new_pdfs/10.1007_s12640-018-9877-3.txt

Neurotoxicity Research (2018) 34:188–197 https://doi.org/10.1007/s12640-018-9877-3

ORIGINAL ARTICLE

Neonatal Exposure to Low-Dose (1.2%) Sevoflurane Increases Rats’ Hippocampal Neurogenesis and Synaptic Plasticity in Later Life

Xi Chen 1 & Xue Zhou 1 & Lu Yang 1 & Xu Miao 1 & Di-Han Lu 1 & Xiao-Yu Yang 1 & Zhi-Bin Zhou 1 & Wen-Bin Kang 1 & Ke-Yu Chen 1 & Li-Hua Zhou 2 & Xia Feng 1

Received: 26 October 2017 / Revised: 7 January 2018 / Accepted: 26 January 2018 / Published online: 9 February 2018 # Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract The increasing usage of general anesthetics on young children and infants has drawn extensive attention to the effects of these drugs on cognitive function later in life. Recent animal studies have revealed improvement in hippocampus-dependent perfor- mance after lower concentrations of sevoflurane exposure. However, the long-term effects of low-dose sevoflurane on the developing brain remain elusive. On postnatal day (P) 7, rats were treated with 1.2% sevoflurane (1.2% sevo group), 2.4% sevoflurane (2.4% sevo group), and air control (C group) for 6 h. On P35–40, rats’ hippocampus-dependent learning and memory was tested using the Morris water maze. Cognition-related and synapse-related proteins in the hippocampus were measured using Western blotting on P35. On the same day, neurogenesis and synapse ultrastructure were evaluated using immunofluorescence and transmission electron microscopy (TEM). On P35, the rats neonatally exposed to 1.2% sevoflurane showed better behavioral results than control rats, but not in the 2.4% sevo group. Exposure to 1.2% sevoflurane increased the number of 5′-bromo-2- deoxyuridine (BrdU)-positive cells in the dentate gyrus and improved both synaptic number and ultrastructure in the hippocam- pus. The expression levels of BDNF, TrkB, postsynaptic density (PSD)-95, and synaptophysin in the hippocampus were also increased in the 1.2% sevo group. In contrast, no significant changes in neurogenesis or synaptic plasticity were observed between the C group and the 2.4% sevo group on P35. These results showed that exposure of the developing brain to a low concentration of sevoflurane for 6 h could promote spatial learning and memory function, along with increased hippocampal neurogenesis and synaptic plasticity, in later life.

Keywords Sevoflurane . Hippocampus . Cognitive function . Neurogenesis . Synaptic plasticity

Introduction

The developing brain is vulnerable to environmental influ- ences including general anesthetics (Loepke and Soriano 2008; Mellon et al. 2007). The widespread and prevalent use

Xi Chen and Xue Zhou contributed equally to this work.

Li-Hua Zhou

zhoulih@mail.sysu.edu.cn

Xia Feng

fengxiar@sina.com

1 Department of Anaesthesiology, The First Affiliated Hospital of Sun

Yat-Sen University, No. 58 Zhongshan Road 2, Guangzhou 510080, Guangdong, People’s Republic of China 2 Department of Anatomy, Zhongshan School of Medicine, Sun

of anesthesia in children makes its safety a major health issue of interest. Sevoflurane is a volatile anesthetic that is common- ly used, particularly in clinical pediatric anesthesia, because it is better tolerated than many other anesthetics and has excel- lent results (Goa et al. 1999). Accumulating evidence suggests that neonatal rodent exposure to higher concentrations of sevoflurane could induce developmental neurotoxicity, in- cluding long-term learning disabilities, degeneration of neu- rons, and impairment of synaptic plasticity (Feng et al. 2012; Ishizeki et al. 2008; Tao et al. 2016). However, recent studies have shown that lower concentrations of sevoflurane have neuroprotective effects (Chen et al. 2015; Payne et al. 2005). These complicated results suggest that different parameters such as concentration, timing, and exposure duration of sevoflurane are critical to the final outcomes and reflect the complexity of the effects on the central nervous system. However, the effect of low-concentration sevoflurane during the neonatal period on learning and memory ability in later life

Yat-Sen University, No. 74 Zhongshan Road 2, Guangzhou 510080, Guangdong, People’s Republic of China

Neurotox Res (2018) 34:188–197

has been unclear. The present work aimed to fill a current gap in the literature by investigating whether exposure to a low concentration (1.2%) of sevoflurane during early postnatal life impacts learning and memory ability later in life.

Here, we exposed rats at postnatal day (P) 7 to air or to 1.2 or 2.4% sevoflurane for 6 h. During the juvenile stage, we compared the effects of lower and higher concentrations of sevoflurane on anesthetic-induced hippocampus-dependent learning and memory ability, neurogenesis, and synaptic plas- ticity in the hippocampal area.

Methods

Ethical Approval

The use of rats in this study was approved by the Institutional Animal Care and Use Committee at Sun Yat-sen University (Guangzhou, China). All experiments were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and ARRIVE guidelines. Sprague-Dawley multiparous dams (n = 31) with litters con- taining male pups (n = 135) were purchased from Experimental Animal Center of Sun Yat-sen University, China. We only used male offspring to exclude the influence of estrogen on the biochemical data and neurocognitive func- tions. The pups from postnatal day 0 (P0) to P20 were housed with the dams in a 12-h:12-h light:dark cycle (light from 07:00 to 19:00), and room temperature (RT) was maintained at 21 ± 1 °C. On P21, the pups were weaned and housed 4–6 per cage in a standard environment.

Anesthesia

SD rats at P7 (weight 14–16 g) were randomly divided into the air-treated control (C group), the 1.2% sevoflurane-exposed (1.2% sevo group), and the 2.4% sevoflurane-exposed (2.4% sevo group). Rats in the 1.2% sevo group and the 2.4% sevo group were placed in a plastic container and exposed to 1.2 or 2.4% sevoflurane continuously for 6 h, using air as a carrier, with a total gas flow of 2 L min−1. A nasopharyngeal airway tube was put in their mouth to prevent apnea and hypoxia when the rats stopped moving in the container. During expo- sure, the temperature inside the container was maintained at 30 °C using an external heating device (NPS-A3 heating de- vice, Midea Co., Guangdong, China) and a hot water bag on the bottom of the container with a constant temperature main- tained between 30 and 35 °C. The concentrations of sevoflurane, oxygen, and carbon dioxide in the chamber were monitored by a gas monitor (Detex-Ohmeda, Louisville, CO, USA). During exposure, an investigator monitored the rats’ spontaneous respiratory frequency and skin color every 5 min to detect any apnea or hypoxia. The rats were immediately

exposed to air and excluded from the experiment if these symptoms were detected. Sevoflurane administration was ter- minated 6 h later, and the rats were exposed only to air. When the rats were moving freely again, they were placed back into their maternal cages. Rats in the C group were exposed to the same container as the rats in the 1.2% sevo and 2.4% sevo group but were exposed to air alone for 6 h.

Arterial Blood Gas Analysis

We performed arterial blood analysis in order to exclude the influence of respiratory or metabolic disorder. The arterial blood samples from the C, 1.2% sevo, and 2.4% sevo groups were obtained from the left cardiac ventricle immediately after removal from the maternal cage (n = 5 in each group) at the end of anesthesia. They were analyzed immediately after col- lection using a blood gas analyzer (Gem Premier 3000, US). We analyzed the pH, arterial carbon dioxide tension (PaCO2), arterial oxygen tension (PaO2), and blood glucose levels of the arterial blood samples.

Morris Water Maze Test

On P35, the rats were tested for spatial learning and memory ability using the Morris water maze (MWM). Three groups of rats (n = 10 in each group, weight 90–100 g) were tested on the MWM, which consists of two different tests including hidden platform acquisition and a probe trial test, at P35– P40 using the Water Maze Tracking System (MT-200; Chengdu, China). A white platform (12 cm diameter) was submerged in a circular pool (160 cm in diameter, 50 cm in height) that was filled with warm water (23 ± 2 °C). The pool, located in a room with no windows, was virtually divided into four quadrants. A video camera connected to the computer running the tracking software was suspended above the pool and captured the rats’ movements for analysis. At P35, before the test, a single habituation trial was performed without the platform; in this trial, the rats were placed in the water for 120 s. In the hidden platform acquisition test, performed at P36–P39, each rat was placed, facing the wall of the pool, in one of the four quadrants and allowed to swim freely in search of the escape platform for a maximum of 120 s. The experi- ment was repeated with four trials per day for four consecutive days. The average escape latency time (latency to reach the platform) was measured to evaluate spatial learning ability. At P40, a probe trial test was performed by removing the plat- form and releasing the rats into the water for 120 s. We calcu- lated the time spent in the quadrant that previously contained the target and the frequency of crossing the former location of the platform. The rats were dried and placed back into a heated cage after completing each test.

189

190

Western Blot Analysis

On P10 and 28, rats (n = 5 in each group at each sacrifice time point) were sacrificed by rapid decapitation, and the bilateral hippocampus areas were harvested and stored at − 80 °C until use. Protein was extracted using RIPA lysis buffer (Keygen Biotech, Nanjing, China). The amount of protein in each hip- pocampal tissues was measured using a protein assay kit (BCA, Pierce, Thermo, USA). Polyacrylamide-SDS gels with an equal amount of 50-μg load in each lane were electropho- resed, and the proteins transferred onto PVDF membranes (Millipore, Carrigtwohill, Ireland). The blots were blocked with 5% skim milk in Tris-buffered saline (150 mM NaCl, 0.1% TWEEN 20, 20 mM Tris, pH 7.4) for 1 h and then incubated overnight at 4 °C with anti-BDNF (1:1000, Novusbio, USA), anti-TrkB (1:800, Millipore, Ireland), anti- postsynaptic density (PSD-95) (1:2000, Abcam, England), and anti-synaptophysin (1:20,000, Abcam, England) primary antibodies. After rinsing, membranes were probed with corre- sponding secondary antibodies at RT for 2 h. Immunoreactive bands were detected with an enhanced chemiluminescence detection system (Bio-Rad, USA). A β-actin antibody (1:1000, ABclonal, China) was used to normalize for sample loading and transfer. The intensities of the bands were densitometrically quantified using ImageJ.

BrdU Injections and Immunofluorescence

For the 5′-bromo-2-deoxyuridine (BrdU) injections, we followed the methods as previously described (Chen et al. 2015; Tozuka et al. 2005). BrdU has been described as a marker of neurogenesis and can incorporate into DNA only during the S-phase of the mitotic process (Kee et al. 2002). BrdU (Sigma, America) was dissolved in normal saline (10 mg mL−1) and injected at a dosage of 300 mg/kg. To investigate the effects of 1.2% sevoflurane on cellular prolif- eration, we performed a single injection of BrdU i.p. 24 h after sevoflurane exposure. Three days later, the rats were perfused, and their brains were processed for immunofluorescence. To investigate the effects of 1.2% sevoflurane on the survival of newborn cells, we performed a single injection of BrdU i.p. 24 h before sevoflurane exposure. Four weeks (28 days) after the BrdU injection, the rats were perfused, and their brains were processed for immunofluorescence.

For morphological examination, rats were deeply anesthe- tized with chloral hydrate at P10 and P35 (n = 5 in each group at each sacrifice time point, 80–100 g) and then transcardially perfused with 0.9% normal saline at RT followed by a fixative solution of 4% paraformaldehyde (Sigma-Aldrich, St. Louis, MO, USA) in 0.1 M PBS (pH 7.4) at 4 °C. The brains were harvested, postfixed in 4% paraformaldehyde for 8 h, and subsequently soaked in 30% sucrose until they sank. Consecutive frozen coronal sections of the hippocampus were

Neurotox Res (2018) 34:188–197

cut at a thickness of 30 μm. Every fifth section of the consec- utive sections was processed by BrdU staining. DNA was first denatured by incubation with 2 N HCl for 30 min at 37 °C followed by a 15-min wash in 0.1 M boric acid (pH 8.5), with three 10-min washes in 0.01 M PBS before each step. The sections were blocked in 3% BSA and 0.4% Triton X-100 for 2 h at RT before being incubated with primary antibody (rat anti-BrdU, 1:200, ab6326, Abcam, UK) in 1% BSA over- night at 4 °C. Then, the sections were incubated with second- ary antibody (Cy3 goat anti-rat IgG, 1:200, KGAB018, Keygentee, China) for 2 h at RT. Fluorescence was detected with a fully automatic fluorescence microscope (Olympus BX63, Japan). An observer who was blinded to group assign- ment was responsible for counting the number of BrdU- positive cells at × 200 magnification. Total cell counts were divided by the total number of sections for analysis.

Transmission Electron Microscopy

TEM was used to assess synaptic plasticity in the hippo- campus after exposure to treatment (n = 5 in each group) at P35. Twenty-eight days after exposure to treatment, the rats were perfused transcardially with 50 mL of 0.9% normal saline, followed by 50 mL of a mixture of 2% paraformaldehyde and 2.5% glutaraldehyde (Sigma- Aldrich, G6257, USA) in 0.1 M PBS. Approximately 1 mm3 of tissue per rat was dissected from the hippocam- pus and fixed in 2% glutaraldehyde for 2 h at 4 °C. The tissues were rinsed in 0.1 M cacodylate buffer and postfixed with 1% osmium tetroxide for 2 h. Then, the tissue was rinsed with distilled water before undergoing dehydration in a graded ethanol series. Subsequently, the tissue was infiltrated overnight at 4 °C using a mixture of half acetone and half resin. The tissue was embedded in resin 24 h later and then cured fully as follows: 37 °C overnight, 45 °C for 12 h, and 60 °C for 24 h. After that, 70-nm sections were cut and stained with 3% uranyl ace- tate for 20 min and 0.5% lead citrate for 5 min. Ultrastructural changes in synapses in the hippocampus were observed under TEM. Five pictures of each subre- gion per ultrathin section (five rats in total per group) were taken at each of two magnifications: × 13,500 and × 37,000. All pictures taken at × 13,500 magnification were used to observe the number of synapses, and all pictures taken at × 37,000 magnification were used to measure the thickness of the postsynaptic density and the width of the synaptic cleft. The number of synapses was expressed as the average number of synapses in each picture taken at × 13,500. The thickness of the postsynap- tic density and the width of the synaptic were expressed as the average values for all synapses in all pictures taken at × 37,000, as described. We measured the distances using the image analysis software ImageJ.

Neurotox Res (2018) 34:188–197

191

Statistical Analysis

The results were expressed as the mean ± standard deviation (SD) for each group. The statistical tests were conducted using the computerized statistical package SPSS 19.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism Software version 5.0 (GraphPad Software, Inc., San Diego, CA, USA). The arterial blood data were analyzed using Student’s t test. One-way ANOVA was used to evaluate differences in the quantities of hippocampal proteins, numbers of BrdU-positive cells and synapses, and ultrastructure parameters of synapses among groups. Unpaired t tests and two-way ANOVA were used to analyze the results of the MWM. Each experiment was per- formed at least three times. A value of P < 0.05 was consid- ered statistically significant.

which indicated that the rats were learning from practice every day. However, from the third training day, the la- tency to locate the hidden platform in the 1.2% sevo group was significantly shorter than that in the C group and of the 2.4% sevo group (all P < 0.001 vs. C group; Fig. 1a). In the probe trial, the time spent in the target quadrant in the 1.2% sevo group (5.22 ± 2.30) was longer than that in the C (4.08 ± 2.50) group and in the 2.4% sevo group (4.19 ± 2.21; P < 0.01, Fig. 1b). Moreover, the frequency of passing through the target quadrant was significantly higher in the 1.2% sevo group (4.50 ± 2.396) than that in the C group (3.17 ± 1.76) and the 2.4% sevo group (2.92 ± 1.53, Fig. 1c). There were no significant differences in the rats’ swimming speeds (C, 23.44 ± 0.99 cm s−1; 1.2% sevo, 24.01 ± 0.81 cm s−1; 2.4% sevo, 24.43 ± 1.02 cm s−1, according to one-way ANOVA, P = 0.63) among the three groups.

Results

Sevoflurane Does Not Cause Respiratory or Metabolic Disorder

1.2% Sevoflurane Increased the Number of BrdU-Positive Cells in DG

During sevoflurane exposure period, none of the rats appeared with apnea or hypoxia. Arterial blood analysis was used to exclude the influence of respiratory or metabolic disorder. No rats died during the exposure period. Compared with the control group, there were no significant changes in the pH, PaCO2, PaO2, or arterial blood glucose levels before or after exposure in the 1.2% sevo group or the 2.4% sevo group (Table 1).

1.2% Sevoflurane Increased Spatial Learning and Memory Development Later in Life

Four weeks after sevoflurane exposure, the spatial learn- ing and memory was measured by the MWM test de- scribed in the BMethods.^ The results (Fig. 1) showed that the latency to find the hidden platform decreased gradu- ally day by day as training progressed in the three groups,

To test the level of neurogenesis in the DG, we used BrdU to label proliferative cells as an indicator of neurogenesis. The im- munofluorescence images showed that BrdU-labeled cells were present in all three groups on both P10 and P35, which means that hippocampal neurogenesis was active in the young rats. In addition, the neurogenesis level was higher on P10 (Fig. 2a–c) than that on P35 (Fig. 2d–f) in all the groups. The data from immunofluorescence staining demonstrated that 1.2% sevoflurane significantly increased the number of BrdU- positive cells on both P10 and P35. Statistical testing showed that the number of BrdU-positive cells was significantly larger in the 1.2% sevo group than that in the C group and the 2.4% sevo group on P10 (C, 254 ± 37.92; 1.2% sevo, 408 ± 35.85; 2.4% sevo, 245 ± 34.24; P < 0.0001, Fig. 2g) and P35 (C, 30.83 ± 2.85; 1.2% sevo, 35.83 ± 2.14; 2.4% sevo, 28.50 ± 2.35; P = 0.0004, Fig. 2h). No significant difference was found between the C group and the 2.4% sevo group.

Table 1 Arterial blood analysis (N = 5 in each group)

Groups

C

1.2% sevo

2.4% sevo

P value

pH PaCO2 (kPa) PaO2 (kPa) Glucose (mmol L−1)

7.40 ± 0.05 3.57 ± 0.38 13.38 ± 0.55 5.6 ± 0.8

7.39 ± 0.08 3.56 ± 0.52 13.42 ± 0.51 5.3 ± 0.7

7.37 ± 0.09 3.58 ± 0.45 13.35 ± 0.60 5.5 ± 0.5

0.70 0.82 0.92 0.69

Neonatal exposure to a low or high concentration of sevoflurane does not lead to significant cardiorespiratory dysfunction. Arterial blood gas analysis revealed no significant difference in any of the measured parameters among the three groups (t test, all P values > 0.05) PaCO2 arterial carbon dioxide tension, PaO2 arterial oxygen tension, Glucose blood glucose levels, C control, 1.2% sevo 1.2% sevoflurane-exposed group, 2.4% sevo 2.4% sevoflurane-exposed group

192

Fig. 1 Exposure to 1.2%, but not 2.4%, sevoflurane in neonatal rats on P7 induces spatial learning and memory changes in the juvenile stage. a In the MWM, the 1.2% sevo group had a significantly shorter latency than the C group to reach the platform. b The numbers of rats that reached the target quadrant within 120 s were significantly increased in the 1.2% sevo group compared with those in the C group. c The frequency of crossing the former location of the target platform within 120 s was significantly increased in the 1.2% sevo group compared with that in the C group. Data are presented as the mean ± SD (n = 10 in each group). *P < 0.05 versus C group; **P < 0.01 versus C group; ***P < 0.001 versus C group. C, control; 1.2% sevo, 1.2% sevoflurane-exposed group; 2.4% sevo, 2.4% sevoflurane-exposed group

Hippocampal Synaptic Changes in Hippocampus

Expression of Synapse-Associated Proteins in the Hippocampus

Two typical synaptic proteins in the hippocampus were visualized by SDS-PAGE and immunoblotting with corresponding antibod- ies for PSD-95 and synaptophysin (SYN). The optical density ratios of the band intensities of the studied proteins normalized to β-actin were expressed as fold changes in the three groups.

Compared with the level of PSD-95 in the hippocampus of the C group, the PSD-95 protein in the hippocampus of the 1.2% sevo group was significantly increased on P35 (158.6% of control, P = 0.0066), but there was no significant change in the 2.4% sevo group (128.5% of control, P = 0.73). Moreover, as with the higher level of PSD-95 observed in the 1.2% sevo group, the SYN protein level in the hippocampus of the 1.2% sevo group at P35 was also increased (231.6% of control, P = 0.0082) compared with that of the C group (Fig. 3). However, the SYN protein levels showed no significant change in the 2.4% sevo group compared with those in the C group (197.6% of control, P = 0.50).

Ultrastructural Changes in Hippocampal Synapses

The number of synapses and the synaptic ultrastructure of the hippocampus were examined using TEM 4 weeks after

Neurotox Res (2018) 34:188–197

sevoflurane exposure. Compared with the C and 2.4% sevo groups, the 1.2% sevo group showed an increase in the num- ber of synapses in the hippocampus (C, 13.0 ± 2.74; 1.2% sevo, 17.4 ± 2.80; 2.4% sevo, 10.80 ± 1.92; P = 0.0042 of control, Fig. 4a–c, g). The statistical analysis showed a signif- icant difference in ultrastructure changes: we found a notice- ably narrower synaptic cleft width and greater PSD thickness in the 1.2% sevo group than in either of the other groups (Table 2). No differences were found between the C and 2.4% sevo groups in the number of synapses, synaptic cleft width, or PSD thickness.

1.2% Sevoflurane Increased the Levels of BDNF and TrkB in the Hippocampus

To assess the neuroprotective effects of low-dose sevoflurane on the developing brain, the levels of the neurogenesis- and synaptic plasticity-related proteins BDNF and TrkB (Lu et al. 2013; Sairanen et al. 2005) (Hariri et al. 2003) were examined by Western blotting at P35. The optical density ratios of the band intensities of the studied proteins normalized to β-actin were expressed as fold changes for the three groups. At P35, the expression levels of BDNF and TrkB protein in the hip- pocampus of the 1.2% sevo group were significantly in- creased compared with those of the C group (BDNF 63.5% of control, P = 0.0063; TrkB 49.8% of control, P = 0.0002) and the 2.4% sevo group (BDNF 54.2% of control, P = 0.90;

Neurotox Res (2018) 34:188–197

Fig. 2 Neonatal exposure to 1.2% sevoflurane increased neuronal neurogenesis in the hippocampal DG on P10 and P35. a–c BrdU- positive cells on P10 in sevoflurane-treated and untreated rats. d–e BrdU-positive cells on P35 in sevoflurane-treated and untreated rats. g,

TrkB 29.8% of control, P = 0.64; Fig. 3a–c). There was no significant difference between the C group and the 2.4% sevo group in these proteins.

Discussion

Sevoflurane is one of the most commonly used anesthetics in neonatal and pediatric anesthesia practice. The safety of the clinical application of sevoflurane in young children is still unclear. The present work aimed to find out whether exposure to a lower concentration (1.2%) of sevoflurane during early postnatal life impacts learning and memory ability later in life. In the current study, we selected 1.2% sevoflurane as the lower dose because this is the lowest sub-anesthetic dose that can prevent rats’ movement in response to a slight stimulus in neonatal rats and did not inhibit respiration. But beyond that, 1.2% sevoflurane is comparable to be used in the clinical setting. Using the MWM, we found that exposure of neonatal rats to 1.2% sevoflurane for 6 h improved their hippocampus- dependent learning and memory ability. In addition, changes were found in the number of BrdU-positive cells in the DG and the number of synapses, synaptic cleft width, and

h Summary data for the experiment at P10 and P35. Scale bar represents 50 μm. ***P < 0.001 versus C group; ****P < 0.0001 versus C group. C, control; 1.2% sevo, 1.2% sevoflurane-exposed group; 2.4% sevo, 2.4% sevoflurane-exposed group

postsynaptic density thickness in the hippocampus of the low-dose group. Our observations indicated that increased neurogenesis and synaptic plasticity in the hippocampus caused by low-dose sevoflurane might induce changes in neu- robehavioral function later in life.

The MWM test in the current study demonstrated that neo- natal exposure to 1.2% sevoflurane for 6 h could facilitate the spatial learning and memory ability of rats later in life. Our result was consistent with previous studies showing that neo- natal exposure to sevoflurane in rodents has no potential to harm their neurobehavioral function in adulthood (Callaway et al. 2012; Liang et al. 2010). More remarkably, Chen et al. found that a subclinical dose of sevoflurane could promote hippocampal neurogenesis in neonatal rats and facilitate den- tate gyrus-dependent learning (Chen et al. 2015). Furthermore, in vitro studies confirmed that a lower dose of sevoflurane could promote the self-renewal capacity and dif- ferentiation of cultured neural cells (Nie et al. 2013; Yang et al. 2017). Our results showed that a higher concentration of sevoflurane had no deleterious effects on learning and mem- ory. However, some studies have found apparently contradic- tory results, indicating that neonatal exposure to a high dose of sevoflurane in rodents and nonhuman primates produces

193

194

Fig. 3 Exposure to 1.2% sevoflurane increased BDNF, TrkB, PSD-95, and SYN levels in the hippocampus. a Representative immunoblots for the expression levels of BDNF and TrkB in the hippocampus 4 weeks after sevoflurane exposure. b, c Quantification of BDNF and TrkB normalized to β-actin (n = 5 per group). d Representative immunoblots for the expression levels of PSD-95 and SYN in the hippocampus 4 weeks after sevoflurane exposure. e, f Quantification of PSD-95 and

neurobehavioral defects persisting into adulthood (Haseneder et al. 2009; Jevtovic-Todorovic et al. 2003). We assume that this discrepancy is due to the use of different animal models and behavioral tests.

Recent studies have confirmed that a subclinical concen- tration of sevoflurane can enhance the proliferation of cultured neural stem cells (NSCs) (Nie et al. 2013). Our immunofluo- rescence histochemistry results also showed that 1.2% sevoflurane exposure increased the number of BrdU-positive cells at both P10 and P35, indicating a positive effect on neurogenesis. This finding is consistent with a previous study, which demonstrated that a sub-anesthetic dose of sevoflurane led to a significant increase in neurogenesis in neonatal rats (Chen et al. 2015). Furthermore, high concentrations and mul- tiple exposures to sevoflurane anesthesia during the neonatal period are considered to be associated with a reduction in neurogenesis (Fang et al. 2017; Lee et al. 2017). These results suggest that sevoflurane exerts dual effects on cognitive func- tion and neurogenesis depending on the dose and duration. In addition, a previous study showed that BDNF plays an impor- tant role in regulating the basal level of neurogenesis in the dentate gyrus (Lee et al. 2002), and those newly generated cells can mature into functional neurons in the mammalian brain (van Praag et al. 2002). In our study, compared with

Neurotox Res (2018) 34:188–197

SYN normalized to β-actin (n = 5 per group). Data are expressed as the mean ± SD. One-way ANOVA: BDNF F = 7.24, P = 0.0063; TrkB F = 17.92, P = 0.0002; PSD-95 F= 7.84, P = 0.0066; SYN F = 7.357, P = 0.0082; *P < 0.05 versus C group, **P < 0.01 versus C group, ***P < 0.001 versus C group, #P < 0.05 versus 2.4% sevo group, and ##P < 0.05 versus 2.4% sevo group. C, control; 1.2% sevo, 1.2% sevoflurane-exposed group; 2.4% sevo, 2.4% sevoflurane-exposed group

expression in the C group and the 2.4% sevo group, hippo- campal BDNF and TrkB protein expression in the 1.2% sevo group was prominently increased after exposure to sevoflurane for 28 days. These results suggest a critical role of BDNF signaling and neurogenesis in hippocampus- dependent learning and memory. In support of this possibility, BDNF signaling has been shown to improve cognitive func- tion (Hariri et al. 2003), and the process of neurogenesis may be a substrate for learning and memory (van Praag et al. 2002). We showed here that the improvement in learning and mem- ory after exposure to low-dose sevoflurane might be linked to neurogenesis via increases in the expression of BDNF and TrkB.

In addition, we observed that 6 h of low-dose sevoflurane exposure could augment the number of synapses and improve the synaptic ultrastructure in the hippocampus. Previous in- vestigations have reported that changes in the number and function of synapses can cause changes in synaptic plasticity and thereby affect learning and memory (Martin et al. 2000). Using TEM to analyze the number and the ultrastructure of synapses in the hippocampus, we found ultrastructural chang- es in hippocampal synapses after low-dose sevoflurane expo- sure. Furthermore, we analyzed the synaptic plasticity by mea- suring two synaptic marker proteins: synaptophysin (a

Neurotox Res (2018) 34:188–197

195

Fig. 4 Effects of sevoflurane on the synaptic ultrastructure of the hippocampus on P35 as visualized by TEM. a–c Representative images show the difference in the number of synapses per unit volume across the three groups (red arrows count the number of synapse). Scale bars = 1 μm. d–f Representative images show the differences in the synaptic

interfaces across the three groups. Scale bars = 200 nm. Magnification is × 13,500 (a–c) and × 37,000 (d–f). Summary data for the experiment are presented in g. **P < 0.001 versus C group; ##P < 0.01 versus 2.4% sevo group. C, control; 1.2% sevo, 1.2% sevoflurane-exposed group; 2.4% sevo, 2.4% sevoflurane-exposed group

presynaptic marker) and PSD-95 (a postsynaptic marker) (Head et al. 2009). The expression levels of synaptophysin and PSD-95 were significantly higher in the 1.2% sevo group than those in the other two groups. Recent studies have dem- onstrated that early exposure to high-dose of sevoflurane can induce neurotoxicity by decreasing the expression of synaptophysin and PSD-95 in the hippocampus (Wang et al. 2013; Zheng et al. 2013) and lead to greater synaptic loss and ultrastructural damage (Amrock et al. 2015). However, these studies noted that high-dose sevoflurane produced neurotoxic effects related to synaptic plasticity damage. Whether the

neuroprotective effect of low-dose sevoflurane is connected to synaptic changes remains unknown. The current study gives us an indication that low-dose sevoflurane exposure exerts a neuroprotective effect on the developing brain and that effect may relate to the improvement in synaptic plasticity.

BDNF is an important, well-studied neurotrophin that carries out a variety of neurotrophic and neuroprotective func- tions in the developing brain (Gray et al. 2013). A consider- able body of research indicates that the role of BDNF signal- ing in hippocampus-dependent learning and memory is

Table 2 Structural parameters of the synaptic interface in the hippocampus (N = 10 synapses)

Groups

C

1.2% sevo

2.4% sevo

P value

PSD thickness (nm) Synaptic cleft width (nm)

32.38 ± 4.69 20.00 ± 1.60

41.39 ± 4.32**** 13.31 ± 1.11****

31.76 ± 3.36 18.64 ± 1.714

P < 0.0001 P < 0.0001

Data are presented as the mean ± SEM N the number of synapses, PSD postsynaptic density ****P < 0.0001 vs. C (one-way ANOVA)

196

important both in humans and in experimental animals (Hariri et al. 2003; Lee et al. 2004; Tyler et al. 2002). Moreover, growing evidence suggests a more nuanced role for BDNF signaling in learning and memory, in which it acts primarily as a facilitator of synaptic plasticity and neuronal survival (Gray et al. 2013). This study showed that BDNF and TrkB protein expression in the hippocampus prominently increased after long-term exposure to low-dose sevoflurane compared with high-dose sevoflurane exposure or no exposure. It is plausible that increased hippocampal expression of BDNF and TrkB may play a mechanistic role in the behavioral per- formance improvement induced by low-dose sevoflurane. Therefore, we hypothesize that the observed improvement in neurogenesis and synaptic plasticity may be connected with BDNF expression and TrkB signaling.

Our study has some limitations. First, our experiment just observed a phenomenon and tendency that the improvement both in cognitive function and in neurogenesis/synaptic plas- ticity followed by 1.2% sevoflurane exposure; we did not demonstrate a definite connection between them. This limita- tion may weaken our evidence regarding the causal link be- tween cognitive function and hippocampal neurogenesis/ synaptic plasticity. This is the first step that we have observed the tendency, but the exact mechanism needs more investiga- tions, which will be also reported by related articles in future. Second, we did not investigate the effects of low-dose sevoflurane on other domains of cognitive function; we fo- cused only on learning and memory function because it is the major domain of cognitive function. However, the data from the current study suggest that low-dose sevoflurane exposure could facilitate learning and memory; this possibility merits further studies to undercover the underlying mechanisms.

In conclusion, our findings demonstrate that neonatal ex- posure to low-dose sevoflurane improves hippocampus- dependent learning and memory later in life. This effect may be connected to improved hippocampal neurogenesis and syn- aptic plasticity. If exposure of young patients to lower doses of sevoflurane can promote learning and memory, the selection of this anesthetic and dose range can serve as a new strategy to improve outcomes for children who must undergo anesthesia. However, further clinical studies will need to confirm this possibility.

Funding Information This work was supported by the grants from the National Natural Science Foundation of China (No. 81571032 to Xia Feng and No. 81701047 to Xue Zhou).

Compliance with Ethical Standards

The use of rats in this study was approved by the Institutional Animal Care and Use Committee at Sun Yat-sen University (Guangzhou, China). All experiments were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and ARRIVE guidelines.

Neurotox Res (2018) 34:188–197

Conflict of Interest The authors declare that they have no conflict of interest.

References

Amrock LG, Starner ML, Murphy KL, Baxter MG (2015) Long-term effects of single or multiple neonatal sevoflurane exposures on rat hippocampal ultrastructure. Anesthesiology 122(1):87–95. https:// doi.org/10.1097/ALN.0000000000000477

Callaway JK, Jones NC, Royse AG, Royse CF (2012) Sevoflurane anes- thesia does not impair acquisition learning or memory in the Morris water maze in young adult and aged rats. Anesthesiology 117(5): 1091–1101. https://doi.org/10.1097/ALN.0b013e31826cb228 Chen C, Shen FY, Zhao X, Zhou T, Xu DJ, Wang ZR, Wang YW (2015) Low-dose sevoflurane promotes hippocampal neurogenesis and fa- cilitates the development of dentate gyrus-dependent learning in neonatal rats. ASN neuro 7(2):175909141557584. https://doi.org/ 10.1177/1759091415575845

Fang F, Song R, Ling X, Peng M, Xue Z, Cang J (2017) Multiple sevoflurane anesthesia in pregnant mice inhibits neurogenesis of fetal hippocampus via repressing transcription factor Pax6. Life Sci 175:16–22. https://doi.org/10.1016/j.lfs.2017.03.003

Feng X, Liu JJ, Zhou X, Song FH, Yang XY, Chen XS, Huang WQ, Zhou LH, Ye JH (2012) Single sevoflurane exposure decreases neuronal nitric oxide synthase levels in the hippocampus of developing rats. Br J Anaesth 109(2):225–233. https://doi.org/10.1093/bja/aes121 Goa KL, Noble S, Spencer CM (1999) Sevoflurane in paediatric anaes-

thesia: a review. Paediatr Drugs 1:127–153

Gray JD, Milner TA, McEwen BS (2013) Dynamic plasticity: the role of glucocorticoids, brain-derived neurotrophic factor and other trophic factors. Neuroscience 239:214–227. https://doi.org/10.1016/j. neuroscience.2012.08.034

Hariri A, Goldberg T, Mattay V, Kolachana B, Callicott J, Egan M, Weinberger D (2003) Brain-derived neurotrophic factor val(66)met polymorphism affects human memory-related hippocampal activity and predicts memory performance. J Neurosci 23:6690–6694 Haseneder R, Kratzer S, von Meyer L, Eder M, Kochs E, Rammes G (2009) Isoflurane and sevoflurane dose-dependently impair hippo- campal long-term potentiation. Eur J Pharmacol 623(1-3):47–51. https://doi.org/10.1016/j.ejphar.2009.09.022

Head E, Corrada MM, Kahle-Wrobleski K, Kim RC, Sarsoza F, Goodus M, Kawas CH (2009) Synaptic proteins, neuropathology and cog- nitive status in the oldest-old. Neurobiol Aging 30(7):1125–1134. https://doi.org/10.1016/j.neurobiolaging.2007.10.001

Ishizeki J, Nishikawa K, Kubo K, Saito S, Goto F (2008) Amnestic concentrations of sevoflurane inhibit synaptic plasticity of hippo- campal CA1 neurons through gamma-aminobutyric acid-mediated mechanisms. Anesthesiology 108:447–456

Jevtovic-Todorovic V, Hartman RE, Izumi Y, Benshoff ND, Dikranian K, Zorumski CF, Olney JW, Wozniak DF (2003) Early exposure to common anesthetic agents causes widespread neurodegeneration in the developing rat brain and persistent learning deficits. J Neurosci : Off J Soc Neurosci 23(3):876–882

Kee N, Sivalingam S, Boonstra R, Wojtowicz JM (2002) The utility of Ki-67 and BrdU as proliferative markers of adult neurogenesis. J Neurosci Methods 115(1):97–105. https://doi.org/10.1016/S0165- 0270(02)00007-9

Lee J, Duan W, Mattson MP (2002) Evidence that brain-derived neuro- trophic factor is required for basal neurogenesis and mediates, in part, the enhancement of neurogenesis by dietary restriction in the hippocampus of adult mice. J Neurochem 82:1367–1375

Neurotox Res (2018) 34:188–197

Lee JL, Everitt BJ, Thomas KL (2004) Independent cellular processes for hippocampal memory consolidation and reconsolidation. Science (New York, NY) 304:839–843

Lee S, Chung W, Park H, Park H, Yoon S, Park S, Park J, Heo JY, Ju X, Yoon SH, Kim YH, Ko Y (2017) Single and multiple sevoflurane exposures during pregnancy and offspring behavior in mice. Paediatr Anaesth 27(7):742–751. https://doi.org/10.1111/pan.13139 Liang G, Ward C, Peng J, Zhao Y, Huang B, Wei H (2010) Isoflurane causes greater neurodegeneration than an equivalent exposure of sevoflurane in the developing brain of neonatal mice. Anesthesiology 112(6):1325–1334. https://doi.org/10.1097/ALN. 0b013e3181d94da5

Loepke AW, Soriano SG (2008) An assessment of the effects of general anesthetics on developing brain structure and neurocognitive func- tion. Anesth Analg 106(6):1681–1707. https://doi.org/10.1213/ane. 0b013e318167ad77

Lu B, Nagappan G, Guan X, Nathan PJ, Wren P (2013) BDNF-based synaptic repair as a disease-modifying strategy for neurodegenera- tive diseases. Nat Rev Neurosci 14(6):401–416. https://doi.org/10. 1038/nrn3505

Martin SJ, Grimwood PD, Morris RG (2000) Synaptic plasticity and memory: an evaluation of the hypothesis. Annu Rev Neurosci 23(1):649–711. https://doi.org/10.1146/annurev.neuro.23.1.649 Mellon RD, Simone AF, Rappaport BA (2007) Use of anesthetic agents in neonates and young children. Anesth Analg 104(3):509–520. https://doi.org/10.1213/01.ane.0000255729.96438.b0

Nie H, Peng Z, Lao N, Dong H, Xiong L (2013) Effects of sevoflurane on self-renewal capacity and differentiation of cultured neural stem cells. Neurochem Res 38:1758–1767

Payne RS, Akca O, Roewer N, Schurr A, Kehl F (2005) Sevoflurane- induced preconditioning protects against cerebral ischemic neuronal damage in rats. Brain Res

Sairanen M, Lucas G, Ernfors P, Castren M, Castren E (2005) Brain- derived neurotrophic factor and antidepressant drugs have different but coordinated effects on neuronal turnover, proliferation, and sur- vival in the adult dentate gyrus. J Neurosci : Off J Soc Neurosci 25: 1089–1094

Tao G, Luo Y, Xue Q, Li G, Tan Y, Xiao J, Yu B (2016) Docosahexaenoic acid rescues synaptogenesis impairment and long-term memory def- icits caused by postnatal multiple sevoflurane exposures. Biomed Res Int 2016:4062579

Tozuka Y, Fukuda S, Namba T, Seki T, Hisatsune T (2005) GABAergic excitation promotes neuronal differentiation in adult hippocampal progenitor cells. Neuron 47:803–815

Tyler WJ, Alonso M, Bramham CR, Pozzo-Miller LD (2002) From ac- quisition to consolidation: on the role of brain-derived neurotrophic factor signaling in hippocampal-dependent learning. Learning Mem (Cold Spring Harbor, NY) 9:224–237

van Praag H, Schinder AF, Christie BR, Toni N, Palmer TD, Gage FH (2002) Functional neurogenesis in the adult hippocampus. Nature 415(6875):1030–1034. https://doi.org/10.1038/4151030a

Wang SQ, Fang F, Xue ZG, Cang J, Zhang XG (2013) Neonatal sevoflurane anesthesia induces long-term memory impairment and decreases hippocampal PSD-95 expression without neuronal loss. Eur Rev Med Pharmacol Sci 17:941–950

Yang Z, Lv J, Li X, Meng Q, Yang Q, Ma W, Li Y, Ke ZJ (2017) Sevoflurane decreases self-renewal capacity and causes c-Jun N- terminal kinase-mediated damage of rat fetal neural stem cells. Sci Rep 7:46304. https://doi.org/10.1038/srep46304

Zheng H, Dong Y, Xu Z, Crosby G, Culley DJ, Zhang Y, Xie Z (2013) Sevoflurane anesthesia in pregnant mice induces neurotoxicity in fetal and offspring mice. Anesthesiology 118(3):516–526. https:// doi.org/10.1097/ALN.0b013e3182834d5d

197