the death of cells presenting the stimulating peptide-MHC class I complexes. MHC class I restricted epitope vaccines have been shown to confer immune protection in animal models. Epitope-based vaccines offer a number of advantages compared with vaccines based on full-length TAAs. For example, peptide vaccines can induce effective immune responses to subdominant epitopes when there is tolerance to a dominant epitope. Furthermore, anchor-modified or heteroclitic peptide analogs can be constructed with the ability to break tolerance and (or) further increase immunogenicity relative to native peptides. Finally, the use of peptides as immunogens also minimizes safety risks potentially associated with the use of intact proteins. In this study, we identified an efficient scaffold for the expression of minigenes encoding T-cell epitopes. To accomplish this aim, we first identified a set of immunogenic peptides within CEA and generated analogs selected on the basis of their class I MHC binding properties, specifically the HLA-A*0201 allelic variant. Prediction and selection of HLA-epitopes relied on specific algorithms that rank potential sequences (within a given protein, in our study CEA) on the basis of their binding properties to the MHC-I epitope-binding pocket. This ranking is particularly relevant for large antigens and can result in long lists of putative epitopes with differing relative scores and affinities. However, higher binding affinity does not necessarily translate into increased immunogenicity in vivo. For this reason, mice transgenic for HLA constitute a powerful means to measure candidate epitope immunogenicity in the native context of human major histocompatibility complexes, an approach conducive to prospective evaluation of vaccination strategies including the identification of novel or enhanced epitopes. The peptides and analogs described herein were selected on the basis of their ability to elicit a maximal tumor-specific immune response, as well as for their minimal potential for eliciting off-target autoimmunity. Specifically, the CEA protein was analyzed using an algorithm that ranked protein fragments based on various factors impacting immunogenicity, including CEA epitope binding affinity for HLA-A*0201, similarity of candidate epitopes to fragments of other human proteins, and amenability to immunogenic enhancement. The program introduced single amino acid substitutions in the synthesis of analogs inducing consistently increased biochemical affinity and stability (; ). Of particular importance and as shown in , these parameters were predictive of the peptide immunogenic efficacy in HHD mice in vivo as well as epitope processing and presentation efficiency by human colon cancer cells. CEA peptide-primed HHD effector T cells were in fact able to lyse only HLA-A2/CEA target cells, albeit to varying degrees. However, it remained to be determined in a DNA vaccine setting how the relative cytotoxicity of polyepitope primed effector cells may be impacted by the CEA epitope expression level or other factors linked to processing machinery. To address this question, we selected the optimized CEA.411V10, 589v10, 682, and 691 epitopes as the antigenic components of the EMM immunization construct and considered the vaccine scaffold structural elements. Previous studies have explored minigene vaccines comprising multiple contiguous minimal murine CTL epitopes or contiguous dominant HLA-A*0201 and HLA-A*11-restricted epitopes from the polymerase, envelope, and core proteins of hepatitis B virus and HIV, together with the PADRE (pan-DR epitope) universal T-cell epitope and an endoplasmic reticulum-translocating signal sequence. However, processing of individual epitopes in these minigenes is not assured due to the non-specific nature of proteasomal processing. The approach of inserting AAY spacers between the epitopes and the use of ubiquitin as a protein-targeting sequence was previously tested in the context of minigenes containing CTL epitopes derived from MPT64 and 38 kDa proteins of . Pitcovski et al. assayed a melanoma DNA immunotherapy encoding a multi-epitope polypeptide having 3 repeats of 4 modified melanoma antigens linked by 5 spacer elements that signal proteasomal cleavage and fused to the LTB enterotoxin as adjuvant. In a separate study, an oral DNA minigene vaccine was also evaluated containing the HIV tat translocation peptide and a spacer (AAA) followed by an HLA-A2-restricted CEA T cell epitope, all inserted into a pCMV vector including an ER signal peptide. In another study, Lu et al., described minigenes having multiple CTL epitopes joined via furin-sensitive linkers and containing the HIV-1 tat sequence. In order to define a “universal” minigene structure, we have systematically explored most of these components in 2 minigenes categories: proteasome-dependent EMMs and furin-dependent EMMs. In the first category, the epitopes were expressed as a polypeptide fused to ubiquitin and spaced by proteasome-sensitive linkers. The second class of minigenes comprised translated immunogenic polypeptides addressed to the secretory pathway (via a TPA leader sequence), followed by furin-mediated processing in the -Golgi. As shown in and , our comparative approach revealed stronger immune responses using furin-based EMMs. In addition, fusion with LTB further enhanced the elicited response whereas other elements, such as the p30 helper epitope and MTS sequence were not found to be necessary. We have previously shown that LTB sequence contains CD4 specific epitopes, therefore it is likely that LTB-fusion stimulates CD4 T helper cells sufficiently to achieve optimal immune response. Lastly, we demonstrate that the relative position of the epitope within the minigene does not influence its immunogenicity. On the other hand, some epitopes, like CEA.411V10, suffer competition with other epitopes both when delivered in the format of a minigene or as a peptide mixture (). Observations that this epitope is poorly immunogenic when co-administered together with other CEA epitopes despite single peptide vaccination efficacy () evinces the occurrence of epitope competition and may represent a potential limitation to our approach. However, combinations of short immunogenic peptides or, alternatively, synthetic long-peptide vaccines are currently being evaluated in clinical trials. One example is the Eastern Cooperative Oncology Group Phase II Trial E1696, in which a mixture of peptides containing multiple epitopes derived from MART-1, gp100, and tyrosinase was administered in patients with metastatic nonresectable melanoma. In another approach, cancer patients are vaccinated in Phase II and III studies with multiple tumor-associated peptides (TUMAPs) isolated from tumor specimens and identified by mass spectrometry. It would be of interest to evaluate whether epitope competition also occurs in patients in these trials. Finally, the best performing EMM (TPA-CEA-furin-LTB) was also evaluated in a tumor challenge study, using the previously established and aggressive B16-CEA/HHD metastatic model in tolerant HHD/CEA transgenic mice. B16-CEA/HHD cells have been previously shown to be recognized by HHD effector T cells stimulated with CEA vaccines. Immunization of recipient mice with our minigene vaccine elicited significant immunogenic protection, thus demonstrating the anticancer therapeutic efficacy of our strategy (refer to ). It remains to be seen, however, how the immune response against each single epitope within the EMM construct contributes to the overall antitumor effect. In conclusion, we have discovered a universal strategy applicable to the design of vaccine minigenes comprising either predicted or experimentally identified epitopes for delivery via DNA-EGT. Our results provide rationale for further studies, including testing this approach in combination with other treatment modalities, such as peptide vaccines or other genetic immunotherapy vectors. HLA-A*0201 (HHD) transgenic mice were bred at Charles River Laboratories and were kindly provided by Dr Lemonnier (Pasteur Institute). These mice are transgenic for the HHD complex (human β2-microglobulin fused to HLA-A*0201 α1 and α2 domain, H-2D α3 domain) and are devoid of H-A2 and murine β2-microglobulin. For this reason the immune response elicited in these mice is specifically restricted to human HLA-A*0201, making this line a suitable model for epitope identification and optimization. HHD/CEA hybrid mice have been obtained by breeding transgenic mice homozygous for CEA with HHD mice. These mice express human CEA antigen presented exclusively by human HLA-A*0201 and represent a unique in vivo animal model to predict and study human immune response of a human CEA–based vaccine. Six to 8-wk-old HHD and HHD/CEA mice were used in this study. At the end of the treatment period and before necropsy, mice were euthanized by compressed CO gas as indicated in the AVMA (American Veterinary Medical Association) Panel on Euthanasia and according to the United Kingdom CO-ordinating Committee of Cancer Research (UKCCCR) guidelines. The experiments were conducted according to EU Directive EC86/609 on the protection of animals used for experimental and other scientific purposes, which was ratified by Italian Legislation with DL no. 116/92 on 19 February, 1992. SW-480, Colo205, Colo705, and Colo201 cells were obtained from American Type Culture Collection (ATCC). B16-HHD/CEA cells were generated by transfecting the murine melanoma cell line B16-F10 (ATCC, cat. 6475) sequentially with 2 plasmids, one encoding CEA (pcDNA3-CEA) and the other HHD (pcDNA3-Hygro-HHD). Cells were maintained in culture at 10% CO in Dulbelco’s modified Eagle’s medium (DMEM), 10% FCS with 1% Pen/Strep, 800 µg/mL G418 and 400 µg/mL hygromycin (Invitrogen). Lyophilized CEA peptides were purchased from Jerini (JPT) and resuspended in DMSO at 40 mg/mL. Pools of peptides of 15 amino acids overlapping by 11 residues were assembled as previously described. Peptides and pools were stored at −80 °C. Seventeen candidate epitopes were selected in silico from CEA on the basis of 3 criteria: predicted binding to MHC-I alleles HLA-A*0201; uniqueness in the human genome; and increased likelihood of natural processing, as previously described. Peptide MHC binding and complex stability was characterized using the iTopia Epitope Discovery System (Beckman Coulter), as previously described. Briefly, the assay utilizes avidin-coated microtiter plates containing biotinylated MHC-I monomers loaded with β2-microglobulin (β2M) and placeholder peptides. The monomer-coated plates, assay buffers, FITC-conjugated anti-HLA-ABC monoclonal antibody (mAb) B9.12.1, β2M and allele-matched positive control peptides were obtained as part of the iTopia kit. Monomer-coated plates were first denatured, releasing the placeholder peptide and leaving only the MHC heavy chain bound to the plate. Test peptides were then introduced under optimal folding conditions, along with the anti-HLA-ABC-FITC monoclonal tracer antibody. Peptides were first evaluated in the initial binding assay (11 μM peptide incubated for 18 h at 21 °C with β2M, anti-MHC mAb and plate-bound MHC heavy chain). Irrespective of the initial binding, peptides were also screened in the stability assay to determine MHC complex half-life (T; time taken for 50% reduction in binding at 37 °C after removing peptide from the assay buffer). Briefly, peptides were incubated overnight at 21 °C in MHC-coated wells. Assay buffer was replaced with fresh buffer containing no peptide. Plates were then incubated at 37 °C and read at multiple time points over a 24 h period. Dissociation rates were calculated by GraphPad Prism software using a single-phase exponential decay equation fitted to data recorded during the first 8 h, as recommended by the manufacturer and for consistency with earlier studies using this assay. All assay plates were read using a Cytofluor II fluorometer (PerSeptive Biosystems). Mice were injected with 50 μg of plasmid DNA in a 50 μL volume into the quadriceps followed by electroporation, as previously described. Peptide vaccination was performed by subcutaneous injection of a mixture of 100 μg peptide, 140 μg HBV-T Helper epitope and 50 μg CpG-ODN (Sigma) in Incomplete Freund’s Adjuvant (IFA) per mouse. The immune response was evaluated 2 to 3 wk after the final treatment. The detection of peripheral immune response was measured by intracellular staining for IFNγ as previously described. Briefly, PBMC or splenocytes harvested from immunized mice (or controls) were resuspended in 0.6 mL RPMI, 10% FCS and incubated with the indicated pool of peptides (5 μg/mL final concentration of each peptide) and brefeldin A (1 μg/mL; BD PharMingen) at 37 °C for 12–16 h. Cells were then washed and stained with CD3, CD4, and CD8 surface antibodies (BD PharMingen). After washing, cells were fixed, permeabilized and incubated with the IFNγ-FITC antibodies (BD PharMingen), fixed with formaldehyde 1% in PBS and analyzed on a FACSCalibur flow cytometer, using CellQuest software (Becton Dickinson). Dimethyl sulfoxide (DMSO) and staphylococcal enterotoxin B (SEB, Sigma) at 10 μg/mL were used as internal negative and positive control of the assay, respectively. Assays were performed according to standard protocols. Briefly, lymphocytes were isolated from harvested spleen of 3 mice per group 5 d after the final vaccination. Cells were resuspended to 2 × 10 cells/mL and were stimulated with CEA peptide pools along with 20 IU/mL recombinant human IL-2 (Sigma). Five days later, these in vitro stimulated cells were used as CTL effector cells, and the CTL activity was determined by a standard 6 h Cr-release cytotoxicity assay using the indicated cell lines as targets. Cr labeled cells were then added to wells for 6 h at an effector to target cell ratio = 100. Specific lysis was calculated as (experimental Cr release − spontaneous Cr release)/(maximal Cr release − spontaneous Cr release) × 100. HHD/CEA mice were injected with 5 × 10 of B16-HHD/CEA cells. This dose of tumor cells is lethal in 100% of mice within 4 to 6 wk after transplant if left untreated. Three weeks after challenge, surface lung metastases were enumerated via optical microscope (Leica). Statistical analyses were performed by Student's 2-tailed test. The data are presented as means ± SD. A value < 0.05 was considered significant.

Plants are major feeding source for insects and other living organisms. Insects feed nearly all parts of plants by different feeding methods and cause severe damages especially to the crop plants. Coleopteran and lepidopteran insects damage crop by direct chewing or biting of different plant parts however homopteran insects cause direct (by sucking phloem sap) as well as indirect damage (by spreading various viral diseases). Plant insect interaction is reported for the past hundreds of years. Both insect and plant exchange chemical and other signals which establish interaction between them. The integrated responses in plants direct the production of phytochemicals against the feeding arthropods. Plants are sessile organism therefore mainly depend on chemical or other phenolic compound for defense against insect attack. They use toxic secondary metabolites, proteins, hormones, and several types of structural and phytochemical changes to repel the insects. Overexpression of some oxidative peroxidases has also been reported as defense responsive system in plants. Several reports are available in which activity of oxidative enzymes like peroxidase found to be increased after insect attack. Reactive oxygen species (ROS) signaling and hormone-signaling pathways are strongly linked with plant–insect interactions. Differential protein expression analysis of rice after infestation with brown plant hopper showed upregulation of numerous peroxidases along with many other enzymes. Further, transcriptomic studies of susceptible and resistant cultivars of and in response to insect attack have shown remarkable changes in oxidative stress related transcripts. Sitka spruce also showed overexpression of many proteins including a number of peroxidases after infestation with budworm and pine weevils. Maffei et al. observed that when Lima bean leaves challenged with the peroxidase activity increased significantly than the unchallenged and mechanically wounded plants. Peroxidases oxidize several compounds by using HO. They are generally heme group containing glycoproteins and divided into acidic, basic and neutral types in plants. Plants peroxidases have many forms, which are encoded by multi gene families. Several utilities of peroxidases have been reported in plants, like degradation of HO, removal of toxic compounds, defense against insect herbivore and many other stress related responses. In present study we have analyzed differential enhancement in peroxidase activity in response to 2 major types of insect pests (chewing and sucking) on 3 agriculturally important crop plants (cotton, tomato and cowpea). The major objective of the present study was to decipher, if any, role of insect attack (chewing or sucking) on stress enzyme (peroxidases) as well as to establish that defensive peroxidase activity enhancement was varied with each crop. Further MS-MS analysis revealed the presence of 4 different forms of peroxidaes in in-gel peroxidase activity and their presence is also varied according to the plant species. Sap was extracted from 3 different plants (cowpea, cotton and tomato) before and after 48 h of insect infestation (chewing type; , sucking type; and ) and total protein content was estimated by Bradford method. Before infestation, quantity of total proteins in the sap of cotton, tomato and cowpea was 133, 69 and 34 µg/ml, respectively. Protein content in all plants was decreased after infestation of chewing as well as sap sucking insects. After infestation with sap sucking insect, protein content in sap of cowpea, cotton and tomato was decreased by 30, 41.66, and 17.70%, respectively (), however in the case of chewing insect it was declined by 26.47, 24.24, and 10.14%, respectively. Bradford assay revealed that, cowpea having maximum amount (3.78 mg/ml) of TSP which was decreased after insect attack (48 h). TSP of Cowpea leaves was decreased by 27.29 and 48.14% after the infestation of sap sucking (2.75 mg/ml) and chewing (1.96 mg/ml) insects, respectively. Further, TSP content of cotton leaves was decreased by 32.8 and 45.8% after the infestation of sap sucking (1.73 mg/ml) and chewing insect (1.38 mg/ml) respectively, over control (2.55 mg/ml). Minimum TSP content was found in tomato plant (2.48 mg/ml) which was also decreased after insect feeding. Feeding of sap sucking insect (1.85mg/ml) decreased TSP content by 25.40% while chewing insect (1.68 mg/ml) decreased by 32.25% (). Sap and TSP of all the 3 plants were analyzed for peroxidase enzyme assay in microtiter plate using TMB/HO as substrate. Peroxidase activities of all three plants followed standard Michaelis–Menten enzyme kinetics. Velocity of the enzyme catalyzed reaction increased continuously by increasing substrate concentration and became constant after a definite point (). GraphPad prism-6 analysis revealed that, K value of cowpea’s enzyme (0.385 ± 0.02 mM) was minimum in all 3 tested enzyme followed by tomato (0.475 ± 0.04 mM) and cotton (0.622 ± 0.03 mM). However, V of peroxidases of cowpea, tomato and cotton plants were 70.81 ± 1.32, 82.18 ± 2.23, and 75.09 ± 1.52 µM min µg , respectively. Turn over efficiency (K) of peroxidase from cotton (~101.3 min) was almost similar to tomato (~100.8 min) but higher then cowpea (~98.21min). Excised protein band was reduced by DTT and then alkylated with iodoacetamide. Afterward, protein was digested with trypsin and peptides were examined on MALDI-TOF-TOF platform. Generated MS/MS data was analyzed on MASCOT search engine via a MASCOT MS/MS Ion Search. Obtained result confirmed that observed band on gel was peroxidase (). MS/MS data of tomato showed significant similarity with two protein i.e., peroxidase 5 precursor of (gi5002348; match score, 252) and peroxidase 53-like protein of (gi356533121; match score, 238) while MS/MS profiles of both cowpea and cotton were significantly matches with cationic peroxidase of (gi577503) and peroxidase of (gi66840760). Peroxidases are the key enzyme of defense related pathways in plants and play core role in response to wide range of pathogens. They also participate in many primary metabolic processes such as auxin metabolism, lignin and suberin formation, cross-linking of cell wall components, phytoalexin synthesis, and in metabolism of ROS and RNS. Peroxidases are coded by a large multigenic family. There are 138 and 73 members of peroxidases reported in rice and , respectively. Present study reports that peroxidases were elicited by the infestation of chewing and sap sucking insects indicating their role in defense against phytophagous insects. Three crop plants (tomato, cowpea and cotton) were taken for the study and peroxidase activity measured after insect feeding in sap protein as well as TSP of plant leaves. Sap of Tomato, Cowpea and Cotton has 69, 34, and 133 µg/ml of protein, respectively. However it is already reported that, xylem sap of apple has 300 µg/ml of protein, whereas peach and pear xylem sap contained approximately 100 µg/ml of protein. TSP in the leaves of tested crop plant (tomato, 2.48%; cowpea, 3.78 mg/ml; cotton, 2.55 mg/ml) is 20–100 fold higher than its sap protein, as the sap is known for less protein content and mainly involved in transport of minerals, amino acids, organic acids, sugars, and sugar alcohol through sieve elements (SEs). SEs are the dead cells and not able to synthesize RNA and proteins. They only bring macromolecules through specialized plasmodesmata connecting SEs to the adjacent companion cells. Protein content in sap as well as leaves are significantly decreased after insect infestation because in biotic stresses plant de-accelerates the rate of protein synthesis and whole translation machinery shifted to produce defense related proteins. It can be the reason that protein content was decreased in both sap and leaves but the peroxidase activity was enhanced. Peroxidases are pathogenesis-related proteins (PRs) and having potential roles in the interaction between plants and insects. Our result suggested that peroxidase activity in sap of plants (tomato, 19.16; cotton, 18.44 and cowpea 24.77 units/µg) is much higher than TSP of leaves (tomato, 2.12; cotton, 2.49 and cowpea 2.01 units/µg), but fold change in activity of peroxidase after insect attack is almost similar in both the cases (Sap and TSP). Approximate ~1.6–~3 fold enrichment in peroxidase activity was observed in both sap and TSP. It suggested that each part of plant respond similarly upon insect attack. Level of peroxidase activity was enhanced after feeding of both chewing and sap sucking insect which showed that the similar pathway was activated in which peroxidase plays a critical role. In-gel assay data revealed that peroxidases are the major protein after insect attack in sap because level of total protein in sap was decreased but level of peroxidase enzyme was increased. Enzyme kinetic study confirmed that peroxidase of all three tested plant follow the Michaelis–Menten enzyme kinetics. The turn over efficiency of cowpea peroxidase (98.21 min) was lowest among three examined plants because it catalyzes reaction with minimum velocity (70.81 ± 1.32µM min µg). This might be a reason for low enhancement of sap peroxidase activity (2.17 fold) in sap sucking insect infested plants. However, maximum turnover efficiency of peroxidase from cotton (101.3 min) was resulted into the highest enhancement in peroxidase activity (2.53 fold) in sap sucking insect infested plants. The results confirmed that the level of peroxidase activity enhancement after insect infestation was directly correlated with the efficiency of enzyme. MS/MS profile of cowpea and cotton showed significant similarity by maximum score with cationic peroxidase of while in tomato it matches with peroxidase-5-precursor of by maximum score. Young et al. also demonstrated that cationic peroxidases were accumulates in xylem vessels during incompatible interactions with in rice. However, it was also reported earlier that overexpression of anionic peroxidase confers resistance against Lepidopteran and Coleopteran insects. Present study suggested that defense mechanism of plants against sap sucking and chewing insect was similarly induced by both the insects and involves peroxidase enzyme as a key player. Peroxidases may be used to raise insect resistant plants against broad range of insects by transgenic approach. Cowpea (), cotton () and tomato () plants were grown in pots containing garden soil under controlled growth chamber conditions (16 h light, 26 ± 2 °C, 75% relative humidity). Three-week-old plants were used for the experiment. Insect were reared in their native host plant. Larvae of were reared on castor while aphids () and whiteflies () were reared on cowpea and cotton plants, respectively. Temperature was maintained at 26 ± 2 °C with 16 h photoperiod and 75% relative humidity. All 3 different crop plants were infested with chewing as well as sap sucking insects, separately. Second instar larvae of (previously reared on castor leaves) were taken as model for chewing insect due to their versatile host range. Whiteflies, one of the most emerging sap sucking insect, were used for cotton and tomato plants while aphids were used for cowpea because of their native host. Sap was extracted from infested and uninfested plants using pressure chamber. Plant’s stems were cut across from lower portion, washed thoroughly with milliQ water, blot dried and fixed in pressure extractor (Model 3005, Soil moisture equipment corp.). Initially low pressure (3 bars) was applied which gradually increased (till 10 bars) using commercial nitrogen gas. The exuding fluid (1.5–2 ml) was collected and stored on ice. One gram leaf tissue per sample was powdered using liquid nitrogen, homogenized in 3 ml extraction buffer (20 mM HEPES, 10% glycerol, 1mM EDTA 100µM PMSF, 5 mM DTT, and 1 mM benzamidine) and centrifuged (16,000 × g, 10 min, 4 °C). Supernatant was collected in fresh tube, dialyzed against 20 mM Tris-Cl and stored at 4 °C for further experiments. Protein concentrations were determined by Bradford method using bovine serum albumin (BSA) as standard. Standard curve was plotted by series of BSA concentrations and used for the calculation of concentration of unknown protein sample. Peroxidase activity was analyzed by spectrophotometer as well as in-gel assay. Protein samples (100 ng) were coated on microtiter plate by overnight incubation at 4 °C. Standard peroxidase enzyme (obtained from Sigma) was also coated at different dilution on Elisa plate in similar way. Plate was washed with water and assay performed by using 100 µl TMB (3, 3′, 5, 5′ Tetramethylbenzidine)/ HO (Hydrogen peroxide) as substrate. Reaction with peroxidase results in the conversion of the substrate into a soluble blue product. The rate of blue color development is a direct measure of the rate of the peroxidase reaction. Reaction was stopped by adding 50 µl of 100 mM HSO after 5 min which transforms blue product into a super-oxidized soluble yellow product. Peroxidase activity was measured by determining the absorbance of yellow product at 450 nm. Increase in absorbance was directly proportional to the peroxidase activity. A standard curve was plotted with the absorbance data of standard peroxidase enzyme and used for the calculation of peroxidase activity in experimental plant samples. In-gel assay was performed on SDS-polyacrylamide gel following the protocol established in our laboratory with slight modification The protein samples were prepared without DTT and heat denaturation, and resolved on 12% SDS-PAGE. Gel was washed twice with 2.5% Triton × 100 and then with distilled water. Peroxidase band was developed by using diaminebenzidine/HO (DAB) substrate according to manufacturer’s protocol (GeneI, Merck). Kinetic parameters (K, V and K) were calculated for peroxidases isolated from the sap of all three tested plants. To compare enzyme activity of peroxidase enzymatic reactions were performed with different concentrations (0.05–4mM) of substrate (Tmb/HO) at regular intervals. The obtained data was analyzed by Michaelis-Menten equation on nonlinear regression using GraphPad Prism-6 software. Peroxidase band developed in in-gel assay was used as marker to excise band from a duplicate gel and used for mass spectrometric analysis. The excised protein band was reduced with 10 mM DTT (56 °C, 30min), alkylated with 50 mM iodoacetamide (room temperature, 30 min) and digested with trypsin (Sequencing grade, Porcine, Promega) in 50 mM ammonium bicarbonate buffer (pH 8.0) 37 °C for overnight. The digested peptides were extracted by sonication in extraction solution (60% ACN and 1% TFA) for several times,dried by centrifugal evaporation and suspended in 5 μl resuspension solution (50% ACN and 0.1% TFA). Suspended peptides (0.5 µl) were spotted on MALDI plate followed by 0.5 μl of CHCA matrix (10 mg/mL in 50% ACN, 0.1% TFA). Spots were dried completely and used for PMF on MALDI-TOF-TOF platform (model 4800, ABsciex). The peptides with higher signal intensity were selected for MS/MS analysis. The spectra obtained from MS/MS were analyzed by searching against the MSDB, Swiss-Prot and NCBI database with the following parameters: fixed precursor ion mass tolerance of 20 ppm, fragment ion mass tolerance of 0.05 Da, calibration error of 0.005 Da, one missed cleavage, carbamidomethylation of cysteines and possible oxidation of methionine. The spectra of common contaminants were removed by searching against contaminant database, and the remaining data were used for the identification of proteins. All the experiments were performed in triplicates and the average was calculated. All the statistical analysis was done on SPSS software.

The genus is one of the largest genera of the monophyletic Brassicaceae family with more than 150 species distributed worldwide. (commonly called as Pepperweed, Pepperwort or Peppergrass) is an invasive plant of western Asia and southeastern Europe and currently distributed from Norway in the west to up to western Himalayas in the east. In high-altitude cold-arid Ladakh area of Western Himalayas, leaves of are consumed as vegetable and salad by the natives. In addition to its agronomic, economic and ecological importance, is also known for its high medicinal value and used in treatment of stomach related disorders. Each mature plant of produces thousands of seeds each year, however only a small percentage of seeds survive. The inherent dormancy in the seeds further delays the onset of advanced generation by a few months. Dormancy may be exogenous or endogenous in nature and a number of molecular to environmental signals may be required to overcome it. However, several physio-chemical pre-germination treatments are known that can break the dormancy by acting as direct or indirect artificial endo/exogenous signals for seed germination. Interestingly, most of these methods are simple, low-cost and can be handled by relatively unskilled manpower. Seed pre-treatments not only improve seed germination rate, they also result into faster and synchronous seed germination and have often reported to improve the matured plant’s ability to tolerate environmental stress. , despite its occurrence in conditions of abiotic stress, has escaped the focus of the scientific community. Till date, there are only a few reports on pre-germination seed treatments or characterization of the stress responsive genes. To improve the efficiency of seed germination and characterize the gene expression behavior of plants away from their natural habitat, we have applied a number of seed pre-treatments and subsequently assessed the performance of plants under drought stress conditions. As little molecular data of exists in public domain, known stress responsive genes from have been cross-amplified in . Pre-treatments were classified on the lines of types of dormancy. Certain treatments like ionic solutions, salts, acids, etc. are primarily used to overcome exogenous dormancy, while phytohormones like GA3, hot and cold water stratifications were generally used to overcome endogenous dormancy. Among the various methods used to overcome exogenous dormancy, seed pre-treatment with salts could improve seed germination rates, as they stick to cell surfaces and thereby induce osmotic pressure on the cytosol. Most saline pre-treatments of the seeds have been based on NaCl and KNO. We found that concentrations of KNO improved both seed germination by 5-fold, as well as the germination velocity (Timson’s index) in comparison to control (). Timson’s index is effective in evaluation of germination considering the time to 50% germination and the final percentage of germination obtained. Pre-treatments with NaCl (50 and 250 mg/L) did not show any major improvement in the germination percentage and seed treatment with 500 mg/L NaCl had rather a negative effect on seed germination. In a previous study from this laboratory, effects of thiourea treatments on germination percentage were found comparable to KNO treatments on seeds of . The same has been found true in the present study for as well (). However, the germination velocity was more profound in case of KNO treatment as compared with thiourea treatments. Pre-soaking of seeds in 98% HSO for 5 min was found as the best pre-germination treatment in terms of obtaining the highest germination percentage (). Shorter soaking times in 98% HSO i.e., 1–2 min. had lesser effect on improvement of germination percentage. For many species, the seeds buried under soil germinate by themselves after remaining dormant for a particular season i.e., winter or summer months, as part of their seasonal patterns of dormancy behavior. Such seeds show a form of dormancy called as endogenous dormancy, which is either controlled by environmental signals or by relative levels of phytohormone concentrations inside the cells. In laboratory, these conditions can be easily mimicked by soaking the seeds in hot or cold water. Alternatively, applications of phytohormones may produce the desired effect. In the present study, hot water treatments had no positive effect on seed germination () and cold water treatments completely inhibited the seed germination, which strongly suggested that dormancy in is controlled by exogenous factors and there is little, if at all, effect of high or low temperatures under natural conditions. Our results were further strengthened by the observation that different GA concentrations had relatively small effect on seed germination (). Among all the treatments described above, concentrations of KNO made most uniform impact on the germination characteristics of seeds. It is well established that its application impact the performance of plants in a positive manner and also increase their water use efficiency (WUE). We transplanted the plantlets obtained after KNO treatment to pots filled with vermiculite and further assessed for their ability to tolerate drought stress. From each set of KNO treated seedlings, three subsets were subjected to drought stress by 5 and 15% PEG. A control set of equal number of seedlings were irrigated with distilled water. These conditions were maintained for 15 d, after which further subsets were drawn and exposed to 5 or 15% PEG and control subset irrigated with water for another 15 d, as detailed in . The state of survivability of the plants was assessed based on leaf counts and the lengths of their midribs. The control set (Experimental set up series 1; ) did not either survive the duration of the experiment or wherever some plants survived, they showed stunted growth. Interestingly, plants showed better adaptability to stress in initial period of stress after transplanting. As the latter 15 d stress of 5% PEG only seemed to have been acclimated by the plants (). The behavior of plants in terms of elongation of aerial parts in response to increasing PEG concentrations vary with species and evidences exist that point such responses to be age-dependent as well. While certain plants like tomato show positive correlation with increase in lengths of aerial parts under increasing PEG concentrations, others like salvia show negative correlation. Interestingly, in case of , 5% PEG treatments showed increase in shoot length, but higher concentrations resulted in antagonist response. In general, the drought stress inhibits cell enlargement and affects various physiological and biochemical processes. Among different concentrations of seed pre-treatments, 0.1% KNO primed seeds showed better adaptability to drought stress in later stages. Drought is a complex stress, in whose response many bio-molecules interact including nucleic acids, proteins, carbohydrates, lipids, hormones, free radicals, etc. In fact, drought affects almost all aspects of plant cell metabolism. As the plant responses to different stresses have cross-talk among themselves at more than one level of metabolism, studies targeted onto any one may have direct implications in understanding other stresses as well. Evidences exist, wherein drought stress has been found related to stresses caused by salinity, temperature extremes, pH extremes, pathological reactions, senescence, growth, development, UV-B damage, wounding, embryogenesis, flowering, signal transduction and so on. Thus, the signals of water scarcity drive the regulation of a number of plant genes. A good number of the genes are induced within a few minutes of receiving the signal and have been called as “early responsive genes.” Transcription factors that further effect expression of other genes are mostly categorized into this category. The late or delayed responsive genes, in contrast are activated by stress more slowly and their expression is sustained. Fourteen primer pairs were designed for known stress responsive genes in to amplify homologous genes in closely related Five of these genes, i.e., zinc finger protein coding gene (), LHCA4 () encoding PSI type IV chlorophyll in , vernalization related gene (), plastid lipid associated protein () and senescence related gene (), were successfully amplified in . In order to assess the efficiency with which KNO treated plants acclimatize to varying degrees of drought stress, the expression levels of each of these five genes were studied based on end-point fluorescence and using cDNA from ten experimental set ups exposed to varying degrees of stress. The zinc finger protein coding gene (ZFAN) acts as a transcription repressor during abiotic stress and is an early responsive gene. The effect of 0.3% KNO could be clearly seen in the ability of the plants to accumulate its transcripts during drought stress induced by PEG treatments (). The vernalization () and genes, known to be responsive to cold stress, were found to have no significant differential expression in response to the drought stress. Only a minute amount of transcript accumulation was recorded for gene in the plants matured out of 0.3% KNO treated seeds, whereas, no net significant difference could be observed in accumulation of gene in differently treated plants samples under study. is known to get up-regulated during abiotic stress. Interestingly, we observed that this gene overexpressed by three times in the plants maturing out of 0.3% KNO treated seeds compared with the plants that were matured out of 0.1% KNO treated seeds (). However, in a given set of water vs. PEG treatment analysis, no change could be observed in the finally accumulated amplified transcripts. is expected to get up-regulated during the stress, but its expression was found more or less constant in all the samples (). The abilities of the plants to perform well under stress after treatment with KNO might have a direct relation with the plant’s ability to utilize the constituent ions (K and NO) separately. At cellular level, Potassium (K) substantially affects enzyme activation, protein synthesis, photosynthesis, stomatal movement and water-relation (turgor regulation and osmotic adjustment), signal transduction, intracellular movement of macromolecules (especially sugars) and many other processes. It is known to enhance plant growth, yield and drought resistance in different crops under water stress conditions. Further, Nitrate ions promote water use efficiency (WUE), photosynthesis and biomass production rates. Applied together, potassium and nitrate ions produce synergistic effect, enhancing survival of seeds and subsequent drought tolerance. Pre-treatment of seeds with KNO (0.1–0.3%) has thus been found an effective method to enhance the germination percentage and ensure its growth under arid conditions. For a cold arid climate like that of Ladakh, where leaves are cooked as vegetables, such practices might be important to ensure the availability of food material for the local population. Similar applications in other food crops can also be applied to ensure their harvest even under conditions of water deficiency, especially in countries with lesser acceptability of genetically modified crops and foods. Mature seeds of L. were collected from Leh, India (34° 8′ 43.43ʺ N, 77° 34ʹ 3.41ʺ, 3505 m asl) and transported to our laboratory in Haldwani, India (29° 13ʹ 11.946ʺ N, 79° 31ʹ 11.9244ʺ E, 443 m asl) in zip lock bags. Seeds were disinfected by immersing in 0.5% sodium hypochlorite solution for 2 min followed by rinsing thoroughly with distilled water four times. Various pre-treatments, as described earlier and depicted in were performed in triplicate sets of 60 seeds each. Emergence of radicle was recognized as the event of seed germination and the plants were grown for 30 d in 16/8 h light/dark cycle at 25°C. Besides final germination percentages, the rate of germination (or germination velocity) was also calculated according to a modified Timson’s velocity index or Timson Index: ΣG/T, where “G” is the percentage of seeds germinated after 2 d interval and “T” is the total time of germination. The overall line of work is described in . Primers of 14 genes () from were used in PCR with cDNA of as template. Out of these, five genes were amplified successfully, which were assessed further using semi-quantitative PCR analysis to study transcript accumulation. gene was used as an internal control for assessment of differential regulation of the five genes under analysis. The PCR reaction was performed in a total volume of 25 µl, which was constituted of 10 pmol of gene-specific primers (), 20 µM of dNTPs mix, 1.5 mM MgCl, 1× polymerase buffer and 1U polymerase (Genei, Merck Millipore). Thermal cycling conditions constituted of initial denaturation of 5 min at 94°C, followed by 35 cycles of denaturation at 94°C for 30 sec, re-annealing for 30 sec at 55°C followed by elongation at 72°C for 30 sec. A final elongation for 5 min was performed at 72°C. Amplicons were run on 2% agarose gel and stained with ethidium bromide (0.1 ppm) in 1× TAE buffer. The gel was documented on a phosphoimager (Typhoon Trio+ Imagers, GE Healthcare) and analyzed using ImageQuant software (GE Healthcare). Molecular sizes and their quantities were calibrated against the known values of various DNA bands of the GeneRuler 100 bp Plus ladder (Fermentas). The background intensity of the gel material was suitably subtracted prior to analysis.

Multiple human tumor viruses maintain their genomes extrachromosomally using host cell machinery for many of their replicative functions (; ; ; ). Epstein–Barr Virus (EBV), which causes both lymphomas and carcinomas, is an extreme example of this cellular parasitism, minimally encoding two -acting elements, a Dyad symmetry (DS) as an origin of DNA synthesis, a Family of Repeats (FR) as a maintenance element, and a -acting protein, EBV nuclear antigen (EBNA1) that binds these DNA elements to foster their activities (). EBV depends on its human host cell for all other functions required for synthesis and partitioning of its plasmids. Studies of EBV plasmids have helped not only to elucidate the mechanisms of their synthesis, but also to understand the cellular activities on which they depend. Examination of EBV's origin of plasmid replication, P that consists of FR plus DS (), has shown, for example, that DS binds EBNA1, and with its particular spacing thereby recruits the origin replication complex (ORC) (; ; ; ; ). P is also a paradigm for a mammalian autonomously replicating sequence (ARS), which can be efficiently introduced and maintained under selection in cells expressing EBNA1. It is not clear, although, how EBV and plasmids derived from it are maintained and partitioned in proliferating cells. EBNA1 has been proposed to mediate these events by binding FR site specifically and tethering the bound plasmids to chromosomal sites (; ), either by associating with specific cellular proteins () or by binding AT-rich cellular DNA through its AT-hook activity (). This proposal would yield random partitioning; a result not obviously compatible with analyses of the rate at which cells lose P vectors following removal of selection and become susceptible to being killed on reapplication of the selective agent (; ). We have resolved this apparent contradiction by monitoring individual EBV plasmids throughout the cell cycle. DNA sequences can be detected in fixed cells by fluorescence hybridization (FISH) and in live cells by their binding fluorescent proteins. We have used both approaches, modifying the latter to optimize its detection of EBV-derived plasmids, without perturbing their replication. used a derivative of the Lac repressor fused at its amino terminus to GFP to detect 256 binding sites for the repressor in the dihydrofolate reductase (DHFR) locus with or without its amplification. This derivative, unlike the wild-type repressor, binds DNA in the presence of the inducer, isopropyl-1-thio-β--galactopyranoside (IPTG). This property posed a problem in our studies, because long-term occupancy of 264 sites on EBV plasmids selects for their integration into host DNA (J Komano and B Sugden, unpublished findings). We therefore constructed and used wild-type Lac repressor fused at its carboxyl terminus to a tandem dimer of RFP () and a nuclear localization signal (NLS), and carried cells in the presence of IPTG to prevent binding of the repressor except when plasmids were visualized. One earlier study visualized EBV-derived plasmids with the GFP-Lac repressor fusion of before the plasmids became established (). EBV plasmids are lost at apparent rates of ⩾25% per cell generation for the first 2 weeks following their introduction into cells, after which they are established and lost at apparent rates of 3–5% per cell generation (). We consequently monitored EBV-derived plasmids in cells after their establishment in order to measure their dynamics at a steady state. Clones of HeLa cells harboring on average two to four copies of EBV-derived plasmids were examined for their individual numbers of plasmids in the absence of selection throughout the cell cycle. These analyses indicated that 84% of all plasmids were duplicated in S phase, and all of these by the end of S phase appeared as colocalized pairs with signal intensities close to two-fold those of the unduplicated plasmids. This colocalization of newly synthesized sister plasmids is without precedent in mammalian cells. A total of 88% of these colocalized pairs partitioned faithfully, separating at anaphase; 12% segregated to only a single daughter cell. The 16% of plasmids that failed to be duplicated segregated randomly, and less than 0.3% of all plasmids detected in G2 failed to being detected in daughter cells following mitosis. These experimentally determined rates were used in computer simulations to predict the evolution of plasmids in cells over time. The distribution of plasmids in a starting population of HeLa cells with an EBV-derived plasmid and in 293 cells with intact EBV were determined, selection removed, and the distributions again measured after 10 and 25 cell generations. These measured distributions were congruent with those predicted by the computer simulations, validating the experimentally determined rates of synthesis and partitioning of the plasmids. The measurements of EBV plasmids were extended to cells under selection. The distribution of genomic EBV in clones of EBV-immortalized B-cells was measured by FISH. These measurements revealed that selection for EBV plasmids yields cells with widely varying numbers of plasmids, but that there is a heritable, mean optimum for a given number for each clone. The measured distributions of plasmids in cells under selection have also been modeled with computer simulations. These combined analyses support a model that reflects the population dynamics of EBV plasmids in proliferating cells. EBV has evolved its replication with minimal viral contributions to balance the gain and loss of its genome in cells to which it provides a selective advantage. We followed the fate of individual EBV plasmids in proliferating cells in order to elucidate their synthesis and partitioning. To do so we first developed and characterized EBV-derived plasmids that can be detected visually in live cells. An EBV-derived plasmid (pLON-33K) containing oriP, 264 copies in tandem of Lac operator (LacO), and a neomycin-resistance gene () was introduced into and established in clones of HeLa cells stably expressing EBNA1 (HeLa-EBNA1). These clones were infected with a retroviral vector expressing wild-type Lac repressor fused to a tandem dimer red fluorescent protein, tdimer2 (wtLacI-tdRFP) and an NLS. Different cells expressed different levels of the fusion protein in their nuclei, producing varied levels of a diffuse background. Plasmids were detected as individual dots by their binding of the fluorescent derivative of the Lac repressor, and confirmed as such by their detection in parallel by FISH. Two individual clones (clone A and B) carrying three to four copies of plasmids on average were isolated and used for further experiments. We confirmed that the EBV-derived DNAs in these clones were maintained extrachromosomally dependent upon EBNA1 by Southern blot analysis, by their loss in the absence of selection, and by their replication being inhibited by a dominant-negative derivative of EBNA1 (). The distributions of the established plasmids in these two cell clones were measured by visualization with wtLacI-tdRFP and by FISH after aphidicolin treatment. Aphidicolin inhibits the replicative DNA polymerases and thereby blocks cells in S phase. Importantly, the distributions of the plasmids in these clones determined by these two different assays ranged from 0 to 30 plasmids per cell and were indistinguishable, validating this visualization of EBV plasmids as a means of enumerating them (). We assayed the number, stability, and distribution of plasmids over time under all the conditions used to visualize them. The presence or absence of IPTG for different times in S, G2/M, or both phases, as well as a double-aphidicolin block had no detectable effect on the maintenance of plasmids or their average numbers per cell (). To characterize the partitioning of an EBV plasmid (pLON-33K) during mitosis in detail, distributions of the plasmids were monitored before and after mitosis in live cells. HeLa-EBNA1 clones carrying the plasmids were synchronized by double-aphidicolin or double-thymidine block in the presence of IPTG and G418. These two treatments efficiently blocked cells at the beginning of S phase. After removal of IPTG and the G418, the cells were released from the second synchronization, incubated, and analyzed twice during 12 h, following S phase at 9 h after release from the block and 3 h later following mitosis (). Only healthy cells as judged by their successfully completing mitosis and having intact peripheries without blebs were studied. The distributions of 370 plasmids in 106 mitoses (1–13 plasmids/cell in G2) were characterized by the measurements of the number of dots and the average fluorescent intensities per pixel of the individual dots following S phase and after the subsequent mitosis (). Of the 370 plasmids examined, 310 appeared in G2 as colocalized pairs (). These 310 colocalized plasmids thus appeared as 155 single signals in G2, which following mitosis yielded 310 signals with intensities close to one-half of that measured in the preceding G2 phase. The remaining 60 plasmids appeared in both G2 and following mitosis as single signals whose intensities decreased only slightly, presumably as a result of photobleaching. Seven of these single plasmids in G2 were the only plasmids in the cell, and were passed onto a single daughter cell. Of the 155 pairs colocalized in G2, 130 (88%) partitioned equally to daughter cells following mitosis; 19 partitioned to a single daughter cell yielding two signals in these cells (the daughter cells receiving these pairs are denoted with asterisks in ); and the partitioning of six could not be determined. The 53 single plasmids accompanying one or more colocalized pairs in G2 partitioned into daughter cells independently of those pairs. For example, 14 cells in G2 had one colocalized pair and one single plasmid. Following mitosis 13 of the resulting pairs of daughter cells had 1 and 2 plasmids, and one pair of daughters had 3 and 0 plasmids. All of the 370 plasmids detected in G2 were found in cells following mitosis, indicating that the rate of loss of EBV plasmids during mitosis is less than 0.3%. Spatial–temporal analysis of 12 colocalized pairs of plasmids during mitosis demonstrated that the plasmids associated with condensed mitotic chromosomes as single signals were segregated at the onset of anaphase, and these pairs partitioned equally to daughter cells. The separated plasmids were frequently localized on the sister chromosomes symmetrically at the end of mitosis (). Wild-type LacI forms a tetramer and can bind to two LacO sites (). It is formally possible that the colocalization of two sister plasmids in G2 reflects their being linked by wtLacI. To test this possibility, we expressed a derivative of wtLacI-tdRFP that can only dimerize and thus bind only one LacO site. Cells were synchronized with aphidicolin, released, and the distributions of 16 plasmids were analyzed before and after mitosis. Of these 16 plasmids, 14 were detected as colocalized pairs (seven signals) in G2, 12 partitioned equally to daughter cells; and two did not (data not shown). The frequency of colocalization of pairs of plasmids bound by a derivative of wtLacI-tdRFP that can only dimerize was the same within experimental error as when the plasmids were bound by the wild-type repressor that can tetramerize (=0.54, Fisher's one-sided exact test), indicating that forming pairs of EBV plasmids in G2 is independent of the repressor's ability to form tetramers. To measure the distributions of plasmids throughout the cell cycle, cells were incubated for 24 h without an imposed synchronization and sampled three times. Cells that rounded up for mitosis yielded daughter cells that adhere and flatten at the beginning of G1. The distribution of the plasmids in these cells and the average fluorescent intensities of the pixels in each dot were measured late in G1, late in S, and early in the subsequent G1 (). The intensities were determined for images from the single z-section having the most intense signal measured without gain, and corrected for background by subtracting the average signal of the surrounding pixels. The intensities in each pixel were less than 6% of the maximum of the 14 bit camera we used and thus in its linear range. A total of 26 of 31 plasmids duplicated in S phase, colocalized as pairs in G2 phase, and partitioned equally during mitosis. Five of the plasmids did not duplicate during S phase, were maintained during G2/M phase, and distributed to one daughter cell. The ratio of the average fluorescent intensities in the plasmids that were duplicated in S phase, colocalized as pairs in G2 phase, and partitioned equally in M phase, increased approximately 1.5-fold during S phase (). On the other hand, the ratio of intensities of the plasmids that failed to duplicate decreased slightly (0.8-fold). These results indicate that 84% of plasmids duplicated successfully, yielded colocalized sister molecules in G2 and early M phase, and usually partitioned faithfully. A total of 16% of the plasmids fail to be synthesized in S phase. This defect in synthesis is the only means by which EBV plasmids are detectably lost from a population of proliferating cells; missegregation, however, can contribute to a daughter cell's failure to acquire a plasmid. None of these 26 colocalized pairs of sister plasmids followed in this experiment missegregated. This level of fidelity is not statistically different from that found in the experiments () using imposed synchronies of 12% missegregation (=0.08, Fisher's exact test). That all 26 colocalized pairs of sister plasmids partitioned without synchrony, faithfully, however, leads us to consider the notion that the imposed synchrony may favor missegregation and thus the 12% is likely an upper limit to this defective partitioning. We predicted the distributions of EBV plasmids for multiple generations in the absence of selection by a computer-based simulation. The distribution of plasmids in a cell clone carried under selection and initially containing 2.0±2.1 copies of plasmids per cell on average was used as starting point for the simulation. Each plasmid in each cell in G1 was given an 84% chance of being duplicated; duplicated plasmids were given an 88% chance of being partitioned equally to two daughter cells and a 12% chance of being partitioned to one daughter cell or the other; unduplicated plasmids were partitioned randomly to either daughter cell. The program was begun with approximately 100 cells having the measured distributions in the analyzed clones and run to simulate 10 and 25 generations. Each run was repeated 40 times with the same initial assumptions, and the 40 results were averaged. We also measured experimentally the distributions of the plasmids in the HeLa clone following removal of selection for 10 and 25 generations. The measured and predicted distributions of EBV plasmids are shown in and closely parallel each other, indicating that the experimentally determined frequencies of duplication and partitioning of the plasmids describe well their fate in proliferating cells in the absence of selection. Similar experiments were performed to test whether the measurements made with pLON-33K in HeLa cells could be extended to intact EBV plasmids in a different host cell, 293. EBV confers a selective advantage on the B-cells it infects, but not on established cells such as 293. Intact EBV plasmids can, however, be maintained in established cells under selection if they encode a drug resistance gene (). The distribution of full-length, genomic EBV plasmids encoding resistance to hygromycin B in a clone of 293 cells was measured with FISH, the hygromycin B removed, the cells propagated, and the distributions of plasmids again determined after 10 and 25 cell doublings. In parallel, a computer-based simulation was run, with the distribution of plasmids determined by FISH before removal of the hygromycin B as the starting point. The rates of duplication and partitioning of plasmids measured in HeLa cells were also used in these simulations. The results predicted by these simulations paralleled the measured distributions at 10 and 25 cell generations (), and indicate that the rates of synthesis and partitioning measured for pLON-33K in HeLa-EBNA1 cells are similar to those of intact EBV in 293 cells. When this program was used in simulations with the same starting distribution of EBV as in but with different, assumed rates for its synthesis or partitioning, the predicted distributions differed () from those measured (). The sensitivity of the simulations to changes in these rates demonstrates the robustness of this computational approach. We determined the distribution of EBV plasmids in the EBV-transformed lymphoblastoid cell clone 721, which requires EBV for its survival and proliferation (). The parental cell population (), which has been propagated for more than 100 generations, was synchronized at G1/early S phase or M phase by treatment with aphidicolin or nocodazole, respectively. The distribution of EBV plasmids in the cell population was analyzed by FISH. The numbers of the viral plasmids in the individual cells varied widely in the population. The measurements of the distribution of the viral plasmids are summarized in . The distribution was broad, with the mean number of plasmids centered on 9.7±4.1 and 19.5±6.3 in aphidicolin- or nocodazole-treated cells, respectively. Four independent subclones were isolated from the parental 721-cell population by limiting dilution, propagated for 25 generations, and their distributions of viral plasmids determined by FISH. The distributions in all four subcloned populations were similarly broad, with the average number of plasmids centered on 6.7–9.6 plasmids in aphidicolin-treated cells, and 11.9–15.1 plasmids in nocodazole-treated cells (), indicating that the broad distribution of viral genomes in 721 cells was generated within only 25 generations. The constancy of these mean numbers of plasmids indicates that this mean number of EBV genomes in 721 cells provides this clone and its descendants an optimal selective advantage. We modeled the behavior of EBV plasmids in 721 cells in order to determine if the frequencies of plasmid duplication and faithful partitioning measured for EBV plasmids in the absence of selection apply to those in 721 cells in which EBV is required both for the survival and proliferation of these cells. In order to model this behavior, it was necessary to understand the relationship between the number of viral genomes in a cell, and the selective advantage those genomes provide it. Two experimental measurements were used to define a cell's dependency on an EBV-derived plasmid for its survival. First, the rate of loss of plasmids from HeLa and 293 cells from which drug selection was removed was measured directly to be 6.7% per generation (). Second, in related, historical experiments in which a selection was removed and after time reapplied, the rate at which cells lost sufficient EBV-derived plasmids to become sensitive to being killed by reapplication of the selective agent has been found to be 3–5% per generation (; ). That these two rates differ indicates that cells dependent on an EBV-derived plasmid lose their selective advantage when their number of plasmids drops below some non-zero, threshold number. Loss of plasmids in cells above this threshold number would not affect the selective advantage conferred on the cell. Because cells dependent on EBV-derived plasmids have a broad distribution of plasmids per cell (); more cells can lose some plasmids per generation than succumb to the selective disadvantage. These considerations allowed us to model the behavior of EBV plasmids with our simulation, based on the initial distribution of the number of EBV genomes in individual cells of the 721 parental population. A simulation was made with the rates for duplication and partitioning found for pLON-33K in the HeLa clones. In this simulation the initial conditions were for a single cell with a specified number of plasmids. The simulation was repeated for the number of times a cell with that given number of plasmids was found in the measured parental population of parental 721 cells (), and repeated for all given numbers in that parental population. These initial conditions thus mimic the cloning of the parental 721 cells that yielded the four subpopulations. An additional parameter was added to simulate a threshold such that five plasmids per cell yielded 100% survival; cells with four plasmids had an 80% chance of survival; the chance for cells with three plasmids was 60%; with two was 40%; with one was 20%; and with none was 0%. We used this parameter of proportional survival because an abrupt threshold with 0% survival yielded an unphysiological distribution of no survivors beyond that threshold. We selected this particular parameter because test simulations in which the threshold for 100% survival was set either to 4 or 6 plasmids per cell, generated distributions with peak values that were either less than or greater than that measured for the average of the 721 subclones (). The results of the simulations were averaged over 40 trials and are shown relative to the average of the distributions measured for the four subclones of 721 cells grown for 25 generations (). The congruence of these measured and simulated distributions is consistent with the rates measured for the duplication and partitioning of EBV-derived plasmids in live HeLa cells applying to EBV genomes in B-cells transformed by EBV. This conclusion was being further substantiated by testing the dependence of this simulation on the rate of plasmid duplication of 84% per cell cycle, derived from measurements of pLON-33K in HeLa-EBNA1 cells by arbitrarily substituting values of 74 and 94% for it, and running the simulations again (as for intact EBV in 293 cells in ). Decreasing or increasing the rate of plasmid duplication used in the simulation shifted the center of the predicted distributions to the lower or higher values, respectively, than that using 84%, which does center on the average of the measured distributions (). EBV uses plasmid replication to maintain itself in proliferating cells. This plasmid replication underlies its pathogenicity in cancers such as Burkitt's lymphoma, where the viral plasmids are present in the proliferating cells characteristic of these diseases; but few viral genes are expressed (; ). We have analyzed the synthesis and partitioning of EBV plasmids in order to shed light on the details of this viral plasmid replicon. One unexpected property of this replicon is the frequency with which it fails to duplicate; 16% of EBV plasmids remain unduplicated in each cell cycle. This frequency is initially surprising, given that EBV's origin of plasmid replication P was identified by an ARS assay in mammalian cells, is a licensed replicon, and is the most efficient mammalian licensed ARS characterized yet (). However, we have found that polymerizing a licensed replicator, Rep, which functions as does DS, but is less efficient, increases Rep's frequency of duplication (; ). There is thus no reason now to expect that DS of P is maximally efficient and is always duplicated in each cell cycle. A second fundamental property of EBV's plasmid replicon uncovered here is the colocalization of sister plasmids on being synthesized, and that this positioning in space is coupled to their non-random partitioning. This colocalization of newly synthesized sister plasmids has not been detected before, probably because detection of established EBV plasmids has been carried out with FISH, which requires denaturation and spreading of DNA, conditions that would be expected to dissociate the tethered complexes. For example, approximately twice the number of EBV plasmids is detected by FISH in clones of cells blocked with nocodazole as with aphidicolin (). Pairs of EBV plasmids are also frequently found in mitosis to be symmetrically positioned on sister chromatids, when detected by FISH () () or by binding tagged EBNA1 (). In 10 mitotic spreads, for example, 60% of the EBV plasmids were detected as colocalized pairs and 97% of all plasmids were overlapped by chromosomal staining (examples shown in ). These measurements are consistent with EBV plasmids being associated with sister chromatids and the spreading forces used to separate mitotic chromosomes also acting to separate some of the colocalized sister plasmids. This colocalization of sister plasmids in conjunction with EBNA1's ability to tether DNA such as FR to which it binds site specifically to chromosomal DNA leads to a plausible model for the mechanism of EBV's partitioning () as follows: in G1, an EBV plasmid is tethered to a chromosomal site probably directly by EBNA1's AT-hook activity. We favor this mechanism rather than the alternative of indirect tethering by EBNA1's binding the cellular protein, EBP2, because derivatives of EBNA1 that cannot bind EBP2 but have AT-hook activity behave as wild-type EBNA1 in supporting replication. We have found that a fusion of the cellular protein, high-mobility group A (HMGA) 1a, to EBNA1's DNA-binding and dimerization domain, which deletes EBNA1's binding site for EBP2 (), or a second derivative of EBNA1 that cannot bind EBP2, 2XLR1 (; ), supports wild-type levels of plasmid synthesis and partitioning (). Both derivatives do have AT-hook activity (); during S phase, the chromosomal site and the tethered EBV undergo synthesis synchronously by their spatial colocalization, placing them in the same replication compartment (; ). This synchrony needs be only within an hour or so, though; synthesis through FR is slow, leading it to being scored as a ‘replication fork barrier' that protracts its own synthesis to be longer than that required to synthesize 160 000 bp of EBV unidirectionally (); newly synthesized EBV sister plasmids bind to the same nearby sites on each sister chromatid, or to the original site if it can accommodate them both. Our measurements indicate that sister plasmids individually associate with sister chromatids 88% of the time. Cohesin holds the sister chromatids together () and we hypothesize indirectly holds the tethered EBV plasmids together within the resolution of fluorescence microscopy. EBV plasmids are found in association with chromosomes early in prophase and embedded in mitotic chromosomes detected by staining for cohesin consistent with this hypothesis (); when the sister chromatids separate at anaphase, the tethered EBV sister plasmids do so too. The third fundamental property revealed in these studies pertains both to EBV-derived plasmids and their host cells. When these plasmids provide their host cell a selective advantage, the number of plasmids in a cell that affords the cell an optimal advantage is a clonally inherited trait. Four subclones of an EBV-immortalized B-cell generate wide distributions of plasmids per cell, but each clonal population retains a similar mean number of plasmids per cell (). This finding is supported by experiments in which the average number of P plasmids per cell was determined by nucleic acid hybridization, hygromycin B or G418, to which the plasmids encoded resistance was removed for 20–30 cell generations, the drug was reapplied, resistant subclones isolated, and the average number of plasmids measured in them (). In these experiments, subclones from five of six different parents had within experimental error the same average number of plasmids as did their parents. This property likely underlies the variation observed in the average number of copies of EBV between different cell lines (), and provides a selection for a given mean number of plasmids per cell in a clonal population of cells. The fourth unexpected, fundamental property of EBV plasmids identified here is that the defects in duplication of these plasmids are approximately balanced by a defect in their partitioning. In each generation, 16% of the plasmids fail to be duplicated, leading to a net loss of 8% of the plasmids from a population of cells. Each generation 12 of the 84% of duplicated plasmids partition unequally, leading to approximately half of these daughter cells (5%) having more plasmids than their parents. Clearly some daughter cells also have fewer plasmids than their parents, and if they have fewer plasmids than their threshold number, they will be at such a selective disadvantage as to be lost from the population. This rate of loss is 3–5% of the cells per generation, while the remaining cells that retain sufficient plasmids can be afforded selective advantages such as a fostering of their survival and proliferation as, for example, in the case of B-cells infected by EBV. The rates of synthesis and partitioning of EBV-derived plasmids are surprisingly efficient when compared with those of derivatives of human chromosomes. These rates for linear replicons engineered from the human X-chromosome are inversely proportional to their length (). Derivatives of the human X-chromosome that are fives times the length of EBV's genome and 25 times that of pLON-33K are lost from human cells in culture at rates of 3–4% per generation, and missegregate at rates up to 16% per generation (). That EBV-derived plasmids have similar rates and are so much smaller in length indicates that the virus has evolved a particularly efficient mechanism to persist in proliferating cells. EBV has evolved a simple, flexible scheme to maintain itself in a proliferating population of cells to which it provides a selective advantage. EBV itself drives proliferation of newly infected B-cells through its LMP1 protein (; ; ); it sustains frank Burkitt's lymphomas by preventing apoptosis and driving proliferation requiring its EBNA1 protein (). These selective advantages when coupled to its mode of synthesis and partitioning, insure that EBV is maintained in infected normal and malignant proliferating B-cells. #text

Many human diseases are caused by or resulted in an abnormal metabolic state such as the high glucose concentration in blood of diabetes patients and the high urine amino-acid level resulted from liver or renal disorders. Metabolic processes are also heavily involved in xenobiotics degradation and drug clearance (). Drug safety is often linked to inhibition of metabolic processes (). Detecting the unusual level of certain specific metabolites in a patient's blood or urine has long been established as an effective method to identify biomarkers for diagnosing particular diseases (; ). Recent rapid developments of advanced metabolomics technology is opening up new horizons, as hundreds or even thousands of metabolites can be measured simultaneously, providing a much more comprehensive assessment of a patient's health status (; ; ). However, for a better and in-depth understanding of the large amounts of data generated from metabolomics, a complete and high-quality human metabolic network is essential. This network links various metabolites by enzyme catalyzed reactions and thus allows us to discover the genetic mechanism, which causes the abnormal state of metabolites by network analysis, and further kinetic modelling. Then drugs, which could target on so far uncharacterised genes/proteins can be developed for disease treatment. Although some human metabolic pathways such as glycolysis and urea cycle has been discovered almost a hundred years ago and many pathways have been extensively studied in biomedical journals and textbooks, a complete whole picture of human metabolic network is still missing. Especially the recent development of genome technology requires us to reconstruct the whole network from genome level to better understand the genetic basis of metabolic organization and regulation. A preliminary metabolic network can be computationally reconstructed easily from the gene annotation information (; ; ). Actually such computational reconstructed networks are available from several metabolic databases such as KEGG and HumanCyc (; ). However, the quality and completeness of such networks are often not very high. They need to be carefully investigated and calibrated through the experimental results reported in literature. This is a very labor-intensive and time consuming (somehow endless) manual process. Even for simple microorganisms such as and , the high-quality metabolic networks reconstruction often take years to finish (; ). In contrast to these microorganisms, which can use only a few substrates such as glucose to synthesis all the metabolites, human (and other animals) requires many essential nutrients for growth. Moreover, human is a multicell, multi-tissue organism with complex networks of interactions between them. The functions of the cells, tissues and organs are well differentiated and the metabolites are transferred within the body through the blood circulation system. Therefore, the human metabolic network is much more complex than those of microorganisms, and has different structural and functional features, which may be representative for higher organisms, especially animals. In this paper we report our ongoing work on human metabolic network reconstruction. We have combined genome reconstruction with reconstruction based on literature to obtain a high-quality human metabolic network with more than 2000 metabolic genes and nearly 3000 metabolic reactions (referred as EHMN: Edinburgh Human Metabolic Network, in the following sections). It allows us to have a coherent picture that could be used in different studies as a reference. To better understand the functional organization of the network, we have reorganized the enzyme reactions into about 70 pathways according to their functional relationships. We further compared our network with other available human networks such as HumanCyc () and the recent reconstruction by Palsson's group (). A bow-tie connectivity structure is rediscovered from a functional rather than structural point of view, and the distribution of disease related metabolic reactions in the bow-tie structure was investigated. The main processes for the reconstruction of the human network are shown in . The first step is the reconstruction of the network solely based on the human gene annotation information. This network is called the genome-based network. This step can be automated and thus the genome-based network can be easily updated with the new annotation information in the databases. Unfortunately, the human gene contents in online databases are often very different. For example, there are more than 38 000 human genes in the NCBI EntrezGene database (), but only about 27 thousand in HGNC (). Moreover, about 3000 genes in HGNC are not in EntrezGene. Therefore, it is very important to integrate information from different databases to get a more complete enzyme gene list for the reconstruction. In our reconstruction, we mainly obtained the enzyme annotation from KEGG, Uniprot and HGNC. Information from NCBI EntrezGene (), Ensembl () and Genecard () databases are also included to provide more complete crosslinks between gene (protein) IDs in different databases and validate the enzyme gene annotation. Special attention was paid on the genes with unclear EC numbers such as 1.-.-.-. The existence of the unclear EC numbers is because only a chemically well-characterized enzyme is assigned an EC number by IUBMB. In the post-genome era, this process is far behind the function annotation of genes, which is mainly based on the DNA sequence. For example, in the UniProt database there are more than 800 proteins annotated with an unclear EC numbers (). For these genes we could not get the reactions catalyzed by its encoded enzymes through the EC numbers. The reactions can only be added directly from the function annotation part in Uniprot and many genes need to be manually examined in literature. Another problem in the reconstruction of the genome-based network is the somehow ambiguous relationships between EC numbers and reactions in the reaction databases. A protein in human may not catalyze a reaction, which is catalyzed by a protein with same EC number in other organisms. For example, the GBA3 gene in human codes for cytosolic beta-glucosidase, which have an EC number 3.2.1.21, whereas in other organisms, proteins with this EC number also function as cellobiase catalyzing the degradation of cellulose. This degradation reaction apparently does not occur in human. Unfortunately there is still no automatic way to obtain the human-specific EC number–reaction relationships. We first used the KEGG ligand database to generate the reaction list, because it is one of the most complete metabolic reaction databases including more than 7000 reactions (). Then the reactions were manually checked to exclude non-human reactions. The second step of the reconstruction is to refine the genome-based network based on information from literature. Fortunately, thanks to the EMP database (personal communication with EMP projects Inc.), we already have a literature based human metabolic network available. The EMP network is a compound centric network reconstructed mainly based on information from literature. It contains more than a thousand compounds and nearly 2000 reactions. We then can compare and integrate the two networks together to obtain a more complete and high-quality human metabolic network. However, this integration process is very time consuming mainly due to the different compound and reaction nomenclature systems used in the two datasets. The reactions and compounds in the genome-based network are mainly from the KEGG ligand database, while EMP has its own nomenclature for reactions and compounds. In order to check if two reactions from the two networks are the same, we need to check if all the compounds in the reaction equations are the same. A straightforward way to match the compounds in different databases is to compare them by name. However, even though both KEGG and EMP databases have a synonym list for each compound, the total number of matching compounds is only about 500 of the more than 2000 compounds in the two networks. This is a surprising result considering that both are human metabolic networks. We have tried different methods to obtain more compound matching relationships such as using synonyms in other compound databases such as PubChem () and ChEBI () to match the compounds in EMP and KEGG, matching compounds by structure and allow fuzzy matching between a generic compound and their specific compounds (for example -glucose and alpha -glucose). Unfortunately only about 300 new matching relationships were obtained through these methods. Based on the matched compounds, we found about 700 matching reactions between the genome-based network and the EMP network, by checking if two reactions have the same reactants. These matching reactions allow us to compare the two networks at a higher pathway level. If a pathway in EMP contains one or more matching reactions with a KEGG pathway, then these two pathways will be functionally linked. We can then compare the two pathways to see if they have some unfound matching reactions or a reaction in one pathway complements a gap in another pathway. This pathway consolidation process can only be done by manual examination of the reactions in the pathways and visual inspection of the pathway maps. However, it is an important step to improve and maintain the quality of the reconstructed network, and for us to better understand the biological function of the large-scale network. As described above, the pathway organization of reactions is an important step in the network reconstruction. However, in comparing the pathways in EMP and KEGG, we found they are organized very differently. In EMP there are more than 300 metabolic pathways for the human metabolic network. Almost 100 of them are very small, containing three or fewer reactions. Moreover, in the pathway maps there is no link to other pathways shown. Therefore, it is very difficult to gain a whole picture of the human metabolic network from so many small pathways. In KEGG, all the reactions from different organisms are organized into about a hundred metabolic pathways. The problems with the KEGG pathways are the following: it is not human specific. In certain pathways only one or two isolated human reactions exist; there is a high overlap between pathways. For example, there are many overlap reactions in the pathways of glutamate metabolism, urea cycle, arginine and proline metabolism. Grouping these pathways as a big pathway would show the functional relationships between the reactions in them much better; the mass flow between a substrate and a product in the pathways is not as clear as those in EMP or Biocyc (). To address these problems, we decided to define a new set of human-specific pathways which are human specific; less overlap between pathways; large enough to include functional related small pathways; including links to other pathway to get a better overview of functional connectivity between pathways. Basically, the small pathways in EMP and KEGG were grouped on the basis of their functional relationships. Some original pathways may also be separated into different new pathways. Altogether, 2823 reactions are included in the network and reorganized into 66 pathways, with the number of reactions between 5 and 142 (there are also more than 300 isolated reactions in the network). The retinol (vitamin A) pathway is shown in as an example. There are much more reactions in our pathways than in the corresponding KEGG pathways. More pathway maps and the whole set of pathways in SBML format can be seen in the (the and , the network and the pathways in SBML format are also available at ). The users can directly open the SBML files in CellDesigner () or other softwares to generate an automatic layout for the pathways. As described in previous studies, the currency metabolites often cause trouble in graph layout of metabolic pathways because they tend to link all the metabolites in a short path. Therefore in the SBML files for the pathways we include only the main compounds in the ‘listofreactants' and ‘listofproducts' section. This makes it possible to quickly generate clear and nice pathway maps from the SBML files. In the process of pathway reorganizing, we noticed that many reactions, especially those related with complex lipid metabolism, are missing in the genome-based network, where the reactions are mainly from KEGG ligand database. For example, the reactions related with omega-3 and omega-6 fatty acid (two essential nutrients for human) metabolism, mono-unsaturated fatty acid metabolism and epoxyeicosatrienoic acids (EETs) metabolism are almost completely missing in KEGG. Due to the great structural variance of complex lipids, the total number of lipid metabolites is more than 8000 () and most of them exist in the human metabolic network. Therefore, even though we already added many lipid pathways from literature, the network is still far from complete. A comprehensive database on lipids and their relating enzymes in various organisms has been developed by the LIPID MAPS Consortium (; ). Based on information in this database and other resources, more lipid-related pathways can be added in the future version of our database. We further compared our network with another computationally reconstructed network in HumanCyc (version 10.6) (). There are 996 reactions in the database, and among them 766 are catalyzed by enzymes. This is only half of the number of reactions in our database. We extracted 976 EC numbers from HumanCyc and compared them with those in our database. We found 151 EC numbers are in HumanCyc but not in our database. We then checked the reactions and proteins corresponding to these EC numbers, aiming to add new reactions to our database. Surprisingly, we found that 116 of the 151 new EC numbers were without any coding gene, but added by the pathway hole filling algorithm used in Pathologic method for the computational reconstruction of metabolic networks in Biocyc (). However, many of them are in pathways where many reactions are without any gene. For example, in the dTDP--rhamnose biosynthesis I pathway, only the reaction catalyzed by 4.2.1.46 is encoded by a human gene. The other three reactions catalyzed by 1.1.1.133, 2.7.7.24 and 5.1.3.13 are all added to complement the pathway. There is even no literature related with these reactions in human. Therefore, we decided not to include these reactions in our network. For the other 35 EC numbers, we examined their corresponding genes and checked how these genes are annotated in other databases and literature. We found that 24 EC numbers unique in HumanCyc are because of wrong annotation of the genes in HumanCyc. For example, among the three genes encoding 2.4.1.87, gta actually functions as a galactosyltransferase activator, CDC2L2 is a galactosyltransferase-associated protein kinase, ENSG00000165196 has already been removed in the latest ENSEMBL database (). For the other 10 EC numbers, four have no reaction or a protein modification reaction, which is currently not included in our network. Therefore, we only need to add reactions for six EC numbers from HumanCyc. Actually some of the reactions are already in our reconstruction, but with a different EC number. Altogether nine reactions were added from HumanCyc. A complete list of the manually examined EC numbers unique in HumanCyc can be seen in . The comparative analysis between our network and HumanCyc indicates from one aspect the importance of integrating information from different databases for network reconstruction, and from another side, the importance of human curation for improving the quality of the computationally reconstructed network. During the review process of the paper, another high-quality human metabolic network reconstructed by Palsson's group (referred as HMN-P below) was published (). We obtained their data from the BiGG database and compared with our network (EHMN). At the gene level, EHMN contains 2322 genes from different databases, HMN-P contains 1496 genes mainly from EntrezGene (actually all genes have EntrezGene ID). The common part is 1069 genes. At enzyme level, in HMN-P only less than half of the genes are assigned EC numbers (total EC numbers less than 500 including unclear EC numbers). In EHMN, all the genes have clear or unclear EC number because we start the reconstruction from such genes. The total number of ECs is more than 800 (excluding unclear ECs). One may argue that in HMN-P ECs are not used to link genes with reactions. However, as a widely used standard for representing metabolic reactions, introducing EC number in the network can greatly simplify the comparative analysis of metabolic networks for that the direct comparison of reaction equations is very difficult due to compound synonyms. At the metabolite level, EHMN contains 2671 compounds, and 1769 of them can be found in KEGG database. HMN-P has 1469 compounds and about a half of them linked to KEGG. For the non-KEGG compounds, in HMN-P only one compound name is given. This makes it difficult to find a matching compound in other databases. As stated previously, we have developed a compound database with synonyms, structure information and IDs in different databases (in ). At the reaction level, HMN-P contains more reactions than EHMN (3731 versus 2823). However, there are 1189 transport reactions and 457 exchange reactions, which are not considered in EHMN because the subcellular location information is still not included. Furthermore, there are 290 repeat reactions in HMN-P, which are the same reaction but in different compartments. Therefore, the number of reactions comparable with EHMN is just 1795. Because of the intrinsic complexity of human cell, it is very difficult to place the reactions into a small number of compartments. Actually we have collected protein location information from different databases and have identified hundreds of cellular locations. We are working on it to develop a GO (Gene Ontology) ()-based hierarchically compartmented human network for the next release. Unlike most microorganisms, which can use simple substrates to produce all the metabolites required for its growth, human requires many essential nutrients obtained directly from the food to maintain a health physiological state. The typical essential nutrients include 10 amino acids, omega-3 and omega-6 fatty acids and various vitamins. To verify the essentiality of these metabolites, we manually examined the pathways related with them and found that there is no pathway for the synthesis of these metabolites from the central metabolites. For a more systematic analysis of the metabolic capacity of the human metabolic network, we start from the central metabolites (in glycolysis, pentose phosphate pathway and TCA cycle pathway) and classify other metabolites as exchangeable (pathways for both the synthesis from central metabolites and the degradation to central metabolites exist), degradable (only degradation pathways exist), synthesizable (only synthesis pathways exist) and isolated (no pathway from/to central metabolites) based on pathway analysis. We found that many of the essential nutrients fall into the isolated metabolites and some essential amino acids are degradable. This metabolite classification is quite similar with the bow-tie structure of metabolic networks discovered previously based on graph analysis (). In the bow-tie structure, all the metabolites in the giant strong component (GSC) can convert to each other (equal to the exchangeable metabolites), the metabolites in the IN subset can convert to metabolites in GSC (degradable), those in the OUT subset can be produced from metabolites in GSC (synthesizable) and those in the isolated subset are not connected with GSC. Here we rediscovery the bow-tie structure by functional analysis of the metabolic network. Because graph is a simplified representation of metabolic networks, which lose some structure information (for example a multiple substrate multiple product reaction just be represent as several links with one substrate and one product), if the bow-tie structure found by graph analysis represents the true biological organization principle of metabolic network is still an open question. Here, based on a functional analysis of the high-quality human metabolic network, we confirmed the bow-tie structure from a biological point of view. Due to the simplification in the graph conversion of a metabolic network, the exact position of certain metabolites in the bow-tie may be different for the two different approaches. However, as a system level macroscopic structure, the bow-tie is true from both structural and biological aspects. This fact strengthens the hypothesis that bow-tie is an important organization principle for complex systems to be robust and flexible (; ). Based on the classification of the metabolites, the reactions can also be classified into four subsets forming a bow-tie structure similar to the bow-tie of the reaction graphs of metabolic networks (). The reactions occur between the exchangeable metabolites are in the GSC. The reactions in a pathway from the degradable metabolites to the exchangeable metabolites form the IN subset. Correspondingly, the reactions in a pathway from the exchangeable metabolites to the synthesizable metabolites form the OUT subset. All the other reactions are in the isolated subset. The bow-tie classification of the reactions is more interesting because the reactions are linked to the genes and proteins, which are the main functional regulation units in the cell. The full classification of the reactions can be seen in the . For the human network, we found that the scale of the isolated subset is often very large (more than one fourth of the whole network). We investigated the metabolites and reactions in the isolated subsets and found that many of them are actually not truly isolated, but can be produced from metabolites in the IN subset and may have important physiological function. As mentioned previously, human requires many essential nutrients for growth. These nutrients are essential because they are used for producing certain metabolites with important physiological functions. For example, the aromatic amino acids are precursors for monoamine hormones and neuron transmitters. These signal metabolites can bind to different protein receptors and then regulate the amount and activity of proteins to change the physiological state. Because many such signal molecules are produced from essential nutrients, which are in the IN subset of the bow-tie, they should be in the isolated subset of the bow-tie structure. To distinguish these metabolites from the real isolated metabolites, we generate a new subset called ‘OUT2' to include the metabolites synthesized from the metabolites in the ‘IN' subset. Correspondingly, an ‘IN2' subset, which contains the metabolites for producing metabolites in ‘OUT' is also added. Therefore a six-subset modified bow-tie structure of the human metabolic network is produced as shown in . The numbers of reactions in the six subsets are shown in . A main objective of human metabolic network analysis is to see how it is related with human disease. More than 10 000 human genes (half of the whole genome) have been reported to be related with one or more human diseases in the OMIM database (). In the human metabolic network, we found that 2215 (of 2823) reactions are catalyzed by enzymes, which are coded by disease related genes. If we exclude the reactions, which are spontaneous or with an unknown gene, the proportion of disease related reactions is even higher at 95% (2215 of 2314 reactions with encoding genes). This surprisingly high value raises a question to us: what does network robustness really mean from a biological aspect? As the most complex organism on the earth, human is expected to have a very robust metabolic network. Actually previous studies on biological network truly suggest that as scale free networks, metabolic networks are robust against random errors from a structural point of view (). However, the result that most reactions are linked with disease genes indicates that the human metabolic network is fragile. We analyzed the distribution of the disease related reactions in the bow-tie structure and the result is shown in . Interestingly the proportion of disease-related reactions in the IN subset is much less than that in the OUT subset, implying that the reactions leading to metabolic products are more fragile than the reactions for the degradation of various substrates. This result looks unusual but understandable from biochemistry. The key function of the central pathways (glycolysis and TCA cycle) is to produce energy and precursors for biosynthesis. Most of the bioproducts can be synthesized from a number of common metabolite precursors in the central pathways, and most of the substrates are also converted to these precursors first for further conversion. Therefore, theoretically one substrate is enough to produce all the necessary products, if it can be converted to these precursors. The existence of multiple pathways for multiple substrates just provides more flexibility to the organism and thus blocking one pathway is unlikely to damage the organism. In contrast, a product is synthesized in an organism often because it has some unique function important for the organism (as a structure molecule or a signal molecule). Hence, a failure in a product synthesis pathway can make the whole system organized improperly, causing a disease. One step further, we may hypothesize that the organization of the bio-products and their synthesis pathways from the common precursors rather than the substrates and their degradation pathways determine the feature of a biosystem. Back to the bow-tie structure, the metabolites and reactions in the Out subset may better define a biosystem than the metabolites and reactions in the IN subset. Further studies on comparative analysis of the metabolic networks of different organisms are needed to validate this hypothesis.

Living cells are self-regulated by interactions between different molecules. Until very recently, most research has focused on transcription regulation interactions and on protein–protein interactions, which in many cases are involved in post-translational regulation. During the last years it has become evident that another type of interaction plays a prominent role in the regulation of cellular processes, manifested by small RNA (sRNA) molecules that base pair with the mRNA and regulate gene expression post-transcriptionally. This mode of regulation was found in both pro- and eukaryotes (for review see ). Although there are differences in the characteristics of the eukaryotic and prokaryotic regulatory RNAs and in the fine-details of their mechanism of action, both exert their regulatory function mostly by base pairing with the mRNA and influencing translation or mRNA stability. It is intriguing to study the properties of this type of regulatory interactions in comparison to the other types of interactions, and to understand their integration in the cellular circuitry. In this paper we focus on bacterial sRNAs, and particularly on regulatory interactions found in , for which most experimental data on sRNAs are available. At present there are about 80 known sRNAs in (for review see ; ). These molecules are 50–400 nucleotides long and many of them are evolutionary conserved (), hinting to their important roles in the cellular mechanisms. Still, for many of the sRNAs, their cellular and molecular functions have not yet been determined. Many of those, for which some functional knowledge has been acquired, were often shown to act as inhibitors of translation by base pairing with the mRNA in the ribosome-binding site (for review see ). However, in there are also a couple of examples where the sRNAs play a role as translational activators, promoting ribosome binding to the mRNA by exposing its binding site (, ; ). In many cases the sRNA–mRNA interactions are assisted by the RNA chaperone Hfq (). The acknowledgment that post-transcriptional regulation by sRNAs is a global phenomenon has raised many interesting questions and speculations regarding their roles in the cellular regulatory networks. It was suggested that it would be cost-effective for the cell to use this mode of regulation, because these molecules are small and are not translated, and therefore the energetic cost of their synthesis is smaller in comparison to synthesis of regulatory proteins (). The ease of synthesis led to the suggestion that it would be beneficial for the cell to use these molecules for quick responses to environmental stresses. In this paper we describe this regulatory mechanism by dynamical simulations, and analyze quantitatively these intuitive conjectures. Furthermore, we compare the properties of post-transcriptional regulation by sRNA–mRNA base pairing to those of transcriptional regulation by protein–DNA interaction and post-translational regulation by protein–protein interaction. We show that there are measurable differences between the three regulation modes and describe the situations when regulation by sRNA is advantageous. The interactions between molecules within the cell can be described as a network in which nodes represent genes (or their products) and edges represent the interactions between them. Recently, a considerable effort has been put in deducing the structure of these networks from experimental data, aiming at a systematic understanding of regulation mechanisms and cell function (; ; ). Here we describe the network of post-transcriptional regulation by sRNAs in , where nodes represent either sRNA genes or their targets, and edges point from sRNA genes to their targets. By integrating this network with the transcription regulation network, we discover intriguing regulatory circuits involving both transcriptional regulation and post-transcriptional regulation. The different properties of transcription regulation and regulation by sRNAs have important implications in these mixed regulatory circuits. We demonstrate this by comparing analogous feed-forward loops that are either composed of transcription regulation or involve also regulation by sRNA. We analyze different types of regulation of gene expression mediated by three different interaction types, protein–DNA, protein–protein and sRNA–mRNA. To this end we described the regulatory mechanisms involving these interactions by mathematical models, followed by simulations, using average kinetic parameters based on experimental data (; ; ). We distinguished between two scenarios. In the first scenario, we assumed that the products of both the regulated gene (target) and the regulator are already present in the cell when an external signal turns on the regulation. In the second scenario, the target protein is already present when an external signal turns on the synthesis of the regulator. For both scenarios we compared the kinetics of regulation mediated by protein–DNA, protein–protein or sRNA–mRNA interaction. We describe in some detail the modeling of regulation by sRNA. Let the sRNA transcription rate be (molecules/second), and the target mRNA transcription rate be (molecules/second). The target mRNAs are translated into proteins at a rate . The degradation rates are , and , for the sRNAs, mRNAs and proteins, respectively. The sRNA base pairs with the target mRNA at a rate α. The base pairing blocks the binding of the ribosome to the mRNA, thus negatively regulating translation. This system is described by the following rate equations: where , and are the number of sRNA, mRNA and protein molecules per cell, respectively. In the analysis below, these equations are solved by direct numerical integration starting from suitable initial conditions, as specified. A similar model was recently used for the analysis of regulation by the sRNA RyhB (). Analogous equations are used in the analysis of transcriptional regulation by protein–DNA interaction and post-translational regulation by protein–protein interaction. The parameters used in the simulations are based on experimental measurements in (; ; ). The transcription rate of mRNAs was taken to be =0.02 (molecules/second). Based on the high abundance of sRNAs, we assumed an average transcription rate of =1 (molecules/second), 50 times faster than that of mRNAs. The high abundance of sRNAs may be due to duplicated copies of their genes (), strong promoters or high stability (). This difference in transcription rates is supported by experimental results obtained with (). The translation rate was taken as =0.01 (s). The degradation rates for sRNAs, mRNAs and proteins were taken as =0.0025, =0.002 and =0.001 (s), respectively. The rate constants for binding of sRNA to mRNA, regulatory protein to promoter and protein to protein were all taken as α=1 (s/molecule). It should be noted that we ran the simulations for a range of biologically relevant parameters around these average values and obtained similar conclusions. In we present for each regulation type the level of the target protein versus time, starting from the time at which the regulation is turned on. At time =0, a sudden change in the external conditions turns on the regulation. In case of transcriptional regulation, the regulatory protein binds to the promoter of the target gene and represses its transcription. In case of post-translational regulation mediated by protein–protein interaction, regulator proteins bind to the target proteins and form complexes, which do not exhibit the activity of the free target proteins (they may be degraded, as in the case of σ, which is targeted to degradation by the binding of DnaKJ proteins; ). In case of post-transcriptional regulation by sRNA, the sRNA molecules bind the transcripts of the target gene and inhibit their translation. In these simulations it is assumed that the complex of regulator and target molecules does not dissociate back to its original components (). We discuss below the case in which such dissociation takes place, and its effects. The two panels in differ in their initial conditions. In both the regulator and the target are already present in the cell when the regulation is turned on due to some external stimulus. In the regulator is initially absent and is produced due to an external stimulus, while the target gene is expressed independent of the stimulus. The first scenario may be regarded as turning the regulator on by a conformational change exerted by the external stimulus (e.g., phosphorylation of OmpR by EnvZ under high osmolarity; ). In the second scenario, the regulator's synthesis is turned on following the stimulus (e.g., induction of synthesis of the sRNA OxyS by OxyR under oxidative stress; ). When both the regulator and the target are present in the cell, protein–protein interaction provides the fastest response to the external stimulus (). The regulator proteins are available to carry out the regulation and they quickly bind to the target proteins and suppress their activity. When the regulation is mediated by sRNA–mRNA base pairing, the sRNA molecules quickly bind to the mRNA molecules and prevent their translation. However, the already present target proteins are active until they degrade. As a result, the regulation by sRNA results in a slower response than that exerted by protein–protein interaction. In case of transcriptional regulation, the regulatory protein binds the promoter of the target and represses its transcription. However, the target proteins that are already present are active until they degrade. Moreover, already transcribed mRNA molecules continue to be translated into proteins until they degrade too. As a result, transcriptional regulation leads to the slowest response. We now turn to analyze the second scenario, in which the regulator is produced in response to the external signal while the target protein is already present. In case of transcription regulation, the regulation process remains virtually the same as in and even slower. This is because at the time of the stimulus the regulatory protein is absent and needs to be transcribed and translated. The post-translational regulation by protein–protein interaction results in a faster response. Once the regulatory proteins are formed, they bind to the target proteins and deactivate them, regardless of the degradation times. However, in this situation, unlike the previous scenario, the regulatory proteins are not available at =0 to carry out the regulation, and therefore the response time depends on their production rate. The response time in case of regulation by sRNA is intermediate. It consists of the time it takes to produce the sRNA molecules and the degradation time of the target proteins that remain after the sRNAs bind to their target mRNAs and suppress their translation. However, since sRNA production rate is extremely fast, the kinetics of the regulation by sRNAs in both scenarios is very similar. It is noteworthy that shortly after the regulation is turned on, no mRNA molecules of the regulatory proteins are present. Thus, the initial production rate of regulatory proteins is much lower than that of sRNAs. As a result, shortly after =0, regulation by sRNA exerts a faster response than regulation by protein–protein interaction. Hence, when the regulator is not present in the cell and a fast response is needed in a short time interval, such as upon an external stress, regulation by sRNA has an advantage over the two other regulation types. Indeed, several of the sRNAs with known functions play a role in response to sudden changes in environmental conditions (). These include OxyS that is induced in response to oxidative stress and regulates ∼40 genes, as suggested by genetic screens (), and RyhB that is induced in response to iron depletion and regulates genes involved in iron metabolism (). Another difference between the various regulation mechanisms is considered below. In case of transcriptional regulation, a single bound repressor is sufficient to shutdown the expression of the target gene. In this case, the regulation effectiveness does not depend on the transcription rate of the target gene. It depends only on the production rate of the regulatory protein and on its binding/dissociation rates to the promoter of the target. Thus, with suitable binding/dissociation rates, transcriptional regulation enables using a protein of low concentration to regulate a protein of high concentration. In case of protein–protein interaction, the regulation effectiveness is determined by the relative production rates of the regulator and target proteins. If the production rate of the regulatory protein is faster than that of the target protein, the regulation will be very effective. On the other hand, when the production rates of these two proteins are comparable, it enables fine-tuning of the regulation strength, which is not possible in transcriptional regulation. A similar property characterizes regulation by sRNA. The regulation effectiveness strongly depends on the relative production rates of the sRNA and the target mRNA. Since the rate of production of sRNAs is up to two orders of magnitude faster than of typical mRNAs, it enables effective regulation. It also enables a single sRNA-encoding gene to regulate dozens of other genes. As long as the sRNA is produced at a faster rate than the combined production rate of all the target mRNAs, the regulation is strong. It gradually weakens when the combined production rate of the target mRNAs exceeds that of the sRNA. As an example, we consider an sRNA-encoding gene that regulates other genes. In this case, the rate equations shown above are modified such that the second and third equations are copied into equations, accounting for the number of sRNA molecules and the number of protein molecules of each of the target genes. In addition, the first equation is modified such that is replaced by the total number of mRNA molecules of all the target genes. For simplicity, the parameter values of all the target genes are taken to be identical. In we present the number of molecules of each of the target proteins versus . In this example, when exceeds 50, the regulation weakens and the number of molecules of each target protein increases. Indeed, there are a few examples where a single sRNA-encoding gene regulates several genes involved in the same physiological process, hinting for the existence of sRNA regulons in accord with the regulons governed by transcriptional regulatory proteins (). Our results suggest that for appropriate relations between the production rates of the regulator sRNA and its target genes in the regulon, the simultaneous regulation of these genes will be very effective. The applicable parameter range for production rates of sRNA and mRNA in suggests that in order to be effective, such a regulon should contain only several dozens of genes. In general, the targets may differ from each other in their transcription and translation rates, as well as in their affinities to the sRNA. These differences may provide a hierarchy of regulation. Kinetic studies indicated that the sRNA–mRNA complexes might dissociate back into their original components (; ), with dissociation rates γ in the range between 0.02 and 0.1 s, which is much faster than the degradation rate of the complex. To address this additional scenario, we added one more equation to the model, which accounts for the copy number of the complex. This equation takes the form d/d=α−(+γ), where is the degradation rate of the complex. For simplicity, we chose the degradation rate of the complex to be equal to that of the free mRNA molecule, namely =. In addition, we added the term +γ to the equations that describe the time derivatives of and . As the dissociation rate increases, the regulation effectiveness is reduced. As a result, there are more mRNA molecules available for translation into proteins, and the protein level increases. In we present the number of the target protein molecules versus the dissociation rate of the complex γ for four different values of the ratio between the production rates of the sRNA and target mRNA, /. When sRNAs are produced much faster than mRNAs, there is a large surplus of sRNAs and the regulation remains strong even when dissociation takes place. However, when the sRNA production rate is close to that of the mRNA, even small dissociation rates significantly weaken the regulation and the protein level increases. Delicate control of the dissociation rate enables fine-tuning and maintenance of the target protein level at a desired steady-state level. Another post-transcriptional regulation mechanism is manifested by mRNA-binding proteins (or metabolites). The rate equations describing this kind of regulation are similar to those describing regulation by sRNA. However, unlike sRNAs, the regulatory proteins do not degrade together with the mRNA. As a result, a smaller copy number of regulatory proteins are sufficient in order to provide strong negative regulation at steady state. However, the transient dynamics of this type of regulation is the same as shown in for regulation by sRNAs. We now consider the recovery of the target gene after the transcription of the regulator is turned off. For concreteness, we focus on regulation by sRNAs, where a single target gene is regulated. We assume that the binding of the sRNA to mRNA is fast, and that the sRNA–mRNA complex does not dissociate. In this analysis, the initial copy number of sRNAs is given by the steady-state result of the rate equation, namely =(−)/. It then decreases according to d/d=−−, giving rise to ()=(e−)/. The translation of the target proteins will resume when all the sRNA molecules are removed at time =ln(/)/, denoted as the recovery time. Our simulations show that for the same parameters as in the recovery time in case of regulation by sRNA is faster than in the case of transcriptional regulation, but somewhat slower than for protein-protein interaction (). Clearly, when the regulation is mediated by sRNA, two parameters determine the recovery time: the ratio between the production rates of the regulatory sRNA and target mRNA; and the degradation rate of the sRNA. The latter has a greater influence on the determination of the recovery time. For example, the recovery time can be made equal to that of transcriptional regulation by either increasing the ratio / by a factor of 5000 or by decreasing the degradation rate of the sRNA by a factor of 3. When the regulatory protein loses its activity without degradation (e.g., by phosphorylation/dephosphorylation), no differences in the kinetics of recovery were observed for the various regulation modes. The analyses were carried out using rate equation models. These equations account for the concentration (average number of molecules per cell) of each component in the circuit, namely mRNA and sRNA molecules, free proteins and proteins that are bound to the promoter site. The model consists of a set of coupled ordinary differential equations, each equation evaluates the time derivative of the concentration of one type of molecule. The model is based on several assumptions made in order to simplify the equations and their analysis. One assumption is that the binding rates of pairs of molecules are diffusion-limited. The transcription rate constants and incorporate all the molecular processes involved in the transcription of the mRNA and sRNA molecules, respectively. The simulation is Markovian, in the sense that it does not include any time delays. A similar assumption regards the translation rates. The sRNA network in was generated using Cytoscape software 2.4.0 ().

To promote the assembly of functional ribonucleoprotein (RNP) complexes, RNA-binding proteins possess several different structural motifs which confer recognition of RNA sequence and structural elements (). The RNA-recognition motif (RRM), which is the most common RNA-binding domain, consists of a βαββαβ-fold in which specific amino acids in the two central β-strands contact bases in a single-stranded (ss)RNA target site (). Additional ssRNA-binding motifs include the KH domain, arginine-rich sequences and zinc-finger related motifs. An important example of the latter motif is found in the TTP/TIS11/ZFP36 proteins which promote the deadenylation and turnover of target mRNAs which contain a 3′ untranslated region (3′ UTR) class II AU-rich element (ARE) (). The NMR structure of a pair of CXCXCXH fingers in the TIS11d tandem zinc finger (TZF) domain protein reveals that these zinc fingers, which are separated by an 18-residue linker, bind to adjacent 5′-UAUU-3′ sequence elements (). For double-stranded (ds)RNAs, the most common motif is the dsRNA-binding domain (dsRBD) which recognizes the A-form helical conformation (). The dsRBD structure, which consists of a αβββα-fold with two highly conserved basic loops, interacts with successive grooves on the RNA helix (,). In addition, some zinc finger (ZF) proteins recognize both sequence and structural elements in their target RNAs, including TFIIIA which simultaneously contacts ssRNA and dsRNA regions in 5S RNA (). Previous studies have suggested that the muscleblind-like (MBNL) family of alternative splicing factors, which possess multiple copies of a TTP/TIS11-related CCCH motif (CXCXCXH), also recognize both ssRNA and dsRNA elements (,). Similar to TIS11d, the CCCH motifs in MBNL proteins are organized as tandem pairs separated by a 14–16-residue linker (). The MBNL proteins were first characterized as factors involved in the pathogenesis of the neuromuscular disease myotonic dystrophy (DM) (). DM is caused by the expansion of structurally similar microsatellites in two functionally unrelated genes. Type 1 (DM1) disease is associated with (CTG) expansions in the 3′-UTR of the gene while (CCTG) expansions in the first intron of 9 results in type 2 (DM2) disease (). The RNA-mediated pathogenesis model for DM suggests that these expansions are toxic at the RNA level because DM1 and DM2 mutant RNAs fold into stable RNA hairpins which bind MBNL proteins with high affinity. MBNL proteins are pre-mRNA alternative splicing factors and loss of MBNL1 activity in DM tissues leads to persistence of fetal splicing patterns in the adult. In support of this model, MBNL1 binds to (CUG) and (CCUG) expansion RNAs (CUG, CCUG) and co-localizes with nuclear RNA (ribonuclear) foci in DM1 and DM2 skeletal muscle and brain, with a concomitant reduction in the nucleoplasmic level (,). Moreover, DM1 ribonuclear foci are lost following siRNA-induced downregulation of expression (). Finally, knockout mice recapitulate several clinical defects characteristic of DM, including myotonia and subcapsular particulate cataracts (). These observations indicate that a critical initiating event in the DM pathogenesis pathway is loss of MBNL function leading to defective pre-mRNA splicing. Apparently inconsistent with this view is a recent finding that several of the cardinal features of DM, including skeletal muscle myotonia and cardiac conduction defects, are recapitulated in a transgenic mouse model which overexpresses an inducible GFP-DMPK 3′-UTR transgene that contains only five CTG repeats (). Surprisingly, these mice do not develop ribonuclear foci and the subcellular localization pattern of Mbnl1 is not altered. This study raises the question of whether foci formation is required for loss of MBNL1 function and highlights the need to further investigate the interactions of muscleblind-like proteins with both normal target and pathogenic RNAs. Here, we report that the MBNL1 protein preferentially recognizes GC-rich RNA helices containing a pyrimidine mismatch on both normal splicing substrates and pathogenic RNAs. Furthermore, MBNL1 binds selectively to the stem region of CUG RNA and can be visualized as a ring-like structure in the electron microscope. This study introduces the possibility that the tandem arrangement of MBNL high affinity binding sites present on CUG RNA results in a stacked ring complex which effectively traps MBNL1 and inhibits its role as an alternative splicing factor during postnatal development. To construct pGEX-6P-His, a XhoI–NotI fragment encoding the His tag from pGEX-MBNL1 was inserted into XhoI–NotI digested pGEX-6P-1. pGEX-6P-MBNL1-His was constructed by inserting a BamHI–XhoI fragment from pGEX-MBNL1 into BamHI–XhoI digested pGEX-6P-His. The MBNL1 N-terminal region (residues 1–253) was amplified using MSS2759 and MSS2760, digested with BamHI and XhoI and inserted into pGEX-6P-His to create pGEX-6P-MBNL1-N-His. For the yeast two-hybrid system, either full length, amino terminal (residues 1–264) or carboxyl terminal (residues 239–382) MBNL1 cDNAs were PCR amplified using the following primers and inserted into pGBKT7 or pGADT7 (Clontech, Mountain View, CA, USA) at SmaI and BamHI sites: (i) full length, MSS1163 (forward primer) and MSS1166 (reverse primer); (ii) N-terminal region, MSS1164 (forward) and MSS1166 (reverse); (iii) C-terminal region, MSS1163 (forward) and MSS1165 (reverse). To create pcDNA3-V5, MSS3045 and MSS3046 were subjected to a 10-cycle PCR reaction: 94°C 30 s, 50°C 20 s, 72°C 20 s. The resulting DNA fragment was gel purified followed by digestion with NheI and BamHI, and inserted into NheI–BamHI digested pcDNA3.1(+) (Invitrogen, Carlsbad, CA, USA). The BamHI–XhoI fragment from pGEX-6P-MBNL1-His was inserted into pcDNA3-V5 at BamHI and XhoI sites to create pcDNA-V5-MBNL1. The BamHI–XhoI fragments from pGEX-6P-MBNL1-His and pGEX-6P-MBNL1-N-His were inserted into pcDNA3.1(+)/myc-His A (Invitrogen) to create pcDNA3-MBNL1-mycHis and pcDNA3-MBNL1-N-mycHis, respectively. The CUGBP1-coding sequence was PCR amplified using MSS2699 and MSS2700 and inserted into pcDNA3.1(+)/myc-His A at BamHI and XhoI sites. The Tnnt3 minigene was prepared by amplifying the mouse genomic region between exons 8 and 9 using MSS1949 and MSS1950 and inserting the PCR product into pSG5(Stratagene, La Jolla, CA, USA) at the EcoRI site. The mutant Tnnt3 minigenes, pSG5-Tnnt3Δ10 and pSG-Tnnt3/gg and pSG-Tnnt3/au, were generated by site-directed mutatgenesis using MSS2129/MSS2130, MSS3131/3132 and MSS3163/MSS3164, respectively. Wild-type pSG5-Tnnt3 (100 ng) (with 125 ng of each of the primers) was subjected to the following PCR reaction: 94°C 30 s, 50°C 1 min, 72°C 8 min, 20 cycles using Pfu DNA polymerase (Stratagene). After DpnI digestion, the PCR product was transformed into DH10B and mutants were identified by plasmid DNA sequencing. Using pSG5-Tnnt3 as a template, PCR fragments generated from primer pairs MSS1865/MSS1879 and MSS1884/MSS1866 were TOPO-cloned into pCR4-TOPO (Invitrogen) to make pTOPO-T5.1 and pTOPO-T5.45, respectively. For the preparation of recombinant proteins, BL21(DE3) RP containing pGEX-6P-1-MBNL1 or pGEX-6P-1-MBNL1-N were grown to OD = 0.5 followed by induction with 1 mM IPTG for 2 h at 30°C. Cells were collected and resuspended in lysis buffer containing 25 mM Tris–Cl, pH 8.0, 0.5 M NaCl, 10 mM imidazole, 2 mM β-mercaptoethanol, 2 mg/ml lysozyme, 10 μg/ml DNase I, 5% glycerol, 0.1% Triton X-100 supplemented with protease inhibitors. The cell suspension was incubated on ice for 30 min with stirring prior to sonication and centrifugation at 12 000. For protein purification, Ni-NTA-Sepharose (Amersham/GE Healthcare, Piscataway, NJ, USA) (12 ml) was incubated with the supernatant for 1 h at 4°C and washed three times with 40 ml of wash buffer containing 25 mM Tris–Cl, pH 8.0, 0.5 M NaCl, 20 mM imidazole, 0.1% Triton X-100, followed by three 10 ml elutions in 25 mM Tris–Cl, pH 8.0, 0.5 M NaCl, 250 mM imidazole, 0.1% Triton X-100. Subsequently, β-mercaptoethanol was added (10 mM final concentration) to the eluate, which was incubated with 2 ml glutathione-Sepharose (Amersham) for 1 h at 4°C. After three washes (10 ml each) of buffer (WB) containing 25 mM Tris–Cl, pH 8.0, 300 mM NaCl, 5 mM β-mercaptoethanol, 0.1% Triton X-100, the glutathione-Sepharose beads were incubated with 4 ml WB containing 40 U of PreScission protease (Amersham) at 4°C overnight. The supernatant was collected following brief centrifugation and concentrated to 1–8 mg/ml. Transcription reactions were carried out in a 50 µl volume which contained 2 μg of each DNA template, 1 mM rNTPs, 3.3 mM guanosine, 60 U of ribonuclease inhibitor RNase Out (Invitrogen), 200 U of T7 RNA polymerase (Ambion, Austin, TX, USA), 10 mM DTT, 40 mM Tris-HCl (pH 7.9), 6 mM MgCl, 2 mM spermidine, 10 mM NaCl. The reaction was performed at 37°C for 2 h, the transcript was then purified on a denaturing 10% polyacrylamide gel and subsequently 5′-end-labeled with T4 polynucleotide kinase and [γP]ATP (3000 Ci/mmol). The labeled RNA was re-purified by electrophoresis on a denaturing 10% polyacrylamide gel. Prior to structure probing, the labeled RNA was subjected to a denaturation/renaturation procedure in a reaction buffer containing 50 mM Tris–HCl (pH 8.0), 60 mM KCl, 15 mM NaCl, 2 mM MgCl by heating the sample at 90°C for 1 min and slowly cooling to 25°C. The RNA sample was then mixed with either a 25-fold molar excess of recombinant MBNL1 in 50 mM Tris–HCl, pH 8.0, 60 mM KCl, 15 mM NaCl, 2 mM MgCl, 2% glycerol, 0.5 mM DTT, 50 μg/ml BSA, or with buffer only (control) and incubated 20 min at 25°C. The final concentration of (CUG) was 20 nM and MBNL1 was 500 nM. Under these reaction conditions >95% of RNA was bound with protein as revealed by filter-binding assays. Additional control samples were prepared by mixing RNA with MBNL1 protein previously denatured by heating at 75°C for 2 min. Limited RNA digestion was initiated by mixing 5 µl of the RNA or RNA/protein sample (25 000 c.p.m.) with 5 µl of a probe solution containing either lead ions or ribonuclease T1 in reaction buffer. The reactions were performed at 25°C for 20 min and stopped by adding 20 volumes of 1× TE buffer followed by phenol/chloroform extraction. Precipitated RNAs were dissolved in a denaturation solution (7.5 M urea and 20 mM EDTA with dyes). To determine the cleavage sites, the products of RNA fragmentation were separated on 10% polyacrylamide gels containing 7.5 M urea, 90 mM Tris-borate buffer and 2 mM EDTA, along with the products of alkaline hydrolysis and limited T1 nuclease digestion of the same RNA. The alkaline hydrolysis ladder was generated by the incubation of the labeled RNA in formamide containing 0.5 mM MgCl at 100°C for 10 min. The partial T1 ribonuclease digestion of RNAs was performed under semi-denaturing conditions (10 mM sodium citrate, pH 5.0; 3.5 M urea) with 0.2 U/µl of the enzyme during incubation at 55°C for 10 min. Electrophoresis was performed at 1800 V (gel dimensions, 30/50 cm). The products of the structure probing reactions were visualized by PhosphorImaging (Storm; Molecular Dynamics, Sunnyvale, CA, USA) and analyzed by ImageQuant 5.2 (Molecular Dynamics). To test for MBNL1–MBNL1 interactions in mammalian cells, HEK293T cells were transfected with 5 μg of pcDNA3-V5-MBNL1 alone as a control or 5 μg of pcDNA3-V5-MBNL1 together with either 5 μg of pcDNA3-MBNL1mycHis or pcDNA3-MBNL1-N-mycHis. Cells were harvested 20–24 h post-transfection by trypsinization followed by neutralization in media contain 10% fetal bovine serum and two washes in 50 mM Tris–HCl, pH 7.4, 150 mM NaCl. Cell pellets were resuspended in 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% IGEPAL with protease inhibitors, and sonicated on ice (3 × 5 s). Cell debris was removed by centrifugation at 16 100 for 10 min at 4°C. Cleared lysates were treated with 200 μg/ml RNase A for 20 min on ice () followed by another 10 min centrifugation. Cleared lysates were mixed with Dynabeads coupled to Protein A (Invitrogen) precoated with rabbit anti-V5 polyclonal antibody (Novus, Littleton, CO, USA) and incubated at 4°C for 2 h. Dynabeads were washed three times with IPP150 buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% IGEPAL) and once with 50 mM Tris–HCl, pH 7.4, 150 mM NaCl. Proteins were dissociated from the beads by heating at 95°C for 2 min in 1× SDS–PAGE sample buffer. Proteins (50% of the immunoprecipitated proteins, 2.5% of the input) were separated on 12.5% SDS–PAGE gels. Immunoblotting was performed using mAb 9E10 (1:1000) or mAb anti-V5 (1:1000, AbD Serotec). The CUG RNA, transcribed from pBC-CTG136 plasmid (R. Osborne, University of Rochester), was incubated with MBNL1 protein at molar ratio of 1:2.5 or 1:10 in buffer containing 16 mM HEPES, 2 mM magnesium acetate, 0.16 mM EDTA, 0.4 mM DTT, 1 mM ATP, 50 mM potassium acetate and 16% glycerol for 30 min at 30°C. The resulting complexes were fixed in 0.6% glutaraldehyde for 5 min at room temperature and subsequently passed over a 2 ml column containing Bio-Gel A-5M (Bio-Rad, Hercules, CA, USA) equilibrated with 0.01 M Tris (pH 7.6) and 0.1 mM EDTA to remove free protein and fixatives. Protein–RNA enriched fractions were incubated with 2.5 mM spermidine and adsorbed on to glow charged carbon coated copper grids and dehydrated in a series of ethanol washes of 25, 50, 75 and 100% ethanol each for 5 min at room temperature. Samples were air dried prior to rotary shadow casting with tungsten. Protein–RNA complexes were visualized using a FEI Tecnai 12 electron microscope (FEI, Hillsboro, OR) at an accelerating voltage of 40 kV and images were captured using a 4K × 4K Gatan CCD camera using plate film or Gatan digital image capturing software (Gatan, Pleasanton, CA, USA). Plate film negatives were scanned using an Imacon scanner and supporting software (Imacon, Redmond, WA). Images were photographed at a magnification of 52K. RNAs for photocrosslinking were uniformly labeled with 40 μCi each of (α-P)-GTP and (α-P)-UTP (800 Ci/mmol) in the presence of 0.5 mM ATP and CTP, 0.02 mM GTP and UTP. Cross-linking was performed by incubating 0.1 pmol RNA with 15 μl of HEK293T whole cell lysate in 25 μl reactions containing 16 mM HEPES–KOH (pH 8.0), 65 mM potassium glutamate, 2 mM Mg(OAC), 0.4 mM DTT, 0.16 mM EDTA, 20 mM creatine phosphate, 2 mM ATP and 16% glycerol (final concentration). Reactions were incubated at 30°C for 15 min, transferred to pre-chilled PCR caps on ice and photocrosslinked in Stratalinker (Stratagene, La Jolla, CA, USA) for 2.5 min (three times) with a 3 min interval between each irradiation. Samples were digested with 5 μg of RNase A for 20 min at 37°C and immunopurified using the anti-myc monoclonal antibody 9E10 pre-coated protein A Sepharose (Amersham). Purified proteins were fractionated on 12.5% SDS–PAGE gels followed by autoradiography. Uniformly labeled RNA was prepared as described previously (). Calibration of the non-specific retention rate of the nitrocellulose filter was performed by incubating 0.01–0.1 nM RNA at 30°C for 30 min in BB [50 mM Tris-HCl (pH 8.0), 40 mM KCl, 20 mM potassium glutamate, 15 mM NaCl, 0.5 mM DTT, 0.05 U/ul RNasin (Promega)] followed by filtration through a Bio-Dot (BioRad) apparatus containing a sandwich of nitrocellulose (BioRad) and Hybond-N plus (Amersham) membranes followed by a single wash step with the same buffer. The membranes were UV-cross-linked, air dried and exposed to a phosphorimager screen. Non-specific retention on the nitrocellulose membrane was undetectable. Binding reactions were set up in the same buffer with 5 pM RNA and 3.13 × 10 M to 1.02 × 10 M of MBNL1/41-His, and incubated at 30°C for 30 min. Each reaction was applied to the Bio-Dot apparatus followed by one wash with binding buffer. Membranes were processed as described above and signals quantified using ImageQuant TL (Amersham). Standard deviations were calculated based on three independent experiments and apparent dissociation constants were calculated using a one-site binding model and GraphPad Prism (v3.00) software. Gel shift assays were performed using modifications of previously described protocols (,). RNA was uniformly labeled with 40 μCi (α-P)-GTP or UTP (800 Ci/mmol) in the presence of 0.5 mM ATP, 0.5 mM CTP, 0.02 mM GTP, 0.02 mM UTP and purified using a 5% denaturing gel containing 8 M urea. Prior to use, purified RNA was heated at 65°C for 5 min in 50 mM Tris–HCl (pH 8.0), 40 mM KCl, 20 mM potassium glutamate, 15 mM NaCl, 0.5 mM DTT, 0.5 U/μl SUPER-asin (Ambion) following by renaturation at RT. Reactions (20 μl) were assembled with 0.1 nM RNA and 0–256 nM protein in 50 mM Tris–HCl, pH 8.0, 40 mM KCl, 20 mM KGlutamate, 15 mM NaCl, 15% glycerol, 0.5 mM DTT, 20 μg/ml acetylated BSA. Alternatively, RNAs were heated using a higher salt concentration in the presence of magnesium (50 mM Tris–HCl (pH 8.0), 60 mM KCl, 120 mM potassium glutamate, 20 mM NaCl, 2 mM magnesium acetate, 0.5 mM DTT, 0.5 U/μl SUPER-asin (Ambion) and then reactions were performed in 50 mM Tris-HCl (pH 8.0), 60 mM KCl, 120 mM potassium glutamate, 20 mM NaCl, 2 mM magnesium acetate, 0.5 mM DTT, 20 μg/ml acetylated BSA. After incubation at 30°C for 30 min, reactions were immediately loaded onto a 4% polyacrylamide gel (80:1) containing 0.5 mM DTT and 5% glycerol which had been pre-run at 150 V for 1–2 h at 4°C. Gels were run in 0.5× TBE (pH 8.3) at 200 V for 2 h, fixed and dried prior to autoradiography. A recent mapping study identified several MBNL1-binding sites containing a core element within human cardiac troponin T (cTNT/TNNT2) intron 4 pre-mRNA immediately upstream of developmentally regulated exon 5 (). Although the major MBNL1-binding sites in chicken cTNT are positioned downstream of exon 5, alignment of the human- and chicken-binding sites revealed a hexanucleotide consensus motif (5′-YGCUU/GY-3′). Neither the human- or chicken-binding sites were located in regions predicted to form secondary structures. Together with prior observations that MBNL proteins are sequestered by CUG and CCUG hairpins in ribonuclear foci in DM cells, this finding suggests that the MBNL proteins bind to both ssRNA and dsRNA structural motifs. To determine if MBNL1 recognizes primarily ssRNA targets in other splicing precursors, we first mapped the MBNL1-binding site on fast skeletal muscle troponin T (Tnnt3) RNAs. This pre-mRNA was selected because earlier studies demonstrated that Tnnt3 fetal (F) exon splicing is particularly sensitive to MBNL1 levels and (,). Mapping was performed using a photocrosslinking protocol in which 293T cells were transfected with protein expression plasmids encoding myc-tagged versions of either CUGBP1, full-length MBNL1 (MBNL1 FL) or the MBNL1 N-terminal region (MBNL1 N) which contains the four CCCH motifs responsible for RNA binding (). Following transfection, cell lysates were incubated with radiolabeled Tnnt3 RNAs encompassing exons 7–9 or different 500–645 nucleotide (nt) subregions designated T1-6 (A), photocrosslinked with UV-light and digested with RNase A. Protein–RNA complexes were then immunopurified with the anti-myc monoclonal antibody (mAb) 9E10 and resolved by SDS–PAGE. Because these studies indicated that only Tnnt3 T5 RNA (500 nt) cross-linked to both MBNL1 FL and MBNL1 N proteins, this region was further subdivided into T5.1–T5.45. Interestingly, only T5.45 RNA (200 nt) cross-linked to CUGBP1 and MBNL1 proteins while T5.1 (125 nt), and the other subregions (T5.2 and T5.3, data not shown), did not (B). In agreement with prior studies, CUGBP1 failed to cross-link to a CUG, (CUG), while both MBNL1 FL and MBNL1 N did (). To further delineate MBNL1-binding sites on Tnnt3, we analyzed MBNL1 cross-linking to a several subregions of T5.45. One region was particularly interesting because it contains a 5′-CGCU-3′ motif which is conserved in the MBNL1-binding site in human cTNT/TNNT2 (). Interestingly, this motif is located within a predicted 18-nt stem-loop structure (A). We first confirmed the existence of this Tnnt3 18-nt hairpin structure using chemical and enzymatic structure probing (A). Cross-linking assays were then used to test whether wild-type and mutant Tnnt3/98 RNA (a 98-nt subregion encompassing 83-nt of the 3′ end of intron 8 and 15-nt of the F exon) was recognized by both CUGBP1 and MBNL1 (B). Concurrently, wild-type or mutant Tnnt3 minigenes were transfected into C2C12 cells to assay whether Tnnt3 splicing remained responsive to MBNL1 overexpression (C). These assays were performed using C2C12 myoblasts since they showed a higher default level of Tnnt3 F exon skipping compared to 293T cells (C, left panel, lane 1 and data not shown). Cross-linking assays confirmed that wild-type Tnnt3/98 (B, left panel) RNA was recognized by both MBNL1 and CUGBP1. Correspondingly, the splicing pattern of the wild-type Tnnt3 minigene was not altered upon CUGBP1 overexpression (C, left panel, compare lanes 1 and 2) while F exon exclusion was enhanced by MBNL1-mycHis (C, left panel, lane 3), consistent with our previous study (). Mutations in the 18-nt hairpin positioned just upstream of F exon (A) were generated to test whether it contains an MBNL1-binding site. The relative position of this hairpin in the F exon 3′ splice site region is significant since a potential binding site for the essential splicing factor U2AF, which binds preferentially to U-rich tracts, lies just upstream of this stem-loop structure and this GC-rich hairpin contains a pyrimidine mismatch reminiscent of the structure of CUG and CCUG RNA hairpins. We first generated a 10-nt deletion within the 18-nt region to eliminate this hairpin (A, deleted nucleotides in gray). As predicted, this mutant showed an impairment of both MBNL1 cross-linking (B, Δ10 mutant) and F exon skipping promoted by MBNL1 overexpression (C, Δ10 mutant). Indeed, the Δ10 deletion eliminated F exon skipping in cells transfected with the Tnnt3 minigene alone or with CUGBP1-mycHis together with the minigene (C, Δ10 mutant, lanes 1–2) suggesting loss of MBNL1 binding and enhanced spliceosome recruitment to the F exon region. The cross-linking and splicing results obtained with the Δ10 mutant indicated that the 18-nt hairpin was a binding site for MBNL1 and that MBNL1–hairpin interactions might promote F exon skipping. Because both sequence and structural elements could contribute to efficient MBNL1 binding, a double C→G and U→G mutant was generated that eliminated the C-C mismatch in the 18-nt hairpin, which also increases the stability of this stem-loop, and furthermore substituted a G for a U in the loop. These mutations are not predicted to alter the overall folding pattern of Tnnt3/98. This double point (gg) mutant showed considerably reduced MBNL1 cross-linking compared to wild-type Tnnt3 (B) confirming that this region was an MBNL1-binding site. Interestingly, a similar decrease in MBNL1 cross-linking activity was also observed for the single C→G stem substitution mutant (data not shown). While loss of MBNL1 binding should promote F exon splicing, inclusion activity was completely eliminated (C, gg mutant). Loss of F exon splicing activity could result from enhanced hairpin stability, or an increase in the purine content of the polypyrimidine (Py) region upstream of the F exon 3′ splice site, and subsequent impairment of spliceosome assembly. Interestingly, another double mutant (G→A and C→U), which preserved the predicted 18-nt stem-loop structure while reducing the GC content in the stem, also showed reduced MBNL1 cross-linking compared to wild type (B, au mutant) while MBNL1-induced F exon skipping activity was reduced similar to the Δ10 mutant (C, au mutant). Overall, these studies indicated that MBNL1 prefers to bind to a GC-rich stem-loop containing a pyrimidine mismatch in a normal splicing target. According to the RNA sequestration model, pathogenic RNAs outcompete normal RNA-binding targets for MBNL1 leading to loss of MBNL-mediated regulation of alternative splicing during postnatal development. However, it is not clear why MBNL1 accumulates on DM1 and DM2 expansion RNAs in ribonuclear foci. The most straightforward explanation is that MBNL1 has a higher affinity for DM pathogenic RNAs compared to its physiological RNA splicing targets. Since the cross-linking of MBNL1 to the Tnnt3 F exon 3′ splice site region and CUG RNAs appeared to be comparable (B), we determined the relative affinities of MBNL1 for (CUG), (CAG) and Tnnt3/5.45 using recombinant MBNL1 protein in filter binding and gel shift assays. The MBNL1 proteins used for this study were either MBNL1 FL or the C-terminal truncation mutant MBNL1 N. For the MBNL1 FL preparation, ∼60% of the purified protein was full length while the MBNL1 N protein preparation was homogeneous as determined by Coomassie blue staining and immunoblot analysis (A). Because MBNL1 shows a temperature-dependent binding profile in cell extracts in the absence of ATP (data not shown), all recombinant protein-binding studies were performed at 30°C to maximize binding while minimizing RNA degradation. As anticipated, MBNL1 FL showed a high affinity in the filter-binding assay for (CUG) ( = 5.3 ± 0.6 nM), but it also showed high affinities for Tnnt3 5.45 (Tnnt5.45) ( = 6.6 ± 0.5 nM) and (CAG) ( = 11.2 ± 1.5 nM) (B). In contrast, Tnnt3/T5.1 RNA, which did not cross-link to MBNL1 (B), bound poorly. The similarities in the binding affinities of MBNL1 for (CUG) and (CAG) accounts for the prior observation that overexpression of either of these repeat RNAs in COS-M6 cells results in the formation of nuclear foci that colocalize with GFP-MBNL1 (). Relatively weak cooperativity was noted for interactions between MBNL1 and (CUG), (CAG) and Tnnt3 T5.45 RNAs (Hill coefficients of 1.62 ± 0.21, 1.67 ± 0.27 and 1.40 ± 0.14, respectively). Comparable affinities were obtained when MBNL1–RNA complexes were analyzed by gel shift analysis (C). Incubation of full-length MBNL1 (MBNL1 FL) with Tnnt3/T5.45 generated several protein–RNA complexes resolved by the polyacrylamide gel whereas significant binding to Tnnt3/T5.1 was only detectable at 256 nM and only gel excluded complexes were observed. Similar complexes were also formed with MBNL1 N although the truncated protein had a higher affinity for Tnnt3/T5.1 RNA and formed fewer gel-resolved complexes with Tnnt3/T5.45 and (CUG) (C). Incubation of MBNL1 with (CUG) also resulted in the formation of several major complexes at protein concentrations (4–16 nM) near the determined by filter binding. At higher MBNL1 FL concentrations (≥64 nM) the majority of the resulting MBNL1–(CUG) complexes migrated at, or near, the top of the gel. Similar gel shift profiles were obtained when RNA–protein complexes were formed in a higher salt buffer containing magnesium (see Methods section) although all values increased ∼4-fold. The striking similarity in the binding affinities of MBNL1 for pathogenic and splicing precursor RNAs prompted us to re-examine the interaction of this splicing factor with CUG RNA. Muscleblind-like proteins were originally characterized as nuclear factors which are recruited by CUG RNAs (). The predicted double-stranded nature of these pathogenic RNAs has been validated by chemical and nuclease mapping, thermal denaturation and electron microscopy (). While the binding of MBNL proteins to CUG RNA is proportional to the predicted stem length, there is currently no direct experimental evidence that MBNL binds directly to the CUG stem or that the RNA structure remains in the hairpin configuration following MBNL binding. Therefore, we initially used chemical and enzymatic structure probing of labeled RNAs to identify MBNL1-binding sites on (CUG) RNA. RNAs were 5′ end-labeled, subjected to either lead ion (Pb)-induced hydrolysis or RNase T1 digestion in the absence or presence of recombinant MBNL1, and the products were fractionated on denaturing polyacrylamide gels. As shown previously, lead ions cleave both ssRNA and relaxed dsRNA structures which yielded a uniform ladder that increased in intensity with increasing lead concentration (, left panel). As anticipated, addition of MBNL1 inhibited strand cleavage in a concentration-dependent manner and densitometry analysis failed to show significant regional differences in the cleavage pattern by MBNL1 suggesting uniform binding of this protein throughout the stem region. In contrast to lead, RNase T1 prefers to cleave after G nucleotides in single-stranded regions. Thus, incubation with RNase T1 resulted in strong cleavage at the terminal loop (, right panel, G26–G29). Interestingly, terminal loop cleavage was unaffected by MBNL1 addition while stem cleavage was uniformly inhibited. We conclude that MBNL1 interacts primarily with the stem region of CUG RNAs. In a previous study, we reported that the MBNL splicing antagonist, CUGBP1, binds to out-of-register ssCUG repeats at the base of CUG RNA hairpins but not the A-form helical region while the dsRBD protein TRBP associates with the stem region (). To confirm that MBNL1 is a stem-binding protein, electron microscopy (EM) was performed using the full-length MBNL1 protein. Purified (CUG) was examined following direct absorption to thin carbon foils, dehydration and tungsten shadowing. In contrast to ssRNA, (CUG) RNA formed rod-like segments as described previously for (CUG) (A–C). To examine the structures of MBNL1–RNA complexes, MBNL1 was incubated with RNAs at two different RNA:protein molar ratios (1:2.5 or 1:10) and subsequently prepared for EM. In the presence of RNA, purified MBNL1 formed a ring-shaped structure with a prominent central cavity and for (CUG)–MBNL1 complexes incubated at a ratio of 1:2.5, ∼70% of the RNAs were bound by one of these MBNL1 rings (D–F). At higher protein levels (1:10), free (CUG) RNA was rarely detectable (6.4% of the RNAs in the field) while >90% of the RNAs were bound by two or more MBNL1 rings (G and H). Also shown is a representative field of dsCUG RNAs and MBNL1–(CUG) complexes (I, RNA:protein is 1:2.5). Although the majority of MBNL1 rings were associated with RNA under our RNA assembly conditions, a few free rings were visualized in the background in the absence of associated (CUG) helices suggesting that MBNL1 may form a ring structure independent of RNA. We failed to visualize rings by negative staining suggesting that this structure is disrupted by the acidic conditions of the negative staining protocol. Interestingly, when MBNL1 N proteins were incubated with (CUG), ring structures were observed although ring size was much less uniform and there was considerably less stacking of MBNL1 N rings compared to MBNL1 FL (data not shown). This observation prompted us to examine whether the C-terminal region was important for MBNL1 homotypic interactions. A number of studies have demonstrated that MBNL1 accumulates in ribonuclear foci together with pathogenic RNAs (,). While additional proteins might also bind to CUG and CCUG RNAs, the number of ribonuclear foci in DM myoblasts declines significantly following loss of MBNL1 suggesting that this protein is required for the formation and/or maintenance of these unusual nuclear structures (). Since many RNA-binding proteins function as components of large multi-subunit complexes () and some of these proteins self-interact via their auxiliary or non-RNA-binding regions (), we tested the possibility that MBNL1 proteins self-associate using the yeast two-hybrid system. Although the amino terminal region of MBNL1 contains all four CH motifs and is responsible for protein–RNA interactions (), very little is known about the function of the C-terminal region. The MBNL1 FL protein showed strong homotypic interactions in this system (A). Although MBNL FL failed to interact with the MBNL1 N-terminal region (1–264), interactions between the full-length protein and C-terminal region (239–382) were readily detectable. This C-terminal region does not contain any known RNA binding, or other, structural motifs. To confirm that MBNL1 homotypic interactions occurred in a mammalian cell context, 293T cells were co-transfected with plasmids which expressed either V5-MBNL1 FL alone, V5-MBNL1 FL and MBNL1 FL-myc or V5-MBNL1 FL and MBNL1 N-myc. Twenty-four hours following transfection, V5-tagged MBNL1 was immunopurified from cell lysates using an anti-V5 antibody and the precipitates were then immunoblotted using either anti-myc or anti-V5 antibodies. In agreement with the two-hybrid analysis, the full-length V5 and myc-tagged proteins were associated while the N-terminal MBNL1 region (MBNL1 N-myc) failed to co-immunopurify with V5-MBNL1 FL (B). Interactions between V5-MBNL1 FL and MBNL1 N-myc were not mediated by RNA tethering since treatment of the cell lysate with RNase A did not affect the amount of MBNL1 N-myc in the V5-MBNL1 FL immunoprecipitate. Although it is possible that MBNL1 interactions could be mediated by other nuclear factors, our demonstration that purified MBNL1 forms a ring-like structure argues against this interpretation. We conclude that MBNL1–MBNL1 interactions occur and these interactions are mediated by the C-terminal region. A number of studies have demonstrated that CUG and CCUG RNAs sequester MBNL proteins in nuclear RNA foci of DM cells (,,,,,). Loss of MBNL1 function by sequestration correlates with the inhibition of alternative splicing regulation for a specific set of developmentally regulated exons that are mis-spliced in DM tissues (,,,). Moreover, knockout mice develop many of the characteristic features associated with DM disease, including myotonia and subcapsular cataracts, while adeno-associated virus (AAV)-mediated Mbnl1 overexpression in transgenic mice expressing a CUG in skeletal muscle reverses myotonia and DM-associated mis-splicing (,). Although there is considerable evidence for this muscleblind loss-of-function model for DM pathogenesis, the molecular basis for MBNL1 sequestration by CUG and CCUG RNAs has not been elucidated. Since mutant RNA expansions must compete with normal pre-mRNA, and possibly mRNA, binding sites for MBNL1 recruitment, effective MBNL1 sequestration might occur if the affinity of this protein for CUG and CCUG RNAs is greater than for its normal splicing targets. The binding analysis presented here does not support this conjecture since MBNL1 also possesses relatively high affinities for CAG and Tnnt3 precursor RNAs. Indeed, these binding studies provide an explanation for the formation of MBNL1-containing ribonuclear foci in cells overexpressing CAG (). Additionally, mapping of a binding site to a stem-loop structure in Tnnt3 intron 8 just upstream of the F exon indicates that RNA recognition by MBNL proteins involves a common interaction mode for both pathogenic and normal pre-mRNAs: recognition of GC-rich hairpins containing pyrimidine mismatches. The importance of RNA secondary structures in inherited disease and alternative splicing has been previously highlighted in frontotemporal dementia with parkinsonism linked to chromosome 17 (FTDP-17) which is caused by mutations in the gene encoding the microtubule-associated protein tau (). Some FTDP-17 mutations destabilize a predicted stem-loop structure, which forms between the 3′end of exon 10 and the 5′ end of the downstream intron, resulting in an increase in U1 snRNP recruitment and E10 inclusion. Interestingly, MAPT exon 10 skipping increases in the DM brain suggesting that MBNL1 promotes exon 10 inclusion during splicing (). The MBNL1-binding preferences shown in this study suggest that this factor may also function as a splicing activator by recognizing RNA stem-loop structures in novel exonic and/or intronic splicing enhancers or by blocking splicing silencer elements by stabilizing overlapping RNA secondary structures. Another interesting question arose when we mapped the MBNL1-binding site on Tnnt3 to an 18-nt stem-loop structure. Our previous study showed that MBNL1 in HeLa nuclear extract cross-links strongly to (CUG) and (CUG) but not to (CUG) (). Based on this observation, we proposed that below a certain length threshold (<20 repeats) the dsCUG helix was unstable in the cell extract and ssCUG was not a binding site for MBNL1. The results reported here support that proposal and demonstrate that MBNL1 is primarily a dsRNA-binding protein which recognizes relatively short GC-rich hairpins if the overall RNA secondary structure is stabilized by additional sequence interactions. This conclusion provides a plausible resolution for the apparently conflicting results that overexpression of a GFP-DMPK 3′-UTR (CTG) transgene results in a DM phenotype () while mice expressing the (human skeletal α-actin containing a 3′-UTR with five CTG repeats) transgene are normal (). For the 3′-UTR, the (CTG) repeat is predicted to interact with an upstream region to form a GC-rich stem interrupted by several U-U and C-C mismatches while this repeat in the HSA 3′-UTR is located in sequential unpaired loops. We postulate that this structural arrangement in the DMPK 3′-UTR promotes MBNL1 sequestration when GFP-DMPK 3′-UTR (CTG) RNA is overexpressed in transgenic mice. Filter binding and gel shift assays indicated that the affinity of MBNL1 for Tnnt3/T5.45 lies between that of (CUG) and (CAG). While overexpression of either CUG or CAG repeats induces the formation of nuclear foci in cell culture, it is interesting to note that we did not observe RNA foci formation following Tnnt3 minigene overexpression. This observation argues that high affinity MBNL1–RNA interactions together with abundant expression of MBNL1 target RNAs is not sufficient for ribonuclear foci formation. Of course, MBNL1 may be cleared from splicing target, but not pathogenic, RNAs during RNA processing and nuclear export or unusual interactions between MBNL proteins and CUG RNA might drive foci formation. In support of the latter possibility, we provide EM evidence that MBNL1 forms a tandem ring structure when bound to CUG RNA. The size of the rings is uniform with a diameter of ∼18 nm and since the MBNL1 isoform employed for these studies is 41 kDa, the ring structure must be an oligomeric complex. At a protein:RNA ratio of 2.5:1, most CUG helices were bound by a single ring while the majority of these hairpins were bound by at least two rings at a higher protein:RNA ratio. It is not clear if the MBNL1 ring contains a hole but if a central cavity exists it might allow threading of dsRNA. When multiple rings were bound to a single RNA molecule they appeared to be tandemly stacked suggesting either a preferential ring-loading site or potential ring–ring interactions. The latter possibility is supported by our finding that MBNL1 self-interacts via its C-terminal region both in the yeast two-hybrid system and in mammalian cells. In contrast, the MBNL1 N-terminal region encompassing the CCCH RNA-binding motifs fails to interact with full-length MBNL1 although RNA-binding activity is comparable to the full-length MBNL1 (data not shown). In this regard, it is interesting to note that the rings formed using MBNL1 N, lacking the C-terminal region, were less uniform in size and there was less ring–ring stacking at higher protein:RNA ratios. A similar situation has been noted for the protein Hfq (Host factor 1) which is a single-strand RNA-binding protein involved in the translational regulation and stability of several RNAs (). As visualized in the EM by negative staining, Hfq forms hexameric rings. Similar to MBNL1, the RNA-binding activity of Hfq resides in the N-terminal region and rings are still formed by a C-terminal truncated protein although they are less stable. Does MBNL1 exist as a ring structure when bound to its normal RNA splicing targets? To address this question, we performed EM analysis of the MBNL1–Tnnt3/T5.45 complex. Although MBNL1 rings were observed, the result was inconclusive due to the difficulty in visualizing small and partially single-stranded Tnnt3/T5.45 RNA by EM. However, MBNL1 forms large complexes with both CUG and Tnnt3/T5.45 RNAs (C) so it is possible that MBNL1 forms a ring structure when bound to splicing regulatory sites. Nevertheless, these normal binding targets do not contain the tandemly arrayed MBNL1-binding sites present on pathogenic CUG RNAs which we propose are essential for multiple ring interactions, MBNL1 sequestration and ribonuclear foci formation in DM cells. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Before a cell divides it must first faithfully copy its DNA so that each daughter cell receives a full complement of genetic information. In the eukaryotic cell cycle chromosomal duplication is restricted to S phase, and chromosomal segregation to M phase, with these phases separated by the gap phases G1 and G2 (). Checkpoints are important for controlling these events, delaying the cell cycle in response to incomplete DNA replication or DNA damage (,), and are essential for the maintenance of genomic integrity and to prevent ploidy changes. Failure to arrest the cell cycle in these situations contributes to cancer development and may cause resistance to standard treatments (). The mitotic cell cycle and checkpoint controls of the fission yeast resemble those of higher organisms with distinct G1-S-G2-M phases and with mammalian homologues identified for the major fission yeast checkpoint genes (). Activation of the cyclin dependent kinase Cdc2 is required for entry into mitosis, and inhibitory phosphorylation on tyrosine 15 by Wee1 and Mik1 maintains the Cdc2-cyclin complex in an inactive state until dephosphorylation occurs through Cdc25 (). Six genes have been identified that are involved in activating the checkpoint response by an S phase block (the S-M checkpoint) or DNA damage, that are collectively referred to as the checkpoint genes () (). Cells mutated in these genes are hypersensitive to DNA damage and to hydroxyurea (HU), an inhibitor of DNA replication (,). Rad3 phosphorylates and activates the Cds1 kinase after a DNA replication block, and also phosphorylates and activates the Chk1 kinase after DNA damage, although Chk1 can substitute for Cds1 in a DNA replication block (,). Checkpoint activation blocks mitotic onset by decreasing the activity of the mitotic Cdc2/cyclin complexes as a consequence of Cdc2 tyrosine 15 phosphorylation. Cds1 also plays a role in recovery from the S phase arrest, and in an HU-induced S phase block cells lacking Cds1 lose viability (,). Both Chk1 and Cds1 phosphorylation and activation are dependent on the presence of Rad3 and other checkpoint Rad proteins (). However, the sequence of upstream events leading to activation of Rad3 is unclear with the identity of the DNA damage/incomplete replication sensors and initial checkpoint activators remaining uncertain. One possibility is replication protein A (RPA) coated single-stranded DNA (ssDNA), which is a common structure generated at sites of DNA damage and is thought to play a role in recruiting two key complexes to damaged DNA: the human ATR (M- and ad3-related) protein complexed with its regulatory partner ATRIP, and the Rad17 complex (,). We set out to find new genes or a new role for known genes involved in checkpoint activation upstream of Rad3. We postulated that overexpression of such genes would ectopically activate the replication and/or damage checkpoint. In the presence of Rad3 this would activate the checkpoint and arrest the cell cycle, but in its absence cells would grow and divide normally. In this study we show that increased levels or stabilization of Cdc18/CDC6, a key factor initiating DNA replication in eukaryotes, induce a Rad3-dependent checkpoint response in the apparent absence of DNA replication structures, and we investigate how this system operates. All experiments were carried out in EMM2 minimal media unless otherwise stated. Growth conditions and DAPI staining are as described previously (). The mutant strain () growing exponentially at 25°C was shifted to 36°C to arrest cells in late G2. The and Δ strains were a gift from Tony Carr, and the Δ strain was constructed for this study. The and Δ were a gift from Shao-Win Wang. The Δ was a gift from Pablo Hernandez. Thiamine-regulatable promoters (,) were repressed using 5 μg/ml of thiamine. Hydroxyurea (HU) was used at 12 mM. Flow cytometry analysis were performed on a Becton–Dickinson FACScan using propidium iodine staining of fixed cells as described previously (). Deletion and genomic tagging were done as described previously (,). lists the -containing strains constructed for this study. A Gateway compatible Lifetech library was constructed from total RNA derived from mitotic, meiotic and shmooing cells in a 2:1:1 ratio, within a Gateway modified version of the ura4 based pRep4X vector (T. Duhig personal communication). The library was transformed into    (CCL3) using the modified Lithium Acetate protocol () (A). Briefly, the transformed cells were plated out and left to grow for 5 days at 25°C under selective conditions in the presence of 15 µM thiamine to suppress gene expression. They were then replica-plated onto selective media, containing Phloxin B (to aid identification of sick elongated cells), at 25 and 36°C in the absence of thiamine to allow expression of the gene controlled by the promoter. After 24 h, 80 000 colonies were screened to identify colonies showing elongation at 25°C and normal growth at 36°C. Eight colonies were identified and plasmids containing , and were recovered, suggesting Rad3-dependent checkpoint roles for these genes. A PCR fragment of the gene encoding the mutant protein (with six threonine residues mutated to alanine, at positions 10, 46, 60, 104, 134 and 374) was used to replace the endogenous gene. The PCR reaction was performed using a plasmid containing the mutant as template and covered the entire open reading frame plus an additional 114 bp upstream and 367 bp downstream (). The 2300 bp PCR product was purified and transformed into a homozygous diploid strain (CCL2) where one copy of the open reading frame was replaced by (). Cells were replica-plated on rich medium containing 5-FOA after overnight incubation on rich medium without selection. Clones with successful recombination at the locus were screened for by PCR and the diploid sporulated to generate haploid mutants. It should be noted that this strain could not be constructed in a background. The gene is linked to the gene, so the mutation here is linked to . In subsequent crosses this linkage was exploited to select for the mutation, which was then confirmed by PCR. The gene was also TAP tagged by gene targeting at the locus as described previously (,). In this case G418 resistance was used to follow the mutation. For western blotting cells were boiled for 6 min after being washed once in STOP buffer (150 mM NaCl, 50 mM NaF, 1 mM NaN, 10 mM EDTA pH 8.0). Protein extracts were prepared using glass beads in HB () and a fastprep machine (Bio101). Protein concentration was determined using the BCA kit from Sigma. About 50 μg was separated on 10% SDS-PAGE and blotted on nitrocellulose membrane (Amersham Hybond ECL) and detected by ECL (Amersham). The antibodies used were PAP (Sigma) at 1:1000, monoclonal anti-HA (Babco) at 1:1000 and polyclonal anti-Cdc18 antibodies () at 1:1000. Approximately 8 × 10 cells were harvested by filtration and washed once with 50 ml of ice-cold nuclear isolation buffer (NIB; 50 mM MOPS pH 7.2, 150 mM KAc, 2 mM MgCl) with 0.1% sodium azide, then washed again with 50 ml of NIB alone. Cell pellets were frozen at −70°C. Genomic DNA was purified from cells as described previously (). Approximately 15 μg of genomic DNA was digested with 80 units of restriction enzymes (HindIII and KpnI) in a 200 μl reaction for 2.5 h at 37°C followed by ethanol precipitation. In the first dimension, DNA was run on a 0.4% agarose gel in the absence of ethidium bromide in TBE, for 24 h at 1 V/cm at room temperature. The second dimension was run at 4°C in TBE buffer (containing 0.3 μg/ml ethidium bromide circulating at 50–100 ml/min) at 6 V/cm, until the arc of linears had migrated 8–10 cm. DNA was transferred to a membrane using standard Southern blotting. Hybridizations were carried out at 68°C with ∼2 × 10 c.p.m. of randomly primed probe. After washing, the membranes were exposed to BioMax film (Kodak). For analysis of whole chromosomes agarose plugs were prepared from mid log phase cells as described previously (). Plugs were loaded in 0.8% pulsed-field certified grade agarose gels (Biorad) and electrophoresed at 14°C in a CHEF-DR III pulsed-field gel apparatus (Biorad) at 2 V/cm for 72 h, using three blocks. Parameters were: Block 1; switch time 1200 s, 96° included angle, run time 24 h; Block 2; switch time 1500 s, 100° included angle, run time 24 h and Block 3; switch time 1800 s, 106° included angle, run time 24 h. For the Sfi1 restriction enzyme digest, the agarose plugs were pre-equilibrated for 30 min in NEB buffer 2 on ice (10 mM NaCl, 5 mM Tris-HCl pH 7.9, 1 mM MgCl, 0.1 mM DTT, 100 μg/ml BSA). Buffer was then replaced, and 100 units of Sfi1 (NEB) added. The plugs were incubated overnight at 50°C and then washed twice for 30 min in TE10X on ice and then loaded onto a 1% pulsed-field certified grade agarose gel (Biorad) in 0.5 × TBE. Electrophoresis was performed at 14°C at 6 V/cm for 24 h, using one block with a switch time of 60–120 s and a 120° angle. All gels were stained with ethidium bromide and visualized using a Dual Intensity Transilluminator. DNA was transferred to a membrane using standard Southern blotting. Hybridizations were carried out at 68°C with ∼2 × 10 c.p.m. of randomly primed probe. After washing, the membranes were exposed to BioMax film (Kodak). To identify genes that play a role in the S-M checkpoint control, we screened for cDNAs which when overexpressed blocked onset of mitosis in a Rad3-dependent manner. For this screen (A), a strain (CCL3) () was used where is active at 25°C and inactive at 36°C. Clones were isolated that led to a cell cycle block and consequent cell elongation at 25°C but not 36°C (B). One clone contained a plasmid expressing the gene and retransformation confirmed that cell elongation was only induced at 25°C when was active. Transformation into a wild type strain (CCL1) and a non-functional strain (CCL6) confirmed that cell elongation was independent of temperature and dependent on Rad3. Over-expression of Cdc18 to high levels induces re-replication resulting in giant nuclei and increased DNA (). Overexpression of the C-terminus of Cdc18 also induces re-replication and a Rad3-dependent block of mitosis, whilst overexpression of the Cdc18 N-terminus induces a Rad3-independent block of mitosis but is unable to induce re-replication. The effect of the Cdc18 N-terminus is thought to be the result of a non-physiological interaction with Cdc2 (see ) (). No re-replication was observed with the newly isolated plasmid as assessed by FACS, which was compared with the FACS profile of pRep3X -induced re-replication (C, D). Possible explanations for the lack of re-replication were that the cDNA was 16 base pairs shorter in the 5′-untranslated region (5′UTR) compared with the previously studied re-replication-inducing cDNA, or that the pRep4X vector was used rather than pRep3X. In support of the 5′UTR having an effect, re-replication was also not observed with the original cDNA isolate, which lacks the 5′UTR (). We speculated that this moderate level of Cdc18 overexpression might induce the S-M checkpoint without re-replication, and to investigate this further sought to increase Cdc18 level in a more controlled manner by modifying Cdc18 stability. Phosphorylation of the 6 CDK consensus sites by Cdc2 regulates the stability of the Cdc18 protein (,,) and over-expression of a mutant lacking these phosphorylation sites (Cdc18-T6A) results in increased re-replication (). We replaced the endogenous gene in a strain by a mutant gene encoding a protein with all 6 CDK sites mutated to alanine expressed by the endogenous promoter (see Materials and Methods section for strain construction). Incubation of this strain (CCL9) at 25°C resulted in cell elongation (A), but not at 36°C. No re-replication was observed by FACS, no enlarged nuclei were observed (A–C), and no replication intermediates were detected by 2D neutral/neutral gel electrophoresis when the strain was arrested at the G2/M transition in a mutant background (CCL13) (D). This result contrasts to that obtained when Cdc18 was overexpressed from pRep3X in a G2 block; in these cells 2D gel electrophoresis clearly demonstrated the presence of replication intermediates (). Western blotting of Cdc18 in the pRep3X , pRep4X and wild type strains demonstrated that re-replication occurred at higher levels of Cdc18 as described previously (), whilst Rad3-dependent arrest in the absence of apparent re-replication occurred at lower levels of Cdc18 (pRep4X and ) (). We conclude that moderately elevated levels of Cdc18, either by overexpression of wild-type Cdc18 or by stabilization of mutant Cdc18-T6A protein levels leads to a Rad3-dependent mitotic block in the apparent absence of re-replication. To investigate activation of the Rad3-dependent checkpoint in the presence of moderately elevated levels of Cdc18, we used a (CCL9) mutant to follow the dynamics of the Rad3-dependent cell cycle block. Colonies growing at 36°C were replica-plated to 25°C. After 5 h at 25°C widespread cell elongation was seen, although by 24 h the cells had reverted to a wild-type length. Cell elongation was investigated further by shifting liquid culture populations of both (CCL9) and control (CCL3) cells from 36 to 25°C, in order to impose the Cdc18-induced Rad3-dependent block, and following them for 7 h. After 7 h at 25°C, mutant septated cells were found to be 60% longer than wild-type septated cells (22.8 μm versus 13.9 μm); there were no differences in length at division at = 0 h. We conclude that the mutant leads to cell elongation and G2 delay. The above results indicate that the elevated level of Cdc18 seen in the mutant may be sending an S-M checkpoint signal in the absence of DNA replication intermediates. If correct, inactivation would abrogate the signal and release the cells from the cell cycle block. To test this, an asynchronous population of (CCL9) cells was blocked for 4 h at 25°C. The temperature was then raised to 36°C for 1 h to inactivate (see schematic in A). As predicted, cells were synchronously released from the cell cycle block and underwent a normal mitosis in the absence of Rad3 (A). This experiment demonstrated that Rad3 is required both to initiate and to maintain the checkpoint signal generated by the elevated level of Cdc18 in the mutant. A synchronous culture of (CCL12) was then used to test for activation of the downstream effector checkpoint kinases Cds1 and Chk1. After inactivation for 1 h at 36°C (see schematic in B), the culture was shifted back to 25°C for 2 h to re-impose the mitotic block. Protein extract samples were taken every 20 min and the phosphorylation status of Cds1 and Chk1 analysed by western blot. After shift back to 25°C when was active, only a small proportion of the Cds1 protein pool showed the altered mobility indicative of its phosphorylated active form () (data not shown). However, most of Chk1 was converted to a slow migrating form corresponding with its phosphorylation and activation () (B), suggesting that Chk1 is likely to be the main effector kinase responsible for the block. Strains were constructed to test whether Chk1 and the other checkpoint proteins were required for the block (CCL15-24). The (CCL9) strain was crossed with Δ, Δ, ΔΔ, Δ, Δ, Δ, Δ, Δ, Δ and Δ strains in a background. Several (also for ΔΔ) isolates of in all checkpoint mutant backgrounds were selected by random spore analysis. These were or , except Δ and Δ, which were only obtained in a background. This result suggested that the Cdc18-T6A-induced checkpoint could only be activated in a Δ or Δ background. To confirm this, the isolates for all the checkpoint mutants were plated at 36°C to inactivate and left to form colonies (72 h). The colonies were then replica-plated and incubated at 25°C before screening for cell elongation at 6 and 24 h. Normal wild type looking cells were seen in and deletion mutants showing that the checkpoint could not be activated in these mutant strains. In contrast, the absence of either Cds1 or its activating partner Mrc1 had no effect on the ability of Cdc18-T6A to activate the checkpoint and the cells became elongated (C). We suggest that an elevated level of Cdc18-T6A activates the checkpoint through the Chk1 kinase, that components of the Rad checkpoint protein network such as Rad3 are required for the cell cycle block, and that Cds1 and Mrc1 are not required for the Cdc18-induced checkpoint. Because Chk1 is the main effector kinase for the DNA damage checkpoint, we investigated whether the Cdc18-T6A protein was acting through either induced DNA damage or effects on S phase. The Cdc18-T6A mutant protein had no detectable effect on exponential cell growth, with the generation time for both control and mutant strains at 36°C being 165 minutes (calculated from logarithmic plots of cell number measurements). To determine its effects on S phase, synchronized cultures of both mutant (CCL13) and wild-type (CCL8) in a Δ:: background were prepared using a block and release procedure. No differences were observed in the timing of S phase initiation and completion, as seen on FACS analysis (A) and with cell number and septation index measurements (B), although the mutant protein was present at an elevated level during S phase and, in contrast to wild type, remained present throughout G2 (C). Next, a mutagenesis assay was performed to see if there was any increase in the forward mutation rate (FMR) of to () in the presence of the Cdc18-T6A mutant protein. (CCL4) and (CCL10) strains were grown to mid log phase and plated out onto YE5S media containing 5FOA for 3 days at 36°C. The FMR per 10 cells was 3.6 for and 2.8 for indicating that the elevated levels of Cdc18 in the mutant do not significantly affect the spontaneous mutation rate. A pulsed-field gel electrophoresis (PFGE) analysis of cells was performed to determine if the mutant Cdc18 protein had any effect on chromosome mobility. Unexpectedly, in the presence of Cdc18-T6A, chromosome III (Chr III) was not visualized with ethidium bromide staining in either cycling (CCL9) or G2 arrested (CCL13) cells (A), despite the normal appearance and behaviour of chromosomes I and II. Southern blotting and probing for chromosome III with both non-origin rDNA and ade6 revealed chromosome III present in the gel as a smear upwards from 3.5 to 7 Mb (B). Chromosome III behaved normally in the (CCL3) and Δ (CCL7) control strains. Synchronized cultures of mutant in a Δ background (CCL13) were prepared using the block and release procedure described above to determine if the chromosome III behaviour changed during the cell cycle. General chromosomal gel entry was reduced as expected during S phase due to the presence of replication intermediates, but chromosome III was not visualized with ethidium bromide staining at any point in the cell cycle (C). Southern blotting and probing for chromosome III with non-origin rDNA revealed a smear throughout the cell cycle (C). We hypothesized that if it was the presence of Cdc18-T6A that changes the size of chromosome III, then elimination of Cdc18-T6A would allow stabilization of the size of chromosome III. The mutant gene was crossed out of a Δ strain (CCL13) (here, the strain is TAP tagged and marked with the gene, this confers G418 resistance and was used to follow presence of the gene) and and Δ strains were selected by random spore analysis. Cycling cells from these strains were analysed by PFGE (A) followed by Southern blotting and rDNA probing (B). In 17 strains analysed (not all data shown), chromosome III was now found to form a discrete band with an approximate 1:1 split between a normal sized chromosome III (in 9 out of the 17) and increased (up to 7 Mb) sized chromosome III (in 8 out of the 17). We selected 4 of these strains (2 with a large chromosome III, and 2 with a normal-sized chromosome III) for further analysis. Each strain was cultured both overnight and for 30 generations before processing the cells for PFGE. Southern blotting and rDNA probing demonstrated that after 30 generations of growth, the normal-sized chromosome IIIs remained unchanged, but in contrast the larger chromosome IIIs returned closer to a normal size (C). In strains derived from the cross that still contained the mutant, chromosome III remained smeared as in the parent strain (data not shown). Thus, we conclude that the Cdc18 phosphorylation mutant protein induces changes in the size of chromosome III. These changes disappear with removal of the mutant Cdc18. It is therefore unlikely that an additional unknown mutation is producing the size variability of chromosome III. Chromosome III contains rDNA repeats found in 915 Kb (containing 73% of the repeats) and a 242 Kb (containing 27% of the repeats) Sfi1 fragments. To determine if changes occurred in this region, an Sfi1 digest of chromosomes derived from the strain (CCL9) in agarose plugs was performed. PFGE was carried out under different conditions, first to resolve smaller fragments (in the Kb range), and then to resolve larger fragments (in the Mb range). This was followed by Southern blotting and probing for chromosome III with non-origin rDNA. Probing for rDNA in the first run (A) showed the loss of both rDNA containing bands, and the second run (B) clearly demonstrated a smear extending from 915 Kb up to the 2 Mb region of the gel. We conclude that there is expansion in the restriction fragments containing the rDNA repeats within chromosome III in the presence of the Cdc18-T6A phosphorylation mutant protein. The rDNA repeat organization strongly influences the organization and localization of the nucleolus (). We used a GFP-tagged nucleolar protein Gar2 (CCL25), which localizes to the rDNA, to further investigate the observed changes in chromosome III (). The cells’ DNA was also DAPI stained to demonstrate the nucleus. In wild-type cells (CCL26), Gar2-GFP occupied a discrete region, distinct from the bulk chromosomal DNA (A). Abnormalities were observed in 150 of 200 mutant cells (75%), including enlarged staining and fragmentation (see arrows in A). These abnormalities were not observed in the wild type strain and suggest that in the presence of Cdc18-T6A there are changes in the nucleolar structure. Cdc18-T6A may activate the checkpoint by stimulating recombination within the rDNA repeats, leading to changes in copy number of the repeats. To see whether recombination was involved in activating the Cdc18-T6A-induced checkpoint, we tested several different situations where recombination frequency in the rDNA repeats was altered. In budding yeast, is necessary for homologous recombination (), and so we looked for the effects of deleting its homologue, , on the Rad3-dependent elongation of the mutant strain (CCL11) (B). We found that there was less elongation of the mutant strain in the Δ background (B), suggesting that the absence of recombination partially suppresses the elongated phenotype. Polar replication fork barriers (RBF) are a conserved feature of rDNA in eukaryotes, and rDNA copy number is maintained through recombination events associated with RBF (,). We speculated that inactivation of the RBF would suppress recombination in the rDNA repeats, and prevent activation of the Cdc18-T6A-induced checkpoint. rDNA contains three RBFs (RBF1, 2 and 3) and the termination protein Reb1 is required for arrest of replication forks at RBF2 and RBF3, but not RFB1 (). We found that deletion of reduced the extent of cell elongation seen with the mutant strain (CCL5) (B). This result suggests that the failure to stall replication forks at the RFB2 and RFB3 sites within the rDNA repeats has an affect on the ability of Cdc18-T6A to delay entry into mitosis. The Rqh1 helicase suppresses inappropriate recombination in fission yeast (), and deletion leads to considerable variation in the size of chromosome III (). strain was not viable, raising the possibility that Cdc18 is driving recombination events, which are lethal in the absence of the Rqh1 helicase. We suggest that recombination is contributing in part to the Cdc18-T6A-induced Rad3-dependent G2 delay, and that reducing rDNA recombination or the cause of recombination (that is replication fork stalling), partially relieves the G2 delay. Our results have shown that the phosphorylation mutant, , has moderately elevated levels of Cdc18 compared to a wild type strain. The presence of Cdc18-T6A leads to a transient activation of a Rad3-dependent checkpoint, blocking the onset of mitosis in the absence of detectable replication intermediates or DNA over replication. We found that moderately overexpressed wild-type Cdc18 also acts upstream of Rad3 in the checkpoint pathway to block the onset of mitosis. Cdc18-T6A acts via Rad3 to bring about Chk1 phosphorylation and activation, and the Rad checkpoint protein network is necessary for this cell cycle block. In contrast, Cds1 and Mrc1 are not required for the cell cycle block. In the absence of Rad3, the cells containing elevated Cdc18 protein levels have the same generation time as those containing wild-type Cdc18, there is no difference in the timing of S phase initiation and completion, and there are no gross effects on mutation rate. We conclude that the increased level of Cdc18 transiently activates the Rad3-dependent checkpoint through Chk1, with no other obvious effects on the cell cycle. However, one unexpected consequence of an elevated Cdc18 level in the mutant was an increase in the size of chromosome III, with expansion of Sfi1 restriction fragments, which contain rDNA repeats and with differences in the nucleolar structure, which contains the rDNA repeats. Cdc18 may be inducing genome wide replication at a low level, which is not detectable by FACS analysis or by 2D DNA gels. In the regions containing the rDNA repeats on chromosome III, the presence of a low level of replication bubbles could increase recombination and unequal cross-over events leading to an increase in size of the rDNA containing restriction fragment. We found that decreasing recombination, either by deleting or inactivation of the RFB, partially suppressed the cell elongation phenotype normally seen with the mutant, whilst increasing recombination, by deletion of in a mutant, causes cell death suggesting that high levels of recombination are lethal. These data suggest that recombination plays some role in activating the checkpoint. However, the reduction in recombination did not completely abolish checkpoint activation. This may be because some recombination still occurs in the mutants tested, or that Cdc18-T6A also activates the checkpoint independently of the effects on chromosome III. It is possible that the effect on rDNA repeat number due to the increased level of Cdc18 in the mutant may act through a mechanism involving cohesin. Proper sister chromatid cohesion enables DSB repair by recombination without any net change in the number of rDNA. Sister chromatids are held together by cohesin, a multi-subunit protein complex, from their synthesis in S phase until anaphase segregation. Cohesin loading in has been shown to be dependent on the formation of the pre-replicative complex: Cdt1, MCM2-7, ORC and CDC6 (). The Sir2 histone deacetylase required for heterochromatin formation acts to silence the rDNA locus, and is a negative regulator of DNA replication in budding yeast () suppressing rDNA copy number changes via effects on cohesin association. Cohesin dissociation leads to sister chromatid misalignment and subsequent upward expansion of the rDNA repeats (). Therefore, in our experiments the increased levels of Cdc18 present after S phase in the mutant could interfere with cohesin association leading to mis-alignment and upward rDNA expansion. It has also been shown in that perturbations in origin firing lead to lesions in the DNA, especially within the rDNA (). PFGE revealed high chromosome XII instability in ORC mutants, and the monitoring of abnormal initiation was found to be rDNA copy number-dependent. These findings led to the suggestion that the rDNA locus plays an important DNA damage surveillance role during the initiation of DNA replication. Although the cell cycle delay seen is likely to be due at least in part to the observed changes in chromosome III, we cannot exclude that Cdc18 has more than one effect and that the elevated level of Cdc18 may also directly activate the cell cycle checkpoint. With this model, activation of the Rad3/Chk1 dependent S-M block would be a direct consequence of elevated Cdc18, and is not due to effects on DNA replication, recombination or damage. Consistent with this view is the fact that cells containing Cdc18-T6A and lacking Rad3 appear to grow normally which would not be the case if damaging changes were occurring to the DNA. It may also be that the Cdc18-T6A mutant protein has other effects independent of Cdc18 stabilization because the protein cannot be phosphorylated. The replication origin has been mapped to a 600 bp region within the non-transcribed spacer region of the ribosomal DNA (rDNA) repeats upstream of the rDNA genes. It has been confirmed that origin activity is restricted to this sequence, and is not detected in other regions of the 10.9 Kb rDNA repeat. This is the most abundant origin in the genome, and hence chosen for the 2D DNA gels performed (,). The same probe was used as to identify the rDNA as a marker of the presence of chromosome III. As the changes in chromosome III appear to localize to the rDNA and low level replication leading to recombination is a possible cause of these changes, the fact that the origin probe did not pick up any replication intermediates in the presence of Cdc18-T6A in G2 arrested cells suggests that any ongoing replication is very low-level indeed. We have shown that a moderately elevated level of Cdc18, generated either by ectopic expression or by modifying Cdc18 stability, activates the Rad3-dependent checkpoint. We propose that this is the result of a low level of Cdc18-induced replication causing subsequent recombination and expansion of the rDNA repeat regions on chromosome III, but suggest other possibilities such as ectopic checkpoint activation in the absence of any replication or a Cdc18-T6A-specific effect on chromosome III.

In the human genome, a large number of genes spans hundreds of kilobases and is regulated by distant enhancers and locus control regions (LCRs) (,). Long-range LCR/enhancer function has been proposed to be mediated by a looping model, which postulates that the enhancer diffuses through the nucleoplasm to loop with and activate the target promoter, while the long intervening DNA that does not participate in enhancer function is looped out (,). Indeed, RNA trap and chromosome conformation capture (3C) techniques have shown recently that enhancers co-localize and thus loop with the distant, -linked genes in the nucleus (). However, co-localization of the enhancer with the promoter represents the final stage of enhancer/promoter interaction. It has not been established whether the enhancer reaches and loops with the distant target promoter by freely diffusing through the nucleoplasm space as postulated by the looping model or the enhancer is guided during its translocation by tracking along the intervening DNA to ultimately reach and loop with the promoter as postulated by a facilitated tracking model (,). We have used the human β-globin gene locus as a model system to study long-range enhancer function. The locus spans 100 kb of DNA and contains the embryonic ε-, the fetal Gγ- and Aγ- and the adult δ- and β-globin genes arranged in the transcriptional order of 5′ ε-Gγ-Aγ-δ-β 3′ (A). The LCR, defined by DNase I hypersensitive sites HS1-5 located far upstream of the globin genes, is absolutely required for transcriptional activation of the globin genes in erythroid cells (). While HS sites 1, 3, 4 and 5 possess weak or no enhancer activity, the HS2 site, located respectively 11 and 55 kb upstream of the ε- and β-globin genes, possesses prominent enhancer activity () and is able to activate transcription of the globin genes over the long distance (). We as well as others have found that in erythroid cells the HS2 enhancer initiates synthesis of intergenic RNAs from multiple sites within and downstream of the enhancer in the direction of the linked promoter and gene (). Interruption of the enhancer-initiated transcription by inserting a transcriptional terminator, the lac operator/repressor complex, in the intervening DNA between the enhancer and the promoter drastically diminishes enhancer function (). These results indicate that the enhancer-assembled protein complex including pol II in tracking and transcribing through the intervening DNA to synthesize intergenic RNAs plays a pivotal role in mediating long-range enhancer function. As opposed to the looping model, our finding was more consistent with a protein-tracking model: The pol II-protein complex assembled by the enhancer tracked from the enhancer through the intervening DNA synthesizing intergenic RNAs to reach the promoter and activate mRNA synthesis, while the enhancer DNA could be immobile since translocation of the enhancer DNA through the nucleoplasm to reach and loop with the promoter did not appear to serve any functional purpose. Insulators in the boundary areas of many eukaryotic gene domains can block enhancer function, when they are inserted between the enhancer and the promoter (). In the present study, we utilized the enhancer-blocking activity of the chicken HS4 insulator in the 5′ boundary of the chicken β−globin gene locus () to resolve the apparently paradoxical findings of enhancer co-localization/looping with the promoter and a tracking and transcription (T&T) mechanism of the enhancer-assembled protein complex. We inserted the chicken insulator in the intervening DNA between the HS2 enhancer and the ε-globin promoter in the 11 kb human ε-globin gene locus. We then analyzed intergenic transcription to determine whether the interposed insulator blocked the enhancer-assembled transcription complex from tracking and transcribing through the intervening DNA to reach and activate the ε-globin promoter. Chromatin immunoprecipitation (ChIP) assays were employed to determine if the HS2 enhancer complex indeed contained pol II and TBP required for intergenic transcription and whether the interposed insulator blocked pol II and TBP from tracking and transcribing through the intervening DNA to reach the promoter. The 3C assays were used to determine if the HS2 enhancer DNA (i) co-localized neither with the intervening DNA nor with the promoter as predicted by the protein tracking model or (ii) co-localized not with the intervening DNA but with the promoter as predicted by the looping model or (iii) co-localized with the intervening DNA and also with the promoter as predicted by the facilitated tracking model. To these ends, we created the LCR(+I) and LCR(−I) cell lines in human erythroid K562 cells using cre-loxp mediated recombination. The integrated LCR(+I) and LCR(−I) plasmids spanned the natural 11 kb human ε-globin gene locus either with or without the chicken HS4 insulator inserted between the HS2 enhancer and the ε-globin promoter. The ε-globin gene locus in the integrated plasmids and the endogenous genome was analyzed by transcription analysis, ChIP and 3C assays. The results showed that the HS2 enhancer complex, containing not only the associated proteins including TBP and pol II but also the enhancer DNA tracked and transcribed through the 10 kb intervening DNA to loop with the ε-globin promoter. The interposed insulator blocked the tracking and transcribing enhancer complex mid-stream and caused an apparent piling-up at the insulator site of both the enhancer DNA and the associated pol II and TBP. This blockage caused a decrease in both the looping frequencies of the enhancer DNA with the downstream intervening DNA and the promoter and the amounts of pol II and TBP delivered to the promoter. As a result, intergenic transcription from the intervening DNA and mRNA synthesis from the -linked gene were greatly diminished. These findings offered an explanation for the apparent paradox of enhancer looping and protein tracking in long-range gene activation and provided the first experimental evidence for a facilitated T&T mechanism of long-range enhancer function. See Supplementary Data. The LCR(+I) DNA was excised from the plasmid by Apal I and MluI digestions. Ten micro gram of the excised DNA, 0.5 μg of pCDneo plasmid and 20 μg of fragmented K562 genomic DNA, serving as spacer DNA to prevent tandem integration of the LCR(+I) DNA (), were electroporated into K562 cells. The electroporated cells were divided into three aliquots and grown in medium containing G418. From the LCR(+I) cell pools, 12 clonal lines were selected and expanded by limit dilutions. The clonal lines were transiently transfected with the expression plasmid CMV-Cre constructed in pEGFP-C1 to delete the insulator. The transfected cells after culture for 24 h were sorted by FACS; those expressing an order of magnitude increase in GFP were expanded into the respective LCR(−I) clonal lines. Southern blots of genomic DNAs and northern blots of RNAs purified with DNase I digestion were performed as described (,). GFP expression was calculated from the FACS dot plots and normalized with respect to the copy number of integrated plasmids; RT–PCRs were performed as described (,). The FACS and RT–PCR data presented were average results of two independent experiments. Transgenic zebrafish were created with procedures as described () and maintained according to the guidelines of the MCG Animal Care and Use Committee. ChIP assays were performed as described (). Chromatin was pulled down with antibodies against phosphorylated pol II, AcH3 and CTCF (Upstate 05-623, 06-599 & 06-917), TBP and nucleophosmin (Santa Cruz sc-273 & sc-5564). The 3C assays were performed as described (,). The data presented were average results of two independent experiments. See Supplementary Data. To determine if the ε-globin gene locus from the LCR HS2 site to the ε-gene was transcribed into polyadenylated RNAs, thus by pol II, we used RT-PCR to map the transcription status of the entire 11 kb locus. Previously, only segments of the ε-globin gene locus had been mapped (,,). A cDNA library was synthesized with an oligo dT primer from the polyadenylated RNAs in total K562 cellular RNAs. The cDNAs were then amplified by 21 primer pairs spanning the entire locus (A). Generation of all 21 RT–PCR bands indicated that the entire locus was transcribed into polyadenylated RNAs by pol II (B). In the transcription profile of the locus in K562 cells, the HS2 and HS1 regions spanned by primer pairs 3 and 8 were transcribed at very low levels (C), probably because the transcription complex assembled at the HS sites efficiently transcribed the DNA located downstream of it. In comparison, the DNA region located 17 kb 5′ of the HS2 enhancer and thus ∼4 kb 5′ of the globin gene locus was transcribed at an even lower level (C). In contrast, the −36 kb DNA region located near the 3′ end of an olfactory receptor gene (ORG) in the 5′ neighboring ORG domain () was transcribed at a relatively high level (C), probably because this ORG was transcribed in K562 cells. To determine if the HS2 enhancer initiated synthesis of one giant, contiguous transcript from the enhancer to the promoter as observed in the avian globin gene locus (), we used longer-range RT–PCR in which the primer pairs were spaced at increasing distances to determine the sizes of the intergenic RNAs. When the cDNA library synthesized by the oligo dT primer from polyadenylatedm RNAs was amplified with such longer-range primer pairs, PCR bands were detected only when the forward and reverse primers were spaced <3 kb apart (D). These bands were thus generated by RNAs shorter than 3 kb, which indicated that the intergenic RNAs were initiated from multiple sites and polyadenylated at sites <3 kb away. We next used directional RT–PCR, in which locus-specific forward or reverse primers instead of oligo dT were used in cDNA synthesis, to determine the direction of intergenic transcription. The results indicated that intergenic RNAs were transcribed in the LCR region exclusively in the sense direction from the HS2 enhancer toward the downstream globin promoter and in the 5.5 kb intervening DNA predominantly in the sense direction (E). Although anti-sense RNAs transcribed from the intervening DNA were detectable, their synthesis was apparently also dependent on the presence of the HS2 enhancer, since the 5.5 kb intervening DNA in the absence of the HS2 enhancer in integrated 5.5-GFP plasmid was not detectably transcribed (B). We do not yet know how and why the HS2-initiated transcription back-tracked to synthesize minor species of anti-sense RNAs from the intervening DNA. However, the results taken together indicated that the 11 kb ε-globin gene locus was transcribed by pol II predominantly in the sense direction from the HS2 enhancer through the intervening DNA to the promoter. Next, we utilized the chicken HS4 insulator to determine whether the insulator inserted between the enhancer and the promoter blocked enhancer function by blocking the directional, intergenic transcription. We first generated the LCR(+I) plasmid, containing a floxed insulator inserted in the 11 kb ε-gene locus between the LCR and the ε-globin promoter coupled to the GFP gene (A). The LCR(+I) plasmid was integrated into K562 cells. Two clonal lines LCR1 and LCR2(+I), containing respectively one and two copies of the integrated plasmid (B), were expanded. The insulator was subsequently deleted from these lines by transiently transfected Cre recombinase to generate the respective derivative lines LCR1 and LCR2(−I) containing a single loxp site in the intervening DNA (A). This strategy ensured that the ε-globin gene loci in the LCR (+I) and (−I) lines were integrated into identical host sites. Therefore, enhancer function in the presence and absence of the insulator could be precisely determined without the unpredictable position-of-integration effect exerted by different host sites on transgene expression. Southern blot showed that the LCR(+I) and (−I) plasmids were indeed integrated into identical host sites, since the integrated LCR(+I) and LCR(−I) plasmids after cleavage at a unique NheI site produced bands of an identical size (B). Comparison of the DNA sequences of PCR fragments amplified respectively across the floxed insulator in integrated LCR(+I) plasmid and the single loxp site in integrated LCR(−I) plasmid (C) showed that the Cre enzyme mediated precise and efficient recombination between the two loxp sites to delete the insulator in integrated LCR(+I) plasmid to generate the integrated LCR(−I) plasmid (see DNA sequences in C and Figure S1). FACS analysis of GFP expression showed that the interposed insulator reduced HS2 enhancer activity by 4–8-fold in the LCR1(+I) and LCR2(+I) lines (D). To determine if blockage of HS2 enhancer function by the interposed insulator was associated with blockage of enhancer-initiated intergenic transcription, we used northern blot and RT–PCR to map the transcription status of the integrated LCR(+I) and (−I) plasmids. Northern blot of RNAs transcribed from the LCR(−I) DNA confirmed RT–PCR analysis () that sense intergenic RNAs were transcribed from multiple sites in the 5.5 kb intervening DNA (B, see multiple bands marked by dots). Synthesis of the intergenic RNAs was driven by the HS2 enhancer, as the 5.5 kb DNA inserted alone in 5.5-GFP plasmid was unable to initiate intergenic transcription or activate GFP mRNA synthesis (B, lane 5.5). In the integrated LCR(+I) DNA, the insulator greatly reduced the levels of the intergenic RNAs and GFP mRNA [B, compare band intensities in (+I) and (−I) lanes], indicating that the interposed insulator obstructed transcription of intergenic RNAs and GFP mRNA. We next used RT–PCR to further confirm the ability of the insulator to block intergenic transcription. In the LCR(+I) plasmid, the RT–PCR band of 810 bp amplified by primer pair 2 spanning the insulator was not detectable [C, LCR2(+I), R lane]. This was not due to the inability of primer pair 2 to amplify a template of this length, as it amplified the 810 bp band from the integrated LCR(+I) DNA [C, LCR2(+I), D lane]. In the LCR(−I) plasmid, the single loxp site did not block transcription; therefore, the anticipated RT–PCR band of 364 bp amplified by primer pair 2 was detectable [C, LCR2(−I), R lane]. Interruption of intergenic transcription by the insulator in the LCR(+I) plasmid correlated with a reduction of ∼8-fold in the level of GFP mRNA [C, p.p.3 (+I) and (−I) lanes]. As the HS2 enhancer was transcribed at comparable levels from the integrated LCR(+I) and (−I) DNAs (C, p.p.1 lanes), the insulator did not inactivate the ability of the enhancer to initiate intergenic transcription. For the RT–PCR analysis, we used only three plasmid-specific primer pairs, as the 11 kb ε-globin gene locus in the integrated plasmid was identical in sequence to the K562 endogenous locus except for these three regions. To carry out a more comprehensive analysis, we generated LCR and LCR(+I) transgenic zebrafish in which the endogenous β-globin gene locus is grossly different from that of humans (A and B). Zebrafish express the general and erythroid transcription factors required for transcriptional regulation of the globin genes () and also the protein factor CTCF (GenBank BI883421) necessary for insulator function (). Analysis by RT–PCR with seven primer pairs distributed throughout the 11 kb locus showed that the interposed insulator reduced by 3–4-fold the level of both intergenic transcription and GFP mRNA (C). Note that primer pair 4 generated an RT–PCR band across the insulator after 43 PCR cycles whereas it generated no band at 35 PCR cycles. This observation indicated that although the interposed insulator blocked progression of pol II, a few pol II molecules could overcome the blockage to transcribe across the insulator. Together, RNA analyses of K562 clonal lines and transgenic zebrafish showed that the interposed insulator blocked enhancer function not by inactivating the enhancer to initiate transcription, as was observed earlier that an interposed drosophila insulator did not inactivate the enhancer (), but the interposed insulator blocked HS2 enhancer function by blocking the progression of the pol II-enhancer complex to transcribe intergenic RNAs (D). Transcription analyses ( and ) indicated that the HS2 enhancer assembled a pol II complex, which transcribed through the ε-globin gene locus to produce the polyadenylated, intergenic RNAs. We next used ChIP assays to determine if pol II, TBP and also acetylated histone 3 (AcH3) were associated with the HS2 enhancer and with the entire transcribed locus and if the interposed insulator, in blocking the pol II-enhancer complex, reduced the levels of pol II, TBP and AcH3 associated with the downstream intervening DNA and ε-globin promoter. In the ChIP assays, we used four plasmid-specific primer pairs: 5′HS2, 5′I, 3′I and εp-GFP (A). The 5′HS2 region was used as the reference for calculating the relative ChIP values of the other three regions. Since the transcriptional levels of the 5′HS2 DNA remained relatively constant in the integrated LCR(+I) and LCR(−I) DNAs (C, bottom panel, p.p.1 lanes and C, lanes 1), the assembly and thus the levels of the proteins associated with the HS2 enhancer complex should remain relatively constant in these different plasmids. The ChIP results showed that in the LCR(−I) plasmid, the levels of pol II, TBP and AcH3 associated with the loxp site in the intervening DNA and the ε-globin promoter were comparable to or higher than those associated with the 5′HS2 region [B, LCR (−I) panel]. Hence, pol II, TBP and AcH3 were associated with the HS2 enhancer as well as with the downstream intervening DNA and the ε-globin promoter in the LCR(−I) plasmid. In contrast, the insulator in the LCR(+I) plasmid drastically reduced the levels of pol II, TBP and AcH3 associated with the 3′I region and the ε-globin promoter [B, LCR(+I) panel]. Interestingly, very high levels of pol II and TBP were associated with the 5′I region [B, LCR(+I) panel], indicating an apparent piling-up of these proteins at the insulator site. To determine if the insulator also reduced the levels of pol II, TBP and AcH3 associated with the intervening DNA further downstream of the insulator, we carried out ChIP assays with three additional primer pairs, 6.3, 7.3 and 8.1, and a reference 5′HS2a primer pair (A). As these four primer pairs amplified the DNA templates in both the integrated plasmids and the endogenous K562 genome, the ChIP data were average levels of proteins associated with the 5.5 kb region in both the integrated plasmid and the endogenous genome. With this in mind, we carried out the ChIP assays first with the LCR(−I) plasmid. The results demonstrated that pol II, TBP and AcH3 were associated with the HS2 enhancer and the 6.3, 7.3 and 8.1 regions in the intervening DNA at comparable levels as in non-transfected K562 cells (C and Figure S2). In contrast, the levels of pol II, TBP and AcH3 associated with these three regions were consistently lower in the LCR(+I) DNA than in the LCR(−I) DNA [compare LCR(−I) and LCR(+I) panels in C and Figure S2A]. Thus, the insulator trapped the enhancer complex and reduced the levels of pol II, TBP and AcH3 associated with the further downstream 5.5 kb intervening DNA. These results indicated that the insulator obstructed a tracking mechanism of pol II and TBP in the enhancer complex. This obstruction caused pol II and TBP to pile up in the region upstream of the insulator and correspondingly reduced the levels of pol II and TBP associated with the downstream intervening DNA and the ε-globin promoter. Consequently, the intergenic DNA and the GFP gene were transcribed at greatly reduced levels and were packaged into relatively inactive chromatin with reduced levels of acetylated histones. As a result, the level of GFP was drastically decreased in the LCR(+I) lines, manifesting the observed blockage of HS2 enhancer function (D). It has been reported that the chicken HS4 insulator binds to CTCF (), which in turn binds to nucleophosmin located on the surface of the nucleolus (). This finding suggests that CTCF through binding to nucleophosmin could tether the insulator to a subcellular location and topologically separate the enhancer and the promoter into independent chromatin domains, thereby blocking direct interaction and thus looping of the enhancer with the promoter (). Hence, the blockage of enhancer-initiated T&T mechanism could be a secondary effect unrelated to the primary mechanism of enhancer blocking by the insulator. To investigate this possibility, we used ChIP to determine if the chicken HS4 insulator in the LCR(+I) plasmid bound to CTCF and nucleophosmin. The results showed that the insulator in the LCR(+I) plasmid bound to CTCF but not to nucleophosmin (D). The inability to detect binding of nucleophosmin to the integrated plasmids was not due to sample preparation or the ChIP protocol we used, since the endogenous LCR HS5 site containing a CTCF binding site () did associate with both CTCF and nucleophosmin (D). Thus, the interposed insulator in the LCR(+I) plasmid blocked HS2 enhancer function not by binding to nucleophosmin and potentially sequestering the enhancer and the promoter into separate topological domains to inhibit direct looping of the enhancer with the promoter but by obstructing a tracking mechanism of the pol II-enhancer complex mid-stream through the intervening DNA. According to the looping model of enhancer function, the HS2 enhancer DNA should co-localize with the ε-globin promoter but not with the intervening DNA. According to the protein tracking model in which the enhancer DNA was stationary but the enhancer-recruited pol II and TBP tracked and synthesized intergenic RNAs through the intervening DNA to reach the promoter and activate mRNA synthesis, the enhancer DNA should co-localize neither with the intervening DNA nor with the promoter. According to the facilitated T&T mechanism of enhancer function, in which the enhancer DNA together with the associated pol II and TBP tracked and transcribed through the intervening DNA to reach and loop with the ε-globin promoter, the enhancer DNA should co-localize temporally with both the intervening DNA and the promoter. To assess which one of these models mediated HS2 enhancer function in the ε-globin gene locus, we used the 3C technique (,) to determine the physical co-localization of the HS2 DNA with the 5.5 kb intervening DNA and the ε-globin promoter. To determine by 3C the co-localization frequencies of the HS2 DNA with the downstream DNAs in the integrated LCR(−I) and LCR(+I) plasmids, the nuclei isolated from the respective LCR2 clonal lines were digested with Xho I and Sal I at sites present only in the plasmid DNA to detect 3C bands produced specifically by the integrated plasmids. To prevent circularization in the ligation step of the Xho I and Sal I cleaved DNA fragments, the chromatin DNA in the nucleus was further fragmented with Bam HI and Bgl II digestions (A and S3). Completeness of restriction enzyme digestions and efficiencies of ligation reactions and of PCR amplification by different primer pairs were determined in control experiments (Figure S3). The 3C results showed that both the LCR(−I) and the LCR(+I) plasmids generated two HS2-GFP co-localization bands (B, HS2-GFP panel, −I and +I lanes). The authenticity of the 3C bands was confirmed by DNA sequencing (data not shown). The longer HS2-GFP band contained at the fusion site extra DNAs from the vector and the 3′ end of HS2. Generation of these two 3C bands indicated that the HS2 enhancer was physically near the GFP gene in both the LCR(−I) and the LCR(+I) DNA. However, the HS2-GFP bands generated by the LCR(+I) DNA were much weaker in intensity than those generated by the LCR(−I) DNA. These weaker bands did not result from a lesser amount of the LCR(+I) DNA template used in the PCR, since the GFP gene in both the LCR(+I) and LCR(−I) DNA templates generated PCR bands of similar intensities (B, G panel, −I and +I lanes). Therefore, the much weaker HS2-GFP bands indicated that the interposed insulator reduced the co-localization frequency between the HS2 enhancer and the GFP gene by ∼70% (C, HS2-GFP bar graph). In both the LCR(−I) and the (+I) DNAs, the HS2 enhancer also co-localized with the intervening DNA 5′ and 3′ of the loxp site or of the insulator (B, V-5′I and HS2-3′I panels). The authenticity of the V-5′I and the HS2-3′I co-localization bands was again confirmed by DNA sequencing (data not shown). In the LCR(−I) DNA, the V-5′I and HS2-3′I bands were generated, however, with six more PCR cycles than those generating the HS2-GFP band (B legend). This difference indicated that co-localization of the HS2 enhancer with the intervening DNA was ∼50-fold more dynamic and transient than the interaction of HS2 with the GFP gene (C, −I lanes). In support of the authenticity of the physical proximity between the enhancer and the intervening DNA, the co-localization frequencies of HS2 and the intervening DNA in the LCR(+I) DNA were distinctly different from those in the LCR(−I) DNA: The V-5′I band was stronger while the HS2-3′I band was weaker in the LCR(+I) DNA than the respective bands in the LCR(−I) DNA (B, compare +I and –I lanes in V-5′I and HS2-3′I panels). Furthermore, the HS2 enhancer did not produce a detectable 3C band with the un-linked CTCF gene (C and F). These results indicated that the HS2 enhancer DNA in the tracking enhancer complex co-localized with the -linked, downstream intervening DNA and the promoter but not with an un-linked gene domain. The interposed insulator obstructed this tracking mechanism of the enhancer DNA, thus causing the enhancer DNA to co-localize with the 5′I region at a higher frequency and with the 3′I region at a lower frequency in the LCR(+I) DNA (C). It could be argued that the V-5′I and HS2-3′I co-localization bands were generated not from enhancer tracking and thus enhancer co-localization with the intervening DNA but from direct interaction of the HS2 enhancer with the loxp site or with the insulator. To examine this possibility, we carried out 3C with the endogenous ε-globin gene locus in non-transfected K562 cells. Following digestions of the K562 nuclei with BamHI and Bgl II enzymes, 3C showed that the endogenous HS2 enhancer co-localized not only with the ε-globin promoter to generate the HS2-εp band but also with the 5.5 kb intervening DNA to generate the HS2-7.7 and the HS2-8.1 bands (D, upper left panel). Amplification of the 3C products by real-time PCR showed that the co-localization frequency between HS2 and the ε-globin promoter was higher than that of HS2 with the intervening DNA (D, bar graph). The increasing co-localization frequencies of the HS2 enhancer with DNA sites at increasing distances away was opposite to the decreasing co-localization frequencies predicted to occur due to random collisions () and indicated the specificity of the interactions between HS2 and the downstream intervening DNA and the ε-globin promoter. Moreover, the co-localization frequency between the HS2 enhancer and the DNA located upstream of the β-LCR or the GATA-1 and NF-YA genes located on different chromosomes were 70–400-fold lower than that between the HS2 enhancer and the ε-globin promoter (E) and thus 30–200-fold lower than that between HS2 and the 7.7 and 8.1 regions in the intervening DNA. These results again indicated the propensity of the enhancer to interact and co-localize with the -linked downstream intervening DNA and the promoter. To determine if the co-localization of the HS2 enhancer with the intervening DNA and the ε-globin promoter was characteristic of an active enhancer, we carried out 3C with non-erythroid HL60 cells in which the HS2 enhancer was inactive (). Both semi-quantitative PCR and real-time PCR of the 3C products showed that the inactive HS2 enhancer in HL60 cells co-localized with the intervening DNA and the ε-globin promoter at frequencies ∼400-fold lower than those in K562 cells (D, lower left panel and HL60 bar graphs). This very low co-localization frequency was not due to defective preparation of the HL60 sample, since within the CTCF gene that is active in both K562 and HL60 cells (), the DNAs in the first and the second introns co-localized with comparable frequencies in K562 and HL60 cells (F). Thus, compared to the active HS2 enhancer in K562 cells, the inactive HS2 enhancer in HL60 cells did not track through and co-localize with the intervening DNA nor did it co-localize and loop with the ε-globin promoter. In this study, we utilized the enhancer blocking activity of the insulator and a combination of RNA, ChIP and 3C analyses to investigate the mechanism of long-range gene activation by the HS2 enhancer in the ε-globin gene locus. Our results provided the first evidence for a facilitated T&T mechanism of long-range enhancer function: The HS2 enhancer complex, containing not only the associated proteins including pol II and TBP but also the enhancer DNA, tracked through the intervening DNA synthesizing overlapping, polyadenylated RNAs to ultimately reach and loop with the ε-globin promoter to activate mRNA synthesis from the -linked gene (). This facilitated T&T mechanism ensured that the HS2 enhancer complex during its translocation through the nucleoplasm space did not haphazardly interact with and activate heterologous promoters and genes located in but activated specifically the -linked globin promoter and gene. The finding that the transcribing enhancer complex contained not only pol II and TBP but also the enhancer DNA indicated that the proteins in the enhancer complex were perhaps unable to self-assemble and required presence of the enhancer DNA to be packaged into a functional, spaceo-specific transcription complex. Analysis by ChIP showed that pol II, TBP and acetylated histones were associated with the HS2 enhancer and also with the intervening DNA and the ε-globin promoter in both the integrated plasmids and the endogenous ε-globin gene locus. Indeed, earlier ChIP studies demonstrated that the HS2 enhancer recruits pol II and transfers it by an undefined mechanism to the downstream globin promoter (). It has been reported that pol II recruits histone acetyltransferase (HAT) to the transcribed locus () and that pol II transcription remodels the chromatin structure of the transcribed locus (). Thus, the HS2 enhancer complex with the associated HAT, in transcribing through the intervening DNA to reach the ε-globin promoter could acetylate the histones and open up the nucleosomal structure of the globin gene domain. Our 3C analysis showed that in erythroid cells in which the ε-globin gene locus was transcriptionally active, the enhancer DNA in both the integrated plasmids and the endogenous genome tracked through and progressively looped with the downstream intervening DNA to ultimately reach and loop with the ε-globin promoter (A–C). However, the HS2 enhancer did not co-localize with DNAs located upstream of the LCR or on different chromosomes nor did it co-localize with the inactive ε-globin promoter in non-erythroid cells. These findings indicated the tissue-specificity and directionality of the facilitated T&T mechanism of long-range enhancer function. The insulator inserted between the HS2 enhancer and the ε-globin promoter blocked mid-stream the tracking and transcribing enhancer complex containing the enhancer DNA and the associated pol II and TBP (D). The enhancer complex thus was stalled at the insulator site in many more cells than in cells in which the enhancer complex could track through and transcribe across the insulator and the downstream intervening DNA to reach and activate the promoter. Therefore, many more molecules of the enhancer DNA and pol II associated with the insulator than with the downstream intervening DNA and the ε-globin promoter. As a result, the intergenic RNAs and mRNA were transcribed at greatly reduced levels and expression of GFP was drastically diminished, thus manifesting blockage of enhancer function by the interposed insulator. The ability of the interposed insulator to trap and obstruct the directional progression of the enhancer complex toward the promoter was consistent with the observed directionality of insulator activity: the insulator blocks enhancer function only when it is inserted between the enhancer and the promoter but not when it is inserted upstream of both the enhancer and the promoter (,). The ability of the insulator to reduce the levels of acetylated histones associated with the downstream intervening DNA and the ε-globin promoter could be interpreted to indicate that the interposed insulator blocked enhancer function by recruiting histone deacetylase, thus causing the nucleosomes of the downstream intervening DNA and the ε-globin promoter to be less acetylated and to exist in an inactive state. This seemed unlikely. The HS4 insulator in the chicken β-globin gene locus has been shown to associate with high levels of histone acetyltransferases (), which are apparently able to acetylate histones and spread an open chromatin structure accessible to exogenous DNase I from the insulator throughout the downstream globin gene locus (). The possibilities that the interposed insulator sequestered the enhancer and promoter into separate chromatin domains () or the insulator decreased the flexibility of the chromatin fiber of the intervening DNA (), thus hindering direct looping of the enhancer with the promoter also did not appear to be the major contributing mechanisms of enhancer blockage. These possibilities were based on the premise that the HS2 enhancer acted primarily by a direct looping mechanism. This premise was not supported by our 3C results. In the human prostate specific antigen (PSA) locus, the enhancer has been suggested to loop directly with the promoter but the pol II recruited by the enhancer has been reported to track through the 6 kb intervening DNA to reach the promoter; yet the intergenic RNAs transcribed by pol II from the intervening DNA were not detected (). Thus, in the PSA locus the enhancer DNA and the associated proteins reached the promoter by separate, un-coordinated routes that did not involve pol II transcription to produce intergenic RNAs. However, in the human histocompatibility and growth hormone gene loci, transcription of the intergenic RNAs has been observed to be associated with long-range enhancer function (,). Large-scale transcription analysis of human chromosomes shows that over 90% of the DNA templates transcribed into RNAs are located in the intergenic regions (). The genome-wide transcription of the intergenic RNAs suggests that a facilitated T&T mechanism may mediate long-range gene activation not only in the globin gene locus but also in many other human gene loci. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Among all three domains of life, gene regulatory systems have evolved that do not require assistance of proteins and that basically act on the level of RNA. So-called ‘riboswitches’ are represented by non-coding regions of mRNA that selectively recognize metabolites (). Depending on metabolite concentration, two mutually exclusive structures are adopted according to the ligand-bound versus unbound state. This structural response is harnessed by the RNAs to lead to the sequestration or accessibility of Shine–Dalgarno sequences, the formation or destabilization of transcription terminator stems, the activation of ribozymes or the activation of splice sites. A riboswitch consists of an aptamer domain that upon ligand binding is stabilized and thereby alters the base-pairing arrangements in the adjoining expression platform. While the aptamer segments are highly conserved in sequence, the expression platforms are variable. The sequences in the platform determine the ‘ON’ or ‘OFF’ character of a riboswitch, meaning that gene expression is either turned on or off in case of ligand binding, and they typically determine the functional level (transcription, translation or splicing). The thiamine pyrophosphate-sensing riboswitch accounts for one of the earliest discovered representatives, and it is most widespread among bacteria, archaea, fungi and plants (). The very motif also exists in tandem riboswitch modules (,), and in all cases, this riboswitch class controls genes that are involved in the synthesis or transport of thiamine and its phosphorylated derivatives. The structure of the TPP-bound aptamer reveals a complex folded RNA in which one subdomain forms an intercalation pocket for the pyrimidine moiety of TPP, whereas another subdomain offers a wider pocket using bivalent metal ions together with water molecules to make contacts to the pyrophosphate moiety of the ligand () (). Despite the scores of detailed structural and biochemical studies on TPP riboswitches, the TPP-induced folding process of its aptamer and its full-length domains has not yet been kinetically investigated by biophysical methods, such as fluorescence spectroscopy. This is mainly due to the fact that selective labeling of RNA with the respective size of 50–200 nt is synthetically highly challenging. Moreover, the ligand TPP and its binding-competent analogs are non-fluorescent and thus cannot be applied for fluorescence studies on ligand-binding kinetics, as has been demonstrated previously in case of FMN () and purine riboswitches (), whose ligand FMN and ligand analog 2-aminopurine provide effective fluorescence emission. In the present study, we have assessed ligand-induced folding of the riboswitch by a fluorescence spectroscopic approach that we have previously applied successfully to rationalize folding of a different class of riboswitches (). For the TPP aptamer domain investigated here, we observe adaptive recognition of the ligand TPP and we reveal the temporal progress of tertiary structure formation. Moreover, while the full-length riboswitch domain remains completely responsive to TPP with kinetic parameters comparable to the aptamer domain alone, ligand binding to RNA variants that represent shorter transcriptional intermediates is hindered. This behavior can be rationalized by competing alternative conformations that are adopted during sequential folding and that are incompetent of ligand binding. The 72, 81, 82, 109, 125 and 151 nt RNAs containing site-specific 2-aminopurine labels were accessible by means of chemical solid-phase synthesis and enzymatic ligation using T4 RNA and/or T4 DNA ligase together with DNA and/or 2′--methyl RNA templates (), as previously described for 2′-methylseleno modified and AP-modified purine riboswitch domains (). All RNAs were purified by anion exchange HPLC under strong denaturating conditions (6 M urea, 80°C). The expected molecular weights of all RNAs were confirmed by LC-ESI mass spectrometry. Yields of typical RNA preparations (chemical synthesis and enzymatic ligation) were 30 nmol (∼800 μg; ∼27 OD) and 6 nmol (∼290 μg; ∼10 OD) of HPLC-purified U62AP and U62AP , respectively. All experiments were measured on a Cary Eclipse spectrometer (Varian, Palo Alto, USA) equipped with a peltier block, a magnetic stirring device and a RX2000 stopped-flow apparatus (Applied Photophysics Ltd., Leatherhead, UK). Our fluorescence-spectroscopic approach to study folding of the TPP riboswitch is based on a recently solved X-ray structure of its ligand-bound aptamer domain (). The structure reveals a number of nucleobases that can be substituted individually by the fluorescent nucleobase analog 2-aminopurine (AP) () without disturbing the overall fold (). Following the criteria of retaining hydrogen-bonding patterns and of maintaining highly conserved sequence portions, we selected seven positions of nucleobases that participate in crucial tertiary structure interactions of the aptamer, and synthesized the corresponding RNA variants, each of them labeled with an individual AP at a particular position (A69, G72, U79, U62, U46, A53, A85; ; B). Based on their fluorescence response upon addition of TPP, binding was confirmed for all variants and binding constants were in the submicromolar range at 25°C ( and Supplementary Figure S1), being well comparable to reported values for TPP riboswitch aptamers (,). Further evidence for the correct functionality of the AP variants came from the observation that their fluorescence emission did not change when thiamine monophosphate or thiamine was added in the same concentration range (10-fold excess over RNA). These control experiments with structurally closely related compounds of TPP confirmed the high specificity of the AP riboswitch variants for their dedicated ligand (Supplementary Figure S2). The concept presented here allowed us to follow using real-time kinetics the ligand-binding induced movement of selected nucleobases in the various subregions of the aptamer—in both a qualitative and quantitative manner. With this basis, we propose a detailed model for the ligand-induced folding process of the aptamer. Additionally, we investigated TPP-induced folding of the full-length riboswitch, as well as of shorter constructs representing transcriptional intermediates. The AP variants (C) provided insights into sequential folding of the riboswitch and into potential alternative secondary structure formation during transcription. The TPP/aptamer complex is organized by two parallel helical domains (P2/J3-2/P3/L3 and P4/P5/L5) connected to a helix (P1) by means of a three-way junction (A). We first explored the interaction which is most distant from the three-way junction, namely the tertiary contacts between L5 and P3. In the ligand-bound state, the loop nucleotide A69(L5) perfectly stacks between nucleotides A70(L5) and C24(P3) (B, left). Upon TPP addition to the free A69AP variant, we expected a prominent fluorescence decrease which is consistent with a movement of the nucleobase into its final stacked position and indeed observed this behavior (B, right). Under pseudo-first-order conditions, the rate constant was determined to be 14.2 ± 0.45 × 10 M s (). Another nucleotide which represented a good candidate for AP replacement in loop L5 was G72. In the ligand-bound state, this nucleotide is directed from loop L5 towards TPP and approaches its thiazol ring to ∼3 Å [see ref. (), Supplementary Figure S3B]. Upon TPP addition to the G72AP variant, we observed a defined fluorescence decrease and determined a rate constant of 12.1 ± 0.73 × 10 M s (). The crystal structure of the TPP/aptamer complex reveals that U62 at the interface of stems P4 and P5 (which shape the binding pocket for the pyrophosphate moiety of TPP) protrudes into solution and is completely ‘unstacked’ from neighboring nucleotides (C). Upon TPP addition to the free U62AP variant, we observed a prominent fluorescence enhancement which is consistent with the local twisting out of this particular nucleotide. Under pseudo-first-order conditions, the rate constant was determined to be 9.26 ± 0.16 × 10 M s (). Also nucleotide U79 which resides in the opposite strand of the P4/P5 interface is forced into an extrahelical position and protrudes into solution upon TPP binding (Supplementary Figure S3C). The rate constant measured for the corresponding U79AP variant was slightly higher ( = 12.0 ± 0.42 × 10 M s) (). The 5′-helical domain consists of stems P2 and P3 with junction J3-2 shaping the binding pocket for the pyrimidine ring of TPP. Junction J3-2 consists of the sequence 5′-UGAGAA adopting a highly complex fold making direct hydrogen-bonding contacts (G40) and direct stacking interactions (G42/pyrimidine/A43) to the pyrimidine ring of TPP. The arrangement is further characterized by non-canonical base pairs, A43·U39 and G42·G19, and by water-mediated hydrogen-bonding networks (A41·G18). In principle, one of the J3-2 nucleotides (A44) could be substituted by AP without major structural impairment. However, we decided not to even slightly perturb this highly conserved recognition element and selected the adjacent single nucleotide bulge U46 (of stem P2) for a replacement by AP to detect conformational changes within the 5′-helical domain. Like U62 and U79, U46 is unstacked in the ligand-bound state and protrudes into solution (Supplementary Figure S3D). Upon TPP addition to the U46AP variant, we observed a defined fluorescence increase and determined a rate constant of 7.53 ± 0.23 × 10 M s (). The three-way junction connecting stems P1/P2/P4 comprises junction J2-4 and is stabilized by two stacked tetrades in the TPP-bound state. A close-up shows that A53 forms a non-canonical base pair with A84 via the Hoogsteen face and that the nucleobase is sandwiched between U52 and G83 [D, ref. ()]. The corresponding A53AP variant shows a weak but defined fluorescence enhancement. We determined the rate constant to be only 3.85 ± 0.14 × 10 M s (). Additionally, the AP replacement at position 85 allowed us to monitor the structuring of the three-way junction via formation of the first base pair of stem P1 (A85:U14) (E). The rate constant of 2.30 ± 0.20 × 10 M s () measured for the corresponding A85AP variant was also significantly smaller compared to the rate constants for AP movements in the helical domains. Among the AP-modified aptamer variants studied, U62AP showed the most prominent fluorescence response during ligand-induced folding. This can be rationalized by the structural change that resembles single nucleotide flipping and unstacks the nucleobase. Because of its high sensitivity and because of its position at the binding site, U62AP represents the most indicative label to assess binding of TPP in a direct manner. In this sense, we synthesized as U62AP variant lacking all of stem P1 except the first potential base pair (A85:U14) (B). This variant did not respond to addition of TPP in 10-fold excess (and it hardly responded to a 100-fold excess of TPP) although all primary recognition elements for TPP are available within the two large helical domains. The simple connection of these domains by junction J2-4 hence does not provide sufficient entropic stabilization to support binding. Our fluorescence study on the TPP aptamer corroborates a folding model that is characterized by the adaptive recognition of TPP. Strategic positioning of AP fluorescent labels allows us to spectroscopically monitor the individual nucleobase movements in the various subregions until the final fold of the aptamer/TPP complex has been adopted. For each AP movement, an individual rate constant has been assessed and we find a remarkable 7-fold differentiation among the values. This range is large when compared with ligand-induced folding of the equally sized adenine riboswitch aptamer that we investigated previously and where rate differentiation was just ∼2-fold (). In the present case of the TPP aptamer, the pronounced differentiation in regional AP folding rates can be well rationalized (). The ‘fast’ rate constants concern AP replacements (G72AP, U62AP, U79AP) very close to the pyrophosphate recognition site in the 3′-helical domain (P4/P5). Variant A69AP—for which the highest rate constant is measured—also resides in the 3′-helical domain (L5) and is a sensor for formation of tertiary contacts between loop L5 (3′-helical domain) and stem P3 (5′-helical domain). The AP label of aptamer variant U46AP resides in the 5′-helical domain (P2, P3, J3-2) very close to the pyrimidine recognition site and is therefore appropriate to reflect its reorganization. U46AP gives a rate constant somewhat slower than G72AP and U79AP, but still comparable to the rate of U62AP at the pyrophosphate recognition site. Variants A53AP and A85AP whose AP replacements are accommodated in the three-way junction (J2-4, P1) lead to a subset of ‘slow’ rate constants; their responses reflect reorganization of J2-4 and completion of stem P1, respectively. In the free RNA aptamer, stem P1 seems to be partly preformed since the corresponding A12AP variant remains unaffected upon TPP addition (Supplementary Figure S3E), suggesting that the base pair AP12:U87 may already exist. In contrast, A85AP results in a defined fluorescence decay that can be attributed to formation of the stem-closing base pair (A85:U14) at the three-way junction. Importantly, variant U62AP which lacks stem P1 () did hardly respond to TPP. Formation of stem P1 is therefore a strict requirement for TPP binding although it is not directly involved in ligand recognition. Taken together, a ligand-induced folding model is proposed where fast recognition of the pyrophosphate moiety of TPP by the 3′-helical domain occurs almost simultaneously with recognition of the pyrimidine moiety by the 5′-helical domain. TPP acts like a clamp between the two large helical domains and supports—on the same timescale—formation of tight hydrogen bonding and stacking networks between interdomain (L5/P3) nucleobases that are distant from the three-way junction. Formation of the three-way junction and closure of stem P1 result from this initial recognition/folding process and require significantly more time to be fully accomplished. Based on this chronological formation of structure interactions, we favor the view that the large 5′- and 3′-helical domains of the free aptamer are preorganized in parallel orientation (). Previous biochemical and structural studies on the TPP riboswitch suggested a model on translational control that is illustrated in . In the absence of TPP, interaction of the anti-Shine–Dalgarno (SD) sequence (orange; nucleotides 108–111) with the anti-anti-SD sequence (red; nts 83–86) allows P8* pairing. This interaction permits the ribosome to access the SD element (blue; nts 126–129), and thus translation is on-regulated. In the presence of TPP, the obligate formation of stem P1 sequesters a portion of the anti-anti-SD element (red; nts 83–86) and by formation of stem P8, the SD-sequence (blue; nts 126–129) becomes inaccessible to the ribosome, and consequently, translation is off-regulated. Very recently, Famulok and coworkers () refined this model and provided evidence by chemical and enzymatic structural probing that the TPP-free full-length riboswitch is able to adopt a secondary structure with an extended 3′-helical domain that involves nucleotides (up to number 125) of the expression platform in base pairing with parts of the aptamer sequence. The structure is characterized by an ACCA tetraloop (nt 96–99) and a 40 nt long unstructured 3′-terminus (up to nucleotide 165) containing the SD sequence. The proposed fold implies that a subtle secondary structure equilibrium (,) exists between the extended 3′-helical domain (with the ACCA loop) and the binding-competent P4/P5/L5 domain required for phosphate recognition of TPP (Supplementary Figure S9). Originally, we considered three AP constructs highly potential for a kinetic assessment of the full-length riboswitch; these variants are U62AP , A12AP and A128AP (C and B). Titration of TPP to the U62AP 151 variant indeed resulted in a pronounced fluorescence increase, with a rate constant almost the same as observed for the U62AP aptamer alone ( = 8.66 ± 0.14 × 10 M s; = 8.13 ± 0.16 × 10 M s) ( and Supplementary Figure S8). The aptamer that is extended by the expression platform therefore remains fully responsive to its ligand. The other full-length variants, A12AP and A128AP , possess single AP labels at positions that are not directly sensitive to TPP binding but that are expected to be sensitive to the proposed TPP-induced formation of stems P1 and P8, respectively. Unfortunately, for both variants, hardly any fluorescence change was detected upon TPP titration. A likely explanation is that the full-length riboswitch construct investigated in this study adopts a conformation that is already very close to the TPP-bound conformation with stems P1/P8 mostly developed as depicted in B (Supplementary Figure S9). An alternative explanation is that the AP modification may reside in two chemical microenvironments that cause it to give a nearly identical fluorescence report although two different overall global folds are involved (see conformations C, D and D·TPP in Supplementary Figure S9). When we investigated shorter than full-length riboswitch constructs which mimic transcriptional intermediates (C and ), the AP approach provided insight into the obligate interplay between stems P1/P8 versus P8* (A). For these experiments, we used the highly sensitive label at position 62 at the TPP recognition site. Variant U62AP lacks the SD element, and not unexpectedly, we observed that TPP binding was significantly hampered as can be deduced from the weak relative fluorescence change (A). This behavior can be rationalized by adoption of an alternative fold characterized through formation of a stable 3′-terminal stem-loop structure involving anti-SD (orange) and anti-anti-SD (red) pairing (C and Supplementary Figure S10). This structure comprises the characteristic ACC(A99)-loop segment corresponding to the fold described by Famulok. As soon as the SD element (blue) becomes available, represented by variant U62AP (D and Supplementary Figure S10), ligand-binding capacity is already significantly restored as reflected in a definite fluorescence increase upon TPP addition (A). This can be understood by the availability of a terminal 3′-sequence stretch, which competes for masking of the anti-SD element (orange), resulting in a shift towards the ligand-binding competent fold (D). When the 3′-sequence stretch is further extended—like in the derivative—almost complete TPP-binding capacity is retained, supported by formation of the larger and hence more stable 3′-stem-loop element (A and E, and Supplementary Figure S10). Taken together, the fluorescence experiments on transcriptional intermediate mimics, U62AP and , well support a finely balanced structure equilibrium that is required for the riboswitch's ability to function as genetic control device. In the present study, we have applied chemical synthesis and enzymatic ligation to obtain a large set of TPP aptamer variants with site-specific 2-aminopurine labels at positions that are non-perturbing with respect to the RNA overall fold. These RNA probes allow to study TPP-induced individual conformational changes in the various regions of the aptamer by their fluorescent response. From the remarkable 7-fold differentiation of the rate constants, it can be deduced that recognition of the pyrophosphate moiety of TPP by the 3′-helical domain (P4/P5) occurs almost simultaneously with recognition of the pyrimidine moiety by the 5′-helical domain (J3-2), thereby tightening the interdomain nucleotide interactions (L5/P3) on the same timescale. Formation of the three-way junction and closure of stem P1 result from this initial recognition/folding process and require significantly more time to be fully accomplished. Moreover, the AP-labeled mimics of transcriptional intermediates provide insight into potential alternative secondary structures that are involved during transcription for the nascent RNA. Importantly, the full-length riboswitch domain recognizes TPP with kinetic parameters as performed by the aptamer alone. The present study on the TPP riboswitch and our previous studies on adenine riboswitches () demonstrate that the AP approach will be a powerful tool to kinetically characterize ligand-binding interactions and ligand-induced folding of other riboswitch classes as well. To reveal the details of riboswitch folding provides a basis for the elucidation of the associated response mechanisms ; folding studies of this kind are therefore of fundamental importance to understand riboswitch function in a comprehensive way (). p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

The -1 programmed ribosomal frameshift (PRF) is a non-conventional translation phenomenon that pertains to a particular change in the reading frame of the messenger RNA (mRNA) induced by a stimulatory signal. This strategy is mainly used by viruses to synthesize the precursor of their enzymes and to maintain a specific ratio between structural and enzymatic proteins (). In addition, -1 PRF is used during the translation of some prokaryotic and eukaryotic mRNAs (). One of the best-known examples of -1 PRF occurs when ribosomes translate the full-length mRNA of the human immunodeficiency virus type 1 (HIV-1) (). A -1 PRF is induced by two -elements within the mRNA: a slippery heptanucleotide X XXY YYZ (X is any nucleotide, Y is either A or U and Z is not a G in eukaryotes; spaces indicate the initial reading frame), where the -1 PRF occurs, and a following specific RNA secondary structure, the so-called stimulatory signal. This RNA structure is often a pseudoknot () but can also be a stem–loop, as it is the case for HIV-1 (), or a three-way stem–loop, as found in bacterial insertion sequences (,). The stimulatory signal controls the -1 PRF efficiency by making the ribosome pause over the slippery sequence (). However, pausing itself is not sufficient to promote -1 PRF (), and it was proposed that the stimulatory signal has a specific interaction with the ribosome. Altering the -1 PRF efficiency impairs viral replication () and it has been observed that even a small change in -1 PRF efficiency substantially handicaps the replication capacity of HIV-1 (). This indicates that the -1 PRF event could serve as a target for the design and development of new antiviral drugs. It is thus important to fully understand the -1 PRF mechanism. Several mechanistic models have been discussed in the literature to account for -1 PRF. Initially, Jacks () proposed a model of simultaneous slippage of the peptidyl-tRNA (pept-tRNA) and the aminoacyl-tRNA (aa-tRNA). In this model (A), pept-tRNA and aa-tRNA bound respectively to the XXY and YYZ codons in the ribosomal P/P and A/A sites unpair from the mRNA (the first and second letters represent, respectively, binding sites on the small and the large ribosomal subunit). The tRNAs and the ribosome shift towards the 5' direction and re-pair to the mRNA in the new reading frame. The Jacks model has been criticized because it ignores the fact that peptide bond formation, which occurs very rapidly (), leaves not much time for the shift to occur after the accommodation of the aa-tRNA in the A/A site. A refinement to the Jacks model had been provided by Plant (), who suggested that a movement of 9 Å of the anticodon loop of the aa-tRNA upon occupancy of the A/A site creates a tension on the mRNA because of the resistance to unwinding of the stimulatory signal. This tension would be relieved by the unpairing of the tRNAs, slippage of the mRNA by one base in the 3′ direction and re-pairing of the tRNAs in the new reading frame. However, the proposed 9 Å displacement of the anticodon loop of the aa-tRNA is not supported by structure analysis and by large-scale molecular dynamics (,). A second model was proposed by Weiss (), where -1 PRF occurs during the translocational step of the elongation cycle. Translocation proceeds in a stepwise manner and requires conformational changes within the ribosome (). In a simplified way, after peptide bond formation, the acceptor stem of the newly deacylated-tRNA (deac-tRNA) and of pept-tRNA move, respectively, from the P to E site and from the A to P site of the large ribosomal subunit. The resulting positions of both tRNAs are thus described as the P/E and A/P sites, respectively. In the next step, the anticodon stem–loop of the tRNAs moves to the E and P sites on the small ribosomal subunit, dragging the mRNA by one codon (,). Weiss () suggested that, when the tRNAs occupy these P/E and A/P sites, they can unpair from the mRNA, move in the 3′ direction with the ribosome and re-pair in the new reading frame. The anticodon stem–loops of the tRNAs then move with the mRNA to the E and P sites of the small ribosomal subunit (B). The analysis of an electron cryo-microscopy (cryo-EM) ribosome-mRNA pseudoknot complex stalled in the process of -1 PRF supports the hypothesis that this event occurs during translocation (), by showing that the elongation factor EF2 (eEF2), the eukaryotic homologue of EF-G, is bound to the stalled complex. However, it was observed that -1 PRF can be affected by mutations in the small subunit rRNA that alter the accommodation of the aa-tRNA in the A/A site () as well as by mutations in the elongation factor 1A (eEF1A) (,), the eukaryotic homologue of EF-Tu, that contributes to the accommodation process. These observations cannot be explained by any of the two previous models, but they are taken into account by a third model, which was initially proposed by Farabaugh () and further supported by experimental data from our group (). This model proposes that the -1 PRF occurs when aa-tRNA and pept-tRNA are located, respectively, in the A/T entry site and in the P/P site (C). Under these conditions, the -1 PRF has more time to take place, compared to the model of Jacks Still, some observations related to -1 PRF cannot be explained by this model. In particular, swapping a fragment of eight nucleotides located immediately 5' from the slippery sequence in the HIV-1 frameshift region with the corresponding fragment in the human T-cell leukaemia virus type 2 (HTLV-2) was found to decrease -1 PRF efficiency (). In this study, we reveal additional aspects that are important for -1 PRF, using the HIV-1 frameshift region as a model. In HIV-1, the slippery sequence is U UUU UUA () and the stimulatory signal is a two-stem helix, where an internal three-nucleotide bulge separates the lower stem from the upper stem–loop () (). This signal is separated from the slippery sequence by one nucleotide. We previously suggested that the lower stem favours a specific interaction between the upper stem–loop and the ribosome, when the stimulatory signal first encounters the ribosome. After this first contact, the ribosome progresses along the mRNA and unwinds the lower stem. Once the slippery sequence occupies the decoding site, the upper stem–loop, which is at a distance of eight nucleotides from this slippery sequence, acts as the effective frameshift stimulatory signal. It allows the -1 PRF to occur, involving the tRNAs interacting with the P- and A-site codons. With use of a dual-luciferase reporter system in which the HIV-1 frameshift region was inserted between the coding sequences for the () and firefly luciferase () such that the Fluc expression requires the -1 PRF of HIV-1, we demonstrated that mutations in the E-site codon can alter the -1 PRF efficiency. The same observation was made for other viral slippery sequences. In addition, since we had recently shown that the -1 PRF of HIV-1 can be recapitulated in and that prokaryotic ribosomes respond in the same manner as eukaryotic ribosomes to mutations in the HIV-1 frameshift stimulatory signal (), we used a -1 PRF system harboured in to undertake a random search of mutations in 16S rRNA that affect -1 PRF. We found three mutations of this kind, all located in the platform region of the small ribosomal subunit. This region is positioned close to the binding site of the E-site tRNA () and is known to be involved in conformational changes of the ribosome during translocation (,,). Based on these findings, we propose a novel model in which -1 PRF is triggered by an incomplete translocation and requires the slippage of not only the tRNAs interacting with the P- and A-site codons, but also of the tRNA occupying the E site. All primers used to generate constructs were purchased from Integrated DNA Technologies Inc. (Coralville, IA, USA). Plasmids used in frameshift assays in mammalian cultured cells are derivatives of pDual-HIV (–1) and plasmids described in Dulude (). The pDual-HIV (–1) plasmid contains the HIV-1 group M subtype B frameshift region inserted between the sequences coding for Rluc and Fluc, such that the coding sequence is in the –1 reading frame relative to the initiator codon. In pDual-HIV, is in frame with the initiator codon, by addition of an adenine 3′ adjacent to the slippery sequence, which is inactivated by mutagenesis to C UUC CUC. The Rluc expression is used to normalize the Fluc expression. The -1 PRF efficiency of each (–1) construct is obtained by dividing the Fluc/Rluc ratio by the corresponding ratio of the frame construct (). The M1 to M7 mutants and C1 to C6 constructs were generated by amplification of the mutated DNA from pDual-HIV (–1), with use of standard PCR procedures and a primer spanning the KpnI site, 5′-GCAGGGGGTACCTGGAAAGGAAGGACACCAAATGAAAGATTGTTCGAGAGACAG GCNNNNNNNNNNGGGAAGATCTGG-3′, (N corresponds to the mutated nucleotides as shown in A, A and B) and a primer spanning the Pfl23II site, 5′-GCCAACCGA ACGGACATTTCG-3′, for the forward and reverse reactions, respectively. The mutated DNA was subcloned into pDual-HIV (–1) previously digested by KpnI and Pfl23II restriction enzymes. Mutations in 16S rRNA (ΔG666, iC739 and G604A) were selected with a specialized bacterial ribosome system (,), using the p3RGFP-HIV (–1) plasmid, described in details elsewhere. Briefly, p3RGFP-HIV (–1) contains the operon under the control of the inducible promoter, as well as two reporter genes coding for the red (RFP) and green (GFP) fluorescent proteins. The DsRed T4 plasmid coding for RFP was a generous gift from Dr B.S. Glick, from the University of Chicago (), and the GFP coding sequence was obtained from the pGFPemd-N1 plasmid, a kind gift from Dr M. Bouvier, from the Université de Montréal. RFP is expressed by conventional translation whereas the HIV-1 frameshift region is inserted in the beginning of the coding sequence of GFP, so that its expression requires a –1 PRF. The ribosome-binding sites (RBS) of the reporters are changed to 5′-AUCCC and the messenger-binding site (MBS) of the 16S rRNA is changed to 5′-GGGAU, so that the reporters are exclusively translated by the ribosomes that contain the plasmid-encoded 16S rRNA (GFP/RFP system). Mutations iG666 and ΔC739 were introduced in the 16S rRNA by amplification of the mutated DNA fragments from p3RGFP-HIV (–1) with a two-step PCR approach encompassing restriction sites BlnI and DraIII, using an overlap extension procedure (). The resultant PCR fragment was subcloned into p3RGFP-HIV (–1) previously digested with the same enzymes. The specialized bacterial dual-luciferase system was created as follows: the to coding sequence encompassing the HIV-1 frameshift region in pDual-HIV (–1) and constructs was amplified by PCR, using the primer spanning the NsiI site, 5′-CTAGAGCCACC ATGCATACCAGCAAGG-3′, and the primer spanning the Pfl23II site, 5′-GTTTCATAGCTTCTGCCAACCGAACG-3′, for the forward and reverse reactions, respectively. The resultant PCR fragments were subcloned into the pRNAluc2 plasmid digested with NsiI and Pfl23II, as described in Bélanger (), generating pDual-HIV/P and (–1) plasmids (where P stands for prokaryote plasmids). The pRNAluc2 plasmid contains the operon under the control of the inducible promoter and a reporter gene coding for Fluc. As in the GFP/RFP system, the RBS of the dual-luciferase reporter and the MBS of the 16S rRNA were mutated so that the dual-luciferase reporter is exclusively translated by ribosomes that contain plasmid-encoded 16S rRNA. The 16S rRNA mutations in p3RGFP-HIV (–1) (ΔG666, iC739, G604A, iG666 and ΔC739) were cloned in pDual-HIV/P and (–1), using two ApaI restriction sites. Random mutations were introduced into the 16S rRNA fragment using a high-copy plasmid (pUC18, Fermentas). A BamHI–SacI fragment encompassing 16S rRNA from p3RGFP-HIV (–1) was cloned into pUC18 digested with the same enzymes, generating pUC18/16S. XL1-Red mutator strain (Stratagene) was used to produce random mutations into 16S rRNA. XL1-Red competent cells were transformed with pUC18/16S as directed by the manufacturer to obtain randomly mutated plasmids (pUC18/R16S, where R stands for randomized 16S rRNA) (,). The procedure was repeated seven times, so that more than 90% of the clones analysed contained a mutation in the 16S rRNA fragment. A second BlnI restriction site located in the beginning of the gene coding for 23S rRNA in p3RGFP-HIV (–1) was mutated to an AflII restriction site, using a standard mutagenesis procedure, so as to conserve a unique BlnI restriction site in the 16S rRNA gene. The 16S rRNA random mutant library from pUC18/R16S was cloned into p3RGFP-HIV (–1), using BlnI–DraIII restriction sites, which generated p3RGFP-HIV/R16S. Each isolated 16S rRNA mutation was re-introduced into p3RGFP-HIV (–1) to verify that the phenotype was conserved. Frameshift assays in eukaryotes were monitored by transient transfection of the dual-luciferase plasmids (pDual-HIV derivatives) into human embryonic kidney 293T cells (HEK293T) maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% (v/v) FBS (Wisent), with 2 × 10 cells/well seeded in 6-well plates. Two μg of plasmids were diluted in 3% of the initial culture volume in fresh serum-free medium and mixed with an equivalent volume containing 5 μg of PEI (polyethylenimine, Polysciences Inc.). The mixture was incubated for 15 min at room temperature prior to the addition to cell culture. Cells were cultured for 48 h before being washed twice with 2 ml of PBS, and lysed with 450 μl of the Cell Passive Lysis Buffer 1X (Promega). Fluc versus Rluc activities of each construct were measured for 10 s as relative light units (RLU) with an EG&G Berthold Lumat LB 9507 luminometer, using a non-commercial dual-luciferase enzyme assay system (). Dual-luciferase and GFP/RFP assays in were conducted as follows: overnight cultures of Top 10 strain (Invitrogen) were transformed with pDual-HIV/P or p3RGFP-HIV derivatives. Plasmids were grown in LB containing 100 μg/ml of ampicillin at 37°C. The cultures were diluted to an absorbance of 0.1 at 600 nm and incubated for 1 h at 37°C. The plasmid-encoded rRNA expression was induced with 1 mM of isopropyl-β--thiogalactopyra noside (IPTG) for 3 h for dual-luciferase assays, and 6 h for GFP/RFP assays at 37°C. The dual-luciferase assays were carried as the luciferase assays described in Bélanger (), except that the cells were lysed with 50 μl of the Cell Passive Lysis Buffer 1X and incubated for 15 min at room temperature. Fluc and Rluc activities were measured as described above. For GFP/RFP assays, 1 ml of cultured cells was centrifuged. The pellet was washed three times with 500 µl of PBS and re-suspended in 200 µl of PBS. Fluorescence was measured with a Fusion Universal Microplate Analyser (Fusion™ α-FP, Packard) at a 485 nm excitation wavelength (bandpass: 20 nm) for both GFP and RFP and at a 535 nm (bandpass: 25 nm) and 580 nm (bandpass: 15 nm) emission wavelength for GFP and RFP, respectively. As mentioned earlier, mutations within the eight nucleotides immediately preceding the slippery sequence can affect the efficiency of -1 PRF (). We may expect that the closer a nucleotide of this region is to the slippery heptamer X XXY YYZ, the higher its influence on -1 PRF. In this study, we focused on the three nucleotides positioned immediately upstream of the slippery heptamer, which we name A, B and C. Together with the slippery heptamer, these nucleotides form the decamer A BCX XXY YYZ, which we call the extended slippery sequence. In the last model describing -1 PRF [C, (,)], the XXY and YYZ codons of the classic slippery sequence (X XXY YYZ) are located in the P and A sites, respectively, when the -1 PRF occurs. The A BC nucleotides that are directly upstream of the slippery sequence are positioned such that the BCX and the ABC codons are located, respectively, in the E site before and after the shift. To investigate whether the identities of nucleotides A BC can affect the -1 PRF, we studied seven mutants of the frameshift region of HIV-1, in which the A BC nucleotide sequence ( U UUU UUA), was replaced by the corresponding sequence found in other viruses. We used a dual-luciferase reporter system () in mammalian cultured cells. In this system, the HIV-1 frameshift region is positioned between and coding sequences, such that Fluc expression requires the –1 PRF of HIV-1. In A, one sees the extended slippery sequences of the wild-type HIV-1 (WT; U AAU UUU UUA) and of seven mutants numbered from M1 to M7. The M1 mutant corresponds to a variant isolated from a protease inhibitor-exposed patient, and M2 is a natural variant of HIV-1 (). In M3 to M7 mutants, nucleotides A BC are taken from HIV type 2 (HIV-2; M3), the giardia virus (M4), the severe acute respiratory syndrome coronavirus (SARS-CoV; M5), the human T-cell lymphotropic virus type 1 (HTLV-1; M6) and the equine infectious anaemia virus (EIAV; M7). The absolute value of the WT -1 PRF efficiency was 8.0 ± 1.0%. In B, a value of 100% is arbitrarily assigned to this efficiency, and the efficiencies of all mutants are shown relative to WT. Among all investigated mutants, M1 is characterized by the lowest level of -1 PRF efficiency, which is only 5% of the WT level. Such a drop of the efficiency can be explained by the fact that M1 is the only variant in which the re-pairing of the pept-tRNA in the new reading frame is not allowed and this variant is most likely very poorly infectious. With M2, M5, M6 and M7 mutants, the -1 PRF efficiency was increased by about 50%. In contrast, with M3, the -1 PRF efficiency decreased slightly to 70% of the wild-type value, while the frameshift level remained unchanged with M4. Although the effects are not dramatic, they are all significant and well-reproducible [with M1, M2, M6 and M7 mutants; values are 0.0001 ( ⩾ 4) and with M5 mutant, -value is 0.001 ( = 4), as determined by Student's -test]. To verify whether the -1 PRF could be also influenced by mutations upstream the A BC nucleotides, we mutated the GCU codon, which is immediately upstream the BCX codon (GCU was exchanged with AAU, UCU, GGU or CAU, so that the A position remained unchanged). These mutations did not alter the -1 PRF efficiency (data not shown). We can conclude that mutations in positions A BC, immediately upstream of the classic slippery sequence, can noticeably affect the -1 PRF efficiency of HIV-1 and, therefore, that the nucleotides located at positions A BC could be involved in the -1 PRF of HIV-1. This was not the case for the nucleotides upstream of A BC. We next investigated whether the identity of the A BC nucleotides is also important for -1 PRF with other viral slippery sequences. To test this, we created various constructs in which the extended slippery sequence of HIV-1 (U AAU UUU UUA) was replaced with other extended viral slippery sequences. The two-stem helix of HIV-1 was used as a stimulatory signal with these different slippery sequences, which were assayed in the eukaryotic dual-luciferase system described above and are shown in A. The C1, C3 and C5 constructs contain the extended slippery sequences found, respectively, in giardia virus (C AUC CCU UUA), EIAV (U CCA AAA AAC) and in SARS-CoV (G UUU UUA AAC). The -1 PRF efficiencies obtained with these three constructs vary between 4% and 5%, about 40–60% of the HIV-1 -1 PRF efficiency (B). The nucleotides at positions A BC in each of the C1, C3 and C5 constructs were mutated to UAA, the nucleotides found at positions A BC upstream of the classic slippery sequence of HIV-1, generating the C2, C4 and C6 chimeric constructs. This decreased -1 PRF efficiency by about 50%, compared to the C1, C3 and C5 constructs, which confirms the observations made with HIV-1 slippery sequence concerning the importance of nucleotides A BC for -1 PRF. It was previously demonstrated that exchanging slippery sequences from different frameshift regions alters the -1 PRF efficiency (,). However, in these studies, only the heptanucleotide slippery sequence was exchanged. B also compares the effect of exchanging either the classic or the extended HIV-1 slippery sequence with other viral slippery sequences on -1 PRF efficiency. When the classic slippery sequence of HIV-1 is replaced with the classic slippery sequence of giardia virus, EIAV or that of SARS-CoV, the -1 PRF efficiency drops to 20–30% when compared to HIV-1 -1 PRF efficiency. When the HIV-1 extended slippery sequence is replaced with the extended slippery sequence found in giardia virus, EIAV or in SARS-CoV, the -1 PRF efficiency also decreases, but much less than when the substitution involves only the classic heptamer, being 50–60% of HIV-1 -1 PRF efficiency. These results confirm that exchanging slippery sequences alters the -1 PRF efficiency, and, in addition, shows that the extent of the decrease depends upon whether the extended slippery sequence or only the classic slippery sequence is exchanged. These results also confirm the importance of positions A BC for -1 PRF. We also attempted to identify rRNA mutants that could interfere with the -1 PRF of HIV-1. So far, the rRNA mutations that were found to influence -1 PRF perturb the accommodation process (). Our aim was to select mutations that influence -1 PRF by affecting the E-site tRNA binding, since we had found that mutations in the E-site codon can alter -1 PRF. Our strategy consisted in using the bacterial ribosome and introducing random mutations in 16S rRNA with the high-mutation-rate XL1-Red mutator strain (see Materials and Methods section for details). A pool of randomly mutated plasmids was obtained, where 90% of the clones analysed contained a mutation in the 16S rRNA. This 16S rRNA random library was then cloned into a plasmid containing reporter genes coding for RFP and GFP, in which RFP is expressed by conventional translation, whereas GFP expression requires the -1 PRF of HIV-1. The RBS of the reporters and the MBS site of the 16S rRNA random library are mutated and remain complementary, so that the reporters are exclusively translated by ribosomes that contain plasmid-encoded 16S rRNA. A small screen was individually performed (about 2000 clones), using the GFP/RFP assay to select clones for which GFP expression was affected but for which conventional translation was only moderately decreased (<50%). Three mutants were selected: mutant ΔG666 with a guanosine deleted from the stretch of guanosines between positions 666 to 671, mutant iC739 with a cytosine inserted into the stretch of cytosines between positions 735 to 739, and mutant G604A with guanosine 604 replaced by an adenosine. A shows the variations in HIV-1 -1 PRF efficiency obtained with each selected mutant relative to wild-type 16S rRNA, which was set arbitrarily at 100%, based on the GFP/RFP assay. ΔG666 decreased -1 PRF efficiency mildly (to 80%), whereas G604A and iC739 increased -1 PRF efficiency to 130 and 200%, respectively. We also created a bacterial dual-luciferase vector, similar to the one used in mammalian cultured cells, but in which the dual-luciferase reporter is exclusively translated by ribosomes containing the plasmid-encoded 16S rRNA. The 16S rRNA mutants were introduced in this vector and -1 PRF efficiencies were assessed from measurements of luciferase activities. With mutant G604A and iC739, the -1 PRF efficiency was increased to about 170%. However, surprisingly, the -1 PRF efficiency was also increased to 150% with mutant ΔG666, despite the fact that it decreased in the GFP/RFP assay. This discrepancy between the GFP/RFP and the bacterial dual-luciferase assay could result from the fact that the HIV-1 frameshift region is located at a short distance from the AUG initiator codon of the GFP reporter in the GFP/RFP system, and that the two reporters are translated separately. In the dual-luciferase assay, the HIV-1 frameshift region is inserted between the two reporter genes, so that the Fluc reporter, whose expression requires the -1 PRF of HIV-1, is synthesized as a Rluc-fusion protein. Although entirely hypothetical, one can propose that the changes in the -1 PRF efficiency triggered by the ΔG666 mutant could cause a drop-off of the short peptide resulting from the -1 PRF, with the GFP/RFP system. This could cause an apparent decrease in -1 PRF efficiency, a situation that does not occur with the dual-luciferase system. Because the ΔG666 and iC739 mutants are located in either strand of the same 16S rRNA helix, we used a standard PCR procedure to create the inverse mutation: iG666 and ΔC739. While iG666 had no effect on the -1 PRF efficiency in both GFP/RFP and dual-luciferase assays (data not shown), ΔC739 increased the -1 PRF efficiency to 160 and 230%, respectively, in both assays. We further investigated whether the 16S rRNA mutations also increase -1 PRF efficiency when combined with other slippery sequences. The iC739 mutation was investigated with the C5 construct that contains the extended slippery sequence found in SARS-CoV and the C6 chimeric construct, where the classic slippery sequence of HIV-1 is replaced with the one found in SARS-CoV, whereas the A BC nucleotides correspond to those found in HIV-1, as described in A. As with the HIV-1 frameshift region, the iC739 mutation increased -1 PRF efficiency with the other frameshift regions to about 150% compared to wild-type 16S rRNA. The locations of the mutations that were selected from the random library are shown in the secondary and tertiary structure of the 16S rRNA (). ΔG666, iG666, ΔC739 and iC739 are located in helix 22 whereas G604A is located in helix 21 of the 16S rRNA. Helices 21 and 22 are part of the central domain of the 16S rRNA, a region that forms the 30S subunit platform and participates in E-tRNA binding (see subsequently). Using a dual-luciferase assay in mammalian cultured cells, we demonstrated that mutating the BCX codon plus the preceding base (A) immediately upstream of HIV-1 classic heptanucleotide slippery sequence affects the -1 PRF efficiency. The BCX and the ABC codons occupy, respectively, the E site before and after the -1 PRF. We found that mutating the ABC nucleotides, located at the same position upstream of other viral slippery sequences also changes the -1 PRF efficiency. Interestingly, from the analysis of 41 eukaryotic virus slippery sequences retrieved from the RECODE database () (), we observed that there is a bias in the identity of the BCX codon located upstream of various slippery sequences: there may be different codons located immediately upstream of a given classic slippery sequence, but they appear to be exclusive for this slippery sequence, e.g. CCU and UUU codons are both found upstream of the slippery sequence U UUA AAC, but are not found upstream of any other viral slippery sequence (). No such bias is observed with the codon immediately 5' to this upstream codon. This observation is in agreement with our findings that mutating the E-site codon changes the -1 PRF efficiency, which is not the case when changing the codon 5' to the E-site codon. Using a specialized bacterial ribosome system, we also showed that mutations located in helices 21 and 22 of the 16S rRNA, in the platform region of the small ribosomal subunit, increase HIV-1 -1 PRF efficiency (). From the analysis of the ribosome crystal structure (,), it can be inferred that these mutations could influence the structure of a region involved in the binding of the tRNA at the E site (). Indeed, the mutations located in helix 22 are close to nucleotides from the 690 and 790 loops that contact the anticodon stem of the tRNA at the E site. These interactions require kink-turn motifs that are maintained by the coaxial helices 21 and 22. Therefore, the mutations that alter the -1 PRF efficiency could be related to the E site. Interestingly, mutations in the small subunit ribosomal proteins S7 and S11, which are located in proximity to the E site, have been found to increase spontaneous frameshift (), further supporting a relationship between this region and frameshift. When 16S rRNA mutations were assessed with different combinations of classic slippery sequences and upstream triplets (ABC positions), the -1 PRF efficiency increased to the same extent. Moreover, the platform region is involved in conformational changes during translocation (,,,). The 16S rRNA mutants could alter -1 PRF efficiency, not only by influencing the binding of tRNA at the E site, but also by interfering with translocation. It is worth mentioning that a 23S rRNA C2394G mutation, that affects translocation, increases spontaneous frameshift (), which also supports a relationship between translocation and frameshift. Also, cycloheximide, a translocation inhibitor (), was found to increase HIV-1 -1 PRF efficiency by about 2-fold (our unpublished data), which supports the involvement of translocation in -1 PRF. From previous observations and our own results, we propose a refinement of the last model describing -1 PRF. In this novel model (), the sequence of events that lead to -1 PRF starts when the BCX and XXY codons (AAU UUU in HIV-1 frameshift region) occupy, respectively, the P and A sites. After peptide bond formation, the newly deac-tRNA and pept-tRNA in the P and A sites become engaged in the translocation process. Translocation begins after the binding of EF-G (or eEF2) associated to a GTP molecule to the ribosome. After GTP hydrolysis, the acceptor stems of the tRNAs move towards the E and P sites, respectively, on the large subunit (). The tRNAs occupy hybrid sites, where the acceptor stems of the tRNAs on the large ribosomal subunit are in an intermediate position between the post- and pre-translocational state (P/E* and A/P*, where the star refers to a transition state). Such intermediate was recently identified using single-turnover rapid kinetics assays () (A). The tRNA anticodon stem–loops then move towards the E and P sites of the small ribosomal subunit, occupying first the intermediate E*/E* and P*/P* sites. The next step is the movement to the E/E and P/P sites, which drags the mRNA a distance of three nucleotides (i.e. one codon), after release of EF-G·GDP from the ribosome. We propose that, for a fraction of ribosomes, the two tRNAs cannot drag the mRNA by three nucleotides, but only by two, due to the presence of the stimulatory signal that is resistant to unwinding. As a consequence, the two tRNAs are blocked in intermediate E*/E* and P*/P* sites and translocation is incomplete (B). This hypothesis mainly relies on the cryo-EM structure from Namy (), showing that the presence of a frameshift stimulatory signal stalls the ribosome in the translocation process. The next step of translation is the arrival of an incoming aa-tRNA bound to EF-Tu (or eEF1A) associated to a GTP molecule. Because of the incomplete translocation, the incoming aa-tRNA occupies an entry site (A*/T*) that differs from the standard A/T entry site (C). Codon–anticodon interactions are dynamic and can break and re-form. We propose that the tRNAs, which are located in intermediate sites and not in their respective standard high-affinity sites on the ribosome, are prone to shift to these standard sites, and, after the shift, re-pair to the mRNA in the -1 reading frame (D). The driving force for the change in the reading frame is therefore the tendency of the tRNAs to occupy their standard binding sites. The slippage of the tRNAs would start with the tRNA located in the E*/E* site, followed by the successive slippage of the tRNAs located in the P*/P* and A*/T* sites, towards the E/E, P/P and A/T sites [see ref. () for the concept of successive tRNA slippage]. The -1 PRF would be followed by the aa-tRNA accommodation in the A/A site, which is coupled with the E-site tRNA ejection from the ribosome (,) (E) and conventional translation would resume with unfolding of the frameshift stimulatory signal. In our model, there are two steps where the ribosome pauses: when it is blocked during translocation and when the three tRNAs shift to their standard sites. As mentioned in the ‘Introduction’ section, Weiss () were the first to suggest that -1 PRF was linked to translocation, but in their model, -1 PRF occurs when the XXY and YYZ codons occupy the P and A sites (B). We suggest that a translocation anomaly leading to -1 PRF occurs at the preceding elongation cycle, when the BCX and XXY codons occupy the P and A sites, which takes into account the fact that the E-site codon participates in -1 PRF. In our model, three tRNAs participate in -1 PRF. The event is triggered by the blockade of the ribosome at a transition state following an incomplete translocation due to the presence of the stimulatory signal. How can we explain that the identity of the BCX codon influences the -1 PRF? It is well known that mutating the classic X XXY YYZ motif prevents the slippage of the tRNAs located in the P and A/T sites, since, after the shift, the tRNAs cannot re-pair to the mRNA in the new reading frame. However, whether there is a codon–anticodon interaction at the E site is still a matter of debate. From the analysis of crystal structures, no codon–anticodon interaction has been observed between the mRNA and the deac-tRNA in the E site (,). In contrast, translational experiments support a codon–anticodon interaction at the E site (,). From our slippery sequences analysis (), it can be observed that, after -1 PRF, only one standard base-pairing is possible between the tRNA and the ABC nucleotides, in most cases. If there is a codon–anticodon interaction at the E site, it could be suggested that base pairs other than Watson–Crick or G-U wobble pairs are tolerated at the E site. Under these relaxed conditions, a major consequence of mutating the BCX codon plus the preceding base is not likely to alter the -1 PRF efficiency by influencing the codon–anticodon interaction of the E-site tRNA in the shifted frame, although minor effects cannot be excluded. We propose that the main consequence of making these changes is to modify the identity of the E-site tRNA. Our hypothesis is that there is a relationship between the structural peculiarities of this tRNA and its capacity to shift from an intermediate to a classic site, since the movement of this tRNA precedes and thus controls the movement of the two other tRNAs. The group of Rousset () had proposed that, in -1 PRF, an E-site tRNA carrying a pseudouridine modification at position 39 was coupled to high -1 PRF efficiency. However, these results could not be reproduced (Rousset J.P. 2006, personal communication). Also, from a search using the tRNA Compilation 2000 database (), our present results do not support any relationship between the presence of a tRNA modification at position 39 and the -1 PRF efficiency (data not shown). In conclusion, we propose a novel model to describe -1 PRF, in which three tRNAs are involved. A detailed understanding of -1 PRF mechanism should contribute to the development of novel anti-frameshift agents that affect the replication capacity of viruses, such as HIV-1.

Providing a detailed description of networks of protein–protein interactions poses a formidable challenge in the post-genomic era (). An initial task in such an endeavor is the identification of interacting protein partners, which can be accomplished using readily available methods such as the yeast two-hybrid system (), tandem affinity purification of protein complexes () and computational predictions (). However, a detailed mapping of the interacting protein interfaces of identified protein partners currently lacks efficient and accessible molecular techniques. Currently, the means to specify regions involved in protein–protein interactions include mutational analyses (e.g. deletion series and alanine scanning), protein footprinting with proteases () or hydroxyl radicals (,), chemical cross-linking (), hydrogen–deuterium exchange experiments () and structural studies by NMR or X-ray crystallography. Each of these methods has certain drawbacks. Whereas some of them rely on time- and labor-consuming production of individual mutant variants and some may lack optimal resolution, others require highly specialized instrumentation and technical skills. Obviously, any methodology that could streamline the process of mapping protein–protein interfaces would be highly beneficial. Transposable elements are indispensable tools in modern genetics, and their ability to insert essentially randomly into DNA enables the generation of exhaustive insertion mutant libraries (). One of the most versatile DNA transposition tools is the reaction derived from bacteriophage Mu transposition (,). This system requires only a simple reaction buffer and three purified macromolecular components: transposon DNA, MuA transposase and target DNA (typically a gene of interest cloned in an appropriate plasmid). The reaction is highly efficient with relatively low target-site selectivity (,). These characteristics make the Mu reaction ideal for the generation of comprehensive mutant DNA libraries usable in a variety of molecular biology applications (). We devised a powerful Mu transposition-derived general strategy to accurately map regions involved in protein–protein interactions. This strategy combines the generation of a pentapeptide insertion mutant library (,), screening for altered protein–protein association on a yeast two-hybrid platform, and parallel analysis of mutant pools using a genetic footprinting technique. To demonstrate the feasibility of the system, we mapped the region in human JFC1 protein that is involved in the interaction with Rab8A. The Rab protein family, which belongs to the Ras superfamily of small GTPases, controls intracellular vesicular transport (). Rab8A appears to participate in polarized transport of proteins through reorganization of microtubules and actin (). JFC1 was identified as a Rab8A-binding partner in a yeast two-hybrid screen (). This protein belongs to the synaptotagmin-like (Slp) protein family, and it contains an amino-terminal conserved Slp homology domain (SHD), including subdomains SHD1 and SHD2 (). The protein also contains two tandem C2 domains () that are involved in Ca-dependent binding of phospholipids, targeting the molecule to the plasma membrane (,). The JFC1/Rab8A interaction has been verified by (co-localization and co-transfection/precipitation) and (pull-down) analyses (). In this study, we initially generated a comprehensive JFC1 mutant library with random five-amino acid insertions. The mutants were then screened in the yeast two-hybrid system and divided into pools on the basis of Rab8A-binding characteristics (strong, weak and no binding). Finally, the respective insertion sites were localized at nucleotide level accuracy by genetic footprinting. Our detailed analysis of the JFC1/Rab8A interaction revealed that the SHD1 region of JFC1 is the main mediator of Rab8A binding. Overall, the strategy provided a convenient general means to accurately map interacting regions in protein partners. The fully optimized system is readily applicable to any protein-encoding gene. Plasmids were isolated using appropriate kits from QIAGEN. Standard DNA techniques, including 5'-labeling with T4 polynucleotide kinase and [γ-P]ATP, were performed as previously described (). The origins of proteins, oligonucleotides, and reagents are listed in Table S1. DNA-modifying enzymes were used as recommended by the supplier. Marker sequencing ladders were each produced by the use of the Sequenase 2.0 sequencing kit (USB) and an appropriate primer. strain DH10B () was grown in Luria Broth (LB) (), and supplementary antibiotics were used at the following concentrations when required: kanamycin (Km, 25 µg/ml) and chloramphenicol (Cm, 10 µg/ml). Transposon cat-Mu(NotI) has been described (). Plasmid pEGFP-C1-JFC1 () contains the JFC1 coding region cloned between the EcoRI and XhoI sites in pEGFP-C1 (Clontech). Plasmid pMPH11 was made from pB42AD (Clontech) as follows: (i) The NotI site was removed by filling-in with Klenow enzyme and dNTPs. (ii) A gene encoding Km resistance (, from Entranceposon-Kan, Finnzymes) was PCR-amplified using the primers HSP464 and HSP465 (Table S2), and the generated PCR fragment was trimmed with ScaI and subsequently cloned (promoter-distal orientation) into the ScaI site of pB42AD. (iii) Prior to cloning, the gene was modified by introducing, via overlap PCR with appropriate primers (), a silent mutation (codon 11, Ser, TCG→TCT) to eliminate a critical XhoI site. Plasmid pMPH11-JFC1 contains the JFC1-encoding EcoRI–XhoI fragment from pEGFP-C1-JFC1 cloned between the respective sites in pMPH11. Plasmids pGildaB-Rab8ΔQ67L and pGildaB-Rab8ΔT22N are versions of pGilda-B (), and they encode the indicated Rab8A variants (,) with the respective genes cloned as previously described (). A JFC1 pentapeptide insertion mutant library was generated using the Mutation Generation System (Finnzymes) as specified by the supplier. This mutagenesis system exploits the MuA transposase-catalyzed transposition reaction () and generates 5-aa insertions in proteins (,). Five standard transposition reactions were performed, each with 300 ng of plasmid pEGFP-C1-JFC1 as a target. Following incubation at 30°C for 3 h, reactions were pooled. Reactions were extracted with phenol and subsequently with chloroform, and DNA was ethanol-precipitated and re-suspended in water (125 µl). Several aliquots (2 µl) were electroporated as previously described () into DH10B competent cells (50 µl) prepared as previously described (). Transposon-containing plasmid clones were selected on LB–Km–Cm plates. Approximately 1.1 × 10 colonies were pooled and grown in LB–Km–Cm medium at 37°C for 3 h. Plasmid DNA from the pool was then isolated, digested with EcoRI and XhoI, and subjected to preparative electrophoresis on a 0.8% Seaplaque GTG agarose gel in TAE buffer (). The 2.9-kb DNA fragment pool, corresponding to transposon insertions into the JFC1-encoding DNA segment, was isolated by electroelution and ligated into the plasmid pMPH11 digested with EcoRI and XhoI, and the ligation mixture was electroporated into DH10B cells as above. Transposon-containing plasmid clones were selected on LB–Km–Cm plates, and plasmid DNA was prepared from ∼3.8 × 10 colonies. The transposon core sequence was then eliminated from the plasmid pool by a cleavage with NotI, followed by preparative electrophoresis on a 1.7% Seaplaque GTG agarose gel and isolation of the plasmid backbone as above and recircularization by ligation at low DNA concentration (1 ng/µl). Ligated plasmids were electroporated into DH10B cells as above, clones were selected on LB–Km plates to generate the final JFC1 insertion mutant library, and DNA was isolated from ∼5.4 × 10 colonies as above. Taken into account the Poisson distribution during sampling, the final insertion library contained 2.4 × 10 independently generated mutants. The library DNA was also electroporated into DH10B cells to generate five sub-libraries (100, 500, 1000, 5000 and 10 000 colonies), and plasmid DNA was isolated as above following selection on LB–Km plates. Cumulative plasmid pools (representing 600, 1600, 6600 and 16 600 colonies) were generated by mixing appropriate library DNA preparations in suitable molar ratios. The yeast two-hybrid–protein interaction screen was performed using the MATCHMAKER LexA Two-Hybrid System (Clontech) according to the manufacturer's specifications. In this system, EGY48 ( -) harboring the plasmid p8op-lacZ serves as a reporter strain, into which interacting protein partners are introduced via successive transformation of appropriate expression plasmids (C). The insertion mutant library was introduced into the above reporter strain, and colonies were pooled based on their color. The pools were grown in SD–Ura–His–Trp medium for 3 h, the plasmid DNA was isolated from the pools with the Qiagen Plasmid Spin mini kit after vortexing with glass beads for 10–30 min in P1 buffer (50 mM Tris–HCl pH 8.0, 10 mM EDTA, 100 μg/ml RNase A), and the DNA was introduced into DH10B by electroporation as described (). The pMPH11-JFC1 mutant plasmids were selected on LB–Km plates. Each non-radioactive PCR reaction (50 μl) contained 50 ng plasmid DNA, 0.25 μM of HSP508 and HSP509 primers, 200 μM each dNTPs and 1 U Vent DNA polymerase (in New England Biolabs’ ThermoPol reaction buffer supplemented with 2 mM MgSO). Non-radioactive PCR conditions consisted of 5 min at 95°C followed by 25 cycles of 1 min at 95°C, 1 min at 60°C and 2 min at 72°C, and finally 5 min at 72°C. Non-radioactive PCR products were electrophoretically purified using the Qiaquick Gel Extraction kit. Each radioactive label-containing PCR reaction (50 μl) contained 50 ng of purified non-radioactive PCR product, 0.25 μM of each primer (biotinylated HSP488 and one radioactively labeled JFC1 gene-specific primer; listed as A-N in Table S2), 200 μM each dNTPs, 1 U DyNAzyme II DNA polymerase (in Finnzymes’ Optimized DyNAzyme reaction buffer). Otherwise PCR conditions were as above except that the extension time at 72°C varied from 45 to 120 s depending on the length of the desired PCR product. Radioactively labeled PCR products were purified using the PCR Clean-up Nucleospin Extract II kit (Macherey-Nagel, Germany) and eluted into 30 μl of 10 mM Tris–HCl, pH 8.5. Streptavidin beads (13 μl per each purified radioactive label-containing PCR reaction) were pre-washed four times in TEN (10 mM Tris–HCl pH 7.5, 1 mM EDTA, 100 mM NaCl), once in 2× binding buffer (10 mM Tris–HCl pH 7.5, 2 mM EDTA, 200 mM NaCl) and finally reconstituted in 30 μl per reaction of 2× binding buffer. Radioactively labeled PCR products (30 μl) were adsorbed to pre-washed streptavidin beads (30 μl) for at least 1 h at RT. The beads were then washed three times with 0.5 ml of TEN (10 mM Tris–HCl pH 7.5, 1 mM EDTA, 1 M NaCl) and twice with 1× restriction enzyme buffer 3 (New England Biolabs), and incubated in 50 μl 1× restriction enzyme buffer 3 containing NotI (25 U) for at least 4 h at 37°C. The beads were removed using a Magnetic Particle Separator (Roche), and the supernatant was purified by centrifugation through a Micro Bio-spin 30 column (Bio-Rad) equilibrated in TEN (10 mM Tris–HCl pH 7.5, 0.5 mM EDTA, 50 mM NaCl) at 3000 r.p.m. for 4 min at RT in a tabletop microcentrifuge. The supernatant was then ethanol-precipitated, and the pellet was re-suspended in 3 μl of TE (10 mM Tris–HCl, pH 7.5, 0.5 mM EDTA) plus 3 μl of 2× formamide loading dye (95% deionized formamide, 10 mM EDTA, 0.1% bromophenol blue, 0.1% xylene cyanol). Samples were analyzed by 7 M urea, 6% polyacrylamide gel electrophoresis as previously described (). The gel was dried at 80°C onto Whatman 3 MM paper, and the bands were visualized by autoradiography using a Fuji BAS 1500 phosphorimager with BAS-Reader 2.9 software (Raytest). Appropriate sequencing reactions and size markers were electrophoresed in parallel lanes for the identification of exact target sites. We devised a general strategy that can be used for fine mapping of a protein region involved in a specific protein–protein interaction. The strategy combines a robust DNA transposition-based insertion mutagenesis system, a visual yeast two-hybrid screen and a high-resolution genetic footprinting technique (A). To validate the methodology, we studied a recently characterized interaction between the human JFC1 and Rab8A proteins. First, we generated a pentapeptide insertion mutant library of JFC1 (B). We then used Rab8A as a bait in yeast two-hybrid analysis to distinguish between those JFC1 variants that were able to interact with Rab8A and those with altered or lost interaction due to a five-amino acid insertion in JFC1 (C). Next, we pooled different clone classes and located the positions of the insertions in each pool using a PCR-based footprinting strategy (D). Finally, the critical insertion sites were mapped to the predicted JFC1 protein structure. In standard yeast two-hybrid protocols, cells are initially selected on the basis of their plasmid content on glucose-containing plates, yielding well-growing yeast colonies. These colonies are subsequently replica-plated onto a two-hybrid screening medium where growth conditions induce the expression of interacting protein partners. In most systems, galactose serves as an inducer via activation of promoters, resulting in blue colonies on X-gal indicator plates upon interaction. We reasoned that it should be possible to combine relatively good growth conditions with conditions that also induce sufficient protein expression, thereby permitting color screening directly on the original transformation plates without prior replica-plating. Thus, we optimized the growth conditions (data not shown) on plates using JFC1/Rab8ΔQ67L and JFC1/Rab8ΔT22N pairs that in conventional two-hybrid assays are known () to yield blue (active GTP-bound form of Rab8A, positive control) and white colonies (inactive GDP-bound form of Rab8A, negative control), respectively. Under the optimized growth conditions in this study [including 100 mM sodium phosphate (pH 7.0), 0.6% glucose, 1.4% galactose and 200 μg/ml X-gal], both deep blue and clean white colonies could readily be obtained with the above two protein pairs, demonstrating that the system is very well suited for distinguishing between interacting and non-interacting protein partners. The insertional pentapeptide mutagenesis strategy based on Mu DNA transposition (,) can be used to generate mutant clone libraries with 100% efficiency, i.e. all the library clones are true insertion mutants and each clone contains only one insertion (B). Such a library, with 5.4 × 10 clones, was generated for JFC1 and analyzed under the optimized yeast two-hybrid conditions (see Materials and Methods section, and C). Approximately 92% of the library clones yielded blue colonies, indicating that most of the generated mutations did not interfere with the analyzed protein–protein interaction. However, ∼5% of the colonies appeared completely white, and 2–3% of the colonies were intermediate in color (pale blue), suggesting that the respective mutations had an effect on the JFC1/Rab8A interaction. To confirm a successful library construction, JFC1-encoding plasmids from 22 blue, 27 pale blue and 30 white colonies were subjected to initial restriction analysis and subsequent sequencing to localize the insertions (see Materials and Methods section). As expected on the basis of our previous study (), each of the analyzed clones contained an accurate 15-bp insertion (Table S3), indicating a high quality library. Most (83%) of the white colony-yielding insertions localized between the nucleotides 116 and 170 of the JFC1 gene, encoding the SHD1 domain close to the N-terminus of JFC1, and most of the pale blue colony-yielding insertions were located in a close proximity to this same region, suggesting that an interacting protein interface can be identified with the chosen strategy. Analysis of mutants as pools is arguably the most effective means to exploit the potential of an insertion mutant library. Thus, we devised a footprinting strategy, by which it should be possible to analyze a large number of mutants simultaneously (D). To test the feasibility of this strategy and to determine the optimal number of pool clones for the analysis, we first made DNA preparations representing five clone pools with cumulative 100, 600, 1600, 6600 and 16 600 member clones (see Materials and Methods section). The entire JFC1-encoding insert was then PCR-amplified from the pooled DNA samples using a vector-specific primer pair. The amplified inserts were then used as template to amplify shorter gene segments using primer pairs consisting of a radioactively labeled and a biotinylated primer (D). Following pull-down with streptavidin-coated beads and subsequent NotI digestion, reaction products were analyzed by denaturing polyacrylamide gel electrophoresis and autoradiography (Figure S1). In this experiment, we used two independent JFC1 gene-specific radioactively labeled primers that hybridized at a 22-nt interval to verify insertion-associated specificity of the data as well as reproducibility of the procedure. In general, the number of different-length radioactive products increased in accordance with an increase in the number of clones present in the DNA sample. The control wild-type JFC1 generated only the full-length PCR product but no radioactive restriction products, indicating that the detected reaction products from mutant pools were insertion specific (Figure S1). In addition, the two independent primers produced identical band patterns, further verifying the specificity and indicating that the entire process can be duplicated independent of the radioactively labeled primer. Altogether, these data showed that the strategy could reliably be used to map insertion sites. All the analyzed plasmid pools generated relatively clear band patterns. However, the presence of a moderate number of pooled clones (e.g. 100 and 600) appeared to be optimally suited for this analysis, as the distribution of radioactive label in these cases was restricted to fewer and thus more intensely labeled reaction products. Having established suitable conditions for comparative parallel analysis of mutant pools, we next compared insertions in plasmids from blue, pale blue and white colonies from the two-hybrid screen. We generated a pool of 174 white and a pool of 35 pale blue colonies. In addition, six pools of 100 blue colonies each were generated. Plasmid DNA was isolated from these pools, and the insertion sites along the entire length of the JFC1 encoding gene were then mapped using 14 JFC1 gene-specific radioactively labeled primers located at ∼120-bp intervals (Figure S2). This strategy produced partially overlapping band patterns in autoradiographs (), and most gene regions were thus analyzed twice, giving rise to high-quality data ( and ). The pool of white colonies produced 48 distinctive bands, and the vast majority of the respective insertions were localized to the same SHD1 domain that was revealed in the above-mentioned sequence analysis of individual clones. The pool of pale blue colonies produced 33 bands with most of the respective insertions within or surrounding the SHD1 domain. The blue colony pools revealed 389 apparent insertions that were scattered relatively evenly along the entire length of the protein; however, with insertions poorly represented within the SHD1 domain (). Several secondary structure prediction programs, including CHOFAS and PELE at Biology WorkBench , suggested a long α-helix within the SHD1 region in JFC1 (aa 32–68, ). JFC1 is structurally similar to Rabphilin-3A () (A), and its interacting partner Rab8A is structurally related to Rab3A () (B). In addition, Rabphilin-3A/Rab3A co-crystal structure is available (). Therefore, we used this structural data to model the SHD1 helix architecture and investigated the distribution and effects of the insertion mutations (C and 5D). The critical insertions were located along the entire length of the helix, and they all seemed to abolish the JFC1/Rab8A interaction by restructuring the helix and ultimately destroying important amino acid side-chain contacts between the two proteins. We described here an efficient strategy to define interacting protein regions for two protein-binding partners. It is based on simultaneous generation of a large number of insertionally mutated protein variants that are screened for an altered protein–protein interaction on a yeast two-hybrid platform and parallel mapping of the respective insertion sites using a PCR-based DNA footprinting strategy with pooled DNA samples. This methodology is general and pinpoints with high precision those protein interfaces that are involved in a specific interaction. As a proof of principle, we analyzed the interaction between the human JFC1 and Rab8A proteins. The Mu transposition-based insertion mutagenesis system generates essentially randomly distributed five-amino acid insertions in proteins. It is accurate and highly efficient, yielding 100% mutants, and can be used to generate exhaustive mutant libraries. Our JFC1 library was composed of 24 000 independent clones, which implies more than 14-fold insertion-per-nucleotide coverage within the JFC1 gene. The accuracy and wide distribution of insertions were evident from the DNA sequencing and footprinting data. We adopted the commonly used yeast two-hybrid system in our strategy but modified it in two critical ways. First, we adjusted the yeast growth medium to promote adequate levels of induced protein expression without overly compromising cell propagation, which enabled us to classify the colonies based on their color on the primary transformation plates. Second, we modified the JFC1-encoding carrier plasmid to include a Km-resistance cassette in order to allow its straightforward transfer into following the screening phase in . The pool size appeared not to be very critical for the described strategy, although those pools that contained a few hundred clones generated the most easily interpretable data. Functional analysis of the JFC1/Rab8A interaction was accomplished by subjecting the selected JFC1 mutant pools to comparative parallel analysis using a PCR-based strategy. Following gel analysis and autoradiography, this type of genetic footprinting assay generates a visual read-out where reciprocal band patterns can be seen between the insertion mutant pools representing unaltered versus altered protein–protein interaction. Previously, genetic footprinting approaches have been used to analyze genes, proteins and entire genomes to identify regions essential for a particular function (,,). To our knowledge, this is the first time that genetic footprinting has been combined with a yeast two-hybrid analysis. However, we note that several pentapeptide insertion mutants have previously been analyzed individually using a yeast two-hybrid platform (,). The footprinting analysis pinpointed a short region in JFC1, in which the insertions disrupted the interaction with Rab8A, indicating that this region must be involved in Rab8A binding. The identified region overlaps with the known SHD1 domain located at the N-terminus. Insertions elsewhere in the protein, including the SHD2 and C2 regions, largely retained the interaction. SHD1 and SHD2 domains constitute an SHD region, also known as the Rab-binding domain (), a common motif among the members of the Slp family. In this protein family, the two domains are typically separated by a Zn-binding motif (), but in JFC1 these domains are directly joined, and the Zn-binding motif is missing. A common feature among the members of the Slp1 family proteins is that they bind Rab27A via the SHD region (). A similar SHD region is present in a number of Rab3A- and/or Rab8A-binding proteins, such as Rabphilin-3A, Noc2, Rim1 and Rim2 (,) (A), suggesting similar binding mode. To model the JFC1/Rab8A interface, we utilized several secondary structure prediction programs and 3D structural information available from the co-crystal complex of Rabphilin-3A and Rab3A (). The co-crystal structure reveals a 34-residue α-helix in Rabphilin-3A that is directly involved in Rab3A binding, and the same helix is known to participate in Rab27A binding (). The SHD1 region of JFC1 involves an analogous predicted α-helix (aa 32–68, ). Insertions that abolished the interaction are located along the entire length of the helix, and we assume that an insertion at any location within the helix would modify the local structure and break critical amino acid side-chain interactions essential for Rab8A binding. The described system is universally suitable for high-resolution mapping of protein–protein interfaces. Any protein-encoding gene cloned in an appropriate vector can be subjected to Mu mutagenesis to yield comprehensive libraries, and the yeast two-hybrid screen can be accomplished with any pair of appropriate plasmids encoding the interacting protein partners. In the two-hybrid system used, the appearance of false positive or false negative colonies is relatively infrequent. This frequency of 1–2% is tolerable and does not interfere with the data interpretation. The benefit of the system is that it allows a comprehensive simultaneous analysis of the entire length of the protein-encoding gene without separately cloning individual mutants. Once the insertion library has been generated, it can be used in multiple screens to identify regions involved in interactions with different protein partners. Overall, the system provides a means for a high-resolution analysis of protein–protein interfaces without the requirement for prior knowledge of the protein structure. The methodology is straightforward and can be applied in any standard molecular biology laboratory, as no special equipment or technical expertise is needed. To provide a proof for the generic nature of the methodology, we have now analyzed an interface between two yeast proteins (Sec1 and Mso1) using identical protocol, and this analysis has revealed novel interacting regions (Weber, M, HT, MP, HS, and Jäntti, J, unpublished data). The next challenge will be linking of the described strategy to high-throughput technologies, including capillary electrophoresis and massive parallel sequencing platforms. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Microscopical detection of chromosomal DNA is an important tool in understanding the function of genomic sequences in the interphase nucleus. In chemically fixed cells or tissues we can visualize any chromosomal domain, including individual gene loci, using standard fluorescence hybridization methods (). However, to observe the dynamics of chromosomes in living cells, such fixation-based techniques are not suitable. Consequently, alternative strategies have been developed allowing monitoring of chromosomal domains in living systems. The first reports on imaging of DNA targets involved systems in which fluorescent proteins, such as GFP, were fused to a DNA-binding domain of a transcription factor, in this case a or repressor (). When introduced into cells, that contain a recombinant locus carrying a series of cognate binding sites, the GFP-tagged transcription factors bound to their targets and enabled localization of the recombinant locus. However, since introduction of an artificial target locus into the genome usually occurs at random positions, the technique is in fact inapplicable for imaging of specific endogenous genomic loci. As an alternative to this inherent problem, GFP fusions have been made with proteins that are known to be present in chromatin. For example, in various organisms GFP-tagged histones have been visualized in chromatin of living cells (). In addition, more specific chromosomal loci were monitored with GFP-fused proteins that are known to localize at specific chromatin domains, such as the centromere (). The techniques mentioned above have clearly demonstrated the intriguing possibilities for visualization of DNA sequences and/or chromatin structures by means of fluorescent proteins. In order to expand the application of chromosome labeling, a larger variety of tools is required. In that respect, recent developments regarding the construction of artificial DNA-binding proteins based on CysHis zinc finger (ZF) domains are of great interest. This rapidly developing technology is based upon the established 3-bp DNA recognition code of CysHis ZFs () and the possibility to construct polydactyl zinc finger (PZF) proteins by fusion of individual ZF moieties in such a manner that the number of fingers fused determines the length of the cognate DNA recognition site. For example, three-fingered (3ZF) PZF domains will recognize 9- bp target sequences (3 × 3 bp), while six-fingered (6ZF) PZF domains in principle interact specifically with 18-bp sites (6 × 3 bp). Although the complete recognition code for all possible 64 3-bp sequences has not yet been elucidated and not all ZF-DNA contacts are of high affinity and specificity, PZF technology already allows construction of artificial transcription factors as well as novel site-specific nucleases (; ). We hypothesized that construction and expression of sequence-specific DNA-binding PZF:GFP proteins should enable live cell imaging of repetitive DNA sequences within a given genome. Here, we present the use of PZF:GFP reporter proteins in the model plant as well as in mouse cells. Centromeric repeats were readily observed in both model species. Furthermore, PZF:GFP content was quantified in living cells via indirect comparison to the fluorescent signal of single eGFP molecules. The sensitivity and further applications of the technique are discussed. PZFs for binding to the selected sequences were constructed in pSKN-SgrAI, as described previously () after which the PZF encoding sequences were cloned as SfiI fragments into the plant expression vector pRF-GFP or pcDNA-GFP (see subsequently). Vector pRF-GFP was obtained by modifications of pGPTV-KAN (), including removal of NotI and SfiI sites from the vector backbone, replacing the promoterless GUS coding sequence by a 1.7-kb XmaI-SacI fragment containing the promoter (), and insertion of sequences providing the vector with an ATG translational start codon, a FLAG tag, a SV40 nuclear localization signal (NLS) and an in frame coding sequence of mGFP6+(HIS)6, an enhanced GFP version (F64L, S65T) with a C-terminal HIS tag (). For expression in mouse cells, an identical GFP-containing reading frame was introduced into the widely used expression vector pcDNA3 (Invitrogen), forming pcDNA-GFP. This involved the removal of the polylinker via HindIII/ApaI digestion, which was replaced by a PCR fragment containing an ATG translational start codon, a FLAG tag, a SV40 NLS, SfiI sites for directional cloning of PZF domains and GFP. Outlines of the constructs are given in . Details of construction and the final vector sequence are available upon request. The construct encoding mRFP:HP1α (a gift of Martijn S. Luijsterburg) was based on the backbone encoding for pEGFP-C1 in which EGFP was replaced by mRFP (). The HP1α sequence () was cloned in the EcoRI/BamHI sites of mRFP-C1, resulting in an mRFP:HP1α fusion protein. The plasmids encoding constructed PZF:GFP fusion proteins were mobilized into strain AGL1 (). PZF:GFP proteins for the centromeric repeat and the 5S rDNA repeat were expressed in ecotype Columbia. Line A of ecotype Zurich (), containing a silent hygromycin resistance locus, was used for experiments involving the targeted PZF. Transformation of was carried out by floral dip () and primary transformants were selected on solidified MA solid medium () without sucrose and containing kanamycin (50 mg/l) for selection and timentin and nystatin (both 100 mg/l) to inhibit bacterial and fungal growth, respectively. Primary transformants were further grown under greenhouse conditions and allowed to self pollinate. NIH 3T3 mouse cells were grown in Dulbecco's modified Eagle medium (DMEM; Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Invitrogen) and 100 units/ml of a penicillin/streptomycin antibiotic mixture (Invitrogen). Culture conditions were 37°C, 5% CO.Cells were co-transfected with plasmid DNA using Lipofectamine 2000 (Invitrogen) according to manufacturer's instructions and were analyzed by confocal microscopy 24–48 h post transfection. For analysis of PZF:GFP expression in , T2 populations expressing : chimeric genes were sown on solidified 1/2 MS medium containing kanamycin (30 mg/l) and sucrose (10 g/l) and grown vertically for 5–8 days prior to analysis via confocal microscopy using either a BIO-RAD confocal microscope equipped with a combined Ar/Kr laser (excitation at 488 nm and emission filter 522DF32) or a Leica confocal setup (see subsequently). Mouse cells were examined 24–48 h post transfection. For DNA staining, DRAQ5 (Biostatus Ltd) was added to the culture medium to a final concentration of 1 μM. All nuclei were recorded with a Zeiss LSM 510 (Zeiss, Oberkochen, Germany) confocal laser scanning microscope. For eGFP, an argon 488 nm laser and a 505–550 nm band pass filter was used. For mRFP, a 543 nm helium-neon laser and a 585–615 nm band pass filter was used. DRAQ5 was recorded using a 633 nm HeNe laser and an LP 650 emission filter. Fluorescent intensities in confocal images of plant root nuclei were quantified in relation to intensities of yellow–green fluorescent styrol spheres with a diameter of 20 nm (FluoSpheres, F8787, Molecular Probes, Leiden, The Netherlands). Nuclei and FluoSpheres were imaged with a confocal microscope (Leica TCS SP, Leica Microsystems GmbH, Heidelberg, Germany) directly one after another to ensure identical imaging conditions. Samples were excited at 488 nm and fluorescent emission was collected with a Leica PlanApo 100 × 1.4 NA objective between 505 and 650 nm using an RSP500 filter. In order to relate these intensities to discrete numbers of GFP molecules, intensities of individual FluoSpheres were compared to single recombinant eGFP (BioVision Inc., Mountain View, CA, USA). The setup for single-molecule imaging has been described in detail (). Briefly, samples were mounted onto an inverted microscope (Axiovert 100TV, Zeiss, Oberkochen, Germany) equipped with a 100× oil-immersion objective (NA 1.4, Zeiss, Oberkochen, Germany) and illuminated for 5 ms by an ArKr-laser (Spectra Physics, Mountain View, CA, USA) at a wavelength of 488 nm. Illumination intensity was set to 2 kW/cm2. This permitted the detection of individual fluorophores by a nitrogen-cooled CCD-camera system (Spec-10:400B/LN, Princeton Instruments, Trenton, NY, USA). Stock solutions of FluoSpheres and eGFP (2% solids and 1 µg/µl, respectively) were diluted 10 and 10 times in PBS, respectively, which contained 1% (w/v) poly vinyl alcohol (PVA, Roth Chemicals, Germany). Samples were then embedded in PVA films on cleaned glass slides by spin coating 50 μl of the dilutions on a spin coater (Model P6700, SCS, Indianapolis, IN, USA) with a two-step protocol: 10 s at 300 r.p.m. and 1 min at 2000 r.p.m. The thickness of these films was on average 1 µm, as revealed by confocal microscopy (). After image acquisition, a 2D Gaussian was fitted to the diffraction limited spots (B) to obtain their integrated intensities. Histograms of the latter were then fitted with a sum of Gaussians of the form: Since the centromeric regions were larger than the confocal volume, their integrated intensities had to be determined from 3D scans. For each confocal slice of a centromeric region, the background signal outside the nucleus was subtracted from the integrated intensity of the region and summed up subsequently. Although by now most of the amino acid sequences needed for the specific recognition of a particular 3-bp triplet by CysHis ZFs are known, most practical applications still heavily lean on the use of the relatively robust ZF-GNN recognition code (,). Also in our laboratory, assembly of PZF domains based on the ZF-GNN code proved to be successful (). To assess the potential of PZFs for live cell imaging, three repetitive loci from the genome varying in length and copy number were chosen to be targeted by PZF modules. Within these loci, we selected the longest non-interrupted (GNN) sequences as PZF binding sites, taking into account that a possible target site overlap should not occur (). The centromeric 180-bp repeat, also known as pAL1, Atcon or 178-bp repeat, was selected because of its high copy number presence exclusively in the 0.6–1.2 Mb centromeric regions of chromosomes (,). Within the 180-bp repeat sequence, a single 9-bp target site 5′-GTTGCGGTT-3′ was chosen for 3ZF PZF design. Considering (i) that the centromeric region also contains multiple copies of the 106B repeat () and (ii) that the exact 9-bp target site for the PZF is not necessarily present in all 180-bp repeats, we estimate that the centromeric region accommodates about 2000 copies of the 9-bp target sequence of the PZF fusion protein, 180:GFP. For the second repeat, we selected 5S ribosomal DNA (5S rDNA). In the accession Columbia, the 5S rDNA repeat predominantly consists of repeating units of 497 bp that are located in the peri-centromeric regions of chromosomes 3, 4 and 5, with an estimated total copy number of about 1000 per genome (,). Within the 497-bp sequence, we selected the sequence 5′-GCCGTGGGT-3′, located upstream of the transcribed region, as a target site for the PZF fusion protein 5S:GFP. We estimate that each of the three major 5S rDNA loci contain 300–350 copies of the 5S repeat. However, due to sequence variation between blocks of repeats at the different loci (), we postulate a maximal number of 200 binding sites per 5S locus for the 5S:GFP fusion protein. The third genomic PZF target that was chosen in consisted of a transgenic locus which contains 5–10 intact copies of a silenced gene, as well as multiple incomplete and rearranged copies (). The repeat locus is visible as a heterochromatic knob on chromosome 1 (Fransz,P., unpublished data) and appears in interphase nuclei as a mini chromocenter (). Within the sequence, it was possible to select a (GNN) target site: 5′-GTCGGAGACGCTGTCGAA-3′. Six-fingered 6ZF moieties generally have a much higher affinity for DNA targets than 3ZF domains (). Anticipating that visualization of a relatively small number of repeating units might prove to be rather demanding, we chose to construct the corresponding 6ZF HPT:GFP in this case. In order to test whether PZF:GFP-mediated chromosomal labeling was also possible in other higher eukaryotes, we constructed a 3ZF PZF domain aimed to bind to the major satellite (MaSat) repeats within the pericentromeric region of mouse chromosomes (). The 234-bp MaSat units, present in 1000–10 000 copies per chromosome () are rather AT-rich, but the majority of them should contain the (GNN) target site 5′-GGCGAGGAA-3′. Transgenes encoding the PZF:GFP fusion proteins 180:GFP, 5S:GFP and HPT:GFP were expressed under the control of the promoter, derived from the ribosomal gene (). This promoter is predominantly active in meristematic tissues, in particular in root tips. A major advantage of the root for live cell imaging is the absence of autofluorescent chloroplasts. An N-terminal SV40 nuclear localization signal ensured targeting of the PZF:GFP fusion proteins to the nucleus. For all constructs, several independent primary transformants were maintained to obtain T2 seeds. Seedlings of the T2 populations were used for fluorescence microscopy analysis. Lines containing a nuclear-targeted GFP construct without a PZF domain were used as a control. All transfortmants, including the GFP control, showed a clear fluorescence in the root tip, especially in the meristematic region (). Close examination revealed a specific nuclear localization of the signals (C–E). Only in the 180:GFP transformants (C and D) we observed bright spots of GFP fluorescence against a faint diffuse fluorescent background. Confocal microscopical examination revealed up to ten 180:GFP-fluorescent spots at the periphery of the nucleus (A). The number and the spatial position of the spots matched very well with the number and position of chromocenters in . These heterochromatic domains accommodate all major repeats including the centromeric 180-bp repeat and other pericentromeric repeats such as transposons (). In order to establish co-localization of 180:GFP with chromocenters, we stained the root tip with propidium iodide. Confocal sectioning clearly demonstrated the co-localization of the 180:GFP spots with propidium iodide positive chromocenters (B and C). This co-localization indicates that the three finger PZF of the 180:GFP construct recognizes the centromeric 180-bp repeats. In support of this, the 180:GFP spots are in a symmetrical position in several juxtaposed cells within a longitudinal cell file (C and D). This is in agreement with the symmetrical position of centromeres in daughter cells immediately after mitosis. Since the 9-bp target sequence is only present as a repetitive array in the centromere region, we conclude that 180:GFP indeed binds to its target. A note of interest is disappearance of the fluorescent spots from the chromocenters, requiring a sequential procedure. This suggests that the binding of the PZFs to the centromere is not stable under these conditions. A brief fixation with formaldehyde could not prevent the dissociation of the 180:GFP molecules from the chromatin. We observed the same feature with other DNA dyes such as DAPI, Hoechst and Sytox® Orange (Invitrogen). Furthermore, the absence of a specific pattern in metaphase cells suggests that the 180:GFP construct does not bind to the centromere during mitosis. The experiments with the 5S:GFP and the HPT:GFP fusion proteins did not generate a specific pattern of the nuclear GFP signal but only a diffuse fluorescent stain similar to the one in control transformants expressing a GFP protein lacking a PZF domain (E). This suggests that either binding of the PZF construct to these targets does not occur or that the number of the target sites is too small to be discriminated from the diffuse background. proved to be an attractive system to investigate the potential of PZF:GFP in live cell imaging of repetitive sequences. Following the experiments in the plant model system, we addressed the question if the same approach is applicable to animals. We therefore used a three finger PZF:GFP to detect the major satellite (MaSat) repeat sequence in mouse. This pericentromeric sequence is present in 1000–10 000 copies of 234 nt units per chromosome () and localizes to mouse chromocenters. These heterochromatin regions are condensed DNA domains that are positive for the heterochromatin protein HP1α (,). The -driven MaSat:GFP fusion protein () was expressed in NIH 3T3 cells. Like in the system, live cell imaging was readily established. Twenty to forty intense fluorescent spots per nucleus were observed for MaSat:GFP. To demonstrate the co-localization of MaSat:GFP with chromocenters, we co-transfected the cells with both MaSat:GFP and a plasmid expressing mRFP:HP1α. In addition, we stained the cells with the vital DNA stain DRAQ5. The GFP spots co-localized with both the mRFP:HP1α and the DRAQ5 positive domains (), demonstrating that MaSat:GFP proteins bind to expected chromosomal targets. We conclude that the PZF-GFP method is applicable to animal cells. Moreover, this is the first report of visualization of the mouse major satellite repeat and hence opens new roads to live cell imaging of chromosomes. In order to further characterize the binding of PZF proteins, the number of 180:GFP molecules per centromere was quantified via the GFP signal. Since the CLSM setup used for observation of living cells did not provide single molecule sensitivity, we first used a single molecule microscopy (SMM) setup () to gauge the fluorescence intensities of 20 nm FluoSpheres against single eGFP molecules. Intensity histograms revealed that a FluoSphere and a single eGFP had an average intensity of 1971 ± 49 and 422 ± 4 counts, respectively (C and D), thus setting the fluorescent signal of a FluoSphere equivalent to that of 4.67 ± 0.12 eGFP molecules. From these calibrations, the number of 180:GFP molecules in centromeric spots as observed with CLSM (such as in ) was calculated by comparing the fluorescent signals from centromeres and FluoSpheres present within the same sample. When measured with the Leica CLSM setup, the signal due to one FluoSphere resulted in 311 ± 2 counts (A) and the average integrated intensity of centromeric spots was found to be 52 357 ± 11 932 counts ( = 46), thus equivalent to 168 ± 38 FluoSpheres and corresponding to a signal of 786 ± 179 eGFP molecules. In the same nuclei, the average integrated intensity of the non- structured background signal within the same volumes as occupied by the centromeric spots was found to be 17 522 ± 6915 counts, corresponding to 263 ± 104 GFP molecules. Assuming the same background also within the centromeres, about 500 (524 ± 207) extra 180:GFP molecules thus seemed to be bound to targets inside each of the centromeres. We thus conclude that a significant fraction of the estimated 2000 binding sites is accessible for 180:GFP binding. The use of FluoSpheres proved to be valuable to quantify GFP concentrations even on microscope setups lacking single molecule sensitivity. In those meristematic cells where 180:GFP-labeled centromeric regions were clearly visible, the nuclear GFP background fluorescence corresponded to a mean GFP concentration of 0.56 ± 0.27 µM. Similar nuclear background values were found for seedlings expressing the 5S:GFP and HPT:GFP constructs. As could be inferred from CLSM pictures and 3D stacks, the average 180:GFP-labeled centromeric region resembled an elliptical cylinder with minor and major diameter and height of 304 ± 47, 792 ± 130, 1038 ± 172 nm, respectively, with an apparent volume of 0.78 ± 0.22 fl. With 786 ± 179 eGFP molecules present within this volume, the apparent eGFP concentration per centromeric region would be 1.66 ± 0.60 µM. The successful visualization of fluorescent signals from the centromeric 180-bp repeats was thus based upon an apparent 3-fold increase in concentration of 180:GFP molecules. Given this fact at a mean 0.56 µM concentration of nuclear PZF:GFP, lack of detectable specific signals due to PZF:GFP binding to less abundantly repeated structures as the 5S rDNA and sequences thus seemed to be a logical consequence of our experimental conditions. When just considering the absolute physical concentration of DNA target sites within a condensed chromatin structure, thus with six nucleosomes and 1200 bp of DNA per 10 nm of the higher order solenoid 30 nm fiber (), repetitive binding sites occurring once per nucleosome have a concentration of ∼1.5 mM. Although such a high concentration suggests that very high signal to noise ratios should be obtained by means of PZF:GFP-mediated repeat labeling, the point spread function of CLSM setups unavoidably generates blurred pictures with volumes that are easily several hundred-fold larger than that of the real physical objects with a diameter of 30 nm. The observed 3-fold difference in fluorescent intensity of the 180:GFP-labeled centromeres compared to the background signal agrees very well with this optical phenomenon. With dissociation constants for 3ZF PZF domains and their cognate 9-bp binding sites ranging between 10 and 75 nM (), a nuclear PZF:GFP concentration around 0.5 µM seems to be a reasonable compromise to ensure that most of the accessible (GNN) sequences can be bound by PZF:GFP molecules while still allowing visualization of highly repetitive DNA sequences. Higher GFP background levels, such as found for transient CaMV35S-driven GFP expression in protoplasts (), should be avoided as they will obscure even the centromeric 180:GFP signals. It is interesting to note that the REPRESSOR:GFP-mediated detection of transgenic loci containing 112–256 copies of the cognate or binding sequences () apparently resulted in a similar signal to noise ratio as observed in our study of 180:GFP-mediated centromere labeling. The repeating units in these cases were 40–50 bp in length, about one quarter of the length of the 180-bp repeat, but in total 4- to 8-fold less abundant per locus. Considering the available data and assuming nuclear PZF:GFP concentrations around 0.5 µM, we propose that it should be possible to detect PZF:GFP-mediated chromosomal labeling when the total number of binding sites at a specific locus divided by the length of the repeating sequence in base pairs is larger than 1. For the 180-bp repeat (2000 sites per locus divided by 180), this tentative formula gives a value of 11.1; for the repressors recognizing their operators, values are between 2.2 and 6.4. For the 5S repeat, even assuming the maximal 200 sites per locus, the value is 0.4, below the imaginary threshold. Obviously, there should be a limit for the length of the repeating unit and this might be chosen at 200–300 bp, about once per one or two nucleosomes. A way to overcome problems with longer and more rare repeating units would be the simultaneous expression of several PZF:GFP fusion proteins, each recognizing a different sequence within the target site of interest, in this way providing the necessary number of binding sites for successful visualization. The more recently elucidated ZF recognition codes for ANN, CNN and some TNN triplets as well (; ), should offer ample opportunities for alternative PZF design, but since the usefulness of these codes is not yet supported by wider experimental evidence we refrained from using them in the present study. A novel method is presented to establish live cell imaging of endogenous repetitive DNA sequences. The method employs sequence-specific polydactyl zinc finger (PZF) domains that can be designed to target any DNA sequence. In combination with a fluorescent tag, the PZF behavior can be monitored . In contrast to the or operator/repressor systems () the presented PZF system does not require the prior introduction of a transgenic target into the genome. PZF-mediated labeling can be applied to detect sequences in plants and animals. Using a 9-bp target, both the 180-bp repeat in and the major satellite repeat in mouse were successfully detected . Already in its present state the method is a valuable and flexible tool to study repetitive sequences and complement existing live cell imaging techniques in a wide variety of organisms. Although target regions containing moderate and low copy repeats were not visualized, refinement of the sensitivity using multiple PZF:GFP proteins and optimization of the protein concentration should enable the detection of theoretically any repetitive DNA sequence.

The fragile X mental retardation syndrome (FXMR/FXS), an X-linked disorder, is the most common cause of inherited mental retardation (). At the molecular level, the progressive expansion of (CGG) repeats and the hypermethylation of the CpG island, in the 5′-untranslated region (5′-UTR) of the gene causes its transcriptional inactivation. The resulting suppression of the encoded protein, named the fragile X mental retardation protein (FMRP), has been shown to be the underlying cause of this syndrome (,). FMRP is a putative nucleocytoplasmic shuttling protein (), found abundantly expressed in the neurons and several studies suggested that this protein participates in the synaptic plasticity of neurons by acting on post-transcriptional control of gene expression (). FMRP has been proposed to regulate the transport and translation of specific messenger RNA targets (mRNA) in a manner critical for neuronal development. It has also been shown that this protein has nucleic acid chaperone properties (). The sequence analysis of FMRP revealed that the 632 amino acid protein contains two types of RNA-binding motifs: two K-homology (KH) domains and one arginine-glycine-glycine rich region (RGG box), suggesting that the protein exerts its function through RNA binding (). FMRP has two autosomal paralogs, the FXR1 and FXR2 proteins (FXR1P and FXR2P), with which it forms the fragile X-related protein family (,). Sequence analysis revealed that the two proteins have ∼60% amino acid identity, with regions of 90% sequence identity to FMRP (,). FXR1P and FXR2P are also cytoplasmic RNA-binding proteins, each containing two KH domains. The FXR2P is divergent from FMRP and FXR1P in the C-terminal region, in that it has a RG cluster instead of an RGG box. The fact that the FXR proteins have been found to be associated predominantly with the ribosomal 60S subunit, and that they have similar RNA-binding domains lead to the suggestion that the FXR1P and FXR2P might compensate for the FMRP function (). However, the comparison of the expression levels of each of these proteins in different tissues and cellular distributions suggests that each of the FXR proteins might have an independent function (). Biochemical studies conducted showed that FMRP uses its RGG box to bind with high affinity to target RNA sequences proposed to contain G quartet structures (). A G quartet is formed from four guanine residues arranged in a planar configuration, which is stabilized by Hoogsteen-type hydrogen bonds. Several such planar structures can stack and are stabilized by potassium or sodium cations, but they do not form in the presence of lithium cations (). The specific mechanism by which FMRP interacts with its mRNA ligands and regulates their translation still remains poorly understood. Human semaphorin 3F (S3F) mRNA has been identified both and as a potential mRNA target of the FMRP (,) and it has been proposed that its interactions with the FMRP RGG box occur in a G quartet-dependent manner (). S3F mRNA encodes for the SEMA 3F protein which belongs to the class 3 semaphorins, a family of secreted and transmembrane signaling molecules that play crucial roles in the nervous (neuronal migration and axon pathfinding), immune and cardiovascular systems. Every member of this family has the 500 amino acid signature, the semaphorin domain (,). SEMA 3F is a putative secreted protein that has been suggested to have chemoattractant and repulsion functions. The expression of the gene has also been reported to suppress tumor formation in nude mice and to cause the alteration of the cellular response to drugs inducing apoptosis (). The goal of this study is to contribute to our understanding of the principles of recognition between FMRP and its RNA target(s), by analyzing its interactions with S3F RNA. We show here that S3F RNA adopts a parallel intramolecular G quadruplex structure and we use thermodynamic methods to determine if the stability and the secondary structure of this RNA are altered by its interactions with the FMRP RGG box. We also investigate the binding of the RGG box and RG cluster of the FXR1P and FXR2P to S3F RNA, in an effort to determine if the recognition of the structural elements in this RNA is unique to the FMRP RGG box. The unlabeled RNA oligonucleotides (S3F-lg, S3F-M2 and Munc-13 site 1) were synthesized by transcription reactions using T7 RNA polymerase (produced in-house), following the procedure by Milligan and Uhlenbeck (). The synthetic DNA templates were purchased from Trilink Biotechnologies, Inc. The RNA oligonucleotides were purified by denaturing gels and electrophoretic elution, followed by extensive dialysis against 10 mM Tris (pH 7.5) or 10 mM cacodylic acid (pH 6.5). S3F-M2 was constructed by introducing two point mutations in S3F-lg to stabilize the stem (A). The FMRP, FXR1P RGG boxes and the FXR2P RG cluster were chemically synthesized and purified by the Peptide Synthesis Unit at the University of Pittsburgh, Center for Biotechnology & Bioengineering. The UV melting curves of the unlabeled S3F-lg, S3F-M2 and of the 2-AP labeled S3F-M2_15AP RNAs were collected using a Varian Cary 3E spectrophotometer equipped with a Peltier cell. The samples were annealed in the standard buffers 10 mM Tris, pH 7.5 or 10 mM cacodylic acid, pH 6.5, containing either 150 mM KCl or 150 mM LiCl. The RNA samples were heated from 20 to 99°C at a rate of 0.2°C/min, recording points every 1°C. Blank samples were treated in the same manner. Depending upon the RNA concentration, the spectral absorbance was measured either at 295 or 305 nm, wavelengths that have been previously identified to be sensitive to G quadruplex dissociation (). To determine if S3F-M2 folds into an intermolecular or intramolecular conformation, the melting temperature of the G quadruplex structure was determined at different RNA concentrations in the range 10–80 µM. The transition of the G quadruplex dissociation in S3F-M2 and S3F-M2_15AP was identified and fitted assuming an independent two state model: The binding of the FMRP RGG box to S3F-M2_15AP was measured by titrating increasing concentrations of the peptide to a fixed concentration of 150 nM S3F-M2_15AP. The same procedure was repeated for the FXR1P RGG box. The binding dissociation constant, , was determined by fitting the binding curves to the equation: The thermodynamic parameters for the FMRP RGG box binding to S3F-M2_15AP were determined by measuring the = 1/ at different temperatures in the range 20–45°C. The standard enthalpy and entropy of binding were determined from the slope and intercept of the graph: The 1D H spectra of S3F-M2 RNA were acquired at 29°C on a 500 MHz Varian Unity Plus spectrometer. The water suppression was accomplished using the jump-and-return pulse sequence () with the maximum of excitation set at 11 p.p.m. S3F-M2 RNA (387 μM) was prepared in 10 mM Tris (pH 7.5) at a 90% HO/10%DO ratio. The melting of the S3F-M2 RNA stem structure was monitored by recording the 1D H NMR spectrum at different temperatures in the range 20–60°C. These experiments were performed on a Bruker AVANCE™ 500 MHz NMR spectrometer. EMSA reactions were performed in a total volume of 15 µl. The RNA:peptide complexes were prepared by mixing the RGG peptides with S3F-M2 or S3F-M2_15AP in 1:1 or 1:2 ratios and resolved on 15% non-denaturing acrylamide gels that were run in the presence of 75 mM KCl, at 35 V. The electrophoretic mobilities of the free RNA and the S3F RNA:peptide complexes were visualized by UV-shadowing at 254 nm, using an AlphaImager HP (AlphaInnotech, Inc.). It has been proposed that the G quartet motif is important in the FMRP recognition of its RNA targets, and S3F mRNA has been identified as a potential target of the protein, based on the fact that its G-rich sequence could fold into this structural motif (). The interactions of FMRP with the semaphorin mRNA fragment used in this study (A) have been visualized in living mammalian cells; moreover, it has been shown that the mutation of the GG doublets proposed to be involved in the G quartet formation abolishes these interactions (), supporting the idea that the S3F RNA recognition occurs in a G quartet-dependent manner. We have expressed and purified a 38-nt RNA, containing the 34-nt G-rich fragment of human S3F mRNA proposed to interact with FMRP, to which four extra nucleotides (GGGA) were added at the 5′-end for transcription purposes () (named S3F-lg; A). To determine if this RNA folds into a G quadruplex structure, we used a combination of CD, fluorescence, NMR and UV spectroscopy techniques. It is well known that the G-rich nucleic acid sequences fold in to G quartets in the presence of cations like K, by forming cation–dipole interactions with the guanine residues (). CD spectroscopy has been extensively used to analyze the G quadruplex structure in DNA and RNA. Typically, there are two types of CD spectra observed for G quadruplexes: type I, which exhibits a positive band ∼265 nm and a negative band ∼240 nm and type II, which exhibits a positive band ∼295 nm and a negative band ∼260 nm (). At least for ‘’ G quadruplexes, there is a strong correlation between the parallel quadruplex and type I CD spectrum (,) and between the antiparallel quadruplex and type II CD spectrum (,). In the case of intramolecular G quadruplexes, there are also examples of parallel type quadruplexes exhibiting type I of CD spectrum () and of antiparallel quadruplexes exhibiting type II of CD spectrum (,,). However, this correlation is not as clearly established in the case of G quadruplexes, since there are not enough high-resolution structures available. Generally, the CD spectroscopy results cannot be exclusively used to assign definitively a particular type of fold to a G quadruplex structure, as there are examples of more complex CD spectra exhibiting the features of both, type I and type II spectra (). The CD spectrum of S3F-lg RNA folded in the presence of K ions is of type I, with a positive band at 263 nm and a negative one at 238 nm (B), confirming the presence of G quartet structural elements in this RNA. This result suggests that the fold of this G quadruplex is of parallel nature, however, as discussed above, this can only be confirmed by the high-resolution structure of this RNA. The UV spectroscopy thermal denaturation profile of S3F-lg measured at 305 nm, shows a characteristic hypochromic transition between 52 and 72°C, corresponding to G quadruplex dissociation () (C), supporting the presence of a G quadruplex structure in this RNA. G-rich sequences are notorious for forming alternate G quartet structures , and to determine if this is true for S3F-lg RNA as well, we used native gel electrophoresis. Two conformations are observed on a 15% native gel performed in the presence of 75 mM KCl at all RNA concentrations investigated (D and data not shown). To obtain higher-resolution information about the structure of S3F-lg RNA, we have used one-dimensional (1D) H NMR spectroscopy. Resonances are present in the 10–12 ppm. proton region corresponding to imino protons involved in G quartets, however, they are very broad, suggesting that this RNA exchanges between different conformations. Surprisingly, we did not observe any imino proton resonances corresponding to Watson–Crick base pairs in the 12–14 ppm. region, indicating that the stem structure proposed in A does not exist in S3F-lg (data not shown). One possible explanation for this finding is that the addition of the extra four nucleotides (GGGA) at the beginning of the S3F-lg sequence might actually promote the folding of this RNA into an alternate G quadruplex structure, since they contribute to the formation of an uninterrupted stretch of 10 purines. In an effort to promote the folding of S3F RNA into a single conformer we have removed its first four GGGA nucleotides [that were added only for transcription purposes ()], and introduced specific point mutations at positions 3 (G to C) and 4 (G to U) and at the complementary positions 31 (U to A) and 32 (U to C), respectively (labeled in blue in A). This mutated RNA, named S3F-M2 RNA, no longer contains a stretch of 10 purines at its 5′ end and of six consecutive guanines in the region proposed to fold into a stem structure (A). We first investigated whether S3F-M2 RNA maintained the ability to form a G quadruplex structure. Upon titration of increasing concentrations of KCl, the CD spectrum of S3F-M2 RNA showed the spectral features of a type I G quadruplex CD spectrum, with a strong positive band at 263 nm and a negative band at 238 nm (C), very similar to that of S3F-lg RNA. The native gel electrophoresis of S3F-M2 RNA indicates that at concentrations <10 µM this RNA exists in a single conformation (black arrow in D, and data not shown), whereas at higher concentrations S3F-M2 adopts more conformations. The 1D H NMR spectrum of S3F-M2 shows resonances corresponding to imino protons involved in G quartets, as well as resonances corresponding to Watson–Crick base pairs, indicating the presence of both a stem and a G quadruplex in the structure of this RNA (E). However, the G quartet imino proton resonances are very broad, consistent with an exchange between the different S3F-M2 conformers formed at the high RNA concentration required when using this technique. The exchange between different RNA conformations is also supported by the Watson–Crick imino protons, whose resonances become much broader upon addition of increasing KCl concentrations. These findings hindered our efforts to pursue high-resolution NMR spectroscopy studies of the S3F-M2 RNA structure. We have employed UV spectroscopy to obtain the thermodynamic parameters of G quartet formation in S3F-M2 RNA. The UV thermal melting profile of 10 µM S3F-M2 RNA folded in the presence of 150 mM KCl shows a hypochromic transition between 38 and 67°C (indicated in red in A), and a hyperchromic transition starting around 75°C. We assign the 38–67°C hypochromic transition, with a melting point ∼52°C, to the S3F-M2 RNA G quadruplex dissociation (). As expected, this transition is absent when the RNA is folded in the presence of 150 mM LiCl (B), since G quartets do not form in the presence of Li ions. We postulated that the hyperchromic transition starting at 75°C corresponds to the S3F-M2 stem structure melting (). To test this hypothesis, we have constructed a S3F RNA from which the stem region has been removed (named S3F-sh). S3F-sh maintains the ability to form a G quadruplex structure, as evidenced by its type I (positive band ∼265 nm and negative band ∼240 nm) CD spectrum and by the presence of G quartet imino proton resonances in its 1D H NMR spectrum (data not shown). In addition, its UV thermal denaturation profile measured at 305 nm shows a 40–65°C hypochromic transition, corresponding to a G quadruplex melting point of ∼52°C (Supplementary Figure 1). A hyperchromic transition starting around 65°C is still present in the UV melting profile of S3F-sh, which lacks a stem structure (Supplementary Figure 1). Thus, it is clear that the origin of the hyperchromic transition observed in the UV melting profile of S3F-M2 RNA, is not the melting of its stem structure. We ruled out the possibility that this hyperchromic transition is due to the RNA degradation at high temperatures, by checking the reversibility of the melting curves of S3F-M2 RNA measured in the range 25–70°C and 25–99°C, respectively (Supplementary and B). One possibility for the presence of the hyperchromic transition in the UV melting curves of S3F-M2, S3F-sh and S3F-lg RNAs could be that all have an uninterrupted stretch of 14 purines (starting at G12 for S3F-M2 RNA- A). Upon the melting of the G quadruplex structure, the liberated rG residues can stack with their rA nearest neighbors, and these rG-rA stacks will melt with increasing temperature, giving rise to the hyperchromic transition observed above 65°C (,). Since we could not determine the transition corresponding to the melting of the S3F-M2 stem structure from its UV melting curve recorded at 305 nm, we used 1D H NMR spectroscopy, since with this technique we can monitor individually the stem structure of S3F-M2 RNA (resonances at 13.3 ppm. and 12.0 ppm. in E). We have determined that both resonances corresponding to Watson–Crick imino protons are no longer present at 50°C, indicating that the stem structure in S3F-M2 RNA is completely melted above this temperature (data not shown). It is interesting to note that the G quadruplex forming sequence of S3F-sh RNA (B, nucleotides 8–27) is wild-type and it is also identical in S3F-lg and S3F-M2 RNAs. Yet, the comparison of the melting points of the G quadruplex structures formed by these three RNAs shows that S3F-sh and S3F-M2 RNA likely form a similar G quadruplex structure (∼52°C), which is different from that formed by S3F-lg ( ∼64°C). This supports the idea that the addition of four extra nucleotides in S3F-lg RNA, promotes the formation of an alternate G quadruplex structure, different from that formed by the wild-type S3F-sh. Moreover, this finding indicates that the mutations introduced in the stem of S3F-M2 RNA do not affect the ability of the G quadruplex forming sequence to fold into a structure similar to that of the wild-type S3F-sh. To determine if S3F-M2 RNA forms an ‘’ or an ‘’ G quartet structure we have measured its melting temperature at various RNA concentrations in the range 10–80 µM. and are the Van't Hoff thermodynamic parameters. For ‘’ species, is independent of the total RNA concentration : Lower concentrations of RNA (<10 µM) favor the formation of a single species, with a melting temperature of ∼52°C (C, blue trace). However, at higher RNA concentrations (>10 µM) a second hypochromic transition appears in the range 63–86°C, corresponding to a new S3F-M2 conformation with a melting temperature of ∼79°C (C, brown trace). These findings are consistent with the native gel electrophoresis results that showed the presence of more conformations in S3F-M2 at RNA concentrations higher than 10 µM (D). The of the 38–67°C hypochromic transition is independent of the RNA concentration (D), indicating that the G quartet conformation formed by S3F-M2 at low RNA concentrations is ‘’. The standard enthalpy, entropy and free energy of G quartet formation in S3F-M2 RNA, which were obtained by fitting the 38–67°C hypochromic transition to Equation () (Materials and Methods section), are summarized in . The values of the thermodynamic parameters for G quartet formation (Δ = −43.1 ± 0.1 kcal/mol and Δ = −3.6 ± 0.1 kcal/mol) are consistent with the presence of two G quartet planes in the structure of S3F-M2 RNA [the enthalpy of formation of a single G quartet plane in an intramolecular G quadruplex, measured in similar experimental conditions, ranges from −18 to −25 kcal/mol ()]. B illustrates a possible S3F-M2 RNA structure consistent with these results, in which two G-tetrads are stacked in a parallel manner. It has been proposed that due to their sequence, the GGA-containing mRNA targets of FMRP might adopt a more complex structure containing a hexad formed by a G quartet flanked by two adenines (), however, our results are not consistent with this proposal. First, the amino protons of two guanines involved in the hexad formation are hydrogen-bonded to adenines at the N7 position, giving rise to sharp resonances in the 1D H NMR spectrum. We do not observe any sharp guanine amino proton resonances in the H NMR spectrum of S3F-M2 RNA. Second, the CD spectrum of S3F-M2 lacks the small positive band at 305 nm that seems to be associated with the presence of these hexads () (Lipay,J. and M.-R.M., unpublished data). To characterize the thermodynamics of FMRP RGG box binding to S3F-M2 RNA, we have employed fluorescence spectroscopy. The S3F-M2 RNA used in this study, named S3F-M2_15AP, was labeled at the 15th position by the highly fluorescent purine analog 2-AP (highlighted in red in B). Based on previous studies in our laboratory, we anticipated that the 2-AP at the 15th position will be a sensitive reporter of the G quartet formation (). The CD spectral features of S3F-M2_15AP indicate that it forms the same type of G quartet structure like S3F-M2 RNA. Moreover, the UV melting profiles of S3F-M2_15AP and S3F-M2 are very similar, indicating that the 2-AP insertion did not cause major perturbations in the structure and stability of S3F-M2_15AP RNA (data not shown). In addition, we determined by EMSA that the FMRP RGG box binds identically to S3F-M2 and S3F-M2_15AP RNAs (A). The steady-state fluorescence of 2-AP is sensitive to stacking interactions, and we expected to observe a change when S3F-M2_15AP is folded in the presence of K versus Li, since the structures formed by the RNA in the presence of these ions are likely very different (compare A and B). As shown in B, the steady-state fluorescence of S3F-M2_15AP increases 5-fold when the RNA is folded in the presence of KCl (forming a G quartet structure in which the 2-AP reporter is located in a G quartet surrounding loop) as compared to the case when it is folded in the presence of LiCl (that does not promote G quartet formation). This result establishes that the 2-AP reporter in S3F-M2_15AP RNA is sensitive to the G quartet structure formation. Next, we measured the binding of the FMRP RGG box to S3F-M2_15AP RNA by titrating increasing concentrations of the FMRP RGG peptide, and monitoring the steady-state fluorescence change of the 2-AP reporter (C). C also shows the results of two negative control experiments: in the first one increasing amounts of the FMRP RGG box were titrated into a solution of Sc1-sh RNA, an RNA previously shown by native gel electrophoresis not to be bound by the FMRP RGG box (,). Sc1-sh RNA forms a G quartet, but lacks a stem structure, and in our experiment we used a 2-AP labeled Sc1-sh RNA in which the 2-AP reporter is located in one of its G quartet surrounding loops (). In the second negative control experiment, we have titrated increasing amounts of the FXR2P RG cluster, a non-binding peptide (see subsequently) to S3F-M2_15AP RNA. A dissociation constant, , of (0.7 ± 0.3) nM was obtained by fitting the FMRP RGG box binding curve to Equation () (Materials and Methods section), indicating that this peptide binds with very high affinity to S3F-M2_15AP RNA. of (−12.5 ± 0.2) kcal/mol. We measured a value smaller by two orders of magnitude than the value of 75 nM reported by Darnell . () for the full-length FMRP or its RGG box binding to S3F RNA. This discrepancy could originate from the fact that Darnell . measured an average value for the FMRP RGG box binding to both S3F conformers that exist even at low RNA concentrations (our native gel electrophoresis results indicate that both S3F-lg conformers are bound by the FMRP RGG box; data not shown). In the case of S3F-M2 RNA we measured the FMRP RGG box binding to the single conformer adopted by this RNA at nanomolar concentrations. Another RNA for which the thermodynamics of FMRP binding has been determined is Sc1 (), a model G quartet forming RNA identified by the SELEX method (). It is interesting to note that the FMRP RGG box binds tighter to S3F-M2 RNA by approximately one order of magnitude: of 0.7 nM for S3F-M2 RNA versus 7 nM for Sc1 RNA, which translates to a difference in the free energy of binding of 1.5 kcal/mol. A comparison of the sequences of these two RNA molecules reveals differences in the G quartet surrounding loops and in the junction connecting the G quartet structure with the stem, which likely account for the difference in their binding by the FMRP RGG box. To define the forces that drive the interactions between the FMRP RGG box and S3F-M2 RNA we have determined the enthalpy and entropy of binding by measuring the association binding constant, = 1/, as a function of temperature. The thermodynamic parameters of binding (summarized in ) were determined from the slope and intercept of the Van't Hoff plot of ln () versus 1/, which is linear when the change in enthalpy (Δ) is independent of temperature [Equation () and D]. −41.4 ± 3.9 kcal/mol, with an unfavorable entropic contribution = −28.9 ± 3.8 kcal/mol. Contributions from hydrogen bonds, van der Waal's or electrostatic interactions are generally associated with the favorable negative enthalpy changes, whereas a decrease in the conformational flexibility or the exposure of hydrophobic residues to the complex surface are associated with unfavorable entropy changes. The association between Sc1 RNA and FMRP RGG box has also been reported to be enthalpically driven, with an unfavorable entropic change (). However, this finding cannot be generalized for all mRNA targets of FMRP since we determined that the FMRP RGG box binding to the microtubule-associated protein, 1B RNA (another proposed G quartet forming RNA target of FMRP) is enthalpically driven only at temperatures higher than 30°C (Menon,L. ., manuscript in preparation). To evaluate the role played by electrostatic interactions in the S3F-M2_15AP RNA:FMRP RGG box recognition, we have determined the association constant = 1/ in the presence of increasing salt concentrations in the range 150–1000 mM. The dependence of on the concentration of monovalent salt concentrations is known as the salt dependence ∂log /∂log [M]. We found that within experimental error does not change in the presence of increasing KCl concentrations in the range 150–1000 mM KCl (E and ), indicating that electrostatic contributions do not play a dominant role in the binding of FMRP RGG box to S3F-M2_15AP RNA. To determine if the FMRP RGG box binding has any influence upon the stability of the G quartet structure of S3F-M2 RNA, we measured its melting temperature when the RNA is in complex with the RGG peptide. The UV melting curve of the S3F-M2 RNA:FMRP RGG box complex shown in E has been corrected by subtracting the UV melting curve of the free FMRP RGG box peptide. The 38–67°C UV hypochromic transition corresponding to G quartet dissociation in the free RNA is now shifted in the range 42–78°C, corresponding to an increase of the G quartet structure from ∼52°C to ∼65°C (). Thus, in a 1:1 ratio, the FMRP RGG box increases the stability of S3F-M2 G quartet structure. It is very interesting to note that upon binding the RGG peptide, a second hypochromic transition appears in the range 79–91°C, indicating that the peptide promotes the formation of an alternate more stable structure of S3F-M2 RNA. The FMRP RGG box binding has also been shown to stabilize the G quartet structure of Sc1 RNA (), however to a different extent: the difference in melting temperatures of the G quartet structure in the free RNA and in the RNA in complex with the RGG peptide is ∼20°C for Sc1 RNA and ∼13°C for S3F-M2 RNA. Thus, in the case of S3F-M2, only a small fraction of the FMRP RGG box binding free energy is used to stabilize the RNA G quartet structure ( and E). In contrast, in the case of Sc1 RNA, a significant fraction of the binding free energy is used to stabilize its G quartet structure (). We inquired next if the recognition of the G quartet structure in S3F-M2 RNA is unique to the FMRP RGG box, or if this RNA is also recognized by the two FMRP autosomal paralogs, the FXR1P and FXR2P. FXR1P has an RGG box different in sequence from that of the FMRP RGG box, whereas the FXR2P has an RG cluster, but not RGG repeats (A). Using EMSA we determined that S3F-M2 RNA is bound by the FXR1P RGG box (A, lanes 5 and 6), but not by the FXR2P RG cluster (A, lanes 3 and 4). To get a quantitative measure of the FXR1P RGG box binding we determined the binding curve by titrating increasing amounts of the peptide into a solution of 150 nM S3F-M2_15AP RNA and monitoring the steady-state fluorescence change of the 2-AP reporter (B, green trace). The of (55.0 ± 3.8) nM, which was determined by fitting the binding curve with Equation (), indicates that the FXR1P RGG box binds with high affinity to S3F-M2_15AP RNA. To determine if the binding of the FMRP and FXR1P RGG boxes to S3F-M2_15AP is specific, we measured their binding either in the presence of a 10-fold excess of a non-specific RNA, Munc13 site1 () or in the presence of 6-fold excess of the FXR2P RG cluster. As shown in B and C, both the FMRP and FXR1P RGG boxes bind specifically to S3F-M2_15AP RNA, since their binding curves are identical in the presence or absence of a large excess of the Munc13 site 1 RNA or of the FXR2 RG cluster (the values are reported in the legend). This result is in contrast to what has been reported for the G quartet-forming Sc1 RNA (), since this RNA is bound specifically only by the FMRP RGG box, but not by the FXR1P RGG box. This difference in specificity of the FXR1P RGG box binding to Sc1 and S3F-M2 RNAs is likely due to structural differences in their G quartet and/or junction regions. Thus, the ability of FXR1P RGG box to bind specifically and with high affinity to the G quartet-forming mRNA targets of FMRP might be modulated by subtle differences in the particular G quartet structure adopted by the RNA. Another level of complexity becomes apparent if one considers the G quartet RNA-binding activity of the full-length FXR1P, as opposed to just of its RGG box domain. Recently, it has been reported that of three different FXR1P isoforms, only one is able to bind specifically to N19 RNA, a G quartet-forming segment of the FMRP mRNA (). Since all these three FXR1P isoforms contain the RGG box domain, it has been suggested that the FXR1P RGG box domain is not sufficient to bind to the G quartet structure and that a 27 amino acid stretch, present only on the FXR1P isoform that binds specifically to N19 RNA, might directly assist the FXR1P RGG box in binding the G quartet, or it might alter the structure of C-terminal portion of FXR1P, thereby allowing binding. We show here that the FXR1P RGG box domain can bind specifically to G quartet forming RNA, so it is more likely that the differences observed in the binding activity of the full-length FXR1P isoforms originate from differences in the C-terminal structure that might have an impact on the accessibility of the RGG box domain. Since the FXR1P RGG box binds with high affinity and specificity to S3F-M2_15AP RNA, we assessed also if this peptide has any effect upon the stability of the RNA G quadruplex structure. D shows that the 38–67°C UV hypochromic transition corresponding to G quartet dissociation in the free RNA is shifted in the range 50–72°C when the RNA is complexed with the FXR1P RGG box, corresponding to a of ∼61°C (). Thus, the FXR1P RGG box binding is also slightly stabilizing the G quadruplex structure of S3F-M2 RNA. Moreover, like the FMRP RGG box, the FXR1 RGG box binding induces the formation of an alternate G quadruplex structure in S3F-M2 RNA, whose dissociation has a hypochromic transition in the range 73–93°C. The FXR2 RG cluster has no effect upon the stability of the S3F-M2 G quadruplex structure () and it does not promote the formation of a secondary alternate structure (E). To investigate the effect of the FMRP and FXR1P RGG boxes on the intramolecular G quartet structure of S3F-M2 RNA, we compared the CD spectra of the free RNA with those of the RNA in complex with the RGG peptides. At a 1:1 ratio of S3F-M2 RNA:RGG peptide, the intensity of the CD band at 263–238 nm was almost unchanged, for both the FMRP and FXR1P RGG boxes (A and B). However, at higher ratios of the RNA:RGG box, both peptides induced the unstacking of the G qudruplex structure of S3F-M2, as reflected by a decrease of intensity of the 263 nm CD band as well as a shift to 265 nm. As a negative control we have also performed the same experiment in the presence of the non-binding FXR2 RG cluster (Supplementary Figure 3) It is interesting to note that the actions of the FMRP and FXR1P RGG peptides are very different: the FXR1P RGG peptide starts to unwind the G quartet RNA structure at only an 1:2 RNA:peptide ratio, whereas this effect occurs for the FMRP RGG box at an 1:6 ratio. At RNA:RGG peptide ratios higher than 1:4 the solutions become turbid, similar to what has been reported for the RGG box of nucleolin protein interactions with MS2 phage RNA (). To rule out the possibility that the RNA is degraded in the presence of the large excess of the RGG peptides, we have treated the solution containing the 1:10 RNA:peptide complexes with proteinase K, which degrades the RGG peptides, and re-acquired its CD spectrum (corrected for the proteinase K contribution). As shown in C and D, the removal of the RGG peptides, allows the free RNA to refold into a G quadruplex structure, indicating that the spectral changes we observed in the presence of a large excess of the RGG peptides are not due to the RNA degradation, but due to the G quadruplex structure unstacking. The unwinding of the S3F-M2 RNA G quartet structure by the FMRP RGG box occurs at high RNA:peptide ratios and it is accompanied by an increase in the solution turbidity; thus it is not clear if this event is biologically significant. However, the finding that the FXR1P RGG box starts to unwind the G quartet structure of S3F-M2 RNA at only an 1:2 RNA:peptide ratio might be relevant considering that the FXR1P has been shown to exist in living cells as a homo-multimeric complex (). p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Alternative acceptors (AA) constitute ∼20% of all conserved alternative splicing events in humans and mice (). During the second transesterification step of the splicing process, the acceptor site, namely the 3′ splice site (3′SS), is selected by the splicing machinery. In mammalians, the 3′SS is defined by a highly conserved AG dinucleotide, a polypyrimidine tract (PPT) located upstream of the splice site and an invariant adenine which is part of the consensus branch point (BP), normally found upstream of the PPT (). Early genomics studies have observed that certain splice site compositions are more probable than others, and these probabilities were used to derive the Shapiro–Senapathy splice site scores (). Other models were further developed to characterize the 3′SS, such as the MAXNET algorithm, which takes into account position dependencies (). Despite the relative weak signal of the 3′SS, the splicing machinery can accurately recognize the authentic splice site from an array of potential sites. In many cases, more than one splice site can be identified, leading to AA which may be regulated in a tissue specific manner (). experimental studies have shown that when two AG sites are placed downstream of the BP and PPT; most often the AG site which is located proximal to the BP is preferred by the splicing machinery (,). Nevertheless, selection of a distal AG site both in constitutive or alternative splicing has routinely been observed (). It was proposed that the selection of a distal splice site can be influenced by its closeness to the BP and the proximal AG site (,). Specifically, it has been shown that AGs which are relatively close to the BP can be bypassed and that close AG dinucleotides are highly competitive for binding to the spliceosome (). In the far end, a number of alternatively spliced exons were found to be characterized by an extremely large distance between the BP and the 3′SS. This region is known as the AG exclusion zone (AGEZ), where AGs are recognized but not utilized by the splicing machinery, possibly repressing downstream splice sites (). Further experiments demonstrated that a proximal AG dinucleotide can be recognized during the first transesterification step, leading to the selection of a neighboring distal splice site, even when the proximal AG is not functional (). In agreement with the observed nucleotide preferences, competition experiments have confirmed that the nucleotide preceding the AG can influence the choice of 3′SS: CAG > TAG > AAG > GAG (). The composition of the 3′SS is thought to be an important component in the process of splice site selection. Recently it was shown that the identity of the nucleotide (N) preceding the invariant AG splice site is also associated with the observed splice site selection in the NAGNAG motif, where the two potential acceptor sites are placed in tandem (,,). In addition, the recognition of the 3′ss during the first () or second () step of transesterification, as well as the tendency of a cryptic splice site to be selected or avoided (), is dependent upon the length and composition of the PPT and the presence of splicing factors that bind -regulatory elements nearby the splice sites (,,). Many regulatory factors have been shown to be involved in splice site selection, e.g. U2AF (,), hSlu7 (), as well as splicing enhancers and silencers such as SF2/ASF and hnRNPA1 (). Experimental studies have shown that AG dinucleotides which are appropriately positioned relative to the BP have an intrinsic potential to become active splice sites (,). What is less clear is how the selection or avoidance of alternative AG dinucleotides is regulated in order to prevent undesired transcripts from being produced. In the last few years, global computational analyses of alternative splicing have been performed, focusing on exon skipping events () and alternative 3′/5′ splice sites (,). Generally, these automatic methods achieved good performance when classifying skipped exons () as well as distant (>50 nt) alternative 3′/5′ sites from constitutive splice sites (), based on sequence features such as splice site strength (,), composition and position of the PPT (,), evolutionary conservation () and frame preservation (,). In addition, several unrelated studies have demonstrated an unusually high level of intronic conservation flanking skipped exons (,,). This property was successfully used for automatic classification of alternative exons (). High levels of intronic conservation were also observed upstream of NAGNAG 3′SS that undergo AS, while intron conservation upstream of constitutively spliced NAGNAGs was generally low (). Though not fully understood, the sequence conservation at the intronic regions flanking alternative spliced events suggests the presence of regulatory elements which are under evolutionary selection. In addition, differences in the overabundance of exonic sequence enhancers (ESE) and exonic sequence silencers (ESS) in the vicinity of alternatively spliced exons compared to constitutively spliced exons were also observed, revealing a complex relationship between splice-site selection and presence of splice-factor binding sites (,). Here we applied a genome-wide approach to analyze human–mouse conserved AA. In order to identify properties which are characteristics of AA, we have analyzed sequence features such as splice site strength, PPT and BP position and composition; intronic evolutionary conservation, ESE/ESS density, GC content and pseudo splice site distribution. We have divided the AA into subgroups according to the distance between the alternative splice sites and compared them to equivalent groups of constitutive acceptors. We have applied both classical statistical analyses on the individual features as well as a machine-learning approach [Support Vector Machine (SVM)] to study the effect of the different features on splicing selection. We show that different splicing patterns can be better differentiated when combining multiple features and that the contribution of the different features to SVM performance varies in relation to the distance between the splice site pairs. Furthermore, we observed an unexpectedly high occurrence of the alternative splicing events in which the PPT was found to overlap the upstream (or proximal) splice site. Overall, the occurrence of multiple PPTs as well as high intronic conservation in the vicinity of the splice sites, are unique properties of AA. Finally, we suggest that the observed differences between the sequence properties of alternative versus constitutive splice sites are indicative of a regulatory mechanism that is involved in the process of splice site selection. We postulate that the process of splice site selection depends on the distance between the competitive splice sites. AA events derived from a database of human–mouse conserved alternative splice sites () were analyzed. In addition, a control set of pseudo acceptors separated from the constitutive splice sites by an AG-depleted region (CA/PA pairs) was extracted. The term ‘pseudo acceptors’ refers to HAG triplets (AAG, TAG or CAG), which were not identified as splice sites based on the existence of EST or mRNA. To avoid the inclusion of alternative splicing events that were undetected in the EST data we required that the PA site (specifically the AG) in each group was not conserved between the species. In addition GAG triplets were not accounted as pseudo acceptors since these rarely serve as splice sites (). We also discarded events in which human–mouse alignments (hg17/mm7) for at least 30 nt upstream to the splice site were not available in the UCSC Genome Browser (ending up with a total of 396 AA pairs and 55,606 CA/PA pairs). #text The region between the distal splice site and 100 nt upstream of the proximal or pseudo splice site was screened for the existence of PPTs. The screening process involved four major steps: For each PPT, the following characteristics were computed: Percent of pyrimidine after extension (defined also as the PPT score), PPT length, the distance of the PPT to the proximal site (PPT∼P) and to the distal site (PPT∼D). In addition, we computed a series of binary features describing whether the PPT is placed upstream, downstream or overlapping the proximal (or pseudo) splice site. xref #text sub italic inline-formula #text T h e f r e q u e n c y o f G a n d C w a s c o m p u t e d f o r e a c h s e q u e n c e a t t h e r e g i o n o f 1 0 0 n t u p s t r e a m o f t h e p r o x i m a l o r t h e p s e u d o s p l i c e s i t e . #text T h e o c c u r r e n c e o f N A G t r i p l e t s i n t h e r e g i o n o f 1 0 0 n t u p s t r e a m o f t h e p r o x i m a l o r t h e p s e u d o s p l i c e s i t e w a s c o m p u t e d . T h e o c c u r r e n c e w a s c a l c u l a t e d s e p a r a t e l y f o r e a c h o f t h e f o u r d i f f e r e n t t r i p l e t s ( A A G , C A G , G A G a n d T A G ) a s w e l l a s f o r a l l H A G ' s ( A A G , C A G a n d T A G ) a s a g r o u p . We performed an -test and a Student's -test (assuming equal variance) on the FAR, MID, CLOSE, NAGNAG-proximal and NAGNAG-distal datasets using the R Stats package (). In the different groups we analyzed the following features: Np, Nd, IC, PPT score, PPT length, PPT∼D, PPT∼P, BS score, BS∼D, BS∼P, ESE/ESS density, HAG and GC (as described above). The significance of the - and -tests were determined using the Westfall–Young method for i-value adjustment (). Briefly, we re-sampled the set of AA and CA/PA pairs and calculate the - or -test -value. The process was carried 1000 times for each test and the minima of the new -value was retained and compared to the original one, namely the -value of the AA versus CA/PA set without re-sampling. If the latter -value was smaller than the minima of the Westfall–Young procedure and <0.05 the result of the test was considered significant. To estimate the sample size we have calculated the power of the -test (1−β) using the R package, only tests which yielded a power ≥0.9 were further considered significant. SVM is a machine-learning algorithm used to detect and exploit complex patterns in data. The SVM is a kernel-based method applying linear classification techniques to non-linear classification problems. It has been widely used to explore biological problems () including alternative splicing (,). In this study, we used the gist-train-svm software with a linear kernel. Input data was normalized by rescaling the columns to values between −1 and 1. All tests were conducted by applying a ‘leave one out’ cross-validation (jackknife) procedure. The following feature sets were used for training the FAR, MID and CLOSE classifiers: PPT parameters (PPT score, PPT length, PPT∼D, PPT∼P, UP, DN, OVLP), splice site parameters (Nd, Np, Nd/Np, STRE), Intronic conservations (IC, IC), frequency of pseudo splice sites (AAG, CAG, GAG, TAG, HAG) and GC content. The UP, DN and OVLP parameters refer to the relative position of the PPT, whether it is placed upstream (UP), downstream (DN) or overlapping (OVLP) the proximal (or pseudo) splice site. For the NAGNAG classifiers the same feature sets were used with the exception of the PPT∼P. The later was exempted since in NAGNAG motifs the PPT∼P is equivalent to the PPT∼D. The SVM performance was evaluated by the ROC (receive operating characteristics) analysis which plots the true positive rate (TPR) versus True negative rate (TNR) for different cutoffs. The AUC (area under curve) was reported for each test. In addition, we calculated the total accuracy (TA), sensitivity (SN), specificity (SP) and the Matthews correlation coefficient (MCC). In an attempt to better understand the mechanism of AA selection, we have analyzed evolutionarily conserved AA pairs in distances ranging from 3 to 100 nt. These include all events in which both the proximal (upstream) 3′SS and a distal (downstream) 3′SS are evidently involved in splicing in both the human and mouse genomes, based on the existence of EST and mRNA transcripts in both species (). The number of conserved AS events represent a lower bound of the AS events in humans, nevertheless, they are expected to be biologically significant (). It is generally accepted that two putative splicing acceptors which are located in close proximity are highly competitive (,,). To test the dependency between the location of the putative splice site and the splice site selection we divided the data into four separate sets according to the distance between the splice site pairs. The groups were designated: FAR, MID, CLOSE and NAGNAG. The FAR, MID and CLOSE groups were composed of AA pairs separated by 40–100, 13–39, 4–12 nt, respectively. The NAGNAG group included only AA pairs placed in tandem. While in the CLOSE and NAGNAG datasets, we expect both the AA to be placed downstream of the BP, which is generally found between 18 and 40 nt upstream of the splice site (,,,), the FAR dataset included sequences in which the first splice site is expected to be upstream of the BP. However, as reported in Gooding (), in some cases the BP could be located upstream of the proximal site also in the FAR group. The borderline cases were grouped together in the MID dataset. In addition, we compiled a series of control sets containing constitutive acceptors (CA) separated from an upstream potential competitor or pseudo acceptor (PA) by an AG-depleted stretch of variable lengths. The sequences for the control set were deliberately selected so the distribution of the distances between the CA and the PA in the control sets will be equivalent to that of the corresponding alternative dataset. In order to study the AA pairs and compare them to 3′SS which are constitutively chosen, we have analyzed a series of intronic properties calculated for each group. Among the features we included were splice site strength, intronic evolutionary conservation (excluding the splice sites), length and score of the PPT and its position relative to the splice sites, BP score and distance to the splice sites, GC content, ESE/ESS density and the occurrence of other AGs dinucleotides (see Methods section). The latter were previously found to affect the recognition of the splice sites when they occur in upstream intronic regions (,). As in the majority of cases [excluding the class of distal BPs ()], the splicing regulatory elements are close to the 3′SS, we restricted the analysis to 100 nt upstream of the proximal site. The above properties were calculated for all sequences in each AA subset and compared to the corresponding subset of CA/PA pairs. In the case of the NAGNAG group, comparisons were conducted against two independent sets of CA/PA pairs: NAGNAG motifs in which the distal site is constitutively spliced and the proximal is a competitive pseudo acceptor (NAGNAG-distal); and the reverse case where the proximal site is constitutively spliced and the distal NAG serves as a pseudo acceptor (NAGNAG-proximal). Although competitive sites are generally placed upstream of the splice site (,), in the unique case of NAGNAG, we chose to test the two control sets since, in tandem acceptors, both the distal and proximal sites were previously suggested to contribute to the competition. Furthermore, these NAGNAG motifs are of special interest as they are widely distributed throughout the human genome (,). For each of the properties analyzed, we have carried out a statistical analysis applied to all datasets pairs (AA versus CA/PA). A summary of the statistical analysis is given in (detailed results are given in Table 1S). Interestingly, in each subgroup (defined by the splice site distance) we found a different set of features that deviated between the AA and CA/PA pairs. For example, splice site features were discriminative only in the NAGNAG group, both when comparing it to the NAGNAG-proximal and to the NAGNAG-distal (Table 1SA and B). This is in agreement with previous studies which observed correlation between the splice site strength and the splicing pattern at the NAGNAG motif (,). In addition, consistent with our previous results (), the intronic conservation in the 100 nt upstream of the proximal splice site was significantly higher in alternative NAGNAGs compared to the NAGNAG-proximal group (-value for -test = 4.−04). Interestingly, the intron conservation did not appear to differ significantly when comparing alternative NAGNAGs to the NAGNAG-distal group. The latter groups both demonstrate a high intronic conservation which may suggest similar regulatory constraints (). Furthermore we observed a significant difference in the GC content between alternative NAGNAGs and the NAGNAG-distal group. Surprisingly, the high GC content in the NAGNAG-distal group was higher than the average GC content found generally upstream of constitutive acceptor sites (Table 1S). Nevertheless, in the CLOSE group in which the splice sites are very close to each other but not adjacent (Table 1SC), the features which were significantly different between AA and CA/PA pairs were the distance of the PPT and BP to the splice sites and the PPT score (which was statistically significant in the CLOSE group when applying the test). Generally, in AA pairs the PPT and BP appeared closer both to the proximal and distal sites and the PPT displayed a wider variance of scores. As in the NAGNAG-distal group, in the CLOSE group we also observed a relatively lower GC content upstream of the AA pairs. In contrast to the NAGNAG group, in the CLOSE group we did not observe significant differences either in the splice site composition or in intronic conservation between AA and CA/PA pairs. In the MID group we observed weaker PPTs (i.e. lower PPT score) in AA compared to CA/PA pairs. In addition, the intronic conservation levels were higher upstream of AA pairs and the AG dinucleotides were slightly underrepresented (Table 1S D). Among the most discriminating features in the FAR group were the score and the relative position of the PPT and BP. These were found to be weaker in AA pairs and farther from the distal splice site. In addition, the intronic conservation was significantly higher in AA pairs and the occurrence of intronic AGs was underrepresented. Overall, our results suggest that only when the two splice sites are placed in tandem (NAGNAG) the splice site composition, namely the identity of the nucleotide preceding the conserved AG, appeared to be discriminative. In contrast, when the distance between splice sites is larger, differences in the PPT composition and the relative location of the PPT and BP to the splice sites seem to play an important role. Consistently, the level of intronic conservation appeared to be discriminative in the MID and FAR groups in which the AA distance is >12 nt (Table 1S, ). However, it is important to note that, in the unique group of the NAGNAG acceptors, the intronic conservation was also found to be statistically significant (though to a lesser extent) only when comparing tandem acceptors to the proximal NAGNAG group and not in comparison to the distal group. Likewise, we observed that the intronic evolutionary conservation in the CLOSE group is relatively high (though not statistically significant) for AA pairs when compared to the average intronic conservation upstream of randomly chosen constitutive splice sites (A). A relative high intronic conservation was also detected in the CA/PA pairs only in the close region adjacent to the pseudo splice site (). The high intronic conservation levels found generally upstream of alternative splice site are consistent with other studies suggesting the existence of regulatory elements involved in splice site selection (). Nevertheless, the relative high intronic conservation close to the splice site in CA/PA pairs, specifically in the CLOSE group, could be due to the presence of regulatory elements which may be involved in controlling constitutive splicing when potential competitors are present. It is important to note that in our dataset the pseudo splice sites (PAs) themselves are not evolutionarily conserved and thus the higher intronic conservation observed cannot be related to an overall high conservation of the site or an alignment artifact. The PPT is a key feature in splicing regulation. Previously, it has been shown that both the composition and the distance of the PPT can influence splice site selection (). The PPT is commonly identified by splicing factors such as the splicing repressor PTB and the splicing enhancer U2AF. Recent reports demonstrate that both factors can compete with each other for binding the PPT (). Although these proteins are considered basic splicing factors, it has been suggested that they also play an important role in the regulation of alternative 3′ splice sites (). As described in the pervious section, we have conducted a comprehensive analysis of the PPT in AA and CA/PA pairs in the different subsets. A and B illustrate the relationship between the distances of the PPT to the distal splice site and the distance between the proximal splice site to the distal splice site in AA pairs and in CA/PA pairs, respectively (in CA/PA pairs CA is equivalent to distal and PA to proximal). Each dot represents the relationship in one sequence. As shown, when the distance between the proximal and distal splice sites is 3 (in the NAGNAG subset) in both alternative and constitutive splice sites a PPT is found anywhere between 0 and 90 nt upstream of the splice sites. However, when the splice sites are not in tandem, in AA pairs the predicted PPT is found close to the distal splice site only when the splice sites are relatively close to each other, separated by <40 nt (A). This is most probably related to the dependency between proximal site and the PPT in the AA pairs. As demonstrated by the dots lying along the diagonal in A, in the majority of AA pairs the predicted PPT falls in close proximity to the proximal splice site. In the CA/PA pairs (B), although we do observe a large proportion of PPTs located in proximity to the distal site we could not detect any clear relationship between the location of the PPT and the PA. To ensure that these results are not due to the smaller sample size of the AA pairs, we have randomly selected from the full set of all constitutive events 1000 sets of equal size to the AA group. For each set we calculated the number of cases in which the PPT was adjacent to the distal site (<5 nt apart) in CA/PA pairs separated by ≤40 and >40 nt and compared it to the distribution in the AA set. We have applied a series of Fisher-exact tests comparing each random set to the AA set and all cases showed a significant difference between the groups (using the Bonferroni correction < 5*10). These results confirmed that the dependency between the ‘distal site-PPT’ distance and the ‘distal-proximal’ distance is restricted to the AA set. Previous studies have suggested that efficient repression by PTB depends on the existence of two binding sites which can mediate the formation of a stem-loop structure by protein–protein interactions between PTB monomers, known as the ‘looping out’ model (). This mechanism was originally suggested in order to explain the regulation of alternative exons; however, BPs where also proposed to be looped out and avoided by the splicing machinery (). In addition, a recent computational study reported the existence of two PPTs flanking the upstream splice site when alternative 3′SS are separated by ≥8 nt (). To further identify features which could be used to differentiate between alternative and constitutive splice acceptors at a genomic level, we searched for the existence of an additional polypyrimidine stretch in the region of 100 nt upstream of the splice site (both surrounding AA and CA/PA pairs). The definition used to automatically assign the second PPT is described in detail in the Methods section. Generally, the assignment of PPTs was done based on their relative size, PPT1 being the longest stretch. Overall, we did not observe a clear difference between the alternative and constitutive splice sites when simply considering the length of PPT2 ( = 0.060) or its relative strength ( = 0.596). It is important to note that in ∼20% of the AA CA/PA pairs we were not able to find an additional PPT, while in 19% of AA and 27% of CA/PA pairs we found both PPTs either upstream or downstream of the splice site (Table S3). Subsequently, we evaluated the relative position of both PPTs in all AA and CA/PA pairs, concentrating on the FAR group. We found that in 53% of the AA pairs, one PPT was found overlapping the splice site and the other one was found upstream of it [PPT-PPT(p)]. In contrast, this pattern accounted for only ∼6% of all CA/PA pairs (). These observations reinforce that the high occurrence of overlap between the proximal splice site and the PPT is not restricted to close splice site pairs, in which the proximal site by default falls within the PPT due to space restriction as in the CLOSE and MID groups. These results are also consistent with previous observations by Dou (). In addition, in most AA pairs analyzed in our study, the PPT that was found to overlap the splice site was the larger one among the two PPTs (Table 4S). Furthermore, we observed that ∼54% of the pseudo splice sites in CA/PA pairs were found to be flanked (but do not overlap) by two PPTs (PPT-p-PPT). This pattern was observed only in ∼17% of the AA pairs (). Generally, in AA pairs the PPTs were always found upstream of the proximal splice site or overlapping the splice site, but never downstream of the proximal site ( and 2S). The fact that PPTs were not detected between AA pairs could be due to the coding potential of this region. Nevertheless, it could suggest that a downstream PPT alone is not capable of regulating an upstream splice site. To test whether the combination of features described above can better separate AA from CA/PA pairs, we have built a SVM classifier for each of the datasets: NAGNAG, CLOSE, MID and FAR. SVM is a supervised machine-learning algorithm which is trained to separate between two sets of data. It has been previously applied to automatically identify alternative exons based on both exonic and intronic properties, including evolutionary conservation, length of PPT and splice sites composition (). In addition, a recent study has applied SVM to differentiate between alternative and constitutive 3′/5′ splice sites based on parameters such as PPT, splice site composition and frame preservation (). Here we applied the SVM algorithm to distinguish alternative versus constitutive 3′ss events in each subset independently. The feature set was composed of intronic features calculated in the previous section including the splice site composition, PPT properties; intronic conservation, pseudo splice site occurrence and GC content (see the Methods section for details). Since the position of the predicted BP relative to splice site was found to be highly correlated with the relative position of the PPT (Pearson correlation ∼0.9), we did not include this parameter in the SVM feature set. Additionally, we have not included the ESE and ESS density as parameters for SVM as they were not found to be statistically significant in any of the subsets. To estimate the performance of our method, we have performed a ‘hold one out’ cross-validation test (also known as the ‘jackknife’ test). For each test, we plotted a receiving operating characteristics (ROC) plot and calculated the area under the curve (AUC), the sensitivity, specificity, total accuracy and The Matthews correlation coefficient (). As shown, the best SVM performance was achieved for the AA FAR versus CA/PA FAR classifier, for which the AUC was 0.94, followed by the MID (AUC = 0.91), the CLOSE (AUC = 0.80) and lastly the NAGNAG-P (AUC = 0.78) ( and ). As shown in the NAGNAG-D classifier, we encountered a notable decrease in SVM performance (AUC = 0.69). These results are in accordance with our previous reports showing similarity in genomic properties between alternative spliced NAGNAG motifs and NAGNAG motif in which the distal site is chosen constitutively (). Overall in agreement with the recent study by Xia () we observe that the performance of the SVM increases as a function of the distance between the splice site pairs. Different from Xia (), in the current study we also obtained a considerably high performance for the classification of splice site pairs which are in close proximity, with sensitivity varying between 60 and 85% in the different subgroups (). The difference in SVM performance obtained for the close sites in the different studies is probably due to the unique features which were selected for the study. As shown in , the features which were found to contribute mostly to SVM performance in the FAR and MID groups were the intronic conservation, (especially in the MID group) and the PPT features, including both the PPT length and the relative distance of the PPT to the splice site. Interestingly, the effect of removing either the PPT or the intronic conservation feature sets from the FAR group was very similar, suggesting that they are both important in the latter group. In contrast, the removal of the conservation set from the MID group had a remarkable effect compared to the removal of the PPT set, suggesting that in this subset the contribution of each of the parameters to the learning process is different. This result could be due to the fact that the MID group includes the borderline sequences and may represent a mixed distribution. In the CLOSE group, the most significant change in the AUC value arose when the PPT features were eliminated. In addition, we observed a lower reduction in SVM performance when eliminating each of the other features from the CLOSE classifier. This indicates that in the CLOSE group (different from the FAR and MID group) the learning process is mostly ruled by a unique feature set. In the NAGNAG-proximal group, the most notable features were the splice sites followed by the intronic conservation. This is in agreement with the statistical analysis results of the current study and previous observations (,). Overall, the feature selection test was consistent with the statistical analysis. Nevertheless, the latter results reinforce that the differences between AS and CS do not rely on unique parameters, but rather a combination of several sequence features. In an attempt to understand the regulation of alternative splice site selection, we have conducted a comprehensive analysis of alternative acceptor (AA) pairs separated by a range of distances. In order to concentrate on functional AA, we have restricted our study to alternative splicing events conserved between humans and mice (). Applying both classical statistics and machine-learning approaches, we demonstrate that a combination of splicing canonical elements found in the introns show major variations between alternative and constitutive acceptors. Most importantly, we find that the ensemble of properties which distinguish between alternative and constitutive splicing strongly depends on the distance between the spice sites. In agreement with previous work, the splice site composition was found to contribute mostly to SVM performance when splice sites are placed in tandem, as in the NAGNAG motif (,,). In addition, we found considerable contribution of the intronic evolutionary conservation levels flanking the NAGNAG motif in comparison to the NAGNAG-proximal control set, but not compared to the NAGNAG-distal set. This is consistent with the high conservation levels that were previously observed both upstream of alterative spliced NAGNAGs and upstream of NAGNAG motifs in which the distal splice site is constitutively chosen (). Furthermore, in this study we observed that when the splice sites are placed nearby (4–12 nt), but not in tandem, there is a relatively high evolutionary conservation level both upstream of alternative splice sites and PAs. In both cases, the conservation was higher than the background conservation level found upstream of constitutive splice sites. The similarity in the intronic properties flanking AA and CA/PA pairs when the splice sites are relatively close was reinforced by the relative low contribution of the intronic conservation feature to the SVM performance in the CLOSE group. Nevertheless, when the distance between the splice sites increased (MID, FAR), we found that elimination of the intronic conservation features strongly affected the SVM performance. It is known that two AG sites placed in close proximity are highly competitive (,,); hence, the relatively high intronic conservation levels observed upstream AA and CA/PA pairs could be indicative of regulatory elements important to avoid the selection of alternative (or pseudo) AG sites that by default would be preferred by the splicing machinery (,). These results are consistent with recent work by Wang , which observed high levels of regulatory elements in the exonic regions between competitive splice sites (). In accordance with a recent report (,), here we have also observed that the most important feature to discriminate between alternative and constitutive acceptors was the PPT. PPT-related features were statistically significant in the CLOSE, MID and FAR groups and were found to play a significant role when combined with other features during the learning process. The fact that removing the PPT had a lesser effect on SVM performance in the MID and FAR compared to the CLOSE group could be related to compensation by other features such as the intronic evolutionary conservation. A striking observation in the current study was the high frequency of PPTs overlapping the proximal splice site, which was predominately observed in AA pairs. This observation coincides with previous studies describing the existence of two PPTs upstream AA pairs separated by ≥8 nt. Different from AAs, pseudo acceptors were found to be mostly located between, but not overlapping two PPTs. This could indicate a mechanism by which the two PTB-binding sites mediate looping-out of the pseudo splice sites. Although the looping out mechanism was originally suggested to explain alternative splice selection (,) our data imply that it could also be involved in avoiding the selection of pseudo splice sites. The high occurrence of AA pairs in which the proximal splice site is located inside the PPT suggests an important contribution of the PPT to the regulation of proximal splice site selection. This is supported by previous studies describing the competition between the splice factor U2AF and PTB for binding to the PPT () and the involvement of both factors in the regulation of alternative 3′ss (). Moreover, it was observed that most of the disease related 3′splice sites are found within the PPT (), indicating that the overlap between the PPT and the 3′SS enhances the likelihood of an AG site to be chosen. Further experimental analysis will be needed to uncover the effect of the overlap between the splice site and the PPT on alternative 3′ss selection. In summary, this study supplies further evidence of the involvement of basal splicing elements in the regulation of alternative splicing. Overall, our results suggest that differences may exist in the regulation of splice site recognition depending on the distance to the neighboring splice site candidates. Generally our findings, which are based on a bioinformatics analysis of only human–mouse conserved AA are in agreement with several experimental studies which have demonstrated that the proximity between AG pairs can affect splice site selection (). p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Transposable elements occur in many bacterial genomes. We can thus not fully understand bacterial genome evolution, unless we understand how such mobile DNA is maintained, and how it spreads among bacterial genomes. Because transposable elements also cause an important public health threat, the spreading of drug-resistance genes among pathogenic bacteria, such understanding may ultimately also shed light on the epidemiology of drug-resistant pathogens. Insertion sequences (ISs) are among the simplest kinds of bacterial mobile DNA. They range in size from 600 to more than 3000 bp and fall into 20 major families (). Most ISs consist of short inverted repeat sequences that flank one or more open reading frames (ORFs), whose products encode the transposase proteins necessary for their transposition. Some but not all ISs transpose into specific target sites. ISs typically generate a direct repeat of their target site after transposition. Transposition is often associated with an increase in IS copy number in a genome. In eukaryotes many non-functional copies of transposable elements can often be passively proliferated using transposase from functional copies (). In contrast, bacterial IS transposition is often tightly regulated, occurs at a very low level, and is often restricted to activity, where a transposase promotes only the transposition of the IS from which it is expressed (). Exceptions exist, for example in the form of very short miniature inverted repeat elements that may only proliferate passively (). In the near future a flood of bacterial genome data will become available. Such data will see many uses in studying IS families, including the identification of functionally important sequences from hundreds of family members, and the reconstruction of the evolutionary history of individual ISs. Efforts like these require a comparison of ISs across (many) different genomes. Such a comparison is hindered by existing IS annotations which may differ greatly among genomes, because they have been produced by different research groups using different tools. In addition, existing annotations provide limited information about sequence elements such as inverted repeats, or about the structure of ISs where the transposase is encoded by more than one open reading frame. With these limitations in mind, we have developed IScan, a software tool that allows a user to identify ISs and their associated direct and inverted repeats automatically, flexibly and in multiple genomes, using a curated reference IS from a database such as ISfinder (). The consistent annotation provided by IScan will greatly aid evolutionary studies. In two analyses that address two different classes of questions, we applied IScan to 438 completely sequenced bacterial genomes and all 20 major IS families. The first set of analyses addresses the biological question: Why is mobile DNA maintained in bacterial genomes? Mobile DNA might be a very effective parasite, a prototypical example of selfish DNA (,), or it might confer benefits to its host. [For example, mobile DNA can mobilize genes for transfer between bacterial strains or species ()]. Despite its long history, this question has not been completely resolved. To find out whether mobile DNA persists because it benefits a host, one needs to understand the dynamics of mobile DNA on evolutionary time scales. Laboratory evolution experiments () are of limited use here. The reason is that the rates at which ISs transpose, are transferred horizontally, and can cause recombinational and other instabilities are so small (,) that even long laboratory evolution experiments may detect IS copy number and position variation, but may not be sufficient to determine whether ISs have net deleterious or beneficial effects. A different approach to understanding the evolutionary dynamics of ISs focuses on the number and distribution of ISs in bacterial populations or closely related bacterial strains (,). Most pertinent studies were carried out before large-scale genome sequence data became available, and are thus very limited. In a recent paper, we overcame some of the limitations of pre-genome work by analyzing the distribution of five major IS families in 202 complete genomes (). This analysis suggested that ISs within a genome have very low nucleotide diversity, cause their host to go extinct on evolutionary time scales, and can only be sustained by horizontal transfer. In other words, ISs are likely to be detrimental to their host in the long run. However, this earlier analysis was also hampered by our reliance on available genome sequence annotations to identify ISs. We here overcome this limitation by our use of IScan to study the distribution and sequence similarity of ISs in more than twice as many genomes and four times as many IS families than in earlier work. The second of our two applications of IScan addresses a methodological rather than a biological question: Is it possible to distinguish functional from non-functional (especially truncated) ISs computationally—without time-consuming experiments—and for hundreds or thousands of ISs? We suggest an approach based on the similarity of IS inverted repeats. IScan is ideal for this approach, because it can calculate various statistical significance measures for inverted repeat similarity. We show that our approach, while certainly not allowing for perfect discrimination, may enrich a dataset for ISs that are likely to be functional. Via the procedure outlined in the Results, IScan produces a file in FastA format. This file contains the following parts for each IS: We used IScan to search for ISs belonging to the 20 major IS families listed in () in 438 curated bacterial genomes (consisting of 790 sequenced DNA molecules) available from GenBank (). The curated query ISs we used were obtained from the IS repository IS Finder ( 11). We retained BLAST hits to IS ORFs with an -value of  ⩽1 and at least 35% amino acid identity to the query sequence. For ISs with more than one ORF, we used the parameter = 50 () to assign ORFs to the same IS. For identification of target direct repeats we used curated data on the length of direct repeats from ISfinder [(), , ()] to define and for each IS family analyzed. To identify inverted and direct repeats we set = = 0, and = = 1.1 × (total IS length – length of IS coding region), where the total and coding region lengths are again derived from information curated for each IS family's reference sequence (). Inverted and direct repeat alignments were performed with the same scoring matrix of 1 for matches, −2 for mismatches and −5 for gaps and gap extensions. We determined various -values (, , , , see Results) that indicate whether the candidate inverted repeats of an IS are statistically significantly similar. We did so by aligning 10 randomly chosen sequence fragment pairs of length    + 1 = + + 1 from the DNA molecule in which the IS was found. Specifically, for , two randomly chosen fragments are aligned against each other, for , two randomly chosen fragments are aligned against the left reference inverted repeat and for (), one randomly chosen fragment is aligned against the left (right) reference inverted repeat. The fraction of these alignments whose score is greater (indicating greater similarity) than the alignment score of the candidate inverted repeat corresponds to the desired -value. For we used Smith–Waterman local alignment, for , and we used clustalw, which implements a global dynamic programming alignment algorithm (). To estimate synonymous and non-synonymous divergence among IS coding regions, we used a previously published tool (). Briefly, the tool uses information from both the DNA and amino acid sequences, and proceeds in three steps. First, it pre-screens related gene pairs using BLASTP () and the Needleman and Wunsch dynamic programming alignment algorithm [Thompson . ()]. Then, it eliminates gene pairs with fewer than 50 alignable amino acid residues and with <50% amino acid identity from further analysis. In the third step, the tool calculates the number of substitutions per synonymous site () and the number of substitutions per non-synonymous site () using the maximum likelihood models of Muse and Gaut () and Goldman and Yang () for the remaining pairs. It uses a simple heuristic test () to determine whether a gene pair has been saturated with synonymous substitutions. For ISs with overlapping ORFs, we merged, for reasons of computational tractability, the overlapping ORFs into one ORF for the calculation of and . (The short overlapping regions are subject to different evolutionary constraints than the non-overlapping regions). Specifically, we calculated the number of nucleotides that overlap in the two ORFs, and eliminated from a sequence containing both ORFs the segment containing the overlap and any additional nucleotides upstream or downstream of the overlapping segment required to retain the reading frames of the two ORFs. On average, IS ORFs were shortened by four nucleotides through this procedure. IScan identifies transposase sequences, inverted repeats and candidate target direct repeats of ISs in complete genomes. IScan is a free open source package developed on a Linux platform and implemented in perl. It is available from the website: . IScan uses a curated reference or query IS (which, in our case, is a representative member of a major IS family; ) to identify other ISs in one or more completely sequenced genomes, or any other DNA molecules. This query sequence contains (i) the amino acid sequences encoded by one or more transposase ORFs, and (ii) the nucleotide sequence of the upstream (IR) and downstream inverted repeat (IR). We note that ISs with two or more transposase ORFs frequently express a single functional transposase through a translational frameshifting mechanism. The extent and length of required sequence similarity to the reference IS are user-specifiable, such that arbitrarily weakly similar ISs or short IS fragments can be identified if needed. IScan identifies ISs in three major steps. We applied IScan to the complete DNA sequences (chromosomes and plasmids) of 438 bacterial genomes, to identify all candidate ISs in the 20 major IS families whose ORFs had at least 35% amino acid sequence identity to a family prototype sequence () over the length of the prototype sequence. This approach yielded 2091 ISs. 95% of them occurred on bacterial chromosomes, and the remainder occurred on plasmids. The length distribution of the ISs we identified is shown in a. In the literature, the lower end of the typical length range for all IS families considered is 540 bp (,). Only 1.2% of the ISs we identified were shorter than 540 bp. The conspicuous peaks in the length distribution come from individual highly abundant ISs, such as IS1, with a highly abundant member of length 695 bp that causes the highest peak in a. Among all ISs we identified, the most abundant IS families are IS1, IS3, IS481 and IS5 (). The distribution of any one IS family is extremely patchy and highly skewed: The vast majority of genomes contains no member of the family; most genomes that contain one member of the family contain only one member; and typically only very few genomes contain a large number of members. We illustrate this distribution in b (note the logarithmic scale on the -axis). The numbers of IS copies within a genome show no strong statistical associations among different IS families. Specifically, the copy numbers of only 10 among 136 possible IS family pairs, show a statistically significant (Bonferroni-corrected = 0.05) Spearman rank order correlation coefficient. All but three of these statistically significant associations vanish, however, if one eliminates genomes from the analysis in which one or both ISs have no copies. The remaining significant associations involve IS1 and IS3 (Spearman's = 0.93; = 16), IS110 and IS4 ( = 0.87; = 13), as well as IS110 and IS3 (Spearman's = 0.83; = 15). However, the genomes that account for this correlation are from the extremely closely related species of and . This indicates that a common evolutionary history rather than similar host preferences is responsible for the co-occurrence of these ISs. It also illustrates that the shared evolutionary history of many sequenced genomes introduces a bias into the data that has to be taken into account when testing certain hypotheses about IS evolution. We next examined the sequence divergence of ISs within and among genomes. Beyond simple nucleotide divergence (a), we determined , the fraction of amino acid replacement substitutions at amino acid replacement sites, and , the fraction of synonymous substitutions at synonymous sites, an indicator of the synonymous divergence within an IS's transposase-coding genes. Synonymous sites are generally under weaker selection than amino acid replacement sites. In addition, because of the low expression level of transposases, the synonymous sites we study are not subject to selection for translation efficiency. These two observations render a better (albeit crude) indicator of the IS's age than (). To estimate and , we used GenomeHistory, a software tool that estimates and using a maximum likelihood method (). We first focused on the sequence divergence of ISs within a family and within genomes. It is very low. b shows a histogram of the distribution of amino acid sequence divergence for pairs of ISs within the same genome. Note the logarithmic scale on the y-axis, indicating a very large number of sequences at low divergence. The mean (median) is 0.012 (0.0019), and its 90th percentile is = 0.0067, even though our approach would readily detect ISs with amino acid sequence divergence up to greater than 1.33.3% of sequence pairs within a genome are completely identical in their amino acid sequence. Synonymous divergence is similarly low (c). Excluding IS pairs with saturated synonymous divergence (1.01% of all pairs), the mean synonymous divergence is = 0.033 with its 90th percentile at = 0.013. More than 60% of all IS pairs within a genome are identical at their synonymous sites, such that the median = 0. shows the mean and SEs of and separately for each IS family (see also ). While amino acid divergence is relatively homogeneous among families, synonymous divergence varies to a greater extent, and it is particularly high for IS5. Most variable is the ratio /, which is normally taken as an indicator of selective constraint on a protein. The mean ratio is / = 0.39 and it varies between / = 0.13 (IS66) and / = 0.86 (IS4). A high ratio / might be taken to indicate the presence of many inactive ISs whose coding region are pseudogenes and thus evolve effectively neutrally ( = ). However, this interpretation would be misleading, because ISs with high / generally show extremely low intragenomic synonymous and non-synonymous divergence. For example, for IS4, where the mean / = 0.86, the mean values of and are an extremely low = 6.3 × 10 and = 7 × 10. Similarly low divergences also hold for other ISs with high / ratios (). Such small values mean that virtually all IS pairs of a family within a genome differ by one or very few nucleotides. At such low divergence, the interpretation of the ratio / as an indicator of selective constraint is not appropriate, because there may be a large amount of stochastic variation in the number of synonymous and non-synonymous nucleotide changes. In this regard, it is also worth mentioning that IS5, which is an extreme outlier with its high mean synonymous divergence of = 0.58 shows a / = 0.22, close to the lower end of the observed range across families. Not unexpectedly, the sequence divergence among pairs of ISs in different genomes is substantially greater. At a mean = 0.3 synonymous divergence of ISs among genomes is almost 10 times greater than within genomes. Ten times more IS pairs in different genomes have saturated synonymous divergence (10.5% as opposed to 1.01% within genomes). The mean non-synonymous divergence = 0.064 (90th percentile = 0.29) is a factor five higher amongst genomes than within genomes. It is nonetheless very low compared to the maximum = 1 our approach could have revealed. This suggests that the IS families we study are well defined on the sequence level. The presence of closely related ISs of the same family in different genomes indicates the importance of horizontal gene transfer in their spreading. For instance, we find a large number of ISs with identical transposase coding regions in different organisms. Many of these pairs stem from species closely related in evolutionary history or life style, such as ISs in the closely associated genera and . However, some of these pairs involve more distantly related organisms, such as the psychrophilic (cold-loving) arctic bacterium on one hand, and the human and on the other hand, which all share identical IS1 elements. We used several different approaches to estimate the ‘quality’ of an IS's inverted repeats, as indicated by their similarity to each other and to a reference sequence. (These approaches are also implemented in IScan and are available to IScan users.) The first approach involves a local dynamic programming alignment () of the sequences immediately upstream and downstream of the coding sequences that contain the inverted repeats. We compared the score of this alignment with that of a large number of alignments of sequence fragment pairs of the same length but chosen from random positions within the same DNA molecule. This allows us to assign a significance threshold (‘left-right’, ) of observing an alignment score as high as that observed between putative IR and IR by chance alone. a shows a histogram of the distribution of these -values together with an indication of the significance threshold = 0.05, as well as the significance threshold = 0.05/2091 = 2.3 × 10. This lower value corresponds to a Bonferroni-corrected = 0.05. It takes into account that we carry out multiple independent tests, but it is excessively conservative. Although a large fraction (55%) of the -value we determined is significant at = 0.05, not one of them is smaller than = 2.3 × 10. In the second approach, we aligned the left inverted repeat of the ‘reference’ sequence we used to identify ISs of a given family with both the left and right (candidate) inverted repeats of the ISs we identified (). For each IS, we evaluated the statistical significance (‘3-way’) of each alignment score by the same randomization approach as above, using a large number of random DNA fragment pairs from the same molecule, and aligning them to the left inverted repeat of the reference sequence. In this analysis, 79.4% of ISs showed < 0.05, and two showed < 2.3 × 10 (b). In a third analysis, we aligned the putative IR of each IS candidate with the IR of the reference IS that we used to identify the IS in the first place. We then carried out the same randomization approach as above to identify the likelihood P (‘left’) to observe such an alignment by chance alone. We repeated this approach for the putative IR to obtain P (‘right’). To obtain a joint significance score for both IR and IR of an IS, we simply calculated the product × . c shows a histogram of ( × ). In this analysis, 90.3% of × -values are smaller than 0.05, and 55.1% values are smaller than the Bonferroni-corrected = 2.3 × 10. Thus, this alignment strategy significantly enriches for ISs with highly similar inverted repeat units. To be sure, this does not demonstrate that ISs with highly similar repeat units are more likely to be functional. The following analysis suggests, however, that this is the case. ISs that have a family member with identical DNA sequence in the same genome are more likely to be functional than ISs for which this does not hold. The reason is that the two identical family members have most likely arisen through transposition, because gene duplication, the other prominent process that could account for the two IS copies, is much less frequent than transposition (). In addition, bacterial IS transposase activity is usually tightly regulated, and often restricted to the IS from which transposase is expressed, such that passive transposition of a defective IS with the aid of an intact ‘helper’ IS may not occur often (,). This means that many ISs in our dataset with an identical IS in the same genome will be functional. If the alignment strategy we pursued above enriched for functional ISs, we would predict that the × values would be significantly lower for ISs with an identical family member in the same genome, than for other ISs. This is exactly what we observe. For example, for ISs with an identical IS in the same genome, the mean ( × ) = 0.019, whereas for other ISs, the mean ( × ) = 0.046. This difference is highly statistically significant as assessed by either a Mann–Whitney U-test ( = 1478; = 613; = 1.2 × 10) or a -test ( <10). In sum, although the inverted repeat units of an IS have limited sequence similarity, it is possible to enrich a data set for likely functional ISs by considering ISs with high × values. We note parenthetically that we also carried out a sequence similarity analysis among transposition target ‘direct’ repeat units generated by those ISs that are known to generate long direct repeats. This analysis showed that the direct target repeat units have too limited sequence similarity to be useful for this or other purposes (data not shown). We have developed IScan, a publicly available tool to identify IS coding regions, and associated sequence elements (direct/inverted repeats). IScan is able to identify ISs with an arbitrary number of ORFs, including ISs with ORFs encoded on both strands. IS annotation in existing genomes may be highly heterogeneous, because different researchers may use different annotation methods. A tool like IScan thus allows the user to create consistent IS annotation with multiple user-specified parameters (repeat length, sequence similarity to a reference family member, etc.) across multiple genomes. This consistency and flexibility is essential for detailed analyses of IS evolution across multiple genomes. Using IScan, we have surveyed 438 bacterial genomes for members of 20 IS families, and studied the similarity in their coding sequences as well as their inverted repeats. Recent other surveys of IS families focused on different aspects of IS biology and studied fewer or different genomes. Specifically, an intriguing analysis () studied ISs in 262 genomes and focused on the question: What determines IS copy numbers in a genome? (Briefly, the answer is genome size.) A short review by Siguier and collaborators () surveys different IS families and examples of IS evolution based on individual case studies. Yet another recent analysis focuses on archaeal ISs (). In contrast to these papers, our work focuses on the sequence divergence of coding regions and inverted repeats in the largest number of genomes analyzed to date. Earlier work on the molecular evolution of ISs dates to the pre-genome era, restricted itself to narrow categories of ISs, or relied on previously available (and heterogeneous) genome sequence annotation for a smaller number of IS families (,). Our current analysis overcomes these limitations by analyzing members of all 20 IS families in an unprecedented (>400) number of genomes. We find that the IS families we analyzed are well defined on the transposase sequence level. Specifically, although our approach would have admitted ISs with as little as 35% amino acid sequence identity to curated reference sequences, the mean amino acid divergence is lower than 7%, and more than 90% of all ISs have more than 70% amino acid identity to the reference. The different IS families show a skewed and patchy distribution, where most genomes carry no members of any given IS family, and a very small number of genomes carry many members [b, see also ()]. This distribution of IS occurrence resembles a similarly skewed distribution of IS occurrence on a much smaller taxonomic resolution, namely for 71 strains (), where many strains carry few or no IS copies. At least two explanations might account for this skewed distribution. One of them involves selection against genomes with high IS copy numbers (see also below), another is an unidentified transposition immunity mechanism that suppresses IS copy number. These causes are non-exclusive and might operate jointly. A strong or highly significant statistical association among IS families of IS copy numbers per genome might indicate that some bacteria are more susceptible to ‘infection’ by ISs in general. However, we do not find strong evidence of such an association for any pair of ISs beyond what would be expected from the shared evolutionary history of many bacterial species. Conversely, some ISs might be more successful than others, in that they can more easily ‘infect’ a larger number of genomes. Different ISs clearly show very different abundances. However, a thorough recent analysis suggests that among many possible factors determining IS abundance, genome size has by far the most important influence (). One biologically significant finding of our survey is the extreme sequence homogeneity of ISs within genomes. This substantially extends earlier, more limited work (,) and demonstrates a consistent pattern across IS families and vast taxonomic scales. This high sequence homogeneity stands in stark contrast to the greater sequence diversity among duplicate genes, another prominent class of repetitive DNA (). Our earlier work suggests that gene conversion is not a likely sole cause of this high homogeneity, because common signatures of gene conversion are absent within IS families (). Instead, this high homogeneity is readily explained by the rapid spreading of ISs within a genome. Consistent with this hypothesis is the observation that transposition and excision rates are very high on the time scale at which DNA sequences change. The high-sequence homogeneity of ISs might be explained by the following evolutionary scenario. After an IS enters a genome, its copy number expands rapidly through transposition (hence the low sequence diversity). Eventually, the IS becomes extinct again from the lineage, mostly due to natural selection, but perhaps aided by excision events. Some time thereafter, it may become reintroduced through horizontal gene transfer. Several other scenarios are not consistent with the data (): If ISs did not go periodically extinct, they would show higher divergence within a genome; if they were not reintroduced by horizontal transfer, bacterial genomes would be devoid of them; and if the net effect of natural selection was an increase in IS copy number, then ISs should be much more diverse within a genome, because they would remain part of the genome indefinitely. The only requirement for this evolutionary scenario is the frequent horizontal transfer of ISs. This is not a problematic requirement, as horizontal gene transfer may account for more than 10% of a bacterial genome's gene content (,). Its likely importance has been noted in an earlier, sequence-limited study on IS evolution in enteric bacteria (). In addition, the mere fact that highly similar ISs of the same family occur in multiple distantly related genomes speaks to the importance of horizontal gene transfer for IS maintenance. Among the methodological questions that a tool like IScan can address is whether rapid, computational, and automatic identification of functional or non-truncated ISs is possible. Aside from coding transposase-coding regions, ISs are typically associated with two sequence elements (direct target repeats and inverted repeats) that might be usable in such an identification. Direct target repeats, however, are of limited use for this purpose. High similarity of direct repeat units might indicate whether the transposition event that produced them occurred recently or a long time ago. However, this criterion only tells us whether the IS in question transposed or inserted successfully, not that it could again do so. In addition, target repeats for members of one IS family are very short, rendering their unambiguous identification difficult. Furthermore, their sequence similarity among different IS family members is highly limited (data not shown). The second class of potential diagnostic sequence features are inverted repeats, except for IS families ISCR and IS91 that do not harbor such repeats. Even though a truncated IS may be only slightly shorter than its intact counterparts, one would expect that truncation is often associated with deletion of one inverted repeat unit. Also, it is reasonable to expect that the inverted repeat units of a functional IS show greater sequence similarity to each other than DNA fragments of the same length but randomly sampled from the same genome. We implemented a statistical test in IScan that relies on a large number of such random fragments, to ask whether IS inverted repeats show such significant similarity. We applied this test ( = 10) to all the ISs we identified in the 438 genomes. When sequence similarity of inverted repeats is assessed through comparison with a reference IS (c), then 90.3% of ISs have inverted repeats with significant similarity at = 0.05. Of these, one would expect a fraction of 0.05 to be false positives, leading to an expected number of ISs with significantly similar inverted repeats of 0.95 × 90.3% = 85.8% ISs. The more stringent Bonferroni-corrected = 2.3 × 10 would yield 55.1% ISs with significantly similar inverted repeats. The inverted repeat -values are significantly greater for ISs with an identical member of the same IS family in the same genome, many of which would be functional. Taken together, these observations suggest that, on one hand, inverted repeats may be highly flexible and cannot always be unambiguously identified. On the other hand, identification of ISs with highly similar inverted repeats may allow enrichment of a dataset with functional ISs, which may facilitate subsequent analyses. The high sequence similarity of ISs within genomes thus has not only biological implications. It also aids in defining a heuristic criterion—perhaps the only one—to identify functional ISs based on sequence data alone. In closing, we note that the applications we illustrated here are only two among many uses that IScan might find. These uses will only increase as more genomes become available, and will include mapping of horizontal transfer histories, as well as transposition sequences within a genome.

In eukaryotic cells, genomic sequences are repetitively and compactly packaged into nucleosomes where they are in close association with histone proteins. In this setting, histone proteins and DNA form tightly knit complexes that sterically occlude DNA from interacting with other proteins (). The stability of the nucleosome controls nearly all DNA-templated biological process, and can be partly modulated using intrinsic regulation mechanisms that rely upon the ability of DNA sequences to create well-ordered, reproducible nucleosome structures (,). In other words, the nucleosome stability, and consequently protein access, is likely encoded in the DNA sequence itself through sequence-dependent DNA bending flexibility, which is inversely related to the energetic cost of forming the nucleosome particles. A variety of experimental tools have been developed to investigate DNA bending flexibility (), including comparative gel electrophoresis, crystallography, electron microscopy, transient electric birefringence and DNA cyclization, as well as more recent single molecule manipulation techniques (). DNA cyclization has distinguished itself from most other approaches by its simplicity, robust theoretical foundation, high sensitivity and accuracy, and the fact that it can be used to provide a simple indication of the local chain stiffness (). In a typical DNA cyclization experiment, DNA is designed with ligatable single-stranded ends. T4 DNA ligase is added to initiate the cyclization reaction. The ratio of equilibrium constants for ligatable unimoleuclar and bimolecular forms are measured under precise conditions to determine the so-called Jacobson-Stockmayer factor, . The experimentally obtained are then fitted to an analytical or numerical model, e.g. the continuous worm-like chain model (WLC) (), to obtain the bending flexibility of the particular sequence used. Recent cyclization data reported by Cloutier and Widom (,), have produced much excitement in the literature because they indicate that short (<100 bp) double stranded DNAs (dsDNA) repeats are dramatically more flexible than anticipated by any current model for DNA bending. Marko and co-workers performed a follow-up theoretical study to explain Cloutier and Widom's experimental results. These authors found that the unusually high cyclization efficiencies reported can be reproduced by a kinkable worm like chain (KWLC) model for the DNA analyte (), though the fundamental origin of kinks in the analytes used in the experiments remains unknown. An even more recent study by Vologodskii and co-workers () has challenged the enhanced DNA flexibility reported by Cloutier and Widom's experiments on different grounds. These authors contend that the exceptionally high cyclization efficiency reported is correlated with the higher than normal T4 ligase concentration used in the measurements. Yuan and co-workers () reported a Fluorescence Resonance Energy Transfer (FRET)-based technique to measure the probability distribution of short dsDNA fragments with and without base pair mismatches. These authors observed that the presence of local unpaired bases in the dsDNA fragment leads to dramatic increases in the probability of highly bent, kinked states implying that local base unpairing can produce the kinked states postulated by Marko and co-workers. Here we utilize fluorescence measurements to test a key assumption about T4 DNA ligase used in cyclization measurements. Specifically, in cyclization measurements T4 ligase is used to perform a single task. Namely, to efficiently trap low-probability, highly bent DNA conformations in which the 3′ hydroxyl end of one dsDNA molecule approaches the 5′ phosphorylated end of the same molecule. The reports by Vologodskii . and by Yuan et al. motivate a perhaps obvious question: Does T4 ligase always restrain itself to the assigned task? To answer this question, DNA molecules containing base analog, which can be used as indicators of the base pair stacking status, are used to evaluate the impact of ligase on DNA conformation under the conditions used for cyclization studies reported in references (,,). DNA oligos with 2-AP inserted at designated positions were custom synthesized by Integrated DNA technologies (Coraville, IA). All oligos used in the study were purified by the manufacturer using high performance liquid chromatography (HPLC); oligo sequences are detailed in the Supplementary Data section, 1a. Unless otherwise specified, the 5′ and 3′ ends of the oligos are comprised of hydroxyl groups. dsDNA were produced by the standard annealing procedure (). For samples containing 2-AP, the ratio of the oligo containing 2-AP to the complementary strand is maintained at 3:4 to minimize the relative amount of 2-AP in the non-helical states. The composition of dsDNA is summarized in the Supplementary Data, Section 1b. All dsDNA fragments contain mutually complementary 4 nt overhangs at both ends. Fluorescence excitation and emission spectra were collected using a fluorophotometer (PTI inc) with a 76 mW source power, 5 nm slit width, and integration time of 0.5 s. Quartz fluorometer cells with 3 mm path length (Starna Cells Inc., CA) were used for all measurements. All spectra reported were corrected in real-time for lamp fluctuations. The fluorescence measurement buffer contains 50 mM Tris-HCl, 10 mM MgCl, 10 mM DTT and 25 μg/ml BSA. The buffer conditions used here are similar to those employed in the standard cyclization assay, except for the absence of ATP (,). To evaluate the effect of T4 DNA ligase on 2-AP fluorescence, solutions containing DNA and T4 DNA ligase (New England Biolabs, MA) at various concentrations were made-up and mixed using a gentle vortexing procedure; solutions were stabilized at room temperature for at least 30 min prior to the measurements. The concentration of T4 DNA ligase is provided by the manufacturer in New England Biolab (NEB) units. The equivalent molar concentration can be quantified using Coomassie Plus Assay (Pierce Biotech, IL). Based on this procedure, the NEB units can be converted to standard molar concentrations using the relation, 1 unit/ml = 0.02 nM (±25%). It should be noted that the original NEB ligase contains 200 μg/ml BSA, the measured ligase concentration must therefore be corrected by subtracting the BSA contribution. Characteristic diffusion times of DNA molecules in the presence of T4 ligase at various concentrations were determined using a home-built fluorescence correlation spectroscopy (FCS) apparatus (). In a typical FCS measurement, fluctuations in the fluorescence intensity are quantified by temporally autocorrelating the intensity signal. Samples containing a fixed concentration of DNA (25 nM) end-labeled with fluorescein were excited with a 488 nm laser, such that only fluorescently labeled DNA molecules or their complexes are visible under the microscope. The recorded autocorrelation function (τ) was analyzed using well-known procedures (,). If the sample contains only one fluorescent diffusive species, (τ) can be written as: The value of ω was characterized using Alexa 488 dye before the measurements, and is taken to be a constant for the subsequent analysis. When the ligase and DNA coexists in solution, DNA-ligase complexes contribute to (τ) as a second diffusive species. The characteristic diffusion time for DNA (τ) can be easily determined from FCS control experiments in which no ligase is added. In the presence of ligase, the autocorrelation function is analyzed with a ‘two-diffusive-species’ model (), in which the diffusion time of DNA (τ) is fixed at the value obtained from the control experiment. The nucleotide 2-AP is a fluorescent analog of guanosine and adenosine and has been used as a site-specific probe of nucleic acid structure and dynamics (,). Fluorescence emission from 2-AP can be readily excited at a wavelength of 320 nm, far away from any DNA absorption, or absorption due to DNA-protein complexes. This feature allows selective excitation of fluorescence from the 2-AP base even in the presence of protein residues. Significantly, the fluorescence emission from a 2-AP nucleotide is highly quenched in the stacked state, and its quantum yield increases approximately 10-fold in the un-stacked state (,). Local conformational changes, especially stacking status changes, can therefore be assessed from the fluorescence of 2-AP, i.e. the 2-AP nucleotide manifests enhanced fluorescence intensity when assuming an extra helical conformation (see inset of ) (). By selectively exciting the sample at 320 nm, the fluorescence intensity increase can be used to report the population of spontaneously ‘flipped-out’ bases. To demonstrate the sensitivity of these measurements, dsDNA containing 2-AP inserted in different sequence contexts were synthesized. In the E89-23 sample (see Supplementary Data 1a), 2-AP is inserted in an AT rich region, while for E89-16 2-AP is located within a random genomic sequence context. The fluorescence emission spectra collected when the two DNA samples, at a concentration of 1.0 μM DNA in buffer, are excited at 320 nm are reported in . The background emission was independently measured using the neat buffer (i.e. in the absence of DNA), and subtracted from the sample spectra. It is apparent from the figure that dsDNA containing 2-AP in the AT rich region (E89-23) exhibits substantially higher fluorescence than the sample where 2-AP is inserted in the random sequence context (E89-16). Increasing the complementary oligo concentration, which does not contain the 2-AP insertion, has virtually no effect on the fluorescence intensity, indicating that the global thermal stability of the as formed dsDNA does not contribute to the observed fluorescence increase. then simultaneously shows that even in the absence of ligase some 2-AP bases are quite exposed and the effect is dramatically larger for ‘softer’ A-T rich sequence domains. The difference of 2-AP fluorescence within different sequence contexts can originate from basic differences in the energy transfer mechanism between stacked bases [E89-23: (AApT); E89-16: (GApT)], (), as well as from differences in the rate of spontaneous base un-pairings, the so-called base flip-out rate. The relatively low base pairing and stacking energies within the AT rich region [the stacking energy difference between E89-23 and E89-16 is estimated to be around 0.2 ∼ 0.3 kcal/mol for regular AA and GA dinucleotides, ()], have been argued to make it easier for a single nucleotide to bypass the energetic barrier, and spontaneously un-pair with its complement (). It has been demonstrated through extensive studies, that the fluorescence of 2-AP can be used as a sensitive reporter of the enhanced base pair un-pairing that would result from such events (). Local base pair flip-out or local defects in short dsDNA fragments can contribute significantly to the measured -factors deduced from cyclization measurements. Here we wish to determine what influence, if any, other species used in these measurements might have on such defects, and thereby the apparent bending stiffness reported by the measurements. The ATP dependent ligation reaction involves three principal steps: (i) Activation of the enzyme through formation of a covalent protein-AMP intermediate, accompanied by release of PPi. (ii) Transfer of the nucleotide to a phosphorylated 5′-end of the nick or the sticky end to produce an inverted (5′)-(5′) phyrophosphate bridge structure. (iii) Catalysis of the transesterification reaction resulting in the joining of the nick and release of free AMP (,). Three components are essential for the successful and timely completion of a ligation reaction, namely, the 5′ phosphorylated end and the presence of ATP and ligase. By controlling the relative abundance of these three elements, the effect of ligase on the DNA conformation at different stages of the reaction can be clarified. For these experiments, the same two 89 bp dsDNA, containing 2-AP as in the previous experiments is used. The 5′ ends of the dsDNA are unphosphorylated to prevent ligation, so the molecular size remains unchanged during the measurement. Fluorescence measurements performed using the buffer containing an equal concentration of ligase are used to quantify any background effects, which are subtracted from the raw spectra to isolate the effect of T4 ligase concentration on DNA fluorescence. a illustrates the effect of increasing the amount of T4 DNA ligase on the fluorescence emission intensity of E89-23. DNA concentration is maintained constant, while the ligase concentration is gradually increased. Background effects are subtracted from all spectra using the procedure outlined in the last section. The spectra shown are averages from at least two runs. It is apparent from the data that as the ligase concentration is increased the fluorescence of 2-AP manifests a corresponding increase. To confirm our findings, excitation spectra are independently measured on the same DNA samples containing varying concentration of ligase. The DNA solution without ligase is treated as background during these measurements, and the final spectra are obtained by subtracting the respective background spectra from the measured sample excitation spectra. The excitation spectra collected at a fixed emission wavelength of 370 nm, should therefore solely reflect any changes to 2-AP fluorescence induced by ligase. b summarizes the main results. It is apparent from this figure that with increased ligase concentration, the protein excitation peak at 280 nm increases, as expected. However, the secondary excitation peak centered at around 310 nm, which originates from direct excitation of 2-AP also changes from insignificant to significant with increasing ligase concentration. This effect has been confirmed using a larger DNA fragment, 94 bp (E94-25), indicating that the observation is quite general. Control experiments using 2-AP inserts in single-stranded DNA indicate that T4 ligase does not induce any changes in 2-AP fluorescence in single stranded DNA. This last observation is significant because it eliminates the trivial possibility that the presence of ligase might change some bulk property of the medium (e.g. its dielectric constant) and thereby alter the quantum yield of 2-AP. These observations can be interpreted either in terms of a higher base pair flip-out rate induced by the ligase or in terms of an increase in the population of dsDNA containing extra helical 2-AP nucleotides induced by the presence of DNA ligase. Either mechanism can in principle provide a physical source for the ‘kinked’ states postulated by Marko and co-workers (), and as such are tempting candidates to explain cyclization data revealing enhanced dsDNA flexibility. It must be kept in mind, however, that the DNA concentration employed in our measurements is 100–1000-fold larger than those normally used in cyclization experiments, and that the ligase concentration is substantially higher than the 100 units/ml value recommended by Du . as a standard example for studies of very short DNA. Implications of both effects (higher DNA concentration and higher than recommended ligase concentration) will be addressed in detail in subsequent sections of the article. The relative enhancement in base pair flip-out, or equivalently the increase in the population of DNA containing local defects, can be directly related to the increase in the fluorescence emission intensity (). Since the quantum yield of 2-AP in its non-stacking status is almost 10 times higher than in the stacked status, the relative amount of DNA with ligase induced 2-AP flip-out, i.e. base pair flip-out, can be estimated to be around one-tenth of the fluorescence emission intensity increase, i.e. the fractional increase referenced to the zero-ligase control. The observed ligase-induced fluorescence enhancement evidently does not exhibit a significant length dependence, which implies that it originates from the general destabilization of bases along the DNA backbone. From the slope of the fitted line in , it is also possible to estimate that the enzyme binds to the 2-AP site or 2-AP containing region with ≈ 50 μM. The apparent for free state ATP, which is structurally very similar to the 2-AP nucleotide, in forming the non-convalent ligase-ATP complex is reported to be below 0.15 μM. The nearly three orders difference in means that the inserted 2-AP nucleotide is well incorporated inside the helix. Results from other experiments using a dsDNA sample in which the 5′ end of the dsDNA is phosphorylated (E89-23p), but in the absence of ATP to prevent further ligation are presented in . The results for E89-23p show similar increases in fluorescence intensity from dsDNA (E89-23p) as the ligase concentration is gradually increased. A noteworthy feature of all of these experiments is that as [Ligase] increases, the enhancement in 2-AP fluorescence seen in the presence of ligase becomes larger. It is well known that to initiate the ligation reaction, ATP must be present in the reaction buffer as an essential cofactor. Specifically, ATP is known to be essential for triggering the first step of the ligation reaction by forming a transient AMP-ligase complex that can bind non-specifically to the DNA target. Also, ATP has been documented as a fluorescence quencher of 2-AP. However, for the low ATP concentrations used in this study, the characteristic emission of the 2-AP is anticipated to be nearly unaffected in the presence of ATP (). Fluorescence measurements performed by simply adding ATP to dsDNA with zero ligase concentration, in fact reveal no measurable effect of ATP on the characteristic fluorescence excitation or emission spectra. Addition of ATP to dsDNA solutions in buffers containing T4 ligase produces a very different effect. This effect is illustrated in where the effect of ATP concentration on ligase-enhanced 2-AP fluorescence is shown. To avoid the potential complication of molecular size change due to the ligation, E89-23 which has hydroxyl groups on both 5′ and 3′ ends are used for these measurements. The fluorescence background is obtained using buffer solutions containing the same concentration of DNA and ATP and subtracted from the sample spectra. It is apparent from that the addition of ATP to E89-23 solutions containing ligase gradually reverses the effect of the ligase on 2-AP fluorescence. The crystal structure of DNA ligases have been studied by several groups (). These studies reveal highly modular structures, all with preserved nucleotide-binding domains or an oligomer-binding (OB) fold. This DNA binding domain is generally positively charged and is thus capable of stabilizing the dsDNA molecules containing transiently unpaired bases by binding non-specifically to the DNA. The ATP binding domain is generally postulated to be close to the DNA binding domain (). The diminishing ‘ligase-enhanced’ fluorescence observed after ATP addition suggests that ATP competes with DNA for binding to the ligase. Two plausible mechanisms could explain the observed phenomena. One is that after ATP binds to the ligase, it triggers a conformational change of the protein and thus precludes the ligase from acting as a stabilizing agent for spontaneously unpaired bases in the dsDNA fragments. The other is that ATP binding enhanced the specific binding of ligase towards its end and thus makes the internal binding on DNA backbone less favorable. FCS measurements were performed at 25°C using a fluorescently labeled 89 bp non-phosphorylated DNA, without 2-AP insertion. The inset in illustrates the effect of ligase concentration on the characteristic diffusion time deduced from the FCS data by fitting the fluorescence correlation function for a single diffusing species. It is evident from the figure that the average diffusion time of the DNA is, significantly increased by the addition of ligase. Furthermore, as the concentration of ligase is increased, the diffusion time at first increases quite strongly but then manifests a weaker but steady increase with ligase concentration, over nearly four decades of concentration. The apparent ligase-induced slowing down of DNA diffusion is consistent, at least qualitatively, with formation of the DNA-ligase complexes hypothesized on the basis of the 2-AP fluorescence data. We emphasize here that the DNA specimen used for FCS measurements do not contain 2-AP, demonstrating that the insertion of 2-AP in the previous fluorescence experiments does not create preferential binding site for the ligase. Also, essentially identical results are obtained in similar experiments using 89 bp DNA sample without 2-AP insertion or 4 nt dangling ends, which indicates that T4 DNA ligase does not preferentially bind to the dangling end of the DNA. More careful scrutiny of the fluorescence correlation function indicates that addition of ligase does not uniformly slow down diffusion of the DNA. Specifically, by fitting the correlation function data to a model for two diffusing species, the ligase is found instead to produce a second slow-diffusing DNA (only DNA is fluorescently labeled) component with a slower characteristic time τ, that increases as the ligase concentration is increased. Significantly, even in buffers with high ligase concentration, the two species model indicates that there is always a significant population of DNA that remains effectively unaffected by the ligase (i.e. τ ≈ τ, irrespective of [Ligase]). Considering the low concentration of DNA used for the measurements the relative abundance of the slower, presumably ligase-complexed, species can be estimated as the relative ratio of slow diffusion () species. shows the effect of ligase concentration on calculated in this manner. , the dissociation constant of the DNA-ligase complex. At the lowest ligase concentration (i.e. ≈ 1.5 μM, while at higher ligase concentrations we obtain ≈ 15 μM. These values are, respectively, 30-times and 3-times smaller than the value calculated previously using the 2-AP fluorescence data. Taken at face value, they suggest that the DNA-ligase complexes are most stable at the lowest ligase concentrations. These conclusions obviously overlook the complicating influence of multiple ligase molecules binding to a single DNA fragment. Specifically, as calculated previously, reflects the dissociation constant when ligase is bound ‘specifically’ to the 2-AP site or the 2-AP incorporated regions. To take into account the possibility of multiple binding sites on a single substrate, relative abundance data are conventionally analyzed using the Hill Function, = [Ligase]/( + [Ligase]).The solid line in , is obtained by fitting this functional form to the experimental data. The fit is evidently not as good as the dual straight line fit, particularly at low [Ligase]. We nonetheless proceed to extract values of ≈ 0.3 ± 0.1 and ≈ (5 ± 16) × 10 from the Hill function fit to the data. A Hill coefficient, , less than unity implies that the binding event is negatively cooperative, while the value of has less physical meaning both due to its large uncertainty and the lack of a direct correlation with the binding constant for the reaction. Based on both analyses, we therefore conclude that ligase has a tendency to bind non-specifically to the DNA and to form a transiently stable complex with a dissociation constant estimated to be 2 – 25 μM, depending on the ligase concentration. It must be noted that the FCS experiments are carried out under conditions with higher ligase concentration as well as with a reaction buffer without added ATP as compared with typical cyclization reaction. The higher ligase concentration is essential for resolving the slow diffusive species, and at lowest ligase concentration being explored, less than 1% of DNA forms complex by binding to the ligase. The effect of ATP is also examined in the FCS measurement. With 1 mM ATP present in the reaction buffer, the binding of the ligase to the DNA molecules is effectively suppressed, exhibiting constant diffusion times independent of the added ligase concentration as indicated in the inset of . The findings reported in the last section suggest that by stabilizing transiently unpaired bases, T4 ligase can dramatically lower the bending stiffness of DNA. This effect is analogous to that produced local defects, e.g. base pair mismatches, gaps and bulges, on the bending stiffness of DNA (,). However, we have already reported that ATP can suppress such non-specific binding of ligase to the DNA backbone, suggesting that the effect can be reversed by employing a sufficiently large ATP concentration in cyclization measurements. To further evaluate this possibility, we now employ the thermodynamic constants obtained in the last section to determine the effect of ATP on the population of DNA containing defects. We begin by setting the ligase concentration to be around 100–400 units/ml (2–8 nM), as commonly used in the cyclization assay. At zero ATP concentration, 10–10 ( ) DNA contains local defects induced and/or stabilized by ligase binding. As the ATP concentration increases, the relative percentage decreases. For example, with 1 mM ATP, as in a typical ligation buffer [ = 0.15 μM, ()], only 10–10 DNA contains the induced defects. This is a very small population of the DNA molecules, and its effect generally negligible when the length of DNA is much larger than its persistence length. However, as the length of DNA decreases, the population of cyclizable species reduces dramatically and finally approaches and becomes even smaller than this value. For example for the 89 bp DNA used in reference () the -factor predicted by the WLC model, with a persistence length of 50 nm, is 10 M, which is equivalent to 10–10 of the total DNA molecules. Since DNA molecules containing local defects stabilized by ligase are expected to have a much higher tendency to be cyclized, the contribution to the -factor measured in the cyclization assay is clearly very important even at the ATP concentrations currently employed. To provide accurate measurements of DNA stiffness using cyclization, our results in fact indicate that for a fixed concentration of ligase, the relative ratio of ATP and DNA must be carefully tuned to prevent non-specific binding of ligase to DNA. However, since ATP is generally consumed during the ligation reaction, reaction time and substrate concentration should also be taken into consideration while selecting the optimal ATP concentration. To illustrate, consider the cyclization measurements conducted on 105 bp DNA by Du and co-workers (). These authors managed to recover the cyclization efficiency predicted by WLC model for 106 bp DNA, using 25 units/ml ligase, 0.025 nM DNA and 1 mM ATP (). Under these conditions, the relative abundance of DNA-ligase complexes is calculated to be 10–10, compared with ∼10 cyclizable DNA relative to all DNA molecules present. It is then possible to estimate that either an ATP concentration of over 10 M or ligase concentration <10 units/ml is needed to fully suppress the contribution of ligase induced DNA destabilization to the measured -factor. Our observations therefore appear to provide a plausible answer to the longstanding controversy as to the variability of bending stiffness values for short DNA fragments reported by various groups (,,). Specifically, our results indicate that the presence of large amounts of ligase in cyclization experiments might well create rather than simply trap highly bent DNA conformations; implying that the cyclization assay may be inherently not suitable for extracting bending stiffness of very short DNA fragments. Using the base analog 2-AP, whose fluorescence is sensitive to its microenvironment, the local conformational change of DNA molecule under standard conditions of the cyclization assay is tested. Existence of T4 DNA ligase enhanced 2-AP fluorescence indicates that ligase is capable of inducing DNA conformation change. The non-specific interaction between DNA and ligase increases the subpopulation of DNA molecules containing local defects, i.e. extra-helical nucleotides, which can lower the apparent stiffness of DNA determined from cyclization data. The presence of ATP can reduce this ‘ligase enhanced’ fluorescence by competitively binding to the functional domain of ligase and thus suppress the DNA conformational change due to the non-specific DNA ligase contact. In performing a meaningful cyclization assay, the ligase and ATP concentration must be selected with extra caution. Besides satisfying the kinetic assumption underlying the analysis (), the ligase induced DNA conformational change must be suppressed efficiently. Performing the reaction in an ATP rich medium seems to be a promising resolution to this problem. But as DNA size reduces, sufficiently high ligase concentration is essential for efficient completion of ligation reaction and the preference of DNA with local defects in the cyclization reaction can shift the binding competition between ATP and DNA towards the DNA side. All of these factors might contribute to the complexity in analyzing the cyclization data of short DNA strands and a measurement of DNA bending stiffness without external agent could be required in resolving this issue. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

The bacteriophage P1 Cre/ site-specific recombination system is widely used as a genetic engineering tool (,) due to its well-defined recognition sequence, lack of any necessary co-factors and efficacy in both bacterial and eukaryotic systems. Since its discovery, Cre/ has been applied to temporal and spatial gene activation/deactivation (), site-specific genomic integration and deletion () as well as the construction of libraries () and cloning strategies (). Interest in expanding and improving the types of reactions that the Cre recombinase can facilitate has led to the discovery of numerous functional recognition sequences for Cre; however, no sequence characteristics have been defined to evaluate what constitutes a functional site other than its need to bind the recombinase and the role spacer compatibility plays in efficient recombination events (,,). The Cre recombinase acts on a 34-bp sequence known as (,). This recognition sequence consists of two 13-bp palindromic arms separated by an 8 bp spacer region (,). In a site-specific recombination reaction two sites are brought together, each arm binding one recombinase monomer (), while strand exchange takes place within the spacer regions (). sequences are maintained following recombination events. This characteristic is responsible for the reversible nature Cre/ reactions. Successful site-specific recombination has been shown to occur between sites with sequences having varying degrees of similarity to (,,,). Beginning with the description of the site (,), several endogenous genomic sequences from different organisms have been discovered that can serve as substrates for Cre-mediated site-specific recombination (,,). Sauer () first identified endogenous cryptic sites in These cryptic sites contained as few as 14 out of 34 bases in common with , and illustrated the apparent importance of the TATA sequence adjacent to the spacer region in functional sites. Later, other functional genomic sites were reported for yeast, human and mouse (,). Together, the description of these functional sites demonstrates flexibility in sequence recognition by the Cre recombinase. The search for endogenous sequences within genomes for use in genomic engineering and gene therapy has been a driving force in the discovery of alternative functional sites. However, diversification and manipulation of site behavior has also identified many alternative substrates for Cre. Albert () were the first to screen for arm mutations that could facilitate stable Cre/ integration reactions; while mutation studies of the spacer (,) documented the effects of single and double base mutations on recombination efficiency. Recently, two different surveys of partially randomized spacer libraries (NNNTANNN) (,) added to the growing list of spacer sequences that are proficient in recombination. Previous studies have expanded the list of sites with which the Cre recombinase is able to recombine without pursuing a full-scale randomized approach to the 34bp sequence. Here, we report on the first randomized sequence studies aimed at defining functional sequences. We have created a randomized spacer library and a randomized arm library, both of which have been tested for functionality in Cre-expressing bacteria and in reactions. The results of these studies indicate that Cre is very flexible in the sequences it can recombine. We show that sites with matching spacers recombine more effectively than sites with non-matching spacers and that the central ‘TA’ is not required for efficient recombination. Our results also indicate that there is a sequence bias in functional arms and that the mode of recombination ( or ) is important when evaluating site function. The spacer library plasmid backbone (pLK) was engineered from the previously described pPG3- plasmid () modified to replace its site with a HindIII–NotI–BglII linker at its XbaI site. Two separate syntheses of the library template (5′ GCGCGAATTCTGCGCATAACTTCGTATANNNNNNNN TATACGAAGTTATAGATAGGACGGATCCAGTTGG 3′) were done in order to compensate for non-random base frequencies in the oligonucleotide manufacturing process. PCR was used to create double-stranded DNA from the templates using primers EcoRI 5′ CCAACTGGATCCGTCCTATCT 3′ and BamHI 5′ GCGCGAATTCTGCGCATAACT 3′. Each template PCR product pool was separately cloned into pLK as an EcoRI/BamHI fragment (B.1). Clones were propagated in strain DH5α under ampicillin selection. Inserts were confirmed by loss of the SacI restriction site. Library clones were chosen at random to go through reiteration of their site. PCR with primers BamHI KAN 5′ GATTTTGAGACACAACGTGGCT 3′ and Hind3 KAN 5′ GCGCGCAAGCTTTTGCCATTCTCACCGGA 3′ was used to amplify the region of the pLK + library plasmid that contained the spacer library sequence and the 3′ end of the kan gene. PCR products were ligated as BamHI/HindIII fragments (613 bp) into their parent clones (B.1). pLK + library + Kan/ clones (B.2) were then propagated in DH5α under kanamycin selection. The pLK + control plasmid was constructed using the same cloning strategy starting with the template primer 5′ GCGCGAATTCTGCGCATAACTTCGT ATAGCATACATTATACGAAGTTATAGATAGGACGGATCCAGTTGG 3′. xref fig #text pLK + library + Kan/ clone DNA was isolated by QIAprep spin miniprep column (Qiagen) and eluted in EB buffer (10 mM Tris–HCl, pH 8.5). Cre reactions were performed with 75 ng of plasmid DNA, 1U of MBP-Cre extract and 3 μl of 10 × Cre buffer (500 mM Tris–HCl, pH 7.5; 330 mM NaCl and 110 mM MgCl) in a 30 µl total volume. Reactions were incubated at 37°C for 15 min and then transferred for storage at −20°C. Recombination was detected through high-speed PCR utilizing iProof High-Fidelity DNA Polymerase (BioRad). A standard 25 μl PCR mix was used containing 2 ng of template DNA taken directly from Cre reactions. The primers used were pLK SYBR R 5′ GAGATAGGGTTGAGTGTTGTTCC 3′ and pLK SYBR L 5′ GACCT ACACCGAACTGAGATACC 3′. Cycling conditions included an initial denaturation at 98°C for 30 s, 17 cycles of 98°C for 10 s, 67°C for 10 s and 72°C for 2 min followed by a final extension of 72°C for 4 min. PCR samples were pre-incubated with SYBR Green I nucleic acid stain and visualized by agarose gel electrophoresis. Gels were scanned via a Storm PhosphorImager system (Molecular Dynamics) and analyzed with ImageQuant 5.1 software. Inverse PCR was used to remove a KpnI site at 623 bp from the pSV-β-Galactosidase Control Vector (Promega) (primers: KpnI invers1 5′ GTACCGGTGGGTGAAGACCAG 3′ and KpnI invers2 5′GAGACCGCCACGGCTTACGGC 3′) (A.1). A linker (annealed primers NBKH5′ 5′ CATGGTTCGAACTGGTACCA 3′ and NBKH3′ 5′ AGCTTGGTACCAGTTCGAAC 3′) was inserted after digesting pSV-β-Gal with NcoI and HindIII. pSV-β-Gal(–KpnI + linker) was modified at its BamHI/PstI sites with a BamHI––NotI–NsiI fragment that was later removed by BamHI/NotI digest and replaced with annealed primers BamXbaNot5 5′ GATCCATAACTTCGTATAATGTATGCTATACGAAGTTATTCTAGAGC 3′ and BamXbaNot3 5′ GGCCGCTCTAGAATAACTTCGTATAGCATACATTATACGAAGTTATG 3′ to form the site. PCR with primers KpnI3X 5′ CAGGTACCATAACTTCGTATAGC 3′ and BstBI 5′ CCCTGTCCT TCGAACTCGAG 3′ was used to create double-stranded DNA from the arm library template oligonucleotide 5′ CAGGTACCATAACTTCGTATAGCATAC ATNNNNNNNNNNNNNCTCGAGTTCGAAGGACAGGG 3′. This PCR product was ligated into pSV-β-Gal(−KpnI + linker) + as a BstBI/KpnI fragment to create pSV-β-Gal + + arm library (A.1) and then transformed either into 294-CRE or DH5α . The control site was added as annealed oligos BstbXhoKpn3 5′CATAACTTCGTATAGCATACATTATACGAAGTTATCTCGAGTT 3′ and BstbXhoLoxPKpn5 5′CGAACTCGAGATAACTTCGTATAATGTATGCTATACGAAGTTATGGTAC 3′ to create the control plasmid pSV-β-Gal + . #text italic xref sub #text All spacer library clones were sequenced with primers PLK 5′ TAAATGAGCATCCATGTTGG 3′ and M13F-pUC(−40) 5′ GTTTTCCCAGTCACGAC 3′. All arm library clones and reconstructed arm library clones were sequenced with the pSVβ 5′ CGACTGGAAAGCGGCAGTG 3′ primer. The arm library site was sequenced with the pQEPromotor 5′ CCCGAAAAGTGCCACCTG 3′ primer. The pMAL-Cre expression plasmid was sequenced with primers pTYB11 #5 5′ GGTCGAAATCAGTGCGTTCG 3′, pTYB11 #4 5′ CGAGTTGATAGCTGGCTGGT 3′ and pTYB11 #3 5′ CGAACGCACTGATTTCGACC 3′. Sequence pools from both the spacer and arm libraries were subject to analysis: where is the observed nucleotide value and is the expected nucleotide value. Active Cre recombinase was purified through a maltose-binding protein (MBP) tag previously described by Kolb and Siddel (). The gene was PCR amplified from the pMC-Cre plasmid () with primers Cre ATG 5′ ATGTCCAATTTACTGACCGTACACC 3′ and Cre2 5′ GGTGGTCTCGAGCTAATCGCCATCTTCCAGCAGGCG 3′, digested with XhoI and cloned in frame into the pMAL –c2x vector (New England Biolabs) digested with SalI. pMAL-Cre was then transformed into K12 ER2508 (New England Biolabs) for expression. For our purposes, 1 U of purified recombinase is defined as the amount of enzyme necessary for maximum recombination after a 15 min incubation at 37°C of either 75 ng of spacer library control (pLK + ) or 0.1 μg of the arm library control plasmid (pSV-β-Gal + ) in a 30 μl total reaction volume. What are the sequence requirements for functional spacers? In order to investigate what constitutes a functional spacer sequence, we created two randomized spacer libraries from template oligonucleotides consisting of two arms separated by an 8 (N) randomized spacer region (A). Since it was already known that the most efficient recombination events usually take place between sites with matching spacer sequences (), we reiterated the library sites in order to test the function between matching spacer pairs (B). Fifty-five clones (40 from group one and 15 from group two) were randomly chosen for reiteration. Following reiteration, clones were sequenced at each site and then passaged through 294-CRE cells, which constitutively express Cre recombinase. Successful site-specific recombination events resulted in the deletion of the kanamycin resistance gene (B). Restriction digests with XmnI linearized both parental (5026 bp) and recombined plasmids (3707 bp). Clones exhibited either complete recombination, a mixture of parental and recombined plasmids, or no recombination product at all ( and ). In addition, two of the clones (#41 & #205) had an unknown product along with their recombination product (data not shown). Sequencing revealed that not all clones contained sites with matching spacers (). A subset of unmatched spacers was cloned for comparison to the matched spacer pairs. The remaining unmatched pairs were the result of the cloning process and were also included for comparison. All clones with full-length matching spacers (33/33) recombined to completion with no parental plasmid detected. Many clones with unmatched spacers also recombined to completion (12/21) or had some visible recombination product (6/21). Clone #23 contained shortened 7 bp matching spacers and no observable recombination product. Overall, this assay for recombination appears to differentiate, in some cases, between matched and unmatched spacer pairs. It does not provide information on the efficiency of recombination between the various spacer pairs, which recombined to completion . χ analysis of all unique library sequences indicated that there were positions in both template libraries that were not randomly represented in terms of nucleotide frequency. Instead of each nucleotide having a frequency of 25% at each of the 8 positions of the spacer, positions 2, 3, 5, 7 and 8 in the first library all had significantly unequal base distribution (Supplementary Figure 1). The second library contained only one position (), which had statistically unequal nucleotide distribution. This means that these libraries do not represent statistically random sequence populations. The non-random base distributions could be a result of the oligonucleotide manufacturing process and/or the manipulations (PCR, cloning and propagation in ) necessary to clone the library oligonucleotide (). The analysis of the spacer library did not demonstrate any differences in recombination based on nucleotide composition at specific positions within the spacer. Therefore, we assayed for recombination in an attempt to better distinguish functional diversity due to variation in spacer sequences. All spacer library clones were tested for function using purified Cre recombinase. Standard reactions were allowed to proceed for 15 min at 37°C. Recombination was detected by PCR followed by SYBR Green I staining of the PCR product and quantitation of band intensities on an agarose gel. contains the averaged results of three separate recombination experiments. The amount of Cre used in these reactions was titrated to give maximum recombination for the control (pLK + ). Due to the reversible nature of the Cre/ system, there is less than 100% recombination for an deletion reaction irrespective of reaction time (,). The 70.4% maximal recombination achieved here mirrors previously published data for Cre reactions (,). Compared to the recombination results in the 294-CRE bacterial system, the system appears to be more stringent in terms of the types of recombination events allowed and the amount of recombination product produced. For example, whereas the range of average recombination for the set of matching spacer clones was 56.7–1.7%, these clones were indistinguishable from one another . Non-matching spacer clones that had functioned did not necessarily function well ; 7 of 12 clones that showed full recombination had <3.0% recombination under our conditions. Therefore, not all clones that appeared to be as functional as the wild-type control were equivalent when compared The control averaged 70.4% recombination while the best library clone averaged 56.7%. Of the clones with full-length matching spacers (), the group averaged 30.5% recombination, while clones with non-matching spacers averaged 4.6%. These results suggest that Cre reactions are much less permissive in the types of spacer pairs that can undergo successful recombination compared to 294-CRE reactions. Matching spacer pairs are favored in recombination events, and while it is difficult to see any difference in recombination across sequences, the environment clearly discriminates between spacer sequences. Further analysis of these sequences is hampered by the small number of clones sampled. Our second basic question addressed what arm sequences the Cre recombinase can use to facilitate site-specific recombination events. Cre is known to bind cooperatively to its recognition sequence (,), and its contact positions within the arms have been mapped (,). By randomizing only one arm out of the four that make-up the Cre/ recombination complex, we could study what minimal functional sequences Cre can co-operatively bind and recombine. In order to easily screen large numbers of randomized arm library sites, we constructed a plasmid in which color selection could be used to visualize recombination events in a bacterial system. A template oligonucleotide containing one arm plus the spacer followed by a 13 N randomized region was used to create the arm library (B). Once the template was made double stranded through PCR, it was cloned into the β-galactosidase reporter construct pSV-β-Gal + . This reporter construct had been modified with a site at the 3′ end of the Z coding region. With the arm library site in place at the 5′ end, the Z gene was flanked by sites (A.1). Following transformation into 294-CRE cells, any site-specific recombination would result in removal of Z and loss of β-galactosidase activity (A.2). Hundreds of colonies were screened by blue/white color selection via seeding of transformations on LB/X-gal agar plates. After overnight growth at 37°C, ∼87.9% of colonies were stained dark blue, 3.5% were white with blue centers (white w/blue) while 8.6% were white only (data not shown). The pSV-β-Gal + control in 294-CRE resulted in white colonies only, while blue-stained colonies were observed for both control and arm library constructs in Cre− DH5α It was assumed that blue colonies would contain only unrecombined, non-functional arm library sites, while white colonies would be the result of successful site-specific recombination between functional sites. White colonies with blue centers were unexpected, but could be explained by a slower rate of recombination, relative to the phenotypically white colonies, for the cells that seeded these colonies. Upon examination of plasmid DNA from each type of colony (), it appeared that all colonies contained DNA that had undergone successful recombination events. Preliminary analysis showed that all plasmid DNAs linearized with XhoI were the size of the deletion product. However, full-length parental product could be visualized for blue colonies after overloading the restriction samples on an agarose gel (data not shown). Retransforming blue colony DNA into DH5α (−Cre) cells resulted in a 1:75 ratio of blue to white colored colonies, pointing to a small population of unrecombined plasmid DNA present after ∼48 h of growth in 294-CRE. In total, 93 clones were sequenced (). Forty-one sequences were from white colonies, 37 were from blue and 15 were from white colonies with blue centers. A random sequence population was derived from the ratios of blue:white:white w/blue colonies (∼37:4:1) observed upon plating of the library (Supplementary Figure 2). χ analysis of the random library sequence population revealed again that the base distributions at several positions of the library template oligonucleotide were not statistically 25% for each nucleotide (Supplementary Figure 3). Positions 1, 2, 4, 5, 8 and 9 had significantly skewed base frequencies. Again, these nucleotide frequencies could be due to the manufacture of the oligonucleotide and/or the cloning process. Despite the non-random nature of the arm library template, it is still possible to analyze the sequences from the efficiently recombined white clones to the overall library population from a statistical standpoint. In order to compare the arm sequences of white clones to that of the random library population, the frequencies for the expected () base distribution in the χ analysis [χ = Σ(( − )/)] were taken directly from the random sequence population calculated above. χ analysis on all 41 white clone arm sequences (Supplementary Figure 3) showed statistically significant nucleotide changes at 6 arm positions. Positions 1, 2, 3, 11, 12 and 13, translating into the first 3 and last 3 arm positions, were statistically different from the random set population. Comparison of the nucleotide frequencies at these 6 positions shows a trend towards the sequence ‘ATA’ at both ends of the library arm for white colony clones (Supplementary Figure 4). This ‘ATA’ sequence is also present in the wild-type site. While 5 of the 6 biased positions are known contacts for the recombinase, position 3 has not been previously identified as an important position for functional arm sequences. These data indicate that there are preferences at the ends of the arm for efficiently recombined sites. Though the total number of sequences in the white colonies with blue centers group is low for statistical analysis, it too shows significant nucleotide distribution changes at positions 11 and 12 (Supplementary Figure 3). Comparison of the nucleotide frequencies at these two positions shows a trend towards the sequence ‘AT’ (Supplementary Figure 5). This, along with the simple calculation of the average number of conserved bases () for each colony type group, points to a continuum for the number of conserved bases in an arm relative to its probable function. The more bases conserved overall, and the more bases conserved at the ends of the arm, the more likely an arm is to recombine efficiently in the 294-CRE system. Several blue and white clones were chosen at random to be tested in Cre reactions. The process of adding back the gene plus the site resulted in a construct (A.3) that was almost identical to the pSV-β-Galactosidase + + arm library construct before it underwent recombination (A.1). reactions were carried out as described for the spacer library except the amount of Cre input was titrated for maximum recombination of 0.1 μg of the arm library control, pSV-β-Gal + . Recombination was detected via color selection following transformation of 2 μl of the reaction into DH5α cells and plating on LB/X-gal agar. On average, a total of 878 colonies were counted for each reconstructed clone encompassing three replicates of their reactions. The control reaction reached a recombination level of 66.0 ± 3.1%. Overall, there was a striking difference in recombination between constructs from white versus blue colonies (). No reconstructed blue clone achieved >0.6% recombination on average, while a full range of total recombination (1.4–54.6%) was observed for white reconstructed clones. These results are more in line with what was predicted for white and blue clones. Clearly, blue clone arm sequences do not support effective recombination , and as might be expected, average fewer conserved bases (2.9, ). Supporting the idea that clones from white colonies with blue centers have characteristics that are part of a continuum between blue and white clones, their average number of conserved bases (4.9) falls between that of the blue and white clones. On average, white clone arms have almost twice as many conserved (both contact, 4.2 and non-contact, 5.4) bases compared to the blue clones. It has already been noted that white clone arms tend to have conserved ‘ATA’ motifs at each end. Drawing precise conclusions about base conservation or preference in relation to percent recombination is precluded by the small overall sample size. However, it is interesting to note that the two best performing arms tested (White #2 and #8) maintained a conserved ‘TATA’ sequence next to the spacer. We have demonstrated the utility of a randomized library approach to the understanding of the nature of the functional domains of the Cre recombinase recognition sequence, as well as the distinct differences in function of sequences between and environments. Although this study did not test a large sampling of all sequences possible in each library (spacer = 4 = 65 536 and arm = 4 = 67 108 864), its results are significant. By electing to randomize whole sections of the site, we have observed a great range of functional sequences, function tied to recombination environment and the first unbiased, systematic evaluation of arm sequences. Two previous studies have concentrated on the sequence of the spacer region and its role in recombination efficiency. Lee and Saito () undertook the task of creating single and double-base mutations of the spacer and noted variability in the recombination efficiency across mutants. Several of their observations are supported by our work such as low or undetectable recombination between unmatched spacers, variable recombination between like mutant spacers, and no mutant spacer with equal or greater recombination efficiency than the wild-type spacer. Our study is, however, in conflict with a published report from Missirlis (). While we agree that mutant spacers and sites with mismatched spacers do recombine, our observations do not substantiate Missirlis 's conclusion that a G-rich spacer is the most favored for recombination or that self-recombination is not significantly greater than recombination with other spacers. Though we cannot rule out the possibility of base preferences within the spacer for efficiently recombined sites, G-rich spacers were not the most effective in our data set. We observed frequency imbalances in our libraries that were the result of the oligonucleotide manufacturing process and/or the cloning of the library. These biases were not from clonal selection since all clones in our study were chosen at random for further manipulation. By sequencing only successful recombination products, Missirlis leave their study open to the possibility that their analyzed pool of products did not originate from a truly random population. Without published analysis of the pool of clones created from the PCR and ligation of the input oligonucleotides, it is unclear whether their observations are the result of nucleotide frequency biases present in the library clones or true base preferences of the Cre recombinase. Second, we believe our observations support the general conclusion that recombination between matching spacers is more efficient and more likely to proceed to completion than recombination between mismatched spacer pairs. Based on our observations, we believe that Missirlis 's conclusions based on equating efficiency of recombination with the number of times a recombination event appears in a shotgun library experiment could be misleading. This could easily be determined by testing Missirlis 's spacer sequences in traditional and assays. In our data set, the non-matching spacer clone (#42) with the highest percent recombination (22.0%) contained sites with homologous spacers with the exception of one position. The remaining unmatched spacer clones all had <20% recombination , compared to 21 of 33 matching spacer clones with >20% recombination. All matched spacer pairs recombined to completion compared to 12 of 21 unmatched. This is the first study to report on a fully randomized (8N) spacer library. Results from this study show there to be great sequence plasticity for successfully recombined spacers. In past work, importance was given to the role of the central ‘TA’ dinucleotide as critical for recombination (,). Our work demonstrates that a central ‘TA’ is not necessary to achieve >45% recombination or complete recombination in 294-CRE. Also, complete identity through the central six bases is not required for recombination as previously thought (). Unfortunately, our sample set was too small to statistically evaluate for base preferences in efficiently recombined spacers. Results from both spacer and arm libraries suggest that recombination in the 294-CRE strain of is robust compared to recombinase reactions. Whether recombination is aided by replication and/or endogenous factors is unclear. This is an important issue to consider when testing for the function of new sites or libraries, as recombination in 294-CRE may not transfer well to the environment. It should also be noted that recombination efficiency in other systems might differ significantly from both the and systems tested here. Though our results from the spacer library study were not unexpected, the fact that our arm library demonstrated effective recombination with the presence of one randomized arm was surprising. It appears that an arm sequence with as few as one conserved base can comprise 1 of the 4 arm sequences necessary for recombination in 294-CRE, and as few as 4 of 13 conserved bases were sufficient for recombination . This could be explained by Cre's cooperative binding properties, where three bound recombinase monomers might be able to recruit the final subunit to the complex leading to recombination in the absence of a fourth arm sequence that has any resemblance to . Our arm library testing system proved useful in the screening of hundreds of clones at once and in identifying the minimal arm sequences required for recombination. Despite not working with a truly random library, statistics reveal nucleotide preferences at 6 out of 13 arm positions for efficiently recombined clones. The ‘ATA’ motif at both ends of the arm was overrepresented in sequences from efficiently recombined clones. Though other studies have looked at the effects of base mutations on the arms (,), Sauer () was the first to speculate on the importance the ‘TATA’ sequence flanking the spacer region. Ours is the first study to find evidence of base preferences for functional arm sequences and disproves the notion that bases at the outer end of the arm are not important for efficient recombination. Clearly, the results from both libraries must be interpreted with the fact that they were tested in the context of the wild-type sequence. The arm library in particular identifies the weakest arm sequences that can be tolerated in an otherwise wild-type reaction. Mutations in the arm or spacer coupled to the fully randomized arm might exert greater selective pressure to conserve bases in the library arm. In the future, we plan to continue to test the function of various site libraries. How mutations in the spacer and in both arms combine to affect function has not been explored. Information from this study and future work would be of use in the expansion of the library of functional sequences, the search for functional endogenous sequences in genomes, finding non-compatible site pairs, and the creation of sites capable of stable recombination events. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

NF-κB/Rel is a critical transcription factor controlling innate immunity, inflammation, cell survival and tumorigenesis (). There are five structurally related members of the mammalian NF-κB/Rel family of proteins: p50, p52, p65/Rel A, Rel B and c-Rel. Genetic analysis has revealed the functional specificities of these NF-κB subunits and shown that the formation of dimers of NF-κB subunits and the exchange of subunits are important for modulating NF-κB activity (). To understand the functional specificity of NF-κB target genes, it is necessary to identify protein–protein as well as DNA–protein interactions involving the NF-κB subunits and its target DNA elements. Although the p65 subunit of NF-κB is the major transcriptional activator of many of NF-κB target genes, other subunits are likely to play important roles in particular NF-κB-mediated processes. For example, the p50 and p52 subunits of NF-κB appear to be important in inflammation and apoptosis (,). In some cases, p50 by itself without a bound p65 subunit, actually activates transcription of target genes. Because it is unusual for proteins without associated transactivation domains to activate transcription, it is likely that transcriptional co-activators are involved in this p50-mediated transcription (). Hence identification of transcriptional co-activators of the p50 subunit should throw light on its target gene specificity. Several natural promoters have been reported to be activated by the p50 subunit but not by the p65 subunit. We chose the C-reactive protein (CRP) promoter as a model for p50-dependent transcription (). CRP is a major human acute-phase protein whose rate of synthesis can increase in inflammatory states (). There is increasing evidence that it is not merely an important risk marker but also has a role in the pathogenesis of atherosclerosis (). It is induced by various cytokines, such as interleukin-6 (IL-6), interleukin-1 (IL-1) and TNF-α (). It was previously proposed that IL-6 stimulates CRP expression, and one of the non-consensus κB sites overlapping the C/EBP-binding site was identified as a p50-responsive element (). Even though it has been proposed that NF-κB p50 subunit plays important roles in CRP expression, the transcriptional activator for p50 protein has not been identified in the cases of these inflammatory signals. β-Catenin is a multifunctional protein that plays critical roles in cell adhesion as well as Wnt-activated tumorigenesis (). When β-catenin levels rise, it accumulates in the nucleus, where it interacts with DNA-bound TCF/LEF family proteins to activate the transcription of various target genes. β-Catenin binds to diverse proteins responsible for specific transcriptional regulation and chromatin remodeling in the nucleus; however, the nuclear protein complexes involving β-catenin remain to be identified. It is especially important to decipher protein–protein interaction on the specific target genes of β-catenin. We searched for a transcriptional co-activator that might provide p50 with a transactivation domain at the CRP promoter and found that β-catenin interacts with the p50 subunit. We also used the technique of Chromatin Conformation Capture (3C) (,) to identify possible interaction sites forming chromatin loops to the CRP promoter. Interestingly, we identified binding sites for TCF/LEF which binds to β-catenin, not in the standard upstream region but in the downstream of the CRP gene (). Most significantly, we also showed that the interaction of the p50 on the CRP promoter and the β-catenin on CRP downstream was involved in CRP activation upon TNF-α treatment. Our findings provide interesting evidence that transcription of the CRP gene depends upon the interaction between p50 and β-catenin and direct long distance interactions of the bound DNA. Cells of 293T were cultured in DMEM containing 10% fetal bovine serum (BRL Life Technology, Inc.). HepG2 cells were cultured in MEM containing 10% fetal bovine serum. Transfection of cells was performed using lipofectAMINE transfection reagent (Invitrogen). pCMV-β (50 ng) was co-transfected in each case to normalize transfection efficiency. After 24 h of incubation, cells were cultured under serum-free conditions for 2 h followed by treatment with 30 ng/ml TNF-α for indicated times. pSUPER-β-catenin (1.5 μg) or pSUPER-p50 was transiently transfected for knock down of p50 and β-catenin protein. After 72 h of transfection, cells were cultured under serum-free conditions for 2 h followed by treatment with 30 ng/ml TNF-α for 8 h before harvest. Recombinant human TNF-α, phosphatase inhibitor cocktail and protease inhibitor cocktail were purchased from Sigma-Aldrich. Restriction and modifying enzymes were from Roche Molecular Biochemicals. Anti-p50 mouse monoclonal antibody (sc-8414) and Anti-TCF-4 rabbit polyclonal antibody (sc-13027) were purchased from Santa Cruz Biotechnology. Anti-FLAG M2 mouse monoclonal antibody (F3165) was from Sigma-Aldrich. Anti-β-catenin mouse monoclonal antibody was from BD Transduction Laboratories. Anti-TCF-1 rabbit polyclonal antibody was generated by KOMA Biotechnology. Human p50 expression clones were kindly provided by Dr Young Mee Kim (Asan Medical Center). Human p65/RelA expression clone were previously reported (). Stable β-catenin (S37A) was previously reported (). FLAG-p50 plasmid was composed of human p50 cDNA (nucleotides 52–1200 of Genbank/Bank Accession NM_003998) in the pCMV-tag2 mammalian expression vector (Stratagene, La Jolla, CA, USA). BAC (bacterial artificial chromosome) clones were purchased from ResGen (Catalog no: RPCI-11.C. Clone ID: RP11-806E24). To make pSUPER-β-catenin and pSUPER-p50, specific oligonucleotides (listed in ) were synthesized (Bioneer) as previously reported (,). To anneal the oligonucleotides, the mixture was incubated at 95°C for 5 min, and was cooled slowly. This mixture was ligated into pSUPER vector (Oligoengine) that had been digested with BglII and HindIII. Whole-cell extracts were centrifuged at 12 000 r.p.m. for 15 min to obtain clear lysates and incubated either with anti-FLAG antibody or with normal rabbit serum for 2 h, followed by the incubation with protein G-Agarose beads (BRL Life Technologies) for 3 h and precipitation. Western blot analysis was previously described (). Cells of 293T were cultured under serum-free conditions for 2 h followed by the treatment with 30 ng/ml TNF-α for 1 h. Cells were fixed in 1% formaldehyde at room temperature and stopped the cross-liking reaction in 125 mM glycine. Cells were lysed with 200 μl lysis buffer [20 mM HEPES (pH 7.9), 10 mM NaCl, 1 mM DTT, 1 mM PMSF, 10 μl protease inhibitor cocktail] for 15 min at 4°C. Nonidet P-40 (5 μl of 0.5%) was added and centrifuged. The pellets were washed with lysis buffer and resuspended in nuclear lysis buffer [20 mM HEPES (pH 7.9), 1.5 mM MgCl, 1 mM EGTA, 1 mM dithiothreitol, 300 μM phenylmethylsulfonyl fluroride, 10% glycerol, 1% Triton X-100, phosphatase inhibitor cocktail and protease inhibitor cocktail] and incubated for 30 min at 4°C. Resuspended cells were sonicated (BRASON sonifier 250, duty cycle 60%, power control 100% 12 sec 2X), centrifuged for 10 min and the supernatants were collected. The chromatin fragments were then cleared with 1 μg of preimmune serum and protein G-Sepharose (30 μl of 50% slurry in 1× PBS) for 2 h at 4°C. Immunoprecipitation was performed for 12–24 h with anti-p50, anti-β-catenin or anti-Tcf-4 antibody at 4°C. Anti-GST or normal rabbit IgG was used as a control. Following immunoprecipitation, 30 μl of protein G-Sepharose was added and incubated for 3 h. Sepharose beads were collected and washed sequentially for 3 min in low salt buffer [0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl (pH 8.1), 150 mM NaCl], high salt buffer [0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl (pH 8.1), 500 mM NaCl] and LiCl wash buffer [0.25 M LiCl, 1% NP-40, 1 mM EDTA, 10 mM Tris-HCl (pH 8.1)] three times. Beads were washed three times with TE buffer and extracted with elution buffer (1% SDS, 0.1 M NaHCO). Elutes were heated at 65°C for 6 h to reverse the formaldehyde cross-linking. DNA fragments were purified by phenol extraction and ethanol precipitation. Purified DNA fragments were amplified with PCR. Primers used in this assay are listed in . Cells of 293T were cultured under serum-free conditions for 2 h followed by treatment with 30 ng/ml TNF-α for 1 h. Cells of 1 × 10 were cross-linked with 2% formaldehyde in 1× PBS (4.5 ml) at room temperature for 10 min. The cross-linking reaction was stopped by adding glycine to a final concentration of 0.125 M. Cells were washed with 1× PBS, resuspended in 1 ml cell lysis buffer [10 mM Tris (pH 8.0), 10 mM NaCl, 0.2% NP-40 and protease inhibitors], and incubated on ice for 10 min. Samples were kept on ice from this point forward. Cell lysis was completed with ten strokes of a Dounce homogenizer (pestle A). Nuclei were washed with 0.5 ml of the 1× BglII restriction enzyme buffer and resuspended in 762 μl of BglII restriction enzyme buffer. SDS was added to a final concentration of 0.1% and incubated for 15–60 min at 37°C while shaking. Triton X-100 was then added to the final concentration of 1%. Digestion was performed with 800 U of BglII at 37°C overnight. The reaction was stopped by adding SDS to a final concentration of 2% and incubated at 65°C for 30 min. The samples (50 μl) were diluted into a 950 μl ligation reaction buffer (5 ng/μl) containing 1% Triton X-100, 1× ligase buffer (NEB) and incubated for 1 h at 37°C. Ligation was started by adding 4000 U of T4 DNA Ligase (NEB) and incubated for 4 h at 16°C. Reactions were stopped by adding EDTA to a concentration of 10 mM. Samples were treated with Proteinase K (100 μg) at 65°C to reverse cross-links, incubated with RNase A (0.5 μg) for 30 min at 37°C, extracted with phenol:chloroform:IAA 2 times, chloroform 2 times, and ethanol precipitated with glycogen. Samples were redissolved in deionized water. Genomic sequences of CRP locus (chr1:156475000-156525000) and their BglII restriction enzyme sites are obtained from UCSC genome browser (). PCR primers are designed by PREMIER biosoft program () and UCSC In-Silico PCR program (). Primers are designed only for non-repetitive sequences. To normalize the differences of the primer efficiency, BAC control template was prepared by digesting with BglII followed by the random ligation. PCR reaction was performed with the control template in parallel with 3C templates. To equalize for the differences in template amount or quality, PCR at the CRP locus were normalized to a control interaction at the GAPDH locus. Primers used in this assay are listed in . -125 CRP luciferase reporter was kindly provided by Dr Irving Kushner (Case Western Reserve University). To generate the chimeric reporter containing the downstream elements of CRP gene, the genomic fragments were cloned on the downstream of luciferase gene. Genomic DNA was prepared from 293T cells and two genomic fragments are amplified with two primer sets (12-1F, 12-2R and 12-2 F, 12-2R in ), which contain one or two TCF/LEF-binding sites. PCR products were cloned in a pCR2.1-TOPO vector (Invitrogen) and digested with BamHI and SalI to clone into -125 CRP Luc reporter. -125 CRP Luc-2 was constructed by cloning a BamHI/Sal fragment having PCR product with 12-2 F and 12-2 R primers, which contains second putative TCF/LEF-binding site within fragment 12. -125 CRP Luc-1-2 was constructed by cloning a BamHI/Sal fragment having PCR product by 12-1 F and 12-2 R primers which contains first and second putative TCF/LEF-binding site within fragment 12. After cloning, -125 CRP Luc reporters were cut by BamHI and used in luciferase assay in some cases. RNA was extracted with Trizol reagent (Invitrogen) and reverse transcribed. Samples were amplified at 95°C for 30 s, 51°C for 30 s, 72°C for 30 s for 30 cycles. The primers for CRP were as follows. F, 5′-CCTATGTATCCCTCAAAGCA-3′ R, 5′-CCCACAGTGTATCCCTTCTT-3′. Since CRP expression is stimulated by various cytokines, we examined how TNF-α induces CRP expression. RT-PCR analysis showed that TNF-α stimulated the transcription of the CRP gene in both 293T and HepG2 cells (A). When we used the -125 CRP Luc reporter that contains the CRP promoter, luciferase activity was also enhanced after TNF-α treatment in two cell lines (data not shown). Since NF-κB is a key mediator of the TNF-α-induced expression of many genes, and the CRP promoter contains a non-consensus κB-binding site (AAACTCCCTTA) on -50 to -40, we tested the effect of overexpressing either the p50 or the p65 subunit of NF-κB on CRP mRNA expression. As shown by RT-PCR analysis (B) and luciferase assays with the -125 CRP promoter (C), overexpression of p50 increased CRP promoter activity in the 293T cell line. On the other hand p65 overexpression did not stimulate the promoter activity. To confirm that the p50 and p65 plasmids possessed the expected biological activity, we determined the luciferase activity transcribed from consensus κB sites (NF-κB Luc) in 293T cells. As expected, overexpression of p65 markedly activated the NF-κB-Luc reporter whereas overexpression of p50 did not (D). This suggests that the TNF-α-induced CRP transcription may be dependent on the p50 subunit of NF-κB acting on the non-consensus κB site in the -125 CRP promoter. Because the p50 subunit of NF-κB lacks a transactivation domain, transcriptional activation by p50 alone is unusual (,). We searched for a putative co-activator that could provide p50 with a transactivation domain during TNF-α signaling. Since TNF-α activates NF-κB signaling and PI3-Kinase/Akt pathway by which β-catenin nuclear translocation is activated, we chose β-catenin as a putative binding partner for NF-κB p50 protein. In addition, because several reports suggested that β-catenin and NF-κB may physically interact in basal level (,), we reasoned that β-catenin could provide the transactivation domain to p50 subunit. We also showed by RT-PCR analysis that β-catenin and p50 acted in concert to enhance the transcription of the endogenous CRP gene in the 293T and HepG2 cell lines (A). These results suggest that β-catenin contributes to the transcriptional activation of NF-κB p50 on the CRP promoter, and may thus act as a transcriptional co-activator of the p50 subunit. To establish the critical roles of the NF-κB p50 subunit and β-catenin in CRP gene transcription, we knocked down the expression of the two proteins using pSUPER-shRNAs in HepG2 cells. To confirm the efficacy of the shRNAs, we performed western blot analyses with anti-p50 and anti-β-catenin antibodies (B). The shRNA for p50 greatly reduced the steady-state level of the p50 precursor protein, p105, and lowered somewhat the level of p50 protein. The shRNA for β-catenin caused significant reduction of the β-catenin protein level. We next tested whether the shRNAs inhibited TNF-α-induced CRP gene expression (C). Knockdown of p50 resulted in a substantial reduction of CRP mRNA, which can be explained by the major roles of the NF-κB p50 in basal and activated CRP transcription. In addition, reduction of CRP mRNA was also observed when shRNA for β-catenin was introduced. These results suggest that NF-κB p50 and β-catenin are important components of TNF-α-induced CRP gene expression. There is some evidence that the p65 subunit of NF-κB interacts physically with β-catenin (,). We tested whether the p50 subunit interacts with β-catenin in the course of TNF-α-induced CRP transcription. We asked if TNF-α promoted the interaction between β-catenin and p50 by performing co-immunoprecipitation assays with FLAG-tagged p50-transfected 293T cells. Even though there is some basal protein–protein interaction between p50 and β-catenin, we found that one hour of TNF-α treatment greatly increased the amount of β-catenin which was co-immunoprecipitated by anti-FLAG antibody (A). Since we had shown that TNF-α activates CRP expression by NF-κB p50 on the CRP promoter, we tested if TNF-α causes recruitment of p50 to the -125 CRP promoter. We performed chromatin immunoprecipitation (ChIP) assays with anti-p50 antibody and primers specific for the -125 CRP promoter. As shown in B, binding of p50 to the -125 CRP promoter was induced by TNF-α treatment. These results indicate that the interaction between p50 and β-catenin is critical for TNF-α-induced CRP gene transcription. Chromosome conformation capture (3C) is a newly developed technique that can effectively map the long distance interactions between promoter and regulatory sites (,,,). Since most activation of gene expression involves changes of chromosome conformation, we used the 3C technique to map TNF-α-induced chromosome interaction sites near the CRP locus. 3C uses sequential nuclear restriction enzyme digestion, ligation and PCR-based quantification of the ligated DNA products to assess the proximity of different regions of a chromosome. Two restriction fragments that are close together in the nucleus display a higher cross-linking frequency and therefore a higher ligation frequency than two fragments that are far apart. A shows the CRP locus and the BglII restriction sites within 50 kb of the locus. BglII was chosen because its sites are evenly spaced along the CRP gene and flanking sequences. The segment between fragments 7 and 8 contains the CRP promoter to which NF-κB p50 binds. To test the effect of TNF-α on the proximity of regulatory elements to the CRP promoter, we used an upstream anchor primer within the promoter fragment in pairwise combinations with primers within the fragments of the CRP gene (B). All the PCR reactions were performed in triplicate and averaged. The 3C analysis showed that, in the absence of TNF-α (open bar), the proximity of each fragment (fragments 1–7) to the CRP promoter decreased with distance. However, after TNF-α treatment (solid bars), the proximity of fragment 4 to the promoter increased (C). We then used the downstream anchor primer in pairwise combination with primers within the BglII fragments and found that TNF-α treatment (solid bar) greatly increased the proximity of fragment 12 (D and E). All the 3C data were normalized by the control BAC template and a GAPDH interaction. These results suggest that TNF-α treatment brings fragments 4 and 12 in contact with the CRP promoter. The induced proximity of chromosomal regions may be caused by interactions between proteins bound to chromosomal sites that are far apart (). We hypothesized that the 3C interaction observed after TNF-α treatment was due to interaction between p50 bound to the -125 CRP promoter (B) and β-catenin bound to the distant regulatory region(s). Since β-catenin itself does not bind to DNA but forms a DNA-bound complex with TCF-4 (), we searched for putative TCF/LEF-binding sites within fragments 4 and 12. Chromatin immunoprecipitation assays were performed with a probe specific for fragment 4 (A). TNF-α treatment did not increase the binding of β-catenin and TCF-4 within fragment 4, which probably reflects the absence of TCF/LEF-binding sites in this fragment. In contrast, six putative TCF/LEF-binding sites were found within fragment 12, and ChIP assays with primers specific for the second putative TCF/LEF-binding sites within fragment 12 showed that the binding of p50, β-catenin and TCF-4 was greatly increased by TNF-α treatment (B). This suggests that the -125 CRP promoter bound p50 and the distal regulatory sites (fragments 12) bound β-catenin are in close proximity. We used fragment 13 as a negative control for the ChIP experiments since the 3C experiment showed that its proximity to the promoter did not increase after TNF-α treatment (C). Since the previous ChIP data suggest that the second putative TCF/LEF-binding site might be responsible for the β-catenin binding to fragment 12, we tested whether the site is responsible for the activation of CRP transcription. We designed the -125 CRP luciferase reporters with the putative TCF/LEF-binding sites from fragment 12 located in the downstream of the gene and named them as -125 CRP Luc-2 and -125 CRP Luc-1-2. B showed that the luciferase activities of the -125 CRP Luc reporters containing one or two putative TCF/LEF-binding site were increased to 2-fold in comparison to that of -125 CRP Luc reporter having no TCF/LEF-binding site. Protein levels of expressed p50 and β-catenin was not different in these samples as shown by immunoblotting with anti-FLAG, anti-β-catenin antibody (B). This result suggests that CRP transcriptional activity is enhanced by the fragment 12 which has putative TCF/LEF-binding sites. We showed here that after TNF-α treatment NF-κB p50 activates CRP expression by associating with β-catenin/TCF bound to a distant downstream region of the CRP gene. The NF-κB p50 subunit lacks a transactivation domain, but we present evidence for a direct interaction between p50 and β-catenin that could provide the p50 with a transactivation domain after TNF-α signaling. We also have evidence that shRNA-mediated suppression of β-catenin significantly reduces TNF-α-induced CRP expression as well as expression of other p50-dependent genes (Choi , unpublished data). These results suggest that TNF-α-induced nuclear translocation of β-catenin could serve as a transcriptional co-activator for the p50 subunit. There is much evidence that PI3-kinase/Akt activation is involved in nuclear translocation of β-catenin following GSK-3β inactivation. It is reported that lipopolysaccharide activates Akt1 in alveolar macrophages resulting in nuclear accumulation and transcriptional activation of β-catenin (). Just as Wnt-activated β-catenin functions as a transcription co-activator for TCF/LEF proteins, activation of β-catenin by PI3-Kinase/Akt is reported to provide a transcriptional co-activator function for the androgen receptor (). We also showed here that β-catenin is involved in the activation of CRP gene with NF-κB p50 protein. If nuclear β-catenin plays a role as a transcriptional co-activator for p50, it has to directly or indirectly interact with the p50 subunit. Several reports have suggested a possible relation between β-catenin and NF-κB (,,) but there has been no definitive evidence that TNF-α induced interaction between them. We showed here for the first time that TNF-α significantly increased the extent of interaction between β-catenin and NF-κB p50 subunit. Full understanding of TNF-α-induced CRP transcription may require the identification of the intermediate proteins as well as signal pathways leading to the strong interaction between NF-κB p50 and β-catenin (). Transcription is mediated by the binding of specific transcription factors to regulatory sequences. Long-range interactions between promoter and regulatory elements are important for high level expression of target genes, and may cause looping of chromatin as well as protein–protein interactions (). Since nuclear β-catenin is engaged in TCF/LEF protein complex (), we screened the genomic sites induced to interact with CRP promoter by 3C technology. We also searched for the TCF/LEF-binding sites and identified a fragment 12 (∼3.3 kb downstream from the promoter) as the binding site of β-catenin by ChIP analysis. If nuclear β-catenin binds TCF/LEF which already occupies a specific sequence in fragment 12, it is likely that TNF-induced binding of β-catenin to the CRP promoter is caused by the looping of chromatin between promoter and the TCF/LEF-binding elements in fragment 12. Our 3C data indicate that the proximity of fragments 4 and 12 to the CRP promoter was increased by TNF-α treatment. However, putative TCF/LEF-binding sites do not exist within fragment 4 and we could not detect any TNF-α-induced binding of β-catenin to fragment 4 by ChIP analysis. This suggests that the interaction between fragment 4 and the CRP promoter may be caused by interaction of p50 with some other protein. Since NF-κB p50 protein may be involved in regulating inflammatory genes such as CRP, we propose here that during TNF-α-induced inflammation, NF-κB p50 induces inflammatory genes by interacting with β-catenin bound to downstream region of the gene. This idea may provide a missing piece of the puzzle posed by the complex cellular NF-κB/β-catenin regulatory network.

PTIP (Pax2 transactivation domain-interacting protein) is a key regulator of cellular responses to DNA damage that is vitally important for cell and organism function. It was originally identified in mice in a two-hybrid screen with the Pax2 transcription factor that regulates embryonic development (). PTIP null embryos do not recapitulate the phenotype of mice lacking Pax2, but instead show very high levels of DNA damage and embryos die at day E8.5 because the DNA damage sustained in S-phase causes a mitotic block (). These observations suggested that PTIP plays an important role in regulating genome stability. Human (h)PTIP is required for survival of cells exposed to ionizing radiation (IR) (,). hPTIP binds to sites of DNA damage (,) and appears to function as an ‘adaptor’ protein in that it is required for IR-induced phosphorylation of a subset of targets of the ATM (taxia elangiectasia-utated) protein kinase (). ATM is a key regulator of cellular responses to double-stand breaks (DSBs). Binding of ATM to sites of DNA damage appears to stimulate ATM kinase activity (), leading to phosphorylation of target proteins at Ser–Gln or Thr–Gln (S/T–Q) motifs (). For example, ATM, and the related kinase ATR, phosphorylate Ser129 of the core histone variant H2AX at sites of DNA damage (), and phospho-H2AX in turn acts as a platform for the recruitment of proteins that are needed to signal and repair DNA damage. The MDC1 protein, for example, has a single pair of BRCT domains that bind to phospho-Ser129 of H2AX, thereby recruiting MDC1 and associated proteins to sites of DNA damage (). BRCT domains, mostly found in pairs, are small modules of ∼100 amino acids () that mediate protein–protein interactions. In some cases these domains recognize phosphorylated epitopes on target proteins (,). hPTIP has a pair of BRCT domains at the N-terminus, and another two pairs at the C-terminus (A). The most extreme C-terminal pair alone was shown to bind to a library of phospho-peptides based on the S/T-Q consensus sequence for phosphorylation by ATM. The optimal sequence bound by this pair of BRCT domains was found to be pS/T-Q-V-F (). hPTIP was shown to interact with 53BP1 after exposure of cells to IR (,). 53BP1 is also an ‘adaptor’ protein for ATM (,) and mice lacking 53BP1 are tumour prone and are hypersensitive to IR (). It was recently shown that 53BP1 is down-regulated in transition from pre-cancerous stage to carcinomas (), and loss of a single 53BP1 allele in mice causes genome instability and lymphoma (). Cells lacking 53BP1 show mild cell cycle checkpoint defects (,,) and a pronounced defect in the repair of a subset of DNA breaks by non-homologous end joining (NHEJ) (). Intriguingly, the interaction of hPTIP and 53BP1 after DNA damage requires ATM-dependent phosphorylation of 53BP1 (,). Translocation of these proteins to sites of DNA damage does not require ATM, however, and is therefore independent of their physical association (). Although it was implied that binding of a phosphorylated residue in 53BP1 that lies in the pS/T-Q-V-F motif, to the most C-terminal BRCT domain pair in hPTIP explained how these proteins interacted (), this was not tested. Thus, the mechanism(s) of interaction of hPTIP and 53BP1 or the significance of their interaction for the DNA damage response is not yet known. In this study, we address these questions. HEK 293 and U2OS cells were grown in DMEM (Gibco BRL) supplemented with 10% fetal bovine serum (FBS, HyClone) and penicillin/streptomycin. Cells were kept at 37°C in a 5% CO atmosphere. Co-immunoprecipitation of FLAG-hPTIP and HA-53BP1 proteins, SDS–PAGE and western blot analysis were carried out as described previously (). The primary antibodies used in this study were: anti-HA (12CA5; Roche), anti-FLAG (M2; Sigma), anti-53BP1, anti-phospho Ser1524 BRCA1 (Bethyl Laboratories), anti-phospho Thr68 Chk2, anti-Chk2 (Cell Signaling Technologies) and anti-BRCA1 (Oncogene Research Products). Antibodies against hPTIP were described previously (). All peptides were synthesized by Dr Graham Bloomberg, University of Bristol. Full-length 53BP1 was amplified from plasmid pCMH6K-53BP1 () with an N-terminal HA tag, sub-cloned into pCR2.1 and cloned into the KpnI and SalI sites of pCMV5. To create the 53BP1 phospho-site mutants shown in , up to five mutations in 53BP1 at a time were introduced into 53BP1 using the QuikChange Multi-Site mutagenesis kit (Stratagene) and PCR reactions were spiked with Ultra DNA polymerase (Stratagene). All plasmids were checked carefully by sequencing the entire insert forwards and backwards. Plasmids expressing hPTIP were described previously (). pCMV5-based plasmids were transfected into HEK293 cells using calcium phosphate. In the siRNA/rescue experiments, 53BP1 siRNA duplexes that recognize nucleotides 84–104 in human 53BP1 (AAGCCAGTTCTAGAGGATGA), or scrambled (SCR) duplexes (100 nM) were co-transfected into cells with 53BP1 siMUT plasmids (0.5 μg) in which 53BP1 bore a mutation at nucleotide T93 that did not affect the coding sequence but made 53BP1 refractory to silencing by siRNA. The calcium phosphate method of transfection of HEK293 cells was used for RNA interference and rescue because this allowed transfection efficiencies of >90% (data not shown). hPTIP depletion by siRNA duplexes was described previously (). The hPTIP BRCT domains were amplified from pCMV5-hPTIP () and cloned into the BamHI site of pGEX6P3. Proteins expression in BL21 (DE3)-RIL cells (Stratagene) was induced by addition of IPTG at 10 μM, when cells were at OD 0.6, at 16°C for 16 h. Cells were lysed in 50 mM Tris/HCl, pH 7.4 containing 1% (w/v) Triton X-100, 150 mM NaCl, 1 mM EGTA, 1 mM EDTA, 0.1% (v/v) 2-mercaptoethanol, 1 mM benzamidine and 0.2 mM phenyl methyl sulphonate. After sonication of cells and centrifugation to remove cell debris, proteins were purified on glutathione-sepharose (AP Biotech) and eluted with a gradient of reduced glutathione (Sigma). Proteins were dialysed free of glutathione and stored frozen at −80°C. Binding was analysed in a BiaCore 3000 system. The relevant biotinylated peptides were bound to an SA sensor chip (GE Healthcare). The indicated concentrations of bacterially expressed wild-type and mutant forms of GST–hPTIP (590–1069) (BRCT pairs C1 + C2) in HBS-EP [HEPES-buffered saline with EDTA and polysorbate 20; 10 mM HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA and 0.005% (v/v) polysorbate 20], were injected over the immobilized peptides at a flow rate of 90 μl/min. Interactions between each peptide and GST–hPTIP pair C1 + C2 (amino acids 590–1069) were analysed and steady-state binding was determined at each concentration. Dissociation of GST–hPTIP pair C1 + C2 forms from each peptide was monitored over 90 s. Regeneration of the sensor chip surface between each injection was performed with three consecutive 5 μl injections of a solution containing 50 mM NaOH and 1 M NaCl. We previously reported that hPTIP interacts with 53BP1 only after DNA damage and that this requires phosphorylation of 53BP1 by ATM. 53BP1 has a total of 32 ATM consensus motifs (S/T-Q) (A). The Ser/Thr residues each of these motifs was mutated to alanine and the resulting mutants were transfected into HEK293 cells and tested for interaction with co-transfected with FLAG-hPTIP. Mutation of the 10 most C-terminal S/T-Q motifs in 53BP1 (termed MUT-C10; A) or mutation of a combination of 17 S/T-Q motifs upstream of the MUT-C10 mutations (termed MUT-N17; A) did not affect its interaction with hPTIP after exposure of cells to IR (B). However, mutation of an additional three residues Ser25, Ser29 and Ser105 (termed MUT-N20), abolished the IR-inducible interaction of 53BP1 with hPTIP (B). Whereas mutation of Ser29, Ser105, Ser6 or Ser13, singly had no effect, mutation of Ser25 of 53BP1 abolished the IR-induced interaction of 53BP1 and hPTIP (C and D). Thus, Ser25 in 53BP1 is essential for the interaction of 53BP1 with hPTIP. It was reported previously that 53BP1 is phosphorylated at Ser25 by ATM (). Consistent with this, we observed that Ser25 is phosphorylated after exposure of HEK 293 cells to IR, but not in cells pre-incubated with the ATM-specific inhibitor KU55933 () (E). Pre-incubation of cells with NU7441 (), a specific inhibitor of DNA-dependent protein kinase (E), or ablation of ATR expression () (data not shown), did not affect IR-induced phosphorylation of 53BP1 Ser25. The 53BP1 Ser25Ala mutant was retained at sites of DNA damage in a manner indistinguishable from the wild-type protein () (F; data not shown). Therefore, the involvement of Ser25 phosphorylation in promoting the interaction of 53BP1 and hPTIP is independent of translocation of 53BP1 to sites of DNA damage. This is consistent with previous reports that ATM is not required for formation of IR-induced nuclear foci by 53BP1 or hPTIP (,). The data in this section demonstrate that Ser25 that is phosphorylated by ATM is essential for the interaction of 53BP1 with hPTIP after DNA damage and that this is independent of 53BP1 translocation to IR-induced foci. We tested if phosphorylated Ser25 of 53BP1 could bind directly to hPTIP. Two peptides corresponding to the sequence around Ser25, in which Ser25 is phosphorylated or not, were synthesized with a biotin moiety at the N-terminal end. As a test for specificity, a peptide in which nearby Ser29 is phosphorylated, and the Ser25/Ser29 diphospho-peptide, were also synthesized. Peptides were immobilized on streptavidin-coated magnetic beads and incubated with cell extracts. As shown in B, endogenous hPTIP was retrieved by the phospho-Ser25 peptide and phospho-Ser25/phospho-Ser29 peptide but not by the unphosphorylated peptide or the phospho-Ser29 peptide or by an unrelated diphospho-peptide from yeast Rad53. Similar results were obtained with full-length FLAG-hPTIP transfected into cells, and with a transfected fragment of hPTIP (amino acids 590–1069) (B), corresponding to the two C-terminal pairs of BRCT domains (pair C1 + pair C2; A), that we previously showed were necessary and sufficient for hPTIP to interact with 53BP1 after DNA damage in cells (). These data demonstrate that hPTIP interacts with phosphorylated Ser25 of 53BP1 and that this is required for the inducible interaction of the two proteins. We assumed that just one of the two C-terminal pairs of hPTIP BRCT domains is responsible for binding phospho-Ser25. To test which of the two pairs was responsible, the two isolated pairs or both pairs together were expressed in bacteria as GST fusion proteins. Neither the first (pair C1) nor the second (pair C2) pair of BRCT domains in isolation bound to the Ser25 phospho-peptide whereas both domains together (pair C1 + pair C2) bound efficiently (C). BiaCore analysis demonstrated that the dissociation constant () for binding of BRCT pair C1 + pair C2 to the phospho-Ser25 peptide was ∼526 nM, whereas the for binding to the non-phospho-peptide was >1000 μM, demonstrating a high degree of phospho-specificity (A; data not shown). The of ∼526 nM is in good agreement with other physiological BRCT–phosphoepitope interactions; the for Ser129 of H2AX and the BRCA1 BRCT domain pair was reported to be 2.2 μM (). BiaCore measurements showed no detectable binding of hPTIP BRCT pairs C1 or C2 in isolation to the phospho-Ser25 peptide (B). The requirement for two BRCT domain pairs for recognition of phospho-Ser25 was surprising because in all reported cases of phospho-epitope recognition by BRCT domains, a single pair is sufficient and also because it was reported that the BRCT domain pair C2 of hPTIP alone bound to a library of degenerate pS/T-Q phosphopeptides. The optimal peptide used in these experiments by Yaffe and colleagues () was biotin-GAAYDI-pSQ-VFPFAKKK. In agreement with these data, even though hPTIP BRCT domain pair C2 did not bind to 53BP1 phospho-Ser25, it bound to this phospho-peptide (termed ‘S-Q-V-F peptide’) with a of ∼273 nM (data not shown), and not to the unphosphorylated version of this peptide (D). These data taken together suggest that BRCT pair C1 + C2 of hPTIP are capable two modes of phospho-epitope recognition: one specific for pS-Q-V-F peptides mediated solely by pair C2, and another mode that is responsible for binding phospho-Ser25 that needs both pair C1 + pair C2. To confirm the requirement for two separate pairs of BRCT domains in hPTIP to bind phospho-53BP1, the effect of mutation of conserved residues in these domains was determined. Inspection of the amino acid sequence of hPTIP revealed that BRCT domain pair C2 contains a small conserved motif that in other proteins recognizes phospho-peptides. MDC1 and BRCA1 each have a single pair of BRCT domains that interact with phospho-peptides (,). Structural analyses showed that Arg1933 of MDC1 and Arg1699 of BRCA1 coordinate phospho-Ser and are essential for phospho-peptide binding (,,). Alignment of each of the two C-terminal pairs of BRCT domains from hPTIP, individually, with the BRCT domain pairs from MDC1 and BRCA1 revealed clear conservation of residues involved in phospho-specific binding in the second pair, but not in the first pair, of hPTIP C-terminal BRCT domains (Figure S1). Arg910 in BRCT pair C2 of hPTIP appears to be equivalent to Arg1699 of BRCA1 and Arg1933 of MDC1 (Figure S1). BiaCore measurements showed that whereas pair C1 + C2 bound to the 53BP1 phospho-Ser25 peptide with a of ∼526 nM, mutation of Arg910 (R910Q) severely reduced binding of BRCT pair C1 + pair C2 of hPTIP to phospho-Ser25 of 53BP1 so that the was so high that it was difficult to determine (A). Almost all BRCT domains have a conserved Trp on the α3 helix () that corresponds to Trp676 in hPTIP BRCT pair C1 and Trp929 in pair C2 (Figure S1). Mutation of either Trp676 (W676A) or Trp929 (W929A) to alanine severely reduced binding to Ser25 phospho-peptide and in each case the was >1000 μM (A). As shown in B, mutation of Trp676, Arg910 or Trp929 abolished binding of hPTIP to 53BP1 in IR-treated cells . These data are consistent with the requirement for two pairs of BRCT domains for hPTIP to recognize phospho-Ser25 of 53BP1. A previous report showed that hPTIP BRCT pair C2 alone fused to GST could retrieve phosphorylated 53BP1 from extracts of irradiated cells (). However, the data presented in this study indicate that two hPTIP BRCT pairs are required to bind to phospho-53BP1, and this discrepancy was investigated. Protein fragments corresponding to the hPTIP BRCT domains were expressed as GST fusions in bacteria, immobilized and incubated with extracts of cells exposed, or not, to IR. As shown in C, BRCT Pairs C1 + C2 of hPTIP retrieved 53BP1 from extracts of irradiated cells much more efficiently than from un-irradiated cells. The isolated BRCT pair C2 of BRCT domains also retrieved 53BP1 from extracts of irradiated cells but with much lower efficiency than hPTIP BRCT pairs C1 + C2 (C), showing that pair C1 is important. Retrieval of 53BP1 from extracts by GST-hPTIP (590–1069) was prevented by mutation of the W676A mutation in pair C1 (C). Therefore, BRCT pair C2 of hPTIP only binds very weakly to phospho-53BP1 in cell extracts compared with a combination of both C-terminal pairs of hPTIP BRCT domains. These data are consistent with our previous finding that both C-terminal pairs of hPTIP BRCT domains are necessary and sufficient to interact with 53BP1 in cells after DNA damage (). We previously showed that both C-terminal pairs of BRCT domains in hPTIP are necessary and sufficient for the formation of nuclear foci by hPTIP after DNA damage (). Since both these domains in hPTIP are also required for binding to phospho-Ser25 of 53BP1, we investigated if binding of the C-terminal domains of hPTIP is required for stable retention of hPTIP at sites of DNA damage. U2OS cells were transiently transfected with plasmids expressing wild-type FLAG-hPTIP and hPTIP bearing the following mutations: Trp676Ala (W676A), Arg910Gln (R910Q) or Trp929Ala (W929A). As shown in A, formation of foci by endogenous 53BP1 was normal in all irradiated cells: this was used as an internal control for focus formation. Wild-type hPTIP formed nuclear foci after exposure of cells to IR (A): almost all cells had >50 foci under these conditions, whereas most cells had <10 foci in untreated cells (B). In contrast, mutation of Trp676 or Trp929 prevented focus formation by hPTIP (A); very few cells had >50 foci after IR, and almost all cells had <10 (B). Mutation of Arg910 had little effect on the ability of hPTIP to form foci after IR and this mutant was difficult to distinguish from wild-type hPTIP in this regard (A and B). However, all three of these mutations—Trp676Ala, Trp929Ala and Arg910Gln—prevent the association of hPTIP with phospho-Ser25 of 53BP1 and (B). Therefore, the two C-terminal BRCT domains of hPTIP are essential for binding to sites of DNA damage but it appears that this is independent of binding to phosphorylated Ser25 of 53BP1. This is consistent with our previous finding that ATM,which catalyses Ser25 phosphorylation, is not required for hPTIP to form nuclear foci after DNA damage (). We next wished to test if the interaction of 53BP1 and hPTIP has important responses to DNA damage and we first tested the effects of mutation of Ser25 of 53BP1. Transfection of HEK293 cells with a siRNA duplex to silence 53BP1 (), but not with a scrambled (SCR) version of this duplex, caused a dramatic reduction in 53BP1 protein levels (A). An assay to allow expression of ectopic HA-tagged 53BP1was then established to allow complementation analysis. This involved making a mutation in (HA-tagged) 53BP1 cDNA on a plasmid at a point (T93) in the target sequence recognized by the siRNA duplex, to render HA-53BP1 refractory to siRNA-mediated silencing. Conditions were established so that HA-53BP1 or HA-53BP1 in which Ser25 was mutated to Ala were expressed at close to endogenous levels, judged by immunoblotting with anti-53BP1 antibodies (A). Immunoprecipitation with anti-HA antibody fully depleted 53BP1 from extracts of cells co-transfected with 53BP1 siRNA and HA-53BP1 (data not shown). This showed that all of the 53BP1 expressed under these conditions corresponded to exogenous 53BP1. 53BP1 is required for ATM-dependent phosphorylation of BRCA1 and CHK2 at low doses of radiation (,,). Consistent with these reports, depletion of 53BP1 caused a severe reduction in phosphorylation of Chk2 at Thr68 and phosphorylation of BRCA1 at Ser1524 (A). Whereas wild-type 53BP1 could rescue IR-induced phosphorylation of Chk2 and 53BP1, mutation of Ser25 to Ala prevented this. We next looked at the effect of mutation of Ser25 on cellular resistance to IR. Depletion of 53BP1 rendered cells hypersensitive to IR (B), consistent with previous reports (,). Treatment of 53BP1-depleted cells with 3 Gy of IR caused an ∼90% decrease in cell viability, in contrast with an almost 40% decrease in viability in cells treated with a scrambled (SCR) RNA duplex (B). 53BP1 in which Ser25 was mutated to alanine was unable to rescue this defect whereas wild-type 53BP1 restored cellular resistance to IR (B). These data show that Ser25 of 53BP1, that mediates the inducible interaction of 53BP1 with hPTIP, is required for 53BP1 to act as an adaptor protein for ATM and for cellular resistance to DSBs. Mutation of Arg910 in hPTIP BRCT pair C2 prevents binding of hPTIP to phospho-Ser25 of 53BP1 and prevents binding of hPTIP to 53BP1after DNA damage in cells, but does not affect hPTIP focus formation. This mutation is therefore useful for probing the consequences of the interaction of hPTIP with 53BP1 phospho-Ser25. hPTIP-specific RNAi was carried out in HEK293 cells and this, like depletion of 53BP1, resulted in a dramatic reduction in the IR-induced phosphorylation of Chk2 at Thr68 and of BRCA1 at Ser1524 (C). Whereas wild-type FLAG-hPTIP could fully rescue these defects, the Arg910Gln (R910Q) mutant could not. Therefore, Arg910 of hPTIP BRCT pair C2 is required for an intact response to DNA damage. These data strongly suggest that recognition of 53BP1 phospho-Ser25 by the C-terminal BRCT domains of hPTIP is important for maintenance of genome stability. The data presented above demonstrate that the two pairs of C-terminal BRCT domains of hPTIP are capable of at least two modes of phospho-epitope recognition and that they control at least two important, but independent, facets of hPTIP function: interaction with phosphorylated 53BP1 and translocation to sites of DNA damage. At the outset of this study, it was known that hPTIP interacts with 53BP1 after DNA damage in an ATM-dependent manner (,) but the mechanisms or significance of this interaction were unclear. In this study, we showed that a single ATM-phosphorylated residue in 53BP1—Ser25—is required for interaction with hPTIP in cells. Mutation of Ser25 did not grossly perturb 53BP1 function since the 53BP1 Ser25Ala mutant protein formed foci after DNA damage in a manner indistinguishable from the wild-type protein () (data not shown). Phosphorylated Ser25 was shown to interact with the two C-terminal pairs (Pair C1 + C2) of hPTIP (). We found that, surprisingly, both of these BRCT pairs are required to bind to the phospho-Ser25 peptide and neither domain alone could bind to this peptide (B). Consistent with these data, mutations of conserved residues in Pair C1 (Trp676) or in Pair C2 (Arg910 or Trp929) severely reduced binding of hPTIP to phospho-Ser25 . Furthermore, mutation of any of these residues abolished the binding of hPTIP to 53BP1 after DNA damage . This is the first reported example of a requirement of two pairs of BRCT domains for binding to a single phospho-epitope. It is not yet clear why two BRCT pairs are required to recognize 53BP1 phospho-Ser25, especially when in BRCA1 and MDC1 (and in other BRCT-proteins), a single BRCT pair in each case is sufficient for interacting with phospho-H2AX and phospho-BACH1, respectively (,). We predict that the small conserved basic patch containing Arg910 in hPTIP pair C2 contacts the phospho-Ser25 of 53BP1 and that pair C1 makes contact with residues nearby, possibly helping to determine the specificity of the interaction. Solving the crystal structure of hPTIP BRCT domain pairs C1 + C2 in complex with the 53BP1 phospho-Ser25 peptide should provide valuable insight into the detailed mechanism of this interaction. The requirement for two pairs of BRCT domains for hPTIP to recognize a single phospsho-epitope—Ser25 of 53BP1—is somewhat surprising since isolated BRCT domain pair C2 of hPTIP was shown previously, in peptide selection experiments, to interact with synthetic phospho-peptides that lie in an pS/T-Q-V-F motif. Ser25 does not lie in this type of motif but there are two serine residues in 53BP1—Ser29 and Ser105—that do conform to the pS/T-Q-V-F motif, are not required for 53BP1to bind hPTIP after DNA damage (C). The observation that hPTIP BRCT pairs C1 + C2 recognize a phospho-peptide different from the pS/T-Q-V-F peptides bound by pair C2 in isolation suggests that there are two modes of phospho-epitope recognition resident in pairs C1 + C2. Again, it would be interesting to compare the crystal structure of hPTIP BRCT pairs C1 + C2 in complex with the two types of phospho-peptide to ascertain whether different modes of interaction are at play. It would be interesting to know if two types of phosphopeptide bind in a mutually exclusive manner or if they can bind to hPTIP simultaneously. We showed previously that both BRCT pairs C1 and C2 are required for translocation of hPTIP to sites of DNA damage. In this study, we showed that mutation of Trp676 in hPTIP BRCT pair C1 abolished formation of IR-induced nuclear foci by hPTIP, as did mutation of Trp929 in pair C2 (). However, the R910Q mutation that also abolished binding of hPTIP to phospho-Ser25 and to 53BP1 in cells, did not affect formation of nuclear foci by hPTIP after IR. Therefore, both C-terminal pairs of BRCT domains in hPTIP appear to be essential for binding of sites of DNA damage but this appears to be independent of their ability to bind phospho-Ser25 of 53BP1. This is consistent with previous reports that ATM, the Ser25 kinase, is not required for binding of hPTIP or 53BP1 to sites of DNA damage (,). At present, the molecular mechanisms governing translocation of hPTIP are not clear and this will be interesting to investigate. Whatever the case, it is clear both C-terminal BRCT domains of hPTIP are required. Mutation of 53BP1 Ser25, that abolished interaction with hPTIP, and mutation of Arg910 of hPTIP that prevents interaction with phospho-Ser25, prevented phosphorylation of BRCA1 and Chk2 by ATM. Thus, 53BP1 must interact with hPTIP to exercise its role as adaptor and to protect cells against DSBs. At present, it is not clear how 53BP1 or hPTIP functions to assist ATM phosphorylation but may involve recruitment of ATM substrates to sites of DNA damage. Mutation of Ser25 also caused cells to become hypersensitive to IR, probably indicative of a DNA repair defect, since 53BP1 are defective in NHEJ of a subset of DSBs in cells (). It would be interesting to know if mutation of Ser25 in mice recapitulates the same spectrum of tumours seen in 53BP1 null mice, and if mutation of Ser25 affects immunoglobulin class switching. Neither 53BP1 nor hPTIP appear to have catalytic activity and probably act as ‘scaffolding’ proteins to recruit and direct ‘effector’ polypeptides to sites of DNA damage. Ultimately it will be vital to identify the effector molecules that hPTIP and 53BP1 bring to sites of DNA damage that facilitate DNA repair and that enable phosphorylation of ATM targets. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Living cells could, at any moment, suffer DNA damage. If damage is left unrepaired, consequent genomic instability can compromise cell survival. Nucleotide excision repair (NER) is a versatile repair pathway that can eliminate a wide variety of lesions, e.g. UV-induced photolesions including cyclobutane pyrimidine dimers (CPD) and 6-4 photoproducts (6-4PP), from the genome of UV exposed cells (). NER includes two distinct subpathways, global genomic repair (GGR) which removes lesions from the entire genome, whereas transcription coupled repair (TCR) eliminates the DNA damage in the transcribed strand of actively transcribing genes (). An autosomal recessive disorder, Xeroderma Pigmentosum (XP) exhibits impaired NER activity. XP patients are classified into seven groups (XP-A to -G), and the defects of corresponding seven genes ( to ) are responsible for the missing NER activity in these XP patients. It is becoming increasingly clear and accepted that NER in mammalian cells is mediated by the sequential assembly of repair proteins at the site of the DNA lesion, rather than by the action of a pre-assembled repairosome (). XPC–hHR23B complex is most likely the initial damage recognition factor when the lesions are situated in the transcriptionally inactive genome or non-transcribed strand of actively transcribed genes (,), whereas XPA–RPA serves an equally critical function in verifying the presence of the DNA lesion (). In addition, due to its high affinity for UV-damaged DNA, the damaged DNA binding protein (DDB) complex has also been implicated in the damage recognition step of GGR. DDB is a heterodimer of DDB1 and DDB2 components. Studies on the role of DDB2 in NER have raised some concerns (). Nevertheless, accumulating evidence has confirmed that DDB2 is undeniably involved in GGR. For example, several studies have shown that the cells from some XP-E patients or DDB2-deficient Chinese hamster V79 cells have a partial deficiency in NER (). Microinjection of the purified DDB complex into XP-E cells reversed the NER defect (). Since NER can be reconstituted with purified components and damaged DNA in the absence of DDB (,), DDB is believed to be relevant only to the NER within the chromatin context. Our previous studies as well as work of other laboratories have clearly shown that DDB2 is a key factor in regulating GGR of CPD, most likely through the recruitment of XPC to the DNA damage sites (). XPC is a 940-amino acid protein, and harbors domains that can bind to damaged DNA and repair factors, e.g. hHR23B, XPB and Centrin 2 (). XPC always exists in a bound form with hHR23B and Centrin 2 in cells. This protein complex actively participates in the process of NER (,,). Although hHR23B contains two ubiquitin-associated domains and one ubiquitin-like domain, it can stabilize XPC and enhance the binding between XPC and damaged DNA (). XPC protein can be modified upon UV irradiation, the modifications include ubiquitylation and sumoylation (,). Interestingly, the ubiquitylation of XPC does not lead to its degradation, but increases the binding of XPC to damaged DNA (). While the role of sumoylated XPC is still unclear, it was speculated to protect XPC from degradation (). DDB2 is required for the UV-induced XPC modifications. Among the modifications, the UV-induced XPC ubiquitylation is regulated by DDB-Cul4A E3 ubiquitin ligase complex comprised of DDB1, DDB2, Cul4A, Roc1 and COP9 signalosome (). DDB–Cul4A complex can ubiquitylate both DDB2 and XPC, but the fates of ubiquitylated DDB2 and XPC appear to be quite different, ubiquitylated DDB2, but not XPC, is subjected to proteasomal degradation (). XPC expression can be induced following UV irradiation through transcriptional activation (). Furthermore, overexpressed exogenous XPC has been found to be intrinsically unstable and is degraded by the proteasome (). Nevertheless, the association with hHR23B protein partly stabilizes these XPC . Similarly, tagged Rad4, the homolog of XPC in yeast, is found to be actively degraded by the 26S proteasome, and the turnover is protected by Rad23 protein (,). However, the endogenous mouse XPC protein is shown to be stable, with a half life of over 6 h (), and the endogenous Rad4 in yeast is also stable in the absence of UV light, but is degraded following UV irradiation (). Our previous study suggested that human XPC undergo degradation following UV irradiation (), and this degradation precedes the XPC induction observed later in the process. In this report, we have demonstrated that XPC is indeed degraded by 26S proteasome upon UV irradiation, and this degradation is independent of ubiquitylation. Furthermore, we provide evidence showing that the subunits of DDB–Cul4A complex differentially affect the UV-induced XPC degradation. Additionally, we have found that K655 residue of XPC protein is intimately involved in the UV-induced modifications as well as degradation of XPC. Elimination of UV-induced XPC degradation impairs the efficient NER of CPD through an effect on the recruitment of XPG to damaged DNA sites. Normal human fibroblasts OSU-2 cells, established and maintained in culture as described earlier (), Li-Fraumeni Syndrome fibroblast strain designated 041 cells (kindly provided by Dr Michael Tainsky, MD Anderson Cancer center, Houston, TX, USA), XP-C (GM15983), HeLa cells with over-expressed FLAG and HA-tagged DDB2 (HeLa-DDB2 cells) (a gift of Dr Yoshihiro Nakatani, Dana-Farber Cancer Institute, Boston, MA, USA) were grown in DMEM supplemented with 10% fetal calf serum (FCS) and antibiotics. XP-A (GM04312) and XPA complemented cells (XP-A+XPA, GM15876), XP-F (GM08437) and XPF complemented cells (XP-F+XPF, kindly provided by Dr Gan Wang, Wayne State University, Detroit, MI, USA), XP-G (XP3BR-SV) and XPG complemented cells (XP-G+XPG, kindly provided by Dr Karlene Cimprich, Stanford University, Stanford, CA, USA) and XP-E (GM01389) cells were grown in MEM supplemented with 10% FCS and antibiotics. Mouse embryo fibroblast ts20 (thermosensitive for E1 ubiquitin-activating enzyme) and its parental cell line A31N (Kindly provided by Dr Harvey L. Ozer, UMDNJ-New Jersey Medical School) were cultured in 50% F-10 + 50% DMEM medium containing 10% FCS and antibiotics. HeLa cells with over-expressed FLAG-HA-DDB2 and V5-His-XPC (HeLa-DDB2-XPC cells) were generated in our lab and cultured in DMEM containing 500 μg/ml G418 (). All cells were cultured at 37°C in a humidified atmosphere of 5% CO except A31N and ts20 cells, which are maintained at 32°C. For overall UV exposure, the cells were washed with PBS, irradiated with varying UV doses and incubated in suitable medium for the desired time period. The irradiation was performed with a germicidal lamp at a dose rate of 0.8 J/m/s as measured by a Kettering model 65 radiometer (Cole Palmer Instrument Co., Vernon Hill, IL, USA). XPC-V5-His and DDB1-V5-His plasmids were generated in our lab. pXPC3 plasmid containing XPC with N-terminal 1–117 amino acids deletion (Δ1–117) was kindly provided by Dr Randy Legerski (The University of Texas MD Anderson Cancer Center, Houston, TX, USA). Cul4A-c-Myc plasmid was kindly provided by Dr Yue Xiong (University of North Carolina, Chapel Hill, NC, USA). DDB2-FLAG plasmid (kindly provided by Dr Gilbert Chu, Stanford University, Stanford, CA, USA) was used to generate point mutants R273H and K244E, and XPC-V5-His plasmid was used to generate point mutants K655A and K917A by QuikChange Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA, USA). The plasmids were transfected into cells either by FuGene 6 (Roche, Indianapolis, IN, USA) or Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacture's instruction. To generate stably transfected cell lines, G418 (500 μg/ml) was added to the medium for selection and resistant colonies confirmed by western blotting. The cells were trypsinized and washed once with PBS. The cell pellets were lysed by boiling for 10 min in a sample buffer (2% SDS, 10% glycerol, 10 mM DTT, 62 mM Tris–HCl pH 6.8, protease inhibitor cocktail). Protein samples were loaded on 8–16% Tris–Glycine gels (Invitrogen) and separated by PAGE. The proteins were then transferred to nitrocellulose membrane, blocked by 5% milk and immunoanalyzed. The antibodies used were, rabbit anti-XPC and rabbit anti-DDB2 (generated in our lab) (), rabbit anti-Cul4A and rabbit anti-DDB1 (a gift from Dr Yue Xiong), goat anti-Lamin B (Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse anti-hHR23B (BD Bioscience, San Jose, CA, USA), rabbit anti-V5 (Bethyl Laboratories, Montgomery, TX, USA) and mouse anti-c-Myc (Invitrogen). XP-C cells growing on glass coverslips were transfected with XPC-V5-His plasmid for 48 h. The cells were then washed with PBS and UV irradiated through an isopore polycarbonate filter (Millipore, Bedford, MA, USA), containing pores of a 5 μm in diameter, as described previously (). The cells were then double stained with rabbit anti-XPC and mouse anti-CPD (TDM-2, MBL International, Woburn, MA, USA), or rabbit anti-XPC and mouse anti-XPA (Lab Vision, Fremont, CA, USA), or rabbit anti-XPC and mouse anti-XPG (Lab Vision) or mouse anti-V5 (to visualize XPC, Invitrogen) and rabbit anti-XPB (Santa Cruz). Fluorescence images were obtained with a Nikon Fluorescence Microscope E80i (Nikon, Tokyo, Japan) fitted with appropriate filters for FITC and Texas Red. The digital images were then captured with a cooled CCD camera and processed with the help of its SPOT software (Diagnostic Instruments, Sterling Heights, MI, USA). GST and the fusion protein GST-hSug1 were expressed in strain DH5α transformed with either pGEX4T-1 or pGEX-hSug1 (kindly provided by Dr Andrew Paterson, The University of Alabama at Birmingham, Birminghan, AL, USA). After purification, GST and GST-hSug1 were separately incubated with glutathione Sepharose 4B beads (Amersham Bioscience, Uppsala, Sweden) at 4°C for 2 h in PBS. The nuclear extract from OSU-2 cells were prepared by incubating OSU-2 cells in nuclear extract (NE) buffer (20 mM HEPES, pH 7.9, 25% glycerol, 0.42 M NaCl, 1.5 mM MgCl, 0.2 mM EDTA, protease inhibitor cocktail) for 20 min and NE was collected by centrifugation. GST or GST-hSug1 bound beads were incubated with either NE, or purified recombinant XPC (a gift of Dr Yue Zou, East Tennessee State University, Johnson City, TN, USA) at 4°C for 2 h in NE buffer. After washing five times with NE buffer, the beads were boiled in 2×SDS loading buffer for 5 min and the supernatant was subjected to western blot analysis. Cul4A and DDB1 siRNA oligonucleotides were synthesized by Dharmacon (Lafayette, CO, USA) in a purified and annealed duplex form. The sequences targeting Cul4A and DDB1 were 5′-GAACAGCGAUCGUAAUCAAUU-3′ and 5′-UAACAUGAGAACUCUUGUC-3′, respectively. Specific and control siRNA transfections were performed with Lipofectamine 2000 (Invitrogen) according to manufacture's instruction. The amount of CPD in DNA was quantified with non-competitive immuno-slot blot assay. Briefly, XP-C cells in 100 mm plates were transiently co-transfected with DDB2 and either empty vector, wild type or K655A XPC mutant. Twenty-four hours post-transfection, cells were split into 60 mm plates and grown for an additional 24 h. After UV exposure (10 J/m) and desired incubation periods, cells were recovered by trypsinization and immediately lysed for DNA isolation. The identical amounts of DNA samples were loaded on nitrocellulose membranes and the amount of CPD was detected with monoclonal anti-CPD antibody (TDM-2). The intensity of each band was determined by laser densitometric scanning and the amount of damage remaining, compared with the initially induced DNA damage, was used to calculate the relative repair rates. Our previous studies have indicated that the level of XPC in cells decreases upon UV irradiation (). To further confirm this phenomenon, we compared the decay rates of XPC in UV- or mock-irradiated normal human fibroblast, OSU-2 cells pre-treated with cycloheximide (CHX), an inhibitor of protein synthesis. A shows that in the absence of new XPC synthesis, XPC protein exhibits a high decay rate following UV irradiation, and the distinct pattern of XPC degradation could be observed as early as 30 min after UV treatment (lanes 5–8). To rule out the possibility that the decreased XPC level at 125 kDa is due to the conversion of XPC to slower migrating modified forms, we over exposed the film to show the corresponding levels of various XPC bands. As shown in A, the modification of XPC is not fully obvious at 2 h time point and yet the level of XPC at 125 kDa is lower than that seen at 1 h time point. We then scanned all bands of XPC and quantified the total XPC amount and normalized by Lamin B level. As shown in A, total XPC amount did decrease with the elongation of incubation time following UV irradiation. This result indicates that the decrease of XPC observed after UV irradiation is most likely due to the protein degradation. Moreover, we treated the cells with proteasome inhibitor MG132 prior to UV irradiation to test the involvement of 26S proteasome in this UV-induced XPC degradation. We found that the UV-induced decrease of XPC levels is promptly inhibited in the presence of MG132 (B, lanes 1 and 2 versus 3 and 4). A similar result regarding the effect of MG132 on the immediate fate of XPC was also seen with repair-deficient XP-A cells (Supplementary Figure 1). These combined data indicate that UV irradiation causes XPC protein degradation proteasome-mediated proteolysis. Moreover, B showed that when protein degradation is inhibited by treatment with MG132, more XPC was detected in UV-treated cells than that in mock-treated cells (lane 3 versus 4). However, when we further inhibited protein synthesis by treatment with CHX, this UV-induced XPC increase did not occur (lane 7 versus 8). These results indicate that both degradation and induction of XPC protein occurs in tandem within UV irradiated cells. Previous studies have implied that UV-induced ubiquitylation of XPC is reversible and does not serve as a signal for degradation (). Therefore, we reasoned that UV-induced XPC degradation observed in our experiments might be independent of ubiquitylation. To test this hypothesis, we analyzed the UV-induced changes of XPC levels in mammalian cells capable of conditional inactivation of E1 enzyme. Ts20 cells growing at permissive 32°C, thus harboring normal E1 activity, exhibit a small extent of XPC degradation upon irradiation (C, lane 5 versus 6). However, when E1 is inactivated by transferring cultures to non-permissive 39°C (,), these cells showed considerable XPC degradation following irradiation (lane 7 versus 8). Exactly the same XPC degradation response is seen in parent control A31N cells at both permissive and non-permissive temperatures (C, lanes 1–4). In essence, the ubiquitylation defect failed to impinge on the protein degradation. These data clearly indicate that UV-induced XPC degradation is independent of ubiquitylation and suggest a direct interaction of XPC with proteasome. To substantiate this idea of direct ubiquitylation-independent interaction, the GST pull down assay was conducted with the whole cell lysates prepared from OSU-2 fibroblasts. We found that recombinant hSug1, a subunit of 19S proteasome, physically binds to XPC protein (D). The interaction was further tested with purified recombinant XPC and hSug1, and the result shows that XPC protein can bind to hSug1 directly (E). Taken together, we believe that UV-induced XPC degradation is independent of ubiquitylation and that XPC can bind to 26S proteasome through direct interaction with hSug1. To explore the relationship between XPC degradation and the NER process, we examined the kinetics of UV-induced XPC degradation in normal human fibroblast as well as human cell lines belonging to different XP complementation groups and the corresponding cell lines corrected for the cognate repair deficiency. As shown in A, normal human fibroblast, OSU-2 cells, showed a significant decrease in XPC levels at 1 h, followed by an increase until it again reached the control levels at ∼8 h after UV irradiation. Meanwhile, all three XP-A, XP-F and XP-G cell lines exhibit the typical XPC degradation upon UV irradiation (B–D), indicating that UV-induced XPC degradation is not affected by the absence of any of these essential repair factors and is independent of the productive cellular excision repair process. In addition, XP-A and XP-F cells exhibit a similar XPC dynamics as that of repair-proficient OSU-2 cells, characterized by a prompt decrease at 1 and 2 h followed by restoration beginning at 4 h following UV irradiation (B and D, lanes 1–5). On the contrary, XP-G cells demonstrated continued XPC degradation without any detectable recovery of XPC at later intervals (C, lanes 1–5). Nevertheless, ectopic expression of XPG in XP-G cells was able to restore the normal XPC dynamics (C, lanes 6–10), indicating that XPG is required for the recovery of XPC protein following repair of UV damage in fully repair-competent cells. Interestingly, the ectopic expression of XPF in XP-F cells prevented the expected XPC decrease observed upon UV treatment (D, lanes 6–10). Since the UV-induced XPC degradation is easily seen in CHX-treated XPF-corrected repair-proficient XP-F cells (Supplementary Figure 2), we conclude that the XPF protein does not interfere with the XPC degradation. Nonetheless, XPF protein could additionally be stimulating the new synthesis of XPC which is also inducible upon UV irradiation. The requirement of DDB2 protein for the UV-induced XPC modifications (ubiquitylation and sumoylation) has previously been reported by our laboratory and others (,). Here, we extend this work by investigating the role of DDB2 in UV-induced XPC degradation. We approached this question by first following the post-irradiation fate of XPC protein in experiments with DDB2-deficient XP-E cells. The results clearly show that UV-induced XPC degradation fails to occur in cells lacking DDB2 (A). Since XP-E cells posed difficulty in transfecting cDNA constructs, we used another DDB2-deficient Chinese Hamster V79 cell line to further observe the effect of restoring DDB2 into these cells on UV-induced XPC degradation. As expected, DDB2 expression restored the XPC degradation following UV irradiation (B). This DDB2-mediated response was more clearly demonstrable in another cell line, 041, that lacks the DDB2 because of the absence of p53 inducer. As shown in C, XPC remains fully intact upon UV irradiation of these cells. However, transient transfection of DDB2 cDNA into these cells restored the normal UV-induced XPC degradation with a distinct dose-response relationship, i.e. greater XPC degradation with higher DDB2 expression. Finally, we tested whether the expression of mutated DDB2 can functionally substitute for the wild-type DDB2. Two XP-E mutations are single amino acid substitutions (K244E and R273H) corresponding to XP-E patients XP82TO and the related individuals XP2RO and XP3RO, respectively (). Extracts from cells of these lines are defective in the ability to bind UV-irradiated DNA fragments (). These two naturally occurring mutants of DDB2, R273H and K244E, along with wild-type DDB2, were separately and stably transfected into 041 cells and evaluated for the fate of XPC. As expected, only the wild-type DDB2 promotes UV-induced XPC degradation (D), which unambiguously indicates that the damaged DNA binding activity of DDB2 is a strict requirement for it to participate in the XPC degradation. DDB-Cul4A E3 ligase is believed to be functionally essential for the XPC ubiquitylation upon UV irradiation (). Furthermore, our results indicate that DDB2, as one of the subunits of this same E3 ligase, is also required for UV-induced XPC degradation. However, the role of other subunits of this E3 ligase in this important cellular process is not known. In order to address this question, we utilized a siRNA-based gene silencing strategy to squelch the activity of individual complex components within cells. As shown in A and B, knocking down the expression of DDB1 or Cul4A in normal human fibroblasts caused an expected inhibition of the UV-induced DDB2 degradation. However, the absence of DDB1 or Cul4A clearly enhanced the XPC degradation. Interestingly, there was a simultaneous reduction in the UV-induced XPC modifications. These results indicate that DDB1 and Cul4A are required for the stabilization of XPC through an influence on its protein modifications. To further confirm this finding, we tested if over-expression of DDB1 and Cul4A can protect XPC from degradation in another cell line, i.e. HeLa-DDB2 cells. It is worthy to note that HeLa cells have both DDB1 and DDB2 () and UV–induced XPC modification and degradation in HeLa cells are similar to those of OSU-2 cells [() and unpublished data]. We over-expressed either V5-tagged DDB1, or c-Myc-tagged Cul4A or both DDB1 and Cul4A in HeLa-DDB2 cells. All transfections involving the over-expression of Cul4A promoted the degradation of DDB2. Consistent with a previous report (), this indicates the normal function of ectopically expressed Cul4A. Interestingly, over-expression of either DDB1 or Cul4A or of both DDB1 and Cul4A components in cells dramatically inhibit UV-induced XPC degradation (C). Taken together, these data indicate that both DDB1 and Cul4A can protect XPC from being degraded upon UV irradiation and this effect is mainly through allowing the modifications of XPC protein. Our previously published studies indicated that UV-induced sumoylation of XPC inhibits its degradation following UV irradiation (), suggesting that sumoylation site in XPC may be involved in XPC degradation. In order to address this question, we needed to determine and manipulate the potential sumoylation sites in XPC protein. The linkage between SUMO and its target proteins occurs through an isopeptide bond between the C-terminal carboxyl group of SUMO and the ɛ-amino group of a lysine residue in the substrate. The majority of the sumoylation sites follow a consensus motif with ψ-K-X-E or ψ-K-X-E/D (), where ψ is a large hydrophobic amino acid, generally isoleucine, leucine or valine; K is the lysine residue that is modified; X is any residue and D or E is an acidic residue. This motif is bound directly by Ubc9, the sole SUMO–conjugating enzyme. We used SUMOplot () to predict the putative sumoylation sites in XPC protein. SUMOplot provides the probability of the SUMO consensus sequence (SUMO-CS) potentially engaged in SUMO attachment. The SUMOplot analysis revealed six putative sumoylation sites in XPC protein, e.g. K81, K89, K113, K183, K655 and K917 (A). In order to assess the valid sumoylation site in XPC, we either mutated the putative lysine to alanine (K655A and K917A), or used an existing 1–117 amino acids deletion XPC construct (pXPC3, Δ1–117) () to experimentally test the possible sumoylation-specific lysine. Since immortalized XP-C (GM15983) cells exhibited reduced DDB2 level, possibly due to disrupted p53 via SV40 large T antigen (data not shown), the XPC constructs were transiently co-transfected with DDB2 into these XP-C cells and the cells were UV irradiated at 20 J/m followed by 1 h for repair. The modified forms of XPC protein were detected by western blot analysis. As shown in B, mutation of K655 to alanine (K655A) produced a XPC that was unable to undergo modifications (lanes 5, 6), whereas mutation of K917 to alanine (K917A) and deletion of 1–117 amino acids had no effect on the protein's modification competence (lanes 7–10). This result indicates that K655 is the site responsible for sumoylation and other modification of XPC protein. Furthermore, we also tested the UV-induced degradation prowess of various XPC forms. Upon UV irradiation, ectopically expressed wild-type XPC could be degraded to the same extent as endogenous XPC, whereas K655A XPC does not undergo any degradation (C, lanes 1 and 2 versus 3 and 4). In contrast, other mutations such as K917A and Δ1-117 did not affect UV-induced XPC degradation (lanes 5–8), suggesting once again that K655 is also an essential residue for the XPC degradation. Since K655A mutation abrogates UV-induced XPC degradation, we used this construct to study the function of XPC degradation in NER following UV irradiation. XPC-Wt or XPC-K655A constructs were transiently co-transfected with DDB2 into XP-C cells, and the characteristics of XPC, e.g. its binding to hHR23B and its recruitment to damaged DNA sites were evaluated. The result indicates that K655A mutation does not affect the complex forming ability of XPC and hHR23B (data not shown). In addition, both XPC-Wt and XPC-K655A could be recruited to CPD sites upon UV irradiation (A). The recruitment of other NER factors, which are placed into the repair complex subsequent to XPC, was also analyzed. TFIIH (XPB) and XPA exhibit the normal recruitment to the UV-damage sites in both XPC-Wt and XPC-K655A expressing cells (B and C). On the other hand, XPG protein, while recruited as normal to the damage sites in XPC-Wt expressing cells, was severely impaired in its damage site recruitment in XPC-K655A transfected cells (D). These results indicate that K655A mutation-induced abrogation of XPC degradation hampers the recruitment of XPG to the damage sites. We also evaluated the effect of XPC-K655A mutation on the efficiency of NER. XP-C cells with transiently expressed XPC-Wt or XPC-K655A were UV irradiated at 10 J/m and allowed to repair for a 24 h period. The CPD remaining in DNA were quantified and the repair rates compared among different cell types. E shows that the expression of XPC-Wt and XPC-K655 is comparable in two transfected cell lines. As expected, CPD were not repaired in XP-C cells transfected with vector alone (F). Moreover, the transfection of XPC-K655A was unable to restore the DNA repair ability of XPC cells like that achieved with the XPC-Wt construct. These data suggests that inhibition of XPC degradation by K655A mutation severely affects the function of XPC in NER. The alterations of XPC levels in cells irradiated with UV have been reported either as no change (), or as an increase (,,). Our previous work, however, detected a decrease in XPC level immediately upon UV irradiation (). Similarly, a study in also demonstrated that Rad4 is degraded upon UV irradiation (). In the present study, we carried out an in-depth mechanistic investigation of the fate of XPC and confirm that the observed decrease of XPC following UV irradiation is a result of active XPC degradation. UV-induced XPC degradation occurs very early and can be seen for more than 2 h. In the meantime, as reported by other groups, XPC expression is also induced so that the new synthesis of XPC becomes an overwhelming event after 4 h and masks the decrease of XPC level invoked earlier. At this point, the cumulative measurement of the dual opposing effects is reflected as a net increase. Importantly, we show that UV-induced XPC degradation is not triggered by the typical protein ubiquitylation process. The mechanistic studies reveal that 26S proteasome can directly bind XPC to affect its degradation. Ubiquitylation and sumoylation of XPC following UV irradiation of cells is already established (,), albeit the nature of the two independent modifications has not been fully resolved. The function of XPC ubiquitylation, which has also been studied extensively , is not for the purpose of its degradation, but to augment DNA binding of XPC. However, the function of XPC sumoylation has so far remained unclear. Since we have found that inhibition of XPC sumoylation increases UV-induced XPC degradation (), it can be surmised that at least one function of XPC sumoylation is to protect XPC from being destroyed. Therefore, XPC undergoes degradation and modifications simultaneously following UV irradiation and in essence the degradation of XPC is intimately regulated by modifications, i.e. more modifications resulting in lesser degradation. With regards irradiation-related XPC protein induction, our data argues that XPG is required for this process because, in the absence of XPG, the level of XPC does not increase following UV irradiation. In addition, the transfection of XPG into XP-G cells restores the XPC increase after 4 h of UV irradiation. Because XP-A and XP-F cells exhibit normal XPC degradation and induction kinetics, we can rule out the possibility that blocking of XPC induction is due to transcription inhibition from un-repaired lesions located in the transcribed strand of the gene. Therefore, XPG may be an important factor in DNA damage-induced XPC expression, and it would be enlightening to unravel the role of XPG in XPC production. The DDB–Cul4A complex is a new class of cullin-containing ubiquitin E3 ligases (). Previous studies have indicated that the DDB–Cul4A E3 ligase regulates the autoubiquitylation and proteolysis of DDB2 in response to DNA damage (,). In addition, DDB–Cul4A complex is also required for UV-induced ubiquitylation of XPC, but this modification does not serve as the signal for proteolysis. Nevertheless, our present study demonstrates that the subunits of this E3 complex, DDB2, DDB1 and Cul4A, also regulate UV-induced XPC degradation. These regulatory events, however, serve different functions. DDB2 is required and promotes XPC degradation upon UV irradiation, whereas DDB1 and Cul4A protect XPC from being degraded. DDB2 has been shown to be a critical factor in the removal of CPD, most likely by allowing the recruitment of XPC to the damage sites (,). For instance, in DDB2-deficient XP-E cells, XPC cannot be recruited to the damage sites and consequently XPC cannot be degraded. In addition, only the wild-type DDB2, but not its mutant forms, has the ability to trigger UV-induced XPC degradation. Since mutated DDB2 cannot bind to UV-damaged DNA, we propose that XPC degradation occurs at the damage sites, and the role of DDB2 in this event is to help promptly recruit XPC to UV lesions. DDB1 and Cul4A have been reported to be involved in the proteolysis of several proteins, such as DDB2 (), p27 () and CDT1 (). However, in this study, we demonstrate that DDB1 and Cul4A did not promote XPC degradation, but instead protect XPC from destruction by the proteasome. In addition, knocking down the expression of either DDB1 or Cul4A impairs UV-induced XPC modifications. In light of the earlier observation that Ubc9 knockdown impaired UV-induced XPC modification while promoting its degradation (), we conclude that both XPC ubiquitylation and sumoylation can prevent XPC degradation upon UV irradiation. The fact that XPC modifications as well as degradation involve the same lysine residue of the XPC protein reinforces this conclusion. In this study, we mutated the XPC K655 to alanine to understand the function of UV-induced XPC degradation in NER. Mutation at this site blocked both UV-induced XPC modifications as well as its degradation. As described above, XPC sumoylation is believed to inhibit XPC degradation while XPC ubiquitylation is shown to enhance the binding of XPC to damaged DNA as well as inhibit XPC degradation. It may be noted that ubiquitylation of XPC was not found to promote the dual incision in a reconstituted NER reaction with purified proteins (). Similarly, inhibition of XPC sumoylation, by knockdown of Ubc9 expression, did not affect the efficiency of NER (). Therefore, it can be reasoned that the observed effect of K655A mutation on DNA repair is a consequence of eliminating its ability to degrade XPC. It has already been reported that during assembly of NER factors, XPC–hHR23B and XPG cannot simultaneously exist in the repair complex and that the entry of XPG into the complex coincides with XPC–hHR23B leaving the complex (). In contrast, XPC–hHR23B and XPA–RPA complexes can simultaneously bind to distorting DNA lesions (). Here, we have provided evidence showing that XPC degradation is a prerequisite for XPG recruitment to the damage sites. If XPC cannot be degraded (as in the case of K655A mutation), the recruitment of XPG to the damage sites is obviously compromised and as a result impairs the efficiency of CPD repair. However, in XP-E cells lacking UV-DDB activity, NER of 6-4PP is almost normal (), even though UV-induced XPC degradation does not occur. This means that XPG can still be recruited to 6-4PP in the absence of XPC degradation. Structural analysis of DNA lesions has revealed that 6-4PP induces significant helix distortion (), including disruption of base pairing and this structural distortion accommodates all needed proteins to allow the required assembly of the repair machinery. Thus, it seems that XPC degradation is not necessary here to make space for incoming XPG. In contrast, the distortion induced by CPD is much less pronounced. The DNA helix distortion induced by CPD could be too subtle to render sufficient space for all the NER factors to simultaneously congregate at the damage site. In this case, XPC–hHR23B complex will have to leave the damage site and, therefore, after serving the damage recognition function XPC is degraded to make the space needed for XPG recruitment. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Recombination-based cloning was developed to address some of the limitations of conventional recombinant DNA techniques. In particular, using conventional methods, DNA sequences must be flanked by compatible restriction sites before they can be sub-cloned into recipient vectors. It was recognized that genome-scale projects would require more efficient cloning methods (). To address this problem, the strategy underlying all recombination-cloning systems is that DNA sequences—typically gene open reading frames—are first introduced into an entry vector to create master clones (,). Entry vectors are engineered to contain site-specific recombination sites that direct gene transfer from the master clone to a panel of expression plasmids. Recombinants are selected by appropriate genetic criteria, yielding constructs in which genes are juxtaposed to novel regulatory elements or fused to epitope or affinity tags. The first recombination cloning method to be devised was the Univector system, in which Cre recombinase () is used to catalyze recombination between a P site on a master clone and a P site on an expression vector, fusing them to form a co-integrate plasmid molecule (). It was also shown that a second recombination event could be employed to resolve plasmid fusions such that only sequences placed between two recombinase sites would be donated to the final recombinant. Subsequent cloning systems have featured this dual recombinase site strategy to effect precise gene transfer. For example, the Creator system offered by Clontech utilizes recombination between two P sites on the master clone and one P site on the expression vector to transfer gene sequences (), while the Gateway system from InVitrogen employs two variant versions of lambda recombinase sites (,). One disadvantage of the Creator and Gateway systems as they are currently used is that both the initial capture of gene sequences and their subsequent transfer to expression vectors is catalyzed by purified recombinases, adding to the cost of these methods. To circumvent this, there has been an interest in developing cloning methods that can be performed directly in bacterial hosts. For example, a cloning system called MAGIC has recently been described that obviates the need for site-specific recombinases entirely, relying instead on endonuclease-targeted bacterial recombination to mobilize gene sequences (). It can be anticipated that further development of cloning methods will improve the speed and efficiency of genome analysis. These methodological advances are important, because biologists are increasingly turning to recombination cloning as a means to systematically process and examine the function of the large numbers of genes. As a testament to this, multiple projects are underway to transfer complete sets of open reading frames from different organisms into recombination vectors to facilitate this experimentation (). We initiated the work in this report because, rather than just transferring genes to different expression plasmids, we wished to use recombineering to analyze large collections of gene mutations. Advances in functional genomics have opened up the possibility of using genome-wide RNAi or chemical genetics screens to assess null phenotypes for novel genes (,). In many cases, however, it would be beneficial to produce and analyze additional types of mutations, either for a more comprehensive genetic analysis or for other types of experiments. Towards this end, we describe a method for constructing large randomly mutagenized gene libraries in recombination entry vectors, and show this approach can be used effectively in conjunction with recombination cloning to identify allelic variants with novel phenotypic traits. In developing this method, it was necessary to establish conditions under which linear DNA molecules flanked by P could be stabilized for site-specific recombination . As an additional application, we show that the ability to stabilize linear DNAs for Cre/ recombination could be exploited as a generalized cloning system. strain BW23474 () [Δ () Δ] was used to propagate plasmids containing the R6Kγ origin. DH5α [ Δ] was used for other plasmid manipulations. BUN10 () [] and AK1 [] were used to examine Univector re-circularization in mutants. AK1 was derived from JC6783 () by disrupting with a cassette. To construct pAK047, a HindIII (filled with T4 DNA polymerase)-MscI fragment encoding Tc from pBR322 was cloned into XhoI treated (filled) pUNI-10 to form pAK004. pAK004 was treated with NdeI and KpnI, releasing a 1484 bp fragment containing a P site at the 3′ end of . This fragment was blunted with T4 DNA polymerase, ligated to itself, and fused with pUNI-10 in an Cre/ UPS reaction (). The resulting recombinant, pAK005, contains flanked by P. The Kn region of pAK005 was replaced with by cloning a SmaI/SacI fragment from pFA6a/kanMX4 () into SacI/PvuII treated pAK005, yielding pAK046. pAK046 was PCR amplified with oligonucleotides 5′-AGCAGATCAGGGATTCACCACTCCAAGAATTGGAGC and 5′-TGCATGGCATAACCAAGCCTATGCCTACAGCATCC, removing the majority of and introducing I-SceI sites (underlined). The fragment was treated with I-SceI and re-ligated to form pAK047. To construct pJBN260, a 435 bp NotI-MscI fragment from pAK047 was cloned into HpaI/PspOMI-treated pDONR221. pJBN250 was constructed by PCR amplifying the region from lambda BstEII DNA standards (New England Biolabs) using oligos 5′-AGAAAGCTTTG TTGTGAGCGGATAACAATTTCACCATATACAT and 5′-GCTAGTGCTAGCGACGAAAGTGATTGCGCCTACCCG. The longer oligo anneals to the 5′ end of (bold), and includes promoter (underlined) and Shine–Delgarno (italicized) sequences. The HindIII and NheI sites incorporated in the oligos were used to clone the resulting fragment into HindIII/NheI-treated pQL269. Details of constructing Univector plasmids for and mutagenesis are available upon request. Error-prone PCR was performed essentially as described (). Approximately 5 ng of template DNA was added to a reaction containing 5 μl 10× Buffer B (10 mM Tris-HCl pH 9.0, 50 mM KCl final concentration), 5 μl 10 pmol/μl JB.45 (final 1 pmol/μl), 5 μl 10 pmol/μl JB.57 (final 1 pmol/μl), 3.5 μl 25 mM MgCl (final 1.75 mM), 0.25 μl of 25 mM MnCl (final 0.125 mM), 4.3 μl of 10 mg/ml BSA, 1 μl of 10 mM dNTPs, 1 μl 10 mM dTTP, 1 μl 10 mM dCTP (final 200 μM dATP, 200 μM dGTP, 400 μM dTTP and 400 μM dCTP), 1.5 μl of 5 U/μl concentration Taq DNA polymerase (New England Biolabs), adjusted with HO to a volume of 50 μl. For more biased nucleotide pools, the concentrations of dTTP and dCTP were adjusted to 600 μM or 1 mM, with 2.0 mM Mg/0.25 mM Mn and 2.5 mM Mg/0.5 mM Mn, respectively. Reactions were amplified using MJ Research PTC-100 thermal cyclers with an initial denaturation step of 2 min 92°C, followed by 35–40 cycles of 10 s 92°C, 1 min 30 s 65°C, 4 min at 72°C and a final 15 min extension at 72°C. JB.45: 5′-TTTCATACACGGTGCCTGACTGCG. JB.57: 5′-AACTGTGAATGCGCAAACCAACCC. To prepare competent cells, 5 ml cultures of BW23474/pJBN250 or DH5α/pJBN250 were incubated overnight in Luria–Bertani broth (LB) supplemented with spectinomycin (40 μg/ml). Overnight cultures were diluted into 500 ml Super Broth (16 g BactoTryptone, 10 g Yeast Extract, 5 g NaCl, 5 ml 1 N NaOH, 500 ml dHO) containing spectinomycin and 300 μM IPTG to induce expression of and . The cultures were then incubated at 37°C until they reached OD 1.0. Culture flasks were cooled in an ice water bath and cells recovered by centrifugation (6000 r.p.m., 5 min, 4°C). The cell pellet was washed sequentially at 4°C with 500 ml 1 mM HEPES, pH 7.0, 250 ml 1 mM HEPES, pH 7.0 and 10 ml 10% glycerol in distilled HO. The final pellet was re-suspended in 1 ml 10% glycerol, frozen in 50 μl aliquots in liquid nitrogen and stored at −80°C. We constructed mutagenized libraries by first precipitating PCR reactions (50 μl) with 500 μl of -butanol. The pellet (10 min, 13 000 r.p.m.) was washed with 70% ethanol and re-suspended in 10–20 μl distilled HO. 1–5 μl was electroporated into competent BW23474/pJBN250 cells. Cells were plated onto LB/kanamycin (50 μg/ml) media and incubated overnight at 42°C. For large libraries, pellets from 10 PCR reactions were pooled in 50 μl HO and 10 separate transformations were plated onto ten 15 cm diameter LB/kanamycin plates. Transformants were scraped into 25 ml LB/kanamycin media and inoculated into two liters of Super Broth supplemented with kanamycin. After a 4 h amplification at 37°C, libraries were prepared using Qiagen Maxi-Prep kits. Fusion libraries were prepared using either UPS reactions () or by co-transformation of DH5α/pJBN250 cells. In our hands, co-transformation typically yielded the largest number of transformants. Approximately 1 μg each of mutagenized library and expression plasmid were electroporated directly into competent DH5α/pJBN250 cells, and recombinants selected on LB/kanamycin plates at 42°C. We occasionally observed fusion libraries becoming contaminated with an apparent deletion form of correct recombinant plasmids. This variant retained the ColE1 origin, Ap and Kn regions, but removed expression plasmid sequences necessary to transform yeast hosts. To minimize this contamination, obviously faster growing colony regions were excised from transformation plates and libraries was prepared directly from recovered transformants without further amplification. Fusion libraries were analyzed in yeast strains derived from CRY1 [] using standard genetic techniques. Further yeast strain information is described in figure legends or is available upon request. For cloning experiments, 20 μl of mini-prep DNA (∼200–300 ng) was restricted to release antibiotic resistance markers flanked by P. The digests were combined with 5 μl of mini-prep DNA for circular target vectors, and DNA mixtures were concentrated by -butanol precipitation. The pellets were washed in 70% ethanol, re-suspended in 10 μl of distilled HO, and 2 μl was transformed into DH5α cells harboring pJBN250. The transformations were allowed to recover for 1 h and plated on antibiotic-containing media at 42°C. Cultures for mini-prep analysis were also cultured at 42°C. was obtained by digesting pJBN240 (a pRS413-derived yeast mini-chromosome harboring ) with DraIII and SacI. The cassette was derived by digesting pAK005 (an intermediate in constructing pAK047) with NotI and XhoI. xref fig #text This report describes a method for constructing randomly mutagenized gene libraries in a format compatible with recombination cloning. Gene sequences are cloned into recombination entry vectors, and the entire master clone is PCR amplified under mutagenic conditions, producing linear plasmid amplicons flanked by P. The mutagenized plasmids are re-circularized through Cre/ recombination by simply transforming them into Cre/Gam-expressing bacterial cells. The resulting libraries can then be processed by recombination cloning into desired expression formats in a similar fashion to other large collections of gene sequences. In the examples provided here, such libraries proved effective in conjunction with the Univector system to isolate mutant forms of the and genes, suggesting it should be possible to apply this approach to other sets of mutations. It is of course possible to construct PCR mutagenized libraries through a number of different approaches, some of which would also be compatible with library transfer by recombination cloning. For example, libraries could be constructed by directly cloning the mutagenized PCR fragments using conventional techniques or by using Topo cloning methods (In Vitrogen). Mutant libraries could also be constructed in Creator vectors using In-Fusion technology (In Vitrogen) or in Gateway constructs (Clontech) using B-tailed PCR products. In comparison, our library construction method has at least two advantages. First, as an cloning procedure, it does not require purchasing charged vectors or recombinase mixtures, and should therefore be comparatively cost-effective for assembling large numbers of mutant libraries. Second, and more importantly from a genetic standpoint, our method is highly efficient, allowing libraries of over a million recombinants to be constructed. The ability to construct such deep libraries should greatly facilitate the identification of rare, informative mutant alleles, especially when a direct selection for the desired mutant phenotypes can be applied. Although we have focused on the Univector system, our library construction method should also be compatible with other recombination cloning strategies. Towards this end, we have constructed a modified Gateway entry vector containing P recombinase sites and tandem P sites. Gene sequences can be introduced into this vector by B/P recombination, mutagenized using Cre/ re-circularization and transferred to Gateway destination plasmids (B). Our method should also be compatible with the MAGIC cloning system (although it would be necessary to remove the I-SceI site between the P repeats) (). In MAGIC, homing endonucleases are used to induce homologous recombination between 50 bp regions located on both recipient and donor vectors, effecting gap repair of the recipient plasmid with donor sequences. It was shown that PCR fragments flanked by these 50 bp regions could recombine with expression plasmids in MAGIC bacterial hosts (). Thus, it is possible to construct PCR-mutagenized libraries in any desired MAGIC expression plasmid. However, if the intent is to create mutagenized libraries in MAGIC entry vectors so that libraries can be transferred to a range of expression constructs, other strategies must be employed. Of necessity, endonuclease sites on MAGIC entry vectors are located outside the homology regions, preventing them from being gapped in a manner compatible with recombination. Therefore, it may be advantageous to use a method similar to ours to generate mutant libraries in MAGIC entry vectors. In addition to intramolecular recombination, our results indicate Gam-mediated RecBCD inhibition enables linear DNA fragments to recombine with P-containing plasmids through a bimolecular reaction. We have also recently found that it is possible to amplify DNA fragments using PCR primers with appended P sites, allowing us to flank specific DNA segments with P repeats (J.A.J. and J.B., unpublished data). In theory it should therefore be possible to perform our mutagenesis procedure on any desired entry clone without having to introduce a tandem P cassette. It should also be possible to utilize Cre/ recombination as a more general cloning procedure. What is principally required for this latter approach is an appropriately engineered capture vector that permits both positive selection for correct recombinants and counter-selection against un-recombined vector molecules. As we show here, the ability to replace genetic markers on the recipient plasmid should facilitate the design of such counter-selection strategies. Ultimately, a general recombination method for cloning linear DNA fragments is desirable, as it would provide an inexpensive alternative to some cloning techniques (such as capturing PCR products) currently performed using reagents. Our intent in developing this library construction technique was to utilize recombineering to expedite the screening of gene mutations. It is worth considering what conditions must be met to perform this type of genetic analysis. Assuming an appropriate expression vehicle is available, a first condition is that recipient cells must be transformed/transfected efficiently. Second, if a direct selection cannot be applied, it must be possible to propagate transformants in a clonal fashion, and a screening procedure devised to identify transformants exhibiting desired phenotypes. Finally, a method must exist to recover mutations of interest. As we show for , both the budding yeast and fission yeast genetic systems meet all these criteria, and it should therefore be possible to use our mutagenesis system to isolate informative mutations from yeast cells under a wide variety of screening conditions. Furthermore, the ability to maintain a permanent mutant library, apply it under different expression conditions, and recover a virtually unlimited number of transformants are significant advantages compared to current yeast gap repair/plasmid shuffle mutagenesis techniques (). Importantly, the use of recombination cloning in conjunction with these mutant libraries is also ideally suited for methodologies where yeast and bacteria are used as surrogate genetic systems, as in identifying interaction-defective alleles in reverse two-hybrid analysis () or proteins with altered properties in display or activity assays (). An ultimate goal would be to develop the procedures established here to a point where mutant libraries could also be productively screened in higher eukaryotes. There are reasons to believe that this should indeed be possible. Retroviral and adenoviral vectors provide extremely efficient gene delivery vehicles, and could presumably be engineered such that mutant libraries could be recombined into these vectors in a format allowing expression in mammalian cells. Cre treatment could then liberate master clones containing mutations of interest in a circular form suitable for recovery in bacteria. We also note that there has been significant progress in developing high-throughput screening methods for plant and animal cells (). Thus, it may prove possible to screen recombination-based mutant libraries in metazoan cells for dominant phenotypic traits. Such mutations, especially in a conditional expression format, would provide valuable reagents for assessing gene function.

Alzheimer's disease (AD) is a progressive age-dependent neurodegenerative disease that leads to cognitive and behavioral impairment. Recent studies show that tissue samples from AD patients have elevated levels of oxidative DNA damage (). A high level of DNA damage can be particularly deleterious in post-mitotic cells because they do not self-renew through cell proliferation. Therefore, oxidative base modifications in nuclear and mitochondrial DNA could lead to selective loss of damaged neurons and may play a significant role in aging and neurodegeneration in mammals (). At present, it is unclear how and why oxidative DNA damage increases in tissues of AD patients; it is also not known whether DNA repair and/or the response to DNA damage play significant roles in the pathogenesis of AD. Base excision repair (BER) is the primary DNA repair pathway for small base modifications such as alkylation, deamination and oxidation, and is thought to play a critical role during development and maintenance of the central nervous system (CNS) (). The first step of BER is the removal of the damaged base by a substrate-specific DNA glycosylase, generating an abasic (AP) site, which is cleaved by an AP lyase or AP endonuclease (i.e. APE1 in human cells). In the most common BER sub-pathway, known as short patch BER, the resulting one base gap is filled in by a DNA polymerase and ligated by a DNA ligase. If the 5′ terminal contains blocking groups, the DNA polymerase can add between 2 and 8 nt, with consequent strand displacement, flap processing and finally ligation. This pathway is known as long-patch BER. In humans, DNA polymerase beta is the major DNA polymerase in both sub-pathways (). Previous studies of BER in AD patients suggested possible changes in expression of BER enzymes. For example, expression of the mitochondrial β-8-oxoG DNA glycosylase (β-OGG1) was reduced in neuronal cytoplasm of affected AD tissue, and was associated with neurofibrillary tangles (NFT), dystrophic neuritis and reactive astrocytes (). Reduced expression of DNA polymerase β (pol β) was reported in midtemporal cortex samples from AD patients (); in contrast, expression of APE1 was higher in affected brain tissue () and in extracts of brain cells from AD patients (). The significance of these observations is not yet known. This study examines BER capacity in brain tissue from sporadic AD patients and normal age-matched controls. BER activities were also assessed in brain tissue from patients with amnestic mild cognitive impairment (MCI), a syndrome associated with a high risk for the development of dementia and AD (). The results indicate that AD is associated with a significant impairment of BER function. The BER impairment was not restricted to damaged brain regions and was also detected in the brains of amnestic MCI patients, where it correlated with NFT pathology, a hallmark of AD and related disorders (). All patients and controls in this study were longitudinally followed with annual neuropsychological testing and physical and neurological examinations. Some late stage AD patients were not tesee in the final phase of their disease. All controls had neuropsychological test scores in the normal range prior to death. The clinical diagnosis of amnestic MCI was made by consensus conference and followed the criteria of Petersen and Morris (). The clinical diagnosis of AD followed the standard accepted criteria (). All AD patients met the National Institute on Aging—Reagan Institute high likelihood guidelines for the neuropathological diagnosis of Alzheimer's disease () after histological and immunohistochemical evaluation of 30 different brain regions. Brain specimens used in this study were obtained from short post-mortem interval (PMI) autopsies of 10 AD (six males, four females), 9 amnestic MCI (two males, seven females) and 10 age-matched normal control subjects (six males, four female). Subject demographic data are shown in . We compared BER activities in affected and unaffected brain regions of AD and control subjects by examination of inferior parietal lobule (IPL) (affected) and cerebellum (CE) (least affected) regions of individual brains. Specimens of IPL and CE were flash frozen in liquid nitrogen at the time of autopsy. Immediately adjacent sections were fixed in 4% formaldehyde for routine histological and immunohistochemical studies. Human brain specimens were suspended in buffer (0.3 g/ml) containing 20 mM HEPES, pH 7.5, 50 mM KCl, 2 mM EGTA and Complete™ protease inhibitor (Roche Applied Sciences, Indianapolis, IN, USA). Tissues were homogenized using a Brinkman Polytron homogenizer for 20 s at setting 4. Lysates were centrifuged at 800 for 10 min to remove large cell debris. The resulting lysates were resuspended (2 mg/ml) in 20 mM HEPES (pH 7.0), 150 mM KCl, 2 mM EGTA, 1% (w/v) CHAPSO (Sigma), and protease inhibitor mixture and incubated at 4°C for 1 h with end-over-end rotation. The lysates were centrifuged at 100 000 for 1 h, and the supernatants were collected. The samples were flash frozen in liquid nitrogen and stored at −80°C. Protein concentration was determined using the Bio-Rad protein assay (Bio-Rad, Hercules, CA, USA). The sequences of the oligonucleotides used in this study are presented in . Oligonucleotides containing 8-oxodG, deoxy-uracil or tetrahydrofuran (THF) (Midland Certified Reagent Company, Midland, TX and Integrated DNA Technologies, Coralville, IA, USA) were 5′-P-labeled by incubating with [γ-P] ATP (PerkinElmer, Boston, MA, USA) in the presence of T4 polynucleotide kinase. Unincorporated free [γ-P] ATP was separated from the reaction mixtures using G25 desalting columns (GE Healthcare Corp., Piscataway, NJ, USA). The P-labeled oligonucleotides were then annealed to the complementary strands in the presence of 100 mM KCl by heating the samples at 90°C for 5 min and allowing them to slowly cool to room temperature. For gap-filling reaction and repair synthesis incorporation, unlabeled substrates were annealed as described above. 8-OxodG incision activity was measured using an oligonucleotide incision assay, as previously described (). The protein concentration of the lysates for all DNA glycosylase assays was adjusted with 20 mM HEPES–KOH (pH 7.4), 1 mM EDTA, 100 mM KCl, 25% glycerol (v/v), 0.015% Triton X-100, 5 mM DTT and protease inhibitors. Incision reactions (20 μl volume) contained 40 mM HEPES–KOH, 5 mM EDTA, 1 mM DTT, 75 mM KCl, 10% glycerol, 95 fmol of P-labeled duplex oligonucleotide. The reactions were incubated at 32°C for 17 h with 16 μg of tissue lysates. The reaction was terminated by the addition of 1 μl each of the following, 5 mg/ml Proteinase K and 10% SDS, and incubated at 55°C for 30 min. The DNA was ethanol-precipitated by the addition of 1 μg of glycogen, 4 μl of 11 M ammonium acetate, and 63 μl ethanol, pelleted, dried and suspended in formamide dye. The samples were resolved in a denaturing 20% polyacrylamide gel containing 7 M urea. After electrophoresis, the gels were visualized using a Molecular Dynamics Phosphoimager (GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA). The images were analyzed using ImageQuant 5.2 software (GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA). Incision activity was calculated as the amount of radioactivity in the band corresponding to the damage-specific cleavage product over the total radioactivity in the lane. Uracil incision activity was measured using a 30-mer oligonucleotide containing a single uracil at position 12 (). Incision reactions (20 μl) containing 70 mM HEPES–KOH (pH 7.4), 5 mM EDTA, 1 mM DTT, 75 mM NaCl, 10% glycerol and 5 μg of protein were incubated for 1 h at 37°C. The reactions were terminated and DNA processed as described for the measurement of activities of the other glycosylases. AP endonuclease 1 (APE1) incision activity was measured using a 28-mer oligonucleotide containing the abasic site analog THF at position 11 (). Samples were diluted in 10 mM HEPES–KOH (pH 7.4) containing 100 mM KCl. Reactions (10 μl) contained 25 mM HEPES–KOH (pH 7.4), 25 mM KCl, 0.1 mg/ml BSA, 5 mM MgCl, 10% glycerol, 0.05% Triton X-100 and 25 ng protein. Reactions were incubated for the indicated duration at 37°C and terminated by the addition of formamide dye and heating at 90°C for 10 min. Samples were resolved, visualized and analyzed as described for the measurement of DNA glycosylase activities. Pol β single nucleotide gap-filling activity was measured using a non-labeled 34-mer duplex oligonucleotide containing a single gap at position 16 (). Samples were diluted in 10 mM Tris–HCl (pH 7.4) containing 100 mM KCl. Reactions (10 μl) contained 50 mM Tris–HCl (pH 7.4), 50 mM KCl, 1 mM DTT, 5 mM MgCl, 5% glycerol, 5 μM dCTP (Roche Applied Sciences, Indianapolis, IN, USA), 1 pmol of duplex gap oligonucleotide, 4 μCi of αP-dCTP (GE Healthcare Corp., Piscataway, NJ, USA) and 1 μg protein. Reactions were incubated at 37°C for 1 h or the indicated duration and terminated by the addition of formamide dye and heating at 90°C for 10 min. Samples were resolved and visualized as described above. Repair synthesis reactions (10 μl) contained 40 mM HEPES (pH 7.6), 0.1 mM EDTA, 5 mM MgCl, 0.2 mg/ml BSA, 20 mM KCl, 1 mM DTT, 40 mM phosphocreatine, 100 μg/ml creatine phosphokinase, 2 mM ATP, 40 μM of each dATP, dTTP, dGTP and 4 μM of dCTP, 0.8 μCi αP-dCTP, 3% glycerol, 80 ng of double-strand U-containing oligonucleotide and 10 μg tissue lysate protein. The reactions were incubated at 37°C for 3 h and terminated by adding 2.5 μg of proteinase K and 0.5 μl of 10% SDS and incubating at 55°C for 30 min. The DNA was precipitated overnight at −20°C after addition of 1 μg glycogen, 4 μl of 11 M ammonium acetate, 60 μl of ethanol. Samples were centrifuged, dried, suspended in 10 μl of formamide loading dye. The gels were resolved and visualized as described earlier. BER activity was quantified as P-dCTP signal strength of the product band relative to control sample #1 (relative activity = 1), after subtracting the background of a reaction without protein. Proteins in tissue lysates (10–20 µg) were separated on 12% Novex® Tris-glycine gels (Invitrogen, Carlsbad, CA, USA) or 12.5% Criterion Tris–HCl gels (Bio-Rad, Hercules, CA, USA), blotted onto a PVDF membrane and blocked for 1 h at room temperature in 5% non-fat dry milk (Bio-Rad, Hercules, CA, USA) in TBST (20 mM Tris—HCl, pH 7.2, 137 mM NaCl, 0.1% Tween-20). Fresh milk-TBST was added with the primary antibody, which was one of the following: rabbit polyclonal anti-UDG (FL-313) (Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse monoclonal anti-human APE1 (Trevigen, Gaithersburg, MD, USA), mouse monoclonal anti-human pol β (Trevigen), rabbit polyclonal anti-beta tubulin (Abcam, Cambridge, MA, USA). Detection was performed with ECL+Plus® (GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA). Blots were quantified using ImageQuant 5.2 software. The results are reported as mean ± SD. Each assay was performed at least twice. Results from control sample #5 and AD sample # 7 in some assays were statistically defined as outliers based on box plots and were therefore excluded from all statistical analyses in this study. The differences among human control and AD or MCI samples were analyzed by the Student's -test, and a < 0.05 was considered statistically significant. Correlation coefficients were calculated using Pearson's correlation test. Trend analysis was performed by calculating the linear contrast using the SAS software version 9.1. To test whether BER activities are altered in human AD brain, BER assays were conducted using brain tissue from short post-mortem interval autopsies of 10 sporadic AD patients and 10 age-matched human controls (). The activity of DNA glycosylases was measured as the incision of a radiolabeled DNA oligonucleotide substrate containing a single lesion, either uracil or 8-oxodG. An oligonucleotide containing the abasic site analog THF was used to measure AP-site incision activity. Incision activity was calculated as the amount of radioactivity in the band corresponding to the damage-specific cleavage product over the total radioactivity in the lane. In the inferior parietal lobule (IPL), uracil incision activity was significantly lower ( = 0.022) in AD samples than in control samples (A). UDG protein level was also lower in AD samples than in control samples ( < 0.001) (D). 8-OxodG incision activity was also significantly lower ( = 0.010) in IPL from AD patients (B). Since the 8-oxodG DNA glycosylase (OGG1) is the main DNA glycosylase for this lesion in human tissues, these results suggest lower abundance or activity of OGG1 in this tissue from the AD patients. However, AP-site incision activity and protein levels were similar in IPL from AD patients and controls (C). Single nucleotide gap-filling capacity in brain tissue was analyzed as the incorporation of a radiolabeled dCTP nucleotide into a 34-mer double-strand substrate containing a single gap. Single nucleotide gap-filling activity was significantly lower ( = 0.006) in IPL from AD patients (E and F). Furthermore, pol β protein level was lower in IPL from AD patients than in control samples ( < 0.0001) (D). Uracil-initiated BER capacity in brain tissue was analyzed as the incorporation of a radiolabeled dCTP nucleotide into an unlabeled 30-mer double-strand substrate containing a U/G base pair. Because the data presented above indicate that pol β and UDG are reduced in IPL from AD patients, it was predicted that total BER capacity would also be reduced. Indeed, the amount of DNA repair synthesis in a uracil-containing double-stranded oligonucleotide () was significantly lower ( = 0.017) in IPL from AD patients than from controls (G and H). Total BER capacity positively correlated with UDG activity ( = 0.87 controls; = 0.92 AD) and pol β activity ( = 0.86 controls; = 0.94 AD) in control and AD lysates (A and B), supporting the idea that the lower BER in the AD lysates was caused by decreased activity of these two enzymes. Furthermore, total BER capacity in controls was inversely correlated with age ( = −0.89), whereas most AD patients had low levels of BER regardless of age ( = −0.25) (C). If lower BER activity is a sensitizing feature in AD rather than an underlying cause of the disease, we might expect similar alterations in BER activities in both affected and unaffected regions of brain from AD patients. This question was examined by comparing BER activities in cerebellum (CE) (least affected) and IPL (affected) regions of individual brains. The results indicated that similar changes in BER function occur in CE and IPL brain regions (A–G). In particular, uracil incision activity and UDG protein level were significantly reduced ( = 0.030 and = 0.032, respectively) in AD CE (A and D). Moreover, 8-oxoG incision activity was significantly reduced ( = 0.003) in AD samples (B). As for the IPL, AP incision activity and APE1 protein level was similar in CE from AD patients and controls (C). Additionally, AD CE lysates had significantly reduced ( = 0.025) single nucleotide gap-filling activity (E and F) and pol β protein level ( = 0.005) (D), although to a lesser extent than IPL. Consequently, total BER capacity was significantly reduced ( = 0.031) in CE from AD patients (G). As observed for BER assays in IPL, total BER capacity correlated positively with UDG ( = 0.72 controls; = 0.67 AD) and pol β ( = 0.31 controls; = 0.57 AD) (A and B) and tended to decrease with the age of controls ( = −0.29), but not with the age of AD patients (C). Recent reports showed increased oxidative DNA damage in leukocytes () and brain specimens () in subjects with amnestic mild cognitive impairment (MCI), a transition phase between normal aging and early dementia and the earliest clinically detectable phase of AD. This suggests that accumulation of DNA damage may be an early event in the progression of AD that could contribute to the pathogenesis of this disease. To test whether loss of BER function occurs in subjects with high risk of developing AD, BER activities were measured in IPL from nine amnestic MCI patients (), and compared to BER activities in AD patients and controls. The results showed a significant linear trend of decrease in uracil incision activity ( = 0.027) (A) and single-nucleotide gap-filling activity ( = 0.010) (C) with the severity of the clinical diagnosis. Median uracil incision (B) and gap-filling (D) activities were 26 and 23% lower in amnestic MCI samples than in controls. Although total BER capacity was not statistically significantly lower in samples from amnestic MCI patients (E), median total BER capacity was reduced by 62% (F). Uracil incision and gap-filling activities also correlated with total BER capacity in IPL from amnestic MCI patients (G). Amyloid β (Aβ) plaques and NFT are hallmarks of AD. Although there was no correlation between BER activities and the number of Aβ plaques in patients with AD or amnestic MCI (data not shown), BER activities were inversely correlated with Braak stage (), a measure of NFT abundance, in patients with amnestic MCI (H). The goal of this study was to determine whether BER dysfunction plays a role in susceptibility to or progression of AD. This question was addressed by measuring BER activities in brain specimens from patients with AD or normal controls. The results indicate that AD is associated with a significant impairment of general BER function. Our findings show that uracil incision activity and UDG protein levels were significantly lower in brains of AD patients than in controls. Uracil accumulates in DNA as a result of spontaneous deamination of cytosine (), generating a U:G mismatch; or incorporation of dUMP during replication (), which results in a U:A base pair. UDG activity decreases rapidly during neuronal development and remains at a low level in adult neurons (), suggesting that uracil might accumulate in DNA of adult neurons and contribute to neuronal aging (). Furthermore, a recent study reported that suppression of UDG expression induced apoptosis in cultured rat hippocampal neurons (), supporting a role for this enzyme in maintaining neuronal viability. Importantly, folic acid deficiency, which has been linked to increased susceptibility to AD (), promoted uracil misincorporation and hypomethylation of DNA in neurons and sensitized them to Aβ toxicity (). It also resulted in increased DNA damage and hippocampal neurodegeneration in APP transgenic mice (). This is consistent with the possibility that reduced uracil incision capacity could sensitize neurons to Aβ toxicity in the brains of AD patients. 8-OxodG incision activity is a primary function of OGG1, a bifunctional DNA glycosylase with a strong glycosylase activity but weak AP lyase activity (,). Lower OGG1 activity was previously observed in nuclear lysates from affected human AD brain regions using a sodium borohydride trapping assay (). This assay detects the covalent complex formed between the AP lyase activity of OGG1 and the abasic site intermediate, and thus measures only the robustness of the AP lyase activity. By employing a DNA cleavage assay we show here that 8-oxodG incision activity of OGG1 is lower in AD extracts independent from its limited AP-lyase activity. The finding that AP-site incision activity and APE1 protein levels were similar in brains of AD patients and controls differs from previous reports of increased APE1 expression in AD (,). However, only expression levels and not APE1 activity was reported in the previous studies. The activity is a more finite determination of function. However, there could be issues with different experimental protocols, such as post-mortem interval and tissue handling. Single nucleotide gap-filling activity and pol β protein level were also significantly reduced in brains of AD patients. DNA pol β protects cells against the cytotoxicity of oxidative DNA damage () and plays a role in genome maintenance in aging and carcinogenesis (). Importantly, mice lacking pol β display neonatal lethality with abnormal neurogenesis characterized by apoptotic cell death in the developing central and peripheral nervous systems, but not in other tissues (). A recent report () on reduced pol β protein levels in AD brains supports our observation. DNA pol β contributes two essential enzymatic activities to BER: a 5′-deoxyribose phosphate (dRP) lyase activity, necessary to remove the dRP intermediate generated by APE1 cleavage of the abasic site, and a nucleotidyl transferase activity that incorporates the correct nucleoside triphosphate in a template-dependent manner (). While we have not directly measured dRP-lyase activity in these samples, the observation of decreased gap filling indicates a likely defect in pol β-catalyzed DNA synthesis in brains of AD patients. Overall BER capacity ultimately determines the efficiency of repair of BER-specific lesions. Our results show that total uracil-initiated BER was significantly lower in brains of patients with AD. Moreover, the finding that total BER capacity correlated positively with UDG and pol β activities in control and AD brains supports the idea that the lower BER was caused by decreased activity of these two enzymes. Notably, total BER capacity was inversely correlated with age of controls, but not with age of AD patients. Instead, the low BER capacity associated with AD regardless of age suggests a premature aging phenotype. It is important to note that the BER defects reported here were not limited to neuropathologically affected regions of AD brains, but instead were apparent in IPL and CE of AD patients. This suggests that BER dysfunction is a general feature of AD brains. This observation also dissociates the reduced BER levels in the IPL from selective loss of neurons in this region, since there is no neuronal cell death in the cerebellum of AD patients. MCI is a syndrome defined as cognitive decline greater than the expected for an individual's age and education level but that does not interfere notably with activities of daily life (). Although some individuals with MCI remain stable or even return to normal over time, more than half progress to dementia within five years. The amnestic subtype of MCI, examined in the present study, has the highest risk of progression to AD. Interestingly, BER activities were reduced in brain tissue from patients with amnestic MCI, a condition also characterized by increased load of oxidative DNA damage (,). This suggests that BER dysfunction, and increased accumulation of oxidative DNA damage, could occur at the earliest stages of dementia and AD. Aβ plaques and NFT are hallmarks of AD. Thus, it is important to ask whether BER dysfunction is associated with these neuropathological features. Gabbita and colleagues () found no correlation between the number of oxidative DNA lesions in AD brain regions and the number of NFT or Aβ plaques. Similarly, BER dysfunction did not correlate with the number of Aβ plaques in this study. However, BER activities and NFT were inversely correlated with Braak stage (), a measure of NFT abundance, in brains of amnestic MCI patients. A similar pattern could not be observed in brains of AD patients because all AD patients in this study were classified in the highest Braak stage (VI). The possible heterogeneity of outcome of amnestic MCI patients supports the finding that the BER deficiency correlates with the NFT pathology. Moreover, since NFT pathology in AD is associated with cognitive decline (), our finding suggests a link between BER capacity and the degree of neurological impairment, as measured by Braak stage. The question of how BER deficiency is involved in the progression of AD has yet to be answered. One possibility is that lack of proficient BER sensitizes neurons to the deleterious effects of Aβ and NFT. It has been speculated that a cause for oxidative DNA damage in AD is the accumulation of Aβ itself. This hypothesis resulted from the observation that Aβ can directly generate hydrogen peroxide through iron and copper ion reduction (,). The combined effect of increased oxidative DNA damage and a significant deficiency in DNA repair could potentially lead to neuronal loss. This may also explain why although BER deficiency was detected in both affected and non-affected regions of AD brains, neuronal loss is limited to areas where Aβ plaques and NFT are present. In summary, this study demonstrates significant BER dysfunction in brains of AD patients, resulting from reduced UDG, OGG1 and pol β activities. Our findings that BER deficiencies were detected in both affected and non-affected brain regions of AD patients suggest that impairment of BER is a general feature of AD brains. We also show that BER activities in amnestic MCI patients inversely correlated with the severity of disease. Together, these findings suggest that defective BER may play an important role in the progression of AD. The results presented here may lead to better understanding of the molecular mechanisms involved in AD, and pave the way to the development of risk assessment tools as well as preventive drug therapy.

The process of linking genotype and phenotype plays a crucial role in understanding the biological processes that contribute to overall cellular, tissue and organism responses, particularly when under a disease state (,). The first and classic example was the discovery of the Huntington gene (), which enabled predictive tests for age of onset and severity of disease to be established. Since then researchers have discovered single-gene lesions for a large number of simple Mendelian traits. It has proved much more difficult, however, to discover genes underlying genetically complex traits which have continuous rather than discrete variation in the phenotype (), since continuous variation is generally a product of small contributions from multiple genes. Over 2000 Quantative Trait Loci (QTL) have been mapped in mice and rats, yet <1% of these have been characterized at the molecular level (). The DNA polymorphism(s) underlying a QTL may be in an exon, which subsequently changes the primary amino acid structure of the gene. In other cases the polymorphism may lie in a regulatory region, possibly several kilobases from the transcription start site, altering the regulation of gene expression or splicing. Tens to hundreds of genes may be under even well-defined QTL. It is therefore vital that the identification, prioritization and functional testing of the polymorphisms identified in relation to the Quantitative Trait gene (QTg) and phenotype are carried out systematically without bias introduced from prior assumptions about candidate genes (). With the advent of microarrays, researchers are able to directly examine the expression of all genes under a QTL and hence examine the effect of regulatory variation directly. This has made it possible to use expert knowledge of the pathways underlying the phenotype to identify a limited number of strong candidate genes (). The scale of data being generated by such high-throughput experiments has led some investigators to follow a hypothesis-driven approach (), where the triage and selection of candidate genes is based on some prior knowledge or assumption. For example de Buhr () selected candidate genes based on their known involvement in the immune response. Although these techniques for candidate gene identification can detect QTg, they run the risk of overlooking genes that have less obvious associations with the phenotype (). The complexity of multigenic traits can also lead to problems when attempting to identify the varied processes involved in the phenotype. For example numerous processes can be involved in the control of parasitic infection, including the ability of the host to kill the parasite, mounting an appropriate immune response, or control of host or parasite induced damage. By making selections based on prior assumptions of what processes may be involved, the pathways, and therefore genes, that may actually be involved in the phenotype can be overlooked or missed entirely. In order to investigate whether such bias could be found within current analysis techniques, we conducted a review of the literature on combined QTL and microarray analyses. This detailed review is available within the Supplementary Data (Supplementray Table 2). Results from this review enabled us to identify the specific issues facing the manual analysis of microarray and QTL data, including the selection of candidate genes and pathways. These are listed below: A further complication is that the use of methods for candidate gene identification are inherently difficult to replicate and are compounded by poor documentation of the methods used to generate and capture the data from such investigations in published literature (). An example is the widespread use of ‘link integration’ () in bioinformatics. This process of hyperlinking through any number of data resources further exacerbates the problem of capturing the methods used for obtaining results since it is often difficult to identify the essential data in the chain of hyperlinked resources. With an ever increasing number of institutes offering programmatic access to their resources in the form of web services (), however, experiments previously conducted manually can now be replaced by automated experiments, capable of processing a far greater volume of data in a systematic and explicit manner. The integration of web services into an automated analysis pipeline or workflow enables the replication of the original chain of processes used in the traditional manual analyses. This is accomplished by connecting the outputs from one such service into the input of another in a consecutive manner. By replicating the original investigation methods in the form of workflows, we are now able to pass data directly from one service to the next without the need for any interaction from researchers. This enables us to process the data in a much more efficient, reliable, un-biased and explicit manner. In this article we propose a methodology that revises the known pathways that intersect a QTL and those derived from a set of differentially expressed genes. This methodology has been implemented systematically through the use of web services and workflows and has been applied to a use case in the mouse, : resistance to African trypanosomiasis. For the purpose of implementing this systematic pathway-driven approach, we have adopted a service-based infrastructure coupled with workflow technology. We chose to use the Taverna workflow workbench () for the means of constructing these workflows. This software was chosen based on previous experience with this workbench within the Manchester based research group. Although this workbench offers many features for workflow construction, the workflows discussed in this article may be constructed using any of the well-established workflow workbenches currently available. This investigation looks at the extent to which workflows will be able to reduce the issues labelled (i) to (vi) listed above, with respect to the manual analysis of gene expression and QTL data. The expression of genes within their biological pathways contributes to the expression of an observed phenotype. By investigating links between genotype and phenotype at the level of biological pathways, it is possible to obtain a global view of the processes which may contribute to the expression of the phenotype (). Additionally, the explicit identification of responding pathways naturally leads to experimental verification in the laboratory. We therefore opted to analyse QTL and gene expression data not directly at the level of genes but at the level of pathways (a). Using this pathway-driven approach provides a driving force for functional discovery rather than gene discovery. In order to determine which genes reside in the QTL region of choice, the physical boundaries of the QTL need to be determined. Each gene is then subsequently annotated with its associated biological pathways, obtained from the KEGG pathway database (). The same process of pathway annotation is also carried out for the genes that are found to be differentially expressed in the microarray study of choice. These two sets of pathway data enable us to obtain a subset of common pathways that contain genes within the QTL region and genes that are differentially expressed in the microarray study. By identifying those pathways common to both microarray and QTL data, we are able to obtain a much richer model of the processes which may be influencing the expression of the phenotype. This process is summarized in b. One drawback of this approach, however, is the reliance on extant pathway annotations for genes identified in the QTL regions and from the microarray studies. However, by explicitly recording the workflow output from KEGG, we can, if required, identify genes that do not have pathways associated with them. For such an approach to be conducted systematically, any web resources used (including their parameters) should be stated explicitly. By passing data from one service to the next in a workflow, vast amounts of data can be analysed with little input required from the user, other than that of parameter configuration. This makes workflow technology an ideal tool for processing high-throughput data in a systematic and explicit manner. In order to determine the genes that lie within Tir1 QTL region, the position of flanking markers used in the original mapping studies were identified in mouse Ensembl () release 40. These were identified as D17Mit29 and D17Mit11 on chromosome 17 (), at 28 394 586 and 38 278 830 bp respectively, within version 40 of the Ensembl Mouse database (NCBI build 36). The position of D17Mit11 was estimated based on close proximity to the gene Crisp2 (Mouse Genome Identifier - MGI:98815). The implementation of the pathway-driven approach consisted of three Taverna workflows. The first workflow constructed, , was implemented to identify genes within a QTL region, and subsequently map them to pathways held in the KEGG pathway database. Lists of genes within a QTL were obtained from Ensembl, together with UniProt () and Entrez gene () identifiers, enabling them to be cross-referenced to KEGG gene and pathway identifiers. A fragment of this workflow can be seen in , which shows the mapping from QTL region (label A) to KEGG gene identifiers (label C). This workflow represents an automated version of the manual methods required to perform such a task, including the process of collating all information into single output files. Additional services were added to format data into the correct input/output style, these services have not been assigned labels in . Analysis of microarray liver sample data from C57BL/6 and A/J mice at Days 3 and 7 post-infection, identified 981 and 1331 probesets that were differentially expressed on the basis of a corrected -test with a -value <0.01 and a log fold change >0.5. We chose to focus on the early time points of Days 3 and 7 for this investigation. This was because the mouse strains used in microarray study showed a strong gene expression response to infection at the early time points in the microarray data compared to that of the later time points. The later time points from the microarray study were found to be the result of secondary effects on infection. Permissive criteria were used at this stage of data analysis in order to reduce the incidence of false negative results (which could result in missing one of the true QTg). Any true negative results would later be discarded on correlation of the pathways with the observed phenotype. These correlations of pathways with the observed phenotype are carried out through traditional and analyses on the candidate QTg. Mining the literature for involvement of the pathways and candidate genes being involved in the expression of the phenotype are also required for hypothesis verification. As a result of permissive criteria being employed, 2312 probesets were chosen for further analysis and annotation with their biological pathways. The second workflow, , provided annotation of microarray probeset identifiers. Ensembl gene identifiers associated with Affymetrix probesets were obtained from Ensembl. These genes were entered into the same annotation workflow as that used for qtl_pathway. The Entrez and UniProt database identifiers were used to map the Affymetrix probeset identifiers to KEGG pathways. Significant problems were encountered when attempting to cross-reference between database identifiers. This has proven to be a considerable barrier in bioinformatics involving distributed resources, including the naming conventions assigned to biological objects (). In an attempt to resolve this, we have provided a single and explicit methodology by which this cross-referencing was done. This methodology is captured within the workflows themselves. To obtain the pathways which both intersect the QTL region and are present in the gene expression data, we used a third workflow, named , to obtain a list of KEGG pathway descriptions. Each of the pathways returned from the workflow were investigated in turn. Lists of intersecting QTL and microarray pathway outputs used in this study are available as Supplementary Data. Details of gene sequencing methods, carried out for validating potential candidate QT genes, are also provided in the Supplementary Data (Supplementary Table 2 and sequencing_methods). Any additional information is available on request. Microarray data used in this investigation is available in ArrayExpress (E-MEXP-1190). shows the set of candidate pathways and QT genes that may be involved in the trypanosomiasis resistance phenotype. From this list it is clear that the complete list of 344 genes, initially identified in the Tir1 QTL region, has been narrowed down significantly to just 32 candidate QT genes. A number of genes identified from these results, are present in multiple pathways from the 87 pathways identified in total. In order to determine the role of each QTg, each of the pathways was associated with the gene expression data using the GenMapp software package (). Those pathways in which a high proportion of component genes showed differential expression following trypanosome challenge were prioritized for further analysis. One such pathway identified was the MapK signalling pathway. There are four genes from the MapK pathway within the Tir1 QTL: Daxx, TNF, Mapk13 and Map14. Of these, Daxx showed the strongest signal of differential expression at early time points. (in Supplementary Data) and TNF has already been shown to be a poor candidate QTg (). Daxx was therefore chosen as the primary candidate QTg, from this pathway, to investigate further. Daxx is widely reported to be an enhancer for apoptosis (,). It is also reported that susceptible mice infected with trypanosomiasis show an increase in apoptosis (). During the acute stage of trypanosome infection, a large number of leucocytes undergo apoptosis, as the immune response is re-modelled to control the infection (). This pathway is therefore an example of the pathway labelled A in a, with the candidate gene being directly related to the QTL region and known, through literature, to be involved in the phenotype. The identification of Daxx as a candidate for the Tir1 QTL gene prompted the re-sequencing of this gene in order to identify polymorphisms that might correlate with the phenotype. Out of the 17 polymorphisms identified, 3 were found to have allele distributions that show a relationship with the phenotype. Two of these three were located in the intronic or 5′ upstream region suggesting a possible affect on splicing or expression. The third mutation to associate with survival time was a three base deletion in exon 5, coding for an aspartic acid [in submission to dbSNP ()]. This deletion of one aspartic acid residue (D) in a poly-aspartic acid tract was identified in BALB/c and A/J whereas no deletion was found the tolerant strain C57BL/6J (). It has been previously noted that Daxx binds to the p53 protein via this acidic region (). The study by Zhao () showed that the deletion of this acidic region abolished the Daxx–p53 interaction. The protein p53 is reported to control cellular apoptosis (,). We therefore hypothesize that a mutation within this acidic region may alter the binding between Daxx and p53 in such a way as to result in a differing apoptosis phenotype between the mouse strains. In order to confirm this effect, however, further and studies are required. Further investigation is also required to confirm the presence of the Daxx gene within the bovine QTL region, and its role as a potential QTg. The use of a hypothesis-driven approach is essential for the construction of a scientifically sound investigation, however, the use of a data-driven approach should also be considered. This would allow the experimental data to evolve in parallel to a given hypothesis to form its own hypotheses regardless of any previous assumptions (), as shown by this case. This method can be used to either confirm or disprove any given hypotheses, compiled by a traditional hypothesis-driven analysis of the data. As such, we propose the use of a combined data and hypothesis-driven analyses of the experimental data. In using the Taverna workflow workbench, we are able to state the services used and the parameters chosen at execution time. Specifying the processes in which these services interact with one another in the native Taverna workflow language, Scufl (), enables researchers to re-use and re-run previously conducted experiments. An additional feature of the Taverna system is the capture of experimental provenance. The workflow parameters and individual results are captured in this execution log where the data obtained from previous analyses can be viewed. In this investigation, we have illustrated how the large-scale analysis of microarray gene expression and quantitative trait data, investigated at the level of biological pathways, enables links between genotype and phenotype to be successfully established. This was implemented systematically through the use of workflows. Our investigation confirmed that non-systematic manual examination of QTL and microarray data does introduce bias into the processes, particularly by discarding candidates in premature filtering of such large datasets. Analysis of the QTL and gene expression data collected under the Wellcome Trust Host-Pathogen project identified a candidate gene, Daxx, which is thought to be strongly associated with resistance to trypanosomiasis infection. The workflows developed in this project are freely available for re-use—either by ourselves or others in future analyses and have been integrated into the myExperiment () project which supports scientific collaboration, discovery and workflow re-use. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Real-time PCR has been increasingly utilized in molecular diagnosis of infectious and genetic diseases and in a wide range of cellular and molecular biology research (). Generation of target-dependent fluorescent signals as amplified products accumulate allows monitoring of reactions in a homogeneous format. Real-time PCR assays can be designed with a broad dynamic range and high sensitivity and specificity. Unlike traditional end-point detection methods, real-time PCR technology does not require post-amplification processing, and therefore is more adaptable to automation and promises greater ease-of-use and reduced risk of contamination (,). Signal generation in real-time PCR reactions is commonly achieved with double-stranded DNA (dsDNA) binding dyes (e.g. SYBR-Green I) or fluorophore-labeled sequence-specific nucleic acid probes (). Sequence-specific probes offer such advantages that they do not bind to background genomic sequences or to non-specific products generated from primer dimer formation during PCR. Depending upon the purpose of a real-time PCR assay, probe design for the target-of-interest often needs to be specifically tailored either to tolerate sequence variations for detection of diverse DNA or RNA sequences (,) or to differentiate between sequence polymorphisms (,). Several technologies have been developed for homogeneous detection of DNA and RNA sequences with nucleic acid probes, including TaqMan probes (,), molecular beacons (), adjacent FRET probes (), scorpion primers (,), Amplifluor primers (), LUX™ primers () and others (,), with the first two being the most frequently utilized. TaqMan probes are single-stranded oligonucleotides commonly labeled with a fluorophore and a quencher attached to the 5′ and 3′ ends of the molecule, respectively. They are designed to bind to template strands during the primer extension steps of PCR cycles where they are cleaved by the 5′ → 3′ exonuclease activity of the DNA polymerase, resulting in the release of the fluorophore and its physical separation from the quencher. Efficiency of fluorescence quenching is consequently reduced, leading to generation of fluorescent signal. The cumulative fluorescence intensity is proportional to the amount of specific product generated. Signal detection during TaqMan PCR typically takes place at the relatively high extension temperature. This can lead to elevated background signal from the probe, negatively impacting fluorescent signal gain (difference in pre- and post-amplification fluorescence). TaqMan technology has been developed as an effective genotyping tool where mismatch discrimination is required (,). However, when mismatch tolerance is the preferred design outcome, the high temperature required for signal generation increases susceptibility to single base mismatches, especially for shorter probes. Longer probes tolerate mismatches better but oftentimes result in increased background fluorescent signal (). Further design limitations arise from the influence of probe position, nucleotide sequence and labeling position on hydrolysis efficiency (,,). Molecular beacons are dual-labeled single-stranded oligonucleic acid probes containing a target-specific loop flanked by short complementary stem sequences at both ends (,). The formation of an intra-molecular stem (or hairpin) brings fluorophore and quencher into close proximity in the absence of target sequence, and as a result, fluorescence is quenched through a contact-mediated mechanism. Hybridization to the target sequence is energetically favored over formation of the hairpin structure. As a result, they undergo conformational rearrangements that force the fluorophore and quencher apart, resulting in an increase in fluorescent signal. Because the hairpin structure generates an alternative energy state that competes with binding between the target and the loop sequences, molecular beacons are destabilized by mismatches in the binding region to a greater extent than linear probes (). This property renders molecular beacons an ideal choice for single nucleotide polymorphism (SNP) detection. Although molecular beacon probes can be designed to enhance mismatch tolerance or to discriminate mismatches (), most optimizations involve changes in either the loop or stem sequence. For single-stranded oligonucleotide probes such as TaqMan probes and molecular beacons, nucleic acid sequence changes represent the primary option for design flexibility. Probe designs with a second oligonucleic acid strand as a competitive binding partner were previously developed to detect nucleic acid hybridization (). More recently, ‘displacement hybridization’ probes have also been introduced to homogeneous real-time PCR (,). These double-stranded linear probes consist of two complementary oligonucleotides, at least one of which acts as a probe for target sequences. The 5′ end of the target binding strand and the 3′ end of the second strand are labeled with a fluorophore and a quencher, respectively so that, in the absence of target sequences, the duplex DNA formation brings fluorophore and quencher together. Binding of target nucleic acids displaces the quencher strand and leads to an increase in fluorescent signal. These double-stranded probes composed of strands of equal length or with limited difference in length were found to exhibit suboptimal reaction kinetics especially for low quantities of target nucleic acid but were ideal for single mismatch differentiation (,). In order to meet the need for a simple and versatile probe design for homogeneous real-time PCR, we developed a novel partially double-stranded linear DNA probe system consisting of a fluorescent-labeled hybridization probe and a shorter complementary quenching oligo. The hybridization stability of the probe-quencher duplex, relative to that of the probe-target duplex was modulated by adjusting (i) the length of target-binding probe strand, (ii) the length of the shorter quencher oligo, (iii) the molar ratio between the two strands and (iv) signal detection temperature. The thermodynamic behavior of such probes can be used to predict their performance in a real-time PCR reaction when probe hybridization is the primary source for signal generation. As a result, pre-amplification signal, signal gain due to target binding, and extent of mismatch discrimination can be reliably controlled and optimized. The use of two DNA strands in the probe provides flexibility in optimization of assay performance. By independently modulating either strand, probes can be designed to maximize mismatch tolerance or discrimination as needed. The mismatch tolerance feature of this probe design was further demonstrated by detection and quantification of genetically diverse HIV-1 variants in the Abbott RealTie HIV-1 assay. A nested set of hybridization probes (HPs) of 20, 25, 31 and 43 nt in length (designated as HP20, HP25, HP31 and HP43) were synthesized with a fluorescein (FAM) label at the 5′ end and a Dabcyl (for HP20, HP25 and HP31) or a BHQ®1-dT (for HP43) at the 3′ end. The 3′ quencher label served two purposes: (i) to prevent unwanted extension during PCR, and (ii) to allow melting analysis of HPs and the hybridization between HPs and targets. Replacing 3′ quencher with a non-quenching moiety has no discernible impact on the performance of real-time PCR (data not shown). All the HPs begin at the same nt position in the target sequence (nt 4749 of HIV-1 subtype B reference genome, HXB2, accession number K03455). The nucleotide sequence for HP20 is: 5′ACAGCAGTACAAATGGCAGT 3′. Longer HPs extends from the 3′ end further into the target sequence. Quenching oligos (QOs) were synthesized with 12, 14 or 16 nt complementary to the 5′ sequence of HP (designated as QO12, QO14 or QO16) and were 3′ labeled with Dabcyl. When used with HP43, QO14 was labeled with BHQ1-dT. An HP and a QO were combined at various ratios (1:1 or 1:3) to form a partially double-stranded linear DNA probe (). In the absence of target, the formation of the HP:QO hybrid brings the quencher and the fluorophore into close proximity, efficiently quenching the fluorescent signal. In the presence of target, the HP preferentially hybridizes to target sequences and, as a result, quencher is separated from fluorophore resulting in an increase in fluorescence emission. Target DNA used in melting temperature measurements was a 48-mer oligonucleotide either with perfect complementarity to the HPs or containing 1, 2, 3 or 4 mismatches in the probe-binding region. Forward and reverse primers for reverse transcription and PCR (RT-PCR) were designed to be upstream and downstream of the probe-binding region (). Oligonucleotides were synthesized by Sigma-Genosys (The Woodlands, TX, USA), TriLink Biotechnologies, Inc. (San Diego, CA, USA) or Abbott Molecular Inc. (Des Plaines, IL, USA). A series of probe designs were created by combining an HP (81 nM for HP20, HP25 and HP31 or 200 nM for HP43) and a QO of variable lengths and molar ratios. Thermal stability of the probes was characterized in melting experiments where fluorescence emission was measured at temperatures ranging from 85 to 10°C. Melting temperature (θ) is defined as the characteristic temperature where HP:QO duplex dissociates or where single-stranded HP undergoes spontaneous and transient uncoiling. was determined as the temperature that corresponds to the maximal absolute first derivative of fluorescent signals, |d/d ( = fluorescence and = temperature). The results of the θ calculation approximate those obtained with the method by Bonnet . (). Binding stability between HPs and targets was also assessed in melting experiments by including target DNA at five times the concentration of HP in the reaction mix. The impact of target mismatches within the probe-binding site was further analyzed by comparing melting profile and θ for a target of perfect complementarity to target sequences with one or more mismatches. Each melting profile was measured in a 96-well PCR plate in a 100 μl reaction containing PCR buffer (potassium acetate-bicine buffer with the same composition as the commercially available and commonly used EZ buffer), MnCl, probes and target DNA (in the case of HP:Target analysis). Thermal cycling was performed in an Mx4000™ multiplex quantitative PCR system (Stratagene, La Jolla, CA, USA) with the following cycling conditions: 1 cycle of denaturation at 95°C for 3 min; 75 cycles of 1 min holding at a range of temperatures from 85 to 10°C with a 1°C decrement per cycle. Fluorescence measurements were recorded during each 1 min hold of the 75 cycles. Targets were prepared by transcription (T7-MEGAscript, Ambion Inc., Austin, TX, USA) of HIV-1 subtype B and non-subtype B sequences cloned into pSP73 vector (Promega, Madison, WI, USA). RNA product was purified with Chroma Spin-100 columns (Clontech, Palo Alto, CA, USA) and the RNA concentration was determined by UV spectrophotometry at 260 nm. The PCR target region corresponds to nt positions 4650–4821 of the HXB2 genome. The sequences for forward and reverse primers are 5′ ATTCCCTACAATCCCCAAAGTCAAGGAGT 3′ and 5′ CCCCTGCACTGTACCCCCCAATCCC 3′, respectively. PCR reactions consist of the same components as in melting reactions together with primers, dNTPs and enzyme. In a real-time RT-PCR reaction, RNA was reverse transcribed into cDNA at 59°C and amplified by PCR in a homogeneous reaction with r DNA Polymerase (Roche Molecular Systems, Inc., Somerville, New Jersey, USA). Amplification reactions were performed in an Applied Biosystems 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) or ABI Prism® 7000 Sequence Detection System (Applied Biosystems) with the following cycling conditions: 1 cycle at 59°C 30 min; 4 cycles at 95°C 40 s and 46°C 30 s; 6 cycles at 92°C 30 s and 60°C 30 s; and 37 cycles at 92°C 30 s, 56°C 30 s and 35°C 40 s. Fluorescence measurements were recorded during the read step (35°C) of the 37 cycles. For the detection temperature study, the 37 read steps were also performed at 56°C. The impact of mismatches on real-time RT-PCR was assessed by comparing signal strengths and threshold cycle (C) values between same amount of mismatch and perfect-match targets. Melting profiles were determined for various probe compositions by monitoring fluorescence at temperatures ranging from 85 to 10°C. Melting experiments were performed in the same buffer and with the same divalent cation concentration as used in PCR reactions. Since HP was labeled with a quenching moiety at the 3′ end to prevent extension, fluorescence without a QO was measured to assess the degree of spontaneous quenching. Without a QO, fluorescence of HP31 was not completely quenched even at 10°C, indicating that spontaneous quenching due to interaction between two labels at the termini is an inefficient process (A). The θ was estimated at 20°C for HP31 and at 32°C for HP20 (A, B, ). Thus, quenching efficiency was inversely correlated with HP length in the absence of QO. Addition of QO16 to HP31 (1:1 ratio) resulted in a substantial increase in the efficiency of quenching (A). Similarly, addition of QO12 to HP20 at a ratio of 1:1 resulted in efficient quenching (θ = 39°C) (B, ). An increase in QO length from 12 nt to 14 or 16 nt resulted in an increased θ from 39°C to 46°C and 52°C, respectively (B, D, ). Increasing QO concentration resulted in an effect similar to that of increasing its length, although to a much lesser degree. For example, adjusting the molar ratio of HP20:QO16 from 1:1 to 1:3 resulted in a 3°C increase in θ (C, ). Changes in the molar ratio of QO12 and QO14 yielded a similar impact on θ (D, ). Of note, at a given length and molar ratio of QO, θ was not affected by the length of HP (). Unlike molecular beacon design where the length of loop significantly influences the closing rate and dissociation equilibrium of the probe (), in a partially double-stranded linear probe, the length of HP does not influence the hybridization stability of HP:QO duplex. To examine the hybridization stability between partially double-stranded DNA probes and perfect-match sequences, thermal denaturation analysis was performed using conditions where target oligonucleotides were at a 5-fold excess relative to HP (A, C). The length of HP was positively correlated with probe-target binding stability; longer HPs exhibited higher θs than shorter ones. For example, θ for HP31 (68°C) was 6°C higher than HP25 (62°C) and 7°C higher than HP20 (61°C) (). For a perfect-match sequence, the presence of QO or an increasing molar ratio had minimal impact on HP:Target melting profiles (A, C, ). Similarly, the length of QO did not exert an appreciable impact on θ for HP:Target hybrid (). A series of double-stranded DNA probes were tested in real-time RT-PCR assays using transcribed HIV-1 RNA sequences as targets. The impact of HP length and HP:QO molar ratio was assessed by measuring threshold cycle (C) values and sensitivity at low target level. B, D and show that C values for both HP20 and HP31 were minimally affected by QO length or concentration, with a maximal C difference of approximately 1 cycle. Detection sensitivity between different probe designs was also evaluated by comparing the detection rate of eight replicates of targets at ∼4 copies per reaction. shows that the detection rate was not adversely impacted when HP was combined with QO of different length or ratios. These data demonstrated that increasing HP:QO stability did not have a negative impact on the binding to a perfect-match target in a real-time RT-PCR reaction. In the absence of QO, HP31 had a much lower signal gain (dRn at the end cycle) than that of HP20, 4 versus 10 (B and D), due to less efficient quenching and elevated background signal associated with longer hybridization probes. When QO16 was included to achieve efficient quenching, the signal gain for HP31 was significantly increased, to the same level as that of HP20. The impact of mismatches on probe binding was studied for multiple probe compositions of various HP and QO lengths and molar ratios. The extent of destabilization caused by mismatches in the target sequence was analyzed by a shift in θ in a melting profile (). For example, a single mismatch located roughly in the middle of HP20 shifted the θ by 9°C (52°C versus 61°C). For the longer probes, HP25 and HP31, a single mismatch shifted the θ by only 5°C while 2 or 3 mismatches shifted the θ by 13°C (for HP25) or 17°C (for HP31), respectively (). Impact of mismatches on real-time RT-PCR for a number of probe designs was also evaluated with C delay and difference in quantitation (). For example, the single mismatch for HP20 caused a 2.5 delay in C and 0.7 log under-quantitation, yet for longer probe HP31, the same mismatch had no significant impact on C and quantitation (). The effect of QO on mismatch tolerance was evaluated. In the absence of QO, binding between the single-mismatch target and HP20 probe can be achieved at room temperature (A). In contrast, the presence of QO16 reduced the amount of probe bound to target by ∼50% (C). Using real-time RT-PCR, the amplification signal for HP20 without QO was only slightly impacted when comparing targets with zero to one mismatch (B). However, in the presence of QO16 at a 1:1 molar ratio, amplification signal was suppressed for the mismatch target (D). Furthermore, addition of QO16 resulted in a larger C delay (3.3 versus 2.5 cycles) and increased under-quantitation (0.9 versus 0.7 log) of the mismatch target as compared to the perfect-match target (). The results indicate that addition of QO can negatively impact mismatch tolerance. The impact of QO concentration, length and HP length on mismatch tolerance was further evaluated. For HP20, decreasing the HP:QO molar ratio resulted in more efficient binding to a mismatch target [compare HP:QO 1:1 (C) to 1:3 (E)] and reduced under-quantitation (D, F, ). Also shortening the QO by replacing QO16 with QO12 improved the binding to the single-mismatch target (E, G), and significantly reduced the extent of under-quantitation (F, H, ). Thus, lowering HP:QO stability facilitated the binding between HP and mismatch target. Similarly, lengthening HP20 to HP31 in the presence of QO16 recovered binding to the single-mismatch target to the same level of perfect-match targets (C, I) and the single mismatch had no significant impact on quantitation (J, ). The melting profiles indicated that the impact of mismatches on fluorescent signal (difference between perfect-match and mismatch signal) was dependent upon the detection temperature. This prompted us to investigate the effect of different read temperatures on mismatch tolerance/discrimination in a real-time RT-PCR reaction. As shown in , a target with a single mismatch at position 12 relative to HP20 was not detected when read at 56°C (between θs of the perfect-match and mismatch targets, 61 and 52°C, respectively), while this mismatch was much better tolerated at 35°C (lower than θs of both perfect-match and mismatch targets). The same mismatch was well tolerated by HP31 even when read at 56°C (lower than θs of both perfect-match and mismatch targets, 68 and 63°C, respectively), consistent with the melting analysis. Based on the principles established above, a long partially double-stranded linear DNA probe (HP43:QO14) was designed for use in the Abbott RealTie HIV-1 assay to quantify HIV-1 viral load. The QO was complementary to the first 14 nt of HP43. The purpose of designing a long HP was two fold: (i) to ensure sufficient binding affinity for perfect-match targets, (ii) to maximize tolerance to mismatches. The length of QO and its molar ratio of 1:4 were chosen to ensure efficient quenching while preserving high sensitivity and mismatch tolerance. Melting analysis was performed to evaluate the impact of 0, 1, 2, 3 or 4 mismatches within the target sequence on performance of the HP43:QO14 probe. As shown in A, in the absence of target, the formation of probe duplex had an θ of 43°C and fluorescent signal was maximally quenched at ∼35°C. In the presence of target, binding of HP43 resulted in a large increase in fluorescent signal at 35°C. The θ for HP43:Target duplex formation (for perfect-match) was calculated to be ∼69°C. Representative naturally occurring mismatches in the probe-binding region shifted θ to lower temperatures, i.e. 67°C or 68°C for 1 mismatch, 64°C or 65°C for 2 mismatches, 60°C for 3 mismatches and 58°C for 4 mismatches, all still well above the of HP:QO. Therefore, at a read temperature of 35°C, HP43 achieved efficient binding to targets containing up to 4 mismatches. Mismatch tolerance of the HP43:QO14 probe was also evaluated in real-time RT-PCR. As shown in B, overall amplification profiles were tightly grouped. The target with 4 mismatches showed a C delay of 2.5 cycles, while Cs for targets with 0–3 mismatches were not significantly different. Signal strength at the end of PCR was similar for all the targets, regardless of number of mismatches. Real-time PCR reactions involve two processes that simultaneously occur and are mechanistically inter-dependent: target amplification and signal generation. It is the combination of these two processes that enables real-time monitoring of PCR product accumulation in a closed-tube format providing the basis for sensitive and reproducible quantification of input samples. Real-time PCR assay design starts with selection of target region and primer sequences as well as optimization of amplification efficiency including reagent composition, concentration and cycling conditions. Signal generation during PCR is generally achieved through hybridization of a fluorescent probe resulting in differential fluorescent signal incurred by the binding or cleavage of the probe. Multiple factors have to be considered when designing a probe within the amplicon region: (i) optimizing sensitivity for detection of low level targets, (ii) maximizing mismatch tolerance or discrimination while not adversely affecting detection of perfect-match sequences and (iii) lowering pre-amplification background signals and improving signal to noise ratio. Optimization of the probe is usually focused on either selecting an optimal probe-binding site or modulating hybridization stability once the former is determined. Binding site choices are often limited by the primer selection and the distribution of mismatches in the target region. Thus, thermodynamic modulation of the probe may further improve the performance attributes of a real-time PCR assay. Several previous studies have investigated the relationship between the thermodynamic properties of different probes and their ability to differentiate between mismatch and perfect-match sequences (,,,). However, the benefits derived from these probe designs are potentially undermined by the inefficient binding to perfect-match sequence because of the intrinsically interdependent and competitive behaviors of the probes. Recently introduced DNA conjugate or nucleoside analogues (e.g. MGB, LNA and PNA) can significantly enhance binding affinity of short probes against perfect match sequences and confer increased mismatch discrimination (). Similar to the above-mentioned approaches, focus was often placed on one particular outcome of real-time PCR (i.e. mismatch tolerance or discrimination), yet assays are almost always designed to meet multiple specifications. Here, we report the development of a partially double-stranded linear DNA probe. Many features of this novel probe design can be thermodynamically modulated with considerable independence. Therefore, the probe can be designed such that an individual performance feature can be improved while others are not negatively impacted. Fluorescent signal of HP, in the absence of QO or target DNA, displayed a single-phase transition when temperature decreased, presumably due to spontaneous coiling of the oligonucleotide that brings quencher and fluorophore into close proximity (). The probability of quenching at each temperature and the signal level during transition is most likely dependent on the distance between the two ends of the primary sequence. Indeed, while HP20 had a θ of ∼32°C and was maximally quenched at ∼20°C, HP31 was quenched by only ∼50% at 20°C (). Since random coiling is also affected by the rigidity of a single-stranded hybridization probe, the actual correlation between melting stability and oligonucleotide length observed in this study may be specific to the chosen oligo length range, sequences, labeling position and divalent cation concentration (,). The quenched signal of HP measured in a melting analysis should be equal to the pre-amplification background signal in a real-time PCR with all other conditions being identical. Data presented in this study demonstrated that the background fluorescence can be effectively adjusted with QO through thermodynamic modulation of double-stranded DNA probe stability, as shown in . Thermodynamic modulation of the probe performance is essentially adjusting the balance between two intrinsically competitive processes: binding of the fluorescent-labeled HP either to the target or to the shorter QO. Since the shorter strand of the DNA duplex determines its hybridization stability, target binding is always more stable with longer HP within a practical range. Similarly, longer QO and higher concentration increases double-stranded probe stability and competes with target for probe binding. Because the design strategy introduced in this study stipulates a longer HP than QO, the HP preferentially binds to the target unless destabilized by the presence of mismatches. When the destabilization is sufficiently large, the level of mismatch tolerance or discrimination can be modulated by the length and/or concentration of the QO. The effectiveness of this modulation depends on the relative thermodynamic stability between the two processes, as measured by θ. When θ of probe melting (θ) is significantly lower than that of target binding (θ), changing probe duplex stability has little impact on target binding and signal generation, as shown in . To the contrary, when θ is significantly higher than θ especially when mismatches are present, the fluorescent signal specific for HP:Target becomes smaller [e.g. HP20:QO16 (1:3) with 1 mismatch target shown in ]. Therefore, desirable mismatch performance can be achieved by adjusting θ for probe binding relative to target binding. Specifically, setting the probe θ above that of mismatch target binding and below that of perfect-match target binding results in mismatch discrimination. Setting the probe θ below that of both mismatch and perfect-match target binding results in mismatch tolerance. This was shown experimentally when HP20 without a quencher oligo was compared with HP20 with quencher oligo of 16 bases. HP20 alone with a lower θ (30°C) than both perfect-match (61°C) and mismatch targets (52°C) tolerated a single mismatch whereas HP20:QO16 (1:3; θ = 55°C) discriminated between perfect-match and mismatch targets (compare A, B with E, F). Similarly, while a single mismatch did not significantly affect binding of HP31 in the presence of QO16 (I, J), three mismatches in the probe-binding region were sufficient to destabilize the binding (; melting data not shown). Under the condition where θ is significantly higher than θ, modulating θ for each binding process by adjusting HP length, QO length, or changing relative concentration between QO and HP will also change the signal gain. The extent of mismatch discrimination can thus be adjusted with precision, which was demonstrated by the comparison between HP20:QO16 and HP31:QO16 (C and I), between HP20:QO16 and HP20:QO12 (E and G) and between 1:1 and 1:3 conditions for HP20:QO16 (C and E). In addition to the consideration of length and concentration of HP and QO, PCR read temperature is another factor that can have a fundamental impact on assay performance. The thermodynamic behaviors of a probe such as target-binding and mismatch tolerance or discrimination are driven by specific detection temperatures. For a given probe design, PCR reactions can be detected at a temperature where there is differential binding between perfect-match and mismatch targets to achieve mismatch discrimination (A, B). Alternatively, a lower read temperature where both mismatch and perfect-match targets bind efficiently to the probe can be used to achieve mismatch tolerance and accurate quantification of diverse sequences (C, D). The relative independence between HP-QO and HP-target binding and between binding to perfect-match and mismatch sequences makes the thermodynamic modulation particularly straightforward for partially double-stranded linear DNA probes. In this probe system, HP can be lengthened or shortened for mismatch tolerance or discrimination without impacting HP:QO stability, as shown in . Moreover, QO length and HP:QO ratio can be adjusted specifically for mismatch tolerance or discrimination without impacting HP binding to perfect-match targets, as shown in and and . This important characteristic adds considerable flexibility and distinguishes partially double-stranded linear DNA probes from other previously described probes. Molecular beacons bind to targets with high specificity due to the formation of stem-loop structure that competes with target binding. Molecular beacons are thus very sensitive to nucleotide variations and can be conveniently optimized for differentiating sequence polymorphisms. However, modulating molecular beacons to tolerate sequence variations is not straightforward because lengthening the loop or weakening the stem can potentially increase background signal and affect sensitivity. TaqMan technology utilizes conditions that are also highly stringent for mismatches in the target-binding region due to the relatively high polymerase extension temperature required for cleavage and signal generation, making it potentially unsuitable for detection of polymorphic sequences. Symmetric or near-symmetric double-stranded probes have been designed for maximal mismatch discrimination (). However, they were also shown to suffer in binding kinetics for perfect-match targets (). Compared with these probe technologies, the partially double-stranded linear DNA probe introduced in this study has the following advantages: (i) ease of design by independently modulating HP and QO, (ii) design flexibility for either mismatch tolerance or discrimination, (iii) ability to modulate probe performance by titration of the quencher oligo, without changing oligo length or nucleotide composition and (iv) detection temperature uncoupled from extension step allowing additional control over mismatch stringency. In conclusion, we introduced in this study a unique and versatile partially double-stranded linear DNA probe design strategy for real-time PCR application. We demonstrated the feasibility of thermodynamic modulation of two probe strands to meet the complex performance requirements for diagnostic assay development. The concept of using a melting model for probe design in a PCR reaction is widely applicable when probe hybridization is involved in a solution-based reaction. The combination of two independent oligonucleotide strands allows for easier optimization of assay performance. Partially double-stranded DNA probes can be modulated to either discriminate mismatches for genotyping or SNP detection or tolerate mismatches as demonstrated with detection of diverse HIV-1 sequences. The mismatch tolerance of this novel class of probes has been demonstrated by several evaluations of Abbott RealTie HIV-1 assay on genetically diverse specimens (,,).

Post-translational modifications of chromatin such as histone and DNA methylation are recognized by epigenetic regulators HP1 (heterochromatin protein 1) and MeCP2 (methyl CpG-binding protein 2) respectively and play an important role in transcriptional regulation. These non-histone chromatin factors read the epigenetic marks and translate them into inactive chromatin states. MeCP2 is a member of a family of proteins, which share a conserved methyl cytosine-binding domain (MBD) that recognizes methylated CpG dinucleotides (). Moreover, MeCP2 contains a nuclear localization signal [NLS; ()] and a transcriptional repression domain (TRD), which binds a corepressor complex containing mSin3a and histone deacetylases [HDACs; ()]. HP1 proteins are conserved from yeast to humans () and recognize histone H3 trimethylated at the lysine 9 position [H3K9Me3; (,)]. In mammals, three isoforms viz α, β, γ have been identified (,). Functionally, three domains have been defined in HP1(s). The chromodomain [CD; ()]and the chromo shadow domain [CSD; ()] are highly conserved and are linked by the poorly conserved hinge domain. The CD has been shown to be important for binding methylated histones, while the CSD is known to interact with several proteins () as well as mediate homo () and heterodimerization of HP1 isoforms (). The hinge domain interacts with DNA () and RNA (). In mouse cells, both HP1 and MeCP2 accumulate at pericentric regions of chromosomes organized into chromocenters, which play an important role in epigenetic gene regulation possibly by creating silencing compartments within the nucleus. Recently, we have shown that the level of MeCP2 as well as of MBD proteins starkly increased during myogenic differentiation concomitant with large-scale chromatin reorganization (). To investigate a potential crosstalk between both epigenetic regulators, we analyzed the amount and localization of HP1 with respect to MBD proteins during cellular differentiation. We found that although the level of HP1 proteins does not change dramatically, there is spatial relocalization of HP1 (especially HP1γ) during myogenesis from a more diffused distribution to a focal enrichment at pericentric heterochromatin. Furthermore, this redistribution to heterochromatin correlates with MeCP2 and MBD1 protein presence. We also demonstrate that HP1 and MeCP2 interact physically with each other, strengthening the argument that they cooperate in the formation of repressive subnuclear compartments involved in epigenetic gene silencing. The following HP1 plasmids were used: GFP-tagged full-length human HP1α/HP1β/HP1γ (); YFP-tagged deletion mutants of human HP1α/HP1β/HP1γ and full-length human HP1α/HP1β tagged with DsRed2 (). To construct a DsRed2 fusion of HP1γ, the BamHI–HindIII fragment of GFP-HP1γ containing HP1γ was subcloned into BglII–HindIII site of pDsRed2-C1 (Clontech). MeCP2 constructs used were GFP/YFP/mRFP1-tagged full-length and deletion mutants of rat MeCP2 (). MeCP2Y.6 and MeCP2G.7 were constructed by subcloning XhoI–HindIII and XhoI–PstI fragments of MeCP2 from MeCP2Y into pEYFP-N1and pEGFP-N1 (Clontech) cut with the same restriction enzymes, respectively. pEGFP-N1 (Clontech) was used as a control. Pmi28 mouse myoblast cells (MB) were cultured as described in (), transfected using Transfectin (Biorad) and differentiated as described before (). Differentiated cultures include syncitial myotubes (MT) and unfused myocytes (MC). HEK293-EBNA human cells (Invitrogen) were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum at 37°C with 5% CO. 4 × 10 HEK293-EBNA cells plated onto 100 mm diameter culture dishes were transfected using PEI (poly-ethyleneimine 25 kDa from Polysciences 1 mg/ml in ddHO, neutralized with HCl). For transfection 500 μl of DMEM without serum, 12 μg of DNA and 50 μl of PEI were mixed well, incubated for 10 min at room temperature, vortexed and added to the cells dropwise. The culture was incubated at 37°C overnight, next day cells were washed in PBS, pelleted and used for co-immunoprecipitation assays. Proliferating and differentiated Pmi28 cultures were fixed in 3.7% formaldehyde/PBS and permeabilized with 0.5% TritonX-100/1XPBS and immunostained as described in (). Primary antibodies used were: mouse monoclonal anti-HP1 isoform-specific antibodies (Chemicon), rabbit polyclonal anti-MeCP2 (Upstate) and anti-MBD1 (Santa Cruz) antibodies. Secondary antibodies used were: anti-mouse IgG-Cy5, anti-rabbit IgG-FITC (Jackson Immuno Research). Samples were counterstained with DAPI and examined on a Zeiss Axiovert 200 using 40× and 63× objectives. Images were acquired with a PCO Sensicam QE cooled CCD camera using Zeiss Axiovision V.3 software and processed with Adobe Photoshop. To quantify the correlation between HP1γ localization at chromocenters and presence of MeCP2 or MBD1, we analyzed 375 MB cells; 71 cells with positive staining for MeCP2; 99 cells with positive staining for MBD1; 125 cells transfected with MeCP2-GFP and 345 MT nuclei from two independent experiments done in triplicate. The mean and SDs were plotted using Microsoft Excel software (). Differentiated and non-differentiated Pmi28 cells were grown on p100 culture dishes, boiled in Laemmli sample buffer and analyzed on western blots (). Immunoprecipitations ( and ) were done as described before (). The following primary antibodies were used: rabbit polyclonal anti-lamin B [kind gift of R.Bastos; ()], rabbit polyclonal anti-H3K9Me3 (Upstate), rabbit polyclonal anti-MeCP2 (Upstate), chromatographically purified rabbit IgG (Organon Teknika), mouse monoclonal anti-HP1α/HP1β/HP1γ (Chemicon), rabbit polyclonal anti-histone H3 (Upstate), mouse monoclonal anti-GFP (Roche), GFP binder (), anti-mRFP1 rabbit polyclonal antiserum. Secondary antibodies used were: anti-mouse IgG HRP (Amersham) and anti-rabbit IgG HRP (Sigma). Immunoreactive signals were visualized using an ECL plus Detection kit (Amersham) and recorded using a luminescence imager (Luminescent Image Analyzer LAS-1000, Fuji). To compare the amounts of the different proteins in proliferating and differentiated myogenic cultures, quantification of the recorded signals was done with the Image Gauge Ver.3.0 software (Fuji). Equal sized boxes were made around the recorded signals and for calculating the background. Integrated pixel intensity was measured for each band and the respective background signal was subtracted. Signals were normalized to the loading control (lamin B or histone H3) and the fold difference between the normalized signals in differentiated versus proliferating cultures was calculated. The mean and SDs were calculated from three independent experiments and plotted using Microsoft Excel software (). During cellular differentiation progressive inactivation of the genome occurs in parallel with the activation of tissue-specific gene expression patterns (). We have shown that the level of methyl CpG-binding protein dramatically increased during muscle differentiation and induced large-scale aggregation of pericentric heterochromatin (). A second major pathway associated with transcriptional silencing is mediated by HP1 binding of histone H3K9Me3. We therefore investigated whether the level of the different HP1 isoforms varied during cellular differentiation using a well-established culture system for myogenesis (A). Pmi28 mouse myoblasts (MB) were induced to differentiate by incubation in horse-serum-containing medium. After three to four days, cells fused to form post-mitotic multinucleated myotubes (MT). These cultures still contained mononucleated not fully differentiated cells termed myocytes (MC). We quantified the level of HP1 in proliferating versus differentiated cell extracts by western blot analysis and normalized it to lamin B level as a loading control for nuclear proteins. The level of HP1α, β, γ remained almost constant during differentiation (B and C). However, the fraction of histone H3 that was trimethylated at lysine 9 position (H3K9Me3) increased about 3-fold in differentiated cells. Previous studies have reported a cell cycle stage and isoform-specific localization of HP1 (). To address this possibility, we examined the in situ localization of the HP1 isoforms as well as H3K9Me3 by immunofluorescence staining during myogenic differentiation. Pericentric heterochromatin organized in chromocenters was highlighted by counterstaining with the DNA dye 4′,6-diamidino-2-phenylindole (DAPI). We found that the level of association of HP1 with pericentric heterochromatin differed between isoforms and changed during differentiation. While HP1α protein could be found accumulated at pericentric heterochromatin in most of the MBs (89%; Supplementary Figure 1), HP1β did not show such an accumulation (data not shown) and HP1γ showed only a weak heterochromatin accumulation in about half of the MBs (61%; ). This weak accumulation was not due to the absence of H3K9Me3, since chromocenters of all MBs stained clearly positive for this histone modification (Supplementary Figure 2) and is consistent with earlier reports showing HP1γ mostly excluded from constitutive heterochromatin (). We can also rule out epitope masking (), as in the same population of MBs, there were cells where HP1γ staining was detected at chromocenters (A magnified nucleus). The fraction of MT nuclei with HP1α and γ accumulated at heterochromatin increased to 100 and 90%, respectively (Supplementary and ). In contrast, upon differentiation there was no major change in the distribution of HP1β (data not shown) even though there was an increase in the level of its binding site H3K9Me3 (). We reasoned therefore, that this increase in heterochromatin association could depend on differentiation-specific factors other than the histone methylation mark . Since MeCP2 and other MBDs are present in a few MB only but increase during differentiation and label almost all chromocenters in MT (), we tested whether the change in heterochromatin association of HP1γ was correlated to MBD protein. Indeed we found a clear correlation of HP1 heterochromatin association in MB and the presence of either MeCP2 or MBD1. Almost all MeCP2 or MBD1 positive MB contained HP1α (100%) and HP1γ (95%) at chromocenters ( and Supplementary Figure 1). Furthermore, 96 and 94% of MB cells ectopically expressing MeCP2-GFP fusion had HP1γ and HP1α accumulation at pericentric heterochromatin (B and Supplementary Figure 1B). Altogether, these data showed that the chromocenter association of HP1 with particular emphasis for HP1γ clearly increased upon myogenic differentiation and was positively correlated with the presence of MeCP2 and MBD1. Since the accumulation of HP1 at chromocenters correlated with the presence of MBD proteins at these sites, we tested whether they could physically interact. HEK293-EBNA cells, which express HP1 proteins, were transfected with plasmids coding for GFP, GFP-tagged MeCP2 or GFP-tagged HP1 (A). Twelve hours later, cells were lysed and immunoprecipitations performed with an anti-GFP-specific antibody fragment [GFP binder; ()]. Input and bound fractions were analyzed on western blots for precipitated GFP-tagged protein (data not shown) and for co-precipitated endogenous HP1γ protein. HP1γ did not bind to GFP alone but was co-precipitated with MeCP2-GFP (B) and the same was true for HP1α and β (data not shown). Since HP1α, β and γ have been shown to form homodimers (,) as well as heterodimers [HP1α-γ; ()], [HP1α-β; ()], we reproduced this data as a positive control for our co-immunoprecipitation conditions. Moreover, the fraction of HP1γ bound to HP1α was comparable with the amount bound to MeCP2 (B). Using a mRFP-tagged MeCP2, we co-immunoprecipitated GFP-tagged HP1α, β and γ (C). MeCP2-GFP proteins could likewise immunoprecipitate DsRed2-tagged HP1s ( and data not shown) showing that the interaction of HP1 with MeCP2 was independent of the tags. Further, we tested whether endogenous HP1 and MeCP2 could interact. We performed immunoprecipitations using anti-MeCP2 antibody on Pmi28 MBs (expressing low level of MeCP2) and MTs (expressing higher level of MeCP2) (). Indeed, the rabbit anti-MeCP2 antibody but not the control rabbit IgG could co-precipitate HP1γ from MT extracts. Finally, to test whether MeCP2 could directly interact with HP1, we used GST pull down assays. Recombinant MeCP2 purified from bacteria was incubated with glutathione agarose coupled GST or GST-HP1γ (Supplementary Figure 3). While no MeCP2 protein was detected in the GST-bound fraction, GST-HP1γ was able to specifically pull down MeCP2. In summary, these results showed that MeCP2 and HP1 interact and at a level comparable to the dimerization of HP1 proteins. The N terminus of HP1 contains the H3K9Me3-binding site () while the C terminus mediates dimerization of HP1 as well as interaction with other proteins (,). To test which domain would be involved in the interaction with MeCP2, we co-transfected HEK293-EBNA cells with plasmids coding for MeCP2-mRFP and with different YFP-tagged deletion constructs of HP1 isoforms coding either for the CD or the CSD. Co-immunoprecipitation assays demonstrated that the CSD of HP1s was necessary and sufficient for binding to MeCP2 ( and data not shown). The CSD of HP1 has previously been shown to be important for the interaction of HP1 with other nuclear proteins (). We then investigated which domain of MeCP2 binds to HP1 by using a series of fluorescently tagged deletion constructs of MeCP2. The results indicate that amino acids 1–55 of MeCP2 are primarily involved in binding HP1 (), though weaker binding could be detected with other regions of MeCP2 as well (Supplementary Figure 4). We conclude that MeCP2 and HP1 interact via the CSD of HP1 and the N-terminal domain of MeCP2. The domains of MeCP2 that have been better functionally characterized are the MBD, the transcriptional repressor domain (TRD) and the overlapping Sin3a co-repressor domain (coRID), all of which are in the central part of MeCP2 (). Our data now implicate the N-terminal region before the MBD in binding to HP1, suggesting a direct physical link between the factors translating DNA and histone methylation. On the one hand, MeCP2 recognizes methyl CpGs and interacts with DNA methyltransferase 1 (). On the other hand, HP1 binds to H3K9Me3 and associates with the histone H3K9 methyltransferase [Suv39h1; ()]. Our data showing that HP1 and MeCP2 interact with each other interconnects these two major epigenetic pathways. Most recently, HP1 was also reported to interact with Dnmt1 (). It is noteworthy that another MBD protein, MBD1 has been reported to interact with HP1α via the MBD (). Since other MBDs ( and Supplementary Figure 1) were also able to enhance the accumulation of HP1 at heterochromatin, any single MBD knockout would not be expected to disrupt it. In line with this, we have previously shown that other MBDs have overlapping functions and knockout of MeCP2 alone did not affect heterochromatin reorganization during myogenic differentiation (). Significantly, we found that the heterochromatin association of HP1γ increased during differentiation and that this was correlated with either MeCP2 or MBD1 presence. The differentiation-specific increase of the MBD proteins could enhance HP1γ binding to constitutive heterochromatin, which would then recruit histone H3K9 methyltransferases leading to higher levels of H3K9 methylation. In Suv39h1/2 double knockout cells where H3K9 methylation at chromocenters is abrogated, MeCP2 still induced clustering (), indicating that its interaction with HP1 is not required for its function in large-scale chromatin organization. We further propose that the multiple interactions of these factors with chromatin and with each other generate subnuclear silencing compartments, which stabilize the differentiated phenotype by reducing transcriptional noise. Individually these interactions are transient but their cumulative effect at heterochromatin increases the local concentration of repressing factors and thereby the efficiency of gene silencing. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Uracil in DNA may arise from the occasional use of dUTP during DNA replication and from spontaneous deamination of cytosine, which is one of the major pro-mutagenic events in DNA. To maintain the integrity of the genetic information, most prokaryotic and eukaryotic cells encode uracil-DNA glycosylases (UDGs). These enzymes recognize and remove uracil residues from DNA by the base excision repair (BER) pathway. In human cells, five distinct UDG activities have been identified namely UNG1, UNG2, TDG, MBD4 and SMUG (). UNG2 is known to enter the nucleus while the isoform UNG1 enters the mitochondria (). Moreover, UNG2 plays an important role in immunoglobulin gene diversification () and is incorporated into virions of the human immunodeficiency virus type-1 (,). Some DNA viruses, such as herpesviruses and poxviruses, also encode a UDG activity. In these instances, the UDG activity appears to have an important role in virus replication (). The first UDG activity reported was purified from cells. Since then, enzymes highly homologous to the archetypal UDG have been purified from numerous organisms, including herpes simplex virus type-1 and human cells (UNG1 and UNG2 enzymes). These UDGs (Family-1) are able to eliminate uracil bases efficiently from both single-stranded (ss) and double-stranded (ds) DNAs regardless of the partner base, U:A or U:G (). However, in some cases, a preference for the ssDNA substrates has been reported (,). Furthermore, a mismatch-specific uracil-DNA glycosylase (MUG) was purified from cells (). This enzyme, which is related to human thymine-DNA glycosylase (TDG) (), is exclusively active against U:G mismatches. Both MUG and TDG are members of the Family-2 UDGs (). During the last years, UDGs are emerging as attractive therapeutic targets due to their role in a wide range of biological processes. Hence, the discovery of small molecules able to inhibit the activity of particular UDGs has a great interest. In addition, the knowledge generated by studying new UDG inhibitors should provide further insights into the process of substrate recognition and catalysis by UDGs. The first natural UDG inhibitor reported was Ugi, a highly acidic protein (84 amino acids) encoded by the phage PBS2, whose DNA genome is unusual in that it contains uracil instead of thymine (). Ugi inactivates Family-1 UDGs from , rat liver, herpes simplex virus, and humans (), but not other DNA glycosylases (). The X-ray crystal structures of Ugi in complex with different UDGs revealed that Ugi mimics electronegative and structural features of duplex DNA (). Some synthetic inhibitors of UDGs have also been described. Among them, uracil derivatives and oligonucleotide-based substrates were shown to inhibit selectively the herpes simplex virus type-1 UDG (). Uracil-based ligands able to inhibit the human UNG2 enzyme have also been designed (). Recently, we reported the identification of a novel natural inhibitor of the UDG (). This inhibitor, named p56, is a small acidic protein (56 amino acids) encoded by the lytic phage ϕ29. Unlike phage PBS2, the DNA genome of ϕ29 does not contain uracil residues. Protein p56 is synthesized upon ϕ29 infection and knocks out a host-encoded BER system that could be harmful for viral replication if uracil residues arise in the replicative intermediates (). In the present work, we have addressed some structural features of protein p56 by sedimentation equilibrium, sedimentation velocity and circular dichroism (CD) spectroscopy. Moreover, using the UDG enzyme, we performed a biochemical characterization of protein p56 as an approach to understand its mechanism of UDG inhibition. Our results revealed that protein p56 blocked the DNA-binding site of UDG. Thus, protein p56 could mimic DNA structural features in order to inhibit UDG. Protein p56 was overproduced in BL21(DE3) cells harbouring plasmid pCR2.1-TOPO.p56, and it was purified following a large-scale purification method as previously described (). Protein p56 concentration was determined either by quantitative amino acid analysis using a Pharmacia-Biochrom 20 Amino Acid Analyzer or by UV absorbance spectroscopy. Amino terminal sequencing of protein p56 was performed by Edman degradation on a Perkin Elmer (Procise 494) Protein Sequencer. Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF, Beckman, Palo Alto, CA, USA) mass spectrometry of purified p56 protein was performed on a Brucker Biflex Instrument (Bruker-Franzen Analytik, Bremen, Germany) using insulin as standard. The spectra (average of 100 shots) were recorded in the linear mode at 19.5 kV. Sedimentation equilibrium experiments were performed at 20°C in an Optima XL-A (Beckman-Coulter) analytical ultracentrifuge equipped with UV-visible optics, using an An60Ti rotor with standard six-channel centrifuge cells (12-mm optical path) and centrepieces of epon charcoal. Protein p56 in buffer A (50 mM Tris-HCl, pH 7.5, 50 mM KCl) was centrifuged at 30 000 r.p.m. until sedimentation equilibrium was reached. Then, absorbance scans were taken at 280 nm. A range of protein concentration from 25 to 500 μM was analysed. In all cases, the baseline signals were measured after high-speed centrifugation (42 000 r.p.m.). Whole-cell apparent weight average molecular weights of p56 were determined using the program EQASSOC (). The partial specific volume of p56 was 0.7331 ml g, calculated from the amino acid composition with the program SEDNTERP (). Sedimentation velocity experiments were carried out at 60 000 r.p.m. and 20°C. Protein p56 (25–200 μM) was equilibrated in buffer A. The sedimentation coefficient for p56 was calculated by direct linear least-squares boundary modelling of the sedimentation velocity data using the program SEDFIT (). The sedimentation coefficients were corrected to standard conditions to get the corresponding value using the SEDNTERP program (). The translational frictional coefficient () of p56 was determined from the molecular mass and sedimentation coefficient of the protein (). The frictional coefficient of the equivalent hydrated sphere () was estimated using a hydration of 0.42 g HO per g protein (). These values allowed us to calculate the translational frictional ratio (/), which in turn gives an estimation of the shape of p56. Before CD analysis, purified p56 was dialysed against 20 mM NaHPO, pH 8.0 and diluted in the same buffer to a protein concentration of 0.7 mg/ml. CD spectra were acquired in a J-720 spectropolarimeter fitted with a peltier temperature control accessory. Far-UV spectrum was recorded in 0.1-mm optical path length quartz cells over a wavelength range from 190 to 260 nm at a temperature of 20°C. The CD spectrum was the average of four accumulations at a scanning speed of 20 nm/min and 1-nm spectral bandwidth. The CD spectrum of the buffer alone was subtracted from the experimental spectrum. To obtain structural information, the CD data were analysed using the following algorithms: CONTINLL () and CDNN (). Temperature-associated changes in the p56 secondary structure were measured by increasing the temperature from 15 to 95°C at two different rates (15 and 45°C h). For temperature scans acquisition, purified p56 was dialysed against 20 mM HEPES, pH 8.0 and diluted in the same buffer to a protein concentration of 0.3 mg/ml. Changes in ellipticity at 218 nm were recorded in a 1-mm optical path length quartz cell. In addition, far-UV CD spectra over a wavelength range from 210 to 260 nm were recorded at temperatures between 20 and 95°C with temperature increments of 10°C. The temperature was allowed to equilibrate for 1 min before each spectrum was acquired. UDG preparations were purchased from New England Biolabs. Then, they were dialysed against buffer B (50 mM Tris-HCl, pH 7.5, 50 mM KCl, 1 mM EDTA, 7 mM β-mercaptoethanol, 5% glycerol) and concentrated using a Microcon microconcentrator 10 (Amicon). To estimate the concentration of UDG, aliquots from the protein preparation were analysed by SDS-PAGE (15% polyacrylamide). Molecular weight markers (Invitrogen) and increasing amounts of a standard protein (ϕ29 single-stranded DNA-binding protein, 13.3 kDa) were run in the same gel. The gel was stained with Coomassie Blue, and scanned by densitometry. Under these conditions, a major band with the mobility expected for UDG (25.7 kDa) was detected. Since some minor bands were visualized, UDG concentration was estimated by comparing the intensity of the UDG bands with that of the standard protein. To measure UDG activity, a 34-mer oligonucleotide containing a single uracil residue at position 16 (ssDNA-U; from Isogen) was used as substrate. It was 5′-labelled with γ-PATP (3000 Ci/mmol) (GE Healthcare) and T4 polynucleotide kinase (New England Biolabs). Reaction mixtures (20 μl) contained increasing amounts of the UDG preparation and the radiolabelled substrate in buffer B. After incubation at 37°C for 8 min, samples were treated with NaOH to a final concentration of 0.2 M, and heated at 90°C for 30 min. Samples were then dried in a Speed Vac, resuspended in 10 μl of formamide loading buffer (95% formamide, 20 mM EDTA, 0.05% xylene cyanol, 0.05% bromophenol blue), and subjected to electrophoresis in 8 M urea/20% polyacrylamide gels. The minimal UDG amount needed to obtain total cleavage of the substrate was used to examine UDG inhibition by p56. The theoretical isoelectric point of UDG and p56 is 6.67 and 4.17, respectively. To examine whether protein p56 was able to interact with UDG, the indicated amounts of both proteins were incubated in buffer B at room temperature for 15 min, kept at 4°C for 15 min, and analysed by basic-native PAGE (16% polyacrylamide). Tris-borate (TBE) buffer, pH 8.3, was used as running buffer. Under these conditions, UDG–p56 complexes were detected. Similar amounts of UDG–p56 complexes were detected when UDG and p56 were incubated at 4, 28 or 37°C for 30 min. Denatured calf thymus DNA immobilized on cellulose (GE Healthcare) was used. The DNA affinity column (150 μl; 1.1 μg DNA/μl) was equilibrated with buffer C (30 mM Tris-HCl, pH 7.5, 1 mM EDTA, 1 mM dithiothreitol). Then, UDG (300 μl; 1.6 μM) was applied to the column under gravity flow. After washing with 1.5 ml of buffer C, UDG bound to the column was eluted with protein p56 (700 μl; 1.35 μM). Elution fractions (100 μl each) were dried in a Speed Vac, and resupended in 10 μl of loading buffer (40 mM Tris-HCl, pH 6.8, 2% SDS, 15% glycerol). Proteins were resolved by SDS-Tricine-PAGE (). To analyse whether UDG–p56 complexes were able to bind to the DNA affinity column, a reaction mixture (900 μl) containing 0.5 μM of UDG and 1 μM of protein p56 in buffer C was incubated at 4°C for 30 min. Under these conditions, UDG–p56 complexes were formed. The reaction mixture was subsequently loaded onto the DNA affinity column (150 μl; 1.1 μg DNA/μl). Column fractions (100 μl each) were collected, dried in a Speed Vac, resuspended in 10 μl of loading buffer, and analysed by SDS-Tricine-PAGE. A ϕ29 DNA region (121 bp) was amplified by the polymerase chain reaction (PCR) using oligonucleotides 5′-CGCATTGTATGAGCTTTCTAGG-3′ and 5′-ATTGTTATATCGTATGAGTCAACAAAATC-3′ as primers. For this amplification, Taq DNA polymerase (New England Biolabs), α-P-dATP (3000 Ci/mmol) (GE Healthcare) and dUTP instead of dTTP were used. Reaction mixtures (20 μl) contained 50 mM Tris-HCl, pH 7.5, 50 mM KCl, 4% glycerol, radiolabelled DNA (6.25 nM) and the indicated amounts of UDG and p56. Samples were kept at 4°C for 20 min, mixed with glycerol to a final concentration of 8%, and then analysed by non-denaturing PAGE (6% polyacrylamide). Gel electrophoresis was performed at 4°C. Gels were vacuum-dried and autoradiographed. Protein p56 from phage ϕ29 was purified to near homogeneity after expression in (). MALDI-TOF mass spectrometric analysis of purified p56 revealed the existence of two forms. The major form had 6573.4 Da, which agreed with the molecular weight of the p56 monomer, as calculated from the DNA sequence of the gene product (6565.3 Da). The minor form (6438 Da) would correspond to p56 lacking the first Met residue, a fact that was confirmed by determination of the N-terminal amino acid sequence of the purified protein. Furthermore, the amino acid analysis of p56 was in agreement with the amino acid composition predicted from the nucleotide sequence (data not shown). Protein p56 has 12 aspartic or glutamic residues and 5 arginine or lysine residues, which results in a low theoretical isoelectric point (4.17). Its molar extinction coefficient was calculated to be 7450 Mcm at 280 nm due to the presence of 5 tyrosine residues. Protein p56, which acts as an inhibitor of the UDG, accumulates throughout the ϕ29 lytic cycle. At early stages of phage infection there are ∼10 copies of p56 per infected cell, and it increases up to ∼10 at late stages (). Therefore, taking into account the cell volume of infected cells (), the intracellular concentration of protein p56 would range from 15 to 150 μM throughout the infective cycle. To determine the oligomerization state of protein p56 in solution, sedimentation equilibrium assays were performed at various p56 concentrations (25–500 μM), and all of them gave a similar sedimentation pattern. At 75 μM, a physiological protein concentration, the experimental data () are best fit to an average molecular mass () of 13 000 ± 360, a value that corresponds with the theoretical mass of a p56 dimer (13 130 Da). Similar average molecular masses were determined when p56 concentrations of 25 μM (12 850 ± 250), 200 μM (13 370 ± 400) and 500 μM (13 110 ± 390) were used. Hence, p56 is a dimeric protein at physiological concentrations. Sedimentation velocity profiles of native protein p56 () fitted well to a single sedimenting species, with a value of 1.6 ± 0.05 and a of 13 500 that corresponds with that obtained from the sedimentation equilibrium and from the theoretical mass of a p56 dimer. No improvement in the best-fit parameters was obtained if more sedimenting species were considered, an indication of sample homogeneity. The frictional ratio (/) calculated from the values obtained in the analytical ultracentrifugation was 1.22 ± 0.05. Therefore, the hydrodynamic behaviour of the p56 dimer deviates from the one corresponding to a rigid spherical particle, which has an / value = 1.0. We conclude that the p56 dimer may have an ellipsoidal shape. To obtain experimental information on the relative amounts of the secondary structural elements of protein p56, far-UV CD spectroscopic analyses were performed. The CD spectrum of p56 in the far-UV region was characterized by two minima around 195 and 215 nm and a maximum at 200 nm, indicative of a protein with low content in α-helices (). The overall secondary structure of p56 was estimated by two deconvolution methods of CD spectra (see Materials and Methods section). The results gave a consensus average of 4.6% α-helices, 39% β-strands and 55.3% assigned to turns and non-regular structures. Both methods gave similar estimations of the content of β-strands and random elements. Temperature-induced changes in the secondary structure of protein p56 were analysed by CD spectroscopy (). Specifically, we monitored the changes in the ellipticity at 218 nm by increasing the temperature from 15 to 95°C. The results showed that the CD of p56 changed as a function of the temperature (A). The temperature CD transition curve of p56 at 218 nm was characterized by a decrease in the ellipticity between 30 and 40°C. Additional changes in the curve region between 40 and 95°C were not observed, and p56 protein remained soluble up to 95°C. Comparison of the spectra acquired at 20 and 60°C (B) showed loss of secondary structure of p56 after the temperature-induced change. This structural change was reversible and independent of the temperature scanning rate (15 or 60°C h; data not shown). Our previous work showed that addition of purified protein p56 to cell extracts inhibited the endogenous UDG activity (). UDG (), which belongs to a family of highly conserved UDGs (Family-1). The activity of UDG is to remove uracil from both ssDNAs and dsDNAs, regardless of the partner base U:A or U:G (). These enzymes hydrolyse the N-glycosidic bond between the uracil residue and the deoxyribose sugar of the DNA backbone, generating an apurinic-apyrimidinic (AP) site. The AP site is further recognized by an AP endonuclease, which cleaves the phophodiester bond of the DNA backbone 5′ to the AP site (). In the absence of an AP endonuclease activity, chemical cleavage of the DNA at the AP site can be achieved by treatment with heat and alkali. To examine whether protein p56 was able to inhibit the UDG activity, UDG (5 nM) and p56 (from 0.04 to 2 μM) were incubated at room temperature for 15 min, and kept at 4°C for 15 min (formation of UDG–p56 complexes). Then, a 34-mer single-stranded oligonucleotide containing a single uracil residue at position 16 (ssDNA-U) was added to the reaction mixture. After 8 min at 37°C, the reactions were treated with NaOH. As shown in , total cleavage of the substrate was detected in the absence of protein p56, whereas nearly 20, 65 and 85% of the substrate remained intact when 0.16, 0.8 and 2 μM of p56, respectively, were used. Inhibition of the UDG by protein p56 was also observed when a 34-bp dsDNA carrying a U:G mismatch at position 16 was used as substrate (data not shown). Therefore, we conclude that protein p56 functions as an inhibitor of the UDG enzyme. Protein p56 targets the UDG enzyme , forming a complex (UDG–p56) that migrated slightly faster than free UDG in a non-denaturing polyacrylamide gel () (). To analyse the stability of the UDG–p56 complex at high urea concentrations, preformed UDG–p56 complexes were incubated with different concentrations of urea (up to 8 M) for 30 min, and then analysed by non-denaturing PAGE (). In this assay, preformed UDG–Ugi complexes were used as control. Like protein p56, Ugi from the phage PBS2 inactivates the UDG (). As expected, the UDG–Ugi complexes were not affected by incubation with 8 M urea (,). The amount of UDG–p56 complexes in the absence of urea (data not shown) was similar to that detected at 4 M urea (). UDG–p56 complexes remained intact up to 6 M urea, indicating that protein p56 forms a tight complex with the UDG enzyme. Moreover, free protein p56 was not denatured with 8 M urea, indicating that p56 is unusually stable at high urea concentrations. To understand the mechanism underlying UDG inhibition by protein p56, we investigated whether protein p56 was able to dissociate preformed UDG–DNA complexes. To this end, affinity chromatography experiments using denatured calf thymus DNA immobilized on cellulose were performed. As shown in A, UDG was able to bind to the DNA affinity column. After the washing steps, protein p56 was applied to the column and the elution fractions (from E1 to E7) were analysed by SDS-Tricine-PAGE. The UDG enzyme eluted with protein p56, indicating that p56 was able to compete with DNA for binding to UDG. Subsequently, we studied whether protein p56 impaired binding of UDG to DNA. In this assay, UDG–p56 complexes were first formed and then they were loaded onto the DNA affinity column. Again, the column fractions were analysed by SDS-Tricine-PAGE (B). Both proteins UDG and p56 co-eluted, indicating that p56 blocked DNA binding by UDG. This result was further confirmed by electrophoretic mobility shift assays (). In this case, a dsDNA fragment (121 bp) containing uracil instead of thymine residues (60% A:U) was incubated with either UDG alone or preformed UDG–p56 complexes. A single band moving slower than free DNA was visualized when the DNA substrate was incubated with free UDG, but not in the samples containing UDG–p56 complexes. In conjunction, these results demonstrate that protein p56 prevents the UDG enzyme from binding to DNA. Ugi functions by mimicking the structure of DNA. Specifically, it mimics phosphate groups and UDG–DNA interactions (hydrogen bonds and hydrophobic contacts) (). Unlike p56, Ugi (84 residues) is a monomer in solution (). An alignment of the primary structures of protein p56 and Ugi is shown in . The regions of p56 spanning from Val12 to Gln23 (region A) and from Leu31 to Leu50 (region B) showed homology to the regions of Ugi spanning from Leu4 to Gln15 and from Leu23 to Val43, respectively. According to the X-ray crystal structure of Ugi in complex with UDG (), several amino acids of the Ugi region spanning residues Leu23 and Val43 contact with UDG. As shown above, protein p56 interacts with UDG and impairs its DNA-binding activity. This effect could be due to an indirect conformational change in the DNA-binding site of UDG. Alternatively, p56 could directly interact with the DNA-binding domain of UDG leading to its physical occlusion. To discriminate between both possibilities, we investigated whether Ugi was able to replace protein p56 already bound to UDG. Thus, preformed UDG–p56 complexes were incubated with Ugi for 10 min at room temperature, and then analysed by non-denaturing PAGE. As shown in , UDG–Ugi but not UDG–p56 complexes were detected, indicating that Ugi was able to replace protein p56. Therefore, protein p56 appears to block the DNA-binding site of UDG rather than to induce an indirect conformational change of the enzyme. Proteins can mimic DNA structures as a mechanism to block DNA-binding enzymes. At present, only a small number of DNA mimic proteins have been discovered. Although this class of proteins are structurally diverse, they tend to resemble some DNA structural features, such as the phosphate backbone of DNA or the hydrogen-bonding properties of the nucleotide bases (,). The Ocr protein from the phage T7 is an example of DNA mimicry. It binds tightly to type I DNA restriction and modification enzymes (). Biophysical and crystallographic studies revealed that the Ocr dimer mimics 24 bp of B-form DNA containing a central bend (,). Moreover, Ugi encoded by the phage PBS2 acts as a UDG inhibitor by mimicking electronegative and structural features of duplex DNA (). A most recent example of DNA mimicry is Mfpa, a protein that binds to DNA gyrase. MfpA exhibits a highly unusual right-handed beta-helix fold (). UDG is the target of p56, an early and small protein (56 amino acids) encoded by phage ϕ29. Inhibition of the host UDG by protein p56 is likely related to the mechanism of ϕ29 DNA replication (). We show in this study that protein p56 has a high content of β-strands (around 40%), and forms dimers in solution at physiological concentrations. In addition, we demonstrate that protein p56 inhibits the UDG enzyme . Several features of protein p56 suggest that it could be a novel naturally occurring DNA mimicry. First, protein p56 competes with DNA for binding to UDG. Second, the interaction between p56 and UDG blocks the interaction between UDG and DNA. Third, the Ugi protein, which interacts with the DNA-binding groove of UDG, is able to replace protein p56 previously bound to the UDG enzyme. Fourth, like Ocr and Ugi, p56 is a highly acidic protein. In most of the known DNA mimic proteins, carboxylates from the side-chains of aspartates and glutamates generate an overall charge distribution that resembles the DNA phosphate backbone (). Further resolution of the 3D structure of both protein p56 and the UDG–p56 complex will reveal whether protein p56 functions as a structural mimic of the DNA substrate recognized by UDG.

Isomerization of uridine into pseudouridine (5-ribosyluracil, Ψ) is a prevalent post-transcriptional modification of cellular RNAs () and this modified residue is frequently found in functionally important RNA regions. Consistently, an important role of Ψ residues was demonstrated for specific codon–anticodon recognition and ribosome function (). Furthermore, one of the highly conserved Ψ residue in U2 snRNA was found to play a key role in pre-mRNA splicing (,). Two types of catalysts can generate Ψ residues at specific sites in a pre-existing polyribonucleotide chains. One type consists in a single protein with an RNA:Ψ-synthase activity. It recognizes specifically one or more RNA segments or motifs and converts the targeted U residues within these segments into Ψ residues. A large panel of such RNA:Ψ-synthases are found in most cell types and they are up to now the unique system detected in bacteria (). These bacterial RNA:Ψ-synthases were the first to be characterized (). The search for their homologs by inspection of various genomes led to the classification of these enzymes into five families, TruA, TruB, TruD, RsuA, RluA (). Based on sequence alignment and 3D structure analysis of the catalytic domains of several RNA:Ψ-synthases, five conserved structural motifs (I, II, IIa, III and IIIa) were identified and they represent the signature for RNA:Ψ-synthases activity (,,). Although only a limited number of amino acids in these motifs are conserved, these motifs have highly conserved conformations in the 3D structures established for RNA:Ψ-synthases (). The few conserved amino acids play important roles for catalysis: a universally conserved aspartate residue in motif II is the essential catalytic amino acid (,). A conserved basic residue (Arg or Lys) in motif III makes a salt bridge with the catalytic Asp residue. A conserved Tyr residue in motif IIa has a stacking interaction with the uracyl residue to be modified in the target RNA (). Two other hydrophobic residues, a leucine in motif IIIa and an isoleucine or a valine residue in motif III, are also highly conserved, but their functional implication in the catalytic mechanism is less understood (,). In the various RNA:Ψ-synthases identified up to now, the conserved catalytic domain is associated with a large variety of additional domains, which are likely involved in the specific recognition of the target RNA motif. One of these domains, the PUA domain (seudoridine and rchaeosine tRNA guanine transglycosylase), has largely been studied. It is found at the C-terminus of members of the TruB family of RNA:Ψ-synthases (). As deduced from the crystal structure of the complex formed between the TruB enzyme and the T stem-loop structure of tRNA (), the PUA domain is expected to play a key role in RNA recognition by interacting with the CCA 3′-terminal sequence of tRNAs (). This proposed interaction of TruB with a sequence common to all tRNAs fits with the general activity of TruB at position 55 of all elongator tRNAs. The second type of catalyst for U to Ψ conversion consists in RNP complexes containing a small RNA with H/ACA motifs (the H/ACA snoRNAs and scaRNAs in Eukarya and H/ACA sRNAs in archaea) and a well-defined set of proteins (). One of these proteins denoted aCBF5 in archaea, Cbf5p in yeast and CBF5/NAP57/Dyskerin in human displays the characteristic motifs I, II, IIa, III and IIIa found in RNA:Ψ-synthases and belongs to the TruB family of RNA:Ψ-synthase (,). In each RNP, the RNA component base pairs with one or more targeted RNA sequences and through this interaction acts as a guide to specify the uridyl residues that will be modified (). Knowledge on the mechanism of the H/ACA RNPs activity is largely based on data obtained with the archaeal H/ACA RNPs that could be reconstituted using an transcribed RNA and recombinant proteins (,). In addition to the protein bearing the signature of RNA:Ψ-synthases, each H/ACA RNP contains three distinct proteins (NOP10, GAR1 and NHP2 in human; Nop10p, Gar1p and Nhp2p in yeast; and aNOP10, aGAR1 and L7Ae in archaea). The three archaeal aCBF5, aNOP10 and aGAR1 proteins form an heterotrimer in solution (,). The 3D structures of the aCBF5-aNOP10 heterodimer and aCBF5-aNOP10-aGAR1 heterotrimer complexes were solved by X-ray crystallography (). Very recently, the crystal structure of an entire archaeal box H/ACA sRNP was determined (). Through reconstitution experiments, we showed that in archaea the aCBF5–aNOP10 complex is the minimal set of proteins required to get an RNA-guided RNA:Ψ-synthase activity (). However, the two other proteins aGAR1 and L7Ae both increase the efficiency of the catalytic reaction. Formation of an H/ACA sRNA–aCBF5–aNOP10 complex depends upon aCBF5 association with the sRNA and the conserved ACA motif in the sRNA plays an essential role for this association (,). More precisely, the PUA domain of aCBF5 interacts with the ACA motif of H/ACA sRNAs (,). According to our previous reconstitution assays, formation of an sRNA–aCBF5–aNOP10 complex is required for binding of the RNA substrate and recent structural studies on NOP10 revealed its capability to bind RNA (). By reconstitution of H/ACA sRNPs with truncated aNOP10 proteins, we showed that its C-terminal half is sufficient for sRNP activity, whereas its N-terminal half reinforces the aCBF5-aNOP10 interaction (). The L7Ae protein binds a K-turn motif present in each stem-loop structures of the archaeal H/ACA sRNAs (,,,). The direct interaction of L7Ae with aNOP10 and the sRNA stabilizes the association of the other proteins with the sRNA (). In light of the higher rate of pseudouridylation observed in the presence of aGAR1, we proposed that protein aGAR1 might enhance the turnover of the reaction catalyzed by the sRNP (). A recent study showed that in addition to its central role in the RNA guided activity of H/ACA sRNPs, the archaeal aCBF5 enzyme has the ability to convert residue U55 of an transcribed tRNA into a Ψ residue (). This activity seems to be increased in the presence of the L7Ae, aNOP10 and aGAR1 set of proteins (). Altogether, these data revealed the ability of aCBF5 to act without guide RNA in case of a tRNA substrate. However, an dual role of aCBF5, as rRNA:Ψ-synthase when included in H/ACA sRNPs and as tRNA:Ψ55-synthase when acting as a free protein, is still not demonstrated since another archaeal RNA:Ψ-synthase PusX/Pus10 is also able to catalyze U to Ψ conversion at position 55 in tRNAs. Nevertheless, the capability of aCBF5 to act as a free protein and in sRNPs represented an opportunity: (i) to delineate the tRNA structures needed to get an activity of aCBF5 in the absence of guide RNA, (ii) to compare the aNOP10 and aGAR1 requirements for the guided and non guided activity of aCBF5, (iii) to compare the aCBF5 amino acid requirement for its guided and non-guided activity. Here, by site-directed mutagenesis of the and tRNAs, we show that a truncated version of tRNA including the acceptor and TΨC stem-loop structures linked by a flexible sequence is sufficient to get aCBF5 activity in the absence of a guide RNA. In the absence of the 3′ terminal CCA, the aCBF5 activity becomes strictly dependent on either aNOP10 or aGAR1. Furthermore, by site-directed mutagenesis of both aNOP10 and aCBF5 we demonstrate distinct amino acid requirements for the RNA-guided and non RNA-guided activities of aCBF5. The results obtained are discussed based on the 3D structures of the archaeal H/ACA sRNP proteins and their complexes. The aCBF5 D82A and CΔ(PUA) protein variants carrying, respectively, an Asp to Ala substitution at position 82 and a truncation of the PUA domain, were previously described (,). The PCR approach was used for site-directed mutagenesis of protein aCBF5 (K53A, H77A and R202A variants) and aNOP10 (Y14A, H31A and P32A variants). The sequence of the oligonucleotides used for the PCR reactions are available upon request. Wild-type and variant aCBF5 and aNOP10 proteins as well as the aGAR1 and TruB proteins were produced as GST fusion proteins as previously described (). The GST moiety was removed by cleavage with the PreScission protease. The various proteins were stored at −80°C in the following buffer 150 mM NaCl, 1 mM EDTA, 1 mM DTT, glycerol 10%, 50 mM Tris–HCl pH 7. Wild-type and mutated tRNA transcripts, the Pab91 H/ACA sRNA and the RNA target RNA-S () were produced by transcription of PCR products. The forward primers used for PCR amplification generated the T7 RNA polymerase promoter. DNA templates for amplifications of sequences coding for the WT yeast tRNA and variants of this tRNA (ΔT-tRNA, ΔD-tRNA,) were a generous gift of C. Florentz (IBMC, Strasbourg). Additional yeast tRNA variants (Δa.c.-tRNA, Δacceptor-tRNA, Δacceptor-tRNA-U55C) were produced by site-directed mutagenesis in the course of the PCR amplification. The sequence coding for the WT tRNA was PCR amplified using genomic DNA from the GE5 strain. Sequences encoding the variants tRNA-ACA and tRNA-ΔCCA were obtained by using a reverse primer carrying the mutated sequence. Sequences encoding the WT semi-tRNA or the variant semi-tRNA-ΔCCA or semi-tRNA-U55C were obtained by hybridization of two complementary oligonucleotides. The other archaeal tRNA variants (tRNA-U55C) were produced by site-directed mutagenesis in the course of a PCR amplification. DNA templates for production of the Pab91 sRNA and its target RNA (RNA-S) were previously described (). The RNA transcripts were purified by denaturing gel electrophoresis. Labeling was done at their 5′ end for EMSA experiments or during transcription by incorporation of [α-P]CTP for the pseudouridylation assays. Conditions for transcription and labeling were previously described (). EMSA were performed in the conditions previously described (,). Briefly, radiolabeled RNAs (50 fmol) were incubated in buffer D [150 mM KCl, 1.5 mM MgCl, 0.2 mM EDTA and 20 mM HEPES (pH 7.9)] with proteins at a 200 nM concentration, at 65°C for 10 min. To test for the association of the Pab91 sRNP with its target RNA (complex CII), 2.5 pmol of RNA-S were added. Complexes were visualized by autoradiography after fractionation by non-denaturing polyacrylamide gel electrophoresis. The tRNA:Ψ55-synthase activities of purified aCBF5 or H/ACA–RNP complexes were measured by the nearest-neighbor approach in the conditions previously described (,). The RNA-guided activity of reconstituted H/ACA sRNPs was measured as previously described by mixing ∼4 pmol of Pab91 sRNA with ∼150 fmol of the [α-P]CTP-labeled target RNA-S (). The non-RNA-guided reaction of aCBF5 on tRNA was tested on ∼50 fmol of [α-P]CTP-labeled tRNAs. For both RNA-guided and non-RNA-guided reactions, samples were pre-incubated at 65°C and the reaction was started by addition of the proteins (200 nM each). Aliquots were collected at different time points, the RNAs were extracted and then digested by RNase T2. The resulting 3′-mononucleotides were fractionated by thin-layer cellulose chromatography. The radioactivity in the spots was quantified using the ImageQuant software after exposure of a phosphorimager screen. The amount of Ψ residue formed was determined taking into account the total number of U residues in the tRNA or RNA-S molecules. Some of the values given in the text are expressed with standard errors of the mean values. All the time-course analyses were repeated at least three times and gave reproducible data. Some control experiments were performed with the TruB recombinant protein. The same procedure was used to measure the activity, except that the reaction was performed at 37°C. We used the -cyclohexyl-′-(2-morpholinoethyl)-carbodiimide metho--toluolsulfonate (CMCT) modification protocol of (), adapted by (). CMCT modifications were performed with 0.5 μg of transcribed archaeal or yeast tRNA. Positions of CMCT modifications were identified by primer extension analysis, using the AMV RT (QBiogene, USA) in the conditions previously described () and the primer oligonucleotides OG4891 for archaeal tRNA and OG5302 for tRNA (A1 and B1). Primers were 5′-end labeled using [γ-P]ATP (3000 Ci/mmol) and T4 polynucleotide kinase. RNA sequencing was done with 4 μg of transcribed RNA [for details on the technique, see ()]. Previous studies revealed a U to Ψ conversion activity of the aCBF5 protein at position 55 in the tRNA(GUC), i.e. a tRNA:Ψ55-synthase activity. Such an activity is enhanced in the presence of the aNOP10, aGAR1 and L7Ae proteins set (). For a deeper understanding of these activation properties, here we quantified the relative stimulation effects of proteins aNOP10, aGAR1 and L7Ae alone and in various combinations, on the tRNA:Ψ55-synthase activity of aCBF5. The assays were performed with the recombinant proteins aCBF5, aNOP10, aGAR1 and L7Ae, and with both tRNA(GUC) and tRNA(GUC) substrates (A1 and B1). We used the nearest neighbor and the RT–CMCT approaches in order to verify that modifications occurred only at position 55 in tRNAs. For the nearest-neighbor approach, both substrates were labeled by incorporation of [α-P]CTP during transcription. After the pseudouridylation reactions, the tRNA transcripts were digested with RNase T2 and the products were fractionated by 2D chromatography (A2 and B2). CMCT modifications followed by alkaline treatments were performed on cold RNA transcripts, and positions of modifications were identified by reverse transcription (A3 and B3) using oligonucleotide primers that were complementary to the 3′ terminal sequence of tRNAs (A1 and B1). The pseudouridylation reaction was performed at 65°C, for 80 min using a 200 nM concentration of each protein. Time point experiments were performed on the tRNA with: (i) aCBF5 alone, (ii) aCBF5 in the presence of aNOP10 or aGAR1, or of both proteins and (iii) aCBF5 in the presence of aNOP10 and L7Ae, or the three L7Ae, aNOP10, aGAR1 proteins. Aliquots of the reaction mixtures were analyzed after 5, 10, 20, 40 and 80 min of incubation (A4 and S1). For comparison, a time point experiment was performed at 37°C with a recombinant TruB enzyme (A2). Altogether, the results obtained demonstrated that protein aCBF5 is able to modify the and tRNA at the same high level after 80 min of incubation (∼0.85 mol ± 0.05 and ∼0.9 mol ± 0.05 of Ψ per mol of tRNA, respectively). In both cases, only a slight increase of the level of modification was observed in the presence of aNOP10 and aGAR1 (A2 and B2). However, as evidenced by the time point experiments, although the kinetics of the reaction with aCBF5 at 65°C is very fast compared to that for TruB at 37°C, it was markedly increased in the presence of aNOP10 or aGAR1. On the contrary, protein L7Ae showed no stimulatory effect (see Supplementary Data, Figure S1), which is in contrast to data obtained for the sRNA-guided activity of aCBF5 (,). The CMCT analysis did confirm that aCBF5 acted only at position 55 in both tRNAs. Finally, we confirmed the general capability of aCBF5 to act at position 55 in tRNAs by using other archaeal and bacterial transcripts as the substrates (data not shown). Our previous data on the sRNA guided activity of aCBF5 revealed a strong importance of the PUA domain of aCBF5 for this activity and a negative effect of mutations in the ACA sequence of the H/ACA sRNA (). The recent crystal structure of an H/ACA sRNP demonstrated a direct contact of the PUA domain with the ACA motif (). Therefore, we tested whether the PUA domain was also required for the activity of aCBF5 at position 55 in tRNAs. As shown in , no U to Ψ conversion occurred in the tRNA when it was incubated with an aCBF5 protein lacking the PUA domain (CΔ(PUA)). Interestingly, this tRNA:Ψ55-synthase activity was partially restored in the presence of the aNOP10 and aGAR1 proteins pair (CΔ(PUA)NG). However, the kinetics was slower and the efficiency of modification after 80 min was lower. The individual aNOP10 and aGAR1 proteins had similar but very low capabilities to restore the tRNA:Ψ55-synthase activity of the truncated aCBF5 protein (). As previous data had revealed the absence of aCBF5 activity on a tRNA lacking the 3′ terminal CCA motif (), we tested whether aNOP10 or aGAR1 or both proteins can compensate for the absence of this motif (). We found in this case that individual aNOP10 or aGAR1 can restore a strong aCBF5 activity (CN and CG, respectively) at position 55 of tRNAs and this activity was reinforced to some extent when the two proteins were present (CNG). Remarkably, modification with the CNG complex was almost complete after a 10-min incubation, even in the absence of the CCA motif (). The truncated tRNA was modified at a slower rate and lower yield, when TruB was used as the catalyst. Taken together, the data revealed a very important role of the PUA domain for the aCBF5 tRNA:Ψ55-synthase activity and showed the capability of the aNOP10–aGAR1 protein pair to compensate partially for its absence. Moreover, the presence of the auxiliary proteins fully restores the aCBF5 activity in the absence of the tRNA CCA 3′ terminal motif. As we found that aNOP10 and aGAR1 can compensate for CCA deletion, we wondered whether they can compensate for more severe truncations in tRNAs. As aCBF5 was acting on the tRNA, and as several variants of this tRNA had already been produced (A1), we tested the capability of aCBF5 (C) and the aCBF5–aNOP10–aGAR1 complex (CNG) to convert U55 into a Ψ residue in these truncated tRNAs (A2). In RNAs ΔT-tRNA, Δa.c.-tRNA and ΔD-tRNA, the T stem-loop, the anticodon stem-loop, and the D stem-loop were deleted, respectively. The acceptor stem was truncated in RNA Δacceptor-tRNA, but the CCA 3′-terminal motif was conserved (A1). Obviously, no Ψ formation was detected in the ΔT-tRNA substrate which was lacking residue U55. Only a very low level of modification was detected in the Δa.c.-tRNA when aCBF5 was used alone, while a high level of modification was detected when aNOP10 and aGAR1 were present in the incubation mixture (A2). Similarly, no modification of the Δacceptor-tRNA substrate was observed when the incubation was performed with aCBF5 alone, whereas almost a complete modification was detected for these two truncated tRNAs in the presence of aNOP10 and aGAR1 (A2). Modification only occurred at position 55 in the truncated substrates since it disappeared after U to C mutation at this position (Δacceptor-tRNA-U55C). Interestingly, the ΔD-tRNA was modified by aCBF5 at a high level in absence of the auxiliary proteins (A2). As expected from previous data (), TruB was able to modify a tRNA lacking the acceptor stem without the need for auxiliary proteins. Based on the activity of aCBF5 alone on the ΔD-tRNA, we concluded that aCBF5 alone, like TruB, does not need the tRNA 3D structure for activity. However, in contrast to TruB, to act alone aCBF5 needs the presence of its target T stem-loop but also the anticodon stem-loop and the acceptor stem. Roovers () had found that aCBF5 does not modify a truncated tRNA that contains the acceptor stem fused to the T stem-loop. We confirmed that for an equivalent substrate, no activity was restored upon addition of the aNOP10–aGAR1 protein pair (data not shown). However, as aCBF5 was active on the yeast ΔD-tRNA substrate, we postulated that flexibility between the acceptor stem and the T stem-loop might be an important feature for aCBF5 activity. Therefore, we produced a truncated RNA substrate (semi-tRNA) by truncation of the tRNA, which contained a UGAC single-stranded sequence acting as a flexible link between the acceptor stem and the T stem-loop (B1). The aCBF5 protein alone had a modest but significant activity on this substrate (mean value <0.2 mol.mol). This activity was abolished in the absence of the 3′ terminal CCA sequence (B2). However, in the presence of the aNOP10–aGAR1 protein pair both the semi-tRNA and semi-tRNA-ΔCCA lacking the 3′ terminal CCA sequence were almost completely modified. As expected, TruB was active on both substrates in the absence of auxiliary proteins. Hence, we concluded that the aCBF5–aNOP10–aGAR1 heterotrimer, but not aCBF5 alone, has catalytic properties similar to that of TruB. As the aNOP10 stimulatory effect on the aCBF5 activity seemed to consist in an increase of the kinetics of the reaction (), and as aNOP10 was known to bind RNA (,), we hypothesized that aNOP10 might facilitate the association of aCBF5 with the tRNA substrate. To test this possibility, we estimated by EMSA experiments the level of complexes formed by the association of aCBF5 with a P-labeled tRNA. The tests were performed in the presence or absence of aNOP10 (). The level of complex formation was increased by a factor of ∼4 in the presence of aNOP10 (, compare lane 2 with lane 3), whereas no significant increase was observed in the presence of aGAR1 (data not shown). As aCBF5 is known to interact with the ACA conserved sequence of H/ACA sRNAs, we compared the affinity of aCBF5 for the WT tRNA to that for a tRNA carrying an ACA triplet instead of the CCA triplet. The affinity of aCBF5 was indeed largely increased in the presence of an ACA triplet (, compare lane 2 with lane 5) and aNOP10 did not increase the level of complex formation in this case (, lane 6). Therefore, the stimulatory effect of aNOP10 on the kinetics of aCBF5 action on tRNAs may in part be due to its capability to facilitate the aCBF5–tRNA interaction. As previously observed for the aCBF5–RNA-guided activity, the stimulatory effect of aGAR1 on the aCBF5 activity is likely of another nature. Based on the 3D structures established for the aCBF5–aNOP10 dimer and the aCBF5–aNOP10–aGAR1 trimer, the direct interactions between some conserved residues in the central flexible linker of aNOP10 (A) with conserved residues of aCBF5 were proposed to influence the sRNA-guided activity of aCBF5 (,). In particular, the aNOP10 residues Y14 and H31 were found to form hydrogen bonds with residues E199 and R201 which are present in the β12 strand of aCBF5 (B) (). These interactions were proposed to influence the orientation of the side-chain of residue R202 that is located on the opposite face of the β12 strand. Interestingly, in the crystal structure of the complex formed between the TruB RNA:Ψ-synthase and a tRNA T stem-loop, the counterpart of residue R202 was proposed to interact with the phosphate on the 5′ side of the targeted uridine (). On the other hand, in the aCBF5–aNOP10 complex, residue P32 in the aNOP10 flexible linker was found to interact by Van der Waals interaction with the conserved residue P54, which is located in motif I of aCBF5 (C). This interaction was proposed to facilitate the formation of an hydrogen bond between residue K53 and the carbonyl oxygen of the catalytic residue D82 (). In light of these structural data, we used alanine substitutions to test for the relative importance of residues Y14, H31 and P32 for the RNA-guided and non-RNA-guided activities of aCBF5. Aiming at this, the stimulatory effects of each aNOP10 variant (Y14A, H31A and P32A) on the aCBF5 activity were tested on both the tRNA lacking the CCA sequence (D) and the RNA-S substrate (RNA target) of the Pab91 sRNPs (F). To this end, sRNPs were reconstituted with each of the variant aNOP10 proteins. Whereas the H31A substitution had only a limited effect on the kinetics and the yield of Ψ55 formation in the archaeal tRNA (D), the Y14A substitution strongly slowed down the kinetics of this modification and the P32A substitution had a negative effect on the modification yield (∼0.7 mol of Ψ per mole of tRNA, instead of ∼0.9 mol of Ψ per mole of tRNA in the presence of the WT aNOP10 protein). Interestingly, the Y14A and P32A variants only had a limited effect on the sRNA-guided activity of aCBF5 (F). The H31A mutation was in contrast strongly deleterious for this activity: only ∼0.2 mol of Ψ per mole of target RNA was obtained after a 80 min incubation. Hence, mutations in aNOP10 have marked different effects on the RNA-guided and non-RNA-guided activities of aCBF5. To try to define the reasons for the observed differences, we tested by EMSA the ability of the aNOP10 variant proteins to reinforce the aCBF5 association with the tRNA and with the Pab91 sRNA and its target RNA sequence (E and G). In agreement with the time-course results, both Y14A and P32A aNOP10 variants were less efficient (∼0.5-fold) than the H31A aNOP10 variant and the WT aNOP10 protein for increasing the aCBF5 association with the tRNA (E). On the contrary, only the H31A variant had a dramatic effect on the yield of complex CII (sRNA-L7Ae-aCBF5-aNOP10-RNA target) formation, suggesting that the H31A mutation has an important negative effect on substrate recruitment in the H/ACA sRNP. Therefore, we concluded that whereas residues Y14 and P32 are highly important for the aCBF5 tRNA:Ψ55-synthase activity, residue H31 plays an essential role in substrate recruitment in the sRNA-guided activity of aCBF5. Next, we tested the role of the aCBF5 residues R202 and K53 whose functions in the catalytic activity were proposed to be modulated by the aNOP10 residues H31 and Y14, and by residue P32, respectively (,). The aCBF5 variants R202A and K53A were produced, and the effect of each substitution was tested on both RNA-guided and non-RNA-guided activities. As evidenced in A1, both R202A and K53A substitutions in aCBF5 abolished its activity at position 55 in the tRNA. However, this activity was fully restored upon addition of aNOP10 and aGAR1 in the incubation mixture (A3). Here again, as evidenced by EMSA, the defect of activity observed for CBF5 alone was at least in part due to its low affinity for the tRNA (A2). The addition of aNOP10 enhanced the association of aCBF5 with the tRNA, which likely explains the restoration of the tRNA:Ψ55-synthase activity. Hence, residues R202 and K53 likely play a role in tRNA association and aNOP10 can compensate for their mutations. In contrast, aNOP10 and aGAR1 were not able to compensate the negative effect of the R202A and K53A mutations in aCBF5 for the Pab91 sRNP activity (B1), and here again, the low activity of the reconstituted sRNPs could be explained by a defect in substrate recruitment as the CII complex was formed at lower yields and displays a peculiar electrophoretic mobility (B2, lanes 7 and 9). Hence, we concluded that residues K53 and R202 are important for tRNA binding in the non-RNA-guided system and for recruitment of the RNA substrate in the RNA-guided system. However, in this latter case, aNOP10 and aGAR1 cannot compensate for the absence of these residues. Therefore, residues R202 and K53 appear to play a more crucial role in the RNA-guided than in the non-RNA-guided system. In the TruB enzyme, residue H43 was proposed to be involved in the flipping of residue U55 out of the helical stack (). The counterpart of residue H43 in aCBF5 is residue H77. As expected, the H77A mutation in aCBF5 almost completely impaired the aCBF5 activity on tRNA (A1) and this is also likely to be explained by the lower affinity of aCBF5 for the tRNA (A2). However, this activity could be restored in the presence of aNOP10 and aGAR1 (A3) and we showed that protein aNOP10 restored an efficient association of aCBF5 with the tRNA (A2). Interestingly, the H77A mutation had a very limited effect on the Pab91 sRNP activity (B1) consistent with an efficient recruitment of the RNA substrate observed for this variant (B2). Hence, we concluded that residue H77 does not play an essential role for the aCBF5 RNA:Ψ-synthase activity, provided that the aNOP10 and aGAR1 proteins are present. e p r e s e n t d a t a s h e d l i g h t o n c o m m o n a n d s p e c i f i c p r o p e r t i e s o f a C B F 5 a n d T r u B . T h e y a l s o b r i n g a c o m p a r a t i v e a n a l y s i s o f t h e s R N A - g u i d e d a n d n o n - g u i d e d a c t i v i t i e s o f a C B F 5 . F i n a l l y , t h e y d e l i n e a t e t h e r o l e o f a N O P 1 0 i n t h e a C B F 5 n o n - g u i d e d a c t i v i t y . p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

DNA replication is a remarkably accurate process. For a genome like that in a human cell, every one of the 6 × 10 nt is replicated once per cell cycle with extremely high fidelity. Moreover, this is achieved on a DNA template that is far from ideal. Cellular DNA is under constant attack from numerous environmental and endogenous agents. While most of the lesions are rapidly repaired, inevitably some are missed and pose as roadblocks for the replication machinery. A major advance in recent years is the recognition that lesion replication is of utmost importance to genome maintenance and requires the concerted action of a myriad of proteins involved in replication, repair, recombination and checkpoint response (). However, the mechanism by which these proteins work together to accomplish lesion replication is still poorly understood, largely due to the lack of an biochemical system that can recapitulate efficiently and faithfully this intricate process. For example, two general mechanisms of lesion replication have been proposed, an error-prone mechanism by the action of low-fidelity translesion DNA polymerases and an error-free mechanism by copying the missing information from the sister chromatid. While there is compelling evidence for the error-prone mechanism, it appears to represent only a minor pathway when compared to homologous recombination in (). To what extent error-prone replication contributes to overall lesion replication in eukaryotes remains to be determined. As to the error-free mechanism, the experimental evidence for its existence is still indirect (). Further complicating the issue is that correct nucleotides can also be inserted opposite many types of lesions by the appropriate translesion DNA polymerases, making it difficult to determine if a true error-free mechanism is involved in replicating a lesion (). To rigorously study the mechanism of lesion replication, we have endeavored to reconstitute lesion replication in the egg extract. In this extract, DNA, usually sperm chromatin, induces nuclear formation around itself and the DNA within the nucleus is replicated once (,). A more recent development is to use the nucleoplasmic extract (NPE) derived from the nuclei reconstituted in the crude egg extract to induce plasmid DNA replication (,). The system is the only system that relies solely on cellular proteins such as ORC and MCMs for replication, recapturing faithfully the complexity and control of eukaryotic replication forks (). This is in sharp contrast to the much simpler replication fork complex built around the large T antigen of the SV40 viral replication system that has previously been used for lesion replication (). In this study, we present evidence to show that a site-specific lesion stalls DNA replication one nucleotide before the lesion in the system. The stalling is transient and the lesion is eventually replicated by both the error-prone and the error-free mechanisms. Crude interphase egg extracts, membrane-free cytosol, and NPE were prepared following the published procedures (,). The parental plasmid, pBS-Trx, was constructed by inserting an NdeI fragment encoding the gene of pET-32a (Novagene) into the NdeI site of pBS- (Stratagene). pBS-Trx DNAs containing a synthetic AP site were constructed as previously described (). Three oligonucleotides, CCGGGTACCΔAGCTCG, CCGGGTACΔGAGCTCG and CCGGGTΔCCGAGCTCG were designed to place the AP site opposite C, G or T (‘Δ’ denotes the synthetic AP site). The dominant negative human APE1 mutant (E96Q/D210N) was kindly provided by D. Wilson. The gene was subcloned into pET32 vector to add a (His) affinity tag at the N terminus. The fusion protein was expressed in BL21 (Rosetta(DE3)) and purified on a Ni column followed by a HiTrap Q column. The cell extract was loaded onto the column at 15 mM imidazole and the column was then washed with buffer containing 70 mM imidazole. dnAPE was eluted with 100 mM imidazole, re-loaded onto a Q column, and eluted with buffer containing NaCl. The peak fractions containing dnAPE was dialyzed against ELB and concentrated with Aimcon Ultra -4 to 5 mg/ml. For a typical replication reaction, 0.25 µl DNA (300 ng/µl) was pre-incubated with 1 µl dnAPE (5 µg/µl) and 2.25 µl ELB buffer (10 mM HEPES (pH 7.5), 250 mM sucrose, 2.5 mM MgCl, 50 mM KCl and 1 mM DTT) at room temperature for 15 min. After addition of 1 µl cytosol and 0.5 µl 10× ATP cocktail (20 mM ATP, 200 mM phosphocreatine and 0.5 mg/ml creative kinase), the reaction was incubated for another 45 min. Finally, 10 µl NPE, 2 µl dnAPE, 2 µl 10× ATP cocktail, 1 µl P dATP and 5 µl ELB were added to accomplish DNA replication. Samples taken at different times were mixed with equal volume of 2× sample (buffer 80 mM Tris–HCl (pH 8), 0.13% phosphoric acid, 8 mM EDTA, 5% SDS, 0.2% bromophenol blue and 10% Ficoll). The volume was brought up to 10 µl with 1× sample buffer and then 1 µl proteinase K (10 mg/ml) was added. After overnight incubation at room temperature, samples were separated on a 1% TAE/agarose gel. DNA was isolated from an agarose gel and purified by Qiagen gel extration columns. pET28a DNA (1 µg total) was used as a carrier DNA during purification. For restriction enzyme digestion analysis of the relaxed and supercoiled replication products, the gel-purified DNA (20 min time point) was digested with various restriction enzymes and then separated on a 1% agarose gel. The gel was first stained by SYBR Gold and then dried for P exposure. For determination of the stalling site, the relaxed product (20 min time point) was first tailed with either dCTP or TTP using terminal deoxytransferase (TdT). PCR was then performed with (dG) or (dA) and an upstream primer (5′ CCGTGTCAAAACTGTCGTC 3′) using Taq DNA polymerase. The products were cloned into pUC19 and introduced into (DH5α) by transformation. The plasmids were isolated from the colonies and the inserts were sequenced. For analysis of the final replication products (after 75 min of incubation in NPE), the purified P-labeled supercoiled DNA was subject to restriction digestion by ClaI or DpnI or both. Half of the DNA was analyzed by TAE agarose gel electrophoresis and the remaining DNA was introduced into Maxi Efficiency DH5α (Invitrogen, CA, USA) by transformation. The plasmids from the transformants were isolated and the AP lesion region was sequenced to determine the nucleotides inserted opposite AP. Data from two independent experiments were used to calculate the average ratio (and absolute deviation) of each nucleotide inserted opposite the AP lesion after replication. For the determination of nucleotides on replicated DNA that still carried the AP lesion, the DNA (after 75 min of incubation in NPE) was digested with DpnI, ClaI and APE (NEB, MA, USA), purified by Qiagen column, and re-digested with KpnI. This DNA was then used as template for PCR with two primers (5′ AGCGAGG AAGCGGAAGAGC 3′ and 5′ TGGTTGCCGCCACT TCACC 3′) that bracket the ClaI and KpnI sites. The PCR product was digested with NdeI and SapI and then subcloned into pUC19 and introduced into DH5α. The DNA was isolated from the transformants and analyzed by restriction enzyme digestion and sequencing. Two reactions, each containing 1 µl of Δ:C DNA (300 ng/µl), were pre-incubated with 13 µl dnAPE (4 µl protein (5 µg/µl) and 9 µl ELB buffer) or 13 µl ELB buffer at room temperature for 15 min. After addition of 4 µl cytosol and 2 µl 10× ATP cocktail, the reactions were incubated for another 65 min. The DNA was purified by Qiagen PCR purification columns and treated with PstI, EarI and wild-type AP endonuclease or buffer. The 3′ EarI recessed ends were then filled in with P TTP by Klenow and the DNA were separated on a 5% urea polyacrylamide gel. The two strands surrounding the ClaI site in the parental plasmid pBS-Trx were prepared by extending primers corresponding to nucleotides 2495–2514 (5′ CTGAGAGTGCACCATATGGC 3′; for copying the AP-carrying strand) or 2849–2870 (5′ GTATTTCACACCGCATATGAGC 3′; for copying the lesion-free strand) with Sequenase in the presence of P-dATP, dGTP, dCTP and TTP. The products were digested with NdeI to generate the 345 bp fragments that contained the hemi-methylated ClaI site. The two NdeI fragments were then digested with ClaI and the products were separated on a 10% polyacrylamide gel and detected by exposure to X-ray film. To reconstitute DNA lesion replication, we replicated a plasmid DNA that carried a site-specific lesion in NPE (A). The lesion chosen is an apurinic/apyrimidinic (AP) site, which is abundant () and known to stall many purified DNA polymerases (). It is non-instructive, so error-free and error-prone replication products can be unambiguously distinguished. Normally, the AP lesion is rapidly repaired in cytosol, prior to the initiation of replication (B). Even depletion of AP endonuclease I (xAPE I) failed to provide significant protection of the AP lesion, most likely due to the presence of another AP endonuclease and/or other base repair pathways in the extract (data not shown). We thus protected the AP lesion with a dominant negative mutant of the human AP endonuclease I (dnAPE) that cannot cleave but still binds tightly to it (). When dnAPE was included in the reaction, the AP site was efficiently protected (>98% by this assay) from repair (B). As the AP endonuclease has no significant affinity for single-strand AP sites [() and data not shown], the mutant protein would fall off the unwound DNA and not by itself pose a hindrance to DNA polymerases. The AP DNA was replicated in NPE supplemented with either dnAPE or buffer (for the replication reactions in this study, dnAPE was included to protect the AP lesion unless otherwise indicated). In the absence of dnAPE (AP lesion repaired), the DNA was gradually replicated and converted to supercoiled plasmids (C). In the presence of dnAPE (AP lesion protected), the supercoiled replication product was still generated, but there was a transient accumulation of the relaxed form. For example, at the 20′ time point, the supercoiled form and the relaxed form were present at similar levels. By 60’, most of the relaxed form was converted into the supercoiled form. The effect was specific for the AP DNA as the non-AP DNA was not affected by dnAPE (D). (The slight excess of the relaxed products in the presence of dnAPE was most likely due to the unavoidable basal level of random AP lesions, formed either by spontaneous base loss or as the intermediates of base excision repair of damaged bases in the extract.) These data suggested that the lesion-free strand was replicated normally and gave rise to the supercoiled product. The AP-carrying strand, in contrast, was temporarily stalled and DNA synthesis restarted downstream, by either the same fork or the opposing fork, forming a gap at or near the lesion (A). To test this hypothesis, we isolated the relaxed and supercoiled products of the 20’ time point from the agarose gel and digested the DNA with various restriction enzymes. As shown in B, the supercoiled product was completely digested by all of the enzymes, but the relaxed product was completely digested by only a subset of the enzymes. Many of the enzymes could not digest well the relaxed product even though they completely digested pET28a plasmid, which was used as the carrier for DNA purification and served as the internal control for restriction digestion. When the digestion pattern was plotted on the plasmid, the enzymes that failed to digest the relaxed product were found to have sites within a small region immediately downstream of the AP site [with the exception of ClaI (see subsequently)]. This observation strongly suggested that most of the relaxed product carried a gap between the EcoRI site immediately 5′ to the AP site and the PvuII site 257 nt downstream of the AP site. The digestion by ClaI, which has a site 886 nt upstream of the lesion, did not conform to the above pattern. While ClaI completely digested the supercoiled product, it was very inefficient in digesting the relaxed product (B). An examination of the sequence revealed that this particular ClaI site overlaps with a GATC dam methylation site and methylation is known to block ClaI digestion. After one round of replication, the two daughter molecules would be hemi-methylated, but at different adenines within the ClaI site (A), and might therefore be differentially digested by ClaI. To test this hypothesis, we used a DNA polymerase to copy the two strands of the NdeI fragment that contains the ClaI site. The two hemi-methylated products were then digested by ClaI. As shown in B, the product copied from the lesion-free strand was digested, but the product from the lesion strand was not. This observation showed that the supercoiled replication product (ClaI sensitive) was exclusively derived from the lesion-free strand (which also suggested that the AP site was not repaired before replication). The gap, on the other hand, was present on the replication product derived from the AP strand (ClaI resistant). We next determined exactly where the AP lesion stalled replication. The strategy was to first add a tail (dC or dT) to the 3′ end of the stalled strand of the gel-purified relaxed DNA (20 min time point) with terminal deoxynucleotidetransferase (TdT) and then use a primer complementary to this tail and another primer further upstream to amplify the intervening region (A). The PCR product was cloned into a vector and introduced into by transformation. The plasmid DNA was isolated from the transformants and the inserts were sequenced. As shown in B, most of the clones from the dT tailing reaction (17/21) ended in GAGCT … T, and most of the clones from the dC tailing reaction (20/26) ended in GAGCTC … C. Combining the two sets of data, it became clear that the major stalling site was one nucleotide before the AP site, indicating that the replication of the lesion was kinetically slow. In addition, this experiment and the one above provided further evidence that replication stalling was caused by the lesion rather than the steric hindrance of the dominant negative mutant APE. A steric hindrance would be extremely unlikely to stall just one template strand and one nucleotide before the lesion. The replication stalling was only temporary and the AP lesion was eventually replicated over, leading to the accumulation of completely replicated, supercoiled product. If the AP lesion was replicated by copying the information from the lesion-free sister chromatid (error-free mechanism), the correct nucleotide would be expected at the position opposite the lesion. In contrast, if the AP lesion was replicated by translesion DNA polymerases (error-prone mechanism), then incorrect nucleotides would be expected. Furthermore, depending on what translesion polymerases were recruited, all 4 nt might be used at random or some nucleotides might be used in preference. To distinguish among these possibilities, we purified the final supercoiled replication products (after 75 min of replication in NPE) from an agarose gel. The DNA was treated with DpnI (to digest any residual unreplicated, fully methylated DNA; the plasmid contains 18 DpnI sites) and ClaI (to digest the product from the lesion-free template strand) and then introduced into by transformation. As a control, DpnI and ClaI were found to have efficiently digested the pET28a carrier DNA, as shown by both DNA agarose gel staining and transformation assay (A and B). The plasmids were isolated from the transformants and sequenced. ( could accurately repair the AP lesion. In a control experiment, 34 transformants from the AP DNA were examined and all were found to be correctly repaired.) In this experiment, among the 44 Δ:C replication products sequenced, 28 had a C, 13 an A, 1 a G and 1 a T inserted opposite the AP site. (C, middle column). A was an incorrect nucleotide, clearly the product of error-prone lesion replication. C was the correct nucleotide, suggesting that the AP lesion might also be replicated by an error-free mechanism, but the error-prone mechanism could not be ruled out because Rev1, a translesion DNA polymerase, is known to insert a C opposite an AP site (). To resolve this uncertainty, we performed a similar analysis on the replication products from a DNA that carried an AP lesion opposite a G (Δ:G). If a true error-free mechanism had been used, then more Gs (and correspondingly fewer Cs) would now be found opposite the AP site. This was indeed the case. In this experiment, among the 39 Δ:G replication products sequenced, 16 had a G, 14 an A, 8 a C and 1 a T inserted opposite the AP site (C, right column). We repeated these experiments and calculated the average ratios of each nucleotide inserted opposite the AP site on Δ:C and Δ:G replication products. As shown in D, the ratios were very different between Δ:C and Δ:G replication products and deviated dramatically from the expected ratios of random insertion. Together, these data strongly suggested that both an error-free mechanism (inserting C for Δ:C and G for Δ:G) and an error-prone mechanism (inserting A and C but rarely T and G for both substrates) were used to replicate the AP lesion. These data provided the first biochemical evidence for the existence of error-free lesion replication, but a mundane explanation was that the correct nucleotide was inserted on DNA whose AP lesion had been repaired before replication. While this seemed very unlikely as AP sites were efficiently protected, we nevertheless attempted to determine if the correct nucleotide was inserted on Δ:G replication products that still carried the AP lesion [95% of the AP strand replication products still carried the AP lesion (Supplementary Figure S2)]. The AP lesion in Δ:G was embedded within a KpnI site, and the nicking of the lesion by AP endonuclease rendered the DNA completely resistant to KpnI digestion (Supplementary Figure S3A). As illustrated in A, we digested the purified replication products with DpnI (to remove any un-replicated DNA; 4 of the 18 DpnI sites in the plasmid lie between ClaI and KpnI), ClaI (to remove the product of the lesion-free strand), APE (to nick the AP site) and finally KpnI (to remove all DNA with an intact KpnI site, including the putative pre-repaired DNA and their replication products). The digested DNA was then used as the template for PCR with two primers that bracketed the ClaI and KpnI sites. The correct nucleotide could only be recovered on PCR products amplified from the newly replicated strand of the DNA that still carried the AP lesion. As a control for digestion efficiency, the DNA purified from a replication reaction containing the normal plasmid pBS-Trx (used in AP DNA construction) did not generate any PCR product after digestion with all four enzymes (Supplementary Figure S3B). In contrast, the DNA purified from the Δ:G replication reaction generated the expected PCR product even after digestion with all four enzymes. This PCR product was subcloned into pUC19 and the DNA isolated from the transformants was analyzed by sequencing. As shown in B, 17 out 50 had the correct nucleotide G, and the remaining clones had mostly A and C but rarely T. In contrast, when the Δ:T DNA was used as the substrate, T was now frequently found opposite the AP lesion, but G became rare (C). Together, the results from these and the above experiments demonstrated that both error-prone and error-free mechanisms were used to replicate the AP lesion. In this study, we have shown that a site-specific AP lesion causes a strong stalling to DNA replication in an system that faithfully and efficiently recapitulates eukaryotic DNA replication. The stalling occurs only on the lesion-carrying template strand but not the lesion-free template strand. The major stalling site is one nucleotide before the lesion and the replication fork complex is slow in filling the position directly opposite the lesion. New DNA synthesis is initiated downstream of the lesion, either by the same fork or by the opposing fork, leading to the formation of a gap immediately downstream of the lesion. The stalling is temporary and the lesion is eventually replicated over. Most importantly, both error-prone and error-free mechanisms are used to replicate the AP lesion. Previous studies that use SV40 viral replication system or purified replication proteins have partially reconstituted some aspects of lesion replication such as fork stalling and translesion synthesis (). The system established in this study is the first to reconstitute with high efficiency all major aspects of lesion replication and has provided some important insights into the mechanism of eukaryotic lesion replication. The AP lesion is the only type of DNA lesions that allows a definitive distinction between error-free products and error-prone products. The drawback is that it is extremely efficiently repaired in extracts and has to be protected. The only way we have found to effectively protect the AP lesion is by the addition of a dominant negative AP endonuclease mutant. An obvious concern is that this strategy might introduce an artifact that the mutant protein itself causes replication stalling. Several observations argue against this possibility. The stalling occurs only on the lesion-carrying template strand but not on the lesion-free strand. As such, the mutant protein does not bind to the AP lesion so tightly that it blocks DNA unwinding. The major stalling site is one nucleotide before the position opposing the AP lesion. This precise position also strongly suggests that the AP lesion rather than steric hindrance from the mutant protein is the cause of stalling. Consistent with this interpretation, many replicative DNA polymerases, including DNA polymerase δ, stall at one nucleotide before the lesion (). In addition, the AP endonuclease is known to have very low affinity for the lesion on single-stranded DNA. Collectively, these observations strongly suggest that the strategy used in this study does recapitulate DNA lesion replication. The error-free mechanism by copying the correct information from the sister chromatid is often invoked as a major pathway for lesion replication, but direct evidence has so far been lacking. In fact, in addition to the error-free mechanism, translesion DNA polymerases can also insert the correct nucleotides opposite certain lesions such as thymine dimers and thymine glycol (). For example, the XP-V gene product was once thought to participate in error-free replication but later shown to insert 2 As opposite a thymine dimer via its translesion polymerization activity (,). AP sites are non-instructional, so the error-free products generated in our system have to somehow copy the correct information from the sister chromatid. This study thus provides the first direct evidence for an error-free mechanism of lesion replication. The correct information might be copied by either replication fork regression or homologous recombination. Future studies with the lesion replication system should help reveal the molecular details of error-free lesion replication. Our data also indicate that the error-prone mechanism can make a significant contribution to lesion DNA replication and that A and C are the major nucleotides inserted opposite an AP lesion. Previous studies in yeast have produced various results ranging from randomly inserted nucleotides () to A (,), C (), G () or T () as the main nucleotide inserted opposite AP sites. However, these studies were not designed to directly examine the replication of a defined AP site by a replication fork complex. Our data show that both the ‘A rule’ and the ‘C rule’ are used in the translesion replication of AP sites in eukaryotes. This conclusion is supported by the enzymatic activities of DNA polymerase Polδ and Rev1, which are respectively capable of inserting A and C opposite of an AP site, and by genetic analysis showing that Polδ and Rev1 are important for MMS-induced (via AP intermediates) mutagenesis in yeast (,). Future studies using the system described here should help reveal what roles the various DNA polymerases play in translesion replication and how the error-free and error-prone mechanisms are controlled. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

In eukaryotic cells, DNA polymerase β (Pol β) is essential for base excision repair (BER) and involved in recombination and drug resistance (,). Pol β expression levels are important for the maintenance of genome integrity. On account of its low fidelity, the overexpression of Pol β, which has been reported in several cancer types, leads to a mutator phenotype associated with genetic instability and decreased sensitivity to anti-cancer chemotherapeutics (). Conversely, cells deficient in DNA polymerase β display hypersensitvity to alkylating agent-induced apoptosis and chromosomal breakage (). Sugo () have demonstrated that Pol β-deficient mice die immediately after birth, due to extensive apoptosis in the developing central and peripheral nervous systems. These data indicate that DNA polymerase β expression level is extremely important and must be tightly regulated. Transcription of the Pol β gene is upregulated by the phosphorylated transcription factor CREB-1 in response to alkylating agent exposure. This upregulation requires the presence of the specific cAMP response element (CRE) in the promoter of the Pol β gene (,). It has also been shown that the Pol β promoter contains a binding site for the transcription factor SP1 (). Although regulation at the transcription level is apparent, expression of the Pol β transcripts might also be controlled at the post-transcriptional level. In rat cells there are two alternatively polyadenylated Pol β transcripts, with 3′UTRs of significantly different lengths (). The expression of the two transcripts is tissue specific. The short transcript is present in most tissues, with significantly higher expression in testis, while the long transcript is expressed mostly in the brain and lungs (). Motifs present in the 3′UTRs of the two transcripts are likely to be responsible for regulation of tissue-specific expression. Sequence of the short 3′UTR is highly similar in the rat and human genes (up to 90% of homology in the conserved region), but in the further region of the long 3′UTR similarity abruptly decreases, which is consistent with the observation that there is only one (short) Pol β transcript in the human cells. Sequence similarity indicates that regulatory motifs present in the short 3′UTR of the rat Pol β transcript may act in the same manner and bind similar factors in human cells. Structure prediction analysis () of the short 3′UTR revealed the presence of a putative hairpin element, ∼50 nt upstream of the polyadenylation sequence and 40–50 nt downstream of the termination codon (). There are several examples of transcripts with similar structures within the 3′UTR, constituting -acting regulatory elements, involved in mRNA localization. Structural motifs (hairpins) present within the 3′UTR of c-fos (), c-myc (), MT-1 [metallothionein-1, ()], slow troponin C () and vimentin () mRNAs have been shown to be involved in targeting these mRNAs to the perinuclear cytoplasm and, presumably, anchoring them in this location by binding to cytoskeletal elements (). Hairpin elements within the 3′UTRs were also reported to stabilize transcripts by binding factors that protect against nuclease cleavage [e.g. binding of IRP protein to the IRE sequence in the 3′UTR of transferrin receptor mRNA, ()] and to re-program translation as in the case of the SECIS element which directs an insertion of selenocysteine into in-frame UGA codons (). In this report, we confirm the existence of the hairpin structure within the 3′UTR of the Pol β mRNA and demonstrate that this element influences the expression of a reporter gene. We describe the identification of a protein factor binding to this motif—Hax-1, an anti-apoptotic, cytoskeleton-related protein, which is known to bind a hairpin structure within the 3′UTR of vimentin mRNA. We demonstrate that binding occurs only for a Hax-1 dimer, though RNA binding is not a prerequisite for the dimerization itself. We confirm the importance of the hairpin structure for binding of Hax-1 by its mutagenic disruption, which impairs the RNA–protein interaction. We also report strong association of Hax-1 with the nuclear matrix, which is a novel finding, consistent with its transcript-binding properties. Taken together, these data suggest that the hairpin element within the Pol β 3′UTR represents a novel motif important for post-transcriptional regulation of expression. The template for transcription encompassing the whole short 3′UTR of rat Polβ (208 nt) was synthesized by PCR with a forward primer containing the T7 RNA polymerase promoter sequence (5′-TAATACGACTCACTATAGGGCCTGCCCCACCCAGGCCT) and reverse primer (5′-AAACCATGGTACTGCGATC). The PCR was performed with the plasmid bearing the Pol β short 3′UTR sequence (pGEM-4Z/H), in the following conditions: 94°C for 1 min followed by 35 cycles at 94°C for 1 s, 60°C for 1 s and 72°C for 30 s. The transcription reaction was carried out in 50 μl containing 20 pmol of PCR product, 500 μM rNTPs, 3.3 mM guanosine, 40 U of ribonuclease inhibitor RNase Out (Invitrogen) and 400 U of T7 RNA polymerase (Ambion). The reaction was carried out at 37°C for 2 h and the transcript was purified from a denaturing 10% polyacrylamide gel, and 5′-end-labeled with T4 polynucleotide kinase and [γP]ATP (4500 Ci/mmol; ICN). The labeled RNA was again purified by electrophoresis on a denaturing 10% polyacrylamide gel. Prior to structure probing reactions, the labeled RNA was subjected to a denaturation and renaturation procedure in a buffer containing 2 mM MgCl, 80 mM NaCl, 20 mM Tris–HCl pH 7.2 by heating the sample at 80°C for 1 min. and then slowly cooling to reaction temperature. Limited RNA digestion was initiated by mixing 5 μl of the RNA sample (50 000 c.p.m.) with 5 μl of a probe solution containing lead ions, nuclease S1 or ribonucleases T1, T2 or Cl3. The reactions were performed at 37°C for 10 min. and stopped by adding an equal volume of stop solution (7.5 M urea and 20 mM EDTA with dyes) and sample freezing. To determine the cleavage sites, the products of the RNA fragmentation reaction along with the products of alkaline hydrolysis and limited T1 nuclease digestion of the same RNA molecule were separated on 10% polyacrylamide gels containing 7.5 M urea, 90 mM Tris-borate buffer and 2 mM EDTA,. The alkaline hydrolysis ladder was generated by the incubation of the labeled RNA in formamide containing 0.5 mM MgCl at 100°C for 10 min. The partial T1 ribonuclease digestion of RNAs was performed under semi-denaturing conditions (10 mM sodium citrate pH 5.0; 3.5 M urea) with 0.2 U/μl of the enzyme and incubation at 55°C for 15 min. Electrophoresis was performed at 1500 V and was followed by autoradiography at −80°C with an intensifying screen. 5′-CCTTTGCTATGTAATTGGGTGTTTTAGGTGATTGCCTCTTC-3′ 5′-GAAGAGGCAATCACCTAAAACACCCAATTACATAGCAAAGG-3′ A fragment of 208 bp, containing the whole short Pol β 3′UTR sequence, generated as described in Structural analysis of RNA, was inserted downstream of Firefly luciferase in the pCMLuc vector (pCM2 derivative, a gift from Dr D.Weil, Institute Andre Lwoff, Villejuif, France) into the EcoRI site present in the polylinker, generating pCMLucH. The same fragment containing a mutation of the hairpin-forming region was cloned into the EcoRI site of pCMLuc, generating pCMLucHmut. The rat hepatoma FTO-2B cell line was grown on DMEM (Invitrogen) supplemented with 10% FBS (Invitrogen). Transfection of FTO-2B was performed using Lipofectamine 2000 (Invitrogen), according to the manufacturer's guidelines. Two hundred nanogram of the appropriate plasmid was used for transfection, performed in 50 μl of medium in a 96-well plate. The activity of Firefly luciferase in lysates prepared from cells transfected with pCMLuc, pCMLucH and pCMLucHmut was measured using a luciferase reporter assay system (Dual-Glo, Promega). Assays were performed according to the manufacturer's instructions. Light emission was measured with LumiCount microplate luminometer (Packard). Transfection efficiencies were normalized by co-transfection with phRL-CMV vector (Promega) containing the luciferase. Total RNA from transfected cells was isolated using the NucleoSpin RNA II kit (Macherey-Nagel). The first strand of cDNA was obtained with SuperScript Reverse Transcriptase (Invitrogen) from 1 μg of RNA, according to the manufacturer's instructions. Primers used in the experiment were designed to amplify 209 bp of the firefly luciferase transcript (forward: 5′-TCGTTGACCGCCTGAAGTCT-3′, reverse: 5′-GGCGACGTAATCCACGATCT-3′) and, as a reference, 232 bp of the luciferase transcript (forward: 5′-TGGAGCCATTCAAGGAGAAG-3′, reverse: 5′-TTCACGAACTCGGTGTTAGG-3′). Quantitative PCR was performed using ABI Prism 7000 Sequence Detection System (Applied Biosystems). SYBR Green PCR Master Mix (Applied Biosystems) was used for detection. PCR was performed in the following conditions: precycling hold at 95°C for 10 min, cycles: 95°C, 30 s, 56°C, 30 s, 72°C, 30 s, up to 40. The ΔΔC method was used for quantity calculations (). The slope of the validation curve was <0.1, which ensures, that target and reference efficiencies are approximately equal. #text xref #text A hybrid RNA construct (pRH5′H) containing the sequence of Pol β 3′UTR (from +3 to −208 downstream of the stop codon) was generated by inserting the Pol β sequence fragment into the unique SmaI site downstream of the phage MS2 sequence in the pRH5′ vector (Invitrogen). The pRH5′H (H for hairpin) and cDNA library plasmids were sequentially transformed using the lithium acetate method () into the yeast R40C-W strain, which contains both HIS3 and β-galactosidase (β-gal) promoters integrated into its genome. Transformed yeast were plated on YC medium lacking tryptophan, uracil and histidine (YC-WUH) and containing 5 mM 3-aminotriazole (3-AT). Transformants were replicated on 25 mM 3-AT and after 3–6 days large colonies were picked for analysis. Selected transformants were screened out by the expression of -galactosidase using the colony lift assay with 5-bromo-4-chloro-3-indolyl---galactopyranoside (X-gal) as the substrate (CLONTECH Yeast Protocols Handbook). To eliminate RNA-independent false positives, plasmids were isolated () from selected blue colonies and used to transform electrocompetent KC8 cells. This bacterial strain enables complementation of yeast auxotrophy marker genes. Colonies with plasmids bearing the yeast TRP1 gene (pYESTrp1 cDNA clones) were isolated on M9 minimal medium lacking tryptophan. Plasmids were isolated, verified by restriction analysis and re-transformed into R40C-W, previously transformed with pRH5′H. Double transformants were checked again by the β-gal colony lift assay. This method of selection eliminated false-positive clones, leaving RNA-dependent positives. Positive clones were sequenced and analyzed by BLAST search. Positive control vectors for the screen: pRH3′/IRE and pYESTrp1/IRP, were taken from RNA-Protein Hybrid Hunter Kit (Invitrogen). Western blotting was performed with ECL Plus (Amersham) according to the manufacturer's instructions. Monoclonal anti-HAX-1 antibody (BD Bioscience) was used in 1:250 dilution. The secondary antibody, goat-anti-mouse (Pierce) was used in 1:2000 dilution. Anti-BclX mouse antibody (Clone 2H12, Sigma) was used in dilution 1:80 and anti-proteasome 20S subunit alpha 1,2,3,5,6,7 (PW8195, Affiniti Research) mouse antibody was used in dilution 1:1000, both with secondary goat-anti-mouse antibody (Pierce) used in dilution 1:15 000. Anti-matrin 3 antibody (a kind gift from Ronald Berezney, State University of New York at Buffalo, Buffalo) was used in 1:1000 dilution with secondary rabbit-anti-chicken antibody (Pierce) used in dilution 1:15 000. Polyclonal rabbit anti-H2B antibody (UPSTATE) was used in dilution 1:5000 with secondary goat-anti-rabbit antibody (BioRad) used in dilution 1:5000. Hax-1 cDNA was amplified with specific primers (forward: 5′- CCAGGATCCGAGCGTCTTTGATCTTTTCCGAGGCT-3′, reverse: 5′-GCTTGTCGACTCGGGACCGAAACCAACGTCCTA-3′) and the PCR product was cloned into pET201 (a gift from Csaba Koncz, Max Planck, Institute for Plant Breeding Research, Cologne). pET201 is a non-commercial vector representing a derivative of pET vector series used for expression of recombinant proteins fused to N-terminal bacterial thioredoxin and C-terminal 6xHis tags. pET201 vector encoding for thioredoxin was used concurrently to produce thioredoxin as a control protein. Bacteria were grown at 37°C to an optical density of 0.5 (OD), followed by induction with 1 mM IPTG for 4 h. Recombinant proteins were purified under native conditions according to the Qiagen protocol for His-tag protein purification. Eluted proteins were analyzed by 12% SDS-PAGE and stained by Coomassie. Western blot analysis was carried out using monoclonal anti-Hax-1 antibody (BD Bioscience). Purified proteins were stored in 50% glycerol at −20°C. The 208 bp of Pol β 3′UTR, encompassing the whole short 3′UTR sequence with the hairpin encoding region, was cloned into pGEM-3Z and pGEM-4Z vectors (Promega) in both orientations, generating a set of templates for transcription of sense and antisense mRNAs. transcription was carried out with polymerases SP6 and T7 (Promega). [α–P]UTP (3000 Ci/mmol) was used for labeling. RNA integrity was controlled by polyacrylamide gel electrophoresis. Labeled transcripts were gel-purified (5% polyacrylamide, 8 M urea). transcribed RNA with heparin (5000 U/ml, Sigma) and 160 U of RNAse inhibitor (RNase OUT, Invitrogen) was incubated on ice for 15 min in cross-link buffer (20 mM Tris, pH 7.9, 2.5 mM MgCl) with 0.5 μg of recombinant Hax-1 protein, BSA (Sigma) or thioredoxin (recombinant thioredoxin purified from bacteria using pET201 vector in the same conditions as Hax-1 protein). Cross-linking was performed on ice for 30 min in the UV Stratalinker 1800 (Stratagene). After cross-linking, the reaction was incubated in the presence of RNase A (final concentration 0.6 μg/ml) for 20 min. in room temperature. Laemmli loading buffer was added and the samples were heated to 95°C for 5 min. Samples were analyzed on SDS-PAGE, with Prestained SDS-PAGE Standards, Kaleidoscope (Bio-Rad). Purified recombinant Hax-1 protein (1 μg for the western blot, indicated amount for silver staining) was suspended in 100 μl of sample buffer (SB: 10 mM HEPES, pH 7.4, 120 mM potassium acetate, 2.5 mM MgCl, 1.6 mM dithiotreitol), incubated for 15 min in room temperature to establish monomer–dimer equilibrium and then incubated for 20 min in 22°C with 0.00125% glutaraldehyde (Sigma). The reaction was stopped by the addition of 50 μl of 3-fold concentrated loading buffer (150 mM Tris pH 6.8, 6 mM EDTA, 6% SDS, 30% glycerol, 1 M urea, 0.003% bromophenol blue). Samples were heated (90°C) and separated on the SDS-PAGE. Proteins were detected by silver staining () and western blot. The lanes in the silver-stained gel were scanned and the areas under the peaks were quantified using Multi-analyst (Bio-Rad). The results were analyzed by nonlinear regression analysis to determine an approximate dimer dissociation constant, as described in (). The filter-binding assay was performed by a modified method described by Wong (). Briefly, radiolabeled, transcribed RNA at a concentration of ∼10 nM was heat denatured, allowed to refold and incubated for 15 min in room temperature in binding buffer (20 mM Tris, pH 7.9, 100 mM KCl, 2.5 mM MgCl) containing heparin (5000 U/ml, Sigma) and 160 U of RNAse inhibitor (RNase OUT, Invitrogen) with purified Hax-1 protein. The reaction mix was loaded onto a presoaked nitrocellulose membrane (0.45 μm, Osmonic Inc.) on top of a nylon membrane (Hybond-N, Amersham) and filtered under pressure in a slot-blot apparatus. Following filtration, each filter was dried and quantitated on a PhosphorImager (BioRad) using QuantityOne software. Dissociation constants () for the RNA–protein complexes were obtained by fitting the empirical data to a sigmoidal curve by nonlinear regression analysis, using Maxima 5.4. Fitting was performed in respect to dimer concentration calculated as a function of the total protein concentration, as in (). Four hundred milligram of pulverized rat testes were homogenized in 1 ml of 2× MEB buffer (100 mM Tris-HCl pH 7.5, 20% glycerol, 4 mM DTT, 10 μl/ml protease inhibitor cocktail [Roche]) in a Potter homogenizer, filtered on 50 μm Nitex membrane (Tetko) and centrifuged at 2000, 4°C. The pellet (nuclear fraction) was washed two times with 0.5 ml of 2× MEB buffer. Nuclei were gently resuspended in 0.5 ml of 2× MEB (small aliquot was analyzed by DAPI staining), passed through a needle and incubated 30 min on ice with 10 μl of DNase I (Warthington). The supernatant was centrifuged at 10 000, 4°C (mitochondrial fraction). After separation of the mitochondrial fraction, the supernatant was centrifuged at 100 000 in an ultracentrifuge and the supernatant (cytoplasmic fraction) was collected. In the first step of our study, we undertook the determination of the structural features of the entire short 3′UTR of the rat DNA polymerase β transcript (208 nt). For this purpose, we used limited fragmentation of the 5′-end-labeled RNA with six well-characterized biochemical structural probes: lead ions and five enzymes (A). Nuclease S1 and lead ions have no documented nucleotide specificity, while ribonuclease T2 and A recognize all nucleotides but have a higher activity for adenosines and pyrimidines, respectively. RNase T1 exhibits specificity exclusively for guanosines and RNase Cl3 digests C-residues only. All the above structural probes recognize flexible and single-stranded regions in the RNA structure. Prior to the probing experiments, the 5′-end labeled transcript of the Pol β short 3′UTR was analyzed by non-denaturing gel electrophoresis. The result of this test suggests that only a single conformer is formed by the analyzed RNA molecule (migrated as a single band on a gel, not shown). Probing data demonstrated that the short 3′UTR of Pol β transcript forms three separate structural modules, designated M1–M3 (B). Module M1 represents a small hairpin structure with a C-rich terminal loop. The 3′-part of this structure is efficiently hydrolyzed with lead ions, even at a low concentration of the probe, which suggests that M1 hairpin has binding capacity for this metal ion. Module M3 contains the polyadenylation signal, followed by the cleavage and polyadenylation site, so only a part of it is present in the mature short Pol β transcript, which implies minor importance of the whole M3 module as a post-transcriptional regulatory element. Module M2 is formed by nucleotides located between bases G29 and C96 of the analyzed transcript and is composed of three helical regions, two 6 bp and one 14 bp in length. Each of these helical regions is resistant to enzymatic digestion and lead ion hydrolysis. The three helixes are separated by two asymmetric, internal loops (b and c), which are mapped very well by all probes used. The longest helical region includes as many as six non-WC, U-G and G-U base pairs. Four of these exist as two tandems: U-G, G-U and U-G, U-G, which are known to be potential metal- or protein-binding sites. The hairpin structure contains a small 3-nt terminal loop (5′-UAU), which is also well recognized by both lead ions and enzymes (A and B). The M2 hairpin structure is conserved among species (), which suggests its significant role in regulation of transcript fate. In order to demonstrate the essential role of the evolutionarily conserved hairpin structure in the regulation of Pol β expression, we carried out mutagenesis to disrupt the M2 hairpin element. Mutagenic primers were designed based on an MFOLD () prediction of the potential effect of nucleotide change on hairpin structure (). Substitution of three C-residues for three G-residues in the stem-forming region (positions 67–69) changed the predicted free energy increment (δ) from −14.1 kcal/mol for the intact structure to −3.6 kcal/mol for the mutant. The mutated sequence (Hmut) was used in luciferase reporter assays and served as a template for transcription to generate mRNA for subsequent crosslink and filter-binding analysis. The influence of the M2 hairpin structure within the short 3′UTR of Pol β mRNA on expression was analyzed utilizing a luciferase reporter system. The rat hepatoma cell line FTO-2B was transfected with reporter constructs bearing the Firefly luciferase gene appended by the Pol β 3′UTR sequence containing the M2 hairpin element, (pCMLucH) and by the same sequence containing a structure-disrupting mutation in the hairpin-forming region (pCMLucHmut). A vector bearing the unmodified luciferase gene served as a control (pCMLuc). To asses mRNA levels of the reporter, quantitative PCR was performed for cDNA preparations obtained from cells subjected to the same transfections as for luciferase assays. The results () show a significant decrease of mRNA levels for the construct with the hairpin structure (H), compared with almost unchanged mRNA levels for the mutated hairpin (Hmut). These differences, however, are not reflected at the protein level: both constructs (H and Hmut) caused an increase in luciferase expression, although only for the Hmut is this increase significant. The relative increase of protein levels in respect to mRNA levels is therefore more than 3-fold for the construct with the intact hairpin, while only a slight relative increase (∼1.7-fold) was observed for the mutated construct. These changes in expression indicate a complex post-transcriptional regulation in which the hairpin element has a key function. In order to identify proteins interacting with the M2 hairpin element present within the 3′UTR of the Pol β transcript, we performed a yeast three-hybrid screen of a rat cDNA library. The IRE–IRP interaction served as a positive control, while negative controls consisted of empty pRH5′ and pYESTrp1 vectors and single transformants of pRH5′H ‘bait’ plasmid or of the protein-hybrid clone isolated from the library. Out of 63 positive clones obtained after the first selection, only 11 were RNA dependent, and these were sequenced and analyzed. Of these, only one was in the proper reading frame, had the correct in-frame orientation and represented a coding sequence—the terminal 622 bp of the rat mRNA(corresponding to the last 150 aa of the protein), (). This clone exhibited strong growth on 25 mM 3-AT medium without histidine and tested positive in the β-gal plate test (). In order to confirm the results of the 3-hybrid screen, analysis of binding was carried out using purified Hax-1 protein. The full-length protein was overexpressed in and purified on a Ni-NTA matrix. UV-cross-linking was performed using the purified recombinant Hax-1 and o transcripts of the hairpin-containing region (H), a control antisense transcript of the same region (A), and the mutated transcript (Hmut). Bovine serum albumin and purified thioredoxin served as controls for interaction specificity. Hax-1 demonstrated a specific interaction only with the hairpin-containing RNA, and this transcript did not interact with either BSA or thioredoxin (). The cross-linked band migrated at a molecular weight of ∼100 kDa, which corresponds to a recombinant Hax-1 dimer (the recombinant protein has a higher molecular mass than endogenous Hax-1 [35 kDa], due to its fusion with the thioredoxin sequence). The protein dimer is likely formed during the UV-cross-linking procedure, as has been previously documented (). No band corresponding in size to monomeric recombinant Hax-1 (. 50 kDa) was observed, which indicates that the protein interacts with RNA exclusively in the form of a dimer. To confirm that Hax-1 forms a dimer, purified recombinant protein was used for chemical cross-linking with glutaraldehyde. As indicated in , monomers with a molecular mass of around 50 kDa were detected, as were dimers with a molecular mass around 100 kDa. Since dimer formation was detected in an experiment performed with purified protein, one can conclude that the RNA molecule is not necessary for dimerization. As deduced from the UV-cross-linking experiments, RNA is bound exclusively by dimeric Hax-1, hence the dimerization rate may represent an important factor for RNA–protein complex formation. Glutaraldehyde cross-linking with increasing protein amounts (A) allowed for the estimation of the monomer–dimer equilibrium dissociation constant () at 13.5 μM ± 5.5. To corroborate the UV-cross-link results, filter-binding assays were performed with increasing amounts of Hax-1 protein and a constant amount of the same test RNAs (hairpin-containing region—H, antisense—A and mutant—Hmut). Data were analyzed on a Klotz plot () and the apparent for each complex was calculated. The fit of a nonlinear binding curve to the experimental data points was best when calculations were performed in respect to dimer concentration, which indicates that the RNA interacts only with a dimeric form of the protein, thus confirming the UV-cross-link results. The approximate established from the titrated glutaraldehyde cross-linking experiments (13.5 μM ± 5.5) was in the same range as the predicted by curve-fitting (fitting performed simultaneously for the H and Hmut data point series in respect to variable and constant values yielded a of 10 μM ± 4). Transcript H showed the greatest affinity for Hax-1, with a of 28 nM ± 7, and significantly lower binding affinity was observed for mutant RNA ( = 170 nM ± 40), and very little affinity for the antisense transcript ( > 1000 nM). While Hax-1 mitochondrial localization has been confirmed in many reports (), this is not obviously consistent with its transcript-binding properties. Several other reports have shown localization of the protein to the endoplasmic reticulum (,,), apical membrane of hepatocytes () and nuclear envelope (). In order to establish Hax-1 cellular localization in rat cells, we performed organellar fractionation and subsequent fractionation of nuclei, followed by SDS-PAGE and western blot. These experiments confirmed the presence of Hax-1 in mitochondria, but also indicated its localization in the nucleus (A) with only traces of Hax-1 detectable in the cytoplasm. Subsequent nuclear fractionation performed with two different methods, revealed that Hax-1 is present in the fraction containing nuclear matrix proteins (B and C), while it was not detected in the other fractions containing soluble and chromatin-associated proteins. To ensure proper quality of nuclear fractions, western blots with appropriate marker proteins were performed. Fraction 1 in both methods contains soluble, chromosomal proteins, released after DNaseI treatment, represented here by histone H2B. These proteins were washed out in subsequent steps of the preparation (fractions 2–3). The nuclear matrix fraction (fraction 4) was probed with a matrix-specific protein matrin 3 antibody (). Close association of Hax-1 with the nuclear matrix sheds new light on its mRNA-binding capacity and may indicate its role in regulation of transcript fate. Stable secondary structures in 3′UTRs have been shown to play a role in mRNA sorting and localization (). Some of them have been also reported to influence mRNA stability (). The existence of a stable structural motif in the 3′UTR of the rat Pol β mRNA was predicted previously () and it was speculated that it may have a regulatory role. In the present work, secondary structure analysis by lead ion hydrolysis and enzymatic digestion revealed the existence of several motifs in the analyzed sequence, namely, a region of strong lead ion binding (M1), a hairpin-forming and highly evolutionarily conserved region (designated as M2) and the region containing polyadenylation sequence (M3). Only M2 is evolutionary conserved, the rest of the untranslated region has high interspecies sequence variation, which suggests lesser functional importance. The sequence in the conserved region exhibits 86.8% identity to the homologous human sequence (100% in the upper stem region), which may indicate a similar role of this element in human cells. We show here that the hairpin structure within the 3′UTR influences the expression of a luciferase reporter gene. Lowering of luciferase mRNA levels for the construct with an intact hairpin structure in contrast with almost unchanged mRNA levels for the mutated structure indicate that the hairpin is in fact an RNA destabilizing element. However, at the protein level, expression for both constructs exceeds the expression of the control. This indicates the presence of at least two regulatory events in which the hairpin structure is involved: (i) mRNA degradation, (ii) enhanced mRNA transport (possibly coupled with mRNA stabilization) and/or enhanced translation. These effects may be the consequence of the competitive binding of -acting factors for the binding site within the hairpin. Thus, identification of the hairpin-binding factors is important for assessment of its physiological role. A yeast three-hybrid screen identified Hax-1 as the binding partner for the hairpin structure of the Pol β 3′UTR. Hax-1 is an RNA-binding protein, known as an anti-apoptotic factor () associated with cytoskeletal proteins and involved in cell migration (,). Hitherto only one RNA target of Hax-1 has been identified: vimentin mRNA (). Data from several reports suggest that Hax-1 binding to the hairpin element within the 3′UTR of the vimentin transcript plays a role in its localization to the perinuclear cytoplasm (,,). The importance of proper vimentin transcript localization is illustrated by the fact that its misdirection alters cell morphology and motility (). The identification of a second RNA target of Hax-1—the hairpin element present in the Pol β transcript—raises the question as to the identity of other mRNA targets of the protein. One may speculate that there is a pool of such mRNAs, especially because vimentin and Pol β are not functionally or evolutionarily related nor are they involved in the same pathway. Comparison of the hairpin motifs in vimentin and Pol β mRNAs did not reveal any significant similarities, which could suggest substrate requirements for Hax-1 binding. The presence of U-rich single-stranded regions (vimentin: AGUUUU in the terminal loop, Pol β: AGUUAU in the internal loop) represents the only similarity between the two structures, but the helical regions adjacent to this U-rich sequence in the Pol β mRNA are not evolutionarily conserved. The lack of similarities suggests that Hax-1-binding mechanism and affinities might be different for these two structures. It is an open question if Hax-1 is in fact a destabilizing factor or if its actions in respect to the Pol β transcript consist of mRNA stabilization, possibly coupled with its transport and/or localized translation. The fact that Hax-1 binds to the instability element (the hairpin) and does not bind to the stable mutated transcript suggests a role in mRNA destabilization. However, data concerning the role of Hax-1 in the regulation of vimentin transcript, indicating that it facilitates its transport to the perinuclear space, are rather contradictory to its potential functions in mRNA degradation. Another possible explanation is that Hax-1 may stabilize otherwise unstable mRNAs and facilitate their transport or enhance the translation rate, conferring elevated luciferase levels in respect to mRNA levels. If this latter case is true, considering that Hax-1 has been identified as the same -acting factor for both vimentin and Pol β mRNAs, the hairpin element within the 3′UTR of the Pol β transcript may also represent a motif directing mRNA to the perinuclear space. Perinuclear localization of certain transcripts, and their subsequent translation at this site, could facilitate an efficient nuclear import of newly synthesized proteins (,,). DNA polymerase β as a nuclear protein could also benefit from such a mechanism. To present a satisfactory explanation of the role of Hax-1 transcript binding in the cell, one has to resolve the question of the subcellular location of Hax-1. Hitherto, Hax-1 has been reported to localize predominantly in the mitochondria (,,) but it has also been detected in the endoplasmic reticulum (,,), apical membrane (), lammelipodia () and nuclear envelope (). In the last case, the presence of Hax-1 in the nuclear envelope was interpreted as a consequence of its association with intracellular membranes (by its putative transmembrane domain), as a continuum of endoplasmic reticulum localization. Our data reveal for the first time, that Hax-1 is associated with the nuclear matrix, which is coherent with its transcript-binding capacity and supports the notion of its role in post-transcriptional regulation. Some new data support Hax-1 association with the nucleus. In a recent report, Kawaguchi () shows that Hax-1 is present in the nucleus of systemic sclerosis fibroblasts (but not in normal fibroblasts) and is involved in pre-IL-1α translocation into the nucleus—a process blocked by inhibition of Hax-1. This activity is contradictory to previously reported Hax-1 involvement in the cytoplasmic retention of IL-1α () and EBNA-LP (). The role of Hax-1 in protein import into the nucleus might suggest that it is shuttling between the nuclear matrix and perinuclear space, transporting different cargo molecules. Participation of the hairpin element in the binding between the Pol β 3′UTR and Hax-1 was demonstrated by UV-cross-linking, in which the transcript with an intact hairpin structure bound to the protein, whereas a transcript with a disrupted hairpin did not. Cross-linking also revealed that Hax-1 binds to mRNA only in the form of a dimer. The presence of a band of a molecular weight of 100 kDa indicates that UV exposure cross-linked a complex consisting of RNA bound to a protein dimer. RNA–monomer complexes were not detected. Data from chemical cross-linking lead us to conclude that the RNA molecule is not necessary for dimerization. However, given that only a small percentage of protein dimerizes in the absence of mRNA, the possibility that RNA binding influences the dimerization rate is tempting, and remains to be assessed. Filter-binding experiments, complementing the cross-links, showed that the binding, though not completely abolished, is substantially weakened for a mutated sequence with a disrupted hairpin structure. Only the C-terminal part of Hax-1 appears to be involved in mRNA binding, since a truncated peptide bearing only the last 150 aa of the protein suffices for binding the hairpin-containing element, as demonstrated in our experiments utilizing the yeast three-hybrid system (). Considering that an interaction with RNA occurs only for dimeric Hax-1, these findings suggest that the domain responsible for the dimerization is also located in the C-terminal part of the protein. From these results, we deduce that BH domains (BH1 and BH2) present in the N-terminal part of the protein do not take part in the dimerization. BH domains and a transmembrane domain (present at the C-terminus of Hax-1) represent the features of Bcl-2 family proteins, but there is no significant sequence homology between these proteins and Hax-1 - only a weak, partial homology to pro-apoptotic Nip3 (). Even though BH domains are known to be important for oligomerization of the proteins from the Bcl-2 family, data seem to exclude the possibility that they are responsible for dimerization of Hax-1. We have demonstrated that the hairpin structure within the 3′UTR of the Pol β mRNA represents a post-transcriptional regulatory element. Hax-1 protein, which binds to this element, appears to be an important -acting factor, though the exact mechanism of Hax-1-mediated regulation remains to be elucidated, and the mechanisms implicating its role in control of mRNA stability, transport and/or localized translation must be verified by subsequent experiments. Hax-1 is a multifunctional protein, active in different cellular compartments and involved in various cellular processes. Attention has been focused on its functions in apoptosis and regulation of cell motility, but it seems that it has a more complex mode of action and plays a regulatory role in the context of its specific mRNA targets.

Alternative pre-mRNA splicing is one of the central mechanisms for the regulation of gene expression in eukaryotic cells. It allows the generation of functionally distinct proteins from a single gene. It has been estimated that 40–60% of human genes are alternatively spliced. Moreover, alternative splicing is often regulated in a cell-type, tissue or developmentally specific manner [for reviews, see ()]. The splicing reaction is carried out by the spliceosome, a large ribonucleoprotein complex containing five small nuclear ribonucleoproteins (snRNPs) and many protein splicing factors. Spliceosome assembly occurs in an ordered manner within each intron. The initial step for spliceosome formation is assembly of early (E) complex (,): U1 snRNP interacts with the 5′ splice site, SF1 (splicing factor 1) binds to the branch point, and the U2AF65/35 heterodimer binds to the pyrimidine tract and the 3′ splice site. In an ATP requiring step, U2 snRNP tightly associates with the branch site, generating the A complex. Subsequently, the U4/U6/U5 tri-snRNPs associate to the A complex to form the B complex. After RNA–RNA rearrangements occur, the catalytically activated spliceosome is formed. During these rearrangements, the U1 and U4 snRNPs dissociate and the U6 snRNA contacts with the 5′ splice site and U2 snRNA. This is the catalytic C complex spliceosome in which the two -esterification reactions of splicing occur, resulting in exon ligation and lariat intron release (). Spliceosome assembly is regulated by several non-spliceosomal RNA-binding proteins, such as SR and hnRNP proteins. SR proteins usually play key roles in constitutive and alternative splicing, by mediating splicing activation via binding to exonic splicing enhancers (ESEs). In contrast, hnRNP proteins act as splicing repressors via binding to exonic splicing silencers (ESSs) and intronic splicing silencers (ISSs) (). These proteins are extensively studied for their effect to spliceosome assembly in alternative splicing, and are thought to affect the initial step of spliceosome assembly, the E complex formation. Recently, several tissue-specific splicing regulators have been reported. For example, a neuron-specific RNA-binding protein, Nova-1, binds to the RNA sequence UCAUY and regulates the alternative splicing of several genes such as glycine receptor a2 (,). The CELF (CUG-BP and ETR3-like factors) family proteins are implicated in regulation of tissue-specific splicing of several genes, including cTNT, IR and α-actinin (). In our previous study, we identified vertebrate homologs of the Fox-1 protein in zebrafish and mouse. Fox-1 is an RNA-binding protein that contains an RNA recognition motif (RRM). In mouse, Fox-1 is expressed in brain, heart and skeletal muscle. Our SELEX experiments showed that zebrafish Fox-1 protein binds specifically to the pentanucleotide GCAUG (). Interestingly, it has been reported that (U)GCAUG is essential for the alternative splicing of several genes (). Furthermore, a recent computational analysis revealed that the UGCAUG element is overrepresented in the downstream introns of neuron-specific exons and is conserved among vertebrate species (). Fox-1 induces muscle-specific exon skipping through binding to the GCAUG repressor element upstream of alternative exon in the human mitochondrial ATP synthase γ subunit (hF1γ) gene (). In the case of calcitonin/CGRP, two copies of UGCAUG in the upstream intron and the regulated exon are essential for the induction of exon skipping by Fox-1 or its paralog Fox-2 (). In contrast, exon inclusion in fibronectin, non-muscle myosin heavy chain (NMHC)-B, c-src and FGFR2, 4.1R is induced by Fox proteins via the (U)GCAUG enhancer element in the downstream intron (,). Thus, in the known cases so far, the (U)GCAUG element that resides in the intron upstream of alternative exon functions as a repressor element, whereas the element that activates exon inclusion is found in the intron downstream of the alternative exon. Thus, it is likely that Fox proteins function as both splicing repressor and activator, depending on where they bind relative to the affected exon. However, little is known about the molecular mechanisms of how Fox proteins regulate such alternative splicing. To examine the molecular mechanism of exon skipping by Fox-1, we studied its effect on the spliceosome assembly using the hF1γ gene as a model. Here we report that Fox-1 induces exon 9 skipping by repressing splicing of the downstream intron 9 via binding to the GCAUG repressor element in intron 8. The splicing efficiency of intron 8 was not affected much by Fox-1 protein. splicing analyses show that Fox-1, by binding to the GCAUG element in intron 8, prevents formation of the pre-spliceosomal E complex onto intron 9. Such repression by Fox-1 represents a novel mechanism for splicing regulation by tissue-specific splicing regulators. In addition, we identified a region of the Fox-1 protein that is required for inducing the exon skipping, suggesting that this region plays a key role in interacting with other splicing factor(s) to regulate alternative splicing. The pCS2+MT mouse Fox-1/A2BP (NM_021477) was described previously (). The coding sequence of mouse Fox-1/A2BP was cloned into pCS2 vector containing Flag peptide (MDYKDDDDK). The pCS2+MT F-A mutant was constructed using chimeric PCR amplification, mutation was induced into the RNP motif of Fox-1 (AAGGGATTTGGTTTCGTAACTTTC to AAGGGATTTGGTGTAACTTTC). For F-A mutant, we used Fox-1-S, F-A-1, F-A-2, Fox-1-AS primers. The hF1γL, hF1γS and hF1γSmt mini-genes were described previously (). To construct the hF1γ 5′SSmt and hF1γBPmt mini-genes, base-substitution were introduced into the 5′ splice site in intron 9 (gtaaagttca to caaaacatca) and the branch point in intron 8 (tcttgac to tcgcgug), respectively, by chimeric PCR amplification. To construct the Ex8-9 and Ex9-10 mini-genes, we used hF1γS mini-gene as a template for PCR amplification, and the amplified fragments were cloned into pCMV sport vector (Life Technologies). The Ex8-9 mt and Ex9-10 mt mini-genes were constructed in the same manner using hF1γSmt mini-gene. CV-1 cells were maintained in DMEM supplemented with 10% FBS. Transfection was performed by the calcium phosphate DNA precipitation method as described previously (). The myc fusion proteins expressed in transfected cells were examined by western blotting using anti-myc anti-body (cMyc 9E10; Santa Cruze Biotechnology). As a loading control of western blotting, U2AF65 was detected by anti-U2AF65 antibody (Sigma). To analyze splicing products from hF1γ mini-gene by RT-PCR, the following F1-2903 and F1-2389 oligonucleotides were used. For hF1γS, hF1γSmt, hF1γmt+3GCAUG, hF1γ 5′SSmt, hF1γ branch mt, Ex8-9 and Ex8-9 mt mini-genes, F1-2903 and T7 primer were used. For Ex9-10 and Ex9-10 mt mini-genes, Ex9-10S and T7 primers were used. Human embryonic kidney cells (HEK293) were grown in DMEM containing 10% FBS. For preparation of nuclear extracts, HEK293 cells grown in 150 mm dishes were transfected with 12 μg plasmids/dish using TransIT-293 Transfection Reagent (Mirus). Nuclear extracts were prepared from HEK293 cells transfected with pCS2 expression plasmids encoding Flag peptide or Flag-tagged mouse Fox-1 protein according to the small-scale nuclear extraction procedure (). Expression of Flag mFox-1 was confirmed by western blotting using anti-Flag tag antibody M2 (Sigma). Pre-mRNAs (2.5 × 10 c.p.m.) were incubated in 5 μl of reaction mixture containing 1.6 mM MgCl, 0.5 mM ATP, 20 mM creatine phosphate and 3 μl of nuclear extracts (1.5 μl HeLa nuclear extract and 1.5 μl transfectant HEK293 nuclear extract). After incubation, the reaction was terminated by treatment with proteinase K at 37°C for 20 min. The splicing products were extracted and separated by electrophoresis on 6% polyacrylamide gels containing 8 M urea and autoradiographed with X-ray film (RX-U, Fuji Photo Film Co.). To analyze spliceosome assembly, pre-mRNAs were incubated under the splicing condition and treated with heparin, and the spliceosomal complex was separated on 4% native polyacrylamide gels. For the E complex assembly analysis, pre-mRNAs were incubated in ATP-depleted nuclear extract without heparin treatment, and the spliceosomal complexes were separated by 1.5% native agarose gel (). In the previous study, we showed that zebrafish Fox-1 protein regulates tissue-specific splicing of several genes via the GCAUG elements, using a heterogeneous system in which zebrafish Fox-1 was expressed in mammalian cells (). In this study, we focused on mouse Fox-1 in order to reveal molecular mechanism of tissue-specific splicing in mammalian cells. As a first step, we attempted to reconfirm whether mouse Fox-1 induces tissue-specific splicing via binding to GCAUG elements using hF1γ gene, as is the case of zebrafish Fox-1. Exon 9 of the hF1γ gene is excluded from the splicing products in a muscle-specific manner (,) and four copies of GCAUG element reside in intron 8 (A). We co-transfected the hF1γ mini-gene constructs with mFox-1 expression plasmids into CV-1 cells, and analyzed RNA products of hF1γ gene by RT-PCR. When hF1γL plasmids were transfected, exclusion and inclusion of exon 9 occur almost equally. In contrast, overexpression of Fox-1 proteins promoted exon 9 exclusion (B, lanes 1 and 2). The F-A mutant, in which an amino acid mutation was introduced in RNP1 of mFox-1, was not able to bind to RNA (data not shown). Fox-1 F-A mutant protein could not induce exon 9 skipping of hF1γ gene (B, lane 3), although the protein was properly expressed and localized to nucleus (C and data not shown). These results indicated that Fox-1 promotes exon 9 skipping of hF1γ in a manner depending on its RNA-binding activity. Next we performed a transfection assay using various hF1γ derivative mini-genes (A). The hF1γS mini-gene, that lacks a large portion of intron 8 and hence contains only a single copy of GCAUG, was transfected with mFox-1 expression plasmids. We found that exon 9 skipping was induced by Fox-1 (B, lanes 4–6). In contrast, when the hF1γSmt mini-gene in which base substitutions were introduced into the GCAUG sequence of hF1γS was transfected, exon 9 skipping was not largely induced by Fox-1 (B, lanes 7–9). Insertion of three copies of GCAUG to the hF1γSmt mini-gene strongly restored induction of exon 9 skipping by Fox-1 protein (B, lanes 10–12). Taken together, we concluded that mouse Fox-1 induces exon 9 skipping via binding to the GCAUG element. In these experiments, however, we found that splicing efficiency between the mini-genes was somehow different. It may be due to RNA context such as a secondary structure. Alternatively, it is possible that the sequence changes in these mini-genes may affect some positive elements present in the wild-type construct. Fox-1 induces exon 9 skipping of hF1γ via binding to GCAUG, but its mechanism of action is unclear. As a first step to understand this, we examined whether the Fox-1 protein regulates the splicing of intron 8, intron 9 or both introns. Two mini-genes, Exon 8-9 and Exon 9-10, containing either intron 8 or 9, respectively, were constructed. The Exon 9-10 mini-gene contains a portion of the preceding intron 8 with the GCAUG element, in addition to exons 9 and 10 and the intervening intron 9. The branch site in intron 8 was disrupted by base substitution mutations (A and B). Since the GCAUG repressor element is located in intron 8, we expected that Fox-1 only would repress the splicing reaction of intron 8. However, transfection experiments showed that when Exon 8-9 pre-mRNA was expressed in CV1 cells, the splicing reaction of intron 8 was not affected much by Fox-1 (A, lanes 1–3). In contrast, surprisingly, splicing of Exon 9–10 pre-mRNA was strongly repressed by Fox-1 (B, lanes 1 and 2). Fox-1 F-A mutant protein could not repress intron 9 splicing (B, lane 3). Mutations to the GCAUG element in the Exon 9-10 mini-gene reduced repression of intron 9 splicing (B, lanes 4 and 5). These results indicated that Fox-1 represses splicing of intron 9, without affecting intron 8 splicing, via the GCAUG element located in intron 8. Next, we examined whether exon 9 skipping is induced by the repression of intron 9 splicing. We constructed two mutants of hF1γ mini-gene to disrupt splicing of either the upstream or downstream intron without affecting exon skipping. BPmt contains the branch point mutation upstream of exon 9 to disrupt intron 8 splicing, while the 5′ splice site of intron 9 was mutated in 5′SSmt (C). When the BPmt mini-gene alone was transfected into CV-1 cells, we detected three kinds of RNA products, corresponding to the unspliced pre-mRNA, intron 9-spliced form and exon 9-skipping form, as expected. When functional Fox-1 protein was co-expressed, the splicing of intron 9 was repressed and exon 9 skipping was induced concomitantly (C, lanes 1–3). In the case of 5′SSmt, we detected only unspliced pre-mRNA and exon 9-skipping products, irrespective of the presence of Fox-1 protein (C, lanes 4–6). Taken together, we conclude that Fox-1 induces exon 9 skipping in hF1γ by repression of intron 9 splicing. To investigate the possible molecular mechanism of repression of intron 9 splicing by Fox-1, we employed an splicing system using two kinds of reporter transcripts. Ex9-10 + 3GCAUG contains three copies of GCAUG, whereas Ex9-10ΔGCAUG does not have any GCAUG element. Nuclear extracts were prepared from the HEK293 cells expressing Flag-tagged mouse Fox-1 protein or Flag peptide alone, and mixed with HeLa nuclear extracts for splicing. Expression of Flag-tagged Fox-1 was confirmed by western blots using α-Flag tag antibody (A). Both the Ex9-10 + 3GCAUG and the Ex9-10ΔGCAUG transcripts were incubated in mock or Fox-1-overexpressed nuclear extracts. splicing showed that Fox-1 repressed the splicing of intron 9 in Ex9-10 + 3GCAUG transcripts (B, lanes 1–5). In contrast, the intron 9 splicing of Ex9-10ΔGCAUG was not repressed by Fox-1 (B, lanes 6–10). These results led us to conclude that the splicing regulation by Fox-1 is faithfully recapitulated by our system. To identify the step at which the splicing reaction of intron 9 is blocked by Fox-1, we analyzed spliceosome assembly on the Ex 9-10 + 3GCAUG and Ex9-10GCAUG pre-mRNAs . Fractionation of splicing reactions by non-denaturing gel electrophoresis can be used to show a well-defined pattern of shifts corresponding to sequential complexes along the assembly pathway. For the spliceosome assembly analysis, Ex9-10 + 3GCAUG or Ex9-10ΔGCAUG transcripts were incubated in Mock extract or Fox-1 extracts in the presence of ATP, and separated on native polyacrylamide gels. We found that spliceosome assembly on Ex9-10 + 3GCAUG transcripts occurred in the mock nuclear extract, although splicing complexes A, B and C could not be well separated in the gel (A, lanes 1 and 2). In contrast, H/E complex seemed to be accumulated in Fox-1 nuclear extracts (A, lanes 3 and 4). These complexes migrated more slowly in the presence of Fox-1. It may suggest that Fox-1 associated with the complexes through binding to Ex9-10 + 3GCAUG pre-mRNA. The H/E complex accumulation was not detected on Ex9-10ΔGCAUG transcripts (A, lanes 5–8). To clearly distinguish the H complex from the E complex, we next resolved the spliceosome E complex using a native agarose gel in the absence of ATP. The formation of E complex is ATP-independent and occurs at 30°C. Moreover, detection of E complex formation requires separation conditions lacking heparin treatment (). Under these conditions, a complex was efficiently assembled on the Ex9-10 + 3GCAUG transcripts in mock extracts. This complex disappeared by addition of heparin, indicating that it is E complex (data not shown). In contrast, E complex assembly was not detected in Fox-1 extract (B, lanes 1–4). In the case of the Ex9-10ΔGCAUG transcript, E complex formation occurs efficiently in both mock and Fox-1 extracts (B, lanes 5–8). These results indicated that Fox-1, by binding to GCAUG element in intron 8, represses intron 9 splicing by blocking formation of the pre-spliceosomal E complex on intron 9. To identify the Fox-1 protein domain required for induction of exon skipping, we created a series of deletion mutants that contain NLS to ensure proper nuclear localization (A). We confirmed expression of these mutant proteins by western blots using anti-myc antibody, although additional slow migrating bands as well as an expected band for ΔC3 protein were observed (C and data not shown). We also confirmed nuclear localization of the proteins by immunofluorescence (data not shown). Previously we reported that the C-terminal region of zebrafish Fox-1 protein, in addition to the RNA-recognition motif (RRM), was required to induce exon 9 skipping of the hF1γ gene (). The amino-terminally truncated and carboxyl-terminally truncated mFox-1 proteins were co-expressed with the hF1γL mini-gene. As a result, the intact Fox-1, ΔN (117–396 aa), ΔC3 (1–338 aa) and ΔC4 (1–355 aa) induce exon 9 skipping of the hF1γ, while ΔC1 (1–307 aa) and ΔC2 (1–326 aa) did not (B, lanes 1–5 and data not shown). These results indicate that the amino-terminal 117 amino acids are dispensable, whereas the carboxyl-terminal amino acids are involved in the repression of intron 9 splicing by Fox-1 protein. In particular, the 326–338 aa portion of the C-terminal region of Fox-1 protein may play a critical role for the negative regulation. Fox-1 protein can act as a negative regulator of alternative splicing via binding to (U)GCAUG repressor elements in upstream introns of the cassette exons. In contrast, all of the (U)GCAUG enhancer elements are found downstream of the regulated exons. Thus, Fox-1 proteins can fuction either positively or negatively, depending on where they bind relative to the affected exon. In this study, we examined the mechanism of exon skipping by Fox-1 using the hF1γ gene as a model. We found that Fox-1 protein induces exon 9 skipping by repressing the splicing of intron 9 via binding to the GCAUG repressor elements in intron 8 (A and C). Our data suggest that, for exon 9 inclusion, intron 9 excision is usually followed by intron 8 splicing of hF1γ pre-mRNA. Interestingly, Fox-1 does not affect splicing of the intron 8 containing the GCAUG repressor element (B), suggesting that Fox-1 dose not interfere with the spliceosome assembly on intron 8. It is very interesting that Fox-1 binds to an intron (intron 8 in the case of F1γ) to repress splicing of another intron (intron 9 of F1γ). Known negative regulators of alternative splicing such as hnRNP A1 and PTB (hnRNP I) inhibit splicing of the intron that they bind or mask the regulated exon via binding to both of the flanking introns (). Tissue-specific splicing regulators Nova and CELF family proteins also repress the splicing of the intron containing their binding sites (,,,). In the cases such as FGFR2 exon IIIb, it was reported that intronic silencers function across the exon (), although the molecular mechanism underlying the regulation remains unclear. Thus, the cross-exon repression by Fox-1 may represent a novel mode of splicing regulation by tissue-specific splicing regulators. It is possible that other determinants, including RNA secondly structure (), may be involved in this type of splicing regulation. Interestingly, Zhou . () recently reported that Fox-1 and Fox-2 proteins bind to two GCAUG elements in exon 4 and its upstream intron of calcitonin/CGRP pre-mRNA, inhibiting splicing of this upstream intron. Thus, it is possible that Fox-1 induces exon skipping by multiple mechanisms. Our splicing analyses showed that Fox-1 protein blocks the pre-spliceosomal E complex formation on intron 9 of hF1γ pre-mRNA ( and ). The E complex contains the U1 snRNP and the spliceosomal proteins SF1 and the U2AF heterodimer. In addition, U2 snRNP is loosely associated with the complex. Kent . () showed that the ATP-independent E′ complex is formed prior to E complex formation, with U1 snRNP and SF1 protein. Since our present experiments could not distinguish the E′ complex from the E complex, we think it possible that Fox-1 blocks the E′ complex formation. Recently, Ule . () showed Nova1 inhibits splicing of an RNA substrate containing Nova1-binding sites (YCAY clusters) by blocking U1 snRNP binding, resulting in the induction of exon skipping. Although Fox-1 does not inhibit the splicing of intron 8, which contains Fox-1-binding sites, it is possible that Fox-1 in some way acts across exon 9 of the hF1γ gene to prevent U1 snRNP assembly at the 5′ splice site in intron 9. It has been shown that components of U1 snRNP are direct targets of several splicing regulator. For example, TIA-1 protein interacts with U1C protein, one of the U1 snRNP components, and recruits U1 snRNP to the 5′ splice site (). The PSI protein represses splicing by interaction with the U1-70K protein (). Notably, it was reported that Fox-1 and Fox-2 interact with U1C protein in a yeast two-hybrid screening (). When Fox-1 protein binds to the (U)GCAUG repressor element upstream of the alternative exon, Fox-1 may repress the splicing of the downstream intron by interacting with U1C protein. Alternatively, Fox-1 may interfere with the interactions between U1 snRNP and U2AF. Izquierdo . () reported that PTB binding to exon inhibits the exon definition. More recently, Sharma . () indicates that PTB binds to the flanking introns of N1 exon, preventing the association of U2AF with U1 snRNP that binds to the 5′ splice site of the downstream intron. PTB protein prevents the assembly of U2AF into the E complex, probably without affecting the binding of U1 snRNP to the 5′ splice site. Although our immunoprecipitation experiments showed that Fox-1 does not interact with U2AF heterodimer (our unpublished data), it is possible that Fox-1 interacts directly or indirectly with U1 snRNP components to prevent the association of U2AF with U1 snRNP. Thus, it will be interesting to study whether Fox-1 blocks association of U1 snRNP and U2AF to intron 9 or the interaction between U1 snRNP and U2AF. In this study, we identified that the carboxyl-terminal region of mouse Fox-1 protein is required for inducing exon skipping. In particular, the 326–338 aa C-terminal region of the protein is essential for induction of exon skipping. Our previous study showed that truncation of the C-terminal 122 residues of zebrafish Fox-1 protein disrupts induction of exon skipping (). Furthermore, Baraniak . () showed that the C-terminal 84 amino acids of the Fox-2 protein are required for the proper regulation of FGFR2 exon choice, while the N-terminal region of its protein is dispensable. These results suggest that Fox proteins interact with some key proteins through the C terminal region, functioning in both positive and negative regulations. Several groups have reported on proteins that interact with Fox protein. Human A2BP/Fox was identified originally as an interacting protein of ataxin-2 protein in yeast two-hybrid screening. The C-terminal region of human A2BP1 is required for strong interaction with ataxin-2 (). The Fyn tyrosine kinase and estrogen receptor-α interact with Fox-1 and Fox-2 (,). It remains to be elucidated whether these proteins are involved in the splicing regulation. Moreover, further identification of interaction partners, including general splicing factors, will be informative to clarify the mechanisms of tissue-specific splicing regulation by Fox proteins.

Riboswitches are RNA molecules located in untranslated regions of several mRNAs which regulate the expression of bacterial genes involved in the biosynthesis, transport or metabolism of small molecules—all of this without the aid of protein cofactors (). These RNA molecules exhibit highly complex structures able to specifically bind cellular metabolites, and following a ligand-induced structural reorganization, to appropriately modulate expression of the associated gene. Riboswitches are made of two distinct domains: an aptamer domain which specifically binds a cognate ligand, and an expression platform which controls gene expression either at the transcriptional or translational level (). Transcriptional and translational controls are respectively performed by modulation of an intrinsic terminator domain or by selective sequestration of the Shine–Dalgarno sequence required for ribosome binding (,). Moreover, it has recently been shown that riboswitches also control mRNA splicing in (). Various riboswitches have been shown to specifically recognize a large variety of ligands, such as adenine (), adenosylcobalamin (), flavin mononucleotide (,), guanine (), glucosamine-6-phosphate (), glycine (), lysine (,), intracellular magnesium (), S-adenosyl methionine (SAM) () and thiamine pyrophosphate (TPP) (,). In addition, a new riboswitch tandem configuration has been shown to detect two different metabolites (SAM and TPP) suggesting that individual riboswitch elements can be assembled to make more complex regulation systems (). The adenine and guanine riboswitches are part of purine-sensing riboswitches, which are highly conserved elements presenting several similarities in their sequence and secondary structure (,,,). Indeed, both purine aptamers are organized around a three-way junction connecting three helices (P1, P2 and P3) where the P1 stem is the only helical region exhibiting some degree of conservation (,). A loop–loop interaction essential for ligand binding is formed between stem-loops P2 and P3 (,). Several purine riboswitch aptamer structures have been determined by X-ray crystallography and exhibit a relatively compact fold characterized by a coaxial stack formed by P1 and P3 helices (,,). In each case, the metabolite is bound in a cavity where it is completely surrounded by RNA contacts, supporting previous studies where a structural reorganization was found in the core region upon ligand binding (,,). The specificity of the ligand interaction results from the formation of a Watson–Crick base pair with nucleotide 74 (). Although adenine and guanine aptamers are structurally very similar, they regulate gene expression differently. Indeed, while the adenine riboswitch positively regulates gene expression upon ligand binding, the guanine riboswitch promotes premature arrest of transcription of the downstream gene in presence of guanine (A). This marked difference is explained by the organization of the riboswitch architecture where the P1 stem is either an antiterminator or anti-antiterminator, depending if it is located in the adenine or guanine riboswitch, respectively. A previously reported consensus sequence of the guanine-sensing aptamer shows that most conserved nucleotides occur within the core region, which is reorganized upon ligand binding (). Indeed, in-line probing assays in absence and presence of the corresponding ligand indicated that the structure of the aptamer undergoes dramatic changes in the core region upon addition of guanine. Considering that the original sequence alignment was performed using purine-sensing aptamer sequences and that we recently observed that ligand binding requirements of adenine aptamers are different from those of purine aptamers (), we have established a new consensus sequence using only guanine riboswitch sequences. We find that positions 24 and 73 display specific sequence requirements and that their identity is related to the riboswitch binding affinity, which is also dependent on the aptamer context. However, a strong nucleotide requirement is observed for position 48 that is totally intolerant to the presence of guanine, which is consistent with our previous observations for the adenine riboswitch (). In addition, RT-qPCR analysis performed on endogenous guanine riboswitches reveals that natural riboswitch variants exhibit significant differences in their propensity to regulate gene expression. A very good correlation is obtained between the observed aptamer–ligand affinity and the regulation efficiency of the corresponding riboswitch. Oligonucleotides were purchased from Sigma Genosys (Canada). Oligonucleotides were purified by denaturing polyacrylamide gel electrophoresis, electroeluted in 8 M ammonium acetate, recovered by ethanol precipitation and dissolved in water. For the production of the guanine riboswitch, DNA templates were prepared by recursive PCR and transcribed using T7 RNA polymerase () in 40 mM Tris–HCl buffer, pH 8.0, containing 0.01% Triton X-100, 20 mM MgCl, 10 mM DTT and 2 mM spermidine. The aptamer was produced from a partial duplex containing the complementary sequence of the aptamer and the T7 promoter sequence. RNA was purified by denaturing PAGE and recovered as described for DNA oligonucleotides. The aptamer sequences used in this study are based on the alignment presented in B where only the region covered by the P1 stem (yellow region) is used and to which a GCG sequence is added to the 5′ side to allow good transcription yield and to minimize the 5′-heterogeneity (). strain 168 cells were grown in minimal media in absence or presence of ligand as described previously (). RNA was extracted from lysate using a Qiagen RNeasy kit, treated with DNase I in presence of RNase inhibitors, and 1 µg was used for reverse transcription with 200 units of MMLV-RT (Promega), using 0.5 µg of the primer required to either reverse transcribe full length or terminated mRNA species. The reaction was performed at 42°C for 1 h. The qPCR amplification step was essentially performed as in a previous study (). Briefly, cDNA was added to the SYBR Green Master Mix (Stratagene) according to the manufacturer's protocol and 0.15 µM of each primer was used. Amplicons were analyzed using the MxPro QPCR software. Where shown, experiments were repeated at least three times to estimate the average and SD of the full length/terminated mRNA ratio. Transcript amounts were normalized to the terminated transcript species. For each riboswitch, three primers were designed either corresponding to the full length or to the terminated mRNA species. Please note that the same 5′ primer is used in both cases. Primer sequences are as follows (5′ to 3′): variant: variant: variant: [5′-P] RNA (final concentration <1 nM) was incubated in presence of 50 mM Tris–HCl (pH 8.0) and 100 mM KCl at 37°C, for 2 min. RNase T1 (1 U) was then added and allowed to react for 2 min. Reactions were stopped by addition of an equal volume of a solution of 97% formamide and 10 mM EDTA. Products were resolved on a 10% denaturing PAGE and dried gels were exposed to Phosphor Imager screens. [5′-P] RNA molecules were incubated for 4 days at 25°C in 50 mM Tris–HCl buffer, pH 8.0, 20 mM MgCl and 100 mM KCl in presence or absence of ligand. The reactions were stopped with a 97% formamide solution containing 10 mM EDTA formamide and samples were separated by electrophoresis in 10% denaturing PAGE and dried gels were exposed to Phosphor Imager screens. Guanine dissociation constants using in-line probing assays were performed according to Mandal (). Briefly, a region showing ligand-dependent protection was quantified (positions 49–53) and normalized against the position 36, which shows no in-line probing change, as previously established (). Guanine concentrations ranging from 1 nM to 10 µM were usually employed but the range was adjusted depending of the aptamer studied. Prior to 2-aminopurine (2AP) fluorescence measurements, RNA was heated for 1 min to 95°C in water and then preincubated at 23°C for 10 min in reaction buffer (see subsequently) to ensure homogeneous folding of RNA species (). Fluorescence spectroscopy was performed on a Quanta Master fluorometer. All data were collected at 25°C in 10 mM MgCl, 50 mM Tris–HCl (pH 8.0) and 100 mM KCl. Spectra were corrected for background, and intensities were determined by integrating data collected over the range 365–475 nm. 2AP was excited at 300 nm to obtain a good separation between the Raman and fluorescence peaks. The fraction of quenched 2AP fluorescence was calculated by monitoring fluorescence emission of free 2AP in solution (50 nM) and titrating with an increasing concentration of RNA. If the total aptamer concentration is in large excess relative to 2AP, then it can be assumed that the concentration of free aptamer is similar to the concentration of total molecules. is a dimensionless constant proportional to the ratio of the quantum yields of the 2AP in the complex and free in solution; it is also less than unity as the quantum yield of free 2AP is higher than in the complex. The parameter is determined, together with , by non-linear least-squares fitting following the Levenberg–Marquardt algorithm and typically corresponds to a value of 0.05. The equation assumes a simple 1:1 stoichiometry between RNA and 2AP, as reported in crystal structures (,). Competition experiments were done using a fixed concentration of RNA (1 µM) and 2AP (50 nM). Where indicated, a competing ligand was also incubated in the mixture at a concentration of 1 µM. For each experiments, at least three measurements were performed to obtain an average value. Data analysis was performed as previously described (,). The consensus sequence of the guanine aptamer was previously generated from the analysis of 32 representatives that were hypothesized to be guanine-sensing riboswitches (). However, subsequent experimental validation demonstrated that three representatives are instead specific to adenine (), thereby generating a suboptimal guanine aptamer consensus sequence. Recent work in our laboratory identified novel ligand binding requirements of the adenine riboswitch, some of which are related to the ligand binding specificity (). Thus, we decided to analyze sequence requirements of the G box motif using only guanine-specific sequences (i.e. containing C74), with a significantly larger set of representatives from the database (). The complete sequence alignment performed with 89 representatives is shown on Figure S1 (Supplementary Data). We used only a subset of these sequences for the alignment shown in B. These sequences were selected because they are representatives of the natural sequence variation observed among all sequences. A consensus secondary structure shows that most nucleotides found in the core region are highly conserved as previously found (C) (). In our alignment, positions 24, 48 and 73 are not as highly conserved as other positions found in the core. Indeed, position 24 conforms to the ‘W’ consensus (adenine or uracil), position 48 conforms to the ‘H’ consensus (every nucleotide but a guanine) and position 73 shows a degree of conservation of only 90% (compared to 95% for other core positions). To validate the consensus structure of the G box motif and to obtain a deeper understanding about the biological effects of core sequence variations found in natural variants, we analyzed the ligand binding properties of selected guanine aptamers. Adenine and guanine aptamer crystal structures display a remarkable similarity about their ligand binding site but they differ in the Watson–Crick base pair that they form with the bound ligand to achieve a high degree of specificity (A) (,,,). Furthermore, the rest of the bound nucleobase makes nearly identical interactions with the aptamer suggesting that structurally similar purine-containing molecules would also make productive RNA-ligand contacts. Indeed, kinetics of ligand binding have been studied in details with the fluorescent purine analog 2AP (,). Upon formation of the 2AP:aptamer complex, 2AP fluorescence becomes significantly quenched as expected from the ligand binding pocket of purine aptamers in which the bound ligand is completely surrounded by RNA (,). Moreover, we have previously shown that the G box domain exhibits ∼10-fold lower affinity toward 2AP when compared to the A box aptamer (). However, upon introduction of a C74U mutation in the guanine aptamer, in order to convert it into an adenine-sensing aptamer, we observed that 2AP binding affinity is nearly identical to the adenine aptamer, indicating that 2AP can be used to detect affinity changes in core sequence variations. Upon visual inspection of the sequence alignment in B, it is immediately apparent that guanine-specific aptamers display nucleotide variations in the core domain. Given that most of these nucleotides are important for making productive interactions for ligand binding, we speculated that aptamers exhibiting core nucleotide variations might display differences in guanine binding, and that this could be reflected in their 2AP affinity. Clearly, although the 2AP binding assay does not give information about guanine binding, it nevertheless allows to directly compare the ligand binding affinities of guanine aptamers. 2AP will be a good alternative to guanine as a ligand only if it is recognized similarly to guanine. To validate our strategy, we employed in-line probing to investigate how similarly the aptamer domain of the guanine riboswitch recognizes 2AP. This method exploits the inherent chemical instability of RNA under physiological conditions due to the spontaneous cleavage of phosphodiester linkages (), which is more pronounced in unstructured regions due to the fact that phosphodiester bonds are free to adopt an in-line conformation that is prevented in the context of a helix. This assay has been extensively used by others and us to monitor conformational changes induced by external ligands on numerous RNA riboswitch domains (,,,,,,). We therefore assessed if guanine and 2AP produced similar cleavage patterns in a guanine-sensing aptamer (B), which should confirm a similar structure between these complexes. In absence of guanine, several cleavage products were observed that all map to formally single-stranded regions. However, a reduction of the extent of cleavage was observed in the core region in presence of guanine (positions 49–53), indicating that the structure is reorganized upon ligand binding, as observed previously for the variant (). When in-line probing experiments were repeated with 2AP instead of guanine as a ligand, a cleavage pattern identical to that observed with guanine was obtained, showing that both purines are recognized in an apparently identical manner by the guanine riboswitch aptamer. These results show that the guanine-sensing aptamer undergoes indistinguishable structural changes in presence of either guanine or 2AP, establishing that 2AP can be used as a ligand binding reporter to study guanine-sensing riboswitches. We performed a 2AP binding assay in which the 2AP fluorescence was monitored as a function of the guanine riboswitch RNA concentration (C). We observed that the guanine riboswitch is able to perform productive 2AP binding. Fluorescence data could be fitted to a simple two-state binding model (see Materials and Methods section), yielding an apparent dissociation constant () value of 6.69 ± 0.91 µM (). This value is in very good agreement with a previous in-line probing study where it was found that the 2AP binding affinity of the complex is ∼10 µM (). To ensure that the decrease in fluorescence does not result from non-specific binding, we engineered a P2 riboswitch mutant in which the formation of the essential loop–loop interaction is not possible. We have previously shown for the adenine riboswitch that this loop–loop interaction is essential for the formation of the aptamer–ligand complex (). As expected, no significant fluorescence quenching was observed for this variant showing that 2AP binds in a specific manner to the aptamer and confirming that the G box riboswitch requires a functional loop–loop interaction to perform ligand binding (C), as previously observed in the context of the isolated aptamer (). To estimate their relative binding affinities, we expanded our 2AP binding analysis to other guanine-sensing riboswitch aptamers (). We first targeted our analysis to the guanine aptamers and from . We observed that apart from , these guanine-sensing aptamers all exhibit a relatively similar 2AP affinity with a ∼8–10 µM (). However, the variant displays a higher 2AP affinity with an increase of 13-fold relative to . The histogram representation (D) with the apparent affinity constant used as a comparison basis shows that although guanine aptamers are very similar in their sequence and secondary structure, they do not all exhibit a similar ligand binding affinity. Additional guanine riboswitches were also monitored for their 2AP binding affinity. Three broad categories were found (D and ). One category has a very poor 2AP affinity (BH- and FN-) while another one (BH-, CA-, CP-, LL-, LM- and OI-) has an affinity similar to the aptamer. The third category (STPY-) exhibited very high 2AP binding affinity, similar to BS-. Taken together, our results indicate that guanine aptamers naturally exhibit variations in their ligand binding affinity that can vary over a range of at least 21-fold (compare BH- and STPY- aptamers). This only represents a lower limit given that the very poor 2AP affinity of BH- and FN- representatives suggests an even larger ligand binding affinity spectrum. According to our 2AP binding analysis, BH- and FN- exhibit undetectable 2AP binding ( > 25 µM). To learn whether these variants are able to perform ligand binding, an in-line probing strategy was employed in which various ligands were tested for their ability to induce a structural change in the core domain of the aptamer. We first investigated the ligand-induced structural changes of the BH- guanine aptamer variant (A). In absence of ligand, cleavage products were observed in the single-stranded region located between P2 and P3 stems (A and B). However, upon incubation with guanine, the extent of cleavage was markedly decreased suggesting that a constrained structure is adopted upon ligand binding. These results are very similar to those previously obtained for the guanine aptamer (). When the experiments were performed in presence of adenine, xanthine or hypoxanthine, no significant reduction of cleavage was observed suggesting a low binding affinity. A previous study with the BS- guanine aptamer has shown that xanthine and hypoxanthine induce RNA structural changes, but to a reduced level compared to guanine (∼10-fold) (). Thus, our in-line probing data indicate that BH- can perform ligand binding but exhibits lower affinity toward xanthine and hypoxanthine compared to BS-, which is in very good agreement with our 2AP binding studies with BS- and BH- (D). Very similar results were obtained for the FN- aptamer (data not shown). Among all aptamers examined, BS- and STPY- are those having the best 2AP binding affinity (). To verify that this high 2AP affinity does not result from a switch in specificity where 2AP is preferred over guanine, competition experiments were performed in which the aptamer was incubated in presence of 2AP and an additional ligand. Under conditions where a high proportion of aptamer-2AP complex is formed (see Materials and Methods section), we incubated the BS- aptamer in presence of 2AP and guanine. 2AP fluorescence monitoring showed that guanine very efficiently competes with 2AP but adenine, xanthine and hypoxanthine do not, indicating that they exhibit a poorer ability to displace 2AP (C). Thus, our results show that the BS- aptamer has a higher affinity toward guanine than adenine, xanthine and hypoxanthine, which is consistent with previous findings for other guanine-specific riboswitches (,). The STPY- variant, also analyzed using this assay, gave very similar results (C), indicating that both the BS- and the STPY- aptamers are guanine-specific, and that they both exhibit ∼15-fold higher ligand affinity compared to the BS- variant. Although 2AP and guanine are recognized similarly by the aptamer domain of the guanine riboswitch (B), 2AP affinity variations observed in this work may not be representatives of affinity variations exhibited toward the natural ligand, guanine. To explore this possibility, a subset of aptamers showing substantial heterogeneity in 2AP affinity was selected and characterized using the in-line probing assay (). This assay may be used to determine ligand dissociation constants by performing the experiments in presence of various concentrations of ligand and by monitoring the variations in spontaneous scission products. Using this procedure, we established an apparent dissociation constant for guanine () of 4.7 ± 3.6 nM for the BS- guanine aptamer (). This is in excellent agreement with the value ∼5 nM obtained by Breaker and coworkers (). Next, we characterized the aptamers BH-, CP- and BS- which show various affinities for 2AP (). As observed using 2AP fluorescence assays, these three aptamers were found to exhibit variations in their affinity toward guanine where values of 414 ± 387 nM, 3.7 ± 0.8 nM and 0.5 ± 0.2 nM where obtained for BH-, CP- and BS-, respectively (). Thus, while BH- exhibits very poor guanine binding, BS- displays a high affinity toward guanine when compared to the subset of studied aptamers, which is in agreement with our 2AP binding analysis (). The CP- and BS- aptamers show very similar affinities for guanine, in agreement with our 2AP binding results (). Taken together, the in-line probing data indicate that guanine riboswitch aptamers can exhibit large variations up to ∼800-fold in their ligand binding affinity. According to our consensus sequence, nucleotides 24, 48 and 73 are the less conserved positions in the aptamer core domain (C). To further investigate nucleotide requirements at these positions for ligand binding, a site-directed mutagenesis approach was employed to analyze those three positions. As a reference, we used the BS- aptamer since it is the best characterized aptamer (,,,,). Position 24 of the guanine aptamer exhibits a ‘W’ consensus where an adenine or a uracil is found (C). To investigate this position, we systematically replaced it with the three other nucleotide possibilities and analyzed the 2AP binding affinity of corresponding variants. None of the engineered aptamers shown detectable 2AP binding activity suggesting that A24 is very important in BS- (A). However, two natural guanine aptamers have variations at position 24 where A24 is either deleted (CA-) or substituted for a uracil (BS-), and can nevertheless form a productive complex with 2AP (A). Thus, this suggests that position 24 is very important for ligand binding in BS- but that the sequence context also plays a critical role in its sequence requirement. In our alignment, the position 48 of the aptamer sequence adopts the ‘H’ consensus where any nucleotide is found but a guanine (C), in contrast to a previous study which reported the non-conservation of that position (). This was later rationalized with purine aptamer crystal structures in which nucleotide 48 is systematically exposed to the solvent (,,,). In addition, recent studies in which a 2AP introduced at position 48 exhibits enhanced fluorescence upon ligand binding, suggested that position 48 becomes exposed to the solvent in the ligand bound state (,). To study the nucleotide requirement at position 48, a guanine was introduced at position 48 in three natural guanine aptamers showing nucleotide variations at this position. Using the 2AP assay, we found that all three guanine-bearing variants displayed a marked reduction in 2AP affinity (B). These results indicate that the presence of a guanine at position 48 significantly perturbs ligand binding in three different sequence contexts. This result is in contrast with our observations at position 24, given that the negative influence of G48 does not depend on the sequence context. A 2AP binding study was also performed to analyze the nucleotide requirement at position 73 (C). Upon introduction of a guanine or a uracil, we observed that the BS- aptamer was no longer able to perform 2AP binding, while the introduction of a cytosine resulted in a higher ligand binding affinity when compared to the BS- sequence. These results are reminiscent to those obtained for position 24 where naturally occurring variants can readily accommodate impeding substitutions (C), indicating that the sequence context is very important for the identity of nucleotide 73. Of all substitutions performed in this work, the introduction of a guanine at position 48 is the only one for which a systematic negative influence is observed (B). To examine the effect of G48 on the local structure of the aptamer, an in-line probing strategy was employed. The wild-type BS- aptamer variant subjected to in-line probing assays showed several cleavage products in the core domain where positions 48, 50 and 51 were strongly cleaved (D). However, upon addition of guanine, a protection of this region was observed suggesting a guanine-dependent reorganization of the core, which is in very good agreement with previous studies (,). However, introduction of G48 in the BS- aptamer (U48G variant) altered the local structure of the core domain where enhanced cleavages were observed for positions 47 and 49, while a reduced scission was observed for position 48 (D). The latter is consistent with our previous results obtained for an U74C adenine aptamer variant (), suggesting that both adenine and guanine aptamers are restricted by similar structural constraints. However, in contrast to what we have observed for the U74C adenine variant, the BS- U48G variant is able to undertake a structural reorganization upon guanine binding albeit to a reduced level (D). According to our ligand binding assays, the affinity spectrum considerably varies among natural aptamers (). To verify if such variations are present among corresponding endogenous guanine riboswitches, an RT-qPCR approach was used to directly assess the relevance of our results in the context of guanine riboswitches. The assay was designed to detect prematurely terminated (OFF state) and full length (ON state) riboswitch mRNA species (A). According to this riboswitch regulation mechanism, it is expected that the full length/terminated ratio should be inversely proportional to the concentration of intracellular guanine. The riboswitch gene expression control was first studied under growth conditions using different guanine concentrations (B). When compared to a control experiment in which no guanine was added into minimal media, no significant full length/terminated ratio change was observed with cells grown in presence of either 0.05 mg/ml or 0.25 mg/ml guanine. However, when a concentration of 0.50 mg/ml or higher was used, the full length/terminated ratio decreased by ∼40%, indicating that the riboswitch induced premature transcription termination, consistent with the presence of an intrinsic terminator motif in the expression platform (B). To our knowledge, this is the first direct demonstration that guanine riboswitches modulate gene expression by controlling premature transcription termination, which is in very good agreement with previously reported β-galactosidase gene expression studies (), and consequently provides further strong evidence in favor of the originally proposed riboswitch regulation mechanism (,). The endogenous riboswitch was also monitored under various guanine concentrations using RT-qPCR (B). In presence of 0.25 mg/ml guanine, a small but significant termination efficiency (∼20%) was detected, which was further increased to ∼80% by incubating at a saturating guanine concentration, indicating that the variant is more efficient than to promote transcription termination . Strikingly, an even higher guanine-dependent transcription termination efficiency was observed when monitoring the riboswitch variant, which yielded ∼55 and ∼98% termination efficiencies at 0.25 mg/ml and saturating guanine concentrations, respectively. These results clearly indicate that guanine riboswitches do not all perform gene regulation with similar efficiencies, which most probably reflects individual regulation requirements of each regulon. Interestingly, the variant exhibits the highest ligand binding affinity () as well as the most efficient riboswitch activity studied in this work (B). Although these results strongly suggest that ligand binding is very important for the riboswitch-mediated gene expression control, alternative factors are likely to play important role(s) in the regulation given that both and exhibit similar ligand binding affinities (), yet they clearly show differences for modulating gene expression levels (B). Here, we study the ligand binding requirements of guanine-sensing riboswitches to establish what are the molecular determinants involved in the formation of the ligand–aptamer complex, and to understand how they are used in the ligand binding process. In contrast to the original study in which the G box motif was described (), the refined G box consensus proposed here is strictly based on 89 guanine-specific aptamers from which several structural characteristics are revealed (Figure S1). For example, compared to other paired regions, the P2 stem shows a slightly higher mismatch rate, suggesting that mutations in this region are more easily accommodated in the tertiary fold. The P1 stem varies from two to nine base pairs throughout the entire sequence alignment suggesting that corresponding expression platforms are very likely to differ concomitantly to properly regulate premature transcription attenuation. Sequences involved in the loop–loop interaction are also conserved for positions known to be involved in platform tetrads (,). The core region of the consensus secondary structure is very conserved, with the exception of positions 24, 48 and 73. From the entire sequence alignment (Figure S1, Supplementary Data), no single base identity co-variation can be established between 24, 48 and 73, nor with other nucleotides, which is most probably because none of these three positions is involved in critical interactions, as observed in crystal structures (,). A24 is located between G72 and A73 and has been proposed to act as a spacer (), position 48 is completely exposed to the solvent, and A73 forms a water-mediated triple with the U22-A52 Watson–Crick base pair. Since nucleotides 24, 48 and 73 are situated relatively in close proximity to the bound ligand, their spatial organization is expected to be dependent upon ligand binding. Indeed, recent fluorescence studies have shown that a 2AP nucleobase introduced at either position 24 or 48 exhibits fluorescence emission changes upon ligand binding (,), in agreement with in-line probing results showing that most of the core region is reorganized upon ligand binding (). Thus, although nucleotides 24, 48 and 73 do not perform direct interactions with conserved nucleotides in the folded state, they are very likely important for the ligand-induced structural reorganization of guanine riboswitches. 2AP is traditionally used as a substitute for adenine, but because 2AP is a purine nucleobase that produces an in-line probing pattern indistinguishable to that obtained in presence of guanine (B), it can also be used as a ligand to study guanine riboswitches. 2AP fluorescence quenching experiments show that the complete riboswitch sequence can efficiently perform ligand binding (C), which is in contrast to what we observed in the context of the adenine riboswitch (), suggesting that both riboswitch regulation mechanisms could operate under different control regimes to achieve gene expression regulation (,). By examining a large array of naturally occurring guanine aptamers, our 2AP data show that an intrinsic large spectrum of ligand affinities is present among them (D), which is also observed when using guanine-induced in-line probing assays (). Interestingly, as previously determined by Breaker and co-workers in the context of BS- (), a systematic variation of ∼10-fold is observed between 2AP and guanine binding affinities () indicating that although 2AP shows weaker affinity compared to guanine, it can nevertheless reliably provide relative ligand binding affinities of guanine aptamers. The sequence alignment in B is arranged to represent 2AP binding affinity variations observed among naturally occurring guanine aptamers. Upon examination of the alignment, the two aptamers showing highest binding affinities (STPY- and BS-) are characterized by the presence of a uracil at position 24 together with a stable P2 helical domain beginning with a G-C or G-U base pair. In addition, a proportional correlation is observed between the stability of the P2 stem and the ligand binding affinity. However, aptamers exhibiting weakest binding activities (position- and FN-) possess an extra nucleotide immediately upstream of the position 73 which could inhibit the insertion of the base 24 between positions 72 and 73. This variation is not present in any other naturally occurring variant (Figure S1, Supplementary Data) suggesting that this insertion is not widespread among riboswitches. Binding activity also appears to be inversely proportional to the stability of the P3 helical domain, which is not always the case since low binding affinity aptamers (BH- and FN-) do not exhibit a stable P3 stem. Taken together, the large affinity binding spectrum suggests that guanine riboswitches do not all respond to similar intracellular guanine concentrations or that various cellular conditions (e.g. ionic strength) may be important for their gene regulation activity. For instance, the operon is found in three different organisms ( and ) and exhibits a variation of ∼14-fold for the formation of the aptamer-2AP complex (). However, this is not always the case since significant differences are found even within the same organism (i.e. ). Therefore, these variations are most likely important to allow the differential control for various bacterial regulons. Site-directed mutagenesis analysis shows that positions 24 and 73 display sequence requirements that are highly dependent on the aptamer variant (A and C). For instance, introduction of either A24U or A73G is highly detrimental for the ligand binding activity of the BS- aptamer. However, given that BS- and LM- contain U24 and G73, respectively, it strongly suggests that the nucleotide identity is highly dependent on the sequence context. Moreover, when comparing BS- and LM- to BS-, it can be observed that they are divergent mostly in their P2 and P3 stem regions where base insertions and deletions occur (B). Since it has been shown that loop–loop formation and ligand binding are related events (,,), it is likely that the structural arrangement of stem-loops P2 and P3 is important for core folding, in which nucleotides 24 and 73 are located. However, this trend is not generalized to the entire sequence alignment (Figure S1, Supplementary Data), suggesting that other structural compensation(s) may be present in other aptamer variants. In addition, the originally proposed non-conserved position 48 displays a remarkable intolerance to guanine at this position given that its introduction in three different aptamers systematically perturbs ligand binding to a high extent (B). This suggests that, in contrast to the other two core positions examined in this study, the negative effect of G48 is dominant over the sequence context (i.e. context-independent). Interestingly, the complete sequence alignment (Figure S1, Supplementary Data) shows that there is no natural aptamer sequence harboring G48. These results are in very good agreement with our previous study where we have shown that, in an U74C purine aptamer mutant, the presence of G48 is highly deleterious for ligand binding (). This was attributed to a putative tertiary interaction occurring between positions 48 and 74 that is taking place in absence of ligand which reduces the propensity of the aptamer to perform efficient ligand binding. However, in contrast to what was observed for the U74C aptamer, the BS- G48-containing guanine aptamer is still competent to perform guanine binding albeit to a reduced level (D). Of all guanine aptamers, the variant exhibits the highest ligand binding affinity (), and interestingly, is the one showing the most efficient riboswitch activity (B). Compared to and variants, the riboswitch exhibits higher activity in guanine concentrations ranging from sub-saturating to saturating levels. Thus, because is the most active riboswitch in saturating conditions, where the riboswitch aptamer is fully bound to the ligand and where premature transcription termination is expected to be primarily dependent on the terminator strength, it suggests that the terminator is more stable than those of and . As expected, when analyzing the predicted relative free energy of each terminator (Δ), it can be observed that exhibits the most stable structure (C), and accordingly, the variant shows a more stable terminator domain compared to , which is in agreement with their respective premature termination efficiencies obtained at saturating guanine concentrations (B). In addition, when analyzing RT-qPCR data obtained at 0.25 mg/ml where saturation is not yet attained and where ligand binding affinity should be important for gene regulation, it can be observed that only and show premature termination, still exhibiting the highest efficiency, which is in agreement with its high ligand binding affinity (). Although the ligand binding affinity is likely to be important for the propensity of the riboswitch to perform gene regulation, subtle differences are present that cannot readily be accounted for. For instance, even though the termination efficiency of is higher than at a sub-saturating guanine concentration (0.25 mg/ml), both aptamers exhibit similar ligand binding affinity (). However, because the ON state antiterminator structure is predicted to be more stable (Δ) than that of , it is possible that the antiterminator might inhibit to a higher degree the formation of the aptamer domain when compared to the variant. This would consequently decrease the propensity of the riboswitch to perform ligand binding, consistent with our RT-qPCR data (B). Interestingly, of all guanine riboswitch-regulated transcripts, is the only one not controlled by the PurR repressor, which regulates transcription initiation (). Thus, it is conceivable that because it does not rely on a transcription initiation regulation mechanism, might exhibit higher activities for both ligand binding as well as premature transcription termination to ensure proper gene expression regulation. Whether this fully explains the very efficient riboswitch activity of will remain to be established but, nevertheless, our study shows that guanine riboswitches exhibit variations in their ligand binding affinities between bacterial hosts and genetic units. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

Affymetrix arrays are widely used for comparing the expression of tens of thousands of genes under different experimental or clinical conditions. The number of probes on these arrays continues to increase: for example, the most recent releases of human chip array HGU133plus holds 54 000 probe sets, representing almost 40 000 genes. Nevertheless, not all genes are expected to be expressed at biologically meaningful or at detectable levels (1–3 RNA copies per cell), as most tissues express only 30–40% of the genes () or, according to a recent estimation, around 10 000–15 000 genes (). Furthermore, among the expressed genes, generally only a very small fraction is expected to be differentially expressed (DE) under different experimental conditions. This situation leads to several problems, including measurement bias, increased potential for false discoveries and reduced sensitivity in detecting DE genes. Measurement bias occurs because arrays with more probes tend to have more spurious hybridizations, particularly through non-specific binding of abundant RNAs from highly expressed genes to the probes associated with under- or un-expressed genes. For these genes, random fluctuation generates spuriously large test statistics, which will then increase the number of false discoveries. Additional problems in real data include an unbalanced proportion of over- and under-expressed genes, especially in laboratory experimental conditions. This may introduce a severe bias in measurements due to the normalization step, which typically assumes that there is a balanced number of over- and under-expressed genes. This bias carries over to the statistical analysis, leading to bias in the estimation of the false discovery rate (FDR), especially among the non-DE genes (). Presently there is no general guidance on whether or not one should filter microarray data, hence many analyses simply include all the genes. Even without the problem of bias in the normalization step, it is intuitively clear that including many non-DE genes in the collection of genes to be tested will reduce the sensitivity in finding DE genes. In technical terms, we say that the non-DE genes contribute to the FDR of the procedure, so filtering out likely non-DE genes prior to statistical comparison will help increase the sensitivity of the procedure. The key idea in gene filtering is to use features of the data that do not directly use the information about the experimental conditions. Many papers have reported filtering based on various approaches, such as average intensity signal (), within-gene signal variance (), percent present-calls (), estimated fold-change or combinations of various methods (). Nevertheless, at present little attention has been devoted to deeper analysis of the raw data and the impact of pre-filtering of genes on the test procedures' performance. In this article, we propose a new algorithm to flexibly filter likely uninformative sets of hybridizations (FLUSH). The method is based on a robust linear model of the probe-level data that captures array and probe set effects. For our purposes, the model yields estimates of array-to-array and residual variation. Probe sets with low array-to-array variation are not likely to carry important biological signal, so they are not likely to be DE and should be filtered out. Furthermore, probe sets with an elevated residual variance typically tend to have inconsistent patterns in the probe-effect across replicate samples of the experiment. These probe sets are mostly associated with un-expressed genes, and again should be filtered out. The FLUSH procedure has been tested on a freely available spike-in experiment as well as on real experimental data on retinal degeneration. We compare the performance of filtered analyses with analyses using unfiltered, presence-filtered, intensity-filtered and variance-filtered data. Eliminating potentially uninformative features reduces bias and increases sensitivity in finding DE genes. Both spike-in data and experimental data were pre-processed, prior to statistical testing, with two of the most widely used procedures for background correction, normalization and expression measure computation, i.e. MAS5 () and RMA (). Expression values were analyzed on a logarithmic scale. For comparison, filtering based on Affymetrix presence-calls was also used, where features with less than 50% presence-calls were excluded (). Data were modeled at the probe level. Each probe set may contain from 8 to 20 pairs of perfect match (PM) and mismatch (MM) probes. The model was fitted on the PM data (on the log scale) after background correction using the so-called ideal mismatch (IMM) () to ensure positive values. The model was fitted through a robust linear model fit through M-Estimation, already implemented by the R package affyPLM (). Usually, the normalization step attempts to remove the technical artifact , such that the remaining signal is the biological effect plus noise. Instead, we will keep the combined technical and biological effects, with the key idea that if the total effect is not significant, then there cannot be any biological signal in the data, which means the gene cannot be DE. So, the uninformative probe sets are those with small array-to-array variations. where is the vector of estimated 's, and is its estimated covariance matrix. These quantities are available from the robust linear model fit. A non-parametric quantile regression smoothing, with a user-specified quantile to be estimated (τ), is fitted on the array effect χ (on the square root scale) as a function of the logarithm of residual standard deviation (SD). It is ‘non-parametric’ in the sense that it is not based on an explicit functional form, but is based on local smoothing of the data. Setting δ = λ leads to a unit weight for probes with low residual SD, and increasing as a function of (). Filtering can be tuned by varying τ, δ and λ. The estimated number of truly differentially expressed genes (TDE), at each FDR level, was computed as 1-FDR multiplied by the number of genes declared significant. Scatterplots of the -statistics versus the logarithm of the standard error with fdr isolines are hereafter called ‘TSE-plots’. Fdr isolines join points with the same fdr value, and are used to show fdr boundaries as a function of varying SD and -statistics (). Plots of the square root of the array effect χ as a function of the logarithm of the residual SD are called ‘RA-plots’. We first summarize () the work-flow in the microarray data analysis using the proposed FLUSH algorithm. Briefly: In a recent experiment Choe () produced a freely available controlled spike-in data set (the ‘Golden Spike’ data set). As a first step, the Golden Spike raw data, briefly described in the ‘Methods’ section, were processed with FLUSH, based on a quantile regression that filtered out 60% of the probe sets (in order to identify features to retain for the subsequent analysis). The whole data set was background-corrected, normalized and summarized using both MAS5 and RMA algorithms; note that this step is not affected by FLUSH. Genes filtered out by the FLUSH procedure were then removed from both the MAS5 and RMA expression matrices. Unlike in Choe , our normalization was based on all features, not just on truly non-DE genes (those with fold change FC=1). We did this because our purpose was to develop a procedure that is applicable to a real experimental setting, where it is impossible to ascertain which genes are present but not differentially expressed. To compare the spike-in versus the control groups, we first computed the standard -statistic and the associated SE, using both MAS5 and RMA expression measures. shows TSE-plots without filtering. Even though all transcripts were designed to be either over-expressed or at constant level, both RMA and MAS5 show a large number of apparently under-expressed features, mainly due to genes with FC close to 1. This problem arises as a consequence of unbalanced over- and under-expressed genes, which leads to biased normalization. Substantial spurious over-expression signal (yellow dots) is evident in both plots, especially in MAS5. This is consistent with previously published analyses that reported a signal content higher than expected (,) and might be due to both non-specific binding and normalization bias. shows the plot of the square-root of the array-effect test statistic as a function of the logarithm of the residual SD—or RA-plot—for the Golden Spike data, showing the array-to-array variability versus residual variance from the probe-level linear model (see ‘Methods’ section). Un-expressed genes showed high residual variance and relatively low array-effects; these correspond to probe sets with inconsistent patterns between replicates. Genes with FC = 1 had low array-effects and relatively low residual variance, but showed some mixing with over-expressed genes. The majority of genes with FC ⩾2 were clearly separated from the cloud of noisy genes. A non-parametric quantile regression smoothing line (see ‘Methods’ section) was fitted using the 60th percentile of array effect as a function of residual SD. As a result of the filtering procedure, a total of 8 400 out of 14 010 features with array effects below the estimated quantile regression line were excluded from further analysis. Given the small sample size, the local fdr estimation through permutation () is not completely trustworthy, but the estimated local fdr can still be used to rank genes. To assess the merits of filtering and to compare the different procedures, we plot in the cumulative number of genes declared DE at increasing values of estimated local fdr, versus the corresponding number of truly DE genes. For the presence-call, intensity and variance filters, we tried to keep almost the same number of genes as for FLUSH. Analyses of unfiltered data suffers from bias as well as large variability due to nonDE genes, resulting in a high number of false discoveries. Presence-call filtering was not able to overcome the biased normalization of non-DE genes, so in this case it performed no better than the unfiltered analysis. The worst performance was demonstrated by the average-intensity and variance filtering, which clearly removed too many truly DE genes. A more restrictive filtering on presence-call was also adopted, selecting features declared present in at least 50% of the samples (). This produced similar results (see S in the Supplementary Report). In contrast, FLUSH filtering reduced bias, by excluding non-DE genes that were falsely declared DE due to imperfect normalization, and clearly increased the sensitivity of the procedure based on both RMA and MAS5 expression values. For RMA analysis with unfiltered genes, the sensitivity was below 60% regardless of the number of genes declared DE; FLUSH procedure increased the sensitivity to over 80% when considering the top ranked 550 genes declared DE and to 90% for up to 465 genes. For MAS5 analysis with unfiltered genes, the sensitivity was mostly below 60%, while after filtering using FLUSH the sensitivity increased to around 80% for the 450 genes declared DE. Interestingly, unfiltered RMA outperforms unfiltered MAS5, which contrasts with Choe (). This might be explained by the different normalization approach, i.e. based on the whole set of genes rather than just the non-DE ones. We used wild-type C3H mice and inbred C3H mutant mice () to serve as an animal model for retinal degenerative diseases. Retina was hybridized to Affymetrix GeneChip Mouse Genome 430 2.0 arrays, which contain 45 101 probe sets for over 39 000 well-characterized genes. As for the Golden Spike experiment, the data were processed using both the MAS5 and RMA algorithms, and genes were filtered using FLUSH. shows the RA-plots of the data. Points were colored according to the quantiles of genes' average expression (on log scale), computed either with the MAS5 or RMA algorithms. Smooth lines (C and D) mark the filtering threshold derived from a quantile regression smoothing using τ = 0.4 and λ = δ = 0.45 [see Equation ()]. As we expected to have relatively few differentially expressed genes in this experiment, we tried to filter most un-informative probe sets. Features lying below the fitted quantile line were filtered out, so that out of 45 101 probe sets, 2 950 were kept. Genes with local fdr () lower than 15% for unfiltered features and 5% for filtered ones were printed with variable point size, depending on local fdr values, with larger points having smaller local fdr. For a sensitivity analysis of the choice of filtering parameters, S in the Supplementary Report shows the RA-plot of the mouse retina data with four different quantile regression lines, derived from different choices of the tuning parameters. For this range of filtering, the results are not sensitive to the choice of filtering parameters. Many more genes were assigned a low local fdr value (⩽0.15) by the local fdr procedure () applied to RMA values compared to MAS5: 73 probe sets showed a local fdr lower or equal to 0.15 for MAS5 expression values, and 1283 for RMA values (A and B). The local fdr estimates for the RMA values are likely biased, as we do not expect to see so many DE genes. It can be seen that many features identified as DE lie in the cloud of probe sets with low array-to-array variability or high residual-variation, and therefore are likely false discoveries. In A and B individual spots are color-coded according to probe set signal intensity; it is worth noting that lower intensity features tend to show lower array effects and higher residual variances compared to the high-intensity features. C and D show the same plots of array effect and log residual of SD after filtering using FLUSH. The FLUSH algorithm enhanced the local fdr estimation both for MAS5 and RMA values by putting more emphasis on genes with higher inter-array variability. Among the FLUSH-filtered DE features we recognized 39 probe sets corresponding to 27 genes known to be regulated during retinal degeneration (). A large majority of these genes are down-regulated, which agrees with the general model of retinal-dysfunction leading to degradation of the photoreceptor layer which can be observed in histological studies in mice (). This set of DE genes contains hallmark genes such as RHO (rhodopsin) or PDE6B (phosphodiesterase 6B). PDE6B was previously found to be mutated in mice (), and thus gives a non-functional gene-product thought responsible for the onset of retinal degeneration. RHO is the main protein involved in detecting light and is highly abundant in photoreceptor cells. Its abundance decreases towards zero during the process of retinal degeneration (). Another key enzyme in the regeneration of visual pigments, RDH12 (retinol dehydrogenase 12), was found significantly down-regulated by both probe sets for this gene. This gene codes for the main enzyme involved in converting 11--retinal to 11--retinol in photoreceptor cells. Also as expected, numerous genes involved in signal transduction and transcriptional regulation were found DE. Among these, CRX (cone-rod homeobox) is known to have a prominent role as photoreceptor-specific transcription factor. In agreement with our model, CRX was down-regulated (,). Several other genes known to play key roles in retinal function and known to be mainly expressed in retina were identified among a stringent selection of 109 probe sets (RMA array-effect χ > 29 with local fdr < 0.027): RSG9 (regulator of G-protein signalling 9) and RPGRIP1 (retinitis pigmentosa GTPase regulator interacting protein 1) both play an important role in regulation of G-proteins and maintaining of their proper function. Also the brain and retina-specific G-protein GNGT1 (guanine nucleotide-binding protein, gamma transducing activity polypeptide 1) was found down-regulated together with MAK (male-germ cell-associated kinase) and CDR2 (cerebellar degeneration-related protein 2). Prdx6 (peroxiredoxin 6), whose gene product is involved in immune-response, was found to be up-regulated, in accordance with the stimulation of stress–response and tissue repair mechanisms due to retinal degeneration. As shown in , all 39 known probe sets are located away from the cloud of noisy genes. Some genes known in the context of retinal degeneration had a very low local fdr, but were located closer to the main bulk of points. One example is Gfap (glial fibrillary acidic protein), a well-characterized marker that is almost exclusively expressed in astrocytes and used to follow the progress of retinal degeneration. Overall, FLUSH outperforms the standard FDR method without any filtering. For the RMA expression data, without any filtering, the median ranking of the standard FDR statistic of the 39 previously mentioned probe sets was 902, i.e. the list of 902 top-ranking genes contained only 19 of the 39 probe sets. Since we did not expect so many DE genes, it was clear that these known probe sets were buried among many non-DE genes. After filtering with FLUSH the median ranking was 281 (data not shown), while using the variance filter, the median ranking was 652; this means that variance filter is worse than FLUSH. Variance filtering applied to MAS5 expression values retained only 12 of the 39 probe sets, which means we are likely to lose a lot of DE genes, so we should not use this filter. To understand what happens, shows the location of the genes retained by variance filtering in the RA plot. Comparison with suggests strongly that a large proportion of these genes are likely unexpressed, and inclusion of such genes leads to loss of sensitivity. Additionally, we compared our results with an alternative filtering based on the GC-RMA algorithm (). Using this approach we typically observe a bimodal histogram of intensity values. Since the first of these peaks is (i) very sharp and (ii) part of the lowest signal intensity, it is tempting to associate this peak with non-expressed genes (data not shown). Removing all probe sets with average signal intensity in the first peak (fixed threshold of 4.8) gave 21 231 probe sets. [These correspond to 11 803 different genes, a slightly larger number than the estimate of 9 100-9 200 genes expressed in the retina ().] This article shows a novel data analytic procedure, called FLUSH, for filtering out potentially uninformative genes in Affymetrix microarrays and selecting features with potentially higher information content. FLUSH is meant as a filtering procedure performed in conjunction with any pre-processing step such as normalization and prior to any statistical or DE analysis. The main motivation is that a large proportion of genes on a microarray are un-expressed or non-DE, and these genes make it harder to detect DE genes, so they should be excluded prior to DE analysis. We have shown that FLUSH performs better than other more filtering methods based on presence-call or signal intensity. To highlight the novel contributions of FLUSH: Conceptually, the filtering method we used here can be adapted to other types of microarray data, such cDNA or bead arrays, as long as there are replications for each gene to allow separation of within- and between-array variance. (The within-array variance is the residual variance.) However, because of the assumed data structure, the specific implementation of the method and the R package reported in this article can be applied only to Affymetrix data. A recent theoretical computation () showed that there is an optimal number of hypotheses to be tested that is limited by the number of samples in the experiment. As seen clearly with the data examples, when the proportion of DE genes is small, they tend to get buried among the non-DE genes, thus increasing the FDR. Filtering out likely non-DE genes is a practical solution to this problem. Our analysis is based on a robust linear model, so it is not affected by outliers generated by some bad samples. Note that the analysis is performed gene by gene, so at any one analysis we expect only a few outliers. Nevertheless, we would recommend that the standard quality control checks for the arrays are followed. We emphasize that our purpose is not to show whether MAS5 or RMA work for Choe's; data, but whether we gain anything by using FLUSH. In principle, FLUSH can be used with any normalization method. With spike-in data such as Choe's;, it is possible to normalize using the ideal FC1-genes, but this is not feasible in real experiments, where finding FC1 genes requires pre-processed data including the normalization step, so there is a vicious cycle between normalization and finding FC1 genes. Normalization with MAS5 or RMA is usually applied to the full set of genes. The so-called ‘housekeeping-gene normalization’ for MAS5, using the set of FC1 genes, was shown to be biased in Ploner (). When the FC1-gene normalization was used for RMA as in Choe's; (), filtering using FLUSH still improves the performance of DE analysis. In most clinical data, the pattern of over/under-expressed genes tends to be balanced. But in lab experiments, e.g. with knock-out mice, an unbalanced proportion of over/under-expressed genes may reasonably happen. Haslett (), for example, reported a relevant bias towards over-expression in muscle-related genes (135 of the 185 declared DE). A similar unbalanced pattern was reported in other works (). Such unbalanced over- and under-expression violates the key assumption of balanced expression for the normalization step in data pre-processing. In this situation, both RMA and MAS5 expression measures will be biased due to imperfect normalization. FDR estimation and the sensitivity of the test will be affected by the bias in the pre-processing procedures. The problem is that, with real data sets, it is not obvious whether all the genes have been properly normalized. Even in clinical data with balanced expression levels, Ploner () showed that the commonly used quantile normalization is biased for low-intensity genes. Existing filtering methods based on Affymetrix presence-calls may be useful for removing noisy signal both for MAS5 and RMA values, but as shown in the Golden Spike data analysis, it cannot overcome all possible biases. The proposed FLUSH algorithm flexibly discards likely uninteresting features in terms of low information content (between-array variability), and lack of consistency among probe-pairs within probe sets (residual variance). Unlike variance filtering, FLUSH operates at the raw un-normalized probe-level data, thus it is not affected by the possible bias due to imperfect normalization. From our experience, low intensity genes tend to have higher residual variability, i.e. more inconsistent hybridization patterns across the experimental replicates. FLUSH can account for intensity, since we can use a flexible weight to penalize high residual variance, which is associated with low intensity features. Filtering genes prior to DE analysis might be viewed with some suspicion, as important differentially regulated features might be lost. There is obviously a sensitivity-specificity trade-off, since without filtering the great amount of spurious signals present in microarray data will make it hard to detect the real information. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

The discovery of chromatin remodelling has revolutionized chromatin research by providing new insights into how the packaging of the eukaryotic genome into nucleosomes participates in gene regulation. Instead of being a static structure, chromatin is now accepted to have a dynamic organization from the nucleosomal level up to higher order structures (). The identification of numerous multiprotein complexes that are involved in histone modification and remodelling of nucleosome arrays has revealed previously unappreciated levels of control over the basic chromatin organization, with defects in these processes leading to inappropriate gene expression and disease. To understand the dynamic mechanisms that generate these specialized chromatin structures and predispose genes to activation or repression, we have focused on the Tup1-Ssn6 co-repressor and the Swi-Snf co-activator whose interplay regulates the balance between repressed and active chromatin structures at a number of yeast genes. Swi-Snf is arguably the best-known example of a chromatin remodelling complex that can act as a transcriptional co-activator (). This large multi-subunit complex is targeted to gene promoters by sequence-specific DNA-binding transcription factors that interact with the Snf5 and Swi1 subunits (,). Swi-Snf utilizes the energy from ATP hydrolysis to alter the structure of chromatin, thereby enhancing nucleosomal DNA accessibility and enabling gene transcription (). Potential mechanisms of action include local DNA deformation resulting in nucleosome sliding, and histone octamer transfer in and in (). Conversely, the Tup1-Ssn6 co-repressor complex has been shown to organize chromatin into a repressive structure, possibly through direct contact with hypoacetylated histones, and in conjunction with histone deacetylases (). The subsets of genes regulated by these complexes overlap at the and genes (). Indeed, these genes represent a paradigm for chromatin-mediated gene regulation and offer a unique insight into the interplay between the two complexes. The traditional model for chromatin remodelling complexes has focused on their activity at gene promoters (,). However, evidence is emerging that remodelling also involves longer-range effects. We have shown that remodelling by Swi-Snf, as well as by Tup1-Ssn6, extends some distance upstream of the gene promoter (). Long-range remodelling has also been shown to occur over the coding regions of genes. For example, Tup1-Ssn6 and Isw2 cooperate to position a regular array of nucleosomes over the promoter and gene-coding region of the repressed gene (). Upon induction, Swi-Snf is required to disrupt this array (). Long-range remodelling by Swi-Snf and Isw1 has also been observed at the gene in a mini chromosome context, where induction of transcription was accompanied by the disruption of nucleosomes over the entire gene sequence and flanking regions (,). An episomal yeast gene has also been shown to be subject to extensive chromatin remodelling over the entire open reading frame (ORF), 5′ and 3′ flanking regions following activation (). In each of the above examples, the data suggests that remodelling is a pre-requisite for transcription, and not a consequence of it. In this study, we have characterized in detail the nature and precise extent of chromatin remodelling in the region upstream of the yeast gene. Using indirect end-labelling and primer extension analyses, we have mapped remodelling events attributable to both Swi-Snf and Tup1-Ssn6 over a 7.5 kb region covering the promoter and far upstream sequences. This extensive nucleosome map reveals how remodelling by both complexes extends as much as 5 kb upstream of . The long-range chromatin remodelling activities reported here support the proposition that it is a general mechanism used by chromatin remodelling complexes such as Swi-Snf and Tup1-Ssn6 to organize extensive chromatin domains in response to cues for transcriptional activation and repression. The yeast strains used were wild-type (wt), AFH41 [S288C a - - ::HIS3 ::LEU2/pUK499 ()]; and its derivatives , AFH44 (AFH41 ); , AFH47 (AFH41 ) and , AFH410 (AFH44 ) (). Yeast were grown in YP supplemented with glucose (2%) to mid-log phase. The cultures were divided into two equal portions and harvested by centrifugation. Cell pellets were washed twice in sterile water and resuspended to the same cell density in fresh YP containing glucose at either 2% (repressed, R) or 0.05% (derepressed, D) (). The cultures were incubated for a further 120 min at 30°C and harvested for either RNA or nuclei preparation. Indirect end-labelling and primer extension analysis were performed as described (). For the preparation of glucose-repressed nuclei, repression was maintained by supplementing all buffers with 2% glucose (). For indirect end-labelling, HindIII-digested DNA was probed with PCR fragments corresponding to base-pairs −1099 to −817, −4215 to −3948 and −7801 to −7519 upstream of the ATG ( SGD ID: S000001424); DraI-, HinfI- and BamHI-digested DNAs were probed with fragments −1736 to −1359, +1 to +285 and −4033 to −4303, respectively and BsrBI-digested DNAs were probed with fragments −2324 to −1899 and −2697 to −2399. For primer extension analysis the primers used correspond to positions −797 to −772 (P1); −3155 to −3131 (P2) and −3587 to −3563 (P3) of the upstream sequence. Band intensities were determined by phosphoimager analysis (FujiFilm FLA2000 FluoroImager). Total RNA was extracted from cells grown as described above and 20 μg samples were analysed by northern blot after electrophoresis in a 1.25% agarose-formaldehyde gel (). and transcripts were analysed with probes corresponding to ORF positions +119 to +1222, +411 to +1421, +26 to +620, +598 to +920, +39 to +351 and +47 to +388, respectively. Total RNA was treated with DNaseI (Promega) and cDNA was generated using a poly-dT primer and Superscript III reverse transcriptase (Invitrogen). Real time PCRs for and were performed with 1/25 dilutions of the cDNA reactions using the SYBR Green Master Mix (SuperArray) in an ABI 9700 PCR machine. RT-PCRs for and cDNAs were performed on a 1/5 cDNA dilution, and undiluted cDNA reactions. Values were normalized to RNA. Real time PCRs were performed in triplicate. The primer sequences used were: , 5′-CCATTGCTATCGCTCCCAAG-3′ and 5′-TGGAGCCAGAGAAAGCACCT-3′; , 5′-GAGGTTGCTGCTTTGGTTATTGA-3′ and 5′-ACCGGCTTTACACATACCAGAAC-3′; , 5′-TCCCAGAAGCCACTCTTGGT-3′ and 5′-AAACCCCAAAGTTCGATCCC-3′; , 5′-GCTCGCTTTGATCTTGACCC-3′ and 5′-TGGAAGACATCTCCCCTAGCA -3′; , 5′- ACTACCCGGCAATCTGCTGT-3′ and 5′-GGAATGACCCTTTCTGGACCA-3′. The gene encodes the enzyme invertase required for sucrose utilization and is subject to glucose repression through the Tup1-Ssn6 complex (,). As previously shown, transferring cells grown at high glucose concentration (repressing, R) to low glucose conditions (derepressing, D) induces transcription (B, lanes 1, 2 and 1C) and a deletion mutation of that cripples the remodelling activity of the Swi-Snf complex, abolishes this induction demonstrating Swi-Snf dependence (B, lanes 3, 4 and 1C) (,). Conversely, deletion of either (B, lanes 5, 6 and 1C) or (data not shown) results in high-level constitutive transcription (,). In the absence of both the Swi-Snf and Tup1-Ssn6 complexes, is also constitutively transcribed, but at lower levels (B, lanes 7, 8 and 1C) (,,). Numerous studies had suggested that upon gene activation or repression chromatin remodelling is limited to the immediate promoter region. However, evidence is emerging that remodelling complexes can in fact operate over longer distances. Indeed, our previous work indicated that this was the case for Swi-Snf and Tup1-Ssn6 regulation of transcription (). If similar long-range remodelling were apparent far upstream of , it would strengthen the case that the ability to organize extensive chromatin domains is a general feature of Swi-Snf and Tup1-Ssn6. However, prior to embarking upon an analysis of chromatin remodelling events upstream of , we first characterized the extent of distal gene activity potentially affecting this region. is located ∼35 kb from the telomere on the left arm of chromosome IX (). Four ORFs are located in the 10 kb region upstream of . Two ORFs (/ and /) are interrupted by a stop codon in S288C strains, while the short has dubious ORF status (A). We analysed these ORFs for transcriptional activity, and determined whether they were subject to regulation by Tup1-Ssn6 or Swi-Snf. From northern blot analysis, only the far upstream ORF yielded a detectable transcript, and this was unaffected in any of the mutant backgrounds under conditions of glucose repression and derepression (data not shown). Further analysis of the remaining genes by quantitative RT-PCR detected no significant transcription (C). We therefore chose the 3′ end of the ORF as the upstream boundary for our mapping analysis. This provided a transcriptionally ‘quiet’ region spanning 7.5 kb upstream of in which to identify chromatin remodelling effects specific to Swi-Snf and Tup1-Ssn6 that could be attributed to without interference from neighbouring gene transcription. We used micrococcal nuclease (MNase) digestion and the methods of indirect end-labelling and primer extension analysis to determine nucleosome positions at the upstream chromosomal locus (). These methods detect translationally positioned nucleosomes by virtue of the protection they afford to nucleosomal DNA against MNase digestion. Nucleosomes are allocated to ∼145 bp regions of protection between strong cut sites in the chromatin cleavage pattern as compared with the corresponding region of digestion in the naked DNA. We mapped 7.5 kb of nucleosome array upstream of the derepressed gene in a variety of strain backgrounds in order to distinguish the effects of different chromatin remodelling activities. Chromatin from and deletion strains revealed remodelling in the absence of the Tup1-Ssn6 and Swi-Snf complexes, respectively. Because these remodelling events could directly reflect the absence of a particular remodelling complex, or the unmasking of underlying activities of the other complex, the double deletion strain was also included in our analyses to distinguish this. Chromatin from wild-type (wt), and nuclei was characterized under glucose derepressed conditions, in which only the strain does not express . To reveal the full impact on the upstream chromatin resulting from Tup1-Ssn6 repression of the gene, we additionally mapped nucleosome positions in a wt strain grown under glucose repressed conditions. Finally, the corresponding 7.5 kb naked DNA control pattern of MNase cutting allowed us to unequivocally identify the sites of nucleosomal protection in the chromatin patterns and to assign nucleosome positions. Multiple short-range maps were required to cover the full 7.5 kb region (). In all regions where mapping was performed from both upstream and downstream restriction sites, the data were consistent. Densitometry traces from each analysis were linearized and combined to produce the composite map of the upstream region that is shown in . Our data confirm previous studies demonstrating chromatin remodelling at the promoter upon glucose derepression in wt cells. Here, strongly positioned nucleosomes in repressed chromatin that occlude the TATA box and upstream activating sequence (UAS) revert to a naked DNA pattern in derepressed chromatin [TATA box: A, B and D, compare wt (R) with wt (D) and DNA; UAS: G and H, compare wt (D) with DNA (N)] (). The remodelling of these nucleosomes on the proximal promoter was attributed to the Swi-Snf complex because this remodelling was absent in derepressed chromatin [B, E and G, H; compare wt (D) and (D)] (,). Under repressing conditions, Tup1-Ssn6 is recruited to the proximal promoter by the DNA-bound Mig1p, Nrg1 and Sko1 repressors in response to signal transduction pathways (,). In glucose derepressed chromatin, we confirm the remodelling of the nucleosome that occludes the TATA box in wt repressed chromatin (C) (,). Our analysis of derepressed chromatin at the TATA box also shows that this remodelling occurs even in the absence of Swi-Snf (and Tup1-Ssn6) [C and F, compare (D) to (D)] (,). These data are consistent with the suggestion that factors other than Swi-Snf may be involved in the remodelling of this nucleosome in the absence of Tup1-Ssn6 (). However, an alternative interpretation is that the presence of the positioned nucleosome at the TATA box is dependent on dominance of the Tup1-Ssn6 complex (as in wt repressed and chromatin), rather than its removal being dependent on the Swi-Snf complex. This would imply a default active chromatin pattern in the absence of both remodelling complexes, matching the gene activity in the double deletion mutant. As we mapped chromatin further upstream from the promoter, we identified two strong MNase cut sites in derepressed chromatin that indicated the presence of a positioned nucleosome at around −1600 bp [ (D): A, B (black gel trace) and C; the two black arrowheads denote the strong cut sites, and the black oval indicates the corresponding positioned nucleosome]. A similar but less distinct pattern was present in wt repressed [wt (R): A, B (white gel trace) and C] and derepressed chromatin (data not shown, see gel trace in ). However, in derepressed wt chromatin, digestion between these cut sites at −1600 bp indicated the loss of nucleosomal protection in this region [wt (D): A, B (upper grey trace) and C; white arrowhead indicates digestion at −1600 bp, and the dashed oval signifies a remodelled nucleosome at this site as compared to (D) and wt (R) chromatin]. This remodelling event is attributable to Swi-Snf, since in the absence of this complex ( deletion) the nucleosome was reinstated. In derepressed chromatin, a digestion pattern equivalent to naked DNA was observed [B, compare (D) lower grey trace to black DNA trace], which was also linked to the presence of the Swi-Snf complex since its absence ( deletion) again reinstated the nucleosomal pattern. In chromatin, the naked DNA-like cutting pattern continued from this point onward towards the coding region over a distance of at least 1500 bp. In contrast, a well-defined nucleosomal array was present in wt repressed chromatin. This was particularly noticeable when the repressed and derepressed wt digests were electrophoresed side by side as bands are narrower and more distinct in wt repressed chromatin [A, compare wt (R) and wt (D)]. Derepressed wild-type and chromatin showed hybrid traces, where some nucleosome boundaries were preserved and others were lost. Most notably, a boundary at −1200 bp is clearly absent in derepressed wt chromatin [wt (D): A, B (upper grey trace) and C; white circle denotes protection from digestion at this site as compared to wt (R) chromatin]. Overall, the data suggests Swi-Snf dependent remodelling dominates this chromatin region following induction. Chromatin remodelling was also observed ∼3.3 kb upstream of the coding region. In wt repressed chromatin, a pattern distinct from the naked DNA control indicated the presence of nucleosomes in this region [A and B, compare wt (R) to DNA]. The pattern included a cut site at −3400 bp that was weak or not observed in or derepressed chromatin, and which most likely represents a DNA region between two positioned nucleosomes (A–C, black arrowhead). Protection of a cut site apparent in naked DNA and the and derepressed chromatin was also evident in the wt repressed strain at −3300 bp (A and B, white arrowhead), indicating a positioned nucleosome at this region [C, wt (R), black oval]. In derepressed wt chromatin, a diffuse cutting pattern was indicative of a less precisely positioned nucleosome occupying this region [data not shown, see wt (D) gel trace in ]. This pattern persisted in derepressed chromatin, suggesting that the remodelling is Swi-Snf-independent [A, (D), 4B (white trace) and 4C (overlapping white ovals)]. By contrast, in derepressed chromatin, the cleavage pattern was similar to that seen in naked DNA [A, B (compare the grey trace to lowermost black trace) and C]. However, when was also deleted in the mutant (data not shown, see trace in ), the pattern reverted to that seen in the derepressed wt chromatin, suggesting that Swi-Snf remodels this nucleosome, but only in the absence of Tup1-Ssn6. Therefore, it appears to be the balance between the positioning activity of Tup1-Ssn6 and the remodelling activity of Swi-Snf that determines the position of this particular nucleosome upon activation. High resolution analysis of the −3500 to −3200 region by primer extension confirmed the naked DNA-like pattern in the absence of Tup1-Ssn6, confirming nucleosome remodelling at this site [D and E, compare (D) to DNA (N)]. Furthermore, by superimposing the indirect end-labelling cleavage patterns (E, white traces) onto the primer extension cleavage patterns (E, black traces), the data from the two techniques show a strong correlation. This also validates the accuracy of the lower-resolution but longer-range indirect end-labelling nucleosome mapping method used primarily in this study. A further instance of remodelling was detected at −2900 bp, where a strong cut site between two positioned nucleosomes was present in the chromatin of all strains but much weaker in derepressed chromatin (A, white circle). As enhanced cutting at −2900 bp was restored in chromatin, the Swi-Snf complex appears responsible for this effect. Proceeding more distally from the promoter, a further instance of remodelling was suggested at −4800 bp, signified by a wide peak of increased cutting in derepressed chromatin (A, black arrowhead and Supplementary Figure S4). In derepressed chromatin, however, protection from digestion at this site was compatible with the presence of a nucleosome, suggesting Swi-Snf was responsible for the subtle remodelling observed in chromatin (A, compare black and white traces). To confirm a role for Swi-Snf in remodelling at this site, we analysed the region in greater detail by indirect end-labelling analysis from a BamH1 restriction site present ∼800 bp upstream of the putative remodelled nucleosome (B and Supplementary Figure S5). In derepressed wt chromatin, nucleosomal protection was weaker at −4800 bp and appeared as a shoulder (B, black arrowhead, black trace), similar to that observed in chromatin, and additional enhanced cutting was also visible at a site −5300 bp upstream of (B, white arrowhead). This enhanced cutting was not evident in the repressed wt or derepressed chromatin patterns (B, grey and white traces) and is compatible with a labile nucleosome being displaced in derepressed wt chromatin or in the absence of . This remodelling event is dependent on the Swi-Snf complex, as deletion of restored the positioned nucleosome. As we mapped nucleosome positions from −5500 bp upstream of towards the transcriptionally active ORF , which formed the 7.5 kb upstream boundary of our mapping analysis, we found over 2 kb of chromatin organized into a very regular array of nucleosomes ( and Supplementary Figure S6). The positions of 12 nucleosomes were easily distinguished by comparison with the naked DNA trace. Significantly, no differences could be detected between different strain backgrounds as the MNase cutting patterns were essentially superimposable. This argues against the occurrence of randomly distributed chromatin remodelling events in the various yeast mutants. Thus, chromatin remodelling activity does not extend further than 5.5 kb upstream of . We have detected and characterized long-range chromatin remodelling by Swi-Snf and Tup1-Ssn6 in the extended upstream region following glucose derepression. The gene is specifically induced by low glucose within a 7.5 kb region free of other transcriptional activity. Our results, summarized in , suggest that chromatin remodelling extends far beyond what is generally considered the promoter region, with the most distant event detected at −4800 bp. Other remodelling events occur at −3300, −2900, −1500, −1100, −500 and −120 bp. Significantly, no instances of remodelling were observed within the ∼2.5 kb of chromatin analysed upstream of this region. This suggests that the characterized long-range chromatin alterations by the Tup1-Ssn6 and Swi-Snf remodelling complexes are linked to their control of transcription. Our data confirm previous studies showing chromatin remodelling of nucleosomes at the TATA box and UAS following derepression (,). However, the observation that remodelling at the TATA box is also found in derepressed chromatin is incompatible with the earlier proposal that the Swi-Snf complex is responsible for this event (,,). Instead, it points to the dependence of the position of this nucleosome on the Tup1-Ssn6 complex. Consistent with this, recent work has shown that Tup1 can regulate Rap1 binding by controlling nucleosome occupancy at some Rap1 binding sites (). However, the involvement of remodelling complexes other than Swi-Snf in disrupting the array also cannot be discounted (). Upstream from the promoter, the less dramatic, but reproducible nucleosome remodelling events in the various strain backgrounds displayed particular characteristics. Thus, instances of remodelling in derepressed chromatin were in most cases reversions to the naked DNA pattern [, (D)]; derepressed chromatin bore the greatest resemblance to the repressed wt pattern [, compare (D) and wt (R)]; whereas derepressed chromatin most resembled wt derepressed chromatin [, compare (D) and wt (D)]. This behaviour parallels the relationship between these strains in terms of activity of the gene (B and C). It is also consistent with our observations at the gene, which suggested that the balance between these antagonizing remodelling activities controls the chromatin organization of this gene (). Hence, depletion of both Tup1-Ssn6 and Swi-Snf has less impact on chromatin structure and gene activity than the absence of a single complex, which mimics the situation where the other complex dominates. The many apparent reversals to the naked DNA pattern in chromatin indicate nucleosome loss or randomization rather than translational rearrangement of nucleosome positions. Nucleosome loss has been shown for the TATA box of the active gene (). Near the coding region, the MNase cleavage pattern of naked DNA bears some similarity to the nucleosomal pattern, most likely because the sequence-specificity of the nuclease reveals a biased nucleotide sequence distribution that is occasionally in phase with the nucleosome array, as has been noted for other genes (,). This might leave less freedom for the repositioning of nucleosomes, although a randomization or mobilization of nucleosome positions might also resemble a naked DNA pattern. The far upstream reversals to the naked DNA pattern in derepressed chromatin are due to Swi-Snf activity, since they are not observed in chromatin [, compare (D) and (D) at −3300 and −4800 bp]. This underlines the roles of the Tup1-Ssn6 and Swi-Snf complexes in respectively organizing and disrupting nucleosome arrays. On this basis, wt repressed and chromatin represent the dominant effects of Tup1-Ssn6 on the nucleosome array, and derepressed wild-type [where Tup1 persists ()] and chromatin represent the dominant role of Swi-Snf. The chromatin shows that, at the gene, the remodelling complexes operate largely within the framework of an array of nucleosome positions predetermined by the DNA sequence (). The upstream sequence includes many transcription factor-binding sites, some of which are unique while others are redundant. However, only a fraction of consensus sites are occupied by their respective binding factors, and this restricts our ability to relate individual changes in the nucleosome array to features in the underlying DNA sequence (,). For example, the DNA sequence at the −4800 bp Swi-Snf-remodelled site harbours a unique Rox1-binding site. Although Rox1 is known to recruit Tup1-Ssn6 to repress hypoxic genes under aerobic conditions, we do not expect that this site is occupied under our conditions (). An important suggestion from our work is that long-range chromatin remodelling may well be a general feature of chromatin modifying complexes such as Tup1-Ssn6 and Swi-Snf. Promoter-centred models of chromatin remodelling reflect the emphasis of gene regulation research on the proximal upstream region of genes. Few studies have investigated the effects of remodelling into intergenic regions. However, it has been found that Swi-Snf-dependent chromatin remodelling extends along the circular chromatin of episomes carrying the gene, including the coding region (,). has also been shown to be subject to remodelling over the entire length of the coding region and promoter (). Furthermore, genome-wide chromatin immunoprecipitation studies have localized the Swi-Snf-related RSC complex at many intergenic locations as well as at promoters (). Our findings, both at and , indicate that remodelling complexes can indeed function at such intergenic regions, and in a manner that can be correlated to the activity of the genes they control. Although remodelling of the nucleosome array upstream of by the Tup1-Ssn6 and Swi-Snf complexes is not restricted to the proximal promoter, it does appear to be confined to within several kilobase upstream of this promoter. It is not clear whether this is a consequence of boundary effects (), or reflects the natural range of direct or indirect nucleosome remodelling effects emanating from a site of complex recruitment, or a distribution of complex-recruitment sites. Nearby genes that are not under the control of these complexes could also conceivably delimit remodelling (). Although the ORFs in the upstream region are not active under the conditions of this study, their bound factors or histone modification patterns might prevent the propagation of -associated nucleosome remodelling effects (). Far upstream instances of chromatin remodelling could equally be a direct consequence of the presence of Tup1-Ssn6 and Swi-Snf complexes targeted to these regions in addition to their documented promoter associations (,). In support of this model are the observations of Swi-Snf complexes forming loops and controlling the helical tension between attachment sites, and of Tup1-Ssn6 showing a continuous association along the chromatin fibre (,,). Alternatively, the remodelling could be at the level of higher order chromatin structure, as has been suggested for Swi-Snf activity (). Finally, recent observations of nuclear relocation of the gene upon activation or repression suggest possible extensive structural rearrangements (). Long-range chromatin remodelling seems at odds with the high gene density in yeast. What are seen at , and here at , may be just two examples demonstrating the potential of these complexes to remodel large domains when the gene under their control is in a less gene-dense region. The dynamic changes to such a large region of chromatin may parallel the extensive range of histone modifications, such as histone acetylation and methylation, that may also be in effect over larger regions of chromatin (). This regional chromatin organization may provide a background essential for gene regulation to take place, and ultimately determine the accessibility of the promoter. p p l e m e n t a r y D a t a a r e a v a i l a b l e a t N A R O n l i n e .

DNA polymerase proofreading removes misincorporated nucleotides at the primer-end (,), which significantly improves the fidelity of DNA replication (). Since increased epithelial tumors are observed in mice that express an exonuclease-deficient DNA polymerase δ, DNA polymerase proofreading is important in preventing mutations that lead to cancer (). DNA polymerase proofreading was first demonstrated to be a major determinant of replication fidelity for the bacteriophage T4 DNA polymerase (,,) and this DNA polymerase continues to be a valuable model for studies of proofreading, especially for Family B DNA polymerases, which include the eukaryotic DNA polymerases δ and ε and several viral DNA polymerases (,). The T4 DNA polymerase proofreading pathway has at least four steps (,). During chromosome replication, the proofreading pathway is initiated in the polymerase active center when an incorrect nucleotide is inserted (step 1), which hinders further primer elongation (,,,). The end of the primer strand is then separated from the template and transferred to the exonuclease active center (step 2), which requires a β hairpin structure in the exonuclease domain to form stable exonuclease complexes (). The terminal nucleotide is cleaved from the primer-end in the exonuclease active center (step 3) and then the trimmed primer-end is returned to the polymerase active center where nucleotide incorporation can resume (step 4). Genetic studies indicate that four of the five protein domains of the T4 DNA polymerase are involved in the proofreading pathway (). The genetic studies are corroborated by structural studies, which find significant conformational differences for the exonuclease, palm and thumb domains in polymerase complexes compared to exonuclease complexes (,). There are still unanswered questions about how the primer-end is shuttled back-and-forth between the polymerase and exonuclease active centers, which we address here. Many DNA polymerases are normally tethered to the DNA by a protein ‘clamp’, which is necessary for processive DNA replication. The phage T4 clamp, the product of gene , is also reported to stimulate proofreading (,), but is the clamp for processive transfer of the primer-end from the polymerase to the exonuclease active center and for transfer of the trimmed primer-end from the exonuclease back to the polymerase active center? We proposed that the clamp is essential for processive proofreading that initiates in the polymerase active center because greater intrinsic processivity in nucleotide incorporation is observed for mutant DNA polymerases that have reduced ability to initiate the proofreading pathway, while reduced processivity in primer extension is detected for mutant DNA polymerases that proofread more (). Intrinsic proofreading, however, is observed for the T4 DNA polymerase and the closely related RB69 DNA polymerase without its clamp (,). Coupled removal of an incorrect nucleotide and primer extension were observed under single turnover conditions in the presence of a heparin trap; however, it is not clear in these experiments if the T4 DNA polymerase first bound the DNA substrate in the polymerase or the exonuclease active center. If the T4 DNA polymerase bound the mismatched DNA initially in the polymerase active center, then the entire proofreading pathway beginning from strand separation and transfer of the primer-end from the polymerase to the exonuclease active center can be carried out without enzyme dissociation. However, if the T4 DNA polymerase can form exonuclease complexes directly without first forming polymerase complexes, then just the steps of hydrolysis and transfer of the trimmed primer-end from the exonuclease to the polymerase active center have been demonstrated to be processive in the absence of the clamp. Another outstanding question is the rate of active site switching. Proofreading during ongoing DNA replication is restricted primarily to incorrect nucleotides at the primer-end because the rate of primer extension for a matched primer terminus is much greater than the rate for initiation of the proofreading pathway, but replicative DNA polymerases have poor ability to extend a mismatched primer terminus, which then tips the balance in favor of proofreading (,,). Thus, there is a kinetic barrier to initiation of the proofreading pathway, which suggests that the rate of polymerase-to-exonuclease active site switching will be relatively slow. In contrast, transfer of the trimmed primer-end from the exonuclease to polymerase active center could be rapid if the corrected primer-end returns to the polymerase active center unassisted (). We have developed a fluorescence assay using the fluorescent adenine base analog 2-aminopurine (2AP) to examine shuttling of the primer-end between the polymerase and exonuclease active centers during the proofreading reaction catalyzed by the T4 DNA polymerase. This assay depends on two observations: (i) T4 DNA polymerase recognizes a terminal 2AP-T base pair as a mismatch and preferentially proofreads the mismatch before primer extension (), which we confirm in experiments reported here and (ii) T4 DNA polymerase forms distinct fluorescent complexes with different levels of fluorescence intensity depending if 2AP is in the or +1 position in the template strand (). Moderately, fluorescent exonuclease complexes are formed preferentially with 2AP in the position of the template strand (A) and highly fluorescent complexes are formed with DNA labeled at the +1 position (B) in which the primer-end is bound in the polymerase active center. Thus, exonucleolytic proofreading of DNA in which 2AP is initially in the position will produce an increase in fluorescence intensity as the moderately fluorescent exonuclease complexes are converted to the highly fluorescent complexes with 2AP in the +1 position. Since the primer-end is initially in the exonuclease active center and is then transferred to the polymerase active center after the terminal nucleotide is removed to form the highly fluorescent +1 complexes, the rate of increase in 2AP fluorescence intensity provides information on the rate of exonuclease-to-polymerase active site switching. We performed this experiment with wild-type and exonuclease-deficient T4 DNA polymerases under pre-steady-state, single-turnover conditions in which heparin was used to trap any free T4 DNA polymerase (). An increase in fluorescence intensity was observed for the wild-type T4 DNA polymerase at the rate of 145 ± 3 s, but not for an exonuclease-deficient T4 DNA polymerase. These results are discussed with respect to the overall proofreading reaction, active site switching, structural implications and replication fidelity of the wild-type and proofreading defective T4 DNA polymerases. Expression, purification and characterization of the wild-type and mutant D112A/E114A- and W213S-DNA polymerases were done as described previously (,). The 2AP-containing DNA substrates are described in . The substrates were prepared as described previously (,,). The 3′ terminus of the template strand of the DNA duplexes was protected from enzyme binding by attachment of a biotin (b) group (BiotinTEG-CPG, Glen Research). The 2AP phosphoramidite was purchased from Glen Research. All oligonucleotides were purified by gel electrophoresis. The primer and template strands were annealed in buffer containing HEPES (pH 7.5) and 50 mM NaCl with a 20% excess of the oligonucleotide without 2AP to ensure complete hybridization of the 2AP containing strand. The non-2AP containing DNA substrates used for were synthesized using standard procedures and purified by gel electrophoresis. The annealing conditions were the same as used for the 2AP-containing oligonucleotides. The template strand was in 20% excess to ensure complete hybridization of the P-labeled primer strand, which was labeled using a standard T4 polynucleotide kinase labeling procedure (). Reaction mixtures (20 μl) contained 50 nM DNA, 150 nM DNA polymerase, 25 mM HEPES (pH 7.5), 50 mM NaCl, 1 mM DTT, 0.5 mM EDTA and 100 μM dNTPs as indicated. The reactions were first pre-incubated at 37°C and then started by the addition of a solution of Mg/heparin to give a final concentration of 8 mM Mg and 1 or 0.1 mg/ml heparin as indicated (Sigma, 3000 average molecular weight from porcine intestinal mucosa). Reactions were stopped after 15 s by addition of 20 μl gel loading solution (95% formamide, 20 mM EDTA, and xylene cyanol and bromphenol blue dyes). Reddy () used heparin at 1 mg/ml, but we find that 0.1 mg/ml is sufficient (). The reaction products were separated on DNA sequencing type gels containing 15% acrylamide and 8 M urea. The P-labeled products were visualized by using a PhosphorImager (Molecular Dynamics, Inc.). The same reaction conditions were used as described above for single-turnover reactions except that the heparin trap was omitted. Stopped-flow experiments were performed with the Applied Photophysics SX.18 MV instrument, which allowed us to determine the pre-steady-state kinetics of selected aspects of the proofreading pathway. Excitation was at 310 nm; a 320 nm cutoff filter was used. The temperature in the sample-handling unit was maintained at 20.0 ± 0.5°C. Reactions were initiated by mixing equal volumes of a solution of T4 DNA polymerase–DNA complexes, which contained 400 nM DNA labeled at the position with 2AP (A), 1000 nM T4 DNA polymerase, 25 mM HEPES (pH 7.5), 50 mM NaCl, 1 mM DTT and 0.5 mM EDTA with a second solution containing 16 mM MgCl, 0.2 mg/ml heparin, 25 mM HEPES (pH 7.5), 50 mM NaCl and 1 mM DTT. After mixing, the final concentrations of reaction components were 200 nM DNA polymerase–DNA complexes, 25 mM HEPES (pH 7.5), 50 mM NaCl, 1 mM DTT, 0.25 mM EDTA, 8 mM MgCl and 0.1 mg/ml heparin. The optimal DNA and enzyme concentrations to ensure full complex formation were determined by titration experiments (). In general, a ⩾2-fold excess of DNA polymerase over DNA produces maximal complex formation for DNA concentrations from 200 to 600 nM. Curves were fit either to single (monophasic) or double (biphasic) exponential equations. Six or more runs were performed with each set of reaction conditions; mean values were calculated. The same reaction conditions were used as described above for single-turnover reactions except that the heparin trap was omitted. Reddy () used heparin to obtain single-turnover (single-encounter) conditions for exonuclease and nucleotide incorporation reactions with the T4 DNA polymerase; any DNA polymerase molecules that dissociate from the DNA template were prevented from rebinding to the DNA substrate by forming stable complexes with heparin. Heparin is indeed a useful ‘trap’ for the T4 DNA polymerase. If 1 mg/ml heparin, the concentration used by Reddy (), is added to exonuclease or primer-extension reactions with the matched DNA substrate (A) the addition of the T4 DNA polymerase, no activity is detected (B, lanes 1 and 3). In the absence of heparin, exonucleolytic degradation (B, lane 2) and full primer extension (B, lane 4) are observed. Single-turnover primer extension reactions with matched and mismatched DNA substrates (A) were carried out with the wild-type T4 DNA polymerase. Reaction components were first pre-incubated in the absence of Mg and then the reactions were initiated by the addition of a solution of Mg/heparin. In the primer extension reaction with the matched DNA substrate in which the only nucleotide provided was dCTP, the wild-type T4 DNA polymerase fully extended most of the primer by two nucleotides; only a small amount of partially extended +1 product was detected (C, lane 1). Although the wild-type T4 DNA polymerase has a potent exonuclease activity, only traces of products less than the length of the primer strand were observed (C, lane 1). Degradation products were detected because the only nucleotide provided in these reactions was dCTP, which means that if there was any primer degradation—first removal of the terminal dTMP, then another dTMP, etc. (A) the primer could not be resynthesized. Thus, the wild-type T4 DNA polymerase formed primarily polymerase complexes with the matched DNA substrate that were poised for nucleotide incorporation rather than exonuclease complexes poised for primer degradation. Primer extension was also detected with the mismatched T-T DNA substrate (A) in which dCTP and dATP were provided (C, lane 2). Because the wild-type T4 DNA polymerase cannot efficiently extend a mismatched primer-end (,,), the primer extension observed with the mismatched DNA substrate must have been preceded by removal of the incorrect terminal dTMP, which was followed by transfer of the trimmed primer-end from the exonuclease to the polymerase active center, incorporation of dAMP and then incorporation of two dCMPs. All steps were performed without dissociation of the DNA polymerase since the heparin trap was present. To further demonstrate that exonucleolytic proofreading of the mismatched primer terminus is required before the primer can be extended, experiments were repeated with the exonuclease deficient D112A/E114A-DNA polymerase under the same conditions used for the wild-type T4 DNA polymerase. The D112A/E114A-DNA polymerase has an alanine substitution for an essential aspartate (D112) residue in the exonuclease active center and, as a consequence, has almost no detectable 3′ → 5′ exonuclease activity (). While the D112A/E114A-DNA polymerase extended the matched DNA substrate under single-turnover conditions (C, lane 5) as efficiently as the wild-type T4 DNA polymerase (C, lane 1), no extension was observed for the mismatched DNA substrate (C, lane 6). We also tested the ability of the W213S-DNA polymerase to carry out primer extension reactions of the matched and mismatched DNA substrates under single-turnover conditions. The W213S-DNA polymerase replicates DNA with reduced fidelity (), which is due to reduced exonuclease activity. Significantly less degradation of single-stranded DNA was observed for the W213S-DNA polymerase compared to the wild-type T4 DNA polymerase in multiple-turnover reactions (; compare wild-type activity in lanes 1 and 2 to that of the W213S-DNA polymerase in lanes 3 and 4). Even less exonuclease activity was detected for the W213S-DNA polymerase on double-stranded DNA (compare wild-type activity in lane 5 to that of the W213S-DNA polymerase in lane 8). In single-turnover reactions with the W213S-DNA polymerase and the matched DNA substrate, dCMP incorporation was observed (C, lane 3), but primer extension was not as efficient as observed for the wild-type and D112A/E114A-DNA polymerases (C, lanes 1 and 5, respectively). In single-turnover reactions with the W213S-DNA polymerase and the mismatched DNA substrate (C, lane 4), a small amount of +2 extension product was detected, which indicates that the W213S-DNA polymerase can catalyze only a very limited processive proofreading-nucleotide incorporation reaction. The DNA substrate labeled with 2AP in the (terminal) position of the primer strand (C) was labeled with P at the 5′-end of the primer strand. Single-turnover experiments were performed as were done for the reactions shown in C except that the concentration of heparin was reduced from 1 to 0.1 mg/ml, which we found to be as effective (). Reactions contained dTTP, dCTP and dATP; thus, successful primer extension will extend the primer by 4 nt. The wild-type T4 DNA polymerase produced the +4 extension product under single turnover conditions with the heparin trap (, lane 2), but much less primer extension was observed for the W213S-DNA polymerase (, lane 1). No primer extension product was detected for the D112A/E114A-DNA polymerase (data not shown), which indicates that extension of the 2AP-T terminal base pair cannot be done under single turnover conditions by this mutant enzyme. Thus, the processive primer extension reaction first required removal of the terminal 2AP nucleotide from the primer-end, then transfer of the trimmed primer-end to the polymerase active center and finally nucleotide incorporation. The reactions shown in demonstrate that the terminal 2AP-T base pair with 2AP in the terminal position of the primer strand is recognized as a mismatch by the T4 DNA polymerase. The same is true if 2AP is in the position in the template strand and T is in the terminal position of the primer strand (,). The DNA substrate labeled with 2AP in the position of the template strand (A) is used in the next experiments. The previous experiments demonstrate that the T4 DNA polymerase can proofread a T-T mismatch (C) and a 2AP-T terminal base pair () and then incorporate nucleotides without dissociating from the DNA substrate; however, it is not possible to determine from these experiments if the proofreading pathway initiated in the polymerase or the exonuclease active center or in both. These experiments also do not provide information about the rate of active site switching. Both questions were addressed by measuring the increase in 2AP fluorescence intensity in stopped-flow experiments for the conversion of the DNA substrate with 2AP in the position in the template strand to the DNA with 2AP in the +1 position (the DNAs are described in A and B, respectively). Two rates were detected for removal of the 2AP nucleotide from the primer-end under multiple turnover conditions for DNA substrates like the DNA described in C (,,); thus, two rates are also expected for the removal of dTMP opposite template 2AP for the DNA substrate described in A. The two rates observed in previous experiments were explained by the proposal that the DNA primer/template exists in two states in solution: (i) an annealed state, which is the substrate used by the T4 DNA polymerase for forming complexes with the primer/template bound in the polymerase active center and (ii) a melted state, which is the preferred substrate for forming exonuclease complexes in which the end of the primer strand is bound in the exonuclease active center (). The faster of the two rates was attributed to proofreading reactions that initiated in the exonuclease active center (activated complexes) and the slower rate was attributed to reactions in which the template/primer was first bound in the polymerase active center and, thus, were initially inactive for the hydrolysis reaction. If two rates are detected in the presence of the heparin trap for removal of dTMP opposite template 2AP, then complexes formed initially with the primer/template in the polymerase active center and complexes formed with the primer-end bound in the exonuclease active center can both support processive proofreading reactions. If a single rate is observed, then only one type of complex can carry out the proofreading reaction processively. For reactions initiating in the exonuclease active center, a rate as fast as 100 to >200 s may be observed, which are the reported rates for hydrolysis of the terminal phosphodiester bond in the exonuclease active center (,). A ⩾10-fold slower rate is expected for proofreading reactions that initiate in the polymerase active center (,,,). A solution of moderately fluorescent complexes was formed with the wild-type T4 DNA polymerase and DNA labeled in the position in the template strand (A). This solution was mixed with an equal volume of a second solution containing Mg/heparin in the stopped-flow, which produced an increase in fluorescence intensity at the observed rate, = 145 ± 3 s (A). The curve was best fit by a single exponential equation. Because a single rate was observed in the range of the reported hydrolysis rate, processive proofreading appears to be detected only for complexes in which the primer-end was bound initially in the exonuclease active center. Thus, the addition of Mg to the preformed complexes triggered the following chain of steps: excision of the terminal phosphodiester bond in the exonuclease active center to remove dTMP and transfer of the trimmed primer-end to the polymerase active center to form the highly fluorescent complexes in which 2AP is now in the +1 position. In experiments with the P-labeled DNAs (C and ), nucleotide incorporation follows return of the trimmed primer-end to the polymerase active center. The highly fluorescent complexes with 2AP in the +1 position are also poised for nucleotide incorporation since these complexes bind the correct nucleotide rapidly within the dead time of the stopped-flow instrument (,). No increase in fluorescence intensity was detected in reactions with the proofreading deficient W213S-DNA polymerase, as expected since this mutant DNA polymerase has only very limited ability to carry out processive proofreading as demonstrated in the primer extension assays (C and ). The above experiments were repeated without the heparin trap. The increase in fluorescence intensity was biphasic in reactions with the wild-type T4 DNA polymerase. The best curve fit was achieved by using a double exponential equation; the faster rate was 106 ± 10 s and the slower rate was 11 ± 1 s (B, ). The two rates indicate that two distinct populations of complexes were formed initially: one population (∼40% of the complexes) can form the highly fluorescent +1 complexes at about the same rate as detected in the presence of the heparin trap and a second population (∼60% of the complexes) that forms the +1 complexes at a 10-fold slower rate (). Since a single 145 s rate was observed in the presence of the heparin trap, only the population of complexes that proofreads at the apparent rate of ∼106 s rate appears to carry out the proofreading reaction processively. Since heparin prevents detection of the slower 11 s rate, complexes responsible for this rate must initially be inactive and can convert to active complexes only by an intervening dissociation, heparin-sensitive step. A single rate of 1.5 ± 0.1 s was observed for the W213S-DNA polymerase (). This slow rate is consistent with the severely reduced ability of this mutant DNA polymerase to degrade single- and double-stranded DNA (). The T4 DNA polymerase and the closely related RB69 DNA polymerase can remove two incorrect nucleotides and then extend the primer terminus under single-turnover conditions in the presence of the heparin trap (,). We repeated the above experiments with the DNA substrate illustrated in D, which has two incorrect G nucleotides at the end of the primer strand and 2AP is in the position in the template strand. Moderately, fluorescent exonuclease complexes are formed with this DNA substrate (,). After removal of the two incorrect G nucleotides, 2AP will be in the +2 position (D). T4 DNA polymerase complexes formed with 2AP in the +2 position of the template strand are only weakly fluorescent (). Thus, removal of the two terminal incorrect G nucleotides is expected to produce an overall decrease in fluorescence intensity, but will there be an intervening increase in fluorescence intensity after removal of the terminal incorrect nucleotide since 2AP will be transiently in the +1 position? Highly fluorescent +1 complexes are not expected to be formed after removal of the first incorrect nucleotide since these complexes are not detected if the terminal base pair is mismatched (). No increase in fluorescence intensity was observed; fluorescence intensity decreased at the rate of 55 ± 2 s (, ). The same rate of decrease in fluorescence intensity was also observed without the heparin trap, which indicates that none of the complexes formed during the process of removing two incorrect nucleotides are sensitive to the heparin trap. The T4 DNA polymerase exonuclease activity degrades single-stranded DNA one nucleotide at a time from the 3′-end; hence, the rate of 55 s for removing two nucleotides is the combined rate for two consecutive excision steps. If removal of the penultimate incorrect nucleotide occurs at the rate of ∼145 s, as determined for removal of a single incorrect nucleotide (A), then it is possible to calculate the overall rate for removal of the terminal incorrect nucleotide by using the following equation: 1/ = 1/ + 1/. By rearranging the equation, 1/ = 1/–1/ = 1/55–1/145. Thus, = 88.5 s. The slower apparent rate for removal of the terminal nucleotide compared to the rate for removal of the second incorrect nucleotide suggests that there are extra steps for removal of the terminal nucleotide. We propose that after removal of the terminal nucleotide, the trimmed primer-end is returned to the polymerase active center. This proposal is reasonable since the correctness of the primer-end can only be examined in the polymerase active center where hydrogen bonding between the terminal base on the primer strand and the complementary template base can be evaluated as well as the geometry of the terminal base pair (,). Such a mechanism must exist in order to explain how exonucleolytic proofreading is limited primarily to the removal of incorrect nucleotides. If the primer-end is found to be incorrect, then the primer-end is returned to the exonuclease active center for a second cycle of excision, and then the further trimmed primer-end is returned to the polymerase active center. We developed a fluorescence assay that uses changes in 2AP fluorescence intensity in 2AP-labeled DNA to monitor the proofreading reaction catalyzed by the T4 DNA polymerase. T4 DNA polymerase forms moderately fluorescent exonuclease complexes with duplex DNA substrates labeled with 2AP in the position of the template strand (A) and highly fluorescent complexes with the primer-end bound in the polymerase active center for DNAs labeled at the +1 position in the template strand (B). Thus, exonucleolytic proofreading of the DNA substrate labeled initially with 2AP in the position in the template strand will produce an increase in fluorescence intensity due to formation of the highly fluorescent complexes with 2AP in the +1 position. The rate of increase in fluorescence intensity is a measure of the overall reaction; under single turnover conditions was 145 ± 3 s (A). We used this assay to confirm the results of Reddy () that the wild-type T4 DNA polymerase can catalyze a processive proofreading reaction without accessory proteins, but only for reactions that initiate in the exonuclease active center. Only a single proofreading rate of ∼145 s was detected in the presence of the heparin trap (A), but two rates of ∼106 and 11 s were detected in the absence of heparin (B). We () and others () proposed that the T4 DNA polymerase can form two types of complexes—[E-D] complexes that are active for hydrolysis of the terminal nucleotide and [E-D] complexes that are inactive for hydrolysis. [E-D] complexes react quickly with Mg to give a burst of product. Since the 145 and 106 s rates are in the range of the reported hydrolysis rate for the T4 DNA polymerase (,), these rates are likely produced from active [E-D] complexes. Under multiple turnover conditions, inactive [E-D] complexes can convert slowly to active [E-D] complexes, at the rate of ∼11 s in experiments reported here (B, ). This slow rate is not detected in the presence of the heparin trap, which indicates that conversion from an inactive to an active state involves enzyme dissociation. This point is discussed again later with respect to the clamped or tethered DNA polymerase. The 2AP fluorescence assay can also be used to determine the rates for active site switching. The rate of increase in fluorescence intensity for conversion of the moderately fluorescent exonuclease complexes with 2AP in the position to the highly fluorescent polymerase complexes with 2AP in the +1 position is a measure of the overall rate for the proofreading pathway that initiates in the exonuclease active center. The terminal phosphodiester bond of the primer strand is hydrolyzed in the exonuclease active center and then the trimmed primer-end is transferred from the exonuclease to the polymerase active center where the highly fluorescent +1 complexes are formed. These steps can be described by the following equation: 1/ = 1/145 = 1/ + 1/ + 1/. The combined rates for exonuclease-to-polymerase transfer of the trimmed primer-end and for formation of the highly fluorescent +1 complexes can be calculated if the hydrolysis rate is known. The hydrolysis rate catalyzed by the T4 DNA polymerase is reported to be ∼100 s for degradation of single-stranded DNA () and from 176 to 228 s () for removal of the 2AP nucleotide from the 3′-end of single-stranded DNA. Because the hydrolysis rate must be faster than the observed overall rate of 145 ± 3 sdetected in our experiments, the true hydrolysis rate is likely closer to ∼200 s, the average rate reported for removal of a terminal 2AP nucleotide (). The combined rates for exonuclease-to-polymerase switching and formation of the highly fluorescent complex with 2AP in the +1 position can be calculated from the following equation: [1/ + 1/] = 1/ −1/ = 1/145−1/200; therefore, [ + ] = 526 s (). Thus, once the hydrolysis reaction takes place, the trimmed primer-end is returned rapidly to the polymerase active center in position to resume nucleotide incorporation. The efficient proofreading reaction that initiates in the exonuclease active center has several implications for understanding proofreading by the T4 DNA polymerase and Family B DNA polymerases in general. First, the template strand is likely bound in the polymerase active center when the primer-end is bound in the exonuclease active center. Intuitively, it makes sense for the template strand to be held in the polymerase active center during proofreading to ensure that the trimmed primer-end will be returned to the polymerase active center in correct alignment, otherwise frameshift mutations will be produced. It is also important that proofreading be limited to only removing incorrect nucleotides in order to prevent gratuitous degradation of the newly synthesized DNA, which would slow DNA replication and waste dNTPs. Severely reduced DNA replication is observed in T4 infections with mutant DNA polymerases that catalyze excessive proofreading (,). These potential problems with proofreading can be reduced if the trimmed primer-end is returned to the polymerase active center in position to resume replication after an incorrect nucleotide is removed. If the primer-end is matched, primer extension will be the favored reaction; however, if the primer-end is not correct or if the primer-end is misaligned, then another cycle of proofreading will be favored over primer extension. Several observations are consistent with the proposal that the template strand is held in the polymerase active center when the primer-end is bound in the exonuclease active center. Moderate fluorescence enhancement is observed for 2AP in the and +1 positions in the template strand in exonuclease complexes (,,), which is the starting point of the fluorescence assay shown in A. The rapid transfer (>500 s) of the trimmed primer-end from the exonuclease to the polymerase active center to form the highly fluorescent +1 complexes (the end point of the fluorescence assay shown in A) is also consistent with the template strand being held in the polymerase active center since the +1 complexes are poised for rapid nucleotide incorporation (,). Furthermore, the ability of the T4 DNA polymerase to remove two incorrect nucleotides under single turnover conditions () suggests that the trimmed primer-end can be efficiently shuttled back-and-forth between the exonuclease and polymerase active centers, which can only reasonably occur if the template strand remains bound in the polymerase active center. How does the T4 DNA polymerase transfer the primer-end between the polymerase and exonuclease active centers? The apparent rapid rate for return of the trimmed primer-end to the polymerase active center—>500 s indicates that exonuclease-to-polymerase switching is rapid once the terminal phosphodiester bond is cleaved. Thus, the primer-end may ‘spring’ back to the polymerase active center unassisted once the terminal phosphodiester bond is cleaved. This proposal is supported by the observation that while a β hairpin structure in the exonuclease domain is important for forming stable exonuclease complexes (,), this structure is not needed for return of the trimmed primer-end to the polymerase active center (). However, deletion of the loop in the β hairpin structure reduces the ability of the RB69 DNA polymerase to remove two incorrect nucleotides (). Thus, the β hairpin may have a role in assisting further strand separation and reformation of exonuclease complexes for the second proofreading cycle. The overall rate for removing two incorrect nucleotides is ∼55 s (, ). The calculated overall rate for removing the first incorrect nucleotide is ∼88.5 s if removal of the second nucleotide occurs at the same rate as removal of a singly incorrect nucleotide—145 s. We propose that the trimmed primer-end is returned to the polymerase active center after removal of an incorrect nucleotide where the accuracy of the primer-end is evaluated based on the ability of the primer-end to form hydrogen bonds with the complementary template bases. Thus, the 88.5 s rate includes transfer of the primer-end back-and-forth between the exonuclease and polymerase active centers plus an intervening evaluation of the primer-end in the polymerase active center. The rate for these combined exo-to-pol/evaluation/pol-to-exo steps can be calculated from the following equation: 1/ = 1/ −1/ = 1/88.5–1/200 = 0.0063; thus, = 159 s (). Processive proofreading for reactions that initiate in the exonuclease active center appear to be limited to removal of two incorrect nucleotides because removal of three incorrect nucleotides is not reported to occur () or to take place less efficiently than removal of one or two incorrect nucleotides (). We conclude from this observation that the removal of a third incorrect nucleotide involves a heparin-sensitive step that is not present for removal of the first two incorrect nucleotides. This step is presumably slower than enzyme dissociation. What is the role of the clamp in proofreading? We could not detect any intrinsic processive proofreading for reactions that initiated in the polymerase active center (A), but proofreading is stimulated by the clamp protein, the product of T4 gene (,). Proofreading could be stimulated if the clamp allows intramolecular polymerase-to-exonuclease switching without dissociation of the DNA polymerase from the DNA. If this is the case, then tethering the DNA polymerase to the DNA allows strand separation and transfer of the primer-end from the polymerase to the exonuclease active center, even if the rate is slower than the rate for enzyme dissociation. Another possibility is that polymerase-to-exonuclease active site switching is intermolecular. Given the efficient ability of the T4 DNA polymerase to proofread mismatched DNAs by forming exonuclease complexes directly without first forming polymerase complexes [A and ; (,,)], the T4 DNA polymerase may normally dissociate from the DNA substrate when a wrong nucleotide is incorporated, even when clamped to the DNA, and then rebind to form exonuclease complexes with the mismatched DNA. The mismatched DNA may be rebound by the same or another DNA polymerase. The proposal of enzyme dissociation has merit because the concept of processivity in DNA replication has recently been redefined for the T4 DNA polymerase. Yang () demonstrated that T4 DNA polymerases exchange during DNA replication and that this exchange requires the clamp. Since the clamp can potentially bind the replicating DNA polymerase and a ‘spare’ DNA polymerase, incorporation of a wrong nucleotide may lead to dissociation of the replicating DNA polymerase and then the spare DNA polymerase forms an exonuclease complex with the mismatched DNA (,). The role of the clamp then is to provide a locally high concentration of spare DNA polymerases at replication forks for exonucleolytic proofreading. This proposal could explain why reduced concentrations of DNA polymerase δ in yeast produces a mutator phenotype (). If reduced concentrations of DNA polymerase δ means that there is not always a tethered spare for exonucleolytic proofreading, then proofreading would be reduced. This situation could also provide increased opportunity for replication by a translesion DNA polymerase such as DNA polymerase ζ, which lacks proofreading activity. This scenario also provides a possible mechanism to explain how DNA polymerase δ can proofread for DNA polymerase α () or for a translesion DNA polymerase to take over replication when DNA damage blocks replicative DNA polymerases (). One last point to consider is what happens with mutant DNA polymerases that have reduced ability to catalyze the exonuclease reaction. The exonuclease deficient D112A/E114A-DNA polymerase has almost no detectable ability to carry out removal of an incorrect terminal nucleotide or to extend the mismatched primer terminus under single-turnover conditions (C, lane 6); the W213S-DNA polymerase has only limited ability to do so (C, lane 4). The W213S-DNA polymerase slowly removed (1.5 s) the terminal dTMP nucleotide from the DNA substrate with 2AP in the position and this activity was observed only in the absence of the heparin trap (). Increased base substitution mutations are observed for both mutant DNA polymerases as expected if proofreading activity is reduced (). Mutant T4 DNA polymerases with reduced ability to catalyze the hydrolysis reaction also produce increased frameshift mutations, which is not observed to the same extent for other mutant DNA polymerases that are defective in proofreading but still retain significant hydrolysis activity (40, unpublished data). One intriguing question is what happens if the terminal nucleotide is removed, as is expected to be the case for the hydrolysis-defective DNA polymerases? Will the uncorrected primer-end be returned to the polymerase active center? If so, will the primer-end be correctly aligned? Since increased frameshift mutagenesis is observed for mutant DNA polymerases that are deficient in cleaving the terminal phosphodiester bond, strand misalignments may be a consequence of aberrant proofreading reactions. The proofreading pathway catalyzed by the bacteriophage T4 DNA polymerase is presented in . The experiments presented in this report begin with preformed exonuclease complexes, but association rates were determined in previous experiments and range from 70 to ∼120 s depending on the DNA sequence, which correspond to bimolecular association rates of 2−4 × 10 M s (,). In pathway I for removal of a single incorrect nucleotide, exonuclease complexes are formed directly in which the primer-end is bound in the exonuclease active center and the template strand is bound in the polymerase active center. Hydrolysis (200 s) and rapid transfer (>500 s) of the trimmed primer-end to the polymerase active center produces a DNA polymerase complex (identified by an asterisk *) that is poised to resume rapid nucleotide incorporation. For DNA substrates with two incorrect nucleotides at the primer-end (pathway II), exonuclease complexes are again formed. The terminal wrong (W) nucleotide is excised, then we propose that the trimmed primer-end is transferred to the polymerase active center (step a) as happens for removal of a single wrong nucleotide. However, since the primer-end still has a wrong nucleotide, the incorrect primer-end is returned to the exonuclease active center (step b) for a second cycle of excision. The overall rate for steps a + b is calculated to be 159 s (). The second wrong nucleotide is then removed as described for pathway I. Intrinsic processive proofreading was not detected for reactions that initiate in the polymerase active center, pathway III. We observed a rate of 11 s in multiple turnover reactions for the DNA substrate used in experiments reported here (), which involves dissociation of the DNA polymerase from a complex that is inactive for the excision reaction and then formation of an active exonuclease complex. Although the apparent rate for initiating the proofreading reaction in the polymerase active center is slow, 11 s, this rate is still much faster than the rate for extending a mismatched primer terminus (), but is slower than the rate for extension of a matched primer terminus (,). Thus, the 11 s rate is a barrier to gratuitous proofreading, but is fast enough to prevent extension of a mismatched primer terminus. We propose that the presence of the clamp will not affect the 11 s rate as the clamp is thought to act only as a tether, but the clamp will allow either the same DNA polymerase that incorporated the incorrect nucleotide or a spare DNA polymerase that is co-tethered to form exonuclease complexes with the mismatched DNA.

RNA·DNA hybrid formation is most often associated with DNA replication, where a role for the hybrid as the primer for DNA synthesis is well known (). Other uses are less clear, and cells possess a host of activities to prevent or remove the hybrids, suggesting that RNA·DNA hybrids may not be generally beneficial. For instance, even simple RNA polymerases such as T7 RNAP separate the template from the nascent RNA, which threads out of the polymerase through a positively charged exit pore (,). In addition, multi-subunit RNA polymerases such as that of have a ‘lid’ element that functions to separate RNA from the RNA·DNA hybrid in the active site, guiding it to the exit pore and inhibiting the formation of long RNA·DNA hybrids even on single-stranded templates (,). RNase H activities that specifically digest the RNA in an RNA·DNA hybrid are ubiquitous in both prokaryotes and eukaryotes (). Also widely expressed are topoisomerases, which can remove the negative supercoils that favor RNA·DNA hybrid formation (). Topoisomerase activity may be particularly crucial to reducing RNA·DNA hybrid formation because a domain of negative supercoiling is generated behind a transcribing RNA polymerase (,). Ribosomes coat the nascent transcript during co-transcriptional translation in bacteria and a faster RNA polymerase such as T7 will outrun the ribosomes leaving the transcript open to RNase digestion (). Bound ribosomes may also discourage re-annealing to the template as translation inhibitors can increase RNA·DNA hybrid formation in bacteria (). In eukaryotes, several different co-transcriptional systems may inhibit re-annealing by coating the nascent transcript as part of their functions in mRNP maturation. Furthermore, chromatin would further limit the accessibility of template DNA to RNA after transcription is completed. In yeast, the THO/TREX system couples transcription elongation to mRNA export (). The THO complex associates with sites of active transcription and recruits mRNA export proteins such as Sub2 and Yra1 to the nascent transcript (,). Many components of the THO/TREX complex are conserved between yeast and higher eukaryotes, but the order of loading is not. Unlike yeast, where the THO complex associated with the transcribing polymerase recruits splicing factors (), the metazoan THO/TREX complex is itself recruited to mRNA during splicing by serine-arginine-rich (SR) splicing factors (). Breakdowns in co-transcriptional RNA processing due to mutations affecting THO/TREX can lead to RNA·DNA hybrid formation, impaired transcription elongation and increased genomic instability in yeast (). Depletion of an SR protein called alternative splicing factor/splicing factor 2 (ASF/SF2) in the avian B-cell line DT40 will also cause RNA·DNA hybrid formation and genome instability (). In both cases, RNA·DNA hybrids link impaired co-transcriptional RNA–protein interactions to genomic instability (,). Some sequences have an intrinsic propensity to form RNA·DNA hybrids that may overcome co-transcriptional mechanisms repressing hybrid formation. For example, immunoglobulin class switch regions have long been known to form RNA·DNA hybrids when transcribed . Notably, the hybrids only form if the sequences are transcribed in the physiological direction to make a purine-enriched transcript (,). The most common inherited ataxia, Friedreich ataxia (FRDA), is caused by expansion of an unstable GAA·TTC trinucleotide repeat within the first intron of the frataxin gene () (). There is a direct relationship between the length of the expanded repeats and disease severity (,), reflecting reduced frataxin expression. However, the causes of GAA·TTC repeat instability are poorly understood and the means by which expanded repeats suppress gene expression are controversial (). Most FRDA patients have intact frataxin-coding sequences, so a refined understanding of factors contributing to the transcript deficit and the repeat instability will aid in the design of therapies. We have previously demonstrated that a long GAA·TTC tract forms a structural impediment to T7 RNA polymerase (,). Here we show that GAA·TTC tracts have an intrinsic length-dependent propensity towards RNA·DNA hybrid formation during transcription, both and in living bacteria cells. Furthermore, we show that an extensive RNA·DNA hybrid is tightly linked to T7 RNA polymerase arrest on templates that contain long GAA·TTC repeats. Uninterrupted (CAG·CTG) and (GAA·TTC) repeats were made by defined stepwise expansion of smaller units cloned into a plasmid (pREX) designed for that purpose. Asymmetric type IIS restriction digestion, fragment purification and ligation to achieve that expansion has been described (). Plasmids bearing a self-cleaving ribozyme 3′ to (CAG·CTG) and (GAA·TTC) repeats have been described (). An SpeI–XbaI fragment was excised from pREX(GAA) and inserted into pBAD18 to make pBAD18(GAA). Plasmids were grown, and experiments were carried out in bacterial strain XL-1 Blue (Stratagene). Vectors for bacterial transcription were under control of the arabinose inducible BAD promoter (). Bacterial strain XL-1 Blue (Stratagene) containing plasmids with zero repeats (pBAD18) or 88 GAA·TTC repeats (pBAD18GAA88) were grown in parallel to mid-log phase. Glucose or arabinose was added to 0.2% for 20 min. Cells were collected by centrifugation at 37°C in pre-warmed holders. RNA and plasmid were rapidly isolated using the GTC-acid phenol method (). Samples were treated with the single-strand specific RNases A and T1 or with the combination of RNases A&T1 and RNase H for 1 h at 32°C. Samples were resolved on a 1% agarose gel and stained with ethidium bromide after electrophoresis. RNA transcription from phage promoters was performed in T7 transcription buffer (50 mM HEPES, pH 8.0, 100 mM NaCl, 20 mM MgCl, 10 mM DTT and 0.5 mM each NTP) supplemented with 200 U/ml RNase Inhibitor (Ambion) in a final volume of 20 μl at 37°C for 20 min. T7 RNA polymerase (Ambion) was used at a final concentration of 1000 U/ml. Template concentrations (usually ∼200 ng/reaction) were estimated by ethidium bromide fluorescence. The 5′ end of the transcript was labeled by including [γ-P]-GTP (6000 Ci/mmol). The gamma phosphate is retained only in the first position. RNase A and T1 (pre-mixed, Ambion) were used at a concentration of 20 μg/ml and 1000 U/ml, respectively, for 1 h at 37°C in TE (10 mM Tris HCl, pH 8.0, 1 mM EDTA) unless otherwise noted. RNase H (US Biochemical Corp.) digestion was performed in the buffer supplied by the manufacturer in the experiments where digestion followed transcription. In most cases, RNase H digestion was performed simultaneously with transcription in the T7 transcription buffer. To make an RNA marker ladder, RNase T1 (Ambion) was titrated on end-labeled transcripts containing (GAA) to obtain an appropriate partial digest. An oligodeoxyribonucleotide with the sequence (5′-TGGACGAGTCTCGAGCAGCTGAAGCTTGCA-3′) that anneals from 71 to 40 bases 3′ of the end of the TRS in transcripts produced by T7 RNAP was synthesized using standard phosphoramidite chemistry (Life Technologies). The oligonucleotide (2 pmol) was end-labeled with [γP]GTP (6000 Ci/mmol, PerkinElmer) using T4 polynucleotide kinase (New England Biolabs) in the supplied buffer. RNA was prepared for primer extension by performing transcription reactions with T7 RNAP on a supercoiled template containing (GAA·TTC) in the presence (2.0 U) or absence of RNase H for 1 h at 37°C in a volume of 25 μl. The reactions were stopped by adding 200 μl stop buffer (5 mM EDTA, 0.5% SDS, 0.2 mg/ml Proteinase K and 10 μg/ml tRNA) followed by incubation at 65°C for 20 min. The RNA was then extracted with 200 μl phenol:chloroform:isoamyl alcohol (25:24:1), followed by ethanol precipitation. Primer extension was performed by taking aliquots of the purified RNA resuspended in HO (10 μl) and adding 2 μl of dilute kinased oligonucleotide (∼0.1 pmol) followed by heating to 70°C for 10 min and then chilled on ice. The Superscript II, RNase H-reverse transcriptase first strand synthesis kit (Life Technologies) was then used following the manufacturers’ protocol. Agarose gel electrophoresis was routinely performed in 1% agarose gels in TAE buffer (40 mM Tris acetate, 2 mM EDTA at pH 8.0). Gels were stained with ethidium bromide after electrophoresis. Radiolabeled reactions were stopped by mixing with 50 μl of stop buffer (98% deionized formamide, 10 mM EDTA and 10 μg/ml tRNA). The samples were ethanol precipitated, resuspended in a denaturing gel loading buffer (98% formamide, 10 mM EDTA and 0.05% bromophenol blue and xylene cyanole) heated to 90°C, and loaded on a pre-warmed 6% polyacrylamide sequencing gel containing 8 M urea. End-labeled pBR322 MspI digest (New England Biolabs) and 10 bp ladder (Life Technologies) were used as size markers. Images of radioactive gels were obtained using FujiFilm type BAS III-S phosphorimaging screens and a FujiFilm BAS 1500 reader. Analysis and quantitation was performed with FujiFilm Image Gauge 3.0 (Mac) software. To investigate how expanded GAA·TTC tracts reduce gene expression we constructed a series of plasmids containing various lengths of the repeat and carried out transcription reactions on these templates . We noted that supercoiled plasmid templates with moderately long GAA·TTC tracts exhibit a more relaxed mobility in agarose gels after transcription if the product transcript contained a GAA tract. For instance, in A, the native gel mobility of supercoiled plasmids carrying 0, 11 or 44 GAA·TTC repeats are shown in the first three lanes. Following transcription by phage T7 RNA polymerase (A, last three lanes), the templates are partially obscured by transcripts, but it is clear that the template with 44 triplet repeats has shifted to a more relaxed mobility (gray arrowhead) rather than supercoiled mobility (black arrow). Treatment of the transcribed samples with RNase H, which specifically degrades the RNA in an RNA·DNA hybrid, returns the transcribed (GAA·TTC) template to supercoiled mobility (B, lane 3) indicating that the source of the mobility change is an RNA·DNA hybrid and not a nick in the template. Conversely, treatment with the single-strand-specific RNases A and T1 after transcription (B, lanes 4–6) does not return the (GAA·TTC) template to supercoiled mobility, but reveals the appearance of a series of partially relaxed conformers (small arrows). Unwinding of the strand displaced by the R-loop relaxes negative supercoils, and the degree of relaxation reflects the length of the RNA·DNA hybrid (shown schematically to the right in B). Different lengths of hybrid remain after single-strand-specific digestion partly because random branch migration can make single-stranded RNA ends available to the nucleases. The examples shown were treated with RNases A and T1 in low salt (TE), which provides a greater range of conformers than treatment in high salt which provides greater RNase protection (data not shown). The templates with 0 or 11 GAA·TTC repeats migrate with mobility expected of supercoiled templates after all treatments. The samples were extracted with phenol prior to electrophoresis, so the altered mobility of the (GAA·TTC) template is unlikely to represent preferential binding of T7 RNA polymerase or RNase A or T1 to the repeat tract. Moreover, neither transcription of these same templates in the opposite direction by SP6 RNA polymerase, nor transcription of templates containing GAA·TTC repeats cloned in the opposite direction by T7 RNA polymerase produced RNA·DNA hybrids (data not shown). The appearance of relaxed conformers that are resistant to single-strand-specific RNases, but resolved by RNase H indicates the presence of an rGAA·dTTC hybrid. We have shown that stable RNA·DNA hybrids form readily during transcription of GAA·TTC repeats and in bacteria. The data indicate a strong propensity for RNA·DNA hybrid formation in the GAA·TTC repeat, link the hybrid to RNA polymerase arrest, and suggest that hybrid formation during transcription is intrinsic to long GAA·TTC repeats. Other sequences that are purine-rich (especially G-rich) on the non-template strand have also been shown to have an intrinsic propensity to form RNA·DNA hybrids. For example, the non-template strands of the immunoglobulin class switch regions have a general purine bias featuring repeated sequences that are 60–70% purine, as well as regions of relatively uninterrupted purine runs (). The immunoglobulin class switch regions have long been known to form RNA·DNA hybrids when transcribed . Moreover, RNA·DNA hybrids form only if the switch regions are transcribed in the physiological direction, suggesting a role for the hybrids in class switch recombination (,). Bisulfite modification of single-stranded cytosines displaced in the R-loop has been used to verify that RNA·DNA hybrids do indeed form during switch region transcription in activated B-cells (). Thus, the propensity that the immunoglobulin class switch regions exhibit for RNA·DNA hybrid formation manages to overcome mechanisms that repress hybrid formation in the cell. While we would like to probe for RNA·DNA hybrid formation in the native GAA·TTC repeat using bisulfite modification, it is not possible due to the lack of cytosines on the non-template strand. Moreover, while RNA·DNA hybrids have been found to extend to some degree both upstream and downstream of the purine-rich immunoglobulin switch region repeats (), we have found that the ends of the GAA·TTC repeat act as a boundary to the hybrid. Consequently, bisulfite modification of DNA flanking the repeat is not likely to be informative. RNA·DNA hybrid formation adds to the already extensive repertoire of non-B DNA structures that can be adopted by GAA·TTC repeats. Like other uninterrupted purine·pyrimidine (R·Y) sequences, the repeat can form intermolecular and intramolecular DNA triple helices of either the R·R·Y or Y·R·Y configuration (). An attractive hypothesis for RNA·DNA hybrid formation is a transient single-stranded state for the template strand when the non-template strand is engaged as the third strand in a triplex (,,). Our working model, shown in schematic form in , assumes that the transcript exits RNAP through the RNA exit pore in the normal way. We posit that negative supercoiling behind the advancing polymerase prompts formation of a short-lived intramolecular R·R·Y triplex in the GAA·TTC tract. This leaves part of the TTC template strand temporarily single stranded. The transcript would then simply re-anneal to the template strand after exiting RNA polymerase, with lessened competition from the non-template strand. Once annealed to the single-stranded region, the transcript can further displace the non-template strand, as shown in panel D of . We have no evidence that the GAA transcript acts as a third strand in a triplex, nor do we think that it invades the duplex DNA independent of the transcribing polymerase. Consequently, while chromatin should inhibit access to the DNA by an exogenous complementary RNA, co-transcriptional RNA·DNA hybrid formation may bypass such inhibition because RNA polymerase has locally dislodged the histones from the template. An alternative model for reduced non-template competition suggested for immunoglobulin RNA·DNA hybrid formation is formation of G-quartet structures by the G-rich non-template strand (). However, G-quartets are unlikely to contribute to hybrid formation in the GAA·TTC repeat. Occupation of the non-template strand by hairpin structures might also be expected to enhance hybrid formation, and CAG·CTG repeats readily form hairpins (). We found that transcription through supercoiled CAG·CTG templates did not lead to extensive RNA·DNA hybrid formation. Thus, RNA·DNA hybrid formation in the FRDA repeat is not simply the consequence of potential structure formation by the non-template strand or the enhanced ability of simple repeat transcripts to re-anneal to the template at almost any point of contact. RNA·DNA hybrid formation is likely to be a combination of several factors. Structure formation by the non-template strand may increase opportunities for the transcript to anneal. The lack of complexity in the repetitive sequence makes annealing simple. Indeed, it is possible that misaligned pairing of the rGAA transcript to the dTTC template contributes to the lack of branch migration beyond the confines of the repeat tract. Finally, the enhanced hybrid strength of rG·dC pairings and the absence of the weaker, disrupting rU·dA hybrids contribute to the stability of the rGAA·dTTC hybrid (). The importance of a purine-rich transcript to hybrid stability is clear when considering transcription of the GAA·TTC repeat in the non-physiological direction. The GAA·TTC repeat has been shown by multiple groups to adopt a stable intramolecular Y·R·Y triplex that leaves a portion of the GAA strand unpaired (,,). When transcribed to make CUU, the GAA·TTC repeat also forms an intramolecular acid stabilized Y·R·Y triplex capable of blocking transcription elongation (,). However, despite the single-stranded length of GAA template strand that is a consequence of an intramolecular triplex, we have found that the CUU transcript does not form stable, or even detectable, RNA·DNA hybrids. This is most likely due to the U-rich nature of the CUU transcript. The rU·dA base pairs are weaker than dT·dA base pairs because uracil lacks the C-5 methyl group of thymine, which is thought to contribute to base-stacking interactions (). In the physiological direction of transcription, we have shown that an extensive RNA·DNA hybrid is tightly linked to T7 RNA polymerase arrest in a TTC repeat template. In the nucleus, RNP formation should generally help block formation of RNA·DNA hybrids. However, in FRDA the GAA repeat within an FXN transcript numbers hundreds to thousands and it is possible that a protein with some specificity for rGAA is simply insufficient to bind it all. Huertas and Aguilera found that the selective transcription elongation impairment seen in mutants of the yeast THO/TREX complex () was due to RNA·DNA hybrid formation (). Moreover, transcription elongation was improved in these yeast mutants if RNase H was over expressed, leading those authors to suggest that impaired elongation was likely due to the polymerase being tethered to an RNA·DNA hybrid (). The mechanism by which GAA·TTC expansions may impede transcription elongation in human cells is still unclear. Most models predict an obstruction to transcription elongation within the repeat, mediated either by DNA structure (), or by altered chromatin (,). Our working model for transcription inhibition differs from most in that it predicts arrest downstream of a structure that includes an RNA·DNA hybrid (,). The results presented in lend further support for that model, and link the RNA·DNA hybrid to transcription arrest . On the other hand, experiments like that shown in suggest that transcription arrest is stochastic, as some transcripts extending beyond the repeat had formed a hybrid within the repeat. Large GAA·TTC expansions in FRDA are unstable between generations (,) and show somatic mosaic expansion in individuals () implying a high degree of instability. While most long trinucleotide repeats are unstable to some degree, with mismatch repair implicated in their continued expansion with age (,), structure formation by GAA·TTC repeats may provide additional pathways to instability. We have shown that RNA·DNA hybrids form even on ‘pre-mutation’ size GAA·TTC repeats of about 40 triplets. This length does not cause FRDA disease symptoms, but is near the threshold of instability in both prokaryotic and eukaryotic systems. Given the recent linking of RNA·DNA hybrid formation to genome instability (), it is plausible that RNA·DNA hybrid formation contributes to GAA·TTC repeat instability in the cell.