Hsp27 functions as a negative regulator of cytochrome c ... - Nature

Oncogene (2000) 19, 1975 ± 1981 ã 2000 Macmillan Publishers Ltd All rights reserved 0950 ± 9232/00 $15.00 www.nature.com/onc

Hsp27 functions as a negative regulator of cytochrome c-dependent activation of procaspase-3 Pramod Pandey1, Rebecca Farber1, Atsuko Nakazawa1, Shailendra Kumar1, Ajit Bharti1, Carlo Nalin2, Ralph Weichselbaum3, Donald Kufe1 and Surender Kharbanda*,1 1

Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts, MA 02115, USA; 2Novartis Pharmaceuticals Corporation, East Hanover, New Jersey, NJ 07936, USA; 3Department of Radiation and Cellular Oncology, University of Chicago, Chicago, Illinois, IL 60637, USA

The release of mitochondrial cytochrome c by genotoxic stress induces the formation of a cytosolic complex with Apaf-1 (mammalian CED4 homolog) and thereby the activation of procaspase-3 (cas-3) and procaspase-9 (cas9). Here we demonstrate that heat-shock protein 27 (Hsp27) inhibits cytochrome c (cyt c)-dependent activation of cas-3. Hsp27 had no eect on cyt c release, Apaf1 and cas-9 activation. By contrast, our results show that Hsp27 associates with cas-3, but not Apaf-1 or cas-9, and inhibits activation of cas-3 by cas-9-mediated proteolysis. Furthermore, the present results demonstrate that immunodepletion of Hsp27 depletes cas-3. Importantly, treatment of cells with DNA damaging agents dissociates the Hsp27/cas-3 complex and relieves inhibition of cas-3 activation. These ®ndings de®ne a novel function for Hsp27 and provide the ®rst evidence that a heat shock protein represses cas-3 activation. Oncogene (2000) 19, 1975 ± 1981. Keywords: Hsp27; cytochrome c; caspase-3; apoptosis Introduction The cellular response to genotoxic stress includes cell cycle arrest and activation of DNA repair. In the event of irreparable DNA damage, cells respond with induction of apoptosis. Apoptosis or programmed cell death, is an active process resulting in characteristic morphological changes of condensed regions of nuclear material, internucleosomal DNA fragmentation and membrane blebbing (Martin and Green, 1995). The induction of apoptosis by diverse stimuli is associated with activation of aspartate-speci®c cysteine proteases (caspases) (Martin and Green, 1995) and cleavage of PARP (Kaufmann et al., 1993), PKCd (Emoto et al., 1995; Ghayur et al., 1996) and other proteins (Widman et al., 1998). Caspases exist in cells as catalytically inactive zymogens composed of three dierent subunits: a prodomain and two catalytic subdomains (Alnemri et al., 1996). Activation of procaspases generally requires cleavage after speci®c Asp residues (Cohen, 1997; Nicholson and Thornberry, 1997). The ®nding that several apoptosis-regulating genes identi®ed in C. elegans have mammalian counterparts has suggested that the basic machinery of apoptosis is

*Correspondence: S Kharbanda Received 22 October 1999; revised 10 February 2000; accepted 15 February 2000

evolutionarily conserved (Vaux, 1997). Three genes, ced-3, ced-4 and ced-9, encode proteins that comprise the apoptotic program in C. elegans (Yuan et al., 1993; Horvitz et al., 1994). The Bcl-2 family of proteins are mammalian homologs of CED-9 (Hengartner and Horvitz, 1994; Vaux et al., 1992). CED-4 is homologous to the recently identi®ed human protein, Apaf1. Recent studies have shown that Apaf-1 binds to cas9 (initiator caspase) in the presence of cyt c and that Apaf-1-mediated oligomerization leads to cas-9 activation (Li et al., 1997). The ®nding that activated cas-9 in turn cleaves and activates cas-3 (executioner caspase) has supported a model in which the Apaf-1/cas-9 complex functions upstream to cas-3 in the apoptotic protease cascade (Li et al., 1997; Xue et al., 1996; Zou et al., 1997). Activation of the Apaf-1/cas-9 complex appears to be a common pathway employed by diverse stimuli that induce apoptosis in mammalian cells (Li et al., 1997). Cas-3 usually exists in the cytoplasm as an inactive precursor that becomes proteolytically activated in cells undergoing apoptosis (Schlegel et al., 1996; Wang et al., 1996). Cas-3 is activated by multiple cleavages of its 32 kDa precursor form to generate the 17/11 kDa or 20/11 kDa active forms (Nicholson et al., 1995; Wang et al., 1996). Other ®ndings that active cas3 cleaves cas-3 has indicated that cas-3 can be activated through autocatalysis (Wang et al., 1996). Taken together, cas-9-mediated proteolysis and/or autocatalysis of cas-3 may provide the signal ampli®cation necessary for rapid and irreversible apoptosis. By contrast, little is known about intracellular factors that negatively regulate activation of cas3. Mammalian Hsp27 belongs to the family of small HSPs that are ubiquitously expressed and exhibit the capacity to form oligomeric structures (Arrigo et al., 1988; Mehlen and Arrigo, 1994). Diverse agents, such as TNFa, IL-1 and oxidative stress have been reported as potent modulators of mammalian Hsp27 phosphorylation (Arrigo and Landry, 1994). Overexpression of Hsp27 has been shown to protect cells against necrosis induced by anti-cancer drugs, TNFa or oxidative stress (Mehlen et al., 1995, 1996a). Mammalian Hsp27 has also been shown to play a role in cell dierentiation (Shakoori et al., 1992). Moreover, recent studies have demonstrated that Hsp27 inhibits apoptosis induced in response to STSP or anti-Fas (Mehlen et al., 1996b). However, the signaling mechanisms involved in the anti-apoptotic function of Hsp27 are presently not understood. The present studies demonstrate that Hsp27 inhibits activation of cas-3 in vitro and in vivo. In contrast to

Hsp27 negatively regulates cyt c-mediated apoptosis P Pandey et al

1976

Apaf-1 and cas-9, Hsp27 binds to cas-3 and blocks activation of cas-3 by cas-9-mediated proteolysis. These ®ndings provide the evidence that a HSP functions as an intracellular repressor of cas-3 activation. Results and discussion Mild heat-shock treatment is associated with induction of Hsp27, Hsp70 and Hsp90 and the inhibition of an apoptotic response to certain types of stress (Zhou et al., 1993; Ito et al., 1997; Mehlen et al., 1997). Whereas other studies have demonstrated that release of cyt c from mitochondria induces a cas-3-mediated apoptotic cascade (Li et al., 1997; Liu et al., 1996; Zou et al., 1997), we asked if heat-shock treatment inhibits cyt cmediated activation of cas-3. As demonstrated previously for other cell types, expression of Hsp27 was induced in 293T cells at 6 ± 8 h after mild heat shock (Figure 1a). Cytosols prepared after 2, 4, 6 and 8 h following heat shock were immunodepleted of cyt c. Cyt c-dependent activation of cas-3 was then analysed after addition of cyt c and dATP to the cytosols. The results demonstrate that addition of cyt c to the depleted cytosols is associated with induction in cleavage of cas-3 (not shown). More importantly, the results demonstrate that, in contrast to cells not exposed to heat, cyt c-dependent cleavage of cas-3 is inhibited in cells with increased levels of Hsp27 (Figure 1b). Together, these ®ndings indicate that induction of HSPs by heat shock treatment is associated with inhibition of cas-3 cleavage. To further demonstrate that Hsp27 inhibits cas-3 activation in a cell free system, cyt c was immunodepleted from cytosols of L929 cells that stably express human Hsp27 (L929/Hsp27) or the empty vector (L929/pRc). Expression of Hsp27 had no detectable eect on the levels of other heat-shock protein, Hsp70 (Landry et al., 1989). Cyt c-dependent activation of cas-3 was then analysed after addition of cyt c/dATP. The results demonstrate that addition of cyt c/dATP to L929/pRc cytosols is associated with activation of cas-3 as assessed by cleavage of the cas-3 substrate, PKCd (Figure 2a). By contrast, there was no detectable cleavage of PKCd when cyt c/dATP was added to L929/Hsp27 cytosols (Figure 2a). To further assess cas-3 activity biochemically, we measured cleavage of an exogenous peptide substrate (DEVDpNA) (Thornberry and Lazebnik 1998; Mancini et al.,

Figure 1 Mild heat-shock treatment inhibits cyt c-dependent activation of cas-3. (a) 293T cells were subjected to 30 min at 428C followed by incubation at 378C for the indicated times. Total lysates were resolved on SDS ± PAGE and analysed by IB with anti-Hsp27. (b) 293T cells were subjected to 30 min at 428C followed by incubation at 378C for the indicated times. Cytosols were prepared and immunodepleted of cyt c (-Cyt c) by IP. Aliquots (50 mg) were then incubated with puri®ed cyt c (1 mg)/ dATP (1 nM) and analysed by IB with anti-cas-3 Oncogene

1998) in cytosols of L929/pRc and L929/Hsp27 cells. The results demonstrate that in the absence of cyt c, background levels of DEVD-pNA cleavage activity was observed but addition of cyt c to L929/pRc cytosols is associated with a marked increase in DEVD-pNA cleavage activity (Figure 2b). By contrast, there was little if any detectable cleavage of DECD-pNA when cyt c was added to L929/Hsp27 cytosols (Figure 2b). Taken together, these ®ndings indicate that Hsp27 inhibits cas-3 activation and thereby functions as a regulator of cyt c-mediated signaling. Previous studies have demonstrated that release of cyt c from mitochondria to cytosol is an important step in inducing cas-3-mediated apoptosis (Li et al., 1997; Liu et al., 1996; Zou et al., 1997). To assess whether Hsp27 functions upstream or downstream to cyt c release from mitochondria, L929/pRc and L929/Hsp27 cells were treated with various agents, such as methylmethane sulphonate (MMS), 1-b-D-arabinofuranosyl-cytosine (ara-C) or staurosporine (STSP), that are known to cause the release of mitochondrial cyt c (Kharbanda et al., 1997a), and cytosols were analysed by immunoblotting (IB) with anti-cyt c. The results demonstrate that MMS treatment of both L929/pRc and L929/Hsp27 cells is associated with the release of cyt c (Figure 3a, upper panel). Purity of the mitochondrial and cytoplasmic fractions was con®rmed by reprobing the blots with antibodies against mitochondrial-speci®c Hsp60 and cytoplasmic-speci®c

Figure 2 Hsp27 inhibits cyt c-mediated activation of cas-3. (a) L929/pRc (pRc) or L929/Hsp27 (Hsp27) cell cytosols were immunodepleted with anti-cyt c (-Cyt c). Aliquots (50 mg) were then incubated with or without cyt c/dATP and analysed by IB with anti-PKCd. (b) L929/pRc or L929/Hsp27 cell cytosols were immunodepleted with anti-cyt c. Depleted extracts were assayed for cleavage of the ¯uorogenic substrate DEVD-pNA. The cas-3 activity is expressed as the DEVD-pNA cleavage (6100 units/mg) (mean+s.d.) of three independent experiments


Actin proteins (Figure 3a, bottom panel and data not shown). Moreover, treatment of L929/Hsp27 cells with ara-C or STSP is associated with partial inhibition of cyt c release to that compared with L929/pRc cells (not shown). Taken together, these ®ndings demonstrate that Hsp27 acts both upstream and/or downstream to cyt c release and that this eect of Hsp27 is stimulusdependent. To assess whether activation of cas-3 was inhibited in similar experimental conditions, L929/pRc and L929/Hsp27 cells were treated with ara-C, STSP or MMS and cas-3 activity was measured by DEVD-pNA cleavage. The results demonstrate that, by contrast to L929/pRc, treatment of L929/Hsp27 cells is associated with a signi®cant inhibition in cas-3 activity (Figure 3b and data not shown). The sensitivity of this cleavage to aldehyde inhibitors (Ac-DEVD-CHO) further con®rms that the activity is due to cas-3 (not shown). To further assess that Hsp27 functions downstream to cyt c release in an in vitro cell free system, cytosols from 293T cells that express basal levels of Hsp27, were immunodepleted of both cyt c and Hsp27. Addition of cyt c to the double-depleted cytosols is

Figure 3 Hsp27 functions downstream to cyt c release in the activation of cas-3. (a) L929/pRc or L929/Hsp27 cells were treated with 1 mM MMS and harvested after 6 h. Cytosols were separated by 12.5% SDS ± PAGE and analysed by IB with anticyt c (upper panel). Cytosols were subjected to IB with anti-Actin (middle panel). Total cell lysates were also analysed by IB with anti-Hsp27 (lower panel). (b) L929/pRc or L929/Hsp27 cells were treated with 10 mM ara-C or 1 mM STSP and harvested after 6 h. Cytosols were assayed for cleavage of the ¯uorogenic substrate DEVD-pNA. The cas-3 activity is expressed as the DEVD-pNA cleavage (6100 units/mg) (mean+s.d.) of three independent experiments

associated with signi®cant inhibition in cas-3 activity, as determined by PKCd cleavage (Figure 4a). By contrast, signi®cant cleavage of PKCd was observed when cyt c was added to cytosols immunodepleted only with cyt c (Figure 4a). Cytosols were immunodepleted of cyt c and Hsp27 and assayed for PKCd cleavage in the presence of cyt c and puri®ed Hsp27. Cytosols immunodepleted of cyt c were also separately incubated with cyt c and puri®ed Hsp27. The results demonstrate that cyt c-mediated activation of cas-3 is partially inhibited by addition of Hsp27 (Figure 4b). In this context, previous studies have shown that the active forms of Hsp27 consists of large non-phosphorylated oligomers (Mehlen et al., 1997). It is thus conceivable that failure of Hsp27 to completely inhibit residual cas-3 activity in these double-depleted cytosols could be due to its inability to form active oligomers in these in vitro conditions. Collectively, these ®ndings indicate that Hsp27 functions as a inhibitor of cas-3 activation and acts downstream to mitochondrial release of cyt c in vitro as well as in vivo. Recent studies have shown that Apaf-1 promotes both activation of cas-9 and cas-9-mediated activation of cas-3 in mammalian cells (Li et al., 1997). These studies have further suggested that Apaf-1 functions as a docking protein for cas-9 and cyt c in mediating cleavage of cas-3 (Li et al., 1997). To determine whether Hsp27 inihibits cyt c-mediated activation of cas-3 by binding to Apaf-1, U-937 cells lysates were subjected to immunoprecipitation (IP) with anti-Apaf-1 and the precipitates were analysed by IB with antiHsp27. The results demonstrate that Apaf-1 is not associated with Hsp27 (Figure 5a, left panel). Similar results were obtained in reciprocal experiments in which anti-Hsp27 IPs were analysed by IB with antiApaf-1 (Figure 5a, right panel). Detection of Apaf-1 in anti-Apaf-1 precipitates demonstrate reactivity with anti-Apaf-1 antibody (Figure 5a). Since Apaf-1 functions as a docking protein for cas-9 in a cyt c-mediated cas-3 activation pathway, we next determined whether Hsp27 binds to cas-9 or acts downstream to cas-9 and thereby inhibits activation of cas-3. To address this issue, anti-cas-9 IPs were analysed by IB with anti-Hsp27. As control, anti-cas9 precipitates were also analysed by IB with anti-cas-9. The results demonstrate that, similar to Apaf-1, cas-9 is also not associated with Hsp27 (Figure 5a). Detection

1977

Figure 4 Depletion of Hsp27 Blocks cyt c-Dependent Activation of cas-3. (a) 293T cell cytosols were immunodepleted of cyt c (Cyt c) (lanes 1 and 2) or cyt c and Hsp27 (-Cyt c, -Hsp27) (lanes 3 and 4). Aliquots (50 mg) from cyt c immunodepleted cytosols were then incubated with buer or cyt c/dATP and analysed by IB with anti-PKCd. (b) 293T cell cytosols were immunodepleted of cyt c (-Cyt c) (lanes 1 and 2) or of both cyt c and Hsp27 (-Cyt c, -Hsp27) (lanes 3 and 4). Aliquots (50 mg) were then incubated with buer or cyt c/dATP with and without Hsp27 (1 mg) and analysed by IB with anti-PKCd Oncogene


1978

of cas-9 in the anti-cas-9 precipitates demonstrate reactivity with anti-cas-9 antibody (data not shown). Furthermore, to determine whether Hsp27 functions downstream to cas-9, in vitro translated [35S]-labeled cas-9 was incubated with L929/pRc or L929/Hsp27 cytosols immunodepleted of cyt c. The results demonstrate that addition of cyt c and incubation for various lengths of time, is associated with cleavage of cas-9 from both cell lines (Figure 5b). Recent studies have shown that cas-9 has an unusually active zymogen that does not require proteolytic cleavage and can be activated without proteolytic prcessing (Stennicke et al., 1999). To further assess cas-9 activity biochemically, we measured cleavage of an exogenous peptide substrate (LEHD-pNA) known to support the activity of cas-9 (Manser et al., 1999), in cytosols of L929/pRc and L929/Hsp27 cells. The results demonstrate that addition of cyt c/dATP is associated with cleavage of LEHD-PNA in cytosols from both cell lines (Figure

Figure 5 Hsp27 functions downstream to Apaf-1 and cas-9 activation. (a) U-937 cell lysates were subjected to IP with antiApaf-1, anti-cas-9 or anti-Hsp27. The resulting protein precipitates were analysed by IB with anti-Hsp27 (left panel). U-937 cell lysates were subjected to IP with anti-Apaf-1 or anti-Hsp27. The resulting precipitates were analysed by IB with anti-Apaf-1 (right panel). (b) L929/pRc (pRc) or L929/Hsp27 (Hsp27) cell cytosols were immunodepleted of cyt c (-Cyt c). Aliquots (50 mg) from cyt c immunodepleted cytosols were then incubated with 35S-labeled in vitro translated cas-9 with (lanes 2, 3, 4, 6, 7 and 8) or without (lanes 1 and 5) cyt c and dATP for 15 min (lanes 2 and 6); 30 min (lanes 3 and 7) and 60 min (lanes 4 and 8). The proteins were then separated by SDS ± PAGE and analysed by autoradiography. (c) L929/pRc (pRc) or L929/Hsp27 (Hsp27) cell cytosols were immunodepleted of cyt c. Aliquots from cyt c immunodepleted cytosols were then incubated with cyt c (+) and dATP and assayed for cleavage of the ¯uorogenic substrate LEHD-pNA. The fold induction in cleavage of LEHD-pNA is expressed as the mean+s.d. of two independent experiments Oncogene

5c). Thus, overexpression of Hsp27 has no apparent eect on cyt c-mediated activation of cas-9. Taken together, these ®ndings indicate that Hsp27 acts downstream to mitochondrial release of cyt c and functions downstream of Apaf-1 and cas-9. Cyt c forms a complex with Apaf-1 and cas-9, and is necessary at least in part for activation of cas-3 (Li et al., 1997; Liu et al., 1996; Zou et al., 1997). Since Hsp27 functions downstream to cyt c, Apaf-1 and cas9, we next determined whether Hsp27 associates with cas-3 and thereby inhibits its activity. To address this issue, cell lysates were subjected to IP with anti-Hsp27 and the IPs were analysed by IB with anti-cas-3. The results demonstrate that, by contrast to Apaf-1 and cas-9, cas-3 associates with Hsp27 (Figure 6a). As a control, anti-JNK1 IPs were devoid of cas-3 (Figure 6a). In the reciprocal experiment, the ®nding that anticas-3 IPs contain Hsp27 provided further support for an interaction between these proteins in U-937 (Figure 6b) and Hela (Figure 6c) cells. Exposure of cells to ionizing radiation (IR) and other DNA-damaging agents is associated with release of cyt c, activation of cas-3 and induction of apoptosis (Kharbanda et al., 1997a). Furthermore, recent studies have shown that cyt c is released from mitochondria of myocytes under ischemic conditions (Narula et al.,

Figure 6 Association of Hsp27 with cas-3 in vivo. (a) U-937 cell lysates were subjected to IP with anti-Hsp27, anti-cas-3 or antiJNK1. The resulting protein precipitates were analysed by IB with anti-cas-3. Total cell lysate (30 mg) was used as a positive control. (b) U-937 cell lysates were subjected to IP with anti-cas-3 or antiShc. The resulting protein precipitates were analysed by IB with anti-Hsp27. Total cell lysate was used as a positive control. (c) HeLa cell lysates were subjected to IP with anti-cas-3, anti-Hsp27 or anti-Shc. The resulting protein precipitates cas-3 were analysed by IB with anti-Hsp27. Total cell lysate was used as a positive control


1999). To determine the eect of IR on the formation Hsp27/cas-3 complexes, U-937 cells were exposed to IR and anti-Hsp27 IPs were analysed by IB with anti-cas3. The results demonstrate that treatment of cells with IR signi®cantly inhibits the interaction between Hsp27 and cas-3 (Figure 7). Similar results were obtained when U-937 cells were treated with other DNA damaging agents, ara-C or CDDP (not shown). Collectively, these results suggest that proapoptotic signals induced by DNA-damage regulate the interaction between Hsp27 and cas-3, and thereby relieve inhibition of cas-9-mediated cas-3 activation. However, our ®ndings do not exclude the possibility that dissociation of Hsp27 from cas-3 in response to stress could also be due to degradation of cas-3, downregulation of cas-3 expression or both. To con®rm that Hsp27/cas-3 complex regulates cyt c-mediated activation of cas-3, cytosols were immunodepleted of Hsp27 and the extracts before (B) and after (A) immunodepletion were analysed by IB with anticas-3. The results demonstrate that, in contrast to extracts before Hsp27 immunodepletion, a signi®cant decrease in the level of cas-3 was observed after immunodepletion of Hsp27 (Figure 8a). The functional signi®cance of the interaction between Hsp27 and cas-3 was further assessed in cytosols depleted of both cyt c and Hsp27. Addition of cyt c/dATP to the depleted cytosol had no apparent eect on cas-3 activation as measured by cleavage of PKCd (Figure 8b). By contrast, addition of puri®ed cas-3 and cyt c was associated with cleavage of PKCd (Figure 8b). The cleavage of PKCd was dependent on the amount of cas-3 protein added to the cytosols. Collectively, our ®ndings demonstrate that Hsp27 acts as a negative regulator of cas-3 activation by functioning downstream to cyt c release, Apaf-1 and cas-9 activation. Moreover, Hsp27 functions as an inhibitor of cas-3 activation by associating with and limiting the availability of cas-3 for cleavage by active cas-9. The release of cyt c from mitochondria is induced by exposure to STSP, diverse DNA damaging agents and ischemia (Kharbanda et al., 1997a; Kluck et al., 1997; Li et al., 1997; Liu et al., 1996; Yang et al., 1997; Zou et al., 1997; Narula et al., 1999). Other studies have shown that cyt c release is inhibited by the anti-apoptotic Bcl-2 and Bcl-xL proteins (Kharbanda et al., 1997a; Yang et al., 1997). The available evidence indicates that cyt c forms a cytosolic complex with

Figure 7 Treatment with IR inhibits association of Hsp27 with cas-3. (a) U-937 cells were exposed to 20 Gy IR and harvested at 6 h. Cytosols were then immunoprecipitated with anti-Hsp27 and analysed by IB with anti-cas-3. (b) The blots were scanned and the percentage inhibition in Hsp27/cas-3 complex formation is expressed as the mean+S.D of three independent experiments

Apaf-1 and cas-9 that contributes to the activation of cas-3 (Li et al., 1997; Liu et al., 1996; Zou et al., 1997). The present ®ndings demonstrate that Hsp27 is also a member of this cytosolic complex and that Hsp27 inhibits cas-3 activation. Of signi®cance is the dissociation that certain apoptotic signals conferred by DNAdamage, i.e., exposure with IR, relieve the interaction of Hsp27 and cas-3, and thereby cas-9-mediated cleavage of cas-3. Hsp27 has been implicated in the development of marginal discs in Drosophila (Pauli et al., 1990). Other studies have implicated Hsp27 in dierentiation of embryonal carcinoma and stem cells (Stahl et al., 1992), B lymphoma cells (Spector et al., 1992), osteoblasts and normal T cells (Chaoufour et al., 1996; Esposito et al., 1994). Our results de®ne a new function for Hsp27 in the regulation of cyt cmediated activation of the cas-3 cascade (Figure 9). These ®ndings provide the evidence for a negative regulator of cyt c and for involvement of a chaperone in the caspase cascade.

1979

Figure 8 Hsp27 Inhibits cyt c-Mediated Activation of cas-3. (a) 293T cell cytosols were immunodepleted of Hsp27. Lysates from B (before) and A (after) immunodepletion were separated by SDS-PAGE and analysed by IB with anti-cas-3. (b) U-937 cell cytosols were immunodepleted of cyt c (-Cyt c) (lanes 1 ± 4) or of both cyt c and Hsp27 (-Cyt c, -Hsp27) (lanes 2 ± 4). Aliquots (50 mg) were then incubated with cyt c/dATP with or without puri®ed cas-3 (lane 3 : 0.25 mg; lane 4 : 0.5 mg) and analysed by IB with anti-PKCd. FL, full length; CF, cleaved fragment

Figure 9 Schematic representation to highlight the inhibitory role of Hsp27 in cyt c-dependent activation of cas-3 Oncogene


1980

Materials and methods Cell culture and reagents U-937 cells were grown in RPMI 1640 supplemented with 10% FBS, 100 units/ml penicillin, 100 mg/ml streptomycin, and 2 mM L-glutamine. HeLa and 293T cells were grown in DMEM containing 10% FBS. L929 and L929/Hsp27 cells were grown in McCoy's 5a media. Cells were treated with 1 mM MMS (Sigma) or 100 mM CDDP (Sigma). Irradiation was performed as described (Kharbanda et al., 1997a). Isolation of cytosols Cytosols were prepared as described (Kharbanda et al., 1997a). Purity of the cytosolic fractions was determined by analysing cytosols by immunobloting with anti-Actin. Immunoprecipitation and immunoblot analysis Total cell lysates were subjected to IP as described previously (Kharbanda et al., 1997b). The sources of antibodies are: anti-Hsp27 (Santa Cruz), anti-Apaf-1 (Zou et al., 1997), anticas-9 (Li et al., 1997), anti-cyt c (Kirken et al., 1995), antiJNK1 (Santa Cruz) or anti-CDK2 (UBI). Cytosols were subjected to IP with anti-cyt c (Kirken et al., 1995). Cyt cimmunodepleted lysates were incubated with or without 1 mg puri®ed cyt c (Sigma) and 1 mM dATP for 30 min to 1 h at 308C in a ®nal volume of 30 ml. The samples were analysed by SDS ± PAGE and IB with anti-cas-3 or anti-PKCd (Santa Cruz). Assays for cas-3 and cas-9 activation L929/pRc and L929/Hsp27 cells were treated with 1 mM MMS, 10 mM ara-C or 1 mM STSP and harvested after 6 h.

Cas-3 activity assays were performed in the presence of DEVD-pNA as a substrate (BioVision, CA, USA). Cas-9 was translated and puri®ed as described (Li et al., 1997). L929/ pRC or L929/Hsp27 cell cytosols were immunodepleted of cyt c by multiple IP. A 3 ml aliquot of the in vitro translated [35S]-labeled cas-9 was incubated with 50 mg aliquots of the cytosols with or without puri®ed cyt c at 308C for 15, 30 and 60 min. Following incubation, samples were analysed by SDS ± PAGE and autoradiography. L929/pRC or L929/ Hsp27 cytosols immunodepleted of cyt c were incubated with LEHD-pNA (BioVision) in the presence of cyt c and assayed for cas-9 activity by manufacturer's protocol. Transient transfections and cell sorting Cells were transiently cotransfected with vector or Hsp27 and pEGFP in the presence of lipofectamine (GIBCO). After 36 h, cells were harvested, sorted and GFP-positive cells were 498%. Sorted cells were treated with 1 mM MMS, harvested after 3 h and the cytosols were analysed by IB with anti-cyt c or anti-PKCd.

Acknowledgments We thank Dr Lawrence Prochaska for providing antibody to cyt c; Drs Michael Bonneli, Robert Blackburn and Yong Lee for providing L929/Hsp27 cell lines and Hsp27 cDNA; Dr X Wang for anti-Apaf-1 and anti-cas-9 antibodies and Dr. Robert Talanian for cas-3 and cas-9 cDNAs. We also thank Andrew Place for technical assistance. This investigation was supported by PHS Grant CA75216 (SK) and CA29431 (DK) awarded by the National Cancer Institute, DHHS.

References Alnemri ES, Livingston DJ, Nicholson DW, Salvesen G, Thornberry NA, Wong WW and Yuan J. (1996). Cell, 87, 171 ± 173. Arrigo A-P and Landry J. (1994). In: Heat Shock Proteins: Structure, Function and Regulation. Morimoto, R., Tissieres A, and Georgopoulos C (eds). Cold Spring Harbor Press: Cold Spring Harbor, New York, pp. 335 ± 373. Arrigo A-P, Suhan JP and Welch WJ. (1988). Mol. Cell. Biol., 8, 5059 ± 5071. Chaoufour S, Mehlen P and Arrigo A-P. (1996). Cell Stress Chaperones, 1, 225 ± 235. Cohen GM. (1997). Biochem. J., 326, 1 ± 16. Emoto Y, Manome G, Meinhardt G, Kisaki H, Kharbanda S, Robertson M, Ghayur T, Wong WW, Weichselbaum R and Kufe D. (1995). EMBO J., 14, 6148 ± 6156. Esposito F, Agosti V, Morrone G, Morra F, Cuomo C, Russo T, Venuta S and Cimino F. (1994). Biochem. J., 301, 649 ± 653. Ghayur T, Hugunin M, Talanian R, Ratnofsky S, Quinlan C, Emoto Y, Pandey P, Kharbanda S, Kamen R, Wong W and Kufe D. (1996). J. Exp. Med., 184, 2399 ± 2404. Hengartner MO and Horvitz RH. (1994). Cell, 76, 665 ± 676. Horvitz HR, Shaham S and Hengartner MO. (1994). Cold Spring Harb. Symp. Quant. Biol., 59, 377 ± 385. Ito H, Okamoto K, Nakayama H, Isobe T and Kato K. (1997). J. Biol. Chem., 272, 29934 ± 29941. Kaufmann SH, Desnoyers S, Ottaviano Y, Davidson NE and Poirier GG. (1993). Cancer Res., 53, 3976 ± 3985. Kharbanda S, Pandey P, Jin S, Bharti A, Weichselbaum R, Weaver D and Kufe D. (1997b). Nature, 386, 732 ± 735.

Oncogene

Kharbanda S, Pandey P, Scho®eld L, Roncinske R, Bharti A, Yuan Z-M, Weichselbaum R, Nalin C and Kufe D. (1997a). Proc. Natl. Acad. Sci. USA, 94, 6939 ± 6942. Kirken RA, Lincoln AJ, Fink PS and Prochaska LJ. (1995). Protein Expression & Puri®cation, 6, 707 ± 715. Kluck RM, Bossy-Wetzel E, Green DR and Newmeyer DD. (1997). Science, 275, 1132 ± 1136. Landry J, Chretien H, Lambart H, Hickey E and Weber LA. (1989). J. Cell Biol., 109, 7 ± 15. Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES and Wang X. (1997). Cell, 91, 479 ± 489. Liu X, Naekyung Kim C, Yang J, Jemmerson R and Wang X. (1996). Cell, 86, 147 ± 157. Mancini M, Nicholson DW, Roy S, Thornberry NA, Peterson EP, Casiola-Rosen LA and Rosen A. (1998). J. Cell Biol., 140, 1485 ± 1495. Manser PW, Bible KC, Martins LM, Kottke TJ, Srinivasula SM, Svingen PA, Chilcote IJ, Basi GS, Tung JS, Krajewski S, Reed JC, Alnemri ES, Earnshaw WC and Kaufmann SH. (1999). J. Biol. Chem., 274, 22635 ± 22645. Martin SJ and Green DR. (1995). Cell, 82, 349 ± 352. Mehlen P and Arrigo AP. (1994). Eur. J. Biochem., 221, 327 ± 334. Mehlen P, Kretz-Remy C, Briolay J, Fostan P, Mirault ME and Arrigo A-P. (1995). Biochem. J., 312, 367 ± 375. Mehlen P, Kretz-Remy C, Preville X and Arrigo AP. (1996a). EMBO J., 15, 2695 ± 2706. Mehlen P, Mehlen A, Godet J and Arrigo A-P. (1997). J. Biol. Chem., 272, 31657 ± 31665. Mehlen P, Schulze-Ostho K and Arrigo A-P. (1996b). J. Biol. Chem., 271, 16510 ± 16514.


Narula J, Pandey P, Arbustini E, Haider N, Narula N, Kolodgie FD, Bello BD, Semigran MJ, Bielsa-Masdeu A, Israels S, Ballester M, Virmani R, Saxena S and Kharbanda S. (1999). Proc. Natl. Acad. Sci. USA., 96, 8144 ± 8149. Nicholson DW, Ali A, Thornberry NA, Vaillancourt JP, Ding CK, Gallant M, Gareau Y, Grin PR, Labelle M, Lazebnik YA, Munday NA, Raju SM, Smulson ME, Yamin T-T, Yu VL and Miller DK. (1995). Nature, 376, 37 ± 43. Nicholson DW and Thornberry NA. (1997). TIBS, 22, 299 ± 306. Pauli D, Tonka C-H, Tisseres A and Arrigo A-P. (1990). J. Cell. Biol., 111, 817 ± 828. Schlegel J, Peters I, Orrenius S, Miller DK, Thornberry NA, Yamin T-T and Nicholson DW. (1996). J. Biol. Chem., 271, 1841 ± 1844. Shakoori AR, Oberdorf AM, Owen TA, Weber LA, Hickey E, Stein JL, Lian JB and Stein GS. (1992). J. Cell Biochem., 48, 277 ± 287. Spector NL, Samson W, Ryan C, Gribben J, Welch WJ and Nadler LM. (1992). J. Immunol., 148, 1668 ± 1673. Stahl J, Wobus AM, Thrig S, Lutsch G and Bielka H. (1992). Dierentiation, 51, 33 ± 37.

Stennicke HR, Deveraux QL, Reed JC, Dixit VM and Salvesen GS. (1999). J. Biol. Chem., 274, 8359 ± 8362. Thornberry NA and Lazebnik Y. (1998). Science, 281, 1312 ± 1316. Vaux DL. (1997). Cell, 90, 389 ± 390. Vaux DL, Weissman IL and Kim SK. (1992). Science, 258, 1955 ± 1957. Wang X, Zelenski NG, Yang J, Sakai J, Brown MS and Goldstein J. (1996). EMBO J., 15, 1012 ± 1020. Widmann C, Gibson S and Johnson GL. (1998). J. Biol. Chem., 273, 7141 ± 7147. Xue D, Shaham S and Horvitz HR. (1996). Genes Dev., 10, 1073 ± 1083. Yang J, Liu X, Bhalla K, Kaekyung Kim C, Ibrado AM, Cai J, Peng T-I, Jones DP and Wang X. (1997). Science, 275, 1129 ± 1132. Yuan J, Shaham S, Ledoux S, Ellis HM and Horvitz HR. (1993). Cell, 75, 641 ± 652. Zhou M, Lambert H and Landry J. (1993). J. Biol. Chem., 268, 35 ± 43. Zou H, Henzel WJ, Liu X, Lutschg A and Wang X. (1997). Cell, 90, 405 ± 413.

1981

Oncogene