Heat Shock Protein 27 Delays Ca²⁺-Induced Cell Death in a Caspase-Dependent and -Independent Manner in Rat Retinal Ganglion Cells

Abstract

purpose. Hsp27 is a well-characterized and studied antiapoptotic protein. A recent study reported that Hsp27 is upregulated in the retina after retinal ischemic preconditioning. The timing of this upregulation of Hsp27 correlates with the protective effects of the treatment. It was the goal of the current study to determine what role Hsp27 plays in this protection.

methods. The rat homologue of Hsp27 (rHsp27) was overexpressed in a transformed rat retinal ganglion cell line and subjected to ischemic stress and calcium overload.

results. The overexpression of rHsp27 increased cell survival and inhibited caspase-3 activation. However, the inhibition of caspase-3 alone had no effect on cell survival. Proteomic analysis after Ca²⁺ overload identified four proteins that were repeatedly associated with rHsp27. These proteins include actin, Hsp70, eEF-1α, and SPIN-2. No association with cytochrome c or any caspase enzymes was detected.

conclusions. The results indicate that Hsp27 protects the retinal cells by both caspase-dependent and -independent mechanisms.

Vision loss resulting from retinal ischemia is associated with several systemic and ocular diseases. ¹ In the retina, ischemia is thought to lead to neuronal cell death by initiating a series of events that include oxidative stress, glutamate excitotoxicity, and intracellular Ca²⁺ overload. ^{2 3 4 5} Attempts to prevent ischemic retinal degeneration have included the use of N-methyl-d-aspartate (NMDA) receptor antagonists, Ca²⁺ channel blockers, and antioxidants. These treatments have had limited success, because they provide incomplete protection and carry the risk of adverse side effects. ^{2 6 7 8 9 10} These problems have caused researchers to look toward endogenous protective mechanisms, such as ischemic preconditioning (IPC) to identify new strategies to prevent or ameliorate ischemic retinal damage. 

Ischemic preconditioning is defined as the ability of a nondamaging ischemic event to protect an organ or tissue from a subsequent damaging ischemic insult. This protective strategy has been demonstrated in the heart, ^{11 12} liver, ^{13 14} kidney, ^{15 16} brain, ^{17 18} and retina. ^{19 20} Although the mechanisms underlying retinal IPC are unknown, the protective effects that occur between 24 and 72 hours after IPC are blocked by pretreatment with the protein synthesis inhibitor cycloheximide. ¹⁹ These data support the idea that increased expression of one or more retinal proteins is responsible for the protective effects in the retina after IPC. Earlier work in this laboratory has shown that rat Hsp27 (rHsp27) is upregulated in the rat retina after IPC. This upregulation appears to be localized to the inner retinal layers, and its expression coincides with the neuroprotective window of IPC. ²¹ It is also important to note that rHsp27 was induced independent of other common stress response heat shock proteins such as Hsp70 and -90. ²¹ Hsp27 is a well characterized stress response and cytoprotective protein and has been shown to be involved in the inhibition of several apoptotic pathways in response to a variety of apoptotic stimuli. ^{22 23 24 25 26 27 28 29} It has been demonstrated that Hsp27 can inhibit apoptosis through both interaction with the cytoskeleton and various caspase effector molecules. ^{22 24 25 26 27 28 29 30} In this article, we provide evidence that the neuroprotective action of rHsp27 involves the inhibition of caspase-3 activation. However, inhibition of caspase-3 alone was not sufficient to account for the neuroprotective action exerted by rHsp27, indicating that rHsp27 also acts by blocking other caspase-independent death pathways. 

Materials and Methods

Cell Culture

The transformed rat retinal ganglion cell line RGC-5, ³¹ was used in these experiments. The RGC-5 cells were grown on polypropylene plates and maintained in DMEM (pH 7.4; Sigma-Aldrich, St. Louis, MO) with 1 g/L glucose, 100 U/mL penicillin, 100 μg/mL streptomycin, and 10% heat-inactivated fetal bovine serum (Sigma-Aldrich) without phenol red. The cells were passaged every 2 to 3 days. For all experiments, cells were used at a confluence of 60% to 80%. 

Stable Overexpression rHsp27 in RGC-5 Cells

Rat Hsp27 cDNA was cloned from rat retina and ligated into the pTracer-CMV (Invitrogen, Carlsbad, CA) expression vector. The vector also contained genes for the green fluorescent protein (GFP) and a Zeocin (Invitrogen)-resistant gene that is under the control of an EM-7 promoter. The RGC-5 cells were passaged 24 hours before transfection and plated in normal DMEM. A lipophilic transfection reagent (Lipofectamine 2000; Invitrogen-Gibco, Grand Island, NY) was used to transfect the DNA constructs into the RGC-5 cells. The circular plasmid DNA was diluted to 1 μg/100 μL in serum-free medium (Opti-MEM; Invitrogen-Gibco) and combined in equal parts with the transfection reagent diluted 5 μL per 100 μL. One hundred microliters of the DNA–transfection reagent mixture was added directly to the RGC-5 cells plated in 1 mL of normal DMEM on six-well plates. The cells were then incubated at 37°C with 5% CO₂ for 48 hours. To select rHsp27-expressing cells, cells were treated with 400 μg/mL Zeocin (Invitrogen). The overexpression of rHsp27 was confirmed by Western blot analysis and GFP flow cytometry (Fig. 1) . The selected cells were maintained in a medium containing 400 μg/mL Zeocin. 

Cellular Stress

Two forms of cellular stress were evaluated: serum, oxygen, and glucose deprivation (SOGD), and intracellular Ca²⁺ overload. Cells were plated 24 hours before the start of stress treatment. To achieve SOGD, DMEM (Sigma-Aldrich) containing no serum or glucose was bubbled with 95% N₂ and 5% CO₂ for at least 60 minutes. The cells were washed twice with prewarmed PBS, to remove all glucose and serum, the SOGD medium was added, and the cells were placed in a hypoxia chamber, which was then flushed for at least 30 minutes with 95% N₂ and 5% CO₂. At the end of the SOGD period, the SOGD medium was replaced with warmed normal DMEM containing 10% serum and glucose, and the cells were allowed to recover for 24 hours. The effects of intracellular Ca²⁺ overload were evaluated by treating the RGC-5 cells with the calcium ionophore A23187 (5 μM; Sigma-Aldrich) for 12 to 48 hours in normal DMEM containing 10% serum. 

Flow Cytometry Cell Viability Assay

RGC-5 cells and RGC5/rHsp27–overexpressing cells (also expressing GFP) were plated together at a 50:50 ratio on 100-mm or six-well tissue culture plates. After stress treatment, cells were washed with PBS and removed from the culture plate with trypsin. The cells were then pelleted, resuspended in complete DMEM, and analyzed for GFP expression by flow cytometry at a wavelength of 488 nm (FACSCalibur analytical flow cytometer; BD Biosciences, Mountain View, CA). The viability of each cell line was calculated by comparing the percentage of viable RGC-5 and RGC5/rHsp27 cells between the two cell lines. Under control conditions, no difference between the number of viable cells from each cell line was measured, and each cell line therefore made up 50% of the total population. After SOGD or calcium overload, resistant cells then made up a larger percentage of viable cells, and the degree of cytoprotection was measured by the difference between cell populations. 

MTT Cell Viability Assay

For selected conditions, cell viability was also assessed with a kit (MTT Cell Proliferation Kit I; Sigma-Aldrich). After treatment, cells were incubated for 4 hours at 37°C in the presence of 0.5 mg/mL MTT salt (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide). Cells were then solubilized with 10% SDS in 10 mM HCl overnight. The resultant samples were read on a spectrophotometer (DU 640; Beckman Coulter, Fullerton, CA) at a wavelength of 560 nm. 

Caspase-3 Assay

Caspase-3 activity was assayed (Apo-ONE Homogenous Caspase-3 Assay; Promega, Madison, WI). Cells were plated in constant numbers in black, clear-bottomed, 96-well plates. In select experiments, assays were performed in the presence of specific cell-permeable and irreversible caspase-3 (DEVD-fmk) and caspase-9 (LEHD-fmk) inhibitors (Enzyme Systems Products, Livermore, CA). At 12, 24, or 36 hours after the addition of A23187, cells were lysed in buffer containing fluorometric caspase-3 substrate (Z-DEVD-R110). Caspase-3 activity was determined by the amount of cleaved R110 fluorophore measured with a fluorescence plate reader (FLUOstar; BMG Labtechnologies, Durham, NC). Each experiment was performed in quadruplicate, and results were expressed as the percentage increase of relative fluorescent units (RFUs) from control untreated levels. 

Immunohistochemistry

Both RGC-5 and RGC-5/rHsp27 cells were grown on collagen-coated slides and allowed to grow for 48 hours in DMEM containing 10% serum. Cells were then treated with A23187 (5 μM), incubated for a further 24 hours, washed with cold PBS, and fixed for 20 minutes in 4% paraformaldehyde in 0.1 M PBS at 4°C. Before immunostaining cells were permeabilized by exposure to 0.3% Triton X-100 for 10 minutes. To visualize rHsp27, cells were incubated overnight with the polyclonal antibody SPA-801 (1:200; StressGen, Victoria, British Columbia, Canada), washed, and incubated for 1 hour with anti-rabbit IgG conjugated to FITC. To visualize actin cytoskeleton and nuclei, cells were exposed to Texas red–conjugated phalloidin (Molecular Probes, Eugene OR) and an antifade reagent containing DAPI (4′,6′-diamino-2-phenylindole; SlowFade; Molecular Probes). Sections were viewed and photographed with a fluorescence microscope (AxioPlan; Carl Zeiss Meditec, Thornwood, NY). 

Identification of rHsp27 Protein Complexes

RGC5/rHsp27 cells were washed twice with PBS and lysed in ice-cold immunoprecipitation (IP) buffer (50 mM HEPES [pH 7.6], 150 mM NaCl, 5 mM EDTA, 10 mM NaF, and 0.1% NP-40). Cell lysates were then transferred to microcentrifuge tubes, and the solution was clarified by centrifugation (15 minutes at 20,000g). An aliquot of the supernatant was used to determine protein concentration with a commercial assay (Bio-Rad, Hercules, CA). Immunoprecipitation was then performed on the remaining supernatant (Seize X Protein Immunoprecipitation kit; Pierce Biotechnology, Rockford, IL). Protein A beads included with the kit were incubated with either preimmune rabbit serum or rabbit polyclonal anti-Hsp25 antibody (StressGen) and cross-linked with disuccinimidyl suberate (DSS). Cell lysates containing 100 to 150 μg of soluble protein complexes were precleared by incubating them overnight at 4°C with cross-linked, preimmune rabbit serum Protein A beads. The lysate was collected and incubated overnight at 4°C with Protein A beads cross-linked with a polyclonal anti-Hsp27 (rodent) antibody (anti-Hsp25; StressGen). The IP complexes were then washed seven times, and proteins were eluted by means of a low-pH elution buffer (Pierce Biotechnology). Eluted proteins were immediately neutralized to pH 8.0 with 1 M Tris (pH 9.0), for later trypsin digestion. Eluted fractions containing detectable levels of rHsp27 by Western blot were combined and subjected to tryptic digestion at 37°C for 18 hours. The tryptic peptides were separated on a 1-mm C18 column (LC Packings, Sunnyvale, CA), using a gradient of 0.02% heptafluorobutyric acid (HFBA) in water as buffer A and 0.02% HFBA in 60% acetonitrile as buffer B, and mass analyzed by electron-spray ionization tandem mass spectrometry (MS-MS; LCQ Classic Mass Spectrometer; Finnigan, San Jose, CA). The HPLC gradient was 2% to 60% buffer B over 120 minutes, 60% to 98% buffer B over 30 minutes, and 98% buffer B for 10 minutes. Raw MS-MS data were analyzed with the Sequest algorithm (Finnigan, San Jose, CA) ³² against an automated rat peptide database. The results were further analyzed using the Peptide-Prophet algorithm. ³³ A list of possible rHsp27 binding proteins was generated using the criteria of: Sequest Xcor value >2.0, the Peptide Prophet minimum probability threshold >0.5, or manual interpretation of the MS-MS data. 

Statistical Analysis

Statistical comparisons were made using the Student’s t-test for nonpaired data or ANOVA using the Dunnett posttest (GraphPad Software, Inc., San Diego, CA). P ≤ 0.05 was considered significant. 

Results

To determine whether Hsp27 expression can protect the RGC-5 retinal ganglion cell line from ischemia-related stress (SOGD or Ca²⁺ overload), a rat Hsp27 (rHsp27) expression vector was transfected into the RGC-5 cell line. Flow cytometry analysis demonstrated that >99% of Zeocin-selected transfected cells were GFP positive. Western blot analysis demonstrated that in the transfected cells, rHsp27 was elevated 25-fold above levels expressed in untransfected control cells (Fig. 1) . Transfection of RGG-5 cells produced a small, but not significant, reduction in rHsp27 expression. No significant change in β-actin expression was measured in any group. Overexpression of rHsp27 in this cell line did not alter cell growth rate or viability (data not shown). 

To assess the protective actions of rHsp27, we exposed cocultures of RGC-5 and RGC-5 cells overexpressing rHsp27 (RGC5/rHsp27) to 2 hours of SOGD and allowed a 24-hour recovery. Flow cytometry analysis showed that overexpression of rHsp27 in these cells improved cell survival by 20% (Figs. 2A 2B) . In separate studies, the MTT viability assay confirmed that overexpression of rHsp27 protected RGC-5 cells from SOGD (Fig. 2C) . Transfection of cells with empty vector did not provide any significant protection from SOGD (data not shown). 

To study the effect of Ca²⁺ overload on these cells, cocultures of RGC-5 and RGC5/rHsp27 cells were treated with A23187 for 12, 24, 36, or 48 hours. Flow cytometry analysis demonstrated that overexpression of rHsp27 significantly increases cell survival after 24, 36, and 48 hours of A23187 treatment (Fig. 3) . 

Figure 4shows the effect of adding A23187 (5 μM) on cellular morphology in RGC-5 and RGC-5/Hsp27 cell lines. In normal medium, RGC-5 cells exhibited no immunoreactivity for rHsp27, and phalloidin staining showed normal actin fibers traversing the cytosol. Incubation of RCG-5 cells with A23187 for 24 hours resulted in collapse of the actin cytoskeleton and rounding in 95% to 98% of all cells. In normal medium, the transformed RGC-5/rHsp27 cells exhibited intense rHsp27 immunoreactivity in all viable cells with normal actin fiber organization. Occasional cells not expressing rHsp27 were observed; however, these cells were rounded in appearance with collapsed cytosol due to the action of the selection antibiotic. Incubation of RGC-5/Hsp27 cells with A23187 produced only limited (20%–30%) cell death as determined by the collapse of the actin cytoskeleton. 

To investigate the interaction between rHsp27 and caspase activation, RGC-5 and RGC5/rHsp27 cells were treated with 5 μM A23187 for 12, 24, or 36 hours, and caspase-3 activation was determined. Figure 5shows that addition of A23187 to RGC-5 cells produced a time-dependent increase in caspase-3 activation. No activation of caspase-3 was measurable after 12 hours of treatment; however, 24 hours after treatment, significant caspase-3 activity was measured and levels continued to increase through 36 hours. The overexpression of rHsp27 blocked activation of caspase-3 in RGC-5 cells. Pretreatment of RGC-5 cells with cell-permeable caspase-3 (2 μM DEVD-FMK) or caspase-9 (2 μM LEHD-FMK) inhibitors also blocked caspase-3 activation. 

In the RGC5/rHsp27 cells, no change in caspase-3activity was observed after treatment with A23187. To determine whether the inhibition of caspase-3 activation alone was sufficient to increase RGC-5 cell viability, we pretreated cells with the caspase-3 inhibitor DEVD-FMK (2 μM) before adding A23187. As shown in Figure 6 , the addition of caspase-3 inhibitor had no effect on cell survival. 

To investigate rHsp27 interaction with other possible protein–protein complexes that may influence cell survival, we used coimmunoprecipitation to isolate all rHsp27 binding complexes after 24 hours of A23187 treatment. The immunoprecipitated complexes were trypsin digested and analyzed by ion-trap mass spectroscopy. Results from this proteomic analysis are listed in Table 1 . The most prevalent peptides sequenced by mass spectroscopy (after rHsp27) were from various actin cytoskeletal molecules, translation elongation factor (EF)-1α, Hsp70, and spindling-like protein 2 (SPIN-2). Several other peptide fragments were also sequenced but occurred only in single experiments. MS-MS data were also specifically searched for proteins in apoptotic pathways that Hsp27 had been thought to interact with (e.g., members of the Bcl-2 family, various caspases, and cytochrome c). The IP-MS technique used in this study did not detect any of these apoptotic proteins. 

Discussion

Retinal ischemia plays a central role in several ocular diseases such as glaucoma, diabetic retinopathy, retinopathy of prematurely, central retinal artery occlusion, and anterior ischemic optic neuropathy. An understanding of the cellular events responsible for the retinal degeneration is paramount in devising therapeutic strategies for these diseases. Ischemic preconditioning has been shown to be an effective strategy to harness the endogenous mechanisms within cells to prevent ischemic damage. However, these mechanisms are not well understood. In the retina, initial studies demonstrate that gene regulation and protein synthesis are necessary for retinal ischemic preconditioning. ¹⁹ In previous studies we have shown that the small heat shock protein rHsp27 is upregulated in the rat retina and that this upregulation is associated in time with the protection of the retina from ischemic injury. ²¹  

The present study was designed to determine whether the upregulation of rHsp27 in the preconditioned retina is in part responsible for the protection offered by IPC. The cell line chosen for these studies was a homogenous rat retinal ganglion cell line derived from embryonic rat retinas that express most ganglion cell–specific cell markers. ³¹ In the retina, ischemia leads to excessive glutamate release and eventual excitotoxicity. One of the main components of this excitotoxicity response is the uncontrolled influx of calcium into the postsynaptic neurons. This influx of calcium triggers several signaling cascades and eventual apoptosis. ^{3 4} Results presented in Figures 2 3 and 4demonstrate that rHsp27 can protect RGC-5 cells from stress associated with ischemia and Ca²⁺ overload. These data support the idea that an increase in Hsp27 expression can protect retinal cells from ischemic injury. Previous studies have shown that Hsp27 upregulation can inhibit the activation of caspase-3 by several apoptotic stimuli. ^{22 23 24 25 26 27 28 29} This action of Hsp27 may result from its interaction with events both upstream and downstream of the mitochondria. Our results demonstrate that A23187 administration to RGC-5 cells activates caspase-3 in a caspase-9–dependent fashion. These data indicate that the main organelle targeted by an influx of Ca²⁺ is the mitochondria. The overexpression of rHsp27 completely blocked this caspase-3 activation induced by 24 hours of treatment with A23187; however, inhibiting caspase-3 with a selective caspase-3 inhibitor did not prevent the cytotoxicity induced by A23187. These results support the idea that rHsp27 can modulate caspase-dependent and -independent apoptotic and necrotic events. If this is the case, then rHsp27 can interact with several of these death pathways, a common early event in the delayed cytotoxicity induced by Ca²⁺ overload, or both. 

Previous studies have provided evidence for the antiapoptotic actions of Hsp27 in response to a variety of insults. Studies have shown that increased levels of Hsp27 can protect cells from mitochondria- and cytochrome c-dependent apoptosis. This was first explained by the ability of Hsp27 to bind to and sequester released cytoplasmic cytochrome c before it combines with Apaf-1 and activates procaspase-9. ^{25 27} In subsequent studies Paul et al. ²⁸ suggest that Hsp27 acts upstream of the mitochondria by possibly altering Bid intracellular relocalization to the mitochondria. This alteration in Bid relocalization is thought to be related to the F-actin stabilizing action of Hsp27. Stabilizing the cytoskeleton with phalloidin or overexpressing Hsp27 has been shown to prevent cytochalasin D–induced F-actin depolymerization and subsequent release of cytochrome c in several cell types. ²⁸ Hsp27 may also interfere with caspase-mediated apoptosis through direct association with activated caspase-3. Studies have reported that small heat shock proteins such as αβ-crystallin and Hsp27 can inhibit both mitochondria- and death receptor–mediated caspase activation. ³⁴ They report that these small Hsp accomplish this by binding to the p24 caspase-3 processing intermediate protein and thereby inhibiting its autocatalytic maturation of active caspase-3, ³⁴ or by the direct association with caspase-3. ³⁵  

The exact role rHsp27 plays in preventing caspase-dependent apoptosis in these RGC-5 cells is still unknown. Given that ischemic stress to the retina causes membrane depolarization, oxidative stress, calcium influx, and metabotropic and ionotropic receptor activation, it is likely that multiple pathways contribute to the eventual apoptotic and necrotic degeneration. If this is the case, there are several places in the apoptosis signaling pathway where Hsp27 could interfere if upregulated. Mitochondria-associated apoptosis can occur through caspase-dependent and -independent mechanisms. We have clearly shown that rHsp27 can block the caspase-dependent pathway. The caspase-independent pathways may involve the release of mitochondrial proteins such as apoptosis-inducing factor (AIF) and EndoG. One study has shown that overexpression of Hsp27 can lead to a decreased release of AIF. ³⁶ Whether rHsp27 directly interacts with these apoptotic proteins in our model is unknown. It is possible that Hsp27 may interact with the release of these proteins from the mitochondria, much as it interferes with the release of cytochrome c. This would explain why overexpression of Hsp27 protects retinal ganglion cells from ischemic death, whereas simply inhibiting caspases has no effect on cell viability. As mentioned earlier, the results of our IP-MS studies did not show any direct interaction with these apoptotic proteins or other proteins in apoptotic pathways that Hsp27 had been thought to interact with (e.g., members of the Bcl-2 family, various caspases, and cytochrome c). Although MS analysis is a very sensitive technique, the failure to detect these proteins may be due to the low amount of total protein used for the immunoprecipitation or to the abundance of these particular proteins being below the detection limits for analysis. However, the strong association with actin implies that one of the major roles of Hsp27 is actin-cytoskeleton stabilization. The association of Hsp27 and actin has been well studied and documented. ^{28 37 38} This interaction may have downstream effects on both caspase-dependent and -independent cell death pathways. Paul et al. ²⁸ demonstrated that overexpression of Hsp27 stabilized the cytoskeleton against cytochalasin D treatment and prevented the release of cytochrome c from the mitochondria. It has been thought that Hsp27 accomplishes this by preventing the translocation of Bid from the cytosol to the mitochondria. It is possible that caspase-independent factors are regulated in a similar manner. 

Among the other proteins revealed by our proteomic analysis was Hsp70. Hsp27 and -70 are generally thought to act independently, but at least one other study has shown an interaction between these proteins. ³⁹ Further investigations are necessary to deterine the combined antiapoptotic effect of Hsp27 and -70. The two additional proteins identified by IP-MS were EF-1α and SPIN-2. The role of EF-1α in apoptosis is somewhat unclear. Studies have shown that it is a proapoptotic factor, ^{40 41} whereas another study reported that it is an antiapoptotic factor. ⁴² SPIN-2 is a newly described nuclear protein that is involved in the regulation of cell-cycle progression and the inhibition of apoptosis. ⁴³ How Hsp27 binds to and interacts with these proteins is still unknown and is an area for future research. 

Further studies are needed to clearly identify the role that Hsp27 plays in preventing ischemia-related cell death. It is apparent, however, that Hsp27 plays a major role in protecting these retinal cells from ischemia-related damage. These findings, when taken together with the fact that Hsp27 is specifically upregulated after retinal IPC and is expressed only during the therapeutic window of IPC, strongly suggests that the upregulation of Hsp27 is a key event during retinal IPC. 

 Supported in part by National Eye Institute Grants EY12231 and EY01560 (J-XM) and EY09741 and EY014793 (CEC); and an unrestricted grant to the Medical University of South Carolina Storm Eye Institute from Research to Prevent Blindness.

 Submitted for publication January 14, 2004; revised October 27, 2004; accepted November 10, 2004.

 Disclosure: N.A. Whitlock, None; K. Lindsey, None; N. Agarwal, None; C.E. Crosson, None; J.-X. Ma, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Stable overexpression of rHsp27 in RGC-5 cells. Transfection efficiency was determined by flow cytometry and GFP expression (A). Overexpression of rHsp27 was confirmed in the RGC-5 cells by Western blot analysis (B). Lane 1: molecular weight marker; lanes 2 to 4: from untransfected cells; lanes 5 to 7: rHsp27 vector-transfected cells; and lanes 8 to 10: RGC-5 cells transfected with empty pTacer vector. Western blot (B) and densitometry analyses (C) show the results of three separate experiments.

Figure 2.

 

Overexpression of rHsp27 in RGC-5 cells increased cell survival after 2 hours of SOGD. RGC-5 and RGC5/rHsp27 cells were cocultured and treated for 2 hours with SOGD insult and allowed to recover for 24 hours before assay. (A) Raw flow cytometry data showing differences in cell survival between RGC-5 and RGC5/rHsp27 cells. The M1 population represents RGC-5 cells, and the M2 population represents RGC5/rHsp27 cells. Flow cytometry data are expressed as the mean ± SE (n = 7) of the percentage of viable cells (B). Flow cytometry results were confirmed by MTT assay analysis on RGC-5 and RGC5/rHsp27 cells grown in separate dishes (C). *Significant difference (P < 0.05) in cell viability between the two cell groups.

Figure 3.

 

Overexpression of rHsp27 in RGC-5 cells increased cell survival after the addition of 5 μM A23187. RGC-5 and RGC5/rHsp27 cells were cocultured and treated with 5 μM A23187 for 12, 24, 36, or 48 hours. Cell survival was then assayed by flow cytometry. Results are expressed as the mean ± SE (n = 6). *Significant difference (P < 0.05) in cell viability between cell groups.

Figure 4.

 

Double-label immunofluorescence of rHsp27 (green) and F-actin (red) in RGC-5 (A, B) and RGC-5/rHsp27 (C, D) cell lines. Expression of rHsp27 was observed only in the RGC-5/rHsp27 cell line. Before the addition of A23187 both cell lines showed normal patterns of F-actin organization in the cytoskeleton. In the RGC-5 cell line (C), the addition of A23187 (5 μM) resulted in collapse of the cytoskeleton in 95% to 98% of the cells. In the RGC-5/rHsp27 cell line (D), the addition of A23187 (5 μM) resulted in collapse of the cytoskeleton in only 25% to 30% of the cells. Original magnification, ×200.

Figure 5.

 

A23187 induced caspase-3 activity in control and RGC-5/rHsp27 cells. Separate cultures of RGC-5 and RGC5/rHsp27 cells were plated and treated with 5 μM A23187 for 12, 24, or 36 hours, and caspase-3 activity was determined. The experiment was also repeated in the presence of either a cell-permeable caspase-3 (DEVD-fmk) or caspase-9 (LEHD-fmk) inhibitor. Experiments were performed three to five times in quadruplicate. *Significant increase (P < 0.05) in caspase-3 activity in A23187-treated RGC-5 cells compared with untreated RGC-5 cells and A23187-treated RGC5/rHsp27 cells.

Figure 6.

 

Pretreatment of RGC-5 cells with a specific cell-permeable caspase-3 inhibitor did not increase cell survival. Cocultures of RGC-5 and RGC5/rHsp27 cells were treated as described earlier (Fig. 5)in either the presence or absence of a cell-permeable caspase-3 inhibitor (DEVD-fmk). Cell survival was assayed using flow cytometry. Results were nonsignificant and are expressed as the mean ± SE (n = 4).

Table 1.

 

Identified Hsp27 Complex Proteins

Protein	Peptides Sequenced	Occurrence Rate
Rat Hsp27	K.AVTQSAEITIPVTFEAR.A	4:4
	R.VSLDVNHFAPEELTVK.T
	R.RVPFSLLR.S
	R.SPSWEPFRDWYPAHSR.L
	R.LFQDQAFGVPR.F
Actin	R.VAPEEHPVLLTEAPLNPK.A	3:4
	R.VAPDEHPILLTEAPLNPK.I
	K.YPIEHGIITNWDDMEKIWHHTFYNELR.V
	R.RTTMEQPKMEMFIGEAAHAR.P
eEF-1α	K.LTNAVPHGWGQKELFIDVLQYVVRK.P	2:4
	R.EHALLAYTLGVK.Q
	R.VETGVLKPGMVVTFAPVNVTTEVK.S
	R.EHAPQAYTLGVK.Q
Hsp70	R.IINEPTAAAIAYGLDKR.E	2:4
	K.VLEDSDLKKSDIDEIVLVGGSTR.I
	R.AKFEELNMDLFR.S
Spin-2	R.KLETLQLDNSDLILFVLLCMLWR.A	2:4
	K.NETCKAAVVEGQATLEVVR.E

 The results from four separate IP experiments were analyzed for common sequenced peptides/proteins. All peptides listed were identified by automated rat database searches using the Sequest algorithm and were further confirmed with Peptide Prophet statistical analysis and manual interpretation of the raw MS-MS data. Results are listed and ranked based on the occurrence rate in the four experiments and on how many unique peptides from each protein were sequenced.