Most in vitro neuroprotection studies with isoflurane have involved cells obtained during the embryonic or early postnatal period. However, in mature rodents, isoflurane neuroprotection does not persist. The authors determined whether neuroprotection of hippocampal slices with isoflurane decreases with aging and is due to decreased intracellular Ca regulation and survival protein phosphorylation.
Hippocampal slices from 5-day-old, 1-month-old, and 19- to 23-month-old rats were deprived of oxygen and glucose for 5-30 min in media bubbled with 1% isoflurane. Cell death was assessed in the CA1, CA3, and dentate regions, and intracellular Ca concentration was measured in CA1 neurons. N-methyl-d-aspartate receptor (NMDAR)-dependent Ca influx was measured and the phosphorylation of NMDARs, and the survival proteins Akt and mitogen-activated protein kinase p42/44 were quantified.
Twenty minutes of oxygen and glucose deprivation killed approximately 40-60% of neurons in CA3 and dentate in all age groups. Isoflurane, 1%, reduced death of CA1, CA3, and dentate neurons in slices from 5-day-old rats but not those from 23-month-old rats. In 5-day slices, isoflurane attenuated NMDAR-mediated Ca influx, whereas in aging slices, Ca influx was increased protein kinase C. In aging slices, isoflurane did not increase the phosphorylation of Akt and p42/44.
Isoflurane neuroprotection of hippocampal slices during oxygen and glucose deprivation decreases with age. Isoflurane does not prevent large increases in intracellular Ca concentration during oxygen and glucose deprivation and does not induce the phosphorylation of the prosurvival proteins in aging slices. A protein kinase C-mediated increase in NMDAR activity may result in increased excitotoxicity and decreased neuroprotection by volatile anesthetics in the aging brain.
MOST ischemic neurologic injury occurs in the elderly, whereas most in vitro studies of neuroprotection have involved cells derived from the embryonic or early postnatal brain. Volatile anesthetics seem to have more efficacy against ischemia-like injuries in these in vitro models1–3than in models of brain ischemia involving mature, intact rodents (for review, see Kawaguchi et al. 4,5). For example, in several studies of long-term outcome after relatively severe cerebral ischemia in intact rats, isoflurane has fleeting neuroprotective effects, apparently delaying, but not ultimately preventing, apoptotic neuronal death.4,5In contrast, isoflurane provides at least 2 weeks of protection in hippocampal slice cultures derived from 7- to 10-day-old rats.2It is not known to what extent the difference in age of these in vitro and in vivo models explain this divergence.
Maturation and aging involve decreases in the intracellular signaling pathways that play critical roles in cellular defense against hypoxic/ischemic injury and in the mechanism of action of neuroprotective anesthetics.6One group of signaling processes that changes substantially with age are those regulated by moderate increases in intracellular Ca2+concentration ([Ca2+]i). For example, the coupling of small to moderate changes in [Ca2+]ito transcription of prosurvival proteins fades with age because of oxidative damage to the Ca2+-binding protein calmodulin with subsequent decreased effectiveness of calmodulin-dependent mitogen-activated protein kinase and transcription factor signaling.7,8Age is also associated with an impaired capacity to recover from intracellular Ca2+loads such as those that occur during hypoxia, ischemia, or neurodegenerative stresses.9
Both acute and preconditioning neuroprotection with isoflurane require that moderate increases in [Ca2+]iare coupled to Ca2+-dependent survival signals including the phosphorylation of mitogen-activated protein kinases and survival-associated proteins such as Akt.1,10Therefore, we hypothesized that the neuroprotective efficacy of isoflurane might decrease with aging and be related to impaired Ca2+homeostasis and to a failure to activate prosurvival mitogen-activated protein kinases and the antiapoptosis factor protein kinase B (Akt).
To explore age-related changes in isoflurane neuroprotection, we used hippocampal slices. Acutely made slices of hippocampus have significant advantages for an in vitro model of brain aging because this model preserves developmental properties and synaptic function that are lost or dedifferentiated in dispersed cell cultures. The model also allowed us to correlate cell death in defined populations of neurons to levels of intracellular Ca2+, the relative activity of N -methyl-d-aspartate (NMDA) receptors (assessed as NMDA-induced increases in [Ca2+]i), and the phosphorylation status of key survival-associated proteins.
Materials and Methods
Preparation of Hippocampal Slices
Hippocampal slices were prepared from 5-day-old, 1-month-old, and 19- to 23-month-old (“aging”) Sprague-Dawley rats using methods approved by the Committee on Animal Research, University of California, San Francisco, California. The younger rats were purchased from Charles River Laboratories (Wilmington, MA), and aging rats were obtained from Zivic Laboratories (Pittsburgh, Pennsylvania). After decapitation during 2–3% halothane anesthesia, the brain was rapidly removed and placed in ice-cold artificial cerebrospinal fluid (aCSF). The aCSF used to prepare and maintain the slices contained 116 mm NaCl, 26.2 mm NaHCO3, 5.4 mm KCl, 1.8 mm CaCl2, 0.9 mm MgCl2, 0.9 mm NaH2PO4, and 5.6 mm glucose, pH 7.4, bubbled with 95% O2–5% CO2. Hippocampi were sectioned with a tissue chopper into 400-μm transverse slices. After isolation, slices remained in oxygenated aCSF at room temperature for approximately 1 h.
In Vitro Oxygen and Glucose Deprivation
Hippocampal slices were exposed to simulated ischemia by transferring them with a large-bore plastic pipette to glass vials containing 20 ml glucose-free aCSF bubbled with 95% N2–5% CO2. The partial pressure of oxygen of this solution was less than 2 mmHg (measured with a Clark oxygen electrode; Cameron Instrument Co., Port Aransas, TX). The vials containing the slices were immersed in a water bath so that aCSF temperature was 37°± 0.2°C, monitored with a thermocouple thermometer. After 5, 10, 20, or 30 min of oxygen and glucose deprivation (OGD), slices were removed and recovered in oxygenated aCSF at 37°C for 1 h. After recovery, hippocampal slices were assigned randomly to morphologic analysis, cell survival assessment, or Western blot analysis. For some slices, 1.0% isoflurane was added to the gas mixtures used to bubble the vials containing the hippocampal slices. A calibrated isoflurane vaporizer (Ohmeda, Tech 3; General Electric Healthcare, Fairfield, CT) was used to prepare these gas mixtures, and 1% isoflurane was continually bubbled during the tests. The gas containing isoflurane was bubbled in the vials 30 min before introduction of the slices.
Cell Survival Analysis
Slices were labeled with the fluorescent viability indicator propidium iodide (PI, 2.3 μm) for 20 min, as described previously.2Cell death was assessed in CA1, CA3, and dentate gyrus regions on three or four different experiments containing 7–11 slices in each treatment group. PI fluorescence emission was measured 5, 10, 15, or 20 min after OGD injury or 20 min after isoflurane exposure using a fluorescence microscope (Nikon, Tokyo, Japan) with a 10× lens coupled to a camera (Diagnostic Imaging, Spot II, San Carlos, CA). PI images were acquired with standardized camera settings, and the optical density was measured with NIH ImageJ software (National Institutes of Health, Bethesda, MD). The regions of interest for cell death analysis were defined separately for each slice and conformed tightly to the margins of the CA1, CA3, or dentate cell body regions. After subtracting the background fluorescence (from control slices not subjected to injury), the results were expressed as percentage of dead cells. The cell death was averaged for a total of 7–11 slices obtained from three or four different rats (each rat comprised a separate experiment, each performed on a different day).
Slices were fixed in 4% paraformaldehyde in buffered saline overnight at 4°C. Further handling included paraffin embedding, sectioning, and staining with hematoxylin and eosin. Paraffin-embedded slices were sectioned in the same transverse direction as the original slice to permit examination of the CA1, CA3, and dentate cell body fields in the same slices. Serial sections 5 μm thick were cut and mounted. Cell death was assessed as the percent of missing cells or vacuolated, grossly disrupted cells.
Measurement of Intracellular Ca2+Concentration
Intracellular Ca2+concentration was measured in separate groups of slices from those in the survival studies. The change of [Ca2+]iin CA1 neurons in slices were measured using the indicator fura-2 AM and a dual excitation fluorescence spectrometer (Photon Technology International, South Brunswick, NJ) coupled to a Nikon Diaphot inverted microscope (Tokyo, Japan). Slices were incubated with 5 μm fura-2 AM for 30 min before measurement. For the measurements, we used a specially constructed low-volume slice array chamber that contained six to eight slices positioned on a mesh support. Fluorescence signals for the determination of [Ca2+]iwere obtained only from the CA1 cell body region by adjusting slit apertures in the photomultiplier tube section of the inverted microscopy apparatus. Measurements of fura fluorescence were obtained in each slice before and after treatments, which were achieved by perfusion application of test treatments or agents (OGD, 100 μm NMDA, the protein kinase C [PKC] inhibitors chelerythrine and calphostin C or 1% isoflurane bubbled perfusate). Relative [Ca2+]iwas determined as background fluorescence–corrected fura fluorescence ratio (510 nm emission, 345/380 nm excitation).
Western Blot Analysis
Slices were frozen at −70°C 1 h after the exposure to the test solutions (e.g. , OGD or OGD plus isoflurane). Frozen tissues were homogenized in ice-cold 50 mm Tris-HCl (pH 7.4) containing 0.1 μm phenylmethylsulfonyl fluoride, 2 μm leupeptin, 1 μm pepstatin, and 0.1% 2-mercaptoethanol. The homogenates were centrifuged at 14,000g for 30 min at 4°C, and the pellet was discarded. The protein in the supernatant was used. Protein samples were loaded (20 μg each) and separated on 7.5% SDS-PAGE, followed by blotting of the proteins to nitrocellulose membrane. The blot was blocked with a buffer consisting of 10 mm Tris-HCl (pH 7.4), 0.15 m NaCl, 2% nonfat milk, 2% bovine serum albumin, and 0.1% Tween-20, for 2 h at room temperature. The blots were then incubated with primary antibody (1:500 dilution) for 2 h at room temperature. Then the blots were incubated with a secondary goat anti-mouse or anti-rabbit immunoglobulin G conjugated with horseradish peroxidase and detected with an enhanced chemiluminescence procedure and photographic film. β-Actin was included as a loading control, and molecular weight standards were included with all blots. The intensities of bands were quantified with NIH ImageJ software; this involved selecting a region of interest and determining the mean gray value. Band intensity was then expressed relative to the density of the band in the control samples.
Reagents and Antibodies
Rat polyclonal antibodies against phospho-Akt Thr 308, phospho-p44/42 Thr 202/Tyr 204, and phospho-NMDA receptor subunit NR1 S896 and S897 were purchased from Cell Signaling (Beverly, MA). Fura-2 AM was obtained from Molecular Probes (Eugene, OR), and PI and the PKC inhibitors chelerythrine and calphostin C were obtained from Sigma-Aldrich (St. Louis, MO).
Results are expressed as mean ± SE. Analysis of variance was used to compare the difference among groups, and the Fisher protected least significant difference statistic was used as a post hoc test. A P value of less than 0.05 was considered statistically significant.
Oxygen and Glucose Deprivation Kills More Neurons in Mature Than Neonatal Hippocampus
Before evaluating the neuroprotective effects of isoflurane, we examined how age influences neuron death in different hippocampal cell regions after periods of OGD. Slices obtained from rats in the neonatal period (5 days of age) were generally more tolerant of OGD than those from mature and aging animals; this was seen in different regions in the slice and for different durations of OGD. Estimates of cell death based on PI fluorescence are shown in figure 1, and representative images of whole-slice PI fluorescence are shown in figure 2. In young adult rats (1 month) and aging rats (19 month), 10 min of OGD killed approximately 55% of CA1, CA3, and dentate neurons. Neuron death was approximately 20–30% in 5-day-old rats for the same duration of OGD (P < 0.05 only for CA3). For the more severe injury of 20 min of OGD, cell death in slices from neonatal rats was statistically less than that in slices from mature/aging animals only in the CA1 region; in the CA3 and dentate gyri, cell death in all age groups was statistically similar. Slices from all age groups exposed to a 30-min period of OGD fell apart; cell death data could not be obtained, and this time point will not be further considered.
Data that compare the accuracy of the PI fluorescence method with histologic estimates of cell death are available only for immature tissue.11Therefore, we compared cell death estimated by the PI method with that of counting damaged/missing neurons in fixed sections of slices stained with hematoxylin and eosin. Regression analysis of this data (fig. 3) suggests that the PI method indicates greater cell death than the histologic method, but that increased death detected by one method correlates with increased death detected by the other.
Isoflurane Reduces Cell Death in Neonatal but Not Mature Hippocampus
For these studies, we used 20 min of OGD because this produced the most similar cell death in slices of all age groups. Figures 4 and 5show that 1% isoflurane present during 20 min of OGD decreases cell death in all cell regions in slices prepared from 5-day-old rats but not in slices from aging rats (23 months).
Histologic staining of slices (fig. 6) confirmed that neuronal damage, indicated by cell swelling and vacuolization, was widespread throughout the CA1, CA3, and dentate regions of slices from the older rats in the OGD and OGD–isoflurane groups. In contrast, in neonatal hippocampus, only the OGD group showed substantial cell swelling, vacuolization, and cell drop-out.
Isoflurane, in Oxygenated Media, Kills Neurons in Slices from Aging Rats
Isoflurane present for 20 min in oxygenated media resulted in significantly increased PI fluorescence in CA1, CA3, and dentate cell regions in slices from 23-month-old animals (figs. 4 and 5). An increase in cell death after exposure to solution bubbled with 1% isoflurane was also clearly seen in slices fixed with paraformaldehyde and stained with hematoxylin and eosin (fig. 6). These indicators of cell death were not seen in the slices from 5-day-old rats exposed to isoflurane.
Survival Signaling Proteins and Isoflurane Protection
Acute and preconditioning neuroprotection of cultured hippocampal slices with 1% isoflurane is associated with increases in levels of the phosphorylated MAP kinase p42/44 (ERK) and Akt (protein kinase B).1,10Accordingly, we performed Western blot analysis of these proteins in slices exposed to isoflurane, OGD, or isoflurane plus OGD. Figures 7A and Bshow that levels of p42/44 decreased after OGD in mature and aging slices but that in neonatal slices, levels of this protein were relatively preserved. Isoflurane present during OGD did not decrease the loss of this immunostaining in either 1-month-old or 23-month-old tissue. Of note, isoflurane present in normally oxygenated media decreased p42/44 levels in hippocampal slices from 23-month-old rats.
Age-related changes in phospho-Akt levels after OGD were also seen (figs. 7A and C). OGD with or without isoflurane was associated with loss of Akt immunoreactivity in slices from 1-month-old and 23-month-old rats, whereas no loss of this protein occurred in slices from neonates; in fact, increased immunostaining was seen in the group of slices exposed to a combination of isoflurane and OGD. Similar to the situation with p42/44, isoflurane present in oxygenated media decreased the level of Akt in slices from 23-month-old rats.
Isoflurane and Intracellular Ca2+Regulation in Aging Neurons
Because isoflurane has salutary effects on intracellular Ca2+regulation in hypoxic neurons from cerebrocortex and hippocampus,12we compared the effects of isoflurane on [Ca2+]iin CA1 neurons in hippocampal slices measured after 10 min of OGD (fig. 8). Compared with slices from 5-day-old rats, OGD in slices from 23-month-old animals resulted in much larger increases in [Ca2+]iat this time point. Further, isoflurane reduced the increase in [Ca2+]ithat occurs during OGD in slices from 5-day-old rats, whereas the increase in [Ca2+]iseen in the isoflurane–OGD group in aging rats was much larger then in the group exposed to OGD alone.
To further examine the effects of isoflurane on [Ca2+]iin CA1 neurons during OGD, we continuously measured [Ca2+]iin a separate group of slices. The averaged traces of [Ca2+]iversus time are shown in figure 9. OGD increased [Ca2+]iin all slices from both age groups. Introduction of isoflurane-bubbled perfusate stabilized [Ca2+]iin the slices from the 5-day-old rats but led to further increases in [Ca2+]iin the slices from the 23-month-old rats.
Effects of Isoflurane on NMDA-mediated Ca2+Influx
To investigate the basis for greater increases in [Ca2+]iin CA1 neurons in hippocampal slices in aging animals, we measured NMDA-induced changes in [Ca2+]iin CA1 neurons. As shown in figure 10, NMDA application caused larger increases in CA1 neurons in slices from 5-day-old rats than in slices from 23-month-old animals. A major determinant of NMDA receptor activity in hypoxic cortical neurons is phosphorylation by PKC at serine 896 of the NR1 subunit; increased phosphorylation at this site is associated with increased receptor activity.13Accordingly, we used Western blots to determine relative levels of phosphorylation at S896. As shown in figure 11, increased phosphorylation of S896 occurred during OGD in mature (1 month) and aging slices. Phosphorylation at S896 during OGD was substantially attenuated by isoflurane in slices from 5-day-old and 30 day-old rats. In contrast, isoflurane did not suppress phosphorylation of this target in the slices from 23-month-old animals. Of note, isoflurane in oxygenated slices increased phosphorylation at S896 in these older slices.
Consistent with a role for PKC in the regulation of NMDA receptor activity by isoflurane, the PKC inhibitor calphostin (1 μm) prevented isoflurane from increasing in NMDA-mediated Ca2+influx in 23-month-old slices (fig. 10). The PKC inhibitor chelerythrine, at two concentrations previously reported to exceed that needed to inhibit PKC activity in neural tissue (10 and 50 μm),14had no effect on isoflurane’s augmentation of NMDA receptor responses in this tissue (data not shown).
The main results of this study show that the vulnerability of hippocampal slices to oxygen and glucose deprivation increases with age and isoflurane is an effective neuroprotectant against this stress only in slices obtained from immature animals. A secondary finding was that aging is associated with decreased capacity of isoflurane to limit increases in [Ca2+]iduring OGD and with a failure of isoflurane to cause changes in the phosphorylation state of prosurvival proteins and NMDA receptors. Because isoflurane apparently does not possess durable neuroprotective qualities in intact mature/aging animals,4,5our findings may partly explain why isoflurane is neuroprotective in in vitro models of ischemia (which are all based on the immature nervous system) but does not yield lasting neuroprotection in older, intact animals.
In vitro models of cerebral ischemia are limited to neonatal preparations because mature tissues, with possible exception of hippocampal slice cultures,15cannot be maintained in vitro for long periods. These in vitro models have uniformly shown varying degrees of protection with isoflurane or other volatile anesthetics.1,2,10,12,16–21In contrast, it is now clear that isoflurane provides transient, not permanent neuroprotection22in intact animal models in which injury and cell loss are followed over longer periods of time than the typical in vitro study.4As far as we are aware, it has not been established whether isoflurane provides long-lasting (weeks to months) neuroprotection in intact, neonatal models of cerebral ischemia.
Our study was designed to test the neuroprotective efficacy of isoflurane against durations of moderately severe simulated ischemia (OGD) that produce similar cell death in neonatal and aging tissue. This was done so that we were not requiring that isoflurane protect against an effectively more damaging injury in aging slices. We also used this design to facilitate comparison with the moderate to severe ischemia used in recent in vivo studies of isoflurane protection (e.g. , Toescu et al. 6). However, there were trends toward more resistance to a 20-min duration of OGD for all cell regions in the slices from 5-day-old animals. Nonetheless, isoflurane produced much greater reduction in cell death than this age effect, and the effects of isoflurane on NMDA receptor–based responses (figs. 8 and 9) were widely divergent in the young and aging slices. Another potential limitation of the study is that 1% isoflurane was used for all age groups, and anesthetic potency increases with age. The effect of isoflurane on NMDA receptors may also vary with age.
Aging-related Changes in Calcium Homeostasis during Oxygen and Glucose Deprivation
Changes in Ca2+homeostasis may contribute substantially to the increased vulnerability of the aging brain to ischemic stress. Our results agree with a large number of previous studies showing that the developing brain is more tolerant of hypoxia or ischemia (reviewed in Bickler and Hansen23) but perhaps more sensitive to excitotoxic drugs and prone to apoptotic cell death.24In CA1 neurons, OGD caused larger increases in [Ca2+]iin 23-month-old slices than in younger slices; even larger differences in [Ca2+]iwere observed when isoflurane was combined with OGD (figs. 8 and 9). The larger and sustained increases in [Ca2+]iare probably more injurious, leading to greater dysfunction and acute and delayed cell death,25and would tend to work against other potential mechanisms of isoflurane protection such as decreases in metabolism, inhibition of excitatory neurotransmitter release, and others.26In view of our findings, it would be of interest to examine whether these other mechanisms of isoflurane neuroprotection also fade with age.
NMDA Receptor Regulation
In the hippocampal slice, most of the increase in [Ca2+]ithat occurs during anoxia or ischemia comes from Ca2+influx via NMDA receptors; increases in [Ca2+]iand cell damage during these stresses are effectively blocked by NMDA receptor antagonists.27Our studies show that isoflurane has a dramatically different effect on NMDA-mediated Ca2+influx in old compared with young slices (fig. 8). In aging slices, it seems that the augmentation of Ca2+influx via NMDA receptors is due to PKC because the augmentation was blocked by 1 μm calphostin, a relatively specific PKC inhibitor. However, the augmentation was not affected by chelerythrine. Calphostin is a more potent inhibitor of PKC-δ than of PKC-γ; the opposite is true of chelerythrine.28A similar differential effect of chelerythrine and calphostin was seen in the phosphoregulation of the metabotropic glutamate receptor.29Although both chelerythrine and calphostin may differentially inhibit a number of the 10 or more isoforms of PKC present in hippocampus,30the results suggest that significant changes occur in the isoforms of PKC that mediate isoflurane-dependent alterations in NMDA receptor activity during maturation and aging. This requires more study.
N -methyl-d-aspartate application to 5-day-old slices caused larger increases in [Ca2+]ithan did similar application to the 23-month-old slices. This is consistent with the high-Ca2+flux NMDA receptors expressed in the developing hippocampus31and with the developing brains sensitivity to excitotoxicity from NMDA receptor agonists.24Probably because the predominant effect of isoflurane was to suppress NMDA receptor–mediated Ca2+increases in slices from 5-day-old slices, calphostin had minimal additional effect on these responses.
Changes in the phosphorylation of Akt and p42/44 were consistent with a neuroprotective effect of isoflurane in the neonatal but not aging hippocampus. We found that levels of Akt and ERK, two proteins associated with survival signaling/inhibition of apoptosis in the hippocampus,32–36were increased by isoflurane neuroprotection. Other studies have reported that Akt and related proteins, such as focal adhesion kinases, are associated with isoflurane protection.1,21Consistent with a critical role for ERK and Akt in cell survival signaling, we found that cell death induced by isoflurane in normally oxygenated slices from aging rats was correlated with loss of phosphorylated ERK and Akt.
In normally oxygenated hippocampal slices from aging rats, 20 min of exposure to 1% isoflurane increased the death of hippocampal neurons in the CA1, CA3, and dentate regions. This apparent toxicity seems consistent with in vivo data from aging rats showing decreased hippocampal function after isoflurane anesthesia37and with a report that 5 h of 1.5% isoflurane causes cell death in hippocampal slice cultures.38Isoflurane toxicity was greatest in the dentate gyrus of aging rats, a finding that may be significant because the dentate is the site of neurogenesis in the hippocampus. Prolonged exposures to volatile anesthetics may be toxic to the developing brain,39whereas ketamine and nitrous oxide have been reported to be more toxic to the aging brain.40
The mechanism for the apparent neurotoxic effect of isoflurane on slices from aging rats was not determined. However, isoflurane exposure resulted both in the loss of phospho-Akt and in significant increases in [Ca2+]i. The increases in [Ca2+]icaused by exposure to isoflurane were larger in aging slices than those in neonatal slices exposed to isoflurane (fig. 8), which may be quite significant because aging neurons may lose the capacity to cope effectively with Ca2+loads or there may be age-related changes in the coupling of changes in [Ca2+]ito neuroprotective or neuro-injurious signaling.6Significant isoflurane toxicity was not seen in immature slices in this study, but the period of observation may have been too short to see apoptotic changes or increased PI staining that might eventually have resulted.
We conclude that in hippocampal slices from aging rats, vulnerability to simulated ischemic injury increases while the neuroprotective effects of isoflurane fade. Loss of isoflurane protection in the aging hippocampus is correlated with alterations in intracellular Ca2+homeostasis and with the phosphorylation of NMDA receptors and survival-associated proteins.