Xenon Pretreatment Attenuates Anesthetic-induced Apoptosis in the Developing Brain in Comparison with Nitrous Oxide and Hypoxia

• ❖ Preconditioning by transient hypoxia or anesthetic exposure can protect the brain and heart from subsequent ischemia

• ❖ Whether preconditioning is effective against cell death from neonatal exposure to anesthetics is not known

• ❖ In neonatal rats, preconditioning with xenon protected against isoflurane-/N²O-induced cell death and learning deficits, and preconditioning with transient hypoxia worsened it

THE ethanol-induced fetal alcohol syndrome provided evidence that anesthetics may be capable of damaging the brain during the synaptogeneic phase of neurodevelopment because of the similarity of their properties at the N -methyl-d-aspartic acid (NMDA) and γ-aminobutyric acid (GABA) A receptors.1–5Jevtovic-Todorovic et al.  6showed that administration of a combination of midazolam, nitrous oxide, and isoflurane to 7-day-old rats produced widespread apoptotic neurodegeneration in the brain and subsequent impairment of memory or learning that persisted into early adulthood.

Drugs acting on NMDA and/or GABA^Areceptors during synaptogenesis, also known as brain growth spurt,7can have such profound effects because the cells, in the developing central nervous system, are differentiating and migrating to target zones and establishing synaptic contact with neighboring neurons, thereby producing functional neuronal circuits. Neurotransmitters, including GABA and glutamate and their receptors, play a fundamental role during this process as they are essential for neuronal migration, dendritic filopodia stabilization, synaptic development, and stabilization.8 

In rodents, synaptogenesis is predominantly a postnatal event and is most intense just before birth and in the first 2 postnatal weeks. In humans, synaptogenesis begins at 6 months' gestation and ends a few years after birth.7We sought to determine whether anesthetic-induced neurotoxicity could be attenuated by preconditioning. Preconditioning (also referred to as pretreatment in this study) is a ubiquitous mechanism whereby an intervention enhances the tolerance of the organism to injury-inducing conditions. We sought to determine whether pretreatment strategies can be effective against anesthetic-induced neurotoxicity. Ischemia has been used as the stimulus in most studies of preconditioning; however, other interventions can also stimulate preconditioning, for example, hyperoxia, oxidative stress, hyperthermia, heat shock, and anesthetics including xenon.9,10We postulated that pretreatment may protect the brain of neonatal rats from anesthetic drug-induced apoptosis.

This study was approved by the Home Office, London, United Kingdom, and conforms to the United Kingdom Animals (Scientific Procedures) Act, 1986.

Previously, we showed that pretreatment with 70% xenon for 2 h can attenuate hypoxic–ischemic-induced injury in a model of asphyxia in rats.10The neuroprotective properties of hypoxic pretreatment (8% oxygen for 2 h) are also well established,11,12which we confirmed in our pilot studies (data not shown).

Seven-day-old Sprague–Dawley rat pups were exposed to 70% xenon or 70% N²O balanced with oxygen or 8% oxygen balanced with nitrogen for 2 h in six purpose-built multi-chambers (1–2 pups/chamber). To reduce the wastage of costly xenon, it was delivered in a customized closed-circuit system, in which carbon dioxide was absorbed by soda lime and water vapor with silica gel. Twenty-four hours after pretreatment, rats were exposed to the anesthetic combination of 70% N²O + 0.75% isoflurane for 6 h. To precisely control the body temperature of the pups, all exposure chambers were partially submerged in a water bath, and the water temperature was adjusted to obtain a desired brain temperature of 37°C. This was performed with one pup brain being planted with a temperature probe, and the brain temperature was measured with a telemetry temperature monitoring system (VitalView; Mini-Mitter, Bend, OR). This pup was excluded for further data analysis. Oxygen, nitrous oxide, isoflurane, and xenon were regularly monitored with a gas monitoring system (Datex-Ohmeda, Bradford, United Kingdon; a 439XE monitor, Air Products, London, United Kingdom) via  a sampling tube inserted into the experimental chamber.

Rats were killed with an intraperitoneal injection of 100 mg/kg⁻¹sodium pentobarbital 48 h after pretreatment and perfused with a fixative.10The brains were removed and kept at 4°C overnight in paraformaldehyde. The fixed brains were then transferred to a solution of 30% sucrose with phosphate buffer and 1% sodium azide and were refrigerated until they were immunostained for the apoptotic marker caspase-3.

Like the Morris water maze, trace fear conditioning is also a standard method to measure hippocampal-dependent memory,13and hence, this paradigm was used in the current study to assess cognitive function. Cognitive assessment was performed in a separate cohort of animals 40 days after anesthetic exposure. The conditioning chamber (30 cm × 24 cm × 21 cm; Med Associates Inc., St. Albans, VT) had a white opaque back wall, aluminum sidewalls, and a clear polycarbonate front door. The conditioning box had a removable grid floor and waste pan. Between each rat, the box was cleaned with an almond-scented solution and dried thoroughly. The grid floor contained 36 stainless steel rods (diameter, 3 mm) spaced 8 mm center to center, which made contact with a circuit board through which a scrambled shock was delivered. During training and context testing, a standard high efficiency particulate air filter provided a background noise of 65 db. On day 1 (41 days after anesthetic exposure) (acquisition), all animals received six cycles of 214 s of the trace fear conditioning paradigm. The tone was presented for 16 s (2 kHz) followed by a trace interval of 18 s and subsequent foot shock (2 s, 0.85 mA). The rats were removed from the conditioning chamber 198 s after the last shock and returned to their home cage. The total time of the acquisition protocol was 26 min. Working memory during acquisition was defined as the time spent immobile after a shock divided by the intertrial interval.

On the next day (context test), the trained rats were placed for 8 min to the same acquisition environment but without unconditional stimulus (the foot shock) or conditional stimulus (the tone). The percentage of time an animal froze during the 8-min observation periods was calculated as the number of observations judged to be freezing divided by the total number of observations in 8 min (i.e. , 60 observations).

On the third day, the tone test was performed. Rats were placed in modified environment (increased lighting, flat floor, and triangular side walls were inserted along with a cinnamon-scented solution) where they were allowed to acclimatize for 192 s before being presented to three cycles of a 16-s tone followed by an intertrial interval of 240 s. Freezing time was automatically scored using VideoFreeze software (Med Associates Inc., Burlington, VT). The overall percentage of freezing time (context results) and the area under curve (tone test) were derived from plots between the percentage of freezing time and trial time in the tone test and were used for statistical comparison.

The overall percentage of freezing time was used to assess hippocampal memory during the context test, whereas for the tone test, the area under curve after tone presentation was derived from the graph, which plotted the percentage of freezing time against trial time.

Rat pups were killed 30 h after pretreatment with or without anesthetic insult. The brains were removed immediately, and the cortices were separated and kept at −80°C. The samples were then lysed in cell lysis buffer (pH 7.5, 20 mm Tris-HCl, 150 mm NaCl, 1 mm Na²DTA, 1 mm ethylene glycol tetraacetic acid, 1%trition, 2.5 mm sodium pyrophosphate, 1 mm β-glycerophosphate, 1 mm Na³VO⁴, 2 mm dl-dithiothreitol, 1 mm phenylmethanesulfonyl, and 1 μg/ml leupeptin). The homogenized sample was centrifuged at 3,000g  for 10 min at 4°C to remove cellular debris. The supernatant was further centrifuged twice, initially at 11,000g  for 20 min at 4°C for Western blotting of Bcl-2 and cleaved caspase-3 and a second time at 20,000g  for 45 min at 4°C to separate cytosolic and mitochondrial material for Western blotting of cytosolic cytochrome C . Protein concentration of the supernatant was determined using the Bradford Protein Assay (Bio-Rad, Herts, United Kingdom) and 20 μg per sample was used for Western blot studies. After electrophoresis, proteins were electrotransferred onto a nitrocellulose membrane (Hybond ECL, Amersham Biosciences, Buckinghamshire, United Kingdom). The membrane was probed with primary antibodies for Bcl-2, cleaved caspase-3 (New England Biolabs, Hitchin, Hertfordshire, United Kingdom), cytochrome C  (BD Biosciences Pharmingen, Oxford, United Kingdom), P53, and mitogen-activated protein kinases (New England Biolabs), followed by relevant species-derived horseradish peroxidise-conjugated secondary antibody for 1 h at room temperature. Thereafter, the membrane was reprobed with housekeeping protein α-tubulin antibody (Sigma, Poole, United Kingdom). The protein bands were visualized with enhanced chemiluminescence (New England Biolabs) with hyperfilm development (Amersham Biosciences). Densitometric analysis with normalization relative to α-tubulin was performed, and data are presented as protein expressed relative to control.

Organotypic hippocampal slice preparation was reported previously.10Briefly, neonatal rat pups at postnatal day 8 were anesthetized using 0.5% isoflurane until quiescent and killed by decapitation. The brains were quickly dissected and placed in an ice cooled (4°C) dissection solution, and excess tissues (cerebellum, olfactory bulbs, and meninges) were removed. The brains were cut into 400-μm saggital slices using a Mclllwain Tissue Chopper (Mickle Laboratory, Cambridge, United Kingdom). The slices with hippocampus were selected and placed on 30-mm-diameter semiporous cell culture inserts (Falcon, Becton Dickinson Labware, Millipore, Bedford, MA). The tissues were then cultured for 24 h in humidified air at 37°C and enriched with 5% CO²to allow them to recover from the trauma induced by the extraction. All subsequent gas exposures for pretreatment and subsequent anesthetic insults were similar to the in vivo  studies (see above, this section) and occurred in a specially constructed exposure chamber.10The gas mixtures consisting of one of the following: 70% xenon, 70% N²O with 5% CO²and 20% O²balanced with nitrogen or oxygen, and 8% oxygen with 5% CO²balanced with nitrogen, were warmed by a water bath and delivered in the headspace above the slices by a standard anesthetic machine at 2–3 l·min⁻¹; the concentrations were closely monitored with an S/5 spirometry module (Datex-Ohmeda, Bradford, United Kingdom) and a 439XE monitor (Air Products, London, United Kingdom). The gases were delivered for 3–4 min; thereafter, the chambers were sealed and placed in an incubator at 37°C for 2 h (Galaxy R Carbon Dioxide Chamber, Wolf Laboratories, York, United Kingdom). After exposure, the slices were returned to the incubator in humidified air at 37°C, enriched with 5% CO², for a further 12 h to allow for caspase-3 expression and were then fixed overnight in 4% paraformaldehyde.

The brain samples from in vivo  experiments or cultured slices from in vitro  experiments were immersed in 30% sucrose with phosphate buffer and 1% sodium azide and refrigerated until they shrunk. They were then frozen and thereafter sliced in a coronal axis into 30-μm-thick sections using a cryostat (Bright Instrument Company Ltd., Huntingdon, United Kingdom). The sections relative to bregma-2 in vivo  and the inner sections in vitro  were stained in situ  with cleaved caspase-3, a well-established marker of neuronal apoptosis in the developing brain.14,15Briefly, the slices were quenched in a solution of 70% methanol and 0.3% hydrogen peroxide for 30 min and then incubated in 3% blocking phosphate-buffered saline solution containing 0.3% Triton-X (Promega Corporation, Madison, WI) and 1% goat serum (Vector Laboratories Inc., Burlingame, CA) for 1 h at room temperature. After overnight incubation at 4°C with rabbit anticleaved caspase-3 antibody (1:1500) (New England Biolabs, Hertfordshire, United Kingdom), the sections were incubated with biotinylated goat antirabbit IgG (1:200; Chemicon International, Temecula, CA) for 1 h at room temperature followed by Avidin-Biotin-peroxidase complex (Vector Laboratories Inc.) for 1 h. The positive caspase-3 cells were visualized with 3,3′-diamino-benzidine solution (Vector Laboratories Inc.). Thereafter, the slices were mounted onto Super Plus-coated glass slides and dried in a dark area. They were then dehydrated with alcohol, cleared with xylene, and covered with glass before light microscopy.

Photomicrographs were taken of the whole hemisphere and hippocampus for in vivo  experiments and of the hippocampal cornu ammonis area for in vitro  experiments with a BX-60 light microscope (Olympus, Southall, United Kingdom) and an Axiocam digital camera (Zeiss, Göttingen, Germany). The images were then printed, and an investigator blinded to the experimental protocol counted the number of 3,3′-diaminobenzidine stained (black) cell bodies in the desired areas. Counting was performed on slices from the whole cortex and and from the hippocampal cornu ammonis 1–3 subregions for in vitro  experiments.

All results are continuous change data, and the normal distribution can be assumed albeit a relatively small sample size. Thus, the results were expressed as mean ± SD. Statistical analysis was performed by one-way analysis of variance followed by post hoc  Student–Newman–Keuls test for comparison where appropriate using an interactive statistical package (London, United Kingdom). P < 0.05 was considered statistically significant.

There were some casapse-3-positive cells in the cortex in the naive control (fig. 1A). However, the combined administration of N²O and isoflurane produced a significant increase in the number of caspase-3-positive cells (fig. 1B). Xenon pretreatment significantly decreased the number of caspase-3 cells induced by the administration of the combination of N²O + isoflurane (fig. 1C); N²O pretreatment did not have any significant effect (fig. 1D). In contrast, hypoxic pretreatment significantly enhanced the cellular apoptosis induced by the combination of N²O + isoflurane (fig. 1E). The pretreatment regimens had no significant effect on baseline cellular apotosis (fig. 1F). Quantitative data analysis of the immunohistochemically stained brain slices showed that hypoxic pretreatment of the rats followed by exposure to N²O + isoflurane produced a significant increase in the number of caspase 3-positive cells in the cortex to 220 ± 22 from 177 ± 18 as observed in the group receiving a combination of N²O + isoflurane only (P < 0.05). Xenon pretreatment significantly decreased it to 61 ± 6 (P < 0.01), whereas N²O pretreatment produced no significant change. A similar pattern of change in cellular apoptosis was found in the hippocampus (fig. 1G). These findings were also confirmed by Western blot analysis (figs. 1H and I).

The neuronal apoptosis seen in the in vivo  experiments could have been due to either the indirect effect of the physiologic disturbances or the inherent effect of anesthetics or to both. To obviate the possible physiologic effects, organotypic hippocampal slices was used, and the data are presented in figure 2. The number of caspase-3 cells was significantly increased to 53 ± 3 after N²O + isoflurane exposure from a baseline level of 29 ± 2 (P < 0.05) (figs. 2A, B, and F). The increase in caspase-3 cells was significantly reduced to 33 ± 4 by xenon pretreatment (P < 0.05) (figs. 2C and F); apoptosis was unchanged by N²O preconditioning (figs. 2D and F). Remarkably, apoptosis was significantly increased by hypoxic pretreatment to 78 ± 7 (P < 0.05) (figs. 2E and F).

Pretreatment with 70% xenon for 2 h followed 24 h later by exposure to N²O + isoflurane triggered a relative up-regulation in the expression of Bcl-2 (1.29 ± 0.05; P < 0.05), measured 24 h after event, compared with the control, which was treated with N²O + isoflurane alone. Pretreatment with either 70% N²O in oxygen (1.15 ± 0.1) or 8% oxygen in nitrogen (1.21 ± 0.39) without exposure to N²O + isoflurane caused no significant change in the expression of Bcl-2 (P > 0.05), but pretreatment with 8% hypoxia followed by exposure to the anesthetic gas mixture caused a significant decrease in the expression of Bcl-2 (0.7 ± 0.03), (P > 0.05) (fig. 3A). Cytochrome C  immunoreactivity was analyzed by Western blotting of the cytosolic fraction of the cortical extracts of rats pups subjected to preconditioning. Pretreatment with 8% oxygen in nitrogen and 70% N²O caused a significant increase in the expression of cytochrome C  to 2.23 ± 0.2 and 1.8 ± 0.1, respectively, relative to the N²O + isoflurane control (P < 0.05 or 0.01) (fig. 3B). Xenon pretreatment caused a decrease in the expression of p53 to 0.79 ± 0.05 relative to N²O + isoflurane control, whereas hypoxic pretreatment induced a relative increase to 1.37 ± 0.1 (P < 0.05), and N²O pretreatment produced no significant change (fig. 3C). There were no significant changes in mitogen-activated protein kinase phosphorylation after any of the above treatments (fig. 3D).

After pretreatment and exposure to the anesthetic gas mixture, cohorts of rats were allowed to live up to the age of 40 days for the assessment of cognitive function using trace fear conditioning. Hypoxia pretreatment followed by N²O + isoflurane administration (11 ± 1; P < 0.05) and N²O + isoflurane administration alone (15 ± 2; P < 0.05) resulted in shorter hippocampal dependent-freezing response time in the context test than xenon pretreatment followed by N²O + isoflurane treatment (24 ± 1). The latter was similar to that observed in naive controls (27 ± 7; P > 0.05) (fig. 4A). The results of the tone test, mainly used to assess amygdala function, showed that the area under curve (arbitrary unit) was higher in the xenon preconditioned group exposed to N²O + isoflurane (243 ± 27) than in the hypoxic preconditioned group exposed to the neurotoxic anesthetic combination (157 ± 25; P < 0.05) and the positive control group that received N²O + isoflurane only (200 ± 26; P < 0.05). The area under curve in the xenon pretreatment followed by N²O + isoflurane group was not significantly different from that of the naive group (261 ± 26) (figs. 4B and C).

In this study, we observed significant increases in caspase-3 levels, both in vitro  and in vivo , in the hippocampus and cortex of 7-day-old rats exposed to 0.7% isoflurane + 70% N²O in oxygen for 6 h at 1 atmospheric ambient pressure compared with naive controls. The observed increase in caspase-3 levels is due to the direct effects of the anesthetic drugs on the neural tissues and not because of altered physiology in neonatal rats, for example, hypoperfusion of the brain or hypoglycemia, because similar results were observed in the organotypic hippocampal slices. In both in vitro  and in vivo  studies, xenon was effective at suppressing N²O + isoflurane-induced apoptosis and subsequent deterioration in cognitive function. Conversely, hypoxic pretreatment exacerbated apoptotic neurodegeneration and worsened cognitive impairment in the rats when they reached adulthood.

The mechanism responsible for the increase in caspase-3 levels in the developing brain during synaptogenesis after exposure to anesthetic drugs remains elusive. Drugs that block NMDA receptors or promote GABA^Areceptor function have been shown to produce neurodegeneration in neonatal animal models.3,16, GABA^Aand NMDA receptors and their agonists play a fundamental role during synaptogenesis as they are essential for migration of neurons to the appropriate regions of the nervous system, dendritic filopodia stabilization, synaptic development, and stabilization.2Any interventions including pharmacologic agents, for example, anesthetics affecting GABA^A- and NMDA-mediated synaptic transmission, are not favorable for normal neuronal development. This hypothesis is supported by a study conducted by Wang and coworkers in cortical neuronal cultures derived from 3-day-old monkeys.17They showed that blockade of NMDA receptors by ketamine for 24 h was associated with an increase in neuronal apoptosis, and when the transcription and translation of NMDA receptors were inhibited by the addition of NMDA receptor NR1subunit antisense oligodeoxy nucleotide to the cultured neurons, this effect was abolished.17Furthermore, cellular ionic and neurotrophin homeostases are also crucial for normal neuronal survival or development. This has been clearly demonstrated by previous elegant studies, showing that excessive calcium release induced cell apoptosis is due to the activation of the endoplasmic reticulum membrane 1,4,5-trisphosphate receptor by isoflurane.18Also, isoflurane reduces tissue plasminogen activator release and subsequently enhances pro-brain-derived neurotrophic factor/p75NTR-mediated apoptosis.19 

In this study, we investigated the therapeutic potential of pretreatment to prevent neuronal apoptosis produced by anesthetic drugs during synaptogenesis. The protective effect of pretreatment with xenon against N²O + isoflurane-induced neuronal apoptosis in neonatal rats could be partly attributed to the drug-inhibiting mitochondria-induced activation of the caspase-3 pathway as xenon up-regulated the expression of antiapoptotic Bcl-2 and down-regulated proapoptotic tumor suppressor transcription factor P53. These were associated with no significant changes at the cytochrome C  level. Xenon pretreatment also caused an increase in the phosphorylation of cyclic adenosine monophosphate response element binding protein transcription factor and brain-derived neurotrophic factor expression.10In contrast to xenon pretreatment, hypoxic pretreatment exacerbated N²O + isoflurane-induced neuronal apoptosis by decreasing the expression of antiapoptotic factor Bcl-2 and increasing the expression of proapoptotic transcription factor p53, resulting in an increase in cytochrome C  release from the mitochondria. However, it should be pointed out that similar to isoflurane, xenon may also have dual effects, that is, neuroprotection and neurotoxicity, depending on the duration of exposure.20Indeed, exposure to xenon for a short period of 2 h, as in our current study, is protective, whereas longer periods of exposure produced mild toxicity to the developing brain.21It was noted in our study that N²O increased both Bcl-2 and cytochrome C . This may be one of the reasons why N²O failed to induce a pretreatment state and also failed to protect anesthetic-induced neurotoxicity although similar to xenon, N²O is also an NMDA receptor antagonist. Interestingly, isoflurane was not neurotoxic to rat fetal brain if it was delivered at the late gestational stage,22whereas isoflurane exposed to postnatal rats even at lower concentrations was still toxic.14Moreover, isoflurane has been shown to protect the brain, heart, lungs, and kidney from ischemic insult at pretreatment settings.23–26It may be postulated that the subject's age and experimental conditions are crucial factors for isoflurane to be protective or toxic. Nevertheless, our result is in keeping with a previous study on rats, which showed that pretreatment with nitrous oxide failed to reduce the extent of myocardial infarction after 25 min of coronary occlusion.27 

The results from trace fear conditioning testing are in line with previous reports of cognitive impairment in rats after exposure to general anesthesia during the neonatal period.6,28,29Although the frontal cortex, hippocampus, periaqueductal gray, and rostral ventral medulla mainly underlie the acquisition of fear memories, hippocampal damage is causally linked to long-term memory deficit.30In our study, rats displayed a normal acquisition response to the several tone-shock pairings using a trace fear conditioning paradigm. The main difference seen between the several treatments was during the context test, which is related to hippocampal function.13Our data showed that the rats pretreated with either nitrous oxide or hypoxia displayed less freezing compared with naive controls and xenon-pretreated animals (fig. 4). These changes were compatible with damage to hippocampus as is shown that neuroapoptosis was increased in the hippocampus of rats (fig. 1).

The relevance to the human species of the aforementioned experimental findings in neonatal rats and of those in many other studies conducted in neonatal animals is being questioned,31the main argument being the lack of strong evidence of its toxicity in neonatal humans. However, Wilder et al.  32in a retrospective study of the medical and educational records of children in Olmsted County showed, using multivariate analysis, that children who had repeated general anesthetics before the age of 4 yr had a higher incidence of learning difficulties compared with those who received only one general anesthetic or no anesthetic. In another retrospective study, Kalkman et al.  33showed that children under 24 months undergoing urologic procedures under general anesthesia tended to show more behavioral problems than those operated after 24 months of age under general anesthesia.

Our experimental model showed that a period of hypoxia followed soon afterward by general anesthesia with an inhalational agent and nitrous oxide during the neonatal period resulted in neuronal death and subsequent cognitive impairment. This unexpected finding may have important clinical implications. For example, neonates born with transposition of great vessels are hypoxic shortly after birth; they require urgent surgery to allow oxygenated blood from the pulmonary circulation to reach the systemic circulation. Babies born with gastroschisis or severe exomphalos also require urgent surgery to repair the abdominal defect. If they are unfortunate enough to experience fetal distress during childbirth, they may be at an increased risk of brain injury after surgical correction during anesthesia with the offending agents. If this finding can be extrapolated to clinical practice, then pediatric anesthesiologists dealing with similar situations need to choose carefully the anesthetic regimen to minimize or even prevent brain injury.