Heat Shock Protein 70 Inhibits Apoptosis Downstream of Cytochrome c Release and Upstream of Caspase-3 Activation*

Heat shock protein 70 (HSP70) has been shown to act as an inhibitor of apoptosis. We have also observed an inhibitory effect of HSP70 on apoptotic cell death both in preheated U937 and stably transfected HSP70-overexpressing U937 (U937/HSP70) cells. However, the molecular mechanism whereby HSP70 prevents apoptosis still remains to be solved. To address this issue, we investigated the effect of HSP70 on apoptotic processes in an in vitro system. Caspase-3 cleavage and DNA fragmentation were detected in cytosolic fractions from normal cells upon addition of dATP, but not from preheated U937 or U937/hsp70 cells. Moreover, the addition of purified recombinant HSP70 to normal cytosolic fractions prevented caspase-3 cleavage and DNA fragmentation, suggesting that HSP70 prevents apoptosis upstream of caspase-3 processing. Because cytochrome c was still released from mitochondria into the cytosol by lethal heat shock despite prevention of caspase-3 activation and cell death in both preheated U937 and U937/hsp70 cells, it was evident that HSP70 acts downstream of cytochrome c release. Results obtained in vitrowith purified deletion mutants of HSP70 showed that the carboxyl one-third region (from amino acids 438 to 641) including the peptide-binding domain and the carboxyl-terminal EEVD sequence was essential to prevent caspase-3 processing. From these results, we conclude that HSP70 acts as a strong suppressor of apoptosis acting downstream of cytochrome c release and upstream of caspase-3 activation.

tors, growth factor deprivation, excessive DNA damage, treatment with chemotherapeutic drugs, or stresses such as heat shock, hyperosmotic shock, or UV irradiation (8 -11).
Apoptosis signals such as Fas ligand and tumor necrosis factor activate procaspase-8 through molecular interactions between components of the death-inducing signaling complex (12)(13). Activated caspase-8 cleaves Bid (Bcl-2 interacting protein), and the carboxyl-terminal domain translocates to mitochondria where it initiates cytochrome c release (14). The release of cytochrome c triggers the formation of a complex containing Apaf1, a mammalian CED-4 homologue, and procaspase-9, which is then autoprocessed and thereby capable of processing downstream effector procaspases such as procaspase-3 (15). The processing of these caspases is followed by the cleavage of apoptotic substrates, leading to the disruption of important cellular processes, changes in cellular and nuclear morphology, and ultimately to cell death (16 -18). Diverse apoptotic signals for caspase activation converge at the mitochondria level, provoking the release of cytochrome c, which participates in the central control or executioner phase of the cell death cascade (19 -21). However, the role of cytochrome c release in apoptosis is still confusing and contradictable, and the mechanism of cytochrome c release has not been elucidated.
Because ischemia and neurodegenerative diseases such as Alzheimer's and Huntington's diseases may result from excessive apoptosis, it may be beneficial to limit apoptosis as a way to manage these diseases (22,23). Cellular and viral proteins such as Bcl-2, CrmA, and IAP 1 might serve as therapeutic agents to inhibit apoptosis, because Bcl-2 prevents cytochrome c release, CrmA inhibits caspase-8, and IAP blocks caspase-3 activation (13). In addition, HSP70 has been suggested as a promising molecule for controlling apoptosis because HSP70overexpressing transgenic mice showed reduced brain and heart ischemia (24,25).
HSPs can be induced by various stresses such as ethanol, amino acid analogues, infection, inhibitors of energy metabolism, and heavy metals (26). In addition to its chaperoning function for folding, transport, and assembly of newly synthesized polypeptides (27,28), HSP70 protects cells from a number of apoptotic stimuli, including heat shock, tumor necrosis factor, growth factor withdrawal, oxidative stress, chemotherapeutic agents, ceramide, and radiation (29 -33). HSP70 prevents caspase-3 and SAPK/JNK activation in heat shock-or ceramide-induced apoptosis (34,35). Despite recent advances, the anti-apoptotic mechanism of HSP70 is still controversial (35,36). Therefore, the aim of this study is to elucidate which step in the apoptosis pathway is affected by HSP70. We used an in vitro apoptosis system as well as preheated and HSP70overexpressing cells. The results demonstrate that HSP70 can inhibit apoptosis downstream of cytochrome c release, but upstream of caspase-3 cleavage, and that the carboxyl-terminal region containing the peptide-binding domain is sufficient to inhibit caspase-3 activation.

EXPERIMENTAL PROCEDURES
Chemicals and Reagents-The deoxynucleotide triphosphates' pepstatin A, leupeptin, and N-acetyl-leucyl-leucyl-norleucina (ALLN) were purchased from Roche Molecular Biochemicals. Phenylmethylsulfonyl fluoride, aprotinin, and bovine heart cytochrome c were obtained from Sigma. The Cpp32/caspase-3 colorimetric protease assay kit was obtained from Medical & Biological Laboratories Co. (Nagoya, Japan). The secondary antibodies, goat anti-rabbit or anti-mouse lgG conjugated to horseradish peroxidase, were obtained from Pierce. The molecular weight standards for SDS-polyacrylamide gel electrophoresis were obtained from Bio-Rad. Enhanced chemiluminescence (ECL) reagent for Western blotting detection was purchased from Amersham Pharmacia Biotech.
Antibodies for immunoblotting were purchased from the following sources. Anti-cytochrome c (7H8.2C12) and cytochrome oxidase II were purchased from PharMingen (San Diego, CA) and Molecular Probes (Eugene, OR), respectively. Anti-caspase-3 (CPP32) and anti-PARP were obtained from Transduction Laboratories (Lexington, KY) and Roche Molecular Biochemicals, respectively. Anti-HSP70 and anti-actin antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).
Cell Cultures and Heat Shock Conditions-U937 cells were grown in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum and 50 g/ml penicillin-streptomycin in a 5% CO 2 incubator. Culture plates containing cells were wrapped tightly with parafilm and immersed in a water bath at the desired temperature. Mild heat shock conditions to induce HSP70 synthesis was 42°C for 30 min, and lethal heat shock conditions were 43°C for 60 or 80 min.
Stable Transfection-To establish a retrovirus-producing cell line, 10 g of MFG.HPS70.puro and MFG.puro plasmids (37) were transfected into the PA317 retrovirus packaging cell line (ATCC; CRL 9078) by thecalcium phosphate precipitation method (38). After 48 h, cells were harvested, and the cells were incubated in the selective media containing 2 g/ml puromycin. The selected cells were maintained as a population and used for retrovirus production. The viral infection was performed as described previously (39). Briefly, the viral supernatant was harvested from the overnight culture of 70% confluent producer cell layers and filtered through a 0.45-m syringe filter (Corning, Corning, NY). The viral supernatant, supplemented with 8 g/ml polybrene (Sigma), was mixed with target cells in a conical tube and incubated for 2 h at 37°C. The cells were then washed two times with PBS. After a 48 h incubation, puromycin (2 g/ml) was added to the culture medium to select the transfected cells. The selection was continued for 10 days after which the cells were maintained as a population and the selected cells were used for the experiments.
Cell Viability and Apoptosis Assay-For quantitative analysis of cell viability, cells were washed once with ice-cold PBS and incubated with annexin V-fluorescein/PI (Roche Molecular Biochemicals) in calciumcontaining HEPES buffer for 15 min at room temperature. The cells were then analyzed with a FACScan machine (FACSCalibur, Becton Dickinson).
Measurement of Caspase-3 Activity-Caspase-3 activity was measured according to the manufacturer's protocol. The cells (5 ϫ 10 6 ) were lysed with 250 l of chilled cell lysis buffer on ice for 10 min. After microcentrifugation (10,000 ϫ g, 1 min, 4°C), the supernatant was used for caspase-3 colorimetric protease assay. 170 g of protein was diluted to 50 l of cell lysis buffer and mixed with 50 l of 2ϫ reaction buffer (containing 10 mM DTT) and 5 l of 4 mM Asp-Glu-Val-Asp-p-nitroanilide (DEVD-pNA) (200 M final concentration). After incubation at 37°C for 2 h, sample were read at 405 nm in a microtiter plate reader.
Preparation of the S-100 Fraction-The S-100 fraction was prepared essentially as described (40) with several modifications. Briefly, cells were washed twice with ice-cold PBS and suspended in 3 volumes of buffer A (20 mM Hepes-KOH, pH 7.5, 10 mM KCl, 1.5 mM MgCl 2 , 1 mM sodium EDTA, 1 mM sodium EGTA, 1 mM DTT), supplemented with protease inhibitors (5 g/ml pepstatin A, 10 g/ml leupeptin, 2 g/ml aprotinin). After sitting on ice for 15 min, the cells were disrupted by douncing 15 times in a 2-ml Wheaton Dounce homogenizer (Kontes Glass Co.). The cell extract was clarified by centrifugation for 10 min at 4°C, and the supernatant was further centrifuged at 10 5 ϫ g for 1 h in a table-top Beckman Ultracentrifuge. This resulting supernatant (S-100 fraction) was stored at Ϫ70°C and used for in vitro apoptosis assays.
Isolation of Mouse Liver Nuclei-Liver nuclei were prepared according to the method of Liu et al. (40). Livers from four FVB strain male mice were rinsed with ice-cold PBS and homogenized in buffer B (10 mM Hepes-KOH, pH 7.6, 2.4 M sucrose, 15 mM KCl, 2 mM sodium EDTA, 0.15 mM spermine, 0.15 mM spermidine, 0.5 mM DTT, and 0.5 mM phenylmethylsulfonyl fluoride) by 10 strokes of a tissue-tearer. The homogenates were centrifuged through 10 ml of buffer B at 25,000 rpm for 1 h in a SW 28 rotor (Backman) at 4°C. The nuclei pellets were resuspended in buffer C (10 mM PIPES, pH 7.4, 80 mM KCl, 20 mM NaCl, 5 mM sodium EGTA, 250 mM sucrose, and 1 mM DTT) at 8.5 ϫ 10 7 nuclei/ml and stored at Ϫ80°C in multiple aliquots.
Purification of HSP70 and Its Deletion Mutants-The full-length human HSP70 gene was cloned into pET30a (Novagen). Several deletion mutants of HSP70 were created and ligated into pET30a as follows: hsp70⌬B lacking the BglII-BglII fragment, hsp70⌬S lacking the SmaI-SmaI fragment, hsp70⌬N lacking the NcoI-NcoI fragment, and hsp70⌬C deleted nucleotides 1756 -1923. All recombinant plasmids were transformed into Escherichia coli BL21. Bacteria were grown at 37°C to an optical density of 0.6, and proteins were induced at 37°C with 1 mM isopropyl-␤-D-thiogalactopyranoside for 2 h. His-tagged HSP70 and its deletion mutant proteins were purified through a nickel affinity column by passing the soluble fractions of cell lysates. Eluted proteins were dialyzed against buffer A.
DNA Fragmentation Assay-To determine the degradation of chromosomal DNA into nucleosome-sized fragments in the in vitro apoptosis system, aliquots of 50 l of cytosolic faction (S-100) and 2 ϫ 10 5 nuclei were incubated at 37°C for 1 h in the presence of dATP. A 500-l aliquot of buffer D (100 mM Tris-HCl, pH 8.5, 5 mM EDTA, 0.2 M NaCl, 0.2% (w/v) SDS, and 0.2 mg/ml proteinase K) was added to the reaction mixture and incubated at 37°C overnight. DNA was obtained by ethanol precipitation, separated on a 1.8% agarose gel, and visualized under UV light.
DNA fragmentation was quantitated as described previously (41) with some modifications. Briefly, exponentially growing cells (2.5 ϫ 10 5 cells/ml) were labeled with 10 Ci/2.5 ϫ 10 6 cells of [methyl-3 H]thymidine for 5 h and washed three times with nonradioactive fresh medium. Labeled cells were exposed to lethal heat shock for different times and returned to 37°C for 12 h. Cells were harvested in a microcentrifuge tube labeled B, and supernatants were stored in a tube labeled S. The cell pellets were suspended with TTE buffer (containing 10 mM Tris-HCl, pH 7.4, 0.2% Triton X-100, and 1 mM EDTA). After 30 min on ice, samples were centrifuged at 14,000 rpm for 10 min, the supernatants were transferred to a tube labeled T, and the pellets were resuspended in STE buffer (containing 1% SDS, 10 mM Tris-HCl, pH 7.4, and 1 mM EDTA) and sonicated for 20 s. The radioactivity from each tube was determined by liquid scintillation counting. The percent DNA fragmentation was calculated as follows: Assay for Cytochrome c Release-After a lethal heat shock treatment, cells were returned to 37°C, and cytochrome c release was analyzed at 12 h. Cells were centrifuged at 600 ϫ g for 10 min at 4°C, and the cell pellets were washed once with ice-cold PBS and resuspended with 5 volumes of buffer A containing 250 mM sucrose. After incubation on ice for 15 min, the cells were homogenized with 8 strokes of a 2-ml Wheaton Dounce homogenizer (Kontes Glass Co.). After the homogenate was centrifuged (750 ϫ g, 10 min at 4°C) twice, the supernatants were centrifuged again at 12,000 ϫ g for 15 min at 4°C. The supernatants were stored at 4°C as cytosolic fractions. The mitochondrial pellets were washed once with PBS, lysed with radioimmunoprecipitation buffer, and after 30 min on ice were centrifuged at 14,000 rpm for 10 min. The resulting mitochondrial fractions and the cytosolic fractions were used for Western blot analysis with an anti-cytochrome c antibody.
Assay for Caspase-3 and PARP Cleavages-Detection of caspase-3 in the in vitro system was carried out as follows. Cytosolic extracts (250 g of protein) from preheated or HSP70-overexpressing cells were mixed with 2 ϫ 10 5 mouse liver nuclei in the presence of 300 or 500 M dATP and incubated for 1 h at 37°C. At the end of the incubation, samples were centrifuged, and aliquots of supernatant (50 g of protein) were mixed with 2ϫ SDS-sample buffer. After boiling for 5 min, samples were subjected to 12% SDS-polyacrylamide gel electrophoresis, and the proteins were transferred to nitrocellulose membranes. The membranes were incubated with blocking buffer (10% non-fat milk, 0.1% Tween 20, 20 mM Tris-HCl, pH 7.4, 150 mM NaCl) for 1 h at room temperature and then with anti-caspase-3 for 1 h at room temperature. The membranes were washed with washing buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% Tween 20, 1% non-fat milk) three times and then incubated for 1 h at room temperature with horseradish peroxidase-conjugated anti-mouse immunoglobulin G. Antibody detection was performed using the ECL detection kit (Amersham Pharmacia Biotech).
To detect PARP cleavages in an intact cell system, whole cell lysate was prepared from cells treated with lethal heat shock, and then immunoblotted with anti-PARP antibody as mentioned above.

Mild Heat Shock Prevents Apoptosis Upstream of Caspase-3 Activation in an in
Vitro System-We first examined the effect of elevated HSP expression on heat resistance in U937 cells. A mild heat shock treatment (42°C, 30 min) that did not induce apoptosis (data not shown) was used to induce HSP70 synthesis. 12 h after the heat treatment, the level of HSP70 in the preheated cells was substantially elevated relative to that of the control cells (Fig. 1A).
When DNA fragmentation was observed in cells exposed to lethal heat shock, the preheated cells, in contrast to the control cells, showed a dramatic level of resistance to DNA fragmentation (Fig. 1B). DNA fragmentation measured in preheated cells was decreased 3.6 -4.0-fold when HSP70 was induced by a mild heat treatment.
To elucidate which step in the apoptotic pathway is regulated by HSP70, we employed an in vitro apoptosis system based on the ability of exogenously added dATP to induce apoptosis in an S-100 fraction containing cytochrome c (40). Extracts prepared from control and preheated cells were incubated with exogenous liver nuclei and dATP. The extent of apoptosis activation was determined by monitoring the proteolytic processing of pro-caspase-3 and the intranucleosomal fragmentation of the mouse liver DNA. Processing of procaspase-3 into the p19 and p17 subunits (the small p12 subunit of caspase-3 was undetectable with the anti-caspase-3 antibody used for this study), which occurred in the extract from control cells, was completely inhibited in the extract from preheated cells (Fig. 1C). Additionally, intranucleosomal fragmentation of the mouse liver DNA was almost completely blocked in the extract from preheated cells, whereas extensive DNA fragmentation occurred in the extract prepared from control cells (Fig. 1D).
HSP70 Is a Major Inhibitor of Apoptosis and Caspase-3 Activation-Because the exposure of cells to mild heat shock can induce the synthesis of several different heat shock proteins (33), we prepared HSP70-overexpressing cells (U937/hsp70) to address the question of whether HSP70 itself inhibits apoptosis in the in vitro system. The level of HSP70 accumulation in U937/hsp70 cells was much higher than in control cells (U937/ puro) and was very similar to that of preheated cells (Fig. 1A). When the percent of DNA fragmentation was observed in U937/hsp70 cells after exposing to lethal heat shock treatments, the U937/hsp70 cells also showed a dramatic level of resistance to DNA fragmentation in contrast to the U937/puro cells (Fig. 2A). These results are very similar to those obtained with preheated cells. Fig. 2, B and C, shows the results of the in vitro apoptosis assay using extracts from U937/hsp70 cells. In contrast to U937/puro, caspase-3 activation and intranucleosomal DNA fragmentation were also prevented in cytosolic extracts of U937/hsp70 cells as shown in preheated cells. This result suggests that HSP70 per se blocks caspase-3 cleavage in heat shock-induced apoptosis.
This antiapoptotic function of HSP70 is a general phenomenon, because we have observed similar results in a different cell line, HL60. When these cells accumulated HSP70 by mild heat shock or by HSP70 transfection, they also showed an inhibition of caspase-3 activation, PARP cleavage, and DNA fragmentation in response to a lethal heat shock compared with control HL60 cells (data not shown); this supports the suggestion that HSP70 is a general inhibitor of apoptosis acting upstream of caspase-3 activation.
As a further test of the ability of HSP70 to function as an inhibitor of apoptosis, the effects of recombinant HSP70 on caspase-3 activation and DNA fragmentation were determined. HSP70 protein was purified using a bacterial expression system. Reaction mixtures containing recombinant HSP70 and cytosolic extracts from control U937 cells were incubated in vitro and examined for caspase-3 activation. Purified recombinant HSP70 clearly inhibited caspase-3 cleavage in comparison with bovine serum albumin (Fig. 3A). Inhibition of DNA fragmentation was also observed in extracts containing recombinant HSP70 but not in extracts containing added bovine serum albumin (Fig. 3B). These results demonstrate that HSP70 itself prevents the activation of key molecules associated with apoptosis in an in vitro system.
Heat Shock Induces Cytochrome c Release from Mitochondria, and This Is Not Blocked by HSP70 -Cytochrome c acts as an important molecule at the early stage of apoptosis pathway. Its release from mitochondria leads to the activation of caspase-9, which then converts procaspase-3 into its active form, resulting in apoptosis. To address the question of whether lethal heat shock initiates apoptosis via cytochrome c release, the effect of heat shock on cytochrome c release was determined. After U937 cells were treated with lethal heat shock, cytochrome c release was analyzed by immunoblot analysis. Cytochrome c was increased in cytosol after lethal heat shock (Fig. 4A). Because HL60 cells also showed cytochrome c release by lethal heat shock (data not shown), this indicates that cytochrome c participates in the executioner phase of apoptotic cell death cascade in response to heat shock.
Since we found that HSP70 blocks apoptosis at some point upstream of caspase-3 activation, we next examined whether HSP70 affects cytochrome c release, which is further upstream step in the apoptosis pathway. To address this question, the presence of cytosolic cytochrome c was measured in preheated U937 and U937/hsp70 cells treated with lethal heat shock. As shown in Fig. 4A, HSP70 had no effect on the release of cytochrome c in lethal heat shocked cells. Because cytochrome c oxidase subunit II, an inner mitochondrial membrane protein (20), was not detected in the cytosolic extracts, it was determined that there was no contamination of intact mitochondria in the cytosolic fractions.
To know exactly the extent of the caspase-3 and PARP cleavages in preheated U937 and U937/hsp70 cells detecting the cytochrome c in the cytosol, we observed caspase-3 and PARP cleavages in lethal heat-shocked cells (Fig. 4A). Despite cytochrome c release by lethal heat shock, the cleavages of procaspase-3 and PARP were protected in both preheated U937 and U937/HSP70 cells in contrast to normal U937 and U937/ puro cells, respectively. At the same time, active caspase-3 capable of cleaving DEVD-pNA was analyzed in the same experiment. Caspase-3 activities in both preheated U937 and U937/HSP70 cells upon exposure to lethal heat shock were very low (Fig. 4B). To clarify whether HSP70 protected the cells from dying when caspase-3 activity was blocked even after cytochrome c release, cell viability was also observed in the same experiment. For measurement of cell viability, annexin V binding and PI uptake were analyzed by flow cytometry in control cells and in preheated U937 cells that were exposed in lethal heat shock (Fig. 4C). In the case of preheated cells, the percent of early or late apoptotic/necrotic cells (annexin V ϩ / PI Ϫ or annexin V ϩ /PI ϩ cells) was clearly decreased in comparison to the control. The extent of apoptotic cell death in U937/hsp70 cells was also compared with that of U937/puro cells by flow cytometry analysis. As shown in Fig. 4C, U937/ hsp70 cells showed substantially less apoptosis than U937/ puro cells. All of the parameters in Fig. 4 were measured in a single experiment with preheated U937 or U937/hsp70 cells, respectively. Taken together, regardless of the inability of HSP70 to block cytochrome c release, HSP70 could interfere with caspase-3 activation and finally cell death.
Peptide-binding Domain and Carboxyl-terminal Region Containing the Last 57 Amino Acids of HSP70 Are Necessary for Inhibition of Apoptosis-To determine which regions of HSP70 might be involved in its anti-apoptotic function, several deletion derivatives of HSP70 were constructed. The regions deleted in the HSP70 mutant proteins HSP70⌬B, HSP70⌬S, HSP70⌬N, and HSP70⌬C are shown in Fig. 5A. HSP70 and mutant proteins were purified as Histidine-tagged fusion proteins using a bacterial expression system, and the purified proteins were analyzed by SDS-polyacrylamide gel electrophoresis (Fig. 5B). The purified recombinant proteins were added to an in vitro apoptosis system to monitor the effects of the HSP70 mutant proteins on caspase-3 activation. Mutant HSP70⌬B and HSP70⌬N proteins prevented caspase-3 cleavage as the full-length HSP70 protein did (Fig. 5C). In contrast, HSP70⌬S and HSP70⌬C mutant proteins had lost the ability to inhibit caspase-3 activation (Fig. 5C). These results suggest that the carboxyl one-third region of HSP70 (amino acids 438 -641), including the peptide-binding domain and EEVD motif, is required to prevent the activation of caspase-3 in vitro.

DISCUSSION
HSPs are evolutionarily conserved molecular chaperones that are essential for the proper folding and assembly of proteins. The proteins are structurally and functionally conserved from prokaryotes to mammals. HSP70 provides protection from elevated temperature and contributes to thermotolerance. This ability of HSP70 to protect cells has recently been shown to be a consequence of inhibition of apoptosis (Refs. [29][30][31][32][33][34][35][36][37]reviewed in Ref. 43). It has been suggested that HSP70 prevents apoptosis by inhibiting the SAPK/JNK signaling cascade (35,44,45). However, in some cases, HSP70-expressing cells were found to be resistant to apoptosis without any effect on SAPK/ JNK signaling (9,35,36). Although survival was shown to correlate with impaired caspase-3 activation, Jää ttelä et al. (36) recently documented an effect of HSP70 downstream of caspase-3. In that study, HSP70 had no effect on SAPK/JNK or caspase-3 activation but did prevent apoptosis induction. It has also been suggested that other heat-inducible factors might reduce SAPK/JNK activation and that HSP70 might act downstream of SAPK/JNK (9). Because the anti-apoptotic mechanism of HSP70 is still controversial, we sought to determine where in the pathway between cytochrome c release and caspase-3 activation HSP70 acts. We obtained evidence that apoptosis induced by heat stress was mediated by the cyto- The presence of cytochrome c in the cytosolic fraction was detected by immunoblot assay. Cytochrome c oxidase subunit II (inner mitochondrial membrane marker) were used to show no contamination of mitochondria in the cytosolic fractions. Caspase-3 and PARP cleavages were also detected by immunoblot assay in a same experiment. An actin protein was used as a control for equal loading of protein. Mit. fr., mitochondrial fractions. B, caspase-3 activity in preheated or U937/hsp70 cells. Caspase-3 activity was measured in the same experiment that detected cytochrome c release. The results presented were reproducible in two separate experiments. C, viability of preheated or U937/hsp70 cells upon exposure to the lethal heat shock. The portion of the cells that was treated with lethal heat shock to determine cytochrome c release and caspase-3 activation was used to measure cell viability. To measure the rate of cell viability, the cells were double-labeled with annexin V and PI and analyzed by a flow cytometry. chrome c pathway. We observed that although HSP70 does not block cytochrome c release, HSP70 clearly prevents caspase-3 activation, PARP cleavage, DNA laddering, and finally apoptotic cell death in both an in vitro system and an intact cell system. That is, HSP70 prevents apoptosis downstream of cytochrome c release and upstream of caspase-3 activation. This kind of mechanism against apoptosis was supported by the report that HSP27 inhibits apoptosis downstream of cytochrome c release without preventing cytochrome c redistribution (46). No correlation has been reported between cytochrome c release and apoptosis, since cytochrome c can also be released in viable cells, whereas cells might undergo apoptosis without releasing cytochrome c (42,47). Our results suggest that HSP70 is a general apoptosis inhibitor similar to such proteins as Bcl-2, IAPs, and CrmA, even though each of them acts on a different stage of apoptosis.
Following the finding that Bcl-2 is an anti-apoptotic protein, several other apoptosis inhibitors have been described. Bcl-2related molecules regulate apoptosis signaling by blocking cytochrome c release from mitochondria and the ensuing activation of caspase-9, as well as affecting calcium homeostasis (43). IAPs bind and inhibit caspase-3 in Fas/caspase-8-induced apoptosis and bind to pro-caspase-9 to prevent its processing in the mitochondrial pathway of apoptosis (48). These IAP family proteins also presumably interfere with apoptotic pathways within an amplification loop between caspase-3 and caspase-9 cleavage (49). The CrmA protein is able to bind to and inhibit caspase-1 and caspase-8, but it does not protect against cell death mediated by other caspases (50). HSP70 has been hypothesized to inhibit apoptosis through its chaperoning function by restoring proteins that have been damaged or unfolded by stresses. This model does not specify a unique target for HSP70. However, HSP70 can suppress activation of apoptosis by non-protein-damaging stimuli, and HSP70 is directly implicated in the regulation of a SAPK/JNK phosphatase activity (51,52), implying that HSP70 might inhibit apoptosis through effects on a specific target protein. This hypothesis is supported by our data, which suggest that the molecule(s) between cytochrome c release and caspase-3 activation might be a target protein of HSP70 in the prevention of apoptosis. Ghibelli et al. (47) insist that cytosolic cytochrome c per se does not necessarily lead to apoptosis, and released cytochrome c might have to wait until the cofactor is available. Therefore, the possibility that HSP70 might act on a cofactor molecule could not be excluded.
We obtained results showing that the peptide-binding domain and carboxyl-terminal region containing the last 57 amino acids of HSP70 are indispensable for the inhibition of apoptotic events in a cell-free system. HSP70 contains three functional regions, the ATP-binding domain, the peptide-binding domain, and the EEVD motif, that are conserved in all of the eukaryotic HSP70 at the carboxyl-terminal end of the protein (53). Whereas HSP70 lacking the ATP-binding domain has been shown to provide thermotolerance, the loss of the peptidebinding domain disrupts the protective effect of HSP70 (54). Because it has been reported that the deletion or mutation of the EEVD motif results in a loss of substrate binding ability, the EEVD motif might be important for thermotolerance (50). Our result is consistent with a recent report showing that the HSP70 mutant lacking the ATP-binding domain was still capable of SAPK/JNK suppression (52). These results strongly suggest that the protein refolding activity of HSP70 is unnecessary for its anti-apoptotic function. From these results, we surmised that HSP70 could directly bind to and inhibit the activity of key molecules acting at the execution phase of apoptosis to prevent cell death. It remains to be determined whether HSP70 directly interacts with a component of the pathway leading from cytochrome c release to caspase-3 activation. Currently, we are investigating this possibility.
It has recently been suggested that other heat shock proteins, HSP27, HSP60, HSP10, and HSP90, are also important regulators of apoptosis (55)(56)(57)(58). In addition to their protective effects, in some situations various HSPs have been found to accelerate apoptotic processes (56 -58). As inappropriate apoptosis is implicated in a number of human diseases, understanding the molecular mechanisms of action of these proteins could offer novel ways of treating apoptosis-related diseases. Several HSPs, especially HSP70, might be useful for the therapeutic treatment of cancer, autoimmune and immunodeficiency diseases, injury after ischemia, and neurodegenerative disorders.