NIH Public Access Author Manuscript J Neurochem. Author manuscript; available in PMC 2009 September 1. NIH-PA Author Manuscript Published in final edited form as: J Neurochem. 2009 June ; 109(5): 1413–1426. doi:10.1111/j.1471-4159.2009.06068.x. Modulation of stress proteins and apoptotic regulators in the anoxia tolerant turtle brain Shailaja Kesaraju*,1, Rainald Schmidt-Kastner2, Howard M. Prentice2, and Sarah L. Milton1 1Department of Biological Sciences, Florida Atlantic University, 777 Glades Road, Boca Raton, Florida, 33431, USA 2College of Biomedical Science, Florida Atlantic University, 777 Glades Road, Boca Raton, Florida, 33431, USA Abstract NIH-PA Author Manuscript Freshwater turtles survive prolonged anoxia and reoxygenation without overt brain damage by well-described physiological processes, but little work has been done to investigate the molecular changes associated with anoxic survival. We examined stress proteins and apoptotic regulators in the turtle during early (1 h) and long-term anoxia (4, 24h) and reoxygenation. Western blot analyses showed changes within the first hour of anoxia; multiple stress proteins (Hsp72, Grp94, Hsp60, Hsp27, and HO-1) increased while apoptotic regulators (Bcl2 and Bax) decreased. Levels of the ER stress protein Grp78 were unchanged. Stress proteins remained elevated in long-term anoxia while the Bcl2/Bax ratio was unaltered. No changes in cleaved caspase-3 levels were observed during anoxia while AIF increased significantly. Furthermore, we found no evidence for the anoxic translocation of Bax from the cytosol to mitochondria, nor movement of AIF between the mitochondria and nucleus. Reoxygenation did not lead to further increases in stress proteins or apoptotic regulators except for HO-1. The apparent protection against cell damage was corroborated with immunohistochemistry, which indicated no overt damage in the turtle brain subjected to anoxia and reoxygenation. The results suggest that molecular adaptations enhance pro-survival mechanisms and suppress apoptotic pathways to confer anoxia tolerance in freshwater turtles. Keywords NIH-PA Author Manuscript Heat shock proteins; turtle; anoxia; reoxygenation; apoptosis; neurons; Trachemys scripta Introduction In mammals, neuronal death due to extended anoxia is inevitable, but some organisms possess the extraordinary ability to survive hours to days of anoxia without apparent damage to the brain. One such anoxia-tolerant species is the freshwater turtle Trachemys scripta elegans. Earlier studies have shown that anoxia tolerance results from the modulation of complex cellular processes including the controlled release and reuptake of neurotransmitters (Milton and Prentice, 2007), arrest of ion channel function (Perez-Pinzon et al., 1992; Bickler et al., 2000), and regulation of MAP kinases (Milton et al., 2008). Studies on preconditioning in rodent models have identified multiple genes/proteins that confer increased tolerance to ischemia (Chen et al., 1996). We have proposed that the constitutive expression and subsequent modulation of such genes/proteins could also play a * To whom correspondence should be addressed Tel: (561) 297-0148 Fax: (561) 297-2749 Email: skesaraj@fau.edu . Kesaraju et al. Page 2 NIH-PA Author Manuscript role in the high tolerance of the turtle to anoxia (Milton and Prentice, 2007). High basal levels and a further upregulation of Hsp72 and Hsp60 (Chang et al., 2000; Ramaglia et al., 2004; Prentice et al., 2004) have been reported in the anoxic turtle brain, while anoxia also increases levels of phosphorylated extracellular regulated kinase (ERK 1/2) and Akt in the turtle brain (Haddad, 2007; Milton et al., 2008). Molecular pathways thought to incur cell death, on the other hand, are suppressed, including activation of the c-jun kinase (JNK) pathway and p38MAPK (Milton et al., 2008). Induction of cytoprotective heat shock proteins (HSPs) in response to physiological stressors (e.g. heat shock, osmotic stress, hypoxia/anoxia) has been well-studied in several model systems which together suggest they shift cellular equilibrium towards survival (Obrenovitch, 2008; Lanneau et al., 2008). Increased expression of Hsp70 is associated with increased ischemic tolerance in preconditioning (Nishi et al., 1993) and permanent cerebral ischemia (van der Weerd et al., 2005). Other HSPs upregulated after ischemia that may play a role in preconditioning include Hsp60, HO-1 (Heme oxygenase 1, also named Hsp32), Hsp27, and the glucose-regulated proteins Grp94 and Grp78 (Kato et al., 1994; Geddes et al., 1996; Chen et al., 1996; Massa et al., 1996; Kirino, 2002; Valentim et al., 2001; Badin et al., 2006; Hwang et al., 2007). NIH-PA Author Manuscript In models of rodent ischemia, however, intrinsic cellular protective responses to ischemiahypoxia cannot be easily separated from secondary reactions to neuronal death. The turtle brain thus presents a unique model to understand the regulation of stress protein expression to promote cell survival in the face of complete anoxia. Histological examination of the turtle brain to determine damage following anoxia or reoxygenation has not been previously performed, on the assumption that turtle survival was accompanied by a lack of cellular damage. In the present study, we determined histologically that the turtle brain remains overtly intact after 24 h of anoxia and examined several stress-related proteins likely to be important in anoxic survival, including Hsp70, Hsp60, Hsp27, HO-1, Grp94 and Grp78, as well as the downstream modulators of apoptosis: Bcl-2, Bax, apoptosis inducing factor (AIF) and caspase-3. The involvement of apoptotic cell death pathways in cerebral ischemia, hypoxia, and neurodegenerative disorders is well studied in rodent models (Cao et al., 2001; Cao et al., 2002; Saito et al., 2005, Mehta et al., 2007), though little work has been done in turtles (Haddad, 2007), who presumably are able to avoid triggering apoptotic pathways. NIH-PA Author Manuscript This study revealed simultaneous regulation of Bcl-2 and Bax, and suggests that apoptotic pathways may be initiated but not executed. We hypothesize that increases in HSPs in the anoxic turtle brain may prevent the activation of key pro-apoptotic regulators such as Bax and caspase-3 (Garrido et al., 1999; Stankiewicz et al., 2005; Matsumori et al., 2006). The present study is the first report to analyze the freshwater turtle’s resistance to anoxia with histological methods and to characterize alterations in the expression of multiple major stress inducible heat shock proteins and apoptotic regulators during different time points in anoxia and in anoxia-reoxygenation. Materials and Methods Induction of anoxia Experiments conducted in this study were approved by the Florida Atlantic University Institutional Animal Care and Use Committee. Freshwater turtles (Trachemys scripta, female) were obtained from commercial suppliers (Clive Longdon, Tallahassee, FL). Animals were kept in freshwater aquaria in an in-lab facility on a 12h light/dark cycle and fed three times weekly with commercial turtle food. J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 3 NIH-PA Author Manuscript To induce anoxia, turtles were placed individually in a tightly sealed container at room temperature (22-23°C) with positive pressure flow-through nitrogen gas (99.99% N2, Air Gas, Miami, FL) for the experimental time period. Animals subjected to reoxygenation were allowed to recover in room air. Control animals (N=5) were utilized directly from the aquaria. For the study of stress proteins, turtles were subjected to anoxia for 1 h, 4 h, or 24 h anoxia in groups of n=4-6. A separate group was exposed to 4 h anoxia followed by 4 h of reoxygenation. The animals were decapitated, and the brains removed and snap frozen in liquid nitrogen in less than 2 min. For histological studies, groups of n=3 turtles were subjected to 24 h anoxia followed by 1 or 3 d of reoxygenation and were then perfusion fixed (as described below). N=5 control animals were fixed in parallel. Protein extraction Proteins were extracted from the frozen brains in cell lysis buffer ( 0.15 M Nacl, 5 mM EDTA pH 8, 1% Triton X100, 10 mM Tris-Cl pH 7.4; with added 5 M DTT, 100 mM PMSF, 5 M mercaptoethanol diluted 1:1000) using a glass homogenizer at 4°C . The homogenate was centrifuged at 13,000 rpm at 4°C for 10 min and the supernatant collected for further analysis. Protein concentrations were determined using a standard BCA assay following the manufacturer’s protocol (Pierce Biotechnology, Inc., Rockville, IL). Cell fractionation NIH-PA Author Manuscript Animals (N=3) were subjected to 4 h of anoxia and 4 h of reoxygenation. N=3 animals were used as controls. Brains were harvested and subcellular fractions were isolated from the brain tissues. Mitochondrial fractions were isolated following the manufacturer’s protocol (Active Motif, Carlsbad, CA). The supernatants obtained through this fractionation were considered as cytosolic fractions. Nuclear extracts were made following the manufacturer’s protocol (Panomics Inc., Redwood, CA). Protein concentrations were assayed through BCA analysis and specific protein levels analyzed by western blot for each fraction. Western blotting NIH-PA Author Manuscript The antibodies used for this study are as follows: rabbit polyclonal anti Hsp70 (SPA-812,), rabbit polyclonal anti Caspase-3 (AAP-113), rabbit polyclonal anti Hsp60 (SPA-805), mouse monoclonal anti HO-1 (OSA-111), rabbit polyclonal anti Hsp25 (SPA-801) and rabbit polyclonal anti Grp78 ( SPA-826) were obtained from Stressgen (Victoria, BC, Canada). Rabbit polyclonal anti-AIF (SC-5586) and mouse monoclonal anti-Bax (SC-7480) were obtained from Santa Cruz Biotechnologies (Santa Cruz, CA). Polyclonal rabbit antiBcl2 (AB1720) was obtained from Chemicon (Temecula, CA). Polyclonal rabbit anti-Grp94 (RB-10642-P1) was obtained from Neomarkers (Union City, CA). All the antibodies except for Hsp70 (1:4000) were diluted at 1:1000 in 2% non-fat milk. All the secondary antibodies (goat anti rabbit and rabbit anti mouse) were obtained from Southern Biotech (Birmingham, AL). Equal amounts of protein (50 μg / lane) were separated on a 15% SDS-Polyacrylamide gel for 1.5 h at 150 V. The proteins were transferred onto a Hybond nitrocellulose membrane (Amersham Biosciences, Pittsburgh, PA) for 1.5 h at 0.3 A. The membrane was blocked in 5 % non-fat milk in TBS (25 mM Tris-Cl, pH 7.5, at 24°C, 150 mM NaCl) for 2 h at room temperature followed by incubation with the respective primary antibody overnight at 4°C. The membranes were washed 3X 10 min in TBS and probed with appropriate HRP conjugated secondary antibody for 2 h at room temperature. Immunoreactive protein bands were then visualized with ECL (Amersham Biosciences). Relative density of the bands was determined using NIH Image J software. Results were normalized to percent of β-actin (utilized as a loading control, no change in anoxia, Prentice et al., 2004; Milton et al., 2008) and relative changes expressed as percent of control (normoxia). J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 4 Perfusion fixation and cryostat sectioning NIH-PA Author Manuscript NIH-PA Author Manuscript Perfusion fixation was carried out to achieve optimal fixation and to avoid the manifestation of neuronal artifacts that impair the evaluation of cell damage. A window was prepared in the ventral shell (plastron) and the heart exposed. A blunt steel cannula was inserted through the tip of the heart, advanced towards the left aorta, i.e. the left-most vessel in the ventral position giving rise to the brachiocephalic artery (Cameron, 1989), and clamped in place. The right atrium was incised to allow for the escape of perfusate. About 200-300 ml of 0.9% saline were slowly injected, followed by 300-400 ml of 4% paraformaldehyde. Brains were postfixed for one day in 4% paraformaldehyde and moved to PBS for storage at 4° C. Following cryoprotection in 20% sucrose for 1- 3 d, brains were immersed into a plastic template filled with Tissue Tek (Sakura) and frozen in 2-methylbutane (Sigma) chilled to — 30-40° C over dry ice. Sections were prepared at 40 μm thickness in the frontal plane (levels A4.8 to P0.8 according to the atlas of Powers and Reiner, 1993) on the cryostat (Microm). Serial sections were taken from the knife and placed into 24-well plates (Costar ® 3526, Corning Inc., Corning, NY) with 2 ml PBS per well. PBS was prepared with 0.5 ml/l antibacterial- antimycotic liquid (10,000 unit/ml penicillin G sodium; 10,000 μg/ml streptomycin sulfate; and 25 μg/ml amphotericin B; (GIBCO, Invitrogen ). The immunohistochemical analysis included regions of the turtle brain between levels A4.8 and A3.2 as defined by Powers and Reiner (1993), e.g. the cortex (cortex medialis, dorsalis and pyriforme), pallial thickening, primordium hippocampi, dorsal ventricular ridge (DVR) of the telencephalon including the core nucleus, paleostriatum augmentatum, globus pallidus, and ventral paleostriatum Other sections were floated onto poly-L-lysine coated slides (Poly-PrepTM slides, Sigma P0425-72EA) and air-dried for cresyl violet staining. Immunohistochemistry NIH-PA Author Manuscript Immunohistochemical labeling was performed using a free-floating method in which relatively thick sections were reproducibly exposed to fixed volumes of antibody solutions (Schmidt-Kastner et al., 1991; Schmidt-Kastner et al., 2005). Reactions were carried out using six sections from 3-4 animals, including a control, in one batch. The peroxidaseantiperoxidase (PAP) method was used due to its high reproducibility and successful prior use on sections of the turtle brain. Sections were moved to PBS containing 0.3% Triton X-100 (PBS-Tr) and 2% normal rabbit or swine serum (matching to the species of the second antibody) for 1 h, and then rinsed in PBS-Tr for 1 h. Reactions in primary antibodies were carried out overnight at +4° C. We used mouse monoclonal antibodies to NeuN (1:250; kind gift of Dr. Mullen, Univ. of Utah) and rabbit antibodies to GFAP (1:500; DAKO, Carpintera, CA). Following a rinse in PBS-Tr for 1 h, sections were reacted with the matched secondary antibodies, i.e. swine anti-rabbit IgG (DAKO) or rabbit anti-mouse IgG (DAKO), diluted 1:100 in PBS-Tr for 1 h. Subsequently, sections were rinsed for 1 h in PBS-Tr and then placed into rabbit-PAP complex (DAKO) or mouse monoclonal PAPcomplex (DAKO) diluted 1:200 in PBS-Tr for 2 h. Following a rinse in PBS for 1 h, peroxidase activity was visualized using 0.05% diaminobenzidine (DAB, Sigma) and 0.01% H2O2 for 5 to 20 min. Following a prolonged rinse in PBS, sections were floated onto polyL-lysine coated glass slides. Dried sections were moved through distilled H2O, 70%, 90% and 100% ethanol, two changes of xylene, and finally coverslipped with DePeX (Sigma). Cresyl violet (CV) staining Cresyl Violet acetate (Sigma C1791-1G) was prepared with 1 g/ l in distilled H2O (0.1%). For delipidation, sections were moved through 70%, 90%, and 100% ethanol, 2 × xylene, 100%, 90%, and 70% ethanol, and then washed in distilled water. Sections were immersed in CV solution for 5 min and quickly moved through distilled H2O, 70%, 90% and 100% ethanol, two changes of xylene, and finally coverslipped with DePeX (Sigma). J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 5 Evaluation of pathology NIH-PA Author Manuscript The analysis was based on CV staining and immunohistochemical labeling for neurons (NeuN) and astrocytes (GFAP), with the presumption that three days is sufficient for the manifestation of serious neuronal damage or infarction in the turtle brain. As a note of caution, protein synthesis and degradation are known to be reduced during anoxia in the turtle brain (Fraser et al., 2001; Storey, 2007), and it is unclear how such changes would influence the manifestation of tissue damage during reoxygenation. A search for selective neuronal necrosis and infarction was first carried out in CV-stained sections. Major structures such as cortex, DVR, core nucleus of the DVR, paleostriatum augmentatum, globus pallidus, septum and ventral areas were examined. The cortex was inspected for interruptions of the pyramidal cell layer, cell loss and gross swelling. Only clear cut signs of ischemic neuronal necrosis such as severe shrinkage and clumping of the nucleus were considered; darkly stained and compressed neurons were not considered as irreversibly damaged. Areas of swelling with decreased background labeling and vacuolization were cautiously considered. NeuN antibodies detect a marker protein for neurons (“NeuN” for neuronal nuclei) (Mullen et al., 1992; Wolf et al., 1995). The antibodies visualize neurons in multiple species (Mullen et al., 1992). Loss of neurons manifests as reduction of the NeuN signal. The basic anatomy of GFAP labeling in the turtle brain has been described (Kalman et al., 1994), whereby labeled glial fibers but not stellate astrocytes were found. GFAP labeled sections were searched for increased signals in reactive astrocytes after 1 and 3 d of reoxygenation (i.e. indicating a reaction to damage). GFAP labeled sections were also inspected for subtle changes of glial processes and for changes in perivascular glia. NIH-PA Author Manuscript Statistical analysis Statistical significance of the data was tested using Minitab 15 (Minitab Inc, State college, PA, USA). One way analysis of variance (ANOVA) was used to compare multiple experimental groups followed by Tukey’s post hoc test. For the reoxygenation experiments after anoxia, data was compared using Student two tailed t-test. All the data are represented as Mean ± SE, and p <0.05 was considered as significant. Results Absence of overt damage to neurons during anoxia and reoxygenation The results of the histological studies on 24 h anoxia and anoxia/recovery animals confirms that anoxia tolerance at which the whole animal survives is indeed accompanied by a lack of overt damage to neurons. Representative images are shown in Fig. 1. NIH-PA Author Manuscript Controls NeuN labeling showed strong nuclear and weak cytoplasmic staining (Fig. 1d). In the cortex, neurons were densely packed into one layer whereas the broad dendritic layer showed few neuronal cell bodies. Different brain nuclei and subregions showed regionally specific densities of neurons, e.g. strong cluster formation was seen in the dorsal ventricular ridge of the telencephalon. GFAP labeling occurred in fine glial processes but not in the cytoplasm of astrocytes (Fig.1g). Radial glial fibers spanned the cortex. Enhanced density of glial fibers was seen in the subpial and subependymal regions. Several large vessels were also outlined by glial processes. CV staining showed neurons and glial cells organized into different layers, nuclei and subregions as described by Powers and Reiner (1993). Densely packed neurons formed a broad layer in the cortex (Fig.1a). Diverse types of neurons were seen in various brain nuclei. A particular feature of the turtle brain is the high density of neurons in many brain areas and the formation of very tight clusters. The cellular features of the turtle brain differ substantially from those of the rat brain and make it difficult to J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 6 NIH-PA Author Manuscript evaluate cell damage. Triangulated and darkly stained neurons were seen in several sections which were considered to be mild fixation artifacts. There were sometimes vacuolated regions around the cortical neurons in CV staining which appeared to be due to mechanical stress on the thin cortical sheath during tissue processing. Subsequently, we surveyed the sections immunostained for NeuN for overt loss of neurons, the GFAP-labeled sections for increases of signals indicative of reactive astrocyte formation, and the CV stained sections for neuronal deficits and gliosis. This approach enabled us to detect infarction and selective neuronal necrosis in circumscribed regions. Anoxia for 24 h NeuN labeling did not show areas of neuronal loss (Fig.1e). Most neurons had normal outlines. GFAP labeling showed swelling of the subpial region in two animals leading to a distortion of the subpial fibers (Fig. 1h). The (open) lumen of the vasculature appeared to be more prominent than in controls. CV staining showed an overall intact architecture with no evidence for infarction or major neuronal loss (Fig. 1b). A line of light labeling separated the neuronal layer from the ependyma in the cortex of two animals. Most neurons showed normal morphology and no signs of widespread shrinkage or nuclear disintegration were found. In areas of light background labeling, neuronal outlines were fuzzy suggesting the presence of edema. Perivascular halos of light staining were noted in basal areas. NIH-PA Author Manuscript Reoxygenation for 1d NIH-PA Author Manuscript Induction of Stress proteins under anoxia NeuN labeling did not reveal neuronal deficits. Most neurons had normal outlines and distinct nuclear labeling. In some regions, neurons appeared to be compressed by small vacuoles. GFAP labeling did not indicate any areas with formation of reactive glial cells, i.e. star-shaped astrocytes were not seen. The overall intensity of labeling of fine glial fibers was variable among brain regions. CV staining showed vacuolization between the ependyma and cortical neuronal layer. Perineuronal vacuolization was also seen in the cortex which lead to some dark staining, but overt ischemic necrosis was not seen. Some fiber rich areas in the basal regions showed vacuolization. Reoxygenation for 3d No areas of neuronal loss or infarction were seen in NeuN-labeled sections (Fig. 1f). The neuronal layer of the cortex was swollen in some but not all animals. GFAP staining showed a regular pattern of glial processes with no evidence for the formation of reactive astrocytes (Fig. 1i). CV staining showed minor swelling of the cortex (Fig. 1c). No infarction was evident. It has been previously suggested that physiological stress in the anoxia tolerant turtle is greatest during the initial transition phase (1-2 h) to the hypometabolic state, associated with a drop in ATP and the specific upregulation of protective MAPK pathways (Lutz and Milton 2004; Milton et al., 2008). The results of this study show increased expression of several heat shock proteins during the initial transition phase which remain elevated throughout the extended maintenance phase of long-term anoxia (Fig. 2). Hsp72 As previously reported by Prentice et al. (2004), Hsp72 was expressed at high basal concentrations in the normoxic turtle brain, in direct contrast to the rodent brain where Hsp72 is typically not detectable in the normal brain unless induced by stress. Hsp72 increased significantly to 207 ± 12% of basal during the initial hour of anoxia and remained elevated at 4 h and 24 h anoxia (Fig. 2a). J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 7 Hsp60 NIH-PA Author Manuscript Exposure to 1 h and 24 h of anoxia resulted in a moderate increases to 111 ± 11% and 119 ± 7% of basal, respectively (Fig. 2b). Hsp60 levels at 4 h anoxia were also elevated, but the difference from control was not statistically significant. HO-1 HO-1 levels were relatively low, though detectable, in normoxic controls. Anoxia-stimulated increases in HO-1, however, were similar to other stress proteins. HO-1 protein levels increased by 1h anoxia (163±11%) and remained elevated through 24 h anoxia, increasing to 228 ± 28 % of basal (Fig. 2c). Hsp27 Relatively low levels of Hsp27 were detected under normoxic conditions; expression of Hsp27 was 1.8-fold to 2-fold over normoxic controls at 1 h and 24 h anoxia (Fig. 2d). Grp78 and Grp94 NIH-PA Author Manuscript To further understand the resistance of freshwater turtles to cell damage in anoxia, we investigated the expression of two ER-specific stress-induced chaperones, Grp78 and Grp94 (Fig. 2e - 2f), which are thought to protect cells against ER-related stress in ischemia (Kudo et al., 2008). Interestingly, the levels of Grp78 were not altered on exposure to anoxia (Fig. 2e). By contrast, Grp94 was strongly and progressively induced through anoxia. Anoxia increased the levels of Grp94 2.5-fold by 1 hr anoxia and four-fold by the end of 24 h of anoxia (Fig. 2f). Bcl2 and Bax NIH-PA Author Manuscript Based on rodent studies, where increased levels of the pro-apoptotic protein Bax and decreased levels of the protective Bcl-2 are associated with apoptosis, we hypothesized that the anoxia resistant turtle brain would instead increase Bcl-2 protein levels and possibly decrease Bax (Zhang et al., 2001) but more complex changes in both Bcl-2 and Bax were encountered. Levels of both proteins varied over the course of 24 h of anoxia, but surprisingly the changes in Bcl-2 and Bax at each time point were in the same direction. The expression levels of both proteins initially decreased significantly by the end of 1h anoxia, with the decrease in Bcl-2 levels slightly greater (to 32% of basal) than decreases in Bax (to 58% of basal expression). The expression of Bcl2 then increased through 4 h anoxia but remained below basal levels, at 67 ± 8% of control, while Bax did not change significantly from levels detected at 1 h anoxia (61 ±12% of normoxic control). By 24 h anoxic exposure, the expression of Bcl2 had increased to 110 ±24 % and Bax to 130 ± 10 % of normoxia. These results suggest that exposure to anoxia and reoxygenation do not alter the overall ratio of Bcl2 to Bax, then, except for an initial decrease in the first hour of anoxic exposure (Fig. 3a-3c). Caspase 3 Levels of cleaved/active caspase-3, a key player in the apoptotic cascade, were unchanged by anoxia. While levels of the inactive zymogen procaspase-3 increased continuously up to 2.3 fold by the end of 24 h anoxia, levels of cleaved (active) caspase-3 did not change significantly through anoxia and reoxygenation (Fig. 4a). Apoptosis Inducing Factor Neuronal death in ischemia also occurs through non-caspase mediated cell death pathways (Cregan et al., 2002 ; Cao et al., 2003) by way of alternate mediators like AIF whose nuclear J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 8 NIH-PA Author Manuscript translocation directs DNA fragmentation (Plesnila et al., 2004). Exposure to 4 h anoxia increased the expression of AIF significantly to 181 ± 29% of the basal level. Protein levels were not significantly different from the control at other time points in anoxia (Fig. 4b). Protein levels during reoxygenation Damage to mammalian systems is also significant upon reoxygenation, due in large part to resulting increases in oxygen free radical production that overwhelm antioxidant protections and trigger further cell death. However, cell death in vitro upon reoxygenation in turtles is low (Milton et al., 2007) presumably due both to high antioxidant levels (Rice et al., 1995; Storey, 2006) and the suppression of ROS formation upon reoxygenation (Milton et al., 2007). As HSPs are also protective upon reperfusion in mammals, it was thus of interest to determine if stress proteins would be further stimulated by reoxygenation in turtles (4 h anoxia/ 4 h recovery). While Hsp72, Hsp60, Hsp27 and Grp94 levels were elevated above controls at 4 h of anoxia, no further upregulation of these proteins was evident following 4 h of reoxygenation (Fig. 5a). In fact, the expression of Hsp72, Hsp60 and Hsp27 tended to return towards basal level after reoxygenation from anoxic increases, though the difference was not significant from 4h anoxia. By contrast, the expression of HO-1 increased further during reoxygenation to 135 ± 3% above 4h anoxic levels, or 2.5 fold above the normoxic basal levels. NIH-PA Author Manuscript The apoptotic regulators also changed, with significant increases in levels of Bcl2, Bax and Procaspase-3 by the end of 4 h of reoxygenation when compared to 4 h of anoxia (Fig. 5 b). The increases in Bcl2 and Bax returned them to basal levels upon reoxygenation, resulting again in a constant Bcl2 to Bax ratio. Procaspase-3 also increased during reoxygenation, though the cleaved form of Caspase-3 remained unchanged. Subcellular distribution of Bax and AIF NIH-PA Author Manuscript Anoxia and reoxygenation do not induce translocation of Bax from cytosol to mitochondria—During the induction of apoptosis, Bax typically moves from the cytosol to the mitochondria (Gross et al., 1998). Accordingly, the intracellular distribution of Bax was analyzed in the cytosolic and mitochondrial fractions of the turtle brain. Cytosolic Bax levels were not altered significantly during anoxia (87±9%) in comparison to the control. During reoxygenation, cytosolic Bax levels increased slightly (113±3%) from control but the change was not significant. Densitometric analysis showed that mitochondrial Bax levels under normoxic conditions are only 29±1.5% of cytosolic levels and do not change significantly in anoxia, at 25±6% of anoxic cytosolic levels. Reoxygenation induced a statistically significant increase in mitochondrial Bax, but levels remained low at 56±4% of cytosolic levels. Anoxia and Reoxygenation do not induce significant translocation of AIF from mitochondria to the nucleus—AIF translocates from the mitochondria to the nucleus in the caspase-independent pathway of apoptosis (Susin et al., 1999). Cytosolic and nuclear translocation of AIF from the mitochondria was evaluated on exposure to anoxia and reoxygenation in the turtle brain. Under normoxic conditions, cytosolic AIF levels were 36±2% of the mitochondrial AIF levels. Nuclear AIF levels were also lower (76±5%) than the mitochondrial levels. Anoxia increased the mitochondrial AIF to 107±3% of control (not significant) but both cytosolic and nuclear levels remained low at 56±2% and 47±6% of mitochondrial levels, respectively. On reoxygenation, mitochondrial AIF expression decreased (to 76±5% of normoxic control) and cytosolic (73±6%) and nuclear distribution increased, with nuclear values returning to normoxic control levels (78±4%). J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 9 Discussion NIH-PA Author Manuscript NIH-PA Author Manuscript We investigated the expression of stress proteins and apoptotic regulators in the anoxiatolerant turtle during anoxia and reoxygenation: in the initial transition phase to anoxic hypometabolism (1 h), during long term anoxia (4 and 24 h), and in the reoxygenation phase. The results of our study show a general pattern of high constitutive levels and further rapid induction of stress proteins within the first hour of anoxia; elevated levels continue through 24 h of anoxia, suggesting activated endogenous protective mechanisms. Previous studies in our laboratory reported the presence of high basal levels and further upregulation of inducible Hsp72 during anoxia, suggesting a degree of constitutive preconditioning in the turtle brain (Prentice et al., 2004). Our present study reconfirms the presence and further anoxia-induced upregulation of Hsp72 as well as reports increases in other stress proteins including Hsp27, Hsp60, HO-1 and Grp94 during anoxia. Unlike the mammalian brain, where Hsp27 is not detectable prior to ischemic stress and requires as long as 24 h to increase significantly (Higashi et al., 1994; Wagstaff et al., 1996), Hsp27 was also detectable under basal conditions in the turtle and increased significantly during the initial hour of anoxia. HSP increases are associated with neuronal protection, and indeed, our histological and immunohistochemical studies indicate that a period of 24 h of anoxia does not lead to overt neuronal loss or infarction despite variable vacuolization in the cortex, nor does 1 to 3 d recovery following anoxia. This is the first study to examine the brain directly for indications of anoxia/reoxygenation damage, confirming histologically the anoxia resistance of the turtle brain. Thus, the changes in stress proteins found here with Western blot analysis occurred in tissue that is going to survive overtly intact for at least three days. Elevated heat shock proteins may protect against apoptosis. Apoptosis is induced by two distinct but interconnected pathways: the extrinsic pathway activated in response to stimulation of death receptors, and the intrinsic pathway initiated by cellular stresses that activate pro-apoptotic members of the Bcl-2 family, which in turn target the mitochondria. In intrinsic apoptosis, pro-apoptotic proteins cause permeabilization of the mitochondria, resulting in the release of apoptogenic factors that activate downstream effector caspases, including caspase-3 (Cao et al., 2001; Cao et al., 2002; Clemons et al., 2005). Caspaseindependent apoptosis may also occur, as through AIF (Cao et al., 2003) or the calpains. In contrast to the significant increases in most HSPs examined, changes indicative of apoptosis were not observed in anoxia or reoxygenation; neither the decrease in the Bcl-2:Bax ratio which occurred in the initial transition to anoxia nor AIF increases were apparently significant enough to trigger apoptosis in the face of elevated intracellular protection. In addition, protection was mediated by increased levels of adenosine (Nilsson and Lutz, 1992) that activated ERK1/2 and Akt (Milton et al., 2008). NIH-PA Author Manuscript Much of this conferred neuroprotection is thought to occur through interactions that block key steps of the intrinsic apoptotic pathways and shift cellular machinery towards cell survival (Abe and Nowak, 2004; Halaby et al., 2004; Zhu et al., 2007), including through the inhibition of the c-jun kinase (JNK) pathway (Gabai et al., 2000; Volloch et al., 2000; Gabai et al., 2002), blocking the translocation of the pro-apoptotic protein Bax to the mitochondria (Wang et al., 1999), and preventing Cyt-c release, caspase-3 activation and the translocation of AIF (Li et al., 2000; Li et al., 2002; Wang et al., 2002; Ruchalski et al., 2006). Such mechanisms allow Hsp72 to be protective even during times of ATP depletion (Wang and Borkan, 1996), which in mammalian cells is also when Bcl-2 levels decrease and Bax increases (Feldenberg et al., 1999). Interestingly, the greatest decrease in Bcl2 and change in the Bcl-2:Bax ratio in the turtle brain occurred in the initial hour of anoxia, when it has been shown that ATP levels temporarily decrease (Lutz et al., 1984). Changes in phosphorylation of both Bcl-2 and Bax by MAP kinases may have lead to the rapid decrease of proteins detectable with the present antibodies (Ashraf et al., 2001). The parallel decline of Bcl-2 and J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 10 Bax early after the onset of anoxia is not paralleled by changes in the ischemic rodent brain, and it appears to be a specific feature of the turtle brain. NIH-PA Author Manuscript Also unlike mammalian models, where increases in many HSPs are delayed, increases in Hsp72, Hsp27, and HO-1 all were apparent in the turtle within the initial hour of anoxia, which makes it likely they could offer neuroprotection concurrent with ATP depletion. It is also during this initial hour, when physiological stress is presumably the greatest, that phospho-Akt levels temporarily increase to over 300% of basal (along with a 6-fold increase in phosphorylated ERK) (Milton et al., 2008). Hsp27’s protective role is thought to occur in part through PI3K activation and Akt increases that inhibit Bax activation (Havasi et al., 2008), while ischemic tolerance conferred by Hsp60 may also be due to its interactions with Bax (Kirchoff et al., 2002, Chandra et al., 2007). Akt and ERK activation have also been associated with HO-1 protection against H2O2-induced apoptosis (Kim et al., 2008). NIH-PA Author Manuscript Overexpression of HO-1 resulted in reduced infarct size in the rat focal ischemic model (Panahian et al., 1999) and neurons overexpressing HO-1 are more resistant to glutamate excitotoxicity (Chen et al., 2000). HO-1 in mammals is mainly induced in astroglial or microglial cells (Geddes et al., 1996), and its synaptic localization led Bechtold and Brown (2000) to suggest that HO-1 may thus be involved in the protection and repair of synapses. The large increases we observed in anoxia and upon reoxygenation, then, are of interest as it has been suggested that the turtle maintains synaptic function during long-term anoxia, with the continued low-level release and reuptake of neurotransmitters by both glia and neurons (Milton et al., 2002; Thompson et al., 2007) and periodic bursts of electrical activity (Fernandes et al., 1997). Hsp27 is also primarily expressed in astroglia post-ischemia (Kato et al., 1995; Currie et al, 2000), and Hsp27 gene transfer offers protection against ischemia (Badin et al., 2006). The protective effect of Hsp27 might be due to its multiple antiapoptotic interactions, e.g. preventing formation of the apoptosome (Garrido et al., 1999) and activation of Bax (Havasi et al., 2008), sequestration of cytochrome c (Bruey et al., 2000), and interactions with Akt (Lannaeu et al., 2008) and Bid (Conconnaon et al., 2001). In whole turtle brain, levels of activated Akt increase in the initial hour of anoxia (Milton et al., 2008), while here we report decreases in Bax. While changes in Akt levels are related to adenosine stimulation, the link between adenosine, Akt, and cellular protection is not necessarily direct, and it is likely that the turtle has multiple and redundant interacting pathways for neuroprotection. The observation that anoxia-induced increases in Hsp27 were among the largest changes detected again suggests that the protection and function of glial cells may be a critical, and previously overlooked, aspect of brain survival in the anoxic turtle. NIH-PA Author Manuscript Involvement of ER stress chaperones in ischemic protection is relatively well established, though reports of Grp78 and Grp94 elevations after ischemia in rodent models are variable (Chen et al, 1996; Hayashi et al., 2003; Kim et al., 2003). Grp94 is again thought to be primarily a glial response (Kim et al., 2003; Bando et al., 2003; Jeon et al., 2004), while Grp78 responses are neuronal (Kudo et al., 2008). In the anoxic turtle brain, Grp94 was upregulated whereas Grp78/BiP did not change, which could be due to regulation by different transcription factors controlling the ER stress response (Schroeder and Kaufman, 2005) or again reflect differences in neuronal and glial responses to anoxia. Further studies on the ER stress response in the turtle brain would be of interest, particularly as the ER is a significant intracellular calcium store. In contrast to large increases at 1h anoxia with further elevation by the end of 24 h of anoxia in the primarily glial (in mammals) Hsp27, Grp94 and HO-1, Hsp72 increases were not observed after 4 h. Preliminary studies with immunohistochemistry indicate neuronal expression of Hsp72 in the normal and anoxic turtle brain (Kesaraju and Schmidt-Kastner, J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 11 NIH-PA Author Manuscript unpublished observations) which is in line with neuronal expression in rodents after ischemia (Vass et al, 1988). Together, then, the data suggest that expression in astrocytes and/or microglial cells is a common denominator for stress genes in mammals which show a greater increase of protein levels during anoxia in the turtle brain. This leads to the intriguing question of whether astrocytes are under increasing metabolic stress in the turtle brain, because their working load for maintaining basal neuron functions must be augmented during anoxia. In mammals, astrocytes provide nutritional support to neurons, and Pellerin and Magistretti (1994) have shown that the uptake of glutamate stimulates astrocytes to shuttle lactate to neurons. The turtle brain is known to undergo swelling during anoxia (Cserr et al., 1988) and much of this swelling can be predicted to reflect intracellular edema in astrocytes (Sykova et al., 1998). The present immunolabeling for GFAP may also be taken to indicate swelling of the cortical astrocytes during anoxia. In view of the lively debate regarding astrocyte-to-neuron transfer of energy-rich metabolites (Meeks and Mennerick, 2003), it is tempting to speculate that astrocytes may continue to support neurons throughout anoxia whereas neurons themselves shut down much of their function. NIH-PA Author Manuscript Pro-caspase-3 levels also increased significantly in the anoxic turtle brain, as has been seen at the mRNA and protein levels in the post-ischemic rodent brain, both in vulnerable neurons and in neurons protected by preconditioning (Cao et al., 2002; Tanaka et al., 2004). Importantly, however, the active cleaved caspase-3 was not elevated in the anoxic turtle brain, a condition also reported in the preconditioned mammalian model (Tanaka et al., 2004). One explanation is that HSPs blocked upstream caspases or calpain during anoxia (Garrido et al, 2001). These initial studies show that the turtle brain expresses regulators of apoptosis under basal conditions, but that protective changes induced by anoxia prevent drastic alterations in Bcl2: Bax ratios and caspase-3 activation. Further support for the contention that the apoptotic cascade is suppressed in the anoxic turtle brain comes from the subcellular fractionation analysis, where we found no evidence for the anoxic translocation of Bax from the cytosol to mitochondria, nor any movement of AIF between the mitochondria and nucleus. We have noted an increase in cell death upon reoxygenation in cultured neurons (Milton et al., 2007), however, and indeed in the whole brain there was a significant increase in mitochondrial Bax upon reoxygenation. The increase in mitochondrial Bax though was not accompanied by increased levels of cleaved caspase-3, nor did AIF appear to shift from the mitochondria to the nucleus, suggesting that the mitochondria are stabilized against apoptosis. Future investigations of apoptosis inhibitors such as XIAP down-stream to the mitochondrial pathway would thus be of interest. NIH-PA Author Manuscript Reoxygenation for 4 h following 4 h anoxia did not further stimulate expression of Hsp70, Hsp60 or Hsp27, nor were there changes in the Bcl2: Bax ratio, indicative again of the stabilization of apoptotic regulators. Additional increases in HO-1 in reoxygenation beyond the already significant increases observed in anoxia suggest involvement in protection against potential oxidative stress. The turtle is endowed with superior protective mechanisms to cope with reoxygenation stress including enhanced antioxidant systems (Rice et al., 1995; Wilmore et al., 1997; Storey, 2006) and suppression of ROS (reactive oxygen species) formation (Milton et al, 2007; Pamenter et al., 2007) and high levels of HO-1 would add further protection. So unlike mammalian systems, where reoxygenation often triggers massive increases in ROS release and cell death, the turtle appears to experience little oxidative stress, and thus the unaltered expression of most stress proteins after reoxygenation would be due to an absence of triggers for stress induction. Alternatively, of course, high “banked” levels of stress proteins may be getting used up as the turtle emerges from its hypometabolic state and returns to basal metabolism, with the requisite upregulation of transcription (Prentice et al., 2003) and protein synthesis (Fraser et J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 12 NIH-PA Author Manuscript al., 2001). The return of elevated stress protein levels exactly to basal, however, argues that this is a controlled return to normoxic status rather than a response to increases in protein synthesis and/or damage, and in turtles protein synthesis likewise returns rapidly to preanoxic levels upon recovery (Fraser et al., 2001). By contrast, hypoxia/ ischemia studies in rodent models show induction of major stress proteins within 1 h -7 d after insult in the salvageable tissue (Chen et al., 1996, Kokubo et a.l, 2003, Nishino and Nowak, 2004) despite an overall reduction in protein synthesis extended till 72 h after the insult (Kokubo et al., 2003). NIH-PA Author Manuscript Taken together, then, the data point to complex molecular mechanisms that result in stable apoptotic machinery and increased stress protein protection (Table I). In the early phase of anoxia, as the brain transitions to a state of deep hypometabolism, a rapid induction of stress proteins and depression of Bcl2 and Bax stabilizes the cell at a time of elevated physiological stress and decreased ATP levels. During the ensuing long term maintenance phase of anoxia, there is a steady increase in overall stress protein levels and a constant low level of pro-apoptotic proteins, tipping the balance towards survival. Although the localization of stress proteins in the turtle brain are not yet known, interestingly, the stress proteins that were most enhanced are the ones mostly localized to glial cells in mammalian studies, suggesting an increased functionality of glial cells for maintenance of neuronal integrity that has not been previously noted. Reoxygenation did not further elevate the stress protein levels, supporting earlier studies indicating that turtles possess innate mechanisms to suppress oxidative stress. A better understanding of the molecular mechanisms that promote survival in the anoxia tolerant brain, honed though millions of years of evolution, may yield new therapeutic targets for mammalian anoxic/ischemic deficits. Acknowledgments This work was funded by an NIH RO-15 grant (grant# 1 R15 NS048909-01) to SLM. References NIH-PA Author Manuscript Abe H, Nowak TS Jr. Induced hippocampal neuron protection in an optimized gerbil ischemia model: insult thresholds for tolerance induction and altered gene expression defined by ischemic depolarization. J. Cereb. Blood Flow Metab. 2004; 24:84–97. [PubMed: 14688620] Ashraf QM, Zanelli SA, Mishra OP, Delivoria-Papadopoulos M. Phosphorylation of Bcl2 and Bax proteins during hypoxia in newborn piglets. Neurochem Res. 2001; 26:1–9. [PubMed: 11358275] Badin RA, Lythgoe MF, Vander Weerd L, Thomas DL, Gadian DG, Latchman DS. Neuroprotective effects of virally delivered HSPs in experimental stroke. J. Cereb. Blood Flow Metab. 2006; 26:371–381. [PubMed: 16079790] Bando Y, Katayama T, Kasai K, Taniguchi M, Tamatani M, Tohyama M. GRP94(94 kDa glucoseregulated protein) suppresses ischemic neuronal cell death against ischemia/reperfusion injury. Eur. J. Neurosci. 2003; 18:829–840. [PubMed: 12925009] Bechtold DA, Brown IR. Heat shock proteins Hsp27 and Hsp32 localize to synaptic sites in the rat cerebellum following hyperthermia. Brain Res. Mol Brain Res. 2000; 75:309–320. [PubMed: 10686353] Bickler PE, Donohoe PH, Buck LT. Hypoxia-induced silencing of NMDA receptors in turtle neurons. J. Neurosci. 2000; 20:3522–3528. [PubMed: 10804192] Bruey JM, Duchase C, Bonniaud P, et al. Hsp27 negatively regulates cell death by interacting with cytochrome c. Nat. Cell Biol. 2000; 2:645–652. [PubMed: 10980706] Cameron, JN. The respiratory physiology of animals. Oxford University Press; New York, Oxford: 1989. Chandra D, Choy G, Tang DG. Cytosolic accumulation of HSP60 during apoptosis with or without apparent mitochondrial release: evidence that its pro-apoptotic or pro-survival functions involve J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 13 NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript differential interactions with caspase-3. J. Biol. Chem. 2007; 282:31289–31301. [PubMed: 17823127] Cao G, Pei W, Lan J, Stetler RA, Luo Y, Nagayama T, Graham SH, Yin XM, Simon RP, Chen J. Caspase —activated DNase DNA fragmentation factor 40 mediates apoptotic DNA fragmentation in transient cerebral ischemia and in neuronal cultures. J. Neurosci. 2001; 21:4678–4690. [PubMed: 11425895] Cao G, Luo Y, Nagayama T, Pei W, Stetler RA, Graham SH, Chen J. Cloning and characterization of rat caspase-9: implications for a role in mediating caspase-3 activation and hippocampal cell death after transient cerebral ischemia. J. Cereb. Blood Flow Metab. 2002; 22:534–546. [PubMed: 11973426] Cao G, Clark RS, Pei W, Yin W, Zhang F, Sun FY, Graham SH, Chen J. Translocation of apoptosisinducing factor in vulnerable neurons after transient ischemia and in neuronal cultures after oxygen-glucose deprivation. J. Cereb. Blood Flow Metab. 2003; 23:1137–1150. [PubMed: 14526224] Chang J, Knowlton AA, Wasser JS. Expression of heat shock proteins in turtle and mammal hearts: relationship to anoxia tolerance. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2000; 278:R209– R214. [PubMed: 10644641] Chen J, Graham SH, Zhu RL, Simon RP. Stress proteins and tolerance to focal cerebral ischemia. J. Cereb. Blood Flow Metab. 1996; 16:566–577. [PubMed: 8964795] Chen K, Gunter K, Maines MD. Neurons overexpressing heme-oxygenase-1 resist oxidative stressmeditated cell death. J. Neurochem . 2000; 75:304–313. [PubMed: 10854275] Clemons NJ, Buzzard K, Steel R, Anderson RL. Hsp72 inhibits Fas-mediated apoptosis upstream of the mitochondria in type II cells. J .Biol. Chem. 2005; 280:9005–9012. [PubMed: 15632129] Conconnaon CG, Orrenius S, Samali A. Hsp27 inhibits cytochrome c-mediated caspase activation by sequestering both pro-caspase-3 and cytochrome c. Gene Expr. 2001; 9:195–201. [PubMed: 11444529] Cregan SP, Fontin A, MacLaurin JG, et al. Apoptosis-inducing factor is involved in the regulation of caspase-independent neuronal cell death. J. Cell Biol. 2002; 158:507–517. [PubMed: 12147675] Cserr HF, DePasquale M, Jackson DC. Brain and cerebrospinal fluid ion composition after long-term anoxia in diving turtles. Am. J. Physiol. 1988; 255:R338–343. [PubMed: 3136673] Currie RW, Ellison JA, White RF, Feuerstein GZ, Wang X, Barone FC. Benign focal ischemic preconditioning induces neuronal Hsp70 and prolonged astrogliosis with expression of Hsp27. Brain Res. 2000; 863:169–181. [PubMed: 10773205] Feldenberg LR, Thevananther S, del Rio M, de Leon M, Devarajan P. Partial ATP depletion induces Fas- and caspase-mediated apoptosis in MDCK cells. Am J Physiol Renal Physiol. 1999; 276:F837–F846. Fernandes JA, Lutz PL, Tannenbaum A, Todorov AT, Liebovitch L, Vertes R. Electroencephalogram activity in the anoxic turtle brain. Am. J. Physiol. 1997; 273:R911–919. [PubMed: 9321867] Fraser KP, Houlihan DF, Lutz PL, Leone-Kabler S, Manuel L, Brechin JG. Complete suppression of protein synthesis during anoxia with no post-anoxia protein synthesis debt in the red-eared slider turtle Trachemys scripta elegans. J. Exp.Biol. 2001; 204:4353–4360. [PubMed: 11815659] Gabai VL, Mabuchi K, Mosser DD, Sherman MY. Hsp72 and stress kinase c-jun N-terminal kinase regulate the bid-dependent pathway in tumor necrosis factor induced apoptosis. Mol. Cell Biol. 2002; 22:3415–3424. [PubMed: 11971973] Gabai VL, Meriin AB, Yaglom JA, Wei JY, Mosser DD, Sherman MY. Suppression of stress kinase JNK is involved in HSP72-medited protection of myogenic cells from transient energy deprivation HSP72 alleviates the stress-induced inhibition of JNK dephosphorylation. J. Biol. Chem. 2000; 275:38088–38094. [PubMed: 10978340] Garrido C, Bruey JM, Fromentin A, Hammann A, Arrigo AP, Solary E. HSP27 inhibits cytochrome cdependent activation of procaspase-9. FASEB J. 1999; 14:2061–2070. [PubMed: 10544189] Garrido C, Gurbuxiani S, Ravagnan L, Kroemer G. Heat shock proteins: endogenous modulators of apoptotic cell death. Biochem. Biophys. Res. Commun. 2001; 286:433–442. [PubMed: 11511077] J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 14 NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript Geddes JW, Pettigrew LC, Holtz ML, Craddock SD, Maines MD. Permanent focal and transient global cerebral ischemia increase glial and neuronal expression of heme oxygenase-1, but not heme oxygenase-2, protein in rat brain. Neurosci. Lett. 1996; 210:205–208. [PubMed: 8805131] Gross A, Jockel J, Wei MC, Korsmeyer SJ. Enforced dimerization of BAX results in its translocation, mitochondrial dysfunction and apoptosis. EMBO J. 1998; 17:3878–3885. [PubMed: 9670005] Haddad JJ. Discordant tissue specific expression of SAPK/MAPK(JNK) —related cofactors in hypoxia and hypoxia/reoxygenation in a model of anoxia-tolerance. Protein Pept. Lett. 2007; 14:373–380. [PubMed: 17504095] Haddad JJ. The role of Bax/Bcl2 and procaspase peptides in hypoxia/reperfusion — dependent regulation of MAPK (ERK): discordant proteomic effect. Protein Pept. Lett. 2007; 14:361–371. [PubMed: 17504094] Halaby IA, Takeda Y, Yufu K, Nowak TS Jr. Pulsinelli WA. Depolarization thresholds for hippocampal damage, ischemic preconditioning, and changes in gene expression after global ischemia in the rat. Neurosci. Lett. 2004; 372:12–16. [PubMed: 15531079] Havasi A, LI Z, Wang Z, Martin JL, Botla V, Ruchalski K, Schwartz JH, Borkan SC. Hsp27 inhibits Bax activation and apoptosis via a phosphatidylinositol 3 —kinase depenedant mechanism. J. Biol. Chem. 2008; 283:12305–12313. [PubMed: 18299320] Hayashi T, Saito A, Okuno S, Ferrand-Drake M, Chan PH. Induction of GRP78 by ischemic preconditioning reduces endoplasmic reticulum stres and prevents delayed neuronal cell death. J. Cereb. Blood Flow Metab. 2003; 8:849–961. Higashi T, Takechi H, Uemura Y, Kikuchi H, Nagata K. Differential induction of mRNA species encoding several classes of stress proteins following focal cerebral ischemia in rats. Brain Res. 1994; 650:239–248. [PubMed: 7953688] Hwang IK, Ahn HC, Yoo KY, Lee JY, Suh HW, Kwon YG, Cho JH, Won MH. Changes in immunoreactivity of HSP60 and its neuroprotective effects in the gerbil hippocampal CA1 region induced by transient ischemia. Exp. Neurol. 2007; 208:247–256. [PubMed: 17931626] Jeon GS, Park SW, Kim DW, Seo JH, Cho J, Lim SY, Kim SD, Cho SS. Glial expression of the 90kDA heat shock protein (HSP90) and the 94 kDA glucose regulated protein (GRP94) following an excitotoxic lesion in the mouse hippocampus. Glia. 2004; 48:250–258. [PubMed: 15390117] Kalman M, Kiss A, Majorossy K. Distribution of glial fibrillary acidic protein — immunopositive structures in the brain of the red-eared freshwater turtle (Pseudemys scripta elegans). Anat. Embryol. (Berl). 1994; 189:421–434. [PubMed: 7522421] Kato H, Liu Y, Kogure K, Kato K. Induction of 27-kDa heat shock protein following cerebral ischemia in a rat model of ischemic tolerance. Brain Res. 1994; 634:235–244. [PubMed: 8131073] Kato H, Kogure K, Liu XH, Araki T, Kato K, Itoyama Y. Immunohistochemical localization of the low molecular weight stress protein HSP27 following focal cerebral ischemia in the rat. Brain Res. 1995; 679:1–7. [PubMed: 7648252] Kim SW, Park S, You KH, Kwon O-Y. Expression of the endoplasmic reticulum chaperone GRP94 gene in ischemic gerbil brain. Z. Naturforsch[C]. 2003; 58:736–739. Kim JW, Li MH, Jang JH, Na HK, Song NY, Lee C, Johnson JA, Surh YJ. 15-Deoxy-Delta(12,14)prostaglandin J(2) rescues PC12 cells from H(2)O(2)-induced apoptosis through Nrf2-mediated upregulation of heme oxygenase-1: Potential roles of Akt and ERK1/2. Biochem Pharmacol. 2008 Kirchoff SR, Gupta S, Knowlton AA. Cytosolic heat shock protein 60, apoptosis, and myocardial injury. Circulation. 2002; 105:2899–2904. [PubMed: 12070120] Kirino T. Ischemic tolerance. J. Cereb. Blood Flow Metab. 2002; 22:1283–1296. [PubMed: 12439285] Kokubo Y, Liu J, Rajdev S, Kayama T, Sharp FR, Weinstein PR. Differential cerebral protein synthesis and heat shock protein 70 expression in the core and penumbra of rat brain after transient focal ischemia. Neurosurgery. 2003; 53:186–190. [PubMed: 12823888] Kudo T, Kanemoto S, Hara H, Morimoto N, Morihara T, Kimura R, Tabita T, Imaizumi K, Takeda M. A molecular chaperone inducer protects neurons from ER stress. Cell Death Differ. 2008; 15:364– 375. [PubMed: 18049481] Lanneau D, Brunet M, Frisan E, Solary E, Fontenary M, Garrido C. Heat shock proteins: essential proteins for apoptosis regulation. J. Cell Mol. Med. 2008; 12:743–761. [PubMed: 18266962] J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 15 NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript Li C, Lee J, Ko Y, Kim J, Seo J. Heat shock protein 70 inhibits apoptosis downstream of cytochrome c release and upstream of caspase 3 activation. J. Biol. Chem. 2000; 275:25665–25671. [PubMed: 10806214] Li F, Mao HP, Ruchalski KL, Wang YH, Choy W, Schwartz JH, Borkan SC. Heat stress prevents mitochondrial injury in ATP-depleted renal epithelial cells. Am. J. Physiol. Cell Physiol. 2002; 283:C917–C926. [PubMed: 12176748] Lutz PL, McMohan P, Rosenthal M, Sick TJ. Relationships between aerobic and anaerobic energy production in turtle brain in situ. Am. J. Physiol. 1984; 247:R740–744. [PubMed: 6093562] Lutz PL, Milton SL. Negotiating brain anoxia survival in the turtle. J Exp Biol. 2004; 207:3141–3147. [PubMed: 15299035] Massa SM, Swanson RA, Sharp FR. The stress gene response in brain. Cerebrovasc. Brain Metab. Rev. 1996; 8:95–158. [PubMed: 8727184] Matsumori Y, Northington FJ, Hong SM, Kayama T, Sheldon RA, Vexler ZS, Ferriero DM, Weinstein PR, Liu J. Reduction of caspase-8 and -9 cleavage is associated with increased c-FLIP and increased binding of Apaf-1 and Hsp70 after neonatal hypoxic/ischemic injury in mice overexpressing Hsp70. Stroke. 2006; 37:507–512. [PubMed: 16397188] Meeks JP, Mennerick S. Feeding hungry neurons: astrocytes deliver food for thought. Neuron. 2003; 37:187–189. [PubMed: 12546812] Mehta SL, Manhas N, Raghubir R. Molecular targets in cerebral ischemia for developing novel therapeutics. Brain Res. Rev. 2007; 54:34–66. [PubMed: 17222914] Milton SL, Thompson JW, Lutz PL. Mechanisms for maintaining extracellular glutamate levels in the anoxic turtle striatum. Am.J. Physiol. Regul. Integr. Comp. Physiol. 2002; 282:R1317–1323. [PubMed: 11959671] Milton SL, Prentice HM. Beyond anoxia: the physiology of metabolic downregulation and recovery in the anoxia-tolerant turtle. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 2007; 147:277–290. [PubMed: 17049896] Milton SL, Nayak G, Kesaraju S, Kara L, Prentice HM. Suppression of reactive oxygen species production enhances neuronal survival in vitro and in vivo in the anoxia-tolerant turtle Trachemys scripta. J.Neurochem. 2007; 101:993–1001. [PubMed: 17326763] Milton SL, Dirk LJ, Kara LF, Prentice HM. Adenosine modulates ERK1/2, PI3K/Akt, and p38MAPK activation in the brain of the anoxia-tolerant turtle Trachemys scripta. J.Cereb. Blood Flow Metab. 2008; 28:1469–1477. [PubMed: 18506199] Mullen RJ, Buck CR, Smith AM. NeuN, a neuronal specific nuclear protein in vertebrates. Development. 1992; 116:201–211. [PubMed: 1483388] Nilsson GE, Lutz PL. Adenosine release in the turtle brain; a possible mechanism for anoxic survival. J. Exp.Biol. 1992; 162:345–351. Nishi S, Taki W, Uemura Y, Higashi T, Kikuchi H, Kudoh H, Satoh M, Nagata K. Ischemic tolerance due to the induction of HSP70 in a rat ischemic circulation model. Brain Res. 1993; 615:281–288. [PubMed: 8364736] Nishino K, Nowak TS Jr. Time course and cellular distribution of hsp27 and hsp72 stress protein expression in a quantitative gerbil model of ischemic injury and tolerance: thresholds for hsp72 induction and hilar lesioning in the context of ischemic preconditioning. J.Cereb.Blood Flow Metab. 2004; 24:167–178. [PubMed: 14747743] Obrenovitch TP. Molecular physiology of preconditioning-induced brain tolerance to ischemia. Physiol. Rev. 2008; 88:211–247. [PubMed: 18195087] Pamenter ME, Richards MD, Buck LT. Anoxia induced changes in reactive oxygen species and cyclic nucleotides in the painted turtle. J. Comp. Physiol. 2007; 177:473–481. Panahian N, Yoshiura M, Maines MD. Overexpression of heme oxygenase 1 in a model of permanent middle cerebral artery occlusion in transgenic mice. J. Neurochem. 1999; 72:1187–1203. [PubMed: 10037492] Pellerin L, Magistretti PJ. Glutamate uptake into astrocytes stimulates aerobic glycolysis: a mechanism coupling neuronal activity to glucose utilization. Proc. Natl. Acad. Sci. U S A. 1994; 91:10625– 10629. [PubMed: 7938003] J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 16 NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript Perez, Pinzon M.A.; Rosenthal, M.; Sick, TJ.; Lutz, PL.; Pablo, J.; Mash, D. Downregulation of sodium channels during anoxia: a putative survival strategy of turle brain. Am. J. Physiol. 1992; 262:R712–715. [PubMed: 1314518] Plesnila N, Zhu C, Culmsee C, Groger M, Moskowitz MA, Blomgren K. Nuclear translocation of apoptosis —inducing factor after focal cerebral ischemia. J. Cereb. Blood Flow Metab. 2004; 24:458–466. [PubMed: 15087715] Powers AS, Reiner A. The distribution of cholinergic neurons in the central nervous system of turtles. Brain Behav. Evol. 1993; 41:326–345. [PubMed: 8324620] Prentice HM, Milton SL, Scheurle D, Lutz PL. Gene transcription of brain voltage — gated potassium channels is reversibly regulated by oxygen supply. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2003; 285:R1317–1321. [PubMed: 14615400] Prentice HM, Milton SL, Scheurle D, Lutz PL. The upregulation of cognate and inducible heat shock proteins in the anoxic turtle brain. J. Cereb. Blood Flow Metab. 2004; 24:826–8. [PubMed: 15241191] Ramaglia V, Buck LT. Time-dependent expression of heat shock proteins 70 and 90 in tissues of the anoxic western painted turtle. J. Exp. Biol. 2004; 207:3775–3784. [PubMed: 15371485] Rice ME, Lee EJ, Choy Y. High levels of ascorbic acid, not glutathione, in the CNS of anoxia-tolerant reptiles contrasted with levels in anoxia-intolerant species. J. Neurochem. 1995; 64:1790–1799. [PubMed: 7891107] Ruchalski K, Mao H, Li Z, Wang Z, Gillers S, Wang Y, Mosser DD, Gabai V, Schwartz JH, Borkan SC. Distinct Hsp70 domains mediate apoptosis —inducing factor release and nuclear accumulation. J. Biol. Chem. 2006; 281:7873–7880. [PubMed: 16407317] Saito A, Tominaga T, Chan PH. Neuroprotective role of neurotrophins: relationship between nerve growth factor and apoptotic cell survival pathway after cerebral ischemia. Curr. Atheroscler. Rep. 2005; 7:268–273. [PubMed: 15975319] Schmidt-Kastner R, Heim C, Sontag KH. Damage of substantia nigra pars reticulata during pilocarpine-induced status epilepticus in the rat: immunohistochemical study of neurons, astrocytes and serum-protein extravasation. Exp. Brain Res. 1991; 86:125–140. [PubMed: 1756784] Schmidt-Kastner R, Aguirre-Chen C, Saul I, Yick L, Hamasaki D, Busto R, Ginsberg MD. Astrocytes react to oligemia in the forebrain induced by chronic bilateral common carotid artery occlusion in rats. Brain Res. 2005; 1052:28–39. [PubMed: 16023090] Schroeder M, Kaufman RJ. The mammalian unfolded protein response. Ann. Rev. Biochem. 2005; 74:739–789. [PubMed: 15952902] Stankiewicz AR, Lachapelle G, Foo CP, Radicioni SM, Mosser DD. Hsp70 inhibits heat-induced apoptosis upstream of mitochondria by preventing Bax translocation. J Biol Chem. 2005; 280:38729–38739. [PubMed: 16172114] Storey KB. Reptile freeze tolerance: metabolism and gene expression. Cryobiology. 2006; 52:1–16. [PubMed: 16321368] Storey KB. Anoxia tolerance in turtles: metabolic regulation and gene expression. Comp. Biochem. Physiol. A. Mol. Integr. Physiol. 2007; 147:263–276. [PubMed: 17035057] Susin SA, Lorenzo HK, ZamZami N, et al. Molecular characterization of mitochondrial apoptosisinducing factor. Nature. 1999; 397(6718):441–446. [PubMed: 9989411] Sykova E, Mazel T, Simonova Z. Diffusion constraints and neuron-glia interaction during aging. Exp. Gerontol. 1998; 33:837–851. [PubMed: 9951627] Tanaka H, Yokoto H, Jover T, Cappuccio I, Calderone A, Simionescu M, Bennett MV, Zukin RS. Ischemic preconditioning : neuronal survival in the face of caspase-3 activation. J. Neurosci. 2004; 17:2750–2759. [PubMed: 15028768] Thompson JW, Prentice HM, Lutz PL. Regulation of extracellular glutamate levels in the long term anoxic turtle striatum: coordinated activity of glutamate transporters, adenosine, K(ATP) (+) channels and GABA. J. Biomed. Sci. 2007; 14:809–817. [PubMed: 17629717] Valentim LM, Geyer AB, Tavares A, Cimarosti H, Worm PV, Rodnight R, Netto CA, Salbego CG. Neuroscience. 2001; 107:43–49. [PubMed: 11744245] J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 17 NIH-PA Author Manuscript NIH-PA Author Manuscript Van der Weerd L, Lythgoe MF, Badin RA, Valentim LM, Akbar MT, de Belleroche JS, Latchman DS, Gadian DG. Neuroprotective effects of HSP70 overexpression after cerebral ischemia-an MRI study. Exp. Neurol. 2005; 195:257–266. [PubMed: 15936758] Vass K, Welch WJ, Nowak TS Jr. Localization of 70 kDa stress protein induction in gerbil brain after ischemia. Acta Neuropathol. 1988; 77:128–135. [PubMed: 3227811] Volloch V, Gabai VL, Rits S, Force T, Sherman MY. HSP72 can protect cells from heat —induced apoptosis by accelerating the inactivation of stress kinase JNK. Cell Stress Chaperones. 2000; 5:139–147. [PubMed: 11147965] Wagstaff MJ, Collaco-Moraes Y, Aspey BS, Coffin RS, Harrison MJ, Latchman DS, de Belleroche JS. Focal cerebral ischemia increases the levels of several classes of heat shock proteins and their corresponding mRNAs. Brain Res. Mol. Brain Res. 1996; 42:236–244. [PubMed: 9013779] Wang YH, Borkan SC. Prior heat stress enhances survival of renal epithelial cells after ATP depletion. Am. J. Physiol. 1996; 270:F1057–1065. [PubMed: 8764325] Wang Y, Knowlton AA, Christensen TG, Shih T, Borkan SC. Prior heat stress inhibits apoptosis in adenosine triphosphate-depleted renal tubular cells. Kidney Int. 1999; 55:2224–2235. [PubMed: 10354271] Wang X, Yang C, Chai J, Shi Y, Xue D. Mechanisms of AIF-mediated apoptotic DNA degradation in Caenorhabditis elegans. Science. 2002; 298:1587–1592. [PubMed: 12446902] Wilmore WG, Storey KB. Antioxidant systems and anoxia tolerance in a freshwater turtle Trachemys scripta elegans. Mol. Cell Biochem. 1997; 170:177–185. [PubMed: 9144333] Wolf HK, Buslei R, Schmidt-Kastner R, Schmidt-Kastner PK, Pietsch T, Wiestler OD, Blumcke I. NeuN: a useful neuronal marker for diagnostic histopathology. J. Histochem. Cytochem. 1995; 44:1167–1171. [PubMed: 8813082] Zhang Z, Sobel RA, Cheng D, Steinberg GK, Yenari MA. Mild hypothermia increases Bcl-2 protein expression following global cerebral ischemia. Brain Res. Mol. Brain Res. 2001; 95:75–85. [PubMed: 11687278] Zhu C, Wang X, Huang Z, Oiu L, Xu F, Vahsen N, Nilsson M, Eriksson PS, Culmsee C, Plesnila N, Kroemer G, Blomgren K. Apoptosis-inducing factor is a major contributor to neuronal loss induced by neonatal cerebral hypoxia-ischemia. Cell Death Differ. 2007; 14:775–784. [PubMed: 17039248] NIH-PA Author Manuscript J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 18 NIH-PA Author Manuscript NIH-PA Author Manuscript Fig 1. Analysis of the cortex of the turtle for neuronal damage. a— c) Cresyl violet staining; d — f) Immunolabeling for the neuronal marker NeuN; and g — i) Immunolabeling for the glial marker GFAP. First column (a, d, g) = control animals; second column (b, e, h) = anoxia; third column (c, f, i) = anoxia followed by three days of survival. Note preservation of neuronal cell band at all time points in cresyl violet staining and NeuN-immunolabeling. GFAP signals are not increased at three days. Magnification bar shown in i) is equivalent to 200 μm for all. NIH-PA Author Manuscript J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 19 NIH-PA Author Manuscript NIH-PA Author Manuscript Fig 2. Multiple stress proteins are upregulated by short- and long-term anoxia in the turtle brain. Representative western blots and densitometric analyses (expressed as % of control) of turtle brain whole cell lysates are shown. For each blot, the order of loading from left is as follows: normoxic control, 1 h anoxia, 4 h anoxia and 24 h anoxia. (a) Hsp72 (b) Hsp60, (c) HO-1 (d) Hsp27 (e) Grp78 (f) Grp94. Data are expressed as mean ± SE. * shows significant difference from normoxic control (p<0.05) . NIH-PA Author Manuscript J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 20 NIH-PA Author Manuscript NIH-PA Author Manuscript Fig 3. Bcl2 family proteins are differentially regulated by short- and long-term anoxia in the turtle brain. Representative western blots and densitometric analyses (expressed as % of control) of turtle brain whole cell lysates are shown. Relative changes in expression of (a) Bcl2 (b) Bax, (c) Ratio of the expression of Bcl2 to Bax obtained from the mean densities NIH-PA Author Manuscript J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 21 NIH-PA Author Manuscript Fig 4. Proteins involved in the regulation of Capase independent and Caspase dependent pathways are differentially expressed by short- and long-term anoxia in the turtle brain Representative western blots of turtle whole brain cell lysates are shown . (a) AIF and ( b) Procaspase 3 and Caspase 3. * shows significant difference from normoxic control (p<0.05) “#” = significant difference from 1 h anoxia. NIH-PA Author Manuscript NIH-PA Author Manuscript J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 22 NIH-PA Author Manuscript Fig 5. Densitometric analysis of protein expression of all the (a) stress proteins and the (b) apoptotic regulators in response to reoxygenation after anoxia. The data are expressed as percent of the 4 h anoxia to analyze the relative change of protein levels in reoxygenation after 4 h anoxia. The data are analyzed by two tailed t-test. *shows significance (p<0.05). “#” = significant difference from 1 h anoxia, $ = significant difference. from 24 h anoxia. NIH-PA Author Manuscript NIH-PA Author Manuscript J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 23 NIH-PA Author Manuscript NIH-PA Author Manuscript Fig 6. Subcellular localization of Bax and AIF. Representative western blots and densitometric analyses. For each blot, the order of loading from left is as follows: normoxic control, 1 h anoxia, 4 h anoxia and 4 h anoxia/ 4h reoxygenation. (a) Relative expression levels of Bax in cytosol and mitochondria expressed as percent of cytosolic normoxic control. * represents significant difference in the mitochondria fraction. (b) Differential levels of AIF in mitochondrial, cytosol and nuclear fractions expressed as percentage of mitochondrial normoxic control. * represents significant difference in cytosolic fraction, # represents significant difference in nuclear fraction. NIH-PA Author Manuscript J Neurochem. Author manuscript; available in PMC 2009 September 1. Kesaraju et al. Page 24 Table I NIH-PA Author Manuscript Overview of the expression of stress proteins and apoptotic regulators in three different phases of anoxic survival in an anoxia tolerant freshwater turtle (After Lutz and Milton, 2004) Downregulation of metabolism (Hypometabolic phase) occurs in the initial transition to full anoxia (1-2 h), followed by the maintenance phase of long term anoxia (> 2h to days); the recovery phase represents a return to basal metabolism during reoxygenation and the potential for oxidative stress. Throughout both stages of hypometabolism , there is an increase in the expression of stress proteins and decrease in the expression of proteins involved in apoptosis represents that may help maintain the structural integrity of the brain and prevent significant damage. expression at the basal level, represents upregulation, represents continued upregulation , represents decreased expression. * represents high constitutive expression of Hsp’s in the turtle brain whereas the protein is undetectable in the mammalian brain under baseline conditions. Hypometabolism Protein Transition (1-2 h anoxia) Maintenanace (h — d anoxia) Recovery (h — d Reox) Hsp72* Grp78 Hsp60 HO-1 NIH-PA Author Manuscript Hsp27* Grp94 Bcl2 Bax Procaspase 3 Caspase 3 NIH-PA Author Manuscript J Neurochem. Author manuscript; available in PMC 2009 September 1.