Activation of the Fas pathway independently of Fas ligand ... - Nature

ã

Oncogene (2001) 20, 1852 ± 1859 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc

Activation of the Fas pathway independently of Fas ligand during apoptosis induced by camptothecin in p53 mutant human colon carcinoma cells Rong-Guang Shao1, Chun-Xia Cao1, Wilberto Nieves-Neira1, Marie-TheÂreÁse Dimanche-Boitrel2, Eric Solary2 and Yves Pommier*,1 1

Laboratory of Molecular Pharmacology, Division of Basic Science, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, MD 20892, USA; 2Department of Biology and Therapy of Cancer, INSERM U517, Faculty of Medicine and Pharmacy, Dijon, France

The present study explored the role of the cell surface receptor Fas (CD95/APO-1) in apoptosis induced by camptothecin (CPT) in the HT29 colon carcinoma cell line. CPT-induced apoptosis was associated with high molecular weight DNA fragmentation as measured by ®lter elution. This fragmentation was inhibited by the caspase inhibitor, z-VAD-fmk and by cycloheximide, which also prevented proteolytic activation of caspase-3 and poly(ADP-ribose)polymerase cleavage. Under such conditions, Fas, Fas ligand, Bax, and p21 expression were increased and Fas recruited the FADD adaptor. Fas expression increase was blocked by cycloheximide but not by z-VAD-fmk, consistent with caspase activation downstream from Fas. Treatment of HT29 cells with FasL or with the CH-11 agonistic anti-Fas antibody potentiated the apoptotic response of cells treated with CPT. The anti-Fas blocking antibody ZB4 and the Fasligand inhibitor failed to protect HT29 cells from CPTinduced apoptosis. Such a protection was obtained by transient expression of constructs encoding a dominantnegative mutant of FADD, FADD in an antisense orientation and E8 or MC159 viral proteins that inhibit Fas-induced apoptosis at the level of FADD and procaspase-8, respectively. Together, these data show that topoisomerase I-mediated DNA damage-induced apoptosis involves activation of the Fas pathway without detectable Fas-ligand requirement in CPT-treated cells. Oncogene (2001) 20, 1852 ± 1859. Keywords: Bax; caspase; FADD; z-VAD; nuclease; programmed cell death Introduction Coupling of the cell surface receptor Fas (CD95/APO1) with its natural ligand, Fas-ligand (FasL) triggers apoptosis very eciently in various cell types (Nagata

*Correspondence: Y Pommier, Laboratory of Molecular Pharmacology, Bldg. 37, Rm. 4E28, NIH, Bethesda, MD 208924255, USA Received 19 May 2000; revised 18 October 2000; accepted 15 January 2001

and Golstein, 1995). This death pathway can also be triggered upon Fas cross-linking with the cytolytic monoclonal antibody CH-11 (Krammer et al., 1994; Nagata and Golstein, 1995; Tanaka et al., 1996). Binding of Fas ligand or cross-linking by CH-11 promotes Fas oligomerization in the plasma membrane. Fas trimer then selectively binds to an adaptor molecule, FADD that recruits and activates caspase-8 (FLICE), which is an upstream caspase that in turn activates intracellular eector caspases such as caspase3 (Apopain/YAMA). The critical role of caspases in Fas-mediated signaling is exempli®ed by the inhibition of Fas-mediated apoptosis by synthetic peptide compounds like z-VAD-fmk, a broad speci®city caspase inhibitor, or YVAD-cmk, a more speci®c caspase-1 and -8 inhibitor (Kuida et al., 1995; Muzio et al., 1996). Fas is often down-modulated in colon carcinoma (Moller et al., 1994). Stimulation of colon carcinoma cells by interferon-g increases Fas expression and sensitizes cells to Fas-induced cell death (Gunthert et al., 1996). Following a fourfold increase of Fas expression in HT29 cells treated with interferon-g, a synergistic eect on Fas-mediated apoptosis is obtained when CH-11 and interferon are combined (Tillman et al., 1998). We have reported (Micheau et al., 1997) that some chemotherapeutic drugs also increase Fas receptor expression at the surface of colon carcinoma cell lines and sensitize cisplatin-treated cells to agonistic anti-Fas antibodies, to soluble Fas ligand, and to allogenic peripheral blood leukocyte-mediated cytotoxicity. The expression and role of Fas-ligand (FasL) in colon carcinoma cells is more controversial. Expression of FasL in tumor cells has been proposed to kill Fas-expressing lymphocyte and NK cells (O'Connell et al., 1996), which might account for the immune tolerance observed in colon cancer. Actually, we have recently challenged this `counterattack hypothesis' by showing that FasL mRNA and protein can be detected in colon cancer cell extracts but the protein is not expressed at the surface of colon carcinoma cells when studied by ¯ow cytometry and confocal microscopy (Favre-Felix et al., 2000). Another controversial issue is the role of the Fas pathway in cytotoxic drug-induced apoptosis. Some

Fas activation in camptothecin-induced apoptosis R-G Shao et al

reports have proposed that, in addition to upregulating Fas expression at the surface of tumor cells, chemotherapeutic agents could also up-regulate FasL expression. These observations suggested the interesting possibility that DNA damage resulting from chemotherapy activates the Fas pathway by inducing a FasL/Fas interaction (Friesen et al., 1996; Houghton et al., 1997). However, antagonistic antibodies that block FasL-mediated apoptosis do not always interfere with drug-induced cell death (Eischen et al., 1997; Villunger et al., 1997; Micheau et al., 1999). We have recently shown in various tumor cell lines that cisplatin and topoisomerase II inhibitors could activate the Fas pathways in a FasL-independent manner (Micheau et al., 1999). The present study extends this latter observation to camptothecin (CPT)-induced apoptosis. CPT derivatives have recently been approved for the treatment of colon and ovarian carcinomas. Camptothecins are also commonly used in experimental model systems to induce and study the apoptosis pathways (Nicholson et al., 1995; Shimizu et al., 1995). Camptothecins are selective inhibitors of mammalian topoisomerase I. They trap topoisomerase I catalytic intermediates complexed with DNA, which are commonly referred to as the topoisomerase I cleavage complexes (Chen and Liu, 1994; Pommier et al., 1998). Hence, camptothecins convert topoisomerase I from an essential enzyme into a DNA damaging protein when replication complexes collide with the drug-stabilized cleavage complexes (Shao et al., 1999; Strumberg et al., 2000). To investigate the role of the Fas-pathway in the mechanisms of cell death induced by CPT, we used the human colon carcinoma HT29 cell line that is p53mutated (Rodrigues et al., 1990) and is routinely used in the NCI Anticancer screen (O'Connor et al., 1997). Our data demonstrate that replication-mediated DNA damage induced by camptothecin treatment enhances both Fas and FasL expression in HT29 cells. Intracellular recruitment of FADD and Fas, rather than FasL coupling appears important for camptothecin-induced apoptosis in HT29 cells.

Results Delayed apoptosis induced by CPT in HT29 cells To determine the kinetics of CPT-induced apoptotic cell death, human colon carcinoma HT29 cells were treated with 1 mM CPT for 3 h and DNA fragmentation was measured by ®lter elution assay at various times after drug removal. Figure 1a shows that DNA fragmentation was detectable after a lag period of 2 days after CPT removal. Conventional agarose gel electrophoresis did not show detectable DNA ladder up to 4 days after CPT removal (Figure 1b). Pulse-®eld agarose gel electrophoresis demonstrated 25 kb DNA fragments 3 ± 4 days after treatment (Figure 1c). These results indicate that CPT treatment produced high molecular weight

1853

Figure 1 High molecular weight DNA fragmentation induced by CPT in HT29 cells. Cells were treated with 1 mM CPT for 3 h. CPT was then removed and cells were incubated in fresh medium for the indicated times. (a) DNA fragmentation was measured by ®lter elution assay. Error bars represent standard error for three independent experiments. Conventional (b) and pulse-®eld (c) agarose gel electrophoresis. DNA fragmentation was visualized after ethidium bromide staining. Representative experiments are shown

(HMW) DNA fragmentation without internucleosomal DNA cleavage in HT29 cells and that the ®lter elution assay (Bertrand et al., 1994, 1995) can be used to detect HMW DNA fragmentation. CPT-induced apoptosis was further examined morphologically. One day after CPT treatment, the nucleus of treated cells was enlarged as cells were arrested in S phase (Figure 2). Three days after CPT treatment, most cells displayed the characteristic morphological features of apoptosis, including chromatin condensation and nuclear fragmentation (Figure 2a). These data indicate that HMW DNA fragmentation corresponds to apoptotic cell death in human colon carcinoma HT29 cells and that apoptosis occurs after an S phase arrest (Goldwasser et al., 1996). Inhibition of CPT-induced DNA fragmentation by cycloheximide and z-VAD-fmk The observation of delayed apoptosis induced by CPT raised the possibility that protein synthesis was required for apoptosis following topoisomerase Imediated DNA damage. Figure 3 shows that the protein synthesis inhibitor cycloheximide inhibited CPT-induced DNA fragmentation, indicating that macromolecular synthesis was required for CPTinduced apoptosis in HT29 cells. Figure 3 also shows that z-VAD-fmk prevented DNA fragmentation induced by CPT demonstrating that caspase activation was involved in CPT-induced apoptosis. Inhibition by z-VAD-fmk or cycloheximide was still observed when z-VAD-fmk or cycloheximide were added 1 day after the CPT treatment. These data indicate that macromolecular synthesis and caspase activation are required to elicit the DNA fragmentation in CPT-treated HT29 cells. Oncogene


1854

Figure 2 Delayed apoptosis induced by CPT in HT29 cells. CPT treatment was with 1 mM for 3 h. (a) Morphological changes. Cells were stained with 4', 6-diamidino-2-phenylindole (DAPI). (b) Cell cycle distribution was determined by ¯ow cytometry after propidium iodide staining: untreated cells: control (left); 1 day after treatment (middle); 3 days after treatment (right)

generates DNA double-strand breaks at the topoisomerase I cleavage complex sites (Strumberg et al., 2000). In order to distinguish which of these steps was required for the induction of apoptosis in HT29 cells, we compared CPT-induced DNA fragmentation in the absence or presence of the DNA polymerase inhibitor, aphidicolin (Strumberg et al., 2000). Figure 3 shows that aphidicolin pretreatment prevented DNA fragmentation induced by CPT. These results indicate that topoisomerase I cleavage complexes are not sucient for induction of apoptosis by CPT and that replication-mediated DNA double-strand breaks initiate the apoptotic response in HT29 cells.

Figure 3 Cycloheximide (CHX), the caspase inhibitor (z-VADfmk) and the DNA polymerase inhibitor, aphidicolin (APH) block CPT-induced DNA fragmentation. HT29 cells were treated with 1 mM CPT for 3 h, after which CPT was removed and fresh medium was added in the presence or absence of 3 mM CHX or 50 mM z-VAD-fmk for 3 days. Aphidicolin (1 mM) was added 15 min before CPT and was removed from cell cultures at the same time as CPT (after 3 h). DNA fragmentation was measured by the ®lter elution assay. Values shown are the mean and standard error of at least three independent experiments

The cytotoxicity of CPT in HT29 cells requires at least two consecutive steps (O'Connor et al., 1991). First, the formation of topoisomerase I cleavage complexes and secondly, ongoing replication that Oncogene

Activation of caspase-3 in response to CPT treatment is inhibited by cycloheximide and z-VAD-fmk We next examined the proteolytic activation of caspase 3 and caspase-mediated cleavage of PARP following CPT treatment by Western blotting. Figure 5 shows that both CHX and z-VAD-fmk prevented caspase-3 activation and PARP cleavage. These results indicate that caspase-3 activation requires protein synthesis in CPT-treated HT29 cells. Induction of Fas and Fas-ligand (FasL) following CPT treatment in HT29 cells We next tested whether Fas and FasL changed after CPT treatment, Figure 5 shows that, when studied by Western blot, both Fas and FasL protein levels


increased 1 ± 2 days after treatment (Figure 5a) and that cycloheximide prevented Fas expression (Figure 5b). z-VAD-fmk only partially aected Fas expression Figure 5b). These results demonstrate that treatment of HT29 cells with CPT induces both Fas and FasL and that caspase activation is downstream of Fas. Figure 6 shows that other DNA damage-inducible genes, Bax and p21CIP1/WAF2 were also induced after CPT treatment. By contrast, anti-apoptotic Bcl-xL decreased after CPT treatment. p53 protein levels were high in the absence of treatment and were not enhanced after CPT treatment, which is consistent with the presence of mutant p53 in HT29 cells (Rodrigues et al., 1990).

Figure 4 Delayed activation of caspase-3 and degradation of PARP induced by CPT in HT29 cells. HT29 cells were treated with 1 mM CPT for 3 h, after which CPT was removed and fresh medium was added in the presence or absence of 3 mM CHX or 50 mM z-VAD-fmk for 3 days. Western blotting for caspase-3 (CPP32/YAMA) and PARP was performed as described in Materials and methods. A representative experiment is shown that was reproducible at least twice

Figure 5 Induction of Fas and Fas-ligand (FasL) in HT29 cells treated with CPT. HT29 cells were treated with 1 mM CPT for 3 h, after which CPT was removed and fresh medium was added in the presence or absence of 3 mM CHX or 50 mM z-VAD-fmk for 3 days. Fas and FasL were measured by Western blotting at the indicated times (a) or 3 days after CPT treatment (b). Representative experiments are shown

1855

Activation of the Fas pathway independently from Fas-ligand in CPT-treated HT29 cells Activation of the Fas signaling pathway is associated with the recruitment of FADD to Fas (Aragane et al., 1998). Cell lysates from CPT-treated HT29 cells were immunoprecipitated using anti-Fas monoclonal antibody and Western blotting was used to detect FADD and Fas proteins. Figure 7 shows that Fas and FADD were recruited after CPT treatment. To determine whether Fas was functional after CPT treatment and whether the complete Fas/FasL pathway was required for CPT-induced apoptosis, FasL, antiFas monoclonal antibody and FasL inhibitor were used. Figure 8a shows that FasL potentiated DNA fragmentation in CPT-treated HT29 cells. By contrast, FasL had no detectable eect in untreated HT29 cells. CH-11, an anti-Fas antibody that induces Fasmediated apoptosis in Fas expressing cells (Tillman et al., 1998), also potentiated CPT-induced apoptosis in HT29 cells (Figure 8a). These data indicated that Fas was functional in CPT-treated cells. We next used blocking anti-Fas monoclonal antibody ZB4 or FasL inhibitor to test whether FasL was involved in CPTinduced apoptosis. Figure 8b shows that ZB4 and FasL inhibitor did not protect HT29 cells from apoptosis induced by CPT. Under the conditions of the experiments shown in Figure 8b, both the blocking anti-Fas monoclonal antibody ZB4 or FasL inhibitor eectively blocked apoptosis in Jurkat cells (data not shown). Transient expression of Fas-pathway inhibitors prevents CPT-induced apoptosis We have previously used transient expression of FADD antisense, FADD dominant negative, MC159 and E8 constructs to demonstrate the role of a Fasligand-independent, FADD-mediated activation of the Fas death pathway in cisplatin-induced cytotoxicity in HT29 cells (Micheau et al., 1999). To determine whether FADD played a role in CPT-induced

Figure 6 Eects of CPT treatment on p53, p21CIP1/WAF1, Bax and Bcl-xL protein levels. Cells were treated with 1 mM CPT for 3 h. CPT was then removed and cells were incubated in fresh medium for the indicated times. Protein levels were detected by Western blotting as described in Materials and methods Oncogene


1856

apoptosis, we repeated these transient transfections. FADD-DN is a dominant negative construct that is capable of blocking Fas signal transduction (Memon et al., 1998). MC159 and E8 are two viral proteins that inhibit Fas-induced apoptosis at the level of FADD and procaspase-8, respectively (Bertin et al., 1997). Transient overexpression of FADD-AS, FADD-DN and MC159 or E8 all prevented CPT-induced apoptosis in HT29 cells (Figure 8c). These transient transfections also prevented CPT-induced cytotoxicity as measured by a methylene blue colorimeteric assay (not shown). Discussion Understanding the mechanisms of anticancer druginduced apoptosis is of principal importance for monitoring cancer response to therapy and for developing novel therapeutic strategies. Carcinoma cell

Figure 7 Recruitment of FADD to Fas in HT29 cells treated with CPT. Cells were treated with 1 mM CPT for 3 h. CPT was then removed and cells were incubated in fresh medium for the indicated times. Cell lysates were immunoprecipitated (IP) using anti-Fas monoclonal antibody (5 mg) at the indicated times after ending the CPT treatment. FADD and Fas were determined by Western blotting

lines are in general more resistant to apoptosis than human leukemia cell lines that undergo rapid apoptosis with internucleosomal DNA fragmentation (Dubrez et al., 1995; Goldwasser et al., 1995). The present study demonstrates that human colon HT29 cells also undergo apoptosis following topoisomerase I-mediated DNA damage in response to CPT treatment. Typical apoptotic morphology was observed as well as caspase3 activation. However, apoptosis was delayed in HT29 cells with a 48 h lag period that coincided with elevation of key apoptotic proteins: Fas, FasL and Bax. Degradation of genomic DNA in the course of apoptosis starts by excision of large DNA fragments of chromatin loop size and is generally followed by internucleosomal cleavage that produces the typical apoptotic DNA ladder. The ®rst cleavage step can be detected in virtually all apoptotic cells, but DNA laddering is not ubiquitously observed (present study) (Khodarev et al., 1998; Oberhammer et al., 1994; Rusnak et al., 1996). Nevertheless, the generation of the 25 ± 50 kbp DNA fragments is sucient to mediate chromatin condensation. Thus, the generation of high molecular weight (HMW) DNA fragmentation is a reliable biochemical marker for apoptosis that can be conveniently detected in cell culture using the ®lter elution assay (Bertrand et al., 1995) or the TUNEL assay. Our data show that CPT-induced HMW DNA fragmentation cannot progress to an internucleosomal DNA ladder in HT29 cells. The requirement for protein synthesis for CPTinduced apoptosis may explain the delayed onset of DNA fragmentation observed in CPT-treated HT29 cells. We previously reported that cycloheximide also prevented the induction of apoptosis by 7-hydroxystaurosporine, UCN-01 (Shao et al., 1997b) or

Figure 8 Fas-ligand (FasL) and anti-Fas agonistic antibody CH-11 potentiate CPT-induced apoptosis in HT29 cells (a). CPTinduced DNA fragmentation is resistant to the anti-Fas antibody ZB4 and FasL inhibitor (b). CPT-induced apoptosis is suppressed by inhibitors of the Fas pathway (c). Cells were treated with 1 mM CPT for 3 h, after which CPT was removed and fresh medium was added in the presence or absence of 100 mg/ml FasL or 100 ng/ml CH-11, or 500 ng/ml ZB4 or 50 mg/ml FasL inhibitor. Percentage of apoptotic cells (Hoechst positive) 48 h after CPT treatment (1 mM). Results of three independent experiments are shown with error bars representing standard error Oncogene


brefeldin A (Shao et al., 1996) in HT29 cells. Hence, HT29 cells provide an interesting experimental system to investigate the death signal protein/pathway(s) required for programmed cell death/apoptosis. Two apoptosis pathways were induced by topoisomerase I-mediated DNA damage in HT29 cells: the Fas/FasL death receptor and the Bax pathways. Interestingly, both the Fas (Owen-Schaub et al., 1995) and the Bax (Miyashita and Reed, 1995; Zhan et al., 1995) genes have been reported to be induced by p53. Our results demonstrate that DNA damage can induce p53-responsive genes in HT29 cells. However, CPT-induced apoptosis is likely to be independent of p53 because p53 is mutated in HT29 cells (O'Connor et al., 1997; Rodrigues et al., 1990) and because p53 protein levels did not accumulate at the time of apoptosis response (Figure 6). Thus, it appears that topoisomerase I-mediated DNA damage can lead to increased protein levels for Fas, FasL, p21CIP1/WAF1 and Bax independently of p53. Because cycloheximide suppressed the Fas elevation, it is likely that Fas synthesis was induced by camptothecin treatment. The elevation of Bax and reduction of Bcl-XL suggests that the mitochondrial death pathway is also activated in HT29 cells treated with CPT. Activation of the Fas pathway by DNA damage did not require caspase activation because the caspase inhibitor z-VAD-fmk did not block Fas induction. Thus, caspase activation is only downstream from Fas activation. The fact that FasL or cytolytic monoclonal antibody CH-11 potentiated CPT-induced apoptosis demonstrate that the Fas pathway was functional in CPT-treated cells (Figure 8). Recruitment of FADD to Fas (Figure 7) is also consistent with functional activation of the Fas pathway in CPT-treated HT29 cells. These results suggest that activation of the Fas pathway could be used to synergize with CPT for the treatment of colon carcinoma. Although FasL was induced by CPT, neither the FasL inhibitor nor the ZB4 blocking antibody could eectively block CPT-induced apoptosis, which suggests that induction of apoptosis was independent of FasL in HT29 cells. Thus, in our study, FasL induction appears futile with respect to activation of the Fas pathway. Two recent reports (Siegel et al., 2000; Chan et al., 2000) demonstrated that assembly of Fas monomers into trimers involves a pre-ligand assembly domain (PLAD) in the extracellular, amino-terminal region of the receptor. These reports also indicate that receptor selfassociation can occur independently of ligand binding. Whether camptothecin and other DNA damaging agents (Eischen et al., 1997; Micheau et al., 1999) in¯uence the spontaneous formation of Fas trimers and/ or increase the recruitment of FADD to Fas spontaneous trimers requires further investigation. Together, our results suggest that CPT-induced DNA damage activates the Fas pathway independently of FasL, which is in agreement with a recent study demonstrating that UV irradiation can directly activate the Fas pathway by directly inducing the recruitment of FADD to Fas, independently of FasL (Aragane et al., 1998).

Figure 9 summarizes our view of the molecular mechanisms/pathways activated during programmed cell death in p53-mutant human colon carcinoma HT29 cells in response to topoisomerase I-mediated DNA damage. It is likely that additional pathways are activated in addition to the Fas pathway, one of them being the induction of Bax that can directly activate mitochondria-mediated cell death.

1857

Materials and methods Drugs, chemicals and antibodies CPT was provided by the Drug Synthesis and Chemistry Branch, DCT, NCI, and stored frozen as a 10 mM stock solution in DMSO. The caspase inhibitor, N-benzyloxycarbony-Val-Ala-Asp(O-methyl)-¯uoromethylketone (z-VADfmk) was purchased from Enzyme System Products (Dublin, CA, USA). Other drugs and reagents, unless otherwise mentioned, were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Anti-CPP32 monoclonal antibody was from Transduction Laboratories (Lexington, KY, USA). Anti-PARP monoclonal antibody was a kind gift from Dr Guy G Poirier (Department of Molecular Endocrinology, Centre Hospitalier de L'UniversiteÂ Laval, Quebec, Canada). Anti-Fas, FasL, Bax, p53, and p21 monoclonal antibodies were from mouse (Oncogene Research Products, Cambridge, MA, USA). AntiBcl-xL monoclonal antibody was from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). FasL, FasL inhibitor, anti-Fas agonistic monoclonal antibody CH-11 and blocking monoclonal antibody ZB4 were purchased from Kamiya Biomedical Company (Seattle, WA, USA). Neutralizing antiFasL monoclonal antibody NOK1 was purchased from Pharmingen (San Diego, CA, USA). [14C]-thymidine (53.6 mCi/mmol) was purchased from New England Nuclear (Boston, MA, USA). Cell culture and DNA labeling Human colon carcinoma HT29 cells were grown at 378C in the presence of 5% CO2 in RPMI 1640 medium supplemented with 5% fetal bovine serum, 2 mM glutamine. Jurkat cells were grown at 378C in the presence of 5% CO2 in RPMI

Figure 9 Schematic representation of CPT-induced apoptotic cell death in human colon carcinoma HT29 cells Oncogene


1858

1640 medium supplemented with 10% fetal bovine serum, 2 mM glutamine. For DNA labeling, the cells were grown with [14C]thymidine (0.02 mCi/ml) for one doubling-time and then chased in isotope-free medium overnight prior to drug treatment. Filter elution assay DNA fragmentation related to apoptosis was measured by ®lter elution (Bertrand et al., 1994, 1995). Brie¯y, each sample was loaded onto vinyl/acrylic copolymers ®lters (0.8 mm pore size, 25 mm diameter, Gelman Sciences Inc., Ann Arbor, MI, USA). Filters were washed with 5 ml of Hanks' balance salt solution (HBSS), and lysed with 5 ml LS10 solution (0.2% sodium sarkosyl, 2 M NaC1, 0.04 M EDTA, pH 10.0) After the lysis had dripped through, ®lters and lysates were washed with 5 ml of 0.02 M Na2EDTA (pH 10.0). Filters were then processed for liquid scintillation counting (Shao et al., 1997b). DNA fragmentation was calculated as the percentage of counts in the lysis fraction plus EDTA wash relative to total intracellular DNA (Shao et al., 1996). DNA gel electrophoresis For conventional gel electrophoresis, cellular DNA was extracted as described previously (Shao et al., 1997b). Brie¯y, cells were resuspended in 30 ml of buer containing 10 mM Tris-HC1, 150 mM NaC1 and 10 mM EDTA, pH 8.0. 1.5 ml of 10% SDS solution was added, followed by 1.5 ml of protease K solution (10 mg/ml in distilled water) and incubated at 508C overnight. One-tenth volume of 3 M sodium acetate and two volumes of cold ethanol were added, mixed and kept on ice for 30 min. After centrifugation at 14 500 g for 15 min at 48C, the pellets were washed twice in cold 70% ethanol and dried at room temperature, followed by RNase (10 mg/ml) digestion at 378C for 1 h. Samples were loaded into 2% agarose gels. Electrophoreses were performed at 0.9 V/cm for 14 ± 16 h. For pulse-®eld gel electrophoresis, DNA samples were prepared by lyzing cells embedded in 1.5% agarose plugs, followed by digestion with proteinase K solution (1 mg/ml ®nal concentration in 0.01 M Tris, 450 mM EDTA and 1% laurylsarcosine) according to the manufacturer's instructions (Bio-Rad Laboratories, Richmond, CA, USA). Electrophoresis was carried out with CHEF-DR II system (Bio-Rad) using 0.56 Tris-Borate buer at 88C for 8 h at 200 V with the pulse time ramped from 1 ± 4 s. The DNA in the gel was visualized with ethidium bromide according to standard procedures. Cell morphology and flow cytometry Cells were harvested, centrifuged, washed with phosphatebuered saline (PBS) and ®xed with 1% paraformaldehyde

followed by 70% ethanol. At the time of staining the cells were resuspended in PBS and stained with 4', 6-diamidino-2phenylindole (DAPI, 2 mg/ml) for 30 min at 378C. Cells were then examined with a ¯uorescent microscope. Cell cycle analyses were performed as described previously (Shao et al., 1997a). Brie¯y, cells were harvested and ®xed in 70% ethanol. Before analysis by ¯ow cytometry, cells were washed with PBS, treated with 1 mg/ml RNase and stained with 50 mg/ml propidium iodide for at least 30 min. DNA content was determined by FACScan ¯ow cytometry (SOBR model, 15 000 cells per sample) (Becton Dickinson Immunocytometry System, San JoseÂ, CA, USA). Western blotting analysis Whole cell lysates were prepared as described previously (Shao et al., 1997b). Protein detection was performed using a protein assay kit according to the manufacturer's instructions (Bio-Rad, Hercules, CA, USA). Proteins were resolved at 125 V on SDS-polyacrylamide gels (NOVEX, San Diego, CA, USA) and electrophoretically transferred to Immobilon membranes (Millipore, Bedford, MA, USA) for 2 h at 30 V. Membranes were blocked overnight in PBS-T (phosphate buered saline-Tween 20) containing 5% nonfat dried milk, probed with primary antibody for 1 h, and with secondary Ab-HRP for an additional hour. Enhanced chemiluminescence (ECL, DuPont NEN, Boston, MA, USA) was used according to the manufacturer's instructions. Transient transfections Vectors used for these transfections have been described elsewhere (Micheau et al., 1999). pCI-MC159 and pCI-E8 were kindly provided by Dr JI Cohen (NIH, Bethesda, MD, USA). PCI-FADD-DN was a kind gift from Dr CM Zacharchuk (NIH, Bethesda, MD, USA). HT29 cells were seeded 24 h before transfection with 2 mg of pCI-neo, pCIMC159, pCIE8, pVI-FADD-DN, pBK-CMV or pBK-CMVFADD-AS using LipofectAMINE Plus (Life Technologies Co., Gaithersburg, MD, USA). These cells were treated or not 16 h after transfection with 1 mM CPT, then analysed for apoptosis 48 h later or for cytotoxicity 72 h later. Cell viability was measured by the use of a previously described methylene blue colorimetric assay (Micheau et al., 1999) whereas the percentage of apoptotic cells was determined 48 h after treatment by trypsinizing the cells, staining these cells with 1 mg/ml Hoechst 33352 for 15 min at 378C and analysing 300 cells by ¯uorescent microscopy.

Abbreviations CPT, camptothecin; CHX, cycloheximide; PARP, poly (ADP-ribose) polymerase.

References Aragane Y, Kulms D, Metze D, Wilkes G, Poppelmann B, Luger TA and Schwarz T. (1998). J. Cell. Biol., 140, 171 ± 182. Bertin J, Armstrong RC, Ottilie S, Martin DA, Wang Y, Banks S, Wang GH, Senkevich TG, Alnemri ES, Moss B, Lenardo MJ, Tomaselli KJ and Cohen JI. (1997). Proc. Natl. Acad. Sci. USA, 94, 1172 ± 1176. Oncogene

Bertrand R, Kohn KW, Solary E and Pommier Y. (1995). Drug Development Res., 34, 138 ± 144. Bertrand R, Solary E, O'Connor P, Kohn KW and Pommier Y. (1994). Exp. Cell Res., 211, 314 ± 321. Chan FK-M, Chun HJ, Zhen L, Siegel RM, Bui KL and Lenardo MJ. (2000). Science, 288, 2351 ± 2354.


Chen AY and Liu LF. (1994). Annu. Rev. Pharmacol. Toxicol., 94, 194 ± 218. Dubrez L, Goldwasser F, Genne P, Pommier Y and Solary E. (1995). Leukemia, 9, 1013 ± 1024. Eischen CM, Kottke TJ, Martins LM, Basi GS, Tung JS, Earnshaw WC, Leibson PJ and Kaufmann SH.(1997). Blood, 90, 935 ± 943. Favre-Felix N, Fromentin A, Hammann A, Solary E, Martin F and Bonnotte B. (2000). J. Immunol, 164, 5023 ± 5027. Friesen C, Herr I, Krammer PH and Debatin KM. (1996). Nat. Med., 2, 574 ± 577. Goldwasser F, Bae IS, Torres K and Pommier Y. (1995). Cancer Res., 55, 2116 ± 2121. Goldwasser F, Shimizu T, Jackman J, Hoki Y, O'Connor PM, Kohn KW and Pommier Y. (1996). Cancer Res., 56, 4430 ± 4437. Gunthert AR, Strater J, von Reyher U, Henne C, Joos S, Koretz K, Moldenhaue RG, Krammer PH and Moller P. (1996). J. Cell. Biol., 134, 1089 ± 1096. Houghton JA, Harwood FG and Tillman DM. (1997). Proc. Natl. Acad. Sci. USA, 94, 8144 ± 8149. Khodarev NN, Sokolova IA and Vaughan AT. (1998). Int. J. Radiat Biol., 73, 455 ± 467. Krammer PH, Dhein J, Walczak H, Behrmann I, Mariani S, Matiba B, Fath M, Daniel PT, Knipping E and Westendorp MO. (1994). Immunol Rev., 142, 175 ± 191. Kuida K, Lippke JA, Ku G, Harding MW, Livingston DJ, Su MS.-S and Flavell RA. (1995). Science, 267, 2000 ± 2006. Memon SA, Hou J, Moreno MB and Zacharchuk CM. (1998). J. Immunol, 160, 2046 ± 2049. Micheau O, Solary E, Hamman A and Dimanche-Boitrel M.T. (1999). J. Biol. Chem., 274, 7987 ± 7992. Micheau O, Solary E, Hammann A, Martin F and Dimanche-Boitrel MT. (1997). J. Natl. Cancer, 89, 783 ± 789. Miyashita T and Reed JC. (1995). Cell, 80, 293 ± 299. Moller P, Koretz K, Leithauser F, Bruderlein S, Henne C, Quentmeier A and Krammer PH. (1994). Int. J. Cancer, 57, 371 ± 377. Muzio M, Chinnaiyan AM, Kischkel FC, O'Rourke K, Shevchenko A, Ni J, Scadi C, Bretz JD, Zhang M, Gentz R, Mann M, Krammer PH, Peter ME and Dixit VM. (1996). Cell, 85, 817 ± 827. Nagata S and Golstein P. (1995). Science, 267, 1449 ± 1456. Nicholson DW, Ali A, Thornberry NA, Vaillancourt JP, Ding CK, Gallant M, Gareau Y, Grin PR, Labelle M, Lazebnik Y, Munday NA, Raju SM, Smulson ME, Yamin T.-T, Yu V and Miller DK. (1995). Nature, 376, 37 ± 43.

O'Connell J, O'Sullivan GC, Collins JK and Shanahan F. (1996) J. Exp. Med., 184, 1075 ± 1082. O'Connor PM, Nieves-Neira W, Kerrigan D, Bertrand R, Goldman J, Kohn KW and Pommier Y. (1991). Cancer Commun., 3, 233 ± 240. O'Connor PM, Jackman J, Bae I, Myers TG, Fan S, Mutoh M, Scudiero DA, Monks A, Sausville EA, Weinstein JN, Friend S, Fornace Jr AJ and Kohn KW. (1997) Cancer Res., 57, 4285 ± 4300. Oberhammer FA, Hochegger K, Froschl G, Tiefenbacher R and Pavelka M. (1994). J. Cell Biol., 126, 827 ± 837. Owen-Schaub LB, Zhang W, Cusack JC, Angelo LS, Santee SM, Fujiwara T, Roth JA, Deisseroth AB, Zhan WW, Kruzel E and Radinsky R. (1995). Mol. Cell. Biol., 15, 3032 ± 3040. Pommier Y, Pourquier P, Fan Y and Strumberg D. (1998). Biochim. Biophys. Acta, 1400, 83 ± 105. Rodrigues NR, Rowan A, Smith MEF, Kerr IB, Bodmer WF, Gannon JV and Lane DP. (1990). Proc. Natl. Acad. Sci. USA, 87, 7555 ± 7559. Rusnak JM, Calmels TP, Hoyt DG, Kondo Y, Yalowich JC and Lazo JS. (1996). Mol. Pharmacol., 49, 244 ± 252. Shao R.-G, Cao C.-X, Zhang H, Kohn KW, Wold MS and Pommier Y. (1999). EMBO J., 18, 1397 ± 1406. Shao RG, Cao CX, Shimizu T, O'Connor PM, Kohn KW and Pommier Y. (1997a). Cancer Res., 57, 4029 ± 4035. Shao RG, Shimizu T and Pommier Y. (1996). Exp. Cell Res., 227, 190 ± 196. Shao RG, Shimizu T and Pommier Y. (1997b). Exp. Cell Res., 234, 388 ± 397. Shimizu T, O'Connor PM, Kohn KW and Pommier Y. (1995). Cancer Res., 55, 228 ± 231. Siegel RM, Frederiksen JK, Zacharias DA, Chan FK-M, Johnson M, Lynch D, Tsien RY and Lenardo MJ. (2000). Science, 288, 2345 ± 2357. Strumberg D, Pilon AA, Smith M, Hickey R, Malkas L and Pommier Y. (2000). Mol. Cell Biol., 20, 3977 ± 3987. Tanaka M, Suda T, Haze K, Nakamura N, Sato K, Kimura F, Motoyosh IK, Mizuke M, Tagawa S, Ohga S, Hatake K, Drummond AH and Nagata S. (1996). Nat. Med., 2, 317 ± 322. Tillman DM, Harwood FG, Gibson AA and Houghton JA. (1998). Cell Death Dier., 5, 450 ± 457. Villunger A, Egle A, Kos M, Hartmann BL, Geley S, Ko¯er R and Greil R. (1997). Cancer Res., 57, 3331 ± 3334. Zhan Q, El-Deiry W, Bae I, Alamo Jr I, Kastan MB, Vogelstein B and Fornace JAJ. (1995). Int. J. Oncol., 6, 937 ± 946.

1859

Oncogene