By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Address correspondence and reprint requests to Guillermina Almazan, Department of Pharmacology and Therapeutics, McGill University, 3655 Promenade Sir William Osler, Room 1321, Montreal, Quebec, H3G 1Y6, Canada. E-mail: firstname.lastname@example.org
Oligodendrocyte progenitors are highly susceptible to various insults. Their limited antioxidant defenses and high levels of apoptotic factors, such as Bax and pro-caspase-3 contribute to their sensitivity. We previously showed that dopamine (DA) is toxic to oligodendrocyte progenitors by inducing superoxide generation, lowering glutathione levels and promoting apoptosis through caspase-3 activation. In contrast, factors that contribute to cell survival and defense against dopamine (DA) toxicity are less studied. Here, we explored the role of two molecules which play important roles in cell survival, namely the heat shock protein 90 (HSP-90) and the protein kinase Akt, using the selective inhibitors, 17-AAG and Akt inhibitor III, respectively. The HSP-90 inhibitor caused a decrease in P-Akt level, induced caspase-3 activation, increased nuclear condensation and caused a loss in cell viability. Furthermore, 17-AAG potentiated DA-induced apoptosis by enhancing caspase-3 activation. In addition, the Akt inhibitor alone exacerbated DA toxicity and in combination with 17-AAG caused synergistic potentiation of DA toxicity by enhancing caspase-3 activation. Together, these results indicate that HSP-90 is essential for oligodendrocyte progenitor survival. Both HSP-90 and Akt play important roles in concert in the defense against DA-induced apoptosis.
In the neonatal white matter, oligodendrocyte progenitors are highly sensitive to signals inducing cell death, both during development and in neurodegenerative diseases (Jelinski et al. 1999; Casaccia-Bonnefil 2000). The hypomyelinating disorder, periventricular leucomalacia (PVL) results from perinatal hypoxic/ischemic insults that produce oxidative stress leading to oligodendrocyte progenitor cell loss (Volpe 2001; Back et al. 2002; Rezaie and Dean 2002; Dewar et al. 2003). Regulating the survival of oligodendrocyte progenitors is critical for their proper development and differentiation into myelin producing cells. Various apoptotic and survival factors may contribute to a delicate balance regulating oligodendrocyte survival and defense against toxic insults. For example, caspase-3 is important for apoptosis of oligodendrocyte progenitors during development, however, contributes to aberrant cell loss in neurodegenerative diseases (De Louw et al. 2002; Knoblach et al. 2005). Furthermore, high levels of iron, a catalyst of oxidative reactions, and inadequate defenses, such as low levels of the antioxidant glutathione are among the factors contributing to the sensitivity of oligodendrocyte progenitors to oxidative stress (Thorburne and Juurlink 1996; Back et al. 1998).
In our previous studies, we showed that DA is toxic to oligodendrocyte progenitors. DA causes superoxide generation and reduces glutathione levels while its toxicity is enhanced in the presence of iron (Khorchid et al. 2002; Hemdan and Almazan 2006, 2007). Furthermore, DA induces an apoptotic mechanism of cell death by activation of caspases 9 and 3, and DNA fragmentation (Khorchid et al. 2002). In contrast, factors that contribute to defense of oligodendrocyte progenitors against oxidative stress have only been minimally studied. We recently demonstrated a role for intracellular glutathione in the cellular defense against DA toxicity, while deficient scavenging of peroxides by glutathione peroxidase is associated with the cells susceptibility to DA toxicity (Hemdan and Almazan 2007). Additionally, DA potently induces expression of heme-oxygenase-1 (HSP-32) (Khorchid et al. 2002), a member of a family of molecular chaperones known as heat shock proteins (HSPs), which are involved in cell survival, proliferation and differentiation (Helmbrecht et al. 2000; Richter-Landsberg and Goldbaum 2003; Beere 2005).
Heat shock proteins contribute to cell survival and protection by regulating the folding and stability of various cellular proteins. The family consists of several members including HSP-32, HSP-72, αB-crystallin, HSP-25 and HSP-90 (Sun and MacRae 2005), some of which are up-regulated in the brain following hypoxic/ischemic injury (Kato et al. 1994; Mariucci et al. 2007). Furthermore, heat shock preconditioning protected mature oligodendocytes from subsequent lethal heat shock treatment (Goldbaum and Richter-Landsberg 2001). However, the roles of HSPs in immature oligodendrocytes have not been addressed.
HSP-90 is a ubiquitous and highly abundant molecule that plays key roles in the regulation of the cell cycle and cell survival in biological systems by interacting with a multitude of client proteins including survival as well as apoptotic factors (Terasawa et al. 2005). HSP-90 can bind to the pro-apoptotic protein Apaf-1, thus preventing apoptosome formation and apoptosis (Pandey et al. 2000). HSP-90 can also inhibit apoptosis by binding and stabilizing the survival factor, Akt (Basso et al. 2002; Georgakis et al. 2006). Akt is a serine-threonine kinase activated by IGF-1 receptor signaling through phosphoinositol-3-kinase (PI3-k) (Datta et al. 1999). IGF-1 receptor signaling has been shown to promote normal oligodendrocyte development in the CNS (Zeger et al. 2007), and Akt protects oligodendrocyte progenitors from growth factor deprivation (Cui et al. 2005), glutamate (Ness et al. 2004) and tumor necrosis factor-α damage (Ye et al. 2007). However, a role in protection against oxidative stress has not been shown.
In this study, we assessed the role of HSP-90 and Akt in oligodendrocyte progenitor survival and defense against DA toxicity using the specific pharmacological inhibitors, 17-AAG (Schulte and Neckers 1998) and AI(III), respectively. Our results indicate that HSP-90 is vital for the survival of oligodendrocyte progenitors, and plays an important role in the defense against DA-mediated apoptosis, by interfering with caspase-3 activation. Furthermore, Akt is involved in the defense of oligodendrocyte progenitors against DA, and the cooperative actions of HSP-90 and Akt account for some of their cytoprotective effects.
Materials and methods
Dulbecco’s modified Eagle medium (DMEM), Ham’s F12 medium, phosphate buffered saline (PBS), Hank’s balanced salt solution, 7.5% bovine serum albumin fraction V, fetal calf serum (FCS), penicillin and streptomycin and 4,6-diamidino-2-phenylindole dihydrochloride (DAPI) mounting medium were from Invitrogen (Burlington, ON, Canada). Nitrocellulose membranes were from Mandel (Guelph, ON, Canada). ECL Western Blotting Detection Kit was from NEB (Oakville, ON, Canada). Platelet-derived growth factor-AA (PDGF-AA) and basic fibroblast growth factor (bFGF) were from PeproTech Inc (Rocky Hill, NJ, USA). Protein assay kit was from BIO-RAD (Mississauga, ON, Canada). 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT), poly-d-lysine, poly-l-ornithine, human transferrin, insulin, HEPES, Triton X-100 and dopamine.HCl (DA) were from Sigma–Aldrich (Oakville, ON, Canada). 17-Allylaminogeldanamycin (17-AAG) and Akt Inhibitor III (AI(III) were from EMD Chemicals (San Diego, CA, USA). Lactate dehydrogenase cytotoxicity detection kit was from Roche Molecular Biomedicals (Laval, QC, Canada). Primary antibodies were obtained from the following suppliers: rabbit polyclonal HSP-25, HSP-90, mouse monoclonal HSP-32, HSP-72, αB-crystallin and α-spectrin antibodies were from Stressgen Biotechnologies (Victoria, BC, Canada); rabbit polyclonal P-Akt (P-Ser), total Akt (T-Akt) and cleaved caspase-3 (17 KDa) antibodies were from Cell Signaling Technology (Beverly, MA, USA), mouse monoclonal myelin basic protein (MBP) antibody was from Cedarlane (Berkeley, CA, USA), and goat monoclonal anti-β-actin antibody from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Secondary horseradish peroxidase-conjugated antibodies were from Southern Biotechnology (Birmingham, AL, USA), Jackson Immunoresearch Laboratories (Bio-Can Scientific, Mississauga, ON, Canada) or BIO-RAD. All other reagents were obtained from VWR (Mont-Royal, QC, Canada) or Fisher (Ottawa, ON, Canada).
Oligodendrocyte Primary Cultures
Primary cultures of oligodendrocytes were prepared from the brains of newborn Sprague–Dawley rats as described before (McCarthy and de Vellis 1980; Almazan et al. 1993). All experiments were approved by the McGill Faculty of Medicine Animal Care Committee in accordance with Canadian Council on Animal Care guidelines. The meninges and blood vessels were removed from the cerebral hemispheres in Ham’s F12 medium. The tissue suspension was passed through a 230 μm nylon mesh and collected by filtration through a 150 μm nylon mesh. The resulting suspension was centrifuged for 10 min at 800 g and then resuspended in DMEM supplemented with 12.5% heat-inactivated fetal calf serum (complete medium). Cells were plated on poly-l-ornithine-pre-coated 80 cm2 flasks and incubated at 37°C with 5% CO2 atmosphere. Culture medium was changed after 3 days and every 2 days thereafter. The initial mixed glial cultures, grown for 9–11 days, were placed on a rotary shaker at 225 rpm at 37°C for 3 h to remove loosely attached macrophages. Oligodendrocyte progentors were detached by shaking for 18 h at 260 rpm. The cells were filtered through a 30 μm nylon mesh and plated on bacterial grade Petri dishes for 3 h. Under these conditions, astrocytes and microglia attached to the plastic surface and oligodendrocyte progenitors remained in suspension. The final cell suspension was plated on multi-well dishes pre-coated with poly-d-lysine at an approximate density of 1.5 × 104/cm2. Oligodendrocyte progenitors were maintained in serum-free medium (SFM) consisting of a 1 : 1 DMEM-F12 mixture, 10 mM HEPES, 0.1% bovine serum albumin, 25 μg/mL human transferrin, 30 nM triiodothyronine, 20 nM hydrocortisone, 20 nM progesterone, 10 nM biotin, 5 μg/mL insulin, 16 μg/mL putrescine and 30 nM selenium. Platelet-derived growth factor-AA (PDGF-AA) and basic fibroblast growth factor (b-FGF) were added at a concentration of 2.5 ng/mL to stimulate proliferation. 95% of the isolated cells were positive for the gangliosides recognized by the monoclonal antibody A2B5, a marker for oligodendrocyte progenitors and < 5% were GalC positive oligodendrocytes, glial fibrillary acidic protein-positive astrocytes or complement type-3-positive microglia (Cohen and Almazan 1994). To obtain differentiated oligodendrocytes, progenitors were maintained in SFM lacking PDGF and b-FGF, but supplemented with 3% calf serum. Mature oligodendrocyte cultures contained more than 90% GalC and MBP-positive cells, while a few progenitors kept dividing and were A2B5+. Treatments were performed in SFM lacking PDGF-AA, bFGF or calf serum. In the case of progenitors, this allows the A2B5+ cells to progress to the O4+ stage.
MTT survival assay
Cell viability was assessed using the MTT assay which measures mitochondrial dehydrogenase activity as described (Hemdan and Almazan 2006). The assay detects cleavage of MTT by active mitochondria in viable cells to an insoluble formazan product. Cultures were incubated with 0.5 mg/mL MTT at 37°C for 3 h after which the formazan crystals were solubilized in an acid-isopropanol mixture. Absorbance was measured at 595 nm with a micro-ELISA spectrophotometer.
Nuclear chromatin condensation was assessed to determine apoptotic cells. Cells grown on glass coverslips were fixed with 4% paraformaldehyde in PBS, then mounted on microscope slides with 4,6-diamidino-2-phenylindole dihydrochloride containing mounting medium. Nuclei were visualized, and photographed by ultraviolet illumination using a fluorescence microscope equipped with an automatic camera. Condensed or fragmented nuclei were counted and expressed as a % of total numbers.
Western blot analysis
Cells were harvested in ice-cold lysis buffer (1% Nonidet P-40, 10% glycerol, 137 mM NaCl, 1 mM phenyl-methyl sulfonyl fluoride, 1 mM aprotinin, 0.1 mM sodium vanadate and 20 mM NaF). The protein concentration in cell lysates was determined with the BIO-RAD Protein Assay Kit. Following addition of 5 × loading buffer (final concentration 2% SDS, 5% glycerol, 5%β-mercaptoethanol, 0.01% bromophenol blue) and boiling for 5 min, aliquots containing 20 μg of protein were resolved by SDS polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. Blots were blocked for 1 h with 5% dry milk in Tris-buffered saline containing 0.1% Tween 20 and then incubated with the following primary antibodies: HSP-90, HSP-25, HSP-32, αB-crystallin, P-Akt, T-Akt, cleaved-caspase-3 (17 KDa) and α-spectrin. The membranes were incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies and immunoreactive bands were visualized by enhanced chemiluminescence (ECL Western Blotting Detection Kit) and quantified by densitometry. Membranes were probed for β-actin to normalize for equal protein loading and transfer.
Data are represented as mean ± SEM of a representative experiment (performed in triplicate), of three that were performed. Statistical significance was determined using one-way anova followed by the Bonferroni test, or by student’s t-test; p-values < 0.05 were considered significant. Significance is indicated in figures by asterisks as follows (*p <0.05, **p <0.01 and ***p <0.001).
HSP-90 is expressed in oligodendrocyte progenitors, while HSP-25, HSP-32 and αB-crystallin appear in differentiating cells
Various heat shock proteins (HSPs) contribute to cell survival by folding and stabilizing signaling proteins. To assess the roles of HSPs in oligodendrocyte progenitor survival and defense against DA toxicity, we determined which HSPs are expressed in oligodendrocyte progenitors and the levels during oligodendrocyte development. We found that HSP-90 is constitutively expressed in oligodendrocyte progenitors and its levels were maintained in differentiating cells (Fig. 1.). In contrast, HSP-25, αB-crystallin and HSP-32 were not detected in oligodendrocyte progenitors, but were expressed at later stages of development. HSP-25 and HSP-32 appeared at 3 days of differentiation (d3) while αB-crystallin appeared at (d6). Both HSP-25 and αB-crystallin reached maximal levels in mature (d12) oligodendrocytes, while HSP-32 was maximal at (d6) but then decreased. As the cultures differentiated, expression of MBP, an abundant structural protein in mature oligodendrocytes and myelin increased in level.
HSP-90 inhibitor (17-AAG) up-regulates HSP-72 in progenitors and mature oligodendrocytes
Since HSP-90 is the only HSP expressed in oligodendrocyte progenitors under basal conditions, we speculated that it may have a fundamental role in their survival and defense against DA toxicity. In cells, HSP-90 is present in a complex with the transcription factor, heat shock factor-1 (HSF-1), which retains HSF-1 in the cytosol. Therefore, inhibiting HSP-90 activity should release HSF-1 allowing it to translocate to the nucleus and transcribe target genes including HSP-72. Thus, we first examined the ability of the HSP-90 inhibitor 17-AAG to inhibit HSP-90 activity in our cultures by assessing its effect on HSP-72 expression. Treatment of oligodendrocyte progenitors with 0.05–1 μM 17-AAG for 25 h induced HSP-72 in a concentration-dependent manner (Fig. 2.), indicating that 17-AAG is pharmacologically active. Furthermore, pre-treatment of cells with 0.05 μM 17-AAG for 1 h followed by 75 μM DA caused a synergistic induction of HSP-72, suggesting an enhancement of DA signaling. To assess whether 17-AAG was effective throughout oligodendrocyte development, it was tested in oligodendrocytes differentiated for 12 days. HSP-72 was similarly induced by identical concentrations as used in progenitors. Increased concentrations, up to 10 μM showed enhanced induction.
17-AAG induces concentration-dependent toxicity in progenitors but not mature oligodendrocytes
To determine whether HSP-90 is involved in cell survival, oligodendrocyte progenitors were treated for 25 h with various concentrations of 17-AAG. Cell survival was assessed by the MTT assay. 17-AAG induced a large concentration-dependent decline in oligodendrocyte progenitor viability (Fig. 3a) indicating that HSP-90 is vital for their survival. Toxicity was observed at concentrations ≥ 0.1 μM. At the lower concentration (0.05 μM), 17-AAG alone did not affect cell survival. These results were compared to mature oligodendrocytes. Surprisingly, concentrations that up-regulated HSP-72 had no effect on cell viability (Fig. 3b), suggesting HSP-90 is not critical for the survival of differentiated oligodendrocytes.
17-AAG exacerbates DA toxicity in oligodendrocyte progenitors
We next assessed whether HSP-90 is also involved in cellular defense against DA toxicity. Cells were pre-treated with 17-AAG for 1 h using 0.05 μM (a concentration shown to have no effect on unstressed cells). This was followed by DA (50, 75 and 150 μM) treatment for 24 h. 17-AAG exacerbated toxicity induced by 75 μM DA, reducing MTT values to a level similar to the 150 μM DA concentration (Fig. 4a). Thus, HSP-90 is involved in the defense of oligodendrocyte progenitors against DA toxicity.
17-AAG alone induces nuclear condensation and potentiates the effect of DA
HSP-90 has been shown to block apoptotic signaling by interacting with various pro-apoptotic (ex. Apaf-1) and anti-apoptotic (ex. Akt) (Pandey et al. 2000; Basso et al. 2002) molecules. Thus, we assessed the effect of 17-AAG on nuclear chromatin condensation/fragmentation, a hallmark of apoptosis. Fig. 4b-(i) is a phase contrast image showing untreated oligodendrocyte progenitors in culture which displayed typical morphology with few processes. Treatment of cells with 0.5 μM 17-AAG for 25 h resulted in the appearance of (11%) condensed nuclei (Fig. 4b-(iv) and Table. 1) compared to untreated cells where most nuclei were intact (Fig. 4b-(ii) and Table. 1), indicating that HSP-90 is required to prevent cells from undergoing apoptosis. Furthermore, pre-treatment of cultures for 1 h with 0.05 μM 17-AAG, a concentration with no effect on its own, followed by 75 μM DA for 24 h increased nuclear condensation to 37.1% compared to 20.9% for DA alone (Fig. 4b-(v) and (vi) and Table. 1).
Table 1. 17-AAG alone induces nuclear condensation and potentiates the effect of dopamine
Condensed nuclei (% of total)
Quantification of DAPI staining is shown. Condensed nuclei are expressed as percentage of total nuclei counted. Data are the mean ± SEM. Statistical differences are Control versus 17-AAG 0.5 μM (p <0.001), Control versus DA 75 μM (p <0.001), DA 75 μM versus DA 75 μM + 17-AAG (p <0.001).
3 ± 0.9
AAG (0.05 μM)
3 ± 1.5
AAG (0.2 μM)
5.4 ± 0.7
AAG (0.5 μM)
11 ± 1.4 ***
DA (75 μM)
20.9 ± 3.4
DA (75 μM) + AAG (0.05 μM)
37.1 ± 5.5***
17-AAG alone causes concentration dependent caspase-3 activation and enhances the effect of DA
HSP-90 can promote cell survival by interacting with various pro-apoptotic molecules preventing their ability to activate caspase-3. Alternatively, HSP-90 may interact with and stabilize Akt, a protein kinase that plays an important role in survival. Akt can inhibit pro-apoptotic molecules such as Bad and caspase-9, and can bind cytochrome c, a mitochondrial protein which gets released to the cytosol during apoptosis. To further examine the protective functions of HSP-90 in our cultures, we assessed the effect of HSP-90 inhibition on caspase-3 activation. Oligodendrocyte progenitors were treated with various concentrations of 17-AAG for 25 h, and Western blotting was performed. 17-AAG (0.2–1 μM) caused a concentration-dependent activation of caspase-3 by cleavage to the active 17 KDa fragment, and induced cleavage of α-spectrin, a downstream target of caspase-3 (Fig. 5a). In addition, as caspase-3 is an important mediator of DA toxicity in progenitors, we speculated that the anti-apoptotic effect of HSP-90 on DA toxicity may involve suppression of caspase-3 activation. Pre-treatment of cells for 1 h with 0.05 μM 17-AAG followed by 75 μM DA for 8 h enhanced DA-induced caspase-3 activation (Fig. 5b). Alone, 0.05 μM 17-AAG did not affect caspase-3 activation nor α-spectrin cleavage. These results suggest that HSP-90 promotes oligodendrocyte progenitor survival and defense against DA toxicity by inhibiting caspase-3 activation.
17-AAG and DA reduce P-Akt levels
Whether HSP-90 inhibition induces caspase-3 activation by inhibiting HSP-90 effects on Akt or other proteins was tested. In various cell lines that over-express HSP-90, its inhibition by 17-AAG has been associated with a decline in the level of P-Akt. In addition, Akt has been previously shown by us and other groups to mediate survival of oligodendrocyte progenitors (Ness et al. 2004; Cui et al. 2005). In contrast, Akt inactivation has been associated with cell death in some systems (Luo et al. 2003). In this study, we assessed whether HSP-90 inhibition results in decreased Akt levels in oligodendrocyte progenitors. Cultures were treated with various concentrations of 17-AAG (0.05, 0.2, 0.5 and 1 μM) for 25 h followed by western blotting. 17-AAG caused a concentration-dependent decrease in the level of P-Akt which was observable with concentrations ≥ 0.2 μM, a highly toxic concentration. T-Akt also decreased but required a higher concentration (1 μM) (Fig. 6a).
An arising question was whether 17-AAG exacerbates DA toxicity by its effect on Akt. To examine this possibility, we first determined whether Akt is inactivated following DA treatment. Oligodendrocyte progenitors were treated with various concentrations of DA (75, 150, 300 and 500 μM) for 16 h. The levels of P-Akt and T-Akt were determined by western blotting. In untreated cells, both P-Akt and T-Akt levels were high, which was expected since the media contains 5 μg/mL insulin which activates Akt through the IGF-1 receptor and its downstream effector PI3-k. DA induced a decrease in the level of P-Akt only with the highly toxic 300 μM concentration (Fig. 6b). The decrease in P-Akt was not accompanied by a decrease in T-Akt. At 500 μM DA, P-Akt, T-Akt and β-actin were all greatly diminished, which may be attributed to the > 80% cell death occurring at this concentration, as shown in our previous report (Hemdan and Almazan 2007).
Treatment of cultures with 17-AAG (1 h, 0.05 μM) prior to 75 μM DA for 8 h (Fig. 6a) or 24 h (data not shown) did not cause a detectable reduction in P-Akt levels. These results suggest that the synergistic action of 17-AAG and DA on OLP toxicity is not directly mediated by Akt inactivation.
17-AAG and DA cause concurrent time-dependent changes in P-Akt and caspase-3 (17 KDa) levels
Time-course analysis was performed to further assess whether the decrease in P-Akt was an early event that could be implicated in caspase-3 activation, or occurs independently of the latter. Cells were treated for different time periods with 17-AAG, DA or their combination. P-Akt and cleaved caspase-3 levels were then determined by western blotting. Treatments with 0.5 μM 17-AAG alone, 300 μM DA and DA (75 μM) + 17-AAG (0.2 μM) all caused an initial increase in P-Akt at 3 h, followed by a decrease at 8, 12, and 24 h. On the other hand, cleaved caspase-3 level was similar to control levels at 3 h, and elevated at 8, 12, and 24 h. (Fig. 7). Thus, under these experimental conditions, inactivation of Akt occurred concurrently with activation of caspase-3.
Akt inhibitor III (AI(III)) exacerbates DA cytotoxicity
Akt may play a protective role by a number of anti-apoptotic mechanisms. To test this, we utilized Akt Inhibitor III (AI(III)), which is a phosphoinositol analog that competes with PIP3 for binding to Akt (Kau et al. 2003). Cells were pre-treated for 1 h with AI(III) (5 μM) prior to exposure to 75 μM DA for 24 h, and cell viability was determined by the MTT assay. AI(III) increased DA toxicity while it had no effect alone (Fig. 8a). These results indicate that Akt is involved in the defense against DA toxicity.
Combination of 17-AAG and AI(III) synergistically potentiates DA cytotoxicity and caspase-3 activation
Since DA toxicity is enhanced by inhibiting either HSP-90 or Akt, we questioned whether there is an interaction between the PI3k/Akt survival pathway and HSP-90 pro-survival mechanisms. Thus, using sublethal concentrations, we simultaneously inhibited HSP-90 (0.025 μM 17-AAG) and Akt (3.5 μM AI(III)) and determined the effect on DA (50 μM) toxicity. Cell viability was assessed after 24 h by MTT reduction, and caspase-3 activation was determined after 8 h by western blotting. Interestingly, while AI(III) alone or in combination with 17-AAG did not affect cell survival compared to untreated cells, 17-AAG + AI(III) synergistically potentiated DA toxicity (Fig. 8b). In addition, 17-AAG and AI(III) alone did not activate caspase-3, however, 17-AAG + AI(III) together activated caspase-3, and when combined with DA potentiated DA-induced caspase-3 activation (Fig. 9). These results suggest an interaction between HSP-90 and Akt survival mechanisms, such that their simultaneous inhibition reduces the threshold for DA toxicity.
Mechanisms that counter apoptosis operate in normal nervous system development and in pathologic states. HSPs are a large group of proteins that regulate cell development and survival, through their actions as molecular chaperones, mediating protein folding (Beere 2005). Various HSPs are up-regulated following cerebral ischemia (Kato et al. 1994; Mariucci et al. 2007), in neurodegenerative diseases (Gordon 2003) and following trauma (Hayes et al. 1995). HSP-90, a constitutive member of this group, is a major survival factor that is over-expressed in tumors (Georgakis et al. 2006). Reducing its levels has been consistently associated with reduced tumor growth (Nimmanapalli et al. 2003; Rahmani et al. 2003; Georgakis et al. 2006). However, its role in the survival of normal cells during development, and in neurodegenerative conditions is less understood.
In this study, we observed that various HSPs (HSP-32, HSP-25 and αB-crystallin) were not detected in oligodendrocyte progenitors, but were expressed in differentiating oligodendrocytes. In contrast, HSP-90 was constitutively expressed at similar levels throughout development. This suggested that the high sensitivity of oligodendrocyte progenitors to toxic stimuli may be partly due to their deficient HSPs expression, and suggested HSP-90 may be vital for their survival. Thus, we assessed the role of HSP-90 in oligodendrocyte progenitor survival and protection against DA toxicity, using a specific inhibitor of HSP-90. Similar experiments were carried out with mature oligodendrocytes to determine whether HSP-90 has the same function throughout differentiation. The induction of HSP-72 by 17-AAG, a competitive inhibitor of ATP-induced HSP-90 client protein folding was used as an indicator of 17-AAG pharmacological effectiveness in our system. An induction of HSP-72 was observed following 17-AAG treatment in both progenitors and mature oligodendrocytes. In addition, combination of 17-AAG and DA further induced HSP-72 in oligodendrocyte progenitors. These results indicate that 17-AAG provided effective HSP-90 inhibition. It may be argued that induction of HSP-72 is expected to protect cells as HSP-72 itself is another HSP that plays an important role in survival of many cells (Yenari et al. 2005). However, although the role of HSP-72 in our system is not clarified, the lack of protection suggests that other, survival effects of HSP-90 are being hampered by its inhibition. HSP-90 mediates its protective effects by binding, folding and stabilizing numerous client proteins involved in the cell cycle, cell survival and apoptosis (Terasawa et al. 2005). Indeed, HSP-90 inhibition induced massive oligodendrocyte progenitor death by apoptosis and enhanced DA-induced toxicity as assessed by several criteria, namely mitochondrial activity, nuclear condensation and caspase-3 activation.
Enhancement of DA-induced cell loss and caspase-3 activation by the HSP-90 inhibitor was detected only using the 75 μM DA concentration. At the lower concentration (50 μM), 17-AAG had no effect, suggesting other antioxidant defenses (Hemdan and Almazan 2007) may be operating or that the level of HSP-90 inhibition is not sufficient to potentiate toxicity under mild oxidative stress. Additionally, at a higher concentration of DA (150 μM), which induces a maximal level of activated caspase-3 as well as cell death by necrosis (Hemdan and Almazan 2006, 2007), 17-AAG was not effective. Collectively, these results indicate that HSP-90 is vital for survival, and is involved in the cellular defense against DA-induced apoptosis but not necrosis. Similar to our results, HSP-90 inhibition increased lipopolysaccharide-induced reactive oxygen species generation and caspase-3 activation in macrophages (Hsu et al. 2007), and enhanced hydrogen peroxide-induced cytochrome c release and nuclear condensation in endothelial cells (Zhang et al. 2005). With the exception of a few cell types, in most other systems including cancer and neural cells, HSP-90 inhibition by 17-AAG or compounds of its class required at least (0.5–3 μM) concentrations to produce effects on cell survival. On the other hand, the massive loss of oligodendrocyte progenitors we observed by a concentration of only 0.1 μM (i.e., ∼ 10-fold less than literature values) suggested HSP-90 plays a central role in their survival. In addition, in 12d differentiated oligodendrocytes, concentrations of 17-AAG up to 25 μM did not induce cell death, suggesting that as oligodendrocyte progenitors differentiate, they may acquire other survival factors or mechanisms, such that HSP-90 is no longer indispensable. The greatly increased expression of αB-crystallin and HSP-25 in mature oligodendrocytes are among the possible contributors to their survival in the face of HSP-90 inhibition. HSPs have been shown in various systems to play redundant roles, and thus may compensate for the absence of each other.
Several studies have revealed that HSP-90 may be involved in inhibition of apoptosis by suppressing cytochrome c-mediated Apaf-1 oligomerization (Pandey et al. 2000), or by stabilizing the survival factor, Akt (Basso et al. 2002). Akt promotes its anti-apoptotic effects by a number of mechanisms including phosphorylating the pro-apoptotic proteins, Bad, forkhead transcription factors and caspase-9, and inhibiting cytochrome c release (Datta et al. 1999; Kennedy et al. 1999; Hirai et al. 2004). Conversely, Akt inactivation by its dephosphorylation and/or degradation has been associated with apoptosis in some systems (Luo et al. 2003). We found that while highly toxic concentrations of 17-AAG and DA caused a decrease in P-Akt level, lower toxic concentrations, alone and when combined did not significantly alter P-Akt. Hence, exacerbation of DA toxicity by HSP-90 inhibition may involve a mechanism other than decreased Akt level. Thus, HSP-90 may interfere with DA toxicity by interactions with other proteins upstream of caspase-3, such as Apaf-1 and cytochrome c. Alternatively, small functional changes in Akt activity which are not detected by western blotting may be occurring with the lower drug concentrations.
Surprisingly, time course analysis showed an early rise in P-Akt level (at 3 h) by higher concentrations of 17-AAG (0.5 μM), DA (300 μM) and their combination (0.2 μM 17-AAG ± 300 μMDA), after which it declined. This may be due to several processes as observed in other studies, such as DA action on DA receptors (Brami-Cherrier et al. 2002), oxidant-induced inactivation of serine/threonine phosphatases involved in Akt dephosphorylation, for example PTEN, by hydrogen peroxide (Leslie et al. 2003) or various effects related to HSP-90 inhibition, and the resulting dysregulation of its client proteins, ex. HSP-25. HSP-25 can be up-regulated by 17-AAG, displayed protection against it (McCollum et al. 2006), and can directly phosphorylate Akt. As well, short term (1 h) treatment by Geldanamycin, and another HSP-90 inhibitor disrupted the ability of the phosphatase, PP2A to dephosphorylate Akt, causing increased Akt phosphorylation (Yun and Matts 2005). Finally, by activating another HSP-90 client protein, Src kinase, Geldanamycin caused transient phosphorylation (15 min to 1 h) of Akt which decreased back to control levels by 2 h (Koga et al. 2006). In our time-dependent experiments, decreases in P-Akt and increases in caspase-3 activity occurred concurrently, and not sequentially, at 8, 12, and 24 h following treatment with high concentrations of 17-AAG, DA and their combination, thus further suggesting that these two events may be related.
However, a clear protective role for Akt against DA toxicity was demonstrated by the observations that Akt inhibitor III alone increased DA-induced cell loss, and in combination with the HSP-90 inhibitor prior to DA treatment caused synergistic cell loss and caspase-3 activation. This suggests that HSP-90 and Akt cooperate in protecting cells from DA. Others have reported synergy in cell killing caused by combinations of HSP-90 inhibitors and inhibitors of other HSP-90 client proteins in various systems which was linked with the enhanced loss of client protein activity (George et al. 2004; Barker et al. 2006; Premkumar et al. 2006; Wang et al. 2006). Although we cannot conclude if a direct or indirect interaction occurs between HSP-90 and Akt, their simultaneous inhibition reduces the threshold of DA toxicity. In line with these results, 17-AAG and the PI3-kinase inhibitor, LY294002 synergistically induced cell death by caspase-3 activation in glioma cells. On the contrary, non-neoplastic astrocytes did not exhibit caspase-3 activation by this treatment (Premkumar et al. 2006).
In contrast to the death-promoting effects of HSP-90 inhibition in oligodendrocyte progenitors, in other neural systems, protection occurred. For example, 17-AAG reduced lipopolysaccharide-induced nitrite and IL-1β production in astrocytes and microglia, and reduced pathology in the multiple sclerosis animal model, experimental allergic encephalomyelitis. These results may be explained by HSP-90 regulation of various genes including NFκB. Thus, disruption of HSP-90 binding to the inhibitor IκK causes IκB up-regulation and results in inhibition of NFκB-induced cytokine production (Dello Russo et al. 2006). In addition, geldanamycin protected against neuronal damage induced by ischemia (Ouyang et al. 2005), huntingtin aggregates (Sittler et al. 2001) and 1-methyl-4-phenyl 1,2,3,6-tetrahydropyridine-induced dopaminergic neurotoxicity (Shen et al. 2005), through induction of HSP-72.
In summary, our results indicate that HSP-90 is vital for oligodendrocyte progenitor survival and defense against DA toxicity. HSP-90 mediates its effects by counteracting the apoptotic process, and caspase-3 activation. The exact mechanism linking HSP-90 inhibition to cell death remains to be identified. However, synergistic enhancement of DA-induced cell death mediated by simultaneous inhibition of Akt and HSP-90 points to an interaction between the HSP-90 and Akt survival pathways. This may involve dysregulation of various anti-apoptotic effects common to the two proteins. Our results have important implications for PVL, since HSPs, including HSP-90 are up-regulated in neural cells (Kawagoe et al. 1993; Kato et al. 1994; Mariucci et al. 2007) and in oligodendrocytes (Jelinski et al. 1999) following hypoxia/ischemia. In addition, IGF-1, the growth factor responsible for Akt activation protects oligodendrocyte progenitors and suppresses caspase-3 activation following cerebral ischemia (Cao et al. 2003; Lin et al. 2005).
This work was funded by operating grants from the Multiple Sclerosis Society of Canada (MSSC) and Canadian Institutes of Health Research to Guillermina Almazan. Sandy Hemdan was supported by a studentship from the MSSC.