Available online at www.sciencedirect.com

Neuropharmacology 53 (2007) 942e950

www.elsevier.com/locate/neuropharm

Effects of fetal exposure to isoflurane on postnatal memory and learning in rats

Yujuan Li a,b, Ge Liang a, Shouping Wang a,b, Qingcheng Meng a, Qiujun Wang a,c, Huafeng Wei a,*

a Department of Anesthesiology and Critical Care, University of Pennsylvania, 305 John Morgan Building, 3620 Hamilton Walk, Philadelphia, PA 19104, USA b Department of Anesthesia, Second Affiliated Hospital of Sun Yat-Sen University, Guangzhou 510120, China c Department of Anesthesiology, The Third Clinical Hospital, Hebei Medical University, Shijiazhuang 050051, China

Received 9 March 2007; received in revised form 9 August 2007; accepted 10 September 2007

Abstract

In a maternal fetal rat model, we investigated the behavioral and neurotoxic effects of fetal exposure to isoflurane. Pregnant rats at gestational day 21 were anesthetized with 1.3% isoflurane for 6 h. Apoptosis was quantified in the hippocampus and cortex at 2 and 18 h after exposure in the fetal brain and in the postnatal day 5 (P5) pup brain. Spatial memory and learning of the fetal exposed pups were examined with the Morris Water Maze at juvenile and adult ages. Rat fetal exposure to isoflurane at pregnancy day 21 through maternal anesthesia significantly decreased spontaneous apoptosis in the hippocampal CA1 region and in the retrosplenial cortex at 2 h after exposure, but not at 18 h or at P5. Fetal ex- posure to isoflurane did not impair subsequent juvenile or adult postnatal spatial reference memory and learning and, in fact, improved spatial memory in the juvenile rat. These results show that isoflurane exposure during late pregnancy is not neurotoxic to the fetal brain and does not impair memory and learning in the juvenile or adult rat. (cid:2) 2007 Elsevier Ltd. All rights reserved.

Keywords: Anesthetics; Fetus; Developing brain; Apoptosis; Memory; Learning

1. Introduction

Anesthetics cause neurotoxicity in a concentration and time dependent manner in in vitro neuronal models (Chang and Chou, 2001; Eckenhoff et al., 2004; Kim et al., 2006; Kvolik et al., 2005; Loop et al., 2005; Matsuoka et al., 2001; Wei et al., 2005; Wise-Faberowski et al., 2005; Xie et al., 2006, 2007). Relatively fewer studies have investigated the neuro- toxic effects of anesthetics in in vivo models. Isoflurane expo- sure at a clinically relevant concentration (0.75%) for 6 h during postnatal development in rats caused persistent memory and learning deficits, which was associated with widespread neuronal apoptosis (Jevtovic-Todorovic et al., 2003; Yon

Corresponding author. Tel.: þ1 215 662 3193; fax: þ1 215 349 5078. E-mail address: weih@uphs.upenn.edu (H. Wei).

et al., 2005). Neurons in the developing brain are specifically vulnerable to isoflurane neurotoxicity (Jevtovic-Todorovic et al., 2003). However, the mechanisms for isoflurane neuro- toxicity are unknown.

Since anesthetics easily cross the placenta, the developing fetal brain will be exposed to inhaled anesthetics, such as iso- flurane, when pregnant women require surgery. In some cases, such as fetal surgery to correct various congenital malforma- tions during mid-gestation (18e25 weeks) (Myers et al., 2002), the fetal brain can be exposed to 2e3 times (2.5e3 minimal alveolar concentration (MAC)) higher than normal concentrations of inhaled anesthetics, in order to relax uterine smooth muscle and provide adequate anesthesia (Cauldwell, 2002; Myers et al., 2002). Although fetal surgery is relatively new, it is a rapidly growing and evolving area, and may be- come standard therapy for most disabling malformations that

0028-3908/$ - see front matter (cid:2) 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropharm.2007.09.005

Y. Li et al. / Neuropharmacology 53 (2007) 942e950

are currently treated in young infants (Goldsmith et al., 1999; Myers et al., 2002). Because most fetal surgeries in humans are performed during mid-gestation, it is important and urgent to know if the anesthetics used cause damage to the develop- ing brain and subsequent postnatal memory problems and learning disabilities.

The aim of the current study was to determine whether exposure to clinically relevant concentrations of isoflurane during prenatal development causes neuronal apoptosis and postnatal learning and memory deficits.

2. Materials and methods

2.1. Animals

Institute of Animal Care and Use Committee (IACUC) at the University of Pennsylvania approved all experimental procedures and protocols used in this study. All efforts were made to minimize the number of animals used and their suffering. SpragueeDawley pregnant rats (Charles River Laboratories, Inc Wilmington, MA) were housed in polypropylene cages and the room temper- ature was maintained at 22 (cid:2)C, with a 12 h lightedark cycle. Pregnant rats at gestation day 21 (E21) were used for all experiments because it approximately corresponds to mid-gestation in human beings according to the theory of brain growth spurt (Dobbing and Sands, 1979; Jevtovic-Todorovic et al., 2003), and is a common time for most fetal surgeries (18e25 weeks) (Myers et al., 2002). We have designed the following three related studies: (1) pilot study; (2) neurodegeneration study; (3) finally a behavioral study. A pilot study was first conducted to find the highest concentration of isoflurane not accompanied by significant arterial blood gas (ABG) and mean arterial blood pressure (MABP) changes in the mothers. A neurodegeneration study was used to determine the appearance of apoptosis by detection of caspase-3 and TUNEL (terminal de- oxynucleotidyl transferase biotin-dUTP nick end labeling) positive cells in the fetal brain (2 and 18 h post-exposure) or neonatal brain at postnatal day 5 (P5). We have chosen the above time points to detect apoptosis in fetal or newborn brains, based on previously published work (Jevtovic-Todorovic et al., 2003). The behavioral study was performed to investigate the effects of fetal exposure to isoflurane on postnatal memory and learning. The pregnant rats used in each study were not reused in the other two studies. Within each study described above, animals were randomly divided into either isoflurane treatment or sham control groups. Pregnant rats in the isoflurane treatment groups inhaled isoflurane for 6 h, while those in the sham control group only inhaled a carrier gas (30% oxygen, balanced with nitrogen) for 6 h under the same experimental

conditions. The distribution of pregnant rats and pups in all three groups is illustrated in Fig. 1.

2.2. Anesthetic exposure

Isoflurane is used clinically at a wide range of concentrations (about 0.2e 3%), depending on the presence of other kinds of anesthetics or narcotics and the type and duration of surgery. As isoflurane neurotoxicity is concentration- dependent (Jevtovic-Todorovic et al., 2003; Wei et al., 2005), a primary goal of this study was to investigate if the highest isoflurane concentration used clin- ically is harmful to the fetal brain. Due to our concern that the physiological side effects of these drugs would contaminate the interpretation, we conducted a pilot study to determine the highest anesthetic concentration we could use without invasive support (tracheal intubation and ventilation) that would not significantly affect arterial blood gas (ABG) and mean arterial blood pressure (MABP) in the mothers, and then used this concentration in the subsequent formal study. We wanted to avoid tracheal intubation, as it could possibly af- fect the hemodynamics of pregnant rats and the apoptosis in the fetal brains. In addition, this makes it more difficult to set up the sham control groups without anesthesia. In the pilot study, five pregnant rats were initially anesthetized with 2% isoflurane in 30% oxygen via a snout cone for approximately 1 h and the right femoral artery was catheterized for blood sample collection and measure- ment of MABP by a pressure transducer/amplifier (AD Instruments Inc., Colorado Springs, CO, USA). The rats were recovered for 2 h and then ex- posed to isoflurane, starting at 1.5% in a humidified carrier gas of 30% oxygen, balance nitrogen for 6 h in a monitored chamber in hood. The pregnant rats breathed spontaneously without intubation or other support while being warmed using a deltaphase isothermal pad (Braintree Scientific Inc, Braintree, MA, USA). The rectal temperature was maintained (Fisher Scientific, Pitts- burgh, PA, USA) at 37 (cid:3) 0.5 (cid:2)C. We monitored isoflurane concentration in the chamber using IR absorbance (Ohmeda 5330, Detex-Ohmeda, Louisville, CO, USA). Arterial blood (0.1 ml) from previously placed femoral arterial catheter was collected and ABG determined every 2 h for up to 6 h by an ABG analyzer (Nova Biomedical, Waltham, MA, USA). Blood glucose was simultaneously measured with a glucometer (ACCU-CHECK Advantage, Roche Diagnostics Corporations, Indianapolis, IN, USA). Control rats were exposed only to humidified 30% O2 balanced by N2 (carrier gas for isoflurane in the treatment group) for 6 h in the same chamber under the same experi- mental conditions as in the treatment group. Because one pregnant rat treated with 1.5% isoflurane showed obvious acidemia (which reversed after termina- tion of anesthesia), we decreased the isoflurane concentration to 1.3%, and subsequently found no significant changes in the ABG or MABP between the treatment group and the sham control group (Table 1). Therefore, 1.3% iso- flurane was used in the ensuing neurodegeneration and behavioral studies.

Total Pregnant Rats (44)

Pilot Study (8)

Neurodegeneration Study (21)

Behavioral Study (15)

Control (4) 1.3% Isoflurane (4)

Control (10)

1.3% Isoflurane (11)

Control (7)

1.3% Isoflurane (8)

2 Hr (Mother 5) (Fetus 12)

18 Hr (Mother 5) (Fetus 14)

2 Hr (Mother 5) (Fetus 13)

18 Hr (Mother 6) (Fetus 13)

Behavioral study Pups (26)

Neurodeg- eneration study P5 (7)

Behavioral study Pups (31)

Neurodeg- eneration study P5 (8)

Fig. 1. Nomogram illustrating distribution of pregnant mother and their fetus or postnatal rats among different studies.

943

944

Y. Li et al. / Neuropharmacology 53 (2007) 942e950

In the behavioral study, pregnant rats were treated with 1.3% isoflurane (n ¼ 8) or carrier gas (sham controls, n ¼ 7) for 6 h. The monitoring was the same as that in the pilot study except that femoral artery catheters were not placed. After the exposures, the animals were returned to their cages and the rat pups were delivered naturally. Four rat pups from each pregnant mother were raised to P28 (Juvenile) and P118 (adult), and then used to deter- mine memory and learning ability with a Morris Water Maze (MWM). Two rat pups from the control group and one from the isoflurane group died unexpect- edly, leaving a total of 26 and 31 rat pups in the control and isoflurane treat- ment groups respectively (Fig. 1).

corresponding to figure 96 of the rat fetal brain atlas (Paxinos et al., 1990) were chosen for detection of apoptosis by caspase-3 immunohistochemistry and TUNEL staining. In the initial examination of brains sections from the neurodegeneration study, we noticed that apoptosis was most apparent in the hippocampus CA1 region and the retrosplenial cortex, and thus we chose these two brain regions to quantify apoptosis.

2.5. Immunohistochemistry for caspase-3

In the fetal brain apoptosis study, pregnant rats were treated with either 1.3% isoflurane or carrier gas for 6 h. At 2 and 18 h after exposure, the rat pups were delivered by C-section under sodium pentobarbital (100 mg/kg, i.p.) anesthesia. The fetal brains were removed and snap frozen for immuno- histochemical analysis. Two fetal brains from each pregnant rat were studied. In addition, newborn brains from the rat pups born to the pregnant rats in the behavioral study group (one pup from each pregnant rat, treatment n ¼ 8, con- trol n ¼ 7) were also obtained at postnatal day 5 and prepared for the apoptosis study at P5 (Fig. 1).

2.3. Measurement of isoflurane concentration in the brain tissues

To confirm that the isoflurane concentration in the fetal brain correlated with the inhaled concentration and brain concentration in the pregnant mothers, we measured the brain isoflurane concentrations in the fetus and the mother simultaneously in one rat. Briefly, after the pregnant rat was ex- posed to 1.3% isoflurane for 6 h, the brains of both mother and fetuses were removed and the brain tissue was immediately placed into 4 ml of 0.02 M phosphate buffer (with 1 mM halothane as internal standard) and homogenized in a glass homogenizer. The homogenate was centrifuged (30,000 (cid:4) g at 4 (cid:2)C for 30 min), the supernatant collected and then loaded onto C18 cartridge that had been conditioned with 2 ml methanol and washed with water, for solid phase extraction. The final sample was eluted with 0.5 ml solution of methanol and 2-propanol (vol:vol 2:1) with 0.1% trifluoroacetic acid. All procedures were performed in the cool room (4 (cid:2)C). A 250 ml aliquot of each final elute was injected into a high performance liquid chromatography (HPLC) system, equipped with a refractive index monitor, for quantitation.

Caspase-3 positive cells were detected using immunohistochemical methods described previously (Gown and Willingham, 2002). Briefly, brain incubated in 3% hydrogen peroxide in methanol for sections were first 20 min to quench endogenous peroxidase activity. Sections were then incu- bated with blocking solution containing 10% normal goat serum in 0.1% phos- phate buffered saline with 0.1% Tween 20 (PBST) for 1 h at room temperature after washing with 0.1% PBST. The anti-activated caspase-3 primary antibody (1/200, Cell Signaling Technology, Inc Danvers, MA, USA) was then applied in blocking solution and incubated at 4 (cid:2)C overnight. Tissue sections were bio- tinylated with goat anti-rabbit antibody (1/200, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) in 0.1% PBST for 40 min, followed by incubation with the avidinebiotinylated peroxidase complex (Vectostain ABC-Kit, Vector Lab, Burlingame, CA, USA) for 40 min. Tissue sections were colorized with diaminobenzidine (DAB, Vector Laboratories, Burlingame, CA, USA) for 8 min and counterstained with modified hematoxylin. Negative control sec- tions were incubated in blocking solution that did not contain primary anti- body. Images were acquired and assessed at 200(cid:4) using IP lab 7.0 software linked to an Olympus IX70 microscope (Olympus Corporation, Japan) equip- ped with a Cooke SensiCam camera (Cooke Corporation, Romulus, MI, USA). Three brain tissue sections at 10 mm corresponding to the Atlas of the Devel- oping Rat Brain, Figure 96 (Paxinos et al., 1990) were chosen from each an- imal and analyzed for caspase-3 positive cells in the two brain regions. Two persons blinded to the treatments counted the total number of caspase-3 pos- itive cells in the hippocampal CA1 region and retrosplenial cortex. The areas of entire hippocampal CA1 region and retrosplenial cortex were defined ac- cording to the Atlas of the Developing Rat Brain, Figure 96 (Paxinos et al., 1990) and the area measured using IPLab Suite v3.7 imaging processing and analysis software (Biovision Technologies, Exton, PA, http://www.Bio- Vis.com). The density of caspase-3 positive cells in a particular brain region was calculated by dividing the number of caspase-3 positive cells by the area of that brain region.

2.4. Tissue preparation

After treatment with isoflurane or carrier gas alone (control), pregnant rats were anesthetized with sodium pentobarbital intraperitoneally (i.p. 100 mg/kg) at either 2 or 18 h after the end of isoflurane exposure, and the fetuses removed by cesarean section. Likewise, postnatal pups at day 5 (P5) were given the same dose of sodium pentobarbital. All fetuses and pups were then perfused transcardially with ice-cold normal saline followed by 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). The brains were then removed and post-fixed overnight in the same fixative at 4 (cid:2)C, and cryoprotected in 30% (wt/vol) sucrose in 0.1 M phosphate buffer (pH 7.4) at 4 (cid:2)C for 24 h. Thereaf- ter, the brains were frozen in isopentane at (cid:5)20 (cid:2)C and stored at (cid:5)80 (cid:2)C until use. Serial coronal sections (10 mm) were cut in a cryostat (DolbeyeJamison Optical Company, Inc., Pottstown, PA, USA), mounted on gelatin-coated slides and stored at (cid:5)80 (cid:2)C. Coronal brain sections from the same brain

2.6. TUNEL for DNA fragmentation

Three brain sections (10 mm) adjacent to the sections used for caspase-3 detection were used for TUNEL staining using the DeadEnd(cid:3) Colorimetric TUNEL System Kit (Promega Corporation, Madison, WI, USA) according to the manufacturer’s protocol (Gavrieli et al., 1992). Briefly, sections were permeabilized by proteinase K solution (20 mg/ml) for 8 min, incubated in equilibration buffer for 10 min and the terminal deoxynucleotidyl transferase (TdT) and biotinylated nucleotide were added to the section and incubated in a humidified chamber at 37 (cid:2)C for 1 h. The reaction was then stopped, fol- lowed by incubation with horseradish peroxidase-labeled streptavidin, colori- zation with DAB/ H2O2 and counterstained with modified hematoxylin. For treated with DNase I positive-controls,

the tissue sections were first

Table 1 Isoflurane (1.3%) did not affect arterial blood pressure and blood gas significantly

Baseline

2 h

4 h

6 h

Control

1.3% Iso

Control

1.3% Iso

Control

1.3% Iso

Control

1.3% Iso

pH PaCO2 PaO2 (cid:5) HCO3 MABP

7.46 (cid:3) 0.02 36.7 (cid:3) 2.80 158 (cid:3) 6.98 25.6 (cid:3) 2.38 117 (cid:3) 7.42

7.46 (cid:3) 0.03 32.8 (cid:3) 0.64 162 (cid:3) 4.67 23.3 (cid:3) 1.57 119 (cid:3) 0.56

7.47 (cid:3) 0.01 33.4 (cid:3) 1.39 149 (cid:3) 9.08 24.4 (cid:3) 1.15 106 (cid:3) 9.61

7.44 (cid:3) 0.02 37.2 (cid:3) 4.33 159 (cid:3) 2.12 25.5 (cid:3) 2.02 90 (cid:3) 6.67

7.45 (cid:3) 0.01 35.8 (cid:3) 3.43 166 (cid:3) 5.45 25.2 (cid:3) 2.84 102 (cid:3) 4.06

7.39 (cid:3) 0.01 39.9 (cid:3) 4.95 156 (cid:3) 8.67 24.2 (cid:3) 2.53 89 (cid:3) 6.68

7.49 (cid:3) 0.00 37.5 (cid:3) 0.90 163 (cid:3) 5.45 28.3 (cid:3) 2.70 106 (cid:3) 4.81

7.39 (cid:3) 0.02 36.6 (cid:3) 2.20 154 (cid:3) 9.43 23 (cid:3) 2.50 89 (cid:3) 7.33

Values are mean (cid:3) SD. n ¼ 4 for each group. Iso, isoflurane; MABP, mean arterial blood pressure.

Y. Li et al. / Neuropharmacology 53 (2007) 942e950

(1000 U/ml, pH 7.6) for 10 min at room temperature to initiate breakdown of DNA. Incubation of sections in reaction buffer without TdT provided negative controls. Images were acquired, and TUNEL quantitation performed as de- scribed above for caspase-3.

rats (P115) received only one block of trials each day for 5 days using a new plat- form location in an effort to increase task difficulty and improve test sensitivity.

2.7.4. Probe trials

2.7. Spatial reference memory and learning performance

2.7.1. Morris Water Maze (MWM)

Pregnant rats were allowed to deliver after the isoflurane treatment and 4 pups per litter (2 females and 2 males) were raised. The body weights of the rat pups were recorded at P0, P3, P5, P11, P17 and P28 to determine growth rate. We determined spatial reference memory and learning with the MWM as reported previously with some modification (Jevtovic-Todorovic et al., 2003). A schematic of the experimental paradigm is shown in Fig. 2. A round, fiberglass pool, 150 cm in diameter and 60 cm in height, was filled with water to a height of 1.5 cm above the top of the movable clear 15 cm diameter platform. The pool was located in a room with numerous visual cues (including com- puters, posters and desks) that remained constant during the studies. Water was kept at 20 (cid:2)C and opacified with titanium dioxide throughout all training and testing. A video tracking system recorded the swimming motions of animals and the data were analyzed using motion-detection software for the MWM (Ac- timetrics Software, Evanston, IL, USA). After every trial, each rat was placed in a holding cage, under an infrared heat lamp, before returning to its regular cage.

2.7.2. Cued trials

Probe trials were conducted after the last place trials for the juveniles (P36) and adults (P119) to evaluate memory retention capabilities. After all rats com- pleted the last place trial on the fifth day, the platform was removed from the water maze and rat was started to swim in the quadrant opposite to one the plat- form was placed before. The rats were allowed to swim for 60 s during each probe trial and the time the rats spent in each quadrant was recorded. The per- centage of the swimming time spent in the target (probe) quadrant where the platform was placed before was calculated. The time spent in the target quadrant compared to other quadrants was an indication of memory retention.

2.7.5. Learning to reach criterion test

After the last probe test for the adult rats, the animals performed the learn- ing to reach criterion test during the next 9 days as described previously (Chen et al., 2000). The experimental procedure was similar to the place trial except that the platform location was changed. For each rat, the platform was moved between nine different locations set up by the computer. Each rat received up to eight trials per day. In order to advance to the next platform location, each rat had to reach the criterion of three successive trials with escape latency of 20 s or less. If a rat reached a criterion in 8 or less trials, a new platform lo- cation would be selected the following day. The numbers of learned platforms and the number of trials used to reach the criteria were recorded and com- pared. The number of platforms learned and the number of trials to reach a cri- terion indicated the learning ability of the rats.

The cued trials were performed only for postnatal rats at P28 and P29 (28 rats in control group and 31 rats in treatment group) to determine whether any non- cognitive performance impairments (e.g. visual impairments and/or swimming difficulties) were present, which might affect performance on the place or probe trials. A white curtain surrounded the pool to prevent confounding visual cues. All rats received 4 trials per day. On each trial, rats were placed in a fixed position in the swimming pool facing the wall and were allowed to swim to a platform with a rod (cue) 20 cm above water level randomly placed in any of the 4 quad- rants of the swimming pool. They were allotted 60 s to find the platform upon which they sat for 30 s before being removed from the pool. If a rat did not find the platform within 60 s, the rat was gently guided to the platform and al- lowed to remain there for 30 s. The time for each rat to reach the cued platform and the swim speed was recorded and the data at P28/29 were analyzed.

2.8. Statistical analysis

To reduce variance from different size litters, we averaged the data from all fetal or postnatal rats from the same mother and considered them as a single sample. Results of weight gain of postnatal rat pups, ABG and MABP of preg- nant rats and place trials of postnatal rats were analyzed using 2-way ANOVA for repeated measurements. Data for immunohistochemistry, TUNEL and other behavioral studies were analyzed using Student’s t-test for comparison of two groups or by ANOVA followed by Fisher’s post hoc multiple compar- ison tests for those with more than two groups. In all experiments, difference were considered statistically significant at P < 0.05.

2.7.3. Place trials

3. Results

After completion of cued trials, we used the same rats to perform the place trials to determine the rat’s ability to learn the spatial relationship between distant cues and the escape platform (submerged, no cue rod), which remained in the same location for all place trials. The starting points were random for each rat. The time to reach the platform was recorded for each trial. The less time it took a rat to reach the platform, the better the learning ability. The juvenile rats (P32) received two blocks of trials (two trials per block, 30 s apart, 60 s max- imum for each trial and 2 h rest between blocks) each day for 5 days. The adult

3.1. Comparison of basic physiological variables and brain isoflurane concentrations between 1.3% isoflurane treatment and sham control groups

In the pilot study, the pregnant rats at E21, initially exposed to 1.5% isoflurane for 3 h, developed respiratory acidosis

E21

P0

P28 P29 P32

P36

P115

P119

P120

P128

Delivery Rat pups

Cued Trials 4 trials per day

Place Trials 2 blocks per day 2 trials per block

Place Trials 1 block per day 2 trials perblock

Learning to Reach Criterion Test

Anesthetic Exposures Pregnant rats at gestation day 21

Probe trial after last Place Trial

Probe trial after last Place Trial

Fig. 2. Schematic time-line of Morris Water Maze tests paradigm. E21, pregnant rats at gestational day 21; P0, postnatal day 0.

945

946

Y. Li et al. / Neuropharmacology 53 (2007) 942e950

(PaCO2 increased from 32 to 55 mmHg) and hypoxia (PaO2 decreased from 179 to 82 mmHg). When the isoflurane con- centration was reduced to 1.3%, for up to 6 h (2, 4 and 6 h), there were no significant changes in any of the ABG variables as compared to the controls (Table 1). MABP decreased slightly beginning at 2 h compared to the same animal’s base- line, but was not significantly different from the control group. Therefore, 1.3% isoflurane for 6 h was used in the subsequent formal study. In addition, there were no significant differences in the blood glucose levels before and after exposure in both the control group (113 (cid:3) 28 mg/dl vs. 116 (cid:3) 29 mg/dl; n ¼ 4; P > 0.05) and the 1.3% isoflurane treatment group (117 (cid:3) 17 mg/dl vs. 118 (cid:3) 14 mg/dl; n ¼ 8; P > 0.05). In the behavioral study, there were no significant differences between the two groups on growth rate measured by weight gain in rats from postnatal day 0 (P0) to P28 (data not shown). The concentration of isoflurane in the brain of a pregnant rat was 0.42 mmol/g after exposure to 1.3% isoflurane for 6 h, which was indistinguishable from that of the fetal brain (0.40 mmol/g) measured at the same time.

3.2. Isoflurane inhibited spontaneous neuronal apoptosis in fetus brains

We determined the degree of apoptosis by counting caspase- 3 positive and TUNEL-positive cells in different brain regions at 2 h and 18 h after isoflurane treatment and then at postnatal day 5. There was spontaneous apoptosis represented by caspase-3 positive or TUNEL positive cells in the developing fetal brain (Figs. 3 and 4). The caspase-3 positive cells were concentrated in the dorsal midline of the fetal brain along its rostralecaudal axis. There were no significant differences bet- ween control and treatment groups in the areas of hippocampus CA1 region and the retrosplenial cortex (data not shown) deter- mined at 2, 18 h and P5. Compared to the sham control group, 1.3% isoflurane treatment significantly decreased spontaneous apoptosis determined by the density of apoptotic cells in both the hippocampus CA1 region and in retrosplenial cortex at 2 h after treatment, but no differences were seen at 18 h after treatment or at P5 (Fig. 3B,C and 4B,C). Isoflurane treatment significantly decreased the density of caspase-3 and TUNEL positive cells by 80% (P < 0.001) and 81% (P < 0.001) respec- tively in the hippocampal CA1 region (Fig. 3B,C) and by 82% (P < 0.001) and 87% (P < 0.01) respectively in retrosplenial cortex (Fig. 4B,C) at 2 h after isoflurane treatment.

reference learning ability in the same animals using the place tri- als, the escape latency to platform was analyzed with two-factor ANOVAwith treatment as a between subjects factor and block as a repeated measure. This analysis yields a main difference in block (P < 0.0001 at P32e36 and P115e119). However, nei- ther the main effect of treatment nor the interaction between treatment and block were significant at P32e36 or at P115e 119 (Fig. 5A,B). The results indicated that the performance dur- ing the place trials improved as training progressed but the there was no significant difference between the two groups at juvenile (P32e36) or at adult (P115e119) ages. We further tested the learning ability in the same adult rats using a more rigorous pro- tocoldthe learning to reach criterion test. They all performed equally well in learning as shown in the numbers of platform reached (Fig. 5C) and took the same number of trials to reach a criterion at each platform (Fig. 5D), suggesting a similar learn- ing ability between the two groups. After the place test, the same postnatal rats were used in a probe test to determine memory re- tention. The juvenile postnatal rats born to the mothers treated with 1.3% isoflurane for 6 h had significantly improved reten- tion of memory (P < 0.05) by spending a greater percentage of time swimming in the probe quadrant as compared with the corresponding control animals (Fig. 6). The swim speed in both groups was not different (data not shown), further indicat- ing the improvement of retention memory was not caused by the rats’ swimming ability as showed in cued trials. However, when the probe test was repeated in adult rats (P119), there was no sig- nificant difference in the percentage of time spent in the probe quadrant between the two groups (Fig. 6). These data suggest that any improvement in memory during juvenile age did not last into the adult age.

4. Discussion

Our initial hypothesis was that isoflurane exposure during pregnancy would cause apoptosis in the fetal brain and subse- quent postnatal memory and learning disabilities. This hypoth- esis was based on recent work which showed isoflurane exposure during early postnatal development resulted in an in- crease in apoptosis and subsequent behavioral impairment (Jevtovic-Todorovic et al., 2003). In contrast, we found that isoflurane exposure during late pregnancy transiently inhibited spontaneous apoptosis in the fetal brain and does not impair memory and learning in the juvenile or adult rat.

3.3. Effect of fetal exposure to isoflurane on memory and learning

Using the MWM, we examined the effect of prenatal isoflur- ane exposure on the memory and learning ability in postnatal rats at different ages. In the cued trials, there was no significant difference in the latency of swimming to a visible platform be- tween the control group of 26 postnatal rats from 7 pregnant mothers (n ¼ 7) and the isoflurane treatment group of 31 postna- tal rats from eight pregnant mothers (n ¼ 8) (18.5 (cid:3) 1.86 s vs. 18 (cid:3) 2.22 s; P > 0.05). When we compared the spatial

The fact that isoflurane did not induce apoptosis in the pre- natal rat brain is consistent with a recent study (McClaine et al., 2005), showing no evidence for a neurotoxic effect of the com- bination of isoflurane/midazolam/sodium thiopental exposure to fetal sheep. Spontaneous apoptosis in the fetal developing brains may be a normal process that shapes the brains as it ma- tures. It seemed that spontaneous apoptosis significantly de- creased as the brain matured especially at postnatal day 5 (Figs. 3 and 4), which was consistent with another report (White and Barone, 2001). The inhibition of spontaneous apoptosis in the fetus developing brains by isoflurane determined 2 h after treatment was not expected. Although the significance of the transient inhibition of spontaneous apoptosis by isoflurane is

Y. Li et al. / Neuropharmacology 53 (2007) 942e950

A

(a)

Control, Caspase-3

(b)

1.3% Isoflurane, Caspase-3

(c)

Control, TUNEL

(d)

1.3% Isoflurane, TUNEL

B

2

m m s

/

l l

250

200

***

Control

1.3% Iso

C

2

m m s

/

250

200

***

Control

1.3% Iso

e C e v

i t i

150

l l

e C e v

150

s o P 3 - e s a p s a C

100

50

i t i

s o P L E N U T

100

50

0

0

2h

18h Hippocampus CA1

P5

2h

18h Hippocampus CA1

P5

Fig. 3. In utero isoflurane inhibited spontaneous apoptosis in the hippocampus CA1 determined 2 h after treatment. (A) Arrows indicate caspase-3 positive cells and TUNEL-labeled cells in the hippocampus CA1 region of normal control fetal brains (Aa and Ac) or 2 h after isoflurane treatment (Ab and Ad) at E21. (B and C). The comparison between the density of caspase-3 positive cells (B) and TUNEL-positive cells (C) in the hippocampus CA1 region at different times after iso- flurane treatment. Data represent mean (cid:3) SE of 12e14 fetus rat brains from 5e6 pregnant mothers in either the control group or the isoflurane treatment group at 2 and 18 h after isoflurane treatment (see Fig. 1 for the detailed distribution of rats). Seven postnatal rats from 7 pregnant mothers (n ¼ 7) and 8 postnatal rats from 8 pregnant mothers (n ¼ 8) from the behavioral study were used for the neurodegeneration study at P5 (see Fig. 1). Data represent mean (cid:3) SE. ***P < 0.001 versus control. Iso, isoflurane. Scale bar 50 mm.

unknown from this study, we speculate this will not affect brain development significantly. Our observation of postnatal rats in the behavioral study did not note obvious physical or behavioral differences between control and isoflurane treatment groups, except the transient memory improvement in the isoflurane treated rats. Nevertheless, the results from this study suggest

that isoflurane exposure to pregnant rats does not induce apo- ptosis in the developing fetal brain and does not impair the memory and learning of their offspring.

Our results would appear to be at odds with those observed by Jevtovic-Todorovic et al. (2003). However, in that study, ne- onates and not fetuses were exposed to isoflurane, and the

947

948

Y. Li et al. / Neuropharmacology 53 (2007) 942e950

A

(a)

Control, Caspase-3

(b)

1.3% Isoflurane, Caspase-3

(c)

Control, TUNEL

(d)

1.3% Isoflurane, TUNEL

B

35

Control

C

35

Control

2

m m / s l l e C e v i t i

30

25

20

***

1.3% Iso

2

m m / s l l e C e v i t i

30

25

20

**

1.3% Iso

s o P 3 - e s a p s a C

15

10

5

s o P L E N U T

15

10

5

0

0

2h

18h Retrosplenial Cortex

P5

2h

18h Retrosplenial Cortex

P5

Fig. 4. Isoflurane inhibited spontaneous apoptosis in the retrosplenial cortex of fetal brains 2 h after treatment. (A) Arrows indicate caspase-3 positive cells and TUNEL-labeled cells in the retrosplenial cortex of normal control fetal brains (Aa and Ac) or 2 h after isoflurane treatment (Ab and Ad) at E21. The density of caspase-3 positive cells (B) and TUNEL-positive cells (C) in the retrosplenial cortex area at different times after isoflurane treatment were compared between the control and 1.3% isoflurane treatment groups. Data represent mean (cid:3) SE of 12e14 postnatal rats from 5e6 pregnant mothers in either control group or the iso- flurane treatment group at 2 and 18 h after isoflurane treatment (see Fig. 1 for the detailed distribution of rats). Seven postnatal rats from 7 pregnant mothers (n ¼ 7) and 8 postnatal rats from 8 pregnant mothers (n ¼ 8) from the behavioral study were used for the neurodegeneration study at P5 (see Fig. 1). Data represent mean (cid:3) SE. ***P < 0.001, **P < 0.01, compared to control. Iso, isoflurane. Scale bar 50 mm.

Y. Li et al. / Neuropharmacology 53 (2007) 942e950

A

50

Control

40B

Control

) s d n o c e S

( y c n e t a L

40

30

20

10

1.3% Iso

) s d n o c e S

(

y c n e t a L

30

20

10

1.3% Iso

0

0

0

1

4 Day of Test (P32-36)

2

3

5

0

1

4 Day of Test (P115-119)

2

3

5

10C m r o f t a P d e n r a e L f o r e b m u N

8

l

6

4

2

D

a

i r e t i r C h c a e R o t

s l a i r T

10

8

6

4

2

Control

1.3% Iso

0

control

1.3% Iso

0

0

1

2

3 platform location

4

5

6

7

8

9

Fig. 5. Isoflurane in utero did not affect postnatal learning ability in juvenile or adult rats. Spatial learning and memory performance were determined using the Morris Water Maze place test paradigm in postnatal juvenile (A) and adult (B) rats and using the learning to reach criteria test in the P120e128 adult (C and D). Data represent mean (cid:3) SE of 26 postnatal rats from 7 pregnant mothers (n ¼ 7) in the control group or 31 postnatal rats from 8 pregnant mothers (n ¼ 8) in the isoflurane treatment group. Iso, isoflurane.

interpretation was that the vulnerability of isoflurane-induced apoptosis in the developing brains is correlated to the period of synaptogenesis (Jevtovic-Todorovic et al., 2003; Yon et al., 2005). If true, then the fetal rat is not expected to be as vulner- able as the neonatal rat, which is consistent with the results of

50



Control

t n a r d a u q e b o r p n

40

30

20

1.3% Iso

i

e m

i t

%

10

0

P36

P119

Day of Test

Fig. 6. Isoflurane significantly increased spatial retention memory in postnatal juvenile but not adult rats. Probe test was performed at postnatal day 36 (P36) and 119 (P119) after the last place trial. Data represent mean (cid:3) SE of 26 post- natal rats born from 7 pregnant mothers (n ¼ 7) in the control group or 31 postnatal rats born from 8 pregnant mothers (n ¼ 8) in the isoflurane treatment group. *P < 0.05 compared to control. Iso, isoflurane.

our study. However, it should be noted that although vulnerabil- ity to isoflurane-induced apoptosis seems to coincide with the period of synaptogenesis (Hansen et al., 2004; Ikonomidou et al., 2001; Jevtovic-Todorovic et al., 2003), there exists no di- rect evidence showing that synaptogenesis is itself altered, or is causally linked to the subsequent cognitive changes. Toxic ef- fects of isoflurane have also been found in the adult and aged brains (Culley et al., 2003, 2004), when of course little in the way of synaptogenesis is occurring.

Isoflurane has long been considered to be cytoprotective against ischemia in the heart and brain (Sakai et al., 2007; Warner, 2000). In previous studies, we and others (Kudo et al., 2001; Wei et al., 2005) have shown that the concentration and time required for isoflurane to induce apoptosis in cultured neurons was greater than that used clinically in patients. It is possible that sub-apoptotic exposure to isoflurane induces pro- tection, analogous to that induced by hypoxia. Accordingly, our preliminary unpublished data have shown that preconditioning of rat primary cortical neurons with 2 MAC isoflurane for 1 h abolished the neurotoxicity induced by a subsequent exposure to 2 MAC isoflurane for 24 h. Thus, it is possible that our use of isoflurane at low ‘‘clinical’’ concentration in the pregnant rat may precondition the fetal brain against spontaneous apo- ptosis in the current study. Therefore, it remains possible that the higher anesthetic concentrations used in fetal surgery (w3% isoflurane) may produce neurotoxicity in the fetal brain.

949

950

Y. Li et al. / Neuropharmacology 53 (2007) 942e950

Other limitations of this study are that we did not expose the mother/fetus to isoflurane at other time points during the pregnancy, so it is possible that we missed the time point when the fetus might be more vulnerable to isoflurane neuro- toxicity. Further, we avoided tracheal intubation in this study in an attempt to minimize confounding variables. This pre- vented us from testing the effects of isoflurane at concentra- tions up to 3% often used during fetal surgery in humans (Cauldwell, 2002; Myers et al., 2002).

cleaved caspase 3. Journal of Histochemistry and Cytochemistry 50, 449e454.

Hansen, H.H., Briem, T., Dzietko, M., Sifringer, M., Voss, A., Rzeski, W., Zdzisinska, B., Thor, F., Heumann, R., Stepulak, A., Bittigau, P., Ikonomidou, C., 2004. Mechanisms leading to disseminated apoptosis fol- lowing NMDA receptor blockade in the developing rat brain. Neurobiol- ogy of Disease 16, 440e453.

Ikonomidou, C., Bittigau, P., Koch, C., Genz, K., Hoerster, F., Felderhoff- Mueser, U., Tenkova, T., Dikranian, K., Olney, J.W., 2001. Neurotransmit- ters and apoptosis in the developing brain. Biochemical Pharmacology 62, 401e405.

In summary, prenatal exposure to isoflurane at a concentra- tion commonly used for the maintenance of general anesthesia during late pregnancy in rats does not appear to be neurotoxic to the fetal brain and does not impair memory and learning in the postnatal juvenile or adult rat.

Acknowledgments

Jevtovic-Todorovic, V., Hartman, R.E.,

Izumi, Y., Benshoff, N.D., Dikranian, K., Zorumski, C.F., Olney, J.W., Wozniak, D.F., 2003. Early ex- posure to common anesthetic agents causes widespread neurodegeneration in the developing rat brain and persistent learning deficits. Journal of Neu- roscience 23, 876e882.

Kim, H., Oh, E., Im, H., Mun, J., Yang, M., Khim, J.Y., Lee, E., Lim, S.H., Kong, M.H., Lee, M., Sul, D., 2006. Oxidative damages in the DNA, lipids, and proteins of rats exposed to isofluranes and alcohols. Toxicology 220, 169e178.

We acknowledge discussions with Drs Roderic Eckenhoff, Randall Pittman and Maryellen Eckenhoff and the technical support of Dr Min Li, University of Pennsylvania. We thank Dr Alex Loeb for editorial assistance. This study was supported by March of Dimes Birth Defects Foundation Research Grant (12-FY05-62, PI: Huafeng Wei), and partially Institute of General Medical supported by the National Science (NIGMS) K08 grant (1-K08-GM-073224-01, PI: Huafeng Wei), and the Research Fund at the Department of Anesthesiology and Critical Care, University of Pennsylvania (PI: Huafeng Wei).

Kudo, M., Aono, M., Lee, Y., Massey, G., Pearlstein, R.D., Warner, D.S., 2001. Effects of volatile anesthetics on N-methyl-D-aspartate excitotoxicity in primary rat neuronaleglial cultures. Anesthesiology 95, 756e765.

Kvolik, S., Glavas-Obrovac, L., Bares, V., Karner, I., 2005. Effects of inhala- tion anesthetics halothane, sevoflurane, and isoflurane on human cell lines. Life Sciences 77, 2369e2383.

Loop, T., Dovi-Akue, D., Frick, M., Roesslein, M., Egger, L., Humar, M., Hoetzel, A., Schmidt, R., Borner, C., Pahl, H.L., Geiger, K.K., Pannen, B.H., 2005. Volatile anesthetics induce caspase-dependent, mito- chondria-mediated apoptosis in human T lymphocytes in vitro. Anesthesi- ology 102, 1147e1157.

Matsuoka, H., Kurosawa, S., Horinouchi, T., Kato, M., Hashimoto, Y., 2001. Inhalation anesthetics induce apoptosis in normal peripheral lymphocytes in vitro. Anesthesiology 95, 1467e1472.

References

Cauldwell, C.B., 2002. Anesthesia for fetal surgery. Anesthesiology Clinics of

McClaine, R.J., Uemura, K., de la Fuente, S.G., Manson, R.J., Booth, J.V., White, W.D., Campbell, K.A., McClaine, D.J., Benni, P.B., Eubanks, W.S., Reynolds, J.D., 2005. General anesthesia improves fetal cerebral oxygenation without evidence of subsequent neuronal injury. Journal of Cerebral Blood Flow and Metabolism 25, 1060e1069.

North America 20, 211e226.

Myers, L.B., Cohen, D., Galinkin, J., Gaiser, R., Kurth, C.D., 2002. Anaesthe-

Chang, Y.C., Chou, M.Y., 2001. Cytotoxicity of halothane on human gingival

sia for fetal surgery. Paediatric Anaesthesia 12, 569e578.

fibroblast cultures in vitro. Journal of Endodontics 27, 82e84.

Paxinos, G., Tork, I., Teccot, L.H., Valentino, K., 1990. Atlas of the Develop-

Chen, G., Chen, K.S., Knox, J., Inglis, J., Bernard, A., Martin, S.J., Justice, A., McConlogue, L., Games, D., Freedman, S.B., Morris, R.G., 2000. A learn- ing deficit related to age and beta-amyloid plaques in a mouse model of Alzheimer’s disease. Nature 408, 975e979.

ing Rat Brain. Academic Press.

Sakai, H., Sheng, H., Yates, R.B., Ishida, K., Pearlstein, R.D., Warner, D.S., 2007. Isoflurane provides long-term protection against focal cerebral ische- mia in the rat. Anesthesiology 106, 92e99.

Culley, D.J., Baxter, M., Yukhananov, R., Crosby, G., 2003. The memory ef- fects of general anesthesia persist for weeks in young and aged rats. Anes- thesia and Analgesia 96, 1004e1009.

Culley, D.J., Baxter, M.G., Yukhananov, R., Crosby, G., 2004. Long-term impairment of acquisition of a spatial memory task following isoflurane- nitrous oxide anesthesia in rats. Anesthesiology 100, 309e314.

Dobbing, J., Sands, J., 1979. Comparative aspects of the brain growth spurt.

Early Human Development 3, 79e83.

Warner, D.S., 2000. Isoflurane neuroprotection e a passing fantasy, again? An-

esthesiology 92, 1226e1228.

Wei, H., Kang, B., Wei, W., Liang, G., Meng, Q.C., Li, Y., Eckenhoff, R.G., 2005. Isoflurane and sevoflurane affect cell survival and BCL-2/BAX ratio differently. Brain Research 1037, 139e147.

White, L.D., Barone, S., 2001. Qualitative and quantitative estimates of apo- ptosis from birth to senescence in the rat brain. Cell Death and Differen- tiation 8, 345e356.

Eckenhoff, R.G., Johansson, J.S., Wei, H., Carnini, A., Kang, B., Wei, W., Pidikiti, R., Keller, J.M., Eckenhoff, M.F., 2004. Inhaled anesthetic en- hancement of amyloid-beta oligomerization and cytotoxicity. Anesthesiol- ogy 101, 703e709.

Wise-Faberowski, L., Zhang, H., Ing, R., Pearlstein, R.D., Warner, D.S., 2005. Isoflurane-induced neuronal degeneration: an evaluation in orga- notypic hippocampal slice cultures. Anesthesia and Analgesia 101, 651e657.

Gavrieli, Y., Sherman, Y., Bensasson, S.A., 1992. Identification of pro- grammed cell-death in situ via specific labeling of nuclear-DNA fragmen- tation. Journal of Cell Biology 119, 493e501.

Goldsmith, M.F., Straus, S.E., Kupfer, C., Lenfant, C., Collins, F., Hodes, R.J., Gordis, E., Fauci, A.S., Alexander, D., Battey, J.F., Slavkin, H.C., Leshner, A.I., Olden, K., Hyman, S.E., Fischbach, G.D., 1999. 2020 vi- sion: NIH heads foresee the future. Journal of the American Medical As- sociation 282, 2287e2290.

Gown, A.M., Willingham, M.C., 2002. Improved detection of apoptotic cells in archival paraffin sections: immunohistochemistry using antibodies to

Xie, Z., Dong, Y., Maeda, U., Alfille, P., Culley, D.J., Crosby, G., Tanzi, R.E., 2006. The common inhalation anesthetic isoflurane induces apoptosis and increases amyloid beta protein levels. Anesthesiology 104, 988e994. Xie, Z., Dong, Y., Maeda, U., Moir, R.D., Xia, W., Culley, D.J., Crosby, G., Tanzi, R.E., 2007. The inhalation anesthetic isoflurane induces a vicious cycle of apoptosis and amyloid beta-protein accumulation. Journal of Neu- roscience 27, 1247e1254.

Yon, J.H., Daniel-Johnson, J., Carter, L.B., Jevtovic-Todorovic, V., 2005. An- esthesia induces neuronal cell death in the developing rat brain via the in- trinsic and extrinsic apoptotic pathways. Neuroscience 135, 815e827.