Activation of c-Abl tyrosine kinase requires caspase activation ... - Nature

Oncogene (1999) 18, 1277 ± 1283 ã 1999 Stockton Press All rights reserved 0950 ± 9232/99 $12.00 http://www.stockton-press.co.uk/onc

Activation of c-Abl tyrosine kinase requires caspase activation and is not involved in JNK/SAPK activation during apoptosis of human monocytic leukemia U937 cells Shingo Dan1, Mikihiko Naito1, Hiroyuki Seimiya2, Atsuo Kizaki1, Tetsuo Mashima1 and Takashi Tsuruo1,2,3 1

Institute of Molecular and Cellular Biosciences, University of Tokyo, 1-1-1, Yayoi, Bunkyo-Ku, Tokyo 113-0032, 2Cancer Chemotherapy Center, Japanese Foundation for Cancer Research, Tokyo 170-8455, Japan

Genotoxic stress triggers the activation of several sensor molecules, such as p53, JNK1/SAPK and c-Abl, and occasionally promotes the cells to apoptosis. We previously reported that JNK1/SAPK regulates genotoxic stress-induced apoptosis in p53-negative U937 cells by activating caspases. c-Abl is expected to act upstream of JNK1/SAPK activation upon treatment with genotoxic stressors, but its involvement in apoptosis development is still unclear. We herein investigated the kinase activities of c-Abl and JNK1/SAPK during apoptosis elicited by genotoxic anticancer drugs and tumor necrosis factor (TNF) in U937 cells and their apoptosis-resistant variant UK711 cells. We found that the activation of JNK1/SAPK and c-Abl correlated well with apoptosis development in these cell lines. Unexpectedly, however, the JNK1/SAPK activation preceded the c-Abl activation. Moreover, the caspase inhibitor Z-Asp suppressed c-Abl activation and the onset of apoptosis but not the JNK1/SAPK activation. Interestingly, c-Abl tyrosine kinase inhibition by CGP 57148 reduced apoptosis without interfering with JNK1/SAPK activation. These results indicate that c-Abl acts not upstream of JNK1/ SAPK but downstream of caspases during the development of p53-independent apoptosis and is possibly involved in accelerating execution of the cell death pathway. Keywords: c-Abl; JNK1/SAPK; caspase; apoptosis

Introduction Genotoxic stresses elicit several cellular responses that lead to cell death. These include the activation of p53, c-Jun N-terminal kinase 1 (JNK1)/stress-activated protein kinase (SAPK) and c-Abl (Liu et al., 1996). p53 is activated by a variety of genotoxic stresses, such as ionizing radiation and DNA-damaging agents, and the activation of p53 by the DNA damage in certain cells can facilitate apoptosis (Kastan et al., 1991; Kuerbitz et al., 1992; Lee and Bernstein, 1993; Lowe et al., 1993). Many cancer cells, including human

3

Correspondence: T Tsuruo Received 7 July 1998; revised 10 September 1998; accepted 10 September 1998

monocytic leukemia U937 cells, however, do not express wild-type p53 but undergo apoptosis in response to genotoxic stresses, suggesting the p53independent pathway of apoptosis (Gunji et al., 1991; Kataoka et al., 1994). JNK1/SAPK, a subfamily of mitogen-activated protein kinases (MAPKs), is activated by ionizing radiation (IR), ultraviolet radiation (UV), anticancer drugs and in¯ammatory cytokines (Derijard et al., 1994; Kyriakis et al., 1994). Previous studies, including ours, provide some evidences that JNK1/ SAPK mediates genotoxic stress-induced apoptosis at a step upstream of caspase activation in p53-positive and -negative cells (Liu et al., 1996; Seimiya et al., 1997). c-Abl was originally identi®ed as a cellular counterpart of the v-Abl transforming gene of Abelson murine leukemia virus (A-MuLV) (de Klein et al., 1982) and is activated by certain DNA damaging agents, such as IR, UV, cisplatin, 1-b-D-arabinofuranosylcytosine (ara-C) and mitomycin C (Kharbanda et al., 1995). Although the physiological function of cAbl has not been clari®ed, the activation of c-Abl tyrosine kinase has been linked to DNA damageinducible pathways and appeared to interact with p53 (Yuan et al., 1996a). c-Abl has been implicated in apoptosis regulation and in cell cycle arrest induced by genotoxic stressors (Huang et al., 1997; Yuan et al., 1997, 1996b). Previous studies have shown that c-Abl is indispensable to the activation of JNK1/SAPK and p38 MAPK in response to genotoxic stress (Kharbanda et al., 1995; Pandey et al., 1996). As mentioned, JNK1/ SAPK mediates genotoxic stress-induced apoptosis by activating caspases, which play a central role during apoptosis development (Seimiya et al., 1997). Based on these observations, the involvement of c-Abl in genotoxic stress-induced apoptosis should be an upstream event of JNK1/SAPK activation, yet the relationship of c-Abl kinase activity with other apoptosis-regulating molecules, such as JNK1/SAPK and caspases, has not been fully explained. To understand the role of c-Abl in DNA damageinduced apoptosis, we studied the c-Abl and JNK1/ SAPK activation in U937 cells and their apoptosisresistant variant UK711 cells. We demonstrate in this paper that activation of c-Abl depended on caspase activation and was not involved in JNK1/SAPK activation during apoptosis of p53-negative U937 cells. We also describe a possible role of c-Abl in accelerated execution of apoptosis.

Caspase mediates c-Abl activation during apoptosis S Dan et al

1278

Results Activation of JNK1/SAPK and c-Abl tyrosine kinase during apoptosis development Human monocytic leukemia U937 cells undergo morphologic apoptosis to form apoptotic bodies when cells were treated with genotoxic agents such as etoposide (Figure 1a). UK711 is a variant cell line that resists etoposide-induced (Figure 1a) but not TNF-a-or anti-Fas antibody-induced apoptosis (Kataoka et al., 1994). Figure 1b shows apoptosis development as determined by ¯ow-cytometric analysis. Upon treatment with etoposide, apoptosis was seen extensively in U937 cells at 4 and 6 h, whereas apoptosis in UK711 cells was much reduced. Since the initial DNA damage in UK711 cells caused by etoposide treatment is comparable to that in U937 cells (Kataoka et al., 1994), cellular responses that result in apoptosis dier between the two cell lines. Cellular responses to genotoxic stresses include activation of c-Abl and JNK1/SAPK, and p53 expression. U937 cells, however, do not express the p53 protein, under genotoxic stress conditions (data not shown). Therefore, we examined the activation of JNK1/SAPK and c-Abl in U937 and UK711 cells after etoposide treatment. GST-c-Jun (1 ± 92) was used as a substrate for JNK1/SAPK immune complex kinase assay (Figure 2, upper panel). In U937 cells, JNK1/ SAPK was greatly activated at 2 h after etoposide treatment, then the kinase activity gradually declined in a time-dependent manner as the cells showed extensive DNA degradation (lower panel). Conversely, JNK1/ SAPK activation was greatly suppressed in UK711 cells after etoposide treatment.

Figure 1 Dierences in sensitivity to etoposide between U937 and UK711. (a) Phase contrast micrographs of U937 and UK711 treated with 10 mg/ml of etoposide for 4 h. (b) U937 and UK711 (16106 cells/ml) were treated with 10 mg/ml of etoposide for the indicated time. Cells were harvested and then stained with propidium iodide. The distribution of apoptotic cells was analysed using a Becton Dickinson FACScan

We next examined c-Abl activation by an immune complex kinase assay using GST-c-Crk (120 ± 225) as a substrate (Figure 2, middle panel). At 3 h following etoposide treatment, the c-Abl kinase was activated, and the maximum activation (approximately ®vefold of the basal level) was observed at 4 h in U937 cells. On the other hand, at 5 h only slight activation of c-Abl was detected in the apoptosis-resistant UK711 cells treated with etoposide. Apoptosis development accompanies activation of a caspase (ICE/CED3 family protease) cascade, and a family member, caspase-3 (CPP32/YAMA/Apopain), was processed to p17 and p12 fragments in U937 cells 3 h after etoposide treatment (Figure 3b). Consistently, DEVD ± MCA cleavage activity dramatically increased in the cytosol of etoposide-treated U937 cells (Figure 3a). In UK711 cells, however, the processing of caspase-3 was not detected, and the DEVD ± MCA cleavage activity was only minimally activated 4 and 5 h after etoposide treatment. These results indicate that JNK1/SAPK is activated prior to c-Abl and caspase-3 activation in U937 cells during etoposide-induced apoptosis. But, activation of JNK1/SAPK, c-Abl and caspase-3 were suppressed in the apoptosis-resistant UK711 cells treated with etoposide. Suppression of c-Abl activation by caspase inhibitor, Z-Asp To study the role of c-Abl activation in the development of apoptosis, U937 cells were treated with etoposide in the presence of Z-Asp, a caspase

Figure 2 The activation of JNK1/SAPK and c-Abl tyrosine kinases in U937 cells, but not in UK711 cells, during the course of etoposide-induced apoptosis. U937 and UK711 cells were treated with 10 mg/ml of etoposide for the indicated times. (Upper) The in vitro kinase activity of JNK1/SAPK upon treatment with etoposide. The cytoplasmic fraction from etoposide-treated cells was immunoprecipitated with anti-JNK1 antibody. JNK1/SAPK kinase reaction was performed by using a bacterially produced GST-c-Jun fusion protein as a substrate in the presence of [g-32P]ATP and visualized by SDS ± PAGE and autoradiography. (Middle) The in vitro kinase activity of c-Abl tyrosine kinase. The nuclear extract from each sample was immunoprecipitated with anti-c-Abl antibody. Kinase assay was performed using a bacterially produced, puri®ed GST-c-Crk (120 ± 225) fusion protein as a substrate in the presence of [g-32P]ATP. (Lower) Oligonucleosomal DNA fragmentation assay was performed using an aliquot of the cell suspension (about 56105 cells) from each sample


inhibitor. As shown in Figure 4, Z-Asp completely suppressed the etoposide-induced c-Abl activation (lanes 3 and 4, middle panel) and apoptosis development, as determined by oligonucleosomal DNA laddering (lower panel). This result indicates that cAbl activation is a downstream event of the caspase activation. In contrast to c-Abl activation, JNK1/ SAPK activation by etoposide was not inhibited by ZAsp co-treatment (lanes 3 and 4, upper panel). Similar results were obtained when U937 cells were treated with the DNA-damaging agent camptothecin (lanes 9 and 10). TNF-a (lane 5) also induced apoptosis in U937 cells (lower panel), activated c-Abl (middle panel), with slight activation of JNK1/SAPK (upper panel). Again, Z-Asp completely suppressed the development of apoptosis and c-Abl activation, whereas it did not inhibit JNK1/SAPK activation (lane 6). We further examined the eect of Z-Asp on c-Abl activation in UK711 cells (Figure 5). As we reported previously, UK711 cells showed resistance to etoposide- and camptothecin-induced apoptosis and showed suppressed activation of c-Abl and JNK1/SAPK. UK711 cells, however, were sensitive to TNF-a-

induced apoptosis and showed activation of c-Abl and JNK1/SAPK after treatment. As in U937 cells, ZAsp completely inhibited the TNF-a-induced apoptosis and c-Abl activation in UK711 cells but not the JNK1/ SAPK activation. These results indicate that c-Abl was

Figure 4 The activation of c-Abl and JNK1/SAPK upon treatment with various apoptosis-inducing agents and the eect of Z-Asp treatment on these cellular responses. U937 cells were treated with 10 mg/ml of etoposide (lanes 3 and 4), 0.25 ng/ml of recombinant human tumor necrosis factor-a (rhTNF-a) plus 1 mg/ ml of cycloheximide (CHX) (lanes 5 and 6), cycloheximide alone (lanes 7 and 8) and 10 mg/ml of camptothecin (CPT) (lanes 9 and 10) in the presence (+) or absence (7) of 100 mg/ml of Z-Asp for 3 h. The kinase assays of JNK1/SAPK (upper) and c-Abl (middle), and DNA fragmentation assay (lower) were carried out as described in Figure 2. The numbers at the bottom of panels are the percentages of subdiploid cells assesed by ¯ow cytometry as in Figure 1

Figure 3 Activation of caspase-3 during etoposide-induced apoptosis. (a) Cleavage activity of DEVD ± MCA, a preferential substrate for caspase-3, was augmented by etoposide treatment in U937 cells but was almost suppressed in UK711 cells. (b) Immunoblot analysis of caspase-3 by using cytoplasmic fraction from each sample. The proform of caspase-3 was cleaved to generate the active form (p17) 3 h or more after treatment with etoposide in U937 cells, but not in UK711 cells (upper). Detection of another processed caspase-3 fragment (p12) upon treatment with etoposide in U937 cells, but not in UK711 cells (lower). The cytoplasmic fraction of each sample was immunoblotted with the antibody speci®c to the amino terminus of cleaved p12 fragment (SGVDD)

Figure 5 The activation of c-Abl in UK711 cells treated with TNF-a. UK711 cells were treated with 10 mg/ml of etoposide (lanes 3 and 4), 0.25 ng/ml of rhTNF-a plus 1 mg/ml of cycloheximide (CHX) (lanes 5 and 6), cycloheximide alone (lanes 7 and 8) and 10 mg/ml of camptothecin (CPT) (lanes 9 and 10) in the presence (+) or absence (7) of 100 mg/ml of ZAsp for 3 h. The experiments are described in Figure 4

1279


1280

activated after caspase activation, irrespective of apoptosis-inducing stimuli in U937 and UK711 cells. Suppression of apoptosis by CGP 57148, an inhibitor of c-Abl tyrosine kinase To study the role of c-Abl activation in the development of apoptosis further, we examined the eect of a selective inhibitor of c-Abl tyrosine kinase, CGP 57148, on etoposide- and TNF-a-induced apoptosis (Figure 6a). In the presence of CGP 57148, emergence of apoptotic cells was signi®cantly suppressed upon treatment with etoposide and TNF-a. CGP 57148 itself neither accumulated subdiploid cells

nor altered the ratio in cell cycle phase of U937 cells (data not shown). In order to con®rm whether CGP 57148 selectively inhibits c-Abl tyrosine kinase activity at the cellular level, we examined the eect of CGP 57148 on the autophosphorylation of c-Abl in the cells (Figure 6b). Treatment of these cells with etoposide and TNF-a resulted in the emergence of a tyrosine-phosphorylated form of c-Abl, which indicated that c-Abl tyrosine kinase was surely activated by treatment with etoposide or TNF-a in vivo. In the presence of CGP 57148, the tyrosine-phosphorylated c-Abl completely disappeared. Thus, CGP 57148 eectively inhibited c-Abl tyrosine autophosphorylation at the cellular level. Immunoblot analysis of nuclear proteins from each sample using anti-phosphotyrosine antibody revealed that CGP 57148 did not alter the pattern of tyrosinephosphorylated protein bands (data not shown), indicating that tyrosine kinase inhibition by CGP 57148 is highly selective to c-Abl. Further, CGP 57148 did not inhibit the DEVD ± MCA cleavage activity of caspase-3 in vitro (data not shown), indicating that apoptosis inhibition by CGP 57148 was not due to its nonspeci®c inhibitory eect on the activated caspase-3, which is believed to be crucial to promote apoptosis. These results suggest that activation of c-Abl tyrosine kinase is involved in accelerated execution of the cellular apoptotic program. Activation of JNK1/SAPK in response to DNA damage is c-Abl independent Previous studies have reported that JNK1/SAPK activation in response to genotoxic stress depends on c-Abl in certain cells, for example NIH3T3 (Kharbanda et al., 1995). In Figure 2, however, JNK1/SAPK activation preceded c-Abl activation, suggesting that the former does not require c-Abl kinase activity in U937 cells. To further con®rm whether JNK1/SAPK activation in response to etoposide treatment was independent of c-Abl kinase activity in U937 cells, we examined the eect of CGP 57148 on JNK1/SAPK activation in the cells by using an antibody speci®c to the active form of JNK1/SAPK (dually phosphorylated at Thr183/Tyr185) (Figure 7). The active form of JNK1/SAPK was observed upon treatment with etoposide, in accordance with the augmentation of in

Figure 6 Suppression of apoptosis development and the autophosphorylation of c-Abl by a speci®c c-Abl tyrosine kinase inhibitor, CGP 57148, in etoposide- and TNF-a-treated U937 cells. (a) Cells were pretreated with or without 10 mM CGP 57148 (as indicated in each lane) for 12 h, then treated with 10 mg/ml of etoposide for 4 h or 0.25 ng/ml of rhTNF-a in the presence of 1 mg/ml of cycloheximide for 3 h. Cells were ®xed and stained with propidium iodide as described in Materials and methods. The emergence of subdiploid cells was measured by ¯ow cytometry, as in Figure 1. (b) Cells were pretreated with or without CGP 57148 as in (a), then treated with 10 mg/ml of etoposide or 0.25 ng/ml of rhTNF-a in the presence of 1 mg/ml of cycloheximide for 3 h. Nuclear proteins extracted from each sample were immunoprecipitated with anti-c-Abl antibody, and the immunoprecipitates were immunoblotted with anti-c-Abl (upper) or anti-phosphotyrosine (lower) antibodies

Figure 7 The amount of active JNK1/SAPK was not aected by c-Abl inhibition. Cells were pretreated with or without CGP 57148, then treated with 10 mg/ml of etoposide for 3 h. The cytosolic fraction from each sample was immunoblotted with anti-JNK1 (upper) or anti-phosphospeci®c SAPK/JNK (Thr183/ Tyr185) (lower) antibodies


vitro kinase activity (Figure 2). However, inhibition of c-Abl kinase by CGP 57148 did not aect activation of JNK1/SAPK in response to etoposide treatment whereas CGP 57148 completely inhibited c-Abl autophosphorylation (Figure 6b, lane 4). Similar result was obtained by taking the time-course experiments of JNK/SAPK immune complex kinase assay (data not shown). This result indicates that upon treatment with etoposide JNK1/SAPK activation is independent of the kinase activity of c-Abl in U937 cells. Discussion We demonstrate in this paper that c-Abl was activated during the development of U937 apoptosis that was elicited by DNA-damaging anticancer drugs and TNFa. Moreover, we found that inhibition of caspases by Z-Asp resulted in simultaneous suppression of c-Abl activation and apoptosis development. In anticancer drug-resistant UK711 cells, c-Abl activation was not observed after treatment with anticancer drugs, while c-Abl was activated when cells underwent apoptosis in the presence of TNF-a. Thus, the caspase-dependent activation of c-Abl kinase correlated well with the development of apoptosis. The activated c-Abl could play a role in amplification of U937 apoptosis, since CGP 57148, a selective inhibitor of Abl tyrosine kinases (Buchdunger et al., 1996), signi®cantly suppressed the etoposide- and TNF-a-induced apoptosis. The result that CGP 57148 completely inhibited the autophosphorylation of c-Abl indicates that it certainly inhibited the c-Abl tyrosine kinase activity in U937 cells. The inhibition of tyrosine kinase by CGP 57148 was highly selective to c-Abl, strongly suggesting that the suppression of U937 apoptosis by CGP 57148 was due to c-Abl tyrosine kinase inhibition. Moreover, previous study have shown that transient expression of c-Abl induced apoptosis in p53 de®cient cells (Yuan et al., 1997), indicating its positive regulatory role in apoptosis development. Nevertheless, involvement of unknown molecules in the suppression of apoptosis by CGP 57148 cannot be ruled out. At least, CGP 57148 did not directly inhibit the activity of caspase-3 like proteases (DEVDase) in vitro (data not shown). On the other hand, the activation of DEVDase in vivo was slightly inhibited upon treatment with CGP 57148 (15 ± 20% inhibition, data not shown), so that the activated c-Abl could in turn enhance the caspase activity to promote the cellular apoptotic program. However, the suppression of apoptosis development by CGP 57148 (50 ± 60% inhibition, Figure 6a) could not fully be explained by the slight inhibition of caspase activity. Previous studies reported that c-Abl activation regulates JNKs in certain cells that have been treated with DNA-damaging agents (Kharbanda et al., 1995). As mentioned, JNK1/SAPK triggers apoptosis in response to DNA-damaging agents by activating ZAsp-sensitive caspases (Seimiya et al., 1997). Therefore, we anticipated that c-Abl activation is an earlier event than JNK1/SAPK activation during etoposide-induced apoptosis of U937 cells. However, so far as we have examined U937 cells, JNK1/SAPK was always activated prior to c-Abl. Furthermore, the c-Abl

inhibitor CGP 57148 did not interfere with JNK1/ SAPK activation while CGP 57148 completely diminished c-Abl autophosphorylation. Thus, the activation of JNK1/SAPK in response to etoposide treatment did not require c-Abl kinase activity in U937 cells. c-Ablindependent activation of JNK1/SAPK was also observed during apoptosis in another human leukemia cell line, HL-60 (data not shown). The dierence between previous reports and our present study could be due to the defect of p53 in U937 and HL-60 cells. p53 is induced by DNA-damaging agents and interacts with c-Abl, resulting in down regulation of CDK2 and in cell cycle arrest in the G1 phase (Yuan et al., 1996a, b). Thus, p53 and c-Abl appeared to be linked to generate cellular responses to DNA-damaging agents. Since many cancer cells do not express p53 or they express mutated p53, apoptosis signaling without wildtype p53 is an important clue to be studied in cancer chemotherapy. We have described a novel mode of c-Abl activation that requires caspase activation during apoptosis development. One possible explanation for this c-Abl activation follows: during the development of apoptosis, caspase-3 is processed to the active form (Figure 3), which in turn cleaves and inactivates the inhibitor of caspase-dependent deoxyribonuclease (ICAD) so that the caspase-dependent deoxyribonuclease (CAD) catalyze apoptotic chromatin degradation (Enari et al., 1998; Sakahira et al., 1998). Since DNA damage activates c-Abl tyrosine kinase (Kharbanda et al., 1995), apoptotic chromatin degradation could cause the activation of c-Abl located in the nucleus. DNAdependent protein kinase (DNA-PK) and ATM gene product, which can directly interact with and activate c-Abl kinase in response to DNA damage (Baskaran et al., 1997; Kharbanda et al., 1997; Shafman et al., 1997), could be involved in this activation of c-Abl. ZAsp inhibits the caspase-3 activity (Mashima et al., 1995, 1997); suppression of olionucleosomal DNA fragmentation and c-Abl activation by Z-Asp could be due to inhibition of the ICAD cleavage. Consistent with this hypothesis, the apoptosis-resistant U937 variant, UK711, which exhibited comparable amounts of DNA-topoisomerase II cleavable complex (Kataoka et al., 1994) but hardly showed caspase activation or DNA degradation in response to etoposide treatment (Figures 1 and 2), exhibited no signi®cant activation of c-Abl (Figure 2). The molecular target(s) for c-Abl tyrosine kinase during apoptosis is still unclear. The previously known substrates for c-Abl are the c-Crk adaptor protein (Ren et al., 1994), RNA polymerase II (Baskaran et al., 1993), DNA-PK (Han et al., 1996; McConnell et al., 1997; Song et al., 1996; Teraoka et al., 1996) and SHPTP1 (Kharbanda et al., 1996), as well as c-Abl itself. The involvement of these substrates of c-Abl in the apoptosis development is under investigation. In summary, we demonstrated that c-Abl is activated after caspase activation during the development of U937 apoptosis and that c-Abl activation could play a role in the ampli®cation of cellular apoptosis program. At present, we do not know how cAbl ampli®es the cellular apoptosis program after caspase activation. Further studies are needed to clarify the involvement of c-Abl tyrosine kinase in the development of apoptosis.

1281


1282

Materials and methods Cell lines, reagents and antibodies U937 and its variant cell line, UK711, are maintained in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum. Etoposide was kindly provided from Bristol-Myers Squibb Co., Ltd (Tokyo, Japan). ZAsp-CH2-DCB was obtained from Funakoshi (Tokyo, Japan). A selective inhibitor of c-Abl tyrosine kinase, CGP 57148, is a generous gift from Novartis Pharma, Inc. (Basel, Switzerland). The antibodies against c-Abl (K-12 and 24-11) and JNK1 (C-17) were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). AntiCPP32 (caspase-3) and anti-phosphotyrosine antibodies were purchased from Transduction Laboratories (Lexington, KY, USA). Phospho-speci®c SAPK/JNK (Thr183/ Tyr185) antibody was obtained from New England Biolabs, Inc. (Beverley, MA, USA). The antibody speci®c to the cleaved form, including the carboxy terminus of caspase-3 (p12), was raised by immunizing a rabbit with a peptide (SGVDD) corresponding to the amino terminus of the cleaved p12 fragment. Flow-cytometric analysis of apoptotic cells Cells were harvested, washed with phosphate-buered saline (PBS) and resuspended in 80% ethanol. The ®xed cells were washed with PBS and resuspended in 1 mg/ml of RNase A in PBS, followed by incubation at 378C for 30 min. Cells were stained in PI solution (50 mg/ml of propidium iodide, 0.1% sodium citrate, 0.1% Nonidet P40) and analysed using a Becton Dickinson FACScan ¯owcytometer with Cell Quest software (Braintree, MA, USA). Production of GST-c-Crk and GST-c-Jun fusion proteins Expression plasmid coding glutathione-S-transferase (GST)-fused c-Crk (pGEX-c-Crk) was constructed by inserting c-Crk (120 ± 225) coding sequence into a pGEX vector (Pharmacia Biotech, Uppsala, Sweden). The cDNA coding c-Crk is a generous gift from Dr Hisamaru Hirai, University of Tokyo. pGEX-c-Jun has been described (Seimiya et al., 1997). GST-fusion protein was produced as described previously, with slight modi®cations (Seimiya et al., 1997). Brie¯y, Escherichia coli carrying a pGEX plasmid was cultured in LB media containing 50 mg/ml of ampicillin to an OD600=0.8 and then isopropyl-b-Dthiogalactopyranoside (IPTG) was added to a ®nal concentration of 0.5 mM. Cells were collected by centrifugation at 50006g and resuspended in the appropriate volume of PBS. Cells were lysed by sonication and centrifuged at 15 0006g for 10 min. The GST-fusion protein in the resulting supernatant was puri®ed by absorbing onto glutathione-conjugated sepharose beads (Pharmacia Biotech) and eluted with elution buer containing 50 mM HEPES ± NaOH and 2 mM glutathione. Puri®ed protein in elution buer was dialysed against 50 mM HEPES ± NaOH containing 0.1 mM phenylmethylsulfonyl ¯uoride (PMSF). Preparation of nuclear extract and cytosolic fraction Cells were washed twice with PBS and resuspended in a hypotonic buer containing 20 mM HEPES ± KOH (pH 7.9), 5 mM KCl, 0.5 mM MgCl2, 0.5 mM dithiothreitol, 10 mM b-glycerophosphate, 1 mM sodium vanadate, 1 mM PMSF, 2 mg/ml of pepstatin A and 2 mg/ml of aprotinin. Cells were homogenized using a glass dounce homogenizer for 20 strokes, and the homogenate was centrifuged at 55006g for 1 min. The pellet was resuspended in a high salt buer containing 0.5 M NaCl, 20 mM HEPES ± NaOH (pH 7.9), 1.5 mM MgCl2,

0.2 mM EDTA, 0.2 mM EGTA, 0.5 mM dithiothreitol, 10 mM b-glycerophosphate, 1 mM sodium vanadate, 25% v/v glycerol, 1 mM PMSF, 2 mg/ml of pepstatin A, and 2 mg/ml of aprotinin. Samples were centrifuged at 100 0006g for 20 min, and the supernatant was collected as nuclear extract. Protein assay was carried out using the Bio-Rad protein assay kit according to the instruction manual. In vitro kinase assay Nuclear extract corresponding to 100 mg of protein was mixed with anti-Ab1 antibody (C-12) in immunoprecipitation buer containing 50 mM Tris-HCl, 150 mM NaCl, 0.1% Nonidet P-40, 1 mM EDTA, 1 mM EGTA, 10 mM bglycerphosphate, 1 mM sodium vanadate, 1 mM dithiothreitol, 1 mM PMSF, 1 mg/ml of pepstatin A, 1 mg/ml of aprotinin. Protein A-sepharose beads were then added, incubated for an additional 1 h and collected by centrifugation. Pelletized beads were washed twice with immunoprecipitation buer and once with kinase buer containing 20 mM HEPES ± KOH and ®nally suspended with reaction mixture containing 20 mCi of [g-32P]ATP and 1 mg of GST-c-Crk as a substrate. Samples were incubated for 30 min, SDS sample buer (10% v/v glycerol, 5% 2-mercaptoethanol, 2% sodium dodecyl sulfate (SDS), 62.5 mM Tris-HCl pH 6.8) was added, and they were boiled for 2 min. Samples were electrophoresed in SDSpolyacrylamide gel and subjected to autoradiography. For JNK1/SAPK kinase assays, a cytosolic fraction was used and mixed with anti-JNK (C-17) antibody. GST-c-Jun was used as a speci®c substrate for JNK1/SAPK. Western blotting Nuclear extracts (for c-Abl) or cytoplasmic fractions (for JNK1/SAPK and caspase-3) were subjected to SDS ± PAGE and transferred to nitrocellulose membrane (Schneider & Schuell, Dasse, Germany). Transferred membranes were probed with primary antibody. Detection was accomplished using an appropriate anti-mouse or rabbit immunoglobulin secondary antibody and the enhanced chemiluminescence detection system (Amersham Japan, Ltd., Tokyo, Japan). Detection of tyrosine-phosphorylated c-Abl Nuclear proteins were immunoprecipitated by using anti-cAbl (24 ± 11) antibodies in the same way as for in vitro kinase assays. The immunoprecipitates were subjected to SDS ± PAGE and immunoblotted with anti-phosphotyrosine or anti-c-Abl (24 ± 11) antibodies. DEVD ± MCA cleavage assay (AMC assay) Cells were harvested, and cell extracts were prepared as described. Brie¯y, 10 mg of the cytosolic fraction from each sample was mixed in 50 ml of ICE buer containing 20 mM HEPES ± NaOH (pH 7.5), 2 mM DTT, 10% v/v glycerol and 1 mM DEVD ± MCA (Peptide Institute, Osaka, Japan) as a substrate. The reaction mixture was incubated at 378C for 30 min, and increase in the AMC ¯uorescence was measured by Hitachi F-2000 ¯uorometer (lex=380 nm, lem=460 nm). DNA fragmentation assay Cells were harvested and treated with proteinase K and RNase A, as previously described. The resulting DNA was electrophoresed in an agarose gel and visualized by staining with ethidium bromide.


Acknowledgements We are grateful to Dr Hisamaru Hirai, the 3rd Department of Internal Medicine, Faculty of Medicine, University of Tokyo, for providing us the cDNA coding c-Crk and Novartis Pharma, Inc. for providing us CGP 57148. This

work was supported by a special grant for Advanced Research on Cancer from the Ministry of Education, Science and Culture, Japan; the Organization for Pharmaceutical Safety and Research (OPSR), Japan; and the Vehicle Racing Commemorative Foundation, Japan.

References Baskaran R, Dahmus ME and Wang JY. (1993). Proc. Natl. Acad. Sci. USA, 90, 11167 ± 11171. Baskaran R, Wood LD, Whitaker LL, Canman CE, Morgan SE, Xu Y, Barlow C, Baltimore D, Wynshaw-Boris A, Kastan MB and Wang JY. (1997). Nature, 387, 516 ± 519. Buchdunger E, Zimmermann J, Mett H, Meyer T, Muller M, Druker BJ and Lydon NB. (1996). Cancer Res., 56, 100 ± 104. de Klein A, van Kessel AG, Grosveld G, Bartram CR, Hagemeijer A, Bootsma D, Spurr NK, Heisterkamp N, Groen J and Stephenson JR. (1982). Nature, 300, 765 ± 767. Derijard B, Hibi M, Wu IH, Barrett T, Su B, Deng T, Karin M and Davis RJ. (1994). Cell, 76, 1025 ± 1037. Enari M, Sakahira H, Yokoyama H, Okawa K, Iwamatsu A and Nagata S. (1998). Nature, 391, 43 ± 50. Gunji H, Kharbanda S and Kufe D. (1991). Cancer Res., 51, 741 ± 743. Han Z, Malik N, Carter T, Reeves WH, Wyche JH and Hendrickson EA. (1996). J. Biol. Chem., 271, 25035 ± 25040. Huang Y, Yuan ZM, Ishiko T, Nakada S, Utsugisawa T, Kato T, Kharbanda S and Kufe DW. (1997). Oncogene, 15, 1947 ± 1952. Kastan MB, Onyekwere O, Sidransky D, Vogelstein B and Craig RW. (1991). Cancer Res., 51, 6304 ± 6311. Kataoka S, Naito M, Tomida A and Tsuruo T. (1994). Exp. Cell Res., 215, 199 ± 205. Kharbanda S, Bharti A, Pei D, Wang J, Pandey P, Ren R, Weichselbaum R, Walsh CT and Kufe D. (1996). Proc. Natl. Acad. Sci. USA, 93, 6898 ± 6901. Kharbanda S, Pandey P, Jin S, Inoue S, Bharti A, Yuan ZM, Weichselbaum R, Weaver D and Kufe D. (1997). Nature, 386, 732 ± 735. Kharbanda S, Ren R, Pandey P, Shafman TD, Feller SM, Weichselbaum RR and Kufe DW. (1995). Nature, 376, 785 ± 788. Kuerbitz SJ, Plunkett BS, Walsh WV and Kastan MB. (1992). Proc. Natl. Acad. Sci. USA, 89, 7491 ± 7495. Kyriakis JM, Banerjee P, Nikolakaki E, Dai T, Rubie EA, Ahmad MF, Avruch J and Woodgett JR. (1994). Nature, 369, 156 ± 160.

Lee JM and Bernstein A. (1993). Proc. Natl. Acad. Sci. USA, 90, 5742 ± 5746. Liu ZG, Baskaran R, Lea-Chou ET, Wood LD, Chen Y, Karin M and Wang JY. (1996). Nature, 384, 273 ± 276. Lowe SW, Schmitt EM, Smith SW, Osborne BA and Jacks T. (1993). Nature, 362, 847 ± 849. Mashima T, Naito M, Fujita N, Noguchi K and Tsuruo T. (1995). Biochem. Biophys. Res. Commun., 217, 1185 ± 1192. Mashima T, Naito M, Noguchi K, Miller DK, Nicholson DW and Tsuruo T. (1997). Oncogene, 14, 1007 ± 1012. McConnell KR, Dynan WS and Hardin JA. (1997). J. Immunol., 158, 2083 ± 2089. Pandey P, Raingeaud J, Kaneki M, Weichselbaum R, Davis RJ, Kufe D and Kharbanda S. (1996). J. Biol. Chem., 271, 23775 ± 23779. Ren R, Ye ZS and Baltimore D. (1994). Genes Dev., 8, 783 ± 795. Sakahira H, Enari M and Nagata S. (1998). Nature, 391, 96 ± 99. Seimiya H, Mashima T, Toho M and Tsuruo T. (1997). J. Biol. Chem., 272, 4631 ± 4636. Shafman T, Khanna KK, Kedar P, Spring K, Kozlov S, Yen T, Hobson K, Gatei M, Zhang N, Watters D, Egerton M, Shiloh Y, Kharbanda S, Kufe D and Lavin MF. (1997). Nature, 387, 520 ± 523. Song Q, Lees-Miller SP, Kumar S, Zhang Z, Chan DW, Smith GC, Jackson SP, Alnemri ES, Litwack G, Khanna KK and Lavin MF. (1996). EMBO J., 15, 3238 ± 3246. Teraoka H, Yumoto Y, Watanabe F, Tsukada K, Suwa A, Enari M and Nagata S. (1996). FEBS Lett., 393, 1 ± 6. Yuan ZM, Huang Y, Ishiko T, Kharbanda S, Weichselbaum R and Kufe D. (1997). Proc. Natl. Acad. Sci. USA, 94, 1437 ± 1440. Yuan ZM, Huang Y, Fan MM, Sawyers C, Kharbanda S and Kufe D. (1996a). J. Biol. Chem., 271, 26457 ± 26460. Yuan ZM, Huang Y, Whang Y, Sawyers C, Weichselbaum R, Kharbanda S and Kufe D. (1996b). Nature, 382, 272 ± 274.

1283