Sevoflurane Exposure during the Critical Period Affects Synaptic Transmission and Mitochondrial Respiration but Not Long-term Behavior in Mice

MANY animal studies have reported that postnatal exposure to anesthetics may be neurotoxic.^1–10  Retrospective clinical studies assessing the neurotoxic effects of anesthetics have reported inconsistent outcomes,^11–13  and several randomized controlled trials are being performed.^14–16  Two trials have recently reported that brief exposures to general anesthesia during infancy do not increase the risk of adverse neurodevelopmental outcome.^14,16  Although reassuring, exposure to anesthetics may have negative effects at older ages, as may relatively longer or multiple exposure to anesthetics during neurodevelopment.

Importantly, rodent studies have reported different neurotoxic consequences from general anesthesia in a neurodevelopmental stage-dependent manner. Most rodent studies have focused on the first postnatal week, showing widespread neurodegeneration after exposure to general anesthetics. However, recent studies focusing on older rodents have reported that anesthesia during the critical synaptogenic period induces dendritic spine formation rather than widespread neuroapoptosis.^6–10 

Dendritic spines are postsynaptic actin-based protrusions that regulate the structure, function, and plasticity of excitatory synapses. Dendritic spine abnormalities and changes in excitatory/inhibitory synaptic transmission during the critical synaptogenic period may be associated with neurodevelopmental disorders, including intellectual disability and autism.^17,18  Since anesthetics induce dendritic spine formation during the critical synaptogenic period, this may explain results indicating an association between general anesthesia and neurodevelopmental disorders.^19–21  However, the effects of these newly developed spines on synaptic transmission and cognitive function have not yet been extensively studied. In addition, the mechanism behind anesthesia-induced spinogenesis during the critical period has not yet been determined. Mitochondrial energy may be essential for supporting the high metabolic requirements of spine formation after general anesthesia.²²  For example, isoflurane has been shown to attenuate ischemic–reperfusion stress by enhancing mitochondrial respiratory chain activity.²³  Because the mechanism by which anesthesia induces neuroapoptosis in rodents during the first postnatal week includes mitochondrial dysfunction, the effects of anesthetics on mitochondrial function may depend on the neurodevelopmental stage. Anesthesia may, therefore, induce widespread neurodegeneration via mitochondrial dysfunction or dendritic spine synthesis via mitochondrial activation.

In order to verify the possibility that early anesthesia may lead to changes in synaptic transmission after dendritic spine formation and induce long-term neurodevelopmental behavior impairments, this study was designed to evaluate the effects of sevoflurane, one of the most widely used anesthetic agents in pediatric patients, in postnatal day (PND) 16 to 17 mice. Also, to better understand the mechanism underlying anesthesia-induced spinogenesis, the effects of sevoflurane exposure on mitochondrial function were analyzed. This study found that exposure to sevoflurane (2.5% for 2 h) during the critical synaptogenic period alters mitochondrial function and induces a transient imbalance of excitatory/inhibitory synaptic transmission sex dependently but does not induce long-term behavioral changes.

C57BL/6J mice were maintained according to the Animal Research Requirements of the Korea Advanced Institute of Science and Technology, Daejeon, South Korea. All experiments were approved by the Committees on Animal Research at Korea Advanced Institute of Science and Technology (KA2010-18) and Chungnam National University Hospital, Daejeon, South Korea (CNUH-014-A0009). Mice were housed in a room maintained at 22°C, with a 12-h light/12-h dark cycle, and fed ad libitum. The mice were weaned at 3 weeks of age and housed in separate cages in groups of three to five.

Male PND 16 to 17 mice were randomly divided into two groups: an anesthesia group and a control group. Each mouse in the anesthesia group was placed in a 1-l plastic chamber and exposed to a constant flow of fresh gas (fraction of inspired oxygen [Fio₂] 1.0, 5 l/min) containing 2.5% sevoflurane for 2 h. Full recovery was confirmed 30 min later, while 100% oxygen was continuously provided. Fresh air was warmed to 36°C and humidified (MR730; Fisher & Paykel Healthcare, New Zealand). Mice in the control group were treated identically but without sevoflurane. Carbon dioxide and sevoflurane were monitored using an S/5 compact anesthetic monitor and an e-CAiO gas analyzer module (Datex-Ohmeda, Finland), respectively.

After 2 h of sevoflurane exposure, the mice were decapitated and trunk blood was collected. Blood gas was analyzed using iSTAT (Abbott, United Kingdom). The results of blood gas analysis are summarized in table 1.

Protein samples were obtained 0, 3, 6, and 9 h after sevoflurane or oxygen exposure. For ethical reasons, mice were exposed to 3% sevoflurane for 5 to 10 min before sampling. The mice were euthanized by decapitation, their brains were removed, and the whole cerebral cortex region was homogenized with a tissue grinder in radio immunoprecipitation assay lysis buffer (ELPIS-BIOTECH, Korea, 100 mM Tris–hydrochloride [pH 8.5], 200 mM NaCl, 5 mM EDTA, and 0.2% sodium dodecyl sulfate), containing a phosphatase and protease inhibitor cocktail, followed by centrifugation of the homogenized tissue samples at 15,000g for 20 min at 4°C. The supernatants were collected, and their protein concentrations were measured using the Bradford assay (Bio-Rad, USA). Samples (20 µg) were electrophoresed on sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels (ELPIS- BIOTECH), and the separated proteins were transferred at 200 mA for 2 h onto nitrocellulose membranes (pore size, 0.2 µm; Amersham Protran®, GE Healthcare, United Kingdom). The membranes were blocked for 1 h with Tris-buffered saline with Tween 20 (made of 10 mM Tris–hydrochloride [pH 7.6], 150 mM NaCl, and 0.1% Tween 20), containing 3% bovine serum albumin (BSA), followed by incubation with primary antibodies and the appropriate secondary antibodies coupled to horseradish peroxidase. Specific antibody-labeled proteins were detected using the enhanced chemiluminescence system (WEST-ZOL plus; iNtRON BioTechnology, Korea). Primary antibodies included antibodies to postsynaptic density 90 (PSD95; Neuromab, USA), NDUFB8 (a mitochondrial complex I subunit; Santa Cruz Biotechnology, USA), poly (adenosine diphosphate [ADP]-ribose) polymerase (Santa Cruz Biotechnology), caspase 3 (Cell Signaling Technology, USA), and actin (Santa Cruz Biotechnology). Antibodies against GluA1 (1193) and GluA2 (1195) have been described previously.²⁴ 

All behavioral tests were performed as previously described,^3,25,26  using age-matched adult male mice, aged 2 to 3 months, during the active (night) period of the circadian cycle. Behavior tests were performed with two cohorts of mice. Light-dark box test, open field test, three-chamber test, and fear test were performed with the first cohort. Elevated plus maze and novel object recognition tests were performed with the second cohort. All procedures were recorded and later analyzed manually or with a tracking software (Ethovision XT; Noldus Information Technology, The Netherlands).

Mice were placed in the center of a lighted compartment (800 to 900 lux), facing away from the dark compartment. Total duration spent in the light chamber (%) was measured for 10 min.

The elevated plus maze consisted of two open arms, two closed arms, and a center zone elevated to a height of 50 cm above the floor. Mice were placed in the center of the maze facing an open arm and were allowed to explore the maze for 10 min.

Mice were placed in the center of an open field box, measuring 40 × 40 × 40 cm. General activity was analyzed for 1 h, with the center zone defined as the area greater than or equal to 10 cm from each wall.

A previously described three-chamber apparatus was used in which each side chamber contained a plastic cage. The experiment consisted of three sessions, each lasting 10 min. During the first session, the mice were confined to the center chamber; during the second session, the mice were allowed to freely explore all three chambers. After these two habituation sessions, a novel object, consisting of a blue cube measuring 3.0 × 3.0 × 4.0 cm, and a novel stranger, consisting of a mouse of the same age and sex, were placed inside each side chamber cage. The subject mouse was allowed to freely explore the apparatus for 10 min. The positions of the object and stranger were alternated between tests to prevent side preference. Preference index (PI) was calculated by using the time spent in each chamber (Mo, time spent in the mouse chamber; Ob, time spent in the object chamber): PI (%) = (Mo − Ob)/(Mo + Ob) × 100.

Mice were placed in the center of a home cage without bedding for 20 min. Cages were changed for each experiment. Grooming was manually measured for 10 min after a 10-min habituation period.

Digging was defined as the coordinate movement of two fore- or hind legs to displace bedding. Mice were placed in a standard home cage with 5 cm of wood-chip bedding and recorded.

Mice were habituated to the open field box for 30 min for two consecutive days. On the third day (sample phase), the mice were placed in the open field box and allowed to explore two identical objects for 10 min. Four hours later (test phase), one of the objects was replaced with a novel object, and the mice were again allowed to explore each object freely for 10 min. Object exploration time was defined as the time each mouse’s nose was within 2 cm of the object. Object replacement was alternated to prevent side preference. PI was calculated with the exploration time (new, exploration time for the novel object; old, exploration time for the previous object): PI (%) = (new − old)/(new + old) × 100.

Fear tests were performed in a conditioning chamber containing a metal grid floor (Coulbourn Instruments, USA) within a sound-attenuating chamber. The conditioned stimulus (CS) was a 20-s, 3-kHz, 80-dB tone, and the unconditioned stimulus was a 1-mA, 1-s electric shock delivered at the end of the CS. After a 5-min habituation period, mice received a CS and unconditioned stimulus three times (60-s interval). Twenty-four hours later, the mice were placed in the same chamber for 5 min (contextual fear test). After an additional 24 h, the mice were placed in a different context (mint-scented, white circular plastic chamber) and exposed to the 3-kHz, 80-dB tone for 3 min after a 5-min habituation period (cue test). Freezing behavior was automatically measured with Freezeframe software (Coulbourn Instruments).

Whole-cell patch-clamp recordings of layer II and III pyramidal neurons in the medial prefrontal cortex (mPFC) were performed 6 h or 5 days after exposure to sevoflurane or oxygen, as described previously.²⁵  Coronal slices of the mPFC (300 µm) were prepared using a VT1200S vibratome (Leica, Switzerland) in an ice-cold dissection buffer (212 mM sucrose, 25 mM NaHCO₃, 5 mM KCl, 1.25 mM NaH₂PO₄, 10 mM d-glucose, 2 mM sodium pyruvate, 1.2 mM sodium ascorbate, 3.5 mM MgCl₂, and 0.5 mM CaCl₂) continuously aerated with 95% O₂/5% CO₂. The slices were transferred to a chamber filled with artificial cerebrospinal fluid (125 mM NaCl, 25 mM NaHCO₃, 2.5 mM KCl, 1.25 mM NaH₂PO₄, 10 mM d-glucose, 1.3 mM MgCl₂, and 2.5 mM CaCl₂), aerated with 95% O₂/5% CO₂, and warmed to 32°C for 30 min for recovery. Glass capillaries were filled with an internal solution (117 mM CsMeSO₄, 10 mM tetraethylammonium chloride, 8 mM NaCl, 10 mM HEPES, 5 mM QX-314-Cl, 4 mM Mg-adenosine triphosphate [ATP], 0.3 mM Na-guanosine triphosphate, and 10 mM EGTA for miniature excitatory postsynaptic current [mEPSC] recordings; 115 mM CsCl, 10 mM tetraethylammonium chloride, 8 mM NaCl, 10 mM HEPES, 5 mM QX-314-Cl, 4 mM Mg-ATP, 0.3 mM Na-guanosine triphosphate, and 10 mM EGTA for miniature inhibitory postsynaptic current [mIPSC] recordings), and recordings were made using a MultiClamp 700B amplifier (Molecular Devices, USA) under visual control (BX50WI; Olympus, Japan). Data were acquired with Clampex 9.2 (Molecular Devices) and analyzed using Clampfit 9 (Molecular Devices).

Mitochondria were isolated from the cerebral cortex as described previously.^27,28  Briefly, the cerebral cortex was homogenized in a mitochondrial isolation buffer (70 mM sucrose, 210 mM mannitol, 5 mM HEPES, 1 mM EGTA, and 0.5% [w/v] fatty acid–free BSA [pH 7.2]) with a Teflon-glass homogenizer (Thomas Scientific, USA). After centrifuging at 600g for 10 min at 4°C and at 17,000g for 10 min at 4°C, the mitochondrial fraction was resuspended in mitochondrial isolation buffer. Protein concentration was measured with the Bradford assay (Bio-Rad). Aliquots containing 20 µg protein were each diluted with 50 µl mitochondrial assay solution (70 mM sucrose, 220 mM mannitol, 10 mM KH₂PO₄, 5 mM MgCl₂, 2 mM HEPES, 1 mM EGTA, 0.2% [w/v] fatty acid–free BSA, 10 mM succinate, and 2 μM rotenone [pH 7.2]) and seeded in an XF-24 plate (Seahorse Bioscience, USA). The plates were centrifuged at 2,000g for 20 min at 4°C using a swinging bucket microplate adaptor (Effendorf, Germany). After adding 450 µl mitochondrial assay buffer, the XF-24 plate (Seahorse Bioscience) was maintained at 37°C for 8 to 10 min and then transferred to the Seahorse XF-24 extracellular flux analyzer (Seahorse Bioscience). Mitochondrial function was determined by measuring the oxygen consumption rate (OCR). Real-time readings were taken for five stages: stage I, basal level; stage II, addition of ADP to measure ATP production; stage III, addition of oligomycin, a mitochondrial oxidative phosphorylation (OXPHOS) complex 5 inhibitor, to measure protein leakage; stage IV, addition of the mitochondrial OXPHOS complex 4 inhibitor carbonyl cyanide m-chlorophenyl hydrazine to measure maximal respiration; and stage V, addition of the mitochondrial OXPHOS complex 3 inhibitor antimycin A to measure nonmitochondrial respiration. OCR was automatically calculated and recorded using the Seahorse XF-24 software (Seahorse Bioscience).

Behavioral studies were performed and analyzed blindly. Since brain slices, Western blot samples, and mitochondrial samples of both groups were prepared simultaneously, these were prepared nonblindly in order to avoid confusion. However, the experiments were all performed and analyzed in a blinded manner. Sample size for tests was based on previous experiences or as previously described.^3,25,29  There are no lost or missing data in our study. All statistical analyses were performed using R statistical software (3.1.2: R Core Team, Austria) and GraphPad Prism 5 software (GraphPad Software, USA). All continuous variables were tested to determine whether they met conditions of normality and homogeneity of variance. Student’s t tests were performed when both conditions were met; Welch t test was performed when homogeneity of variance was unmet, and the Kruskal–Wallis test was performed if normality was unmet. Repeatedly measured data were analyzed via mixed effect modeling with a fixed effect for slope. Correlated random intercept and random slope term were incorporated into the model. P < 0.05 was considered statistically significant. Bonferroni post hoc testing was used for multiple comparisons. See the table (Supplemental Digital Content 1, https://links.lww.com/ALN/B349) for summary of statistics.

Although several studies have shown that brief to moderate exposure to sevoflurane induces dendritic spine formation,^6–10  the effects of spinogenesis on synaptic transmission had not been analyzed. We, therefore, evaluated the acute and long-term effects of sevoflurane on synaptic transmission by measuring mEPSCs/mIPSCs of layer II and III pyramidal neurons in the mPFC. Brain slices of the mPFC were prepared 6 h or 5 days after sevoflurane exposure. Although there was no change in mEPSC amplitude, its frequency was significantly increased 6 h after exposure to anesthesia (fig. 1, A and B), indicating that sevoflurane induces functional dendritic spines, leading to increased excitatory synaptic transmission. Unexpectedly, the frequency of mIPSCs was significantly decreased 6 h after sevoflurane exposure (fig. 1, C and D). However, recordings of mEPSCs/mIPSCs 5 days after sevoflurane exposure were comparable, indicating that sevoflurane does not have long-term effects on synaptic transmission (fig. 1, E to H).

Expression levels of PSD-95, a scaffold protein abundantly expressed in the postsynapse (fig. 2, A to E), and subunits of the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor, an ionotropic transmembrane glutamate receptor (fig. 2, A to D, F, and G), were analyzed over time with protein samples of the whole cerebral cortex after exposure to sevoflurane. The expression level of subunit GluA2 was significantly increased 3 and 6 h after sevoflurane exposure.

Anesthetic agents have been reported to induce widespread neuronal cell death during the prenatal and immediate postnatal stages but not during later stages of development. Exposure of PND 16 to 17 mice to sevoflurane did not increase the expression of cell death markers such as cleaved caspase 3 and cleaved poly (ADP-ribose) polymerase (fig. 2, A to D).

Changes in dendritic spines and imbalances in excitatory/inhibitory synaptic transmission during the critical period have been shown to be associated with neurodevelopmental disorders. A series of experiments were, therefore, performed to evaluate the long-term behavioral effects of sevoflurane exposure. Sevoflurane-treated mice showed normal activity and anxiety levels in the open field test (fig. 3, A and B). Their anxiety level was comparable to that of the control group in the light-dark box and elevated plus maze tests (fig. 3, C and D). The sociability level, which has been shown to decrease in autism model mice, was also normal in the three-chamber test (fig. 3, E to G). Learning and memory were normal in the novel object recognition and fear chamber tests (fig. 3, H to K).

Although the mechanism underlying neuronal cell death during the prenatal and immediate postnatal periods remains unclear, anesthesia-induced mitochondrial dysfunction and consequent overproduction of reactive oxygen species have been reported to play important roles. However, the effects of anesthetic agents on mitochondrial function at the age of PND 16 and 17 have not been determined. Because neuronal cell death does not occur during this period, mitochondrial function may be less affected. As dendritic mitochondria were shown to be deeply involved with the process of dendritic spine formation,²²  mitochondrial function was evaluated by measuring OCR after sevoflurane exposure (fig. 4, A to H). OCR was significantly reduced immediately after anesthesia but recovered 3 h later (fig. 4, A to D). Interestingly, 3 h after sevoflurane exposure, OCR was significantly increased by treatment with ADP (stage II; fig. 4, C and D). Even basal OCR was significantly increased 9 h after anesthesia (fig. 4, G and H). Time-dependent changes in mitochondrial activity were also analyzed by measuring the expression level of NDUFB8, a subunit of OXPHOS complex 1. Although anesthesia induced a similar trend of expression level changes after anesthesia, these changes were not statistically significant (fig. 4, I and J).

Male mice have been suggested to be more sensitive to the neurotoxic effects of anesthetics.³⁰  Also, interpretation of behavioral studies can be complex in female mice due to hormonal factors.³¹  However, in order to make a general conclusion about how sevoflurane affects neurodevelopment, female mice were also analyzed after an identical sevoflurane exposure. Unexpectedly, unlike male mice, there was no increase in mEPSC frequency 6 h after sevoflurane exposure (fig. 5, A and B). However, whole cerebral cortex samples of female mice show identical increases of excitatory synaptic protein expression 6 h after sevoflurane exposure (fig. 5, C and D). Most interestingly, the frequency of mIPSC was significantly increased 6 h after sevoflurane exposure (fig. 5, E and F), opposite to that of male mice that displayed a significant decrease in mIPSC frequency (fig. 1B). Such changes in synaptic transmission did not affect adult female behavior in the open field test, three-chamber test, and fear chamber test (fig. 5, G to L).

While initial animal studies focused on the early postnatal brain, many studies now focus on a later stage, which may represent the critical period in humans.^6–10  These studies suggest that anesthetic agents induce dendritic spine formation, not widespread neurodegeneration. However, most of these studies focused on structural and morphologic changes in dendrites. To date, few studies have evaluated the function and long-term behavioral consequences of such spinogenesis.

Dendritic spines are actin-based protrusions on which most excitatory synapses form.³²  Dendritic spines can be distinguished from filopodia, another type of dendritic protrusion that does not form excitatory synapses, by their characteristic morphology. Most studies evaluating dendritic changes after exposure to anesthetics have, therefore, assessed dendritic spines morphologically.^7,10  Electron microscopy showed the presence of excitatory synapses in propofol-induced dendritic spines, suggesting that these newly formed spines were functional.⁷  Although functionality has been assessed indirectly through calcium imaging,⁹  overall changes in excitatory synaptic transmission resulting from anesthesia-induced spinogenesis have not been determined. By analyzing the mEPSCs in layer II and III pyramidal neurons in the mPFC, we found that excitatory synaptic transmission significantly increased 6 h after exposure of PND 16 to 17 mice to sevoflurane in male mice. This increase was not long-lasting as synaptic transmission had normalized when analyzed 5 days after exposure to anesthesia. This finding is in good agreement with a recent study, which found no long-term changes in synaptic transmission 3 months after exposure to sevoflurane for 30 min.⁸  Unexpectedly, female mice did not show increased excitatory synaptic transmission. However, the expression level of synaptic proteins was also significantly increased after sevoflurane exposure in female mice. Sevoflurane exposure also increased the expression level of PSD-95 in the thalamus, a region vulnerable to anesthesia-induced neurodegeneration in both male and female mice (Supplemental Digital Content 2, https://links.lww.com/ALN/B350). Additional female mice studies focusing on different cell layers of the cortex or different brain regions are needed to explain the discrepancy between electrophysiology and Western blot results.

Inhibitory synapses have been known to exist mostly on the dendritic shaft. However, recent studies now show that 30% inhibitory synapses reside on dendritic spines in adult mice.³³  Thus, in order to evaluate the effects of early anesthesia on synaptic transmission, we also analyzed changes in mIPSCs. Our results show that inhibitory synaptic transmission is also transiently altered after sevoflurane exposure. Even more interesting was the fact that such changes were observed in the opposite direction between male and female mice (fig. 1, C and D and fig. 5, E and F). Previous studies have shown that sexually dimorphic gene expression and gonadal steroid hormones promote sex differences during neurodevelopment.³⁴  Such differences have been suggested to be involved with the sex-dependent incidence of neurodevelopmental disorders.³⁵  Although additional studies are needed to confirm sexual differences of anesthetic neurotoxicity, it is possible that such sex-dependent differences in neurodevelopment may also affect the consequences of anesthetic neurotoxicity.

Although it is reassuring that changes in excitatory/inhibitory synaptic transmission are not long lasting, even transient changes during the critical period may affect neurodevelopment.¹⁸  Several clinical studies have found that early exposure to anesthetic agents may influence neurodevelopment.^21,36  In contrast, a recent animal study found that a 30-min exposure to sevoflurane did not induce deficiencies in memory.⁸  Our results show that a 2-h sevoflurane exposure does not induce long-term behavioral changes in general activity, anxiety, sociability, learning, and memory. However, as clinical studies have suggested that multiple exposures to anesthesia may have a greater effect on development, studies assessing the consequences of multiple exposures are needed.^19,37,38 

Studies have also focused on the mechanisms underlying the inadvertent effects of general anesthetics in the developing brain. Although still unclear, mitochondrial respiratory chain dysfunction and the increase in reactive oxygen species resulting from dysfunctions in mitochondrial complexes I and III may play important roles in the widespread neuroapoptosis that occurs when rodents are exposed to anesthesia during the first postnatal week.^39,40  Anesthesia-induced neurodegeneration may, therefore, be prevented by mitochondrial protection.^41,42  However, anesthetics have also been shown to increase mitochondrial complex activity. For example, exposure to propofol has been shown to increase mitochondrial complex I activity, leading to cardioprotection,⁴³  and exposure to isoflurane protects cardiac muscle from ischemia-induced damage by enhancing mitochondrial complex activity.²³  To date, no study has assessed the effects of anesthetics on mitochondrial function in PND 16 to 17 mice, a period when anesthesia has been shown to induce dendritic spine formation. Dendritic mitochondria play a vital role in spinogenesis, with mitochondrial fission facilitating the dendritic localization of mitochondria.^22,44  Fissional changes in mitochondria increase the number of mitochondria, satisfying the high energy demand during spinogenesis. Anesthetics may actually increase dendritic mitochondrial activity, thus allowing spinogenesis to occur. Our measurements of mitochondrial OCRs show that exposure to sevoflurane during the critical period induced time-dependent changes in mitochondrial activity in male mice. Mitochondria are hypoactive directly after anesthesia but become significantly more active within hours. Further studies are required to determine the precise role of mitochondrial activity after sevoflurane exposure during the critical synaptogenic period.

This study has several limitations, including the lack of respiratory control during sevoflurane exposure. Although we were able to avoid hypoxia with an Fio₂ of 1.0, blood gas analysis showed a significant increase in carbon dioxide and a reduction in pH. Anesthesia-induced alterations of systemic homeostasis in small rodents have been shown to cause long-lasting cognitive dysfunctions.⁴⁵  Although our results are in agreement with studies showing normal physiologic parameters during anesthesia,^6,8  with no increases in neuronal cell death and cognitive impairment, sevoflurane-induced loss of systemic homeostasis may have affected other parameters, such as mitochondrial activity. Another limiting factor is the use a high Fio₂ (1.0) during sevoflurane exposure. Since high concentrations of oxygen itself can be toxic,⁴⁶  they may have affected the changes during anesthesia.

In conclusion, this study shows that sevoflurane exposure during the critical period induces mitochondrial hyperactivity and an acute, short-term imbalance of excitatory/inhibitory synaptic transmission without long-lasting behavioral consequences. Further studies are needed to confirm the sexual differences and to define the role of mitochondrial activity changes during anesthesia-induced spine formation.

Supported by the National Research Foundation of Korea (grant nos. NRF-2015R1C1A1A01054659, 2014R1A1A1037655, and HRF-2014R1A2A1A11051231; Daejeon, Korea) and the research fund of Chungnam National University (Daejeon, Korea).

The authors declare no competing interests.