P53-Mediated Upregulation of Dcr1 Impairs Oxaliplatin/TRAIL-Induced
Total Page:16
File Type:pdf, Size:1020Kb
Oncogene (2008) 27, 4161–4171 & 2008 Macmillan Publishers Limited All rights reserved 0950-9232/08 $30.00 www.nature.com/onc ORIGINAL ARTICLE p53-Mediated upregulation of DcR1 impairsoxaliplatin/TRAIL-induced synergistic anti-tumour potential in colon cancer cells F Toscano1,3, Z El Fajoui1,3, F Gay1, N Lalaoui2, B Parmentier1, J-A Chayvialle1, J-Y Scoazec1, O Micheau2, J Abello1 and J-C Saurin1 1INSERM, U865, Institut Fe´de´ratif de Recherche Lyon Est, Lyon, F-69372, France; Univ. Lyon 1, Faculte´ Lae¨nnec, Lyon, F-69008, France and 2INSERM, U866, Dijon, F21079, France; Univ. Bourgogne, Dijon, F-21079, France Oxaliplatin hasemerged asa major chemotherapeutic Introduction drug in the treatment of advanced colorectal cancer, yet like most conventional cancer therapeutics, its efficacy is Colorectal cancer is one of the leading causes of cancer- often compromised due to p53 mutations. Unlike oxali- related deathworldwide (van Geelen et al., 2006). After platin, tumour necrosis factor-related apoptosis-inducing surgery, up to 50% of patients will relapse and die from ligand (TRAIL) triggersapoptosisin a p53-independent metastatic disease (Gramont, 2005). The development of manner, and chemotherapy isknown to overcome tumour colorectal carcinomas is caused by multiple genetic resistance to TRAIL-induced cell death in most cancer changes, including frequent inactivating mutations of cells. Using a panel of colon cancer cell lines, we assessed the p53 gene (Grady and Markowitz, 2002). The wild- the ability of oxaliplatin to sensitize to TRAIL-induced type p53 gene plays a key role by inducing cell cycle apoptosis. We demonstrate that while both drugs addi- arrest and activating the intrinsic apoptosis pathway in tively or synergistically induced apoptosis in almost all response to a large variety of chemotherapeutic agents cell lines tested, p53 wild-type colon cancer cells such as (van Geelen et al., 2006). HCT116, LS513 or LS174T remained resistant. Impaired Although oxaliplatin-based drug combinations have TRAIL-induced cell death resulted from a strong p53 improved response rates in the treatment of advanced dependent, oxaliplatin-mediated, DcR1 receptor expres- colorectal cancer, chemoresistance remains a major sion increase. According to our finding, downregulation of problem. It is therefore important to identify prognostic DcR1 using siRNA, in p53 wild-type colon cancer cells, factors that can help to develop new patient-tailored restored oxaliplatin/TRAIL synergistic apoptotic activity. treatment strategies. We previously showed that On the contrary, exogenous DcR1 overexpression in oxaliplatin, used in a clinically relevant manner, mainly SW480, a p53-mutated cell line, abolished the synergy acted in a cytostatic fashion (Toscano et al., 2007), between the two drugs. Altogether we demonstrate for although in some studies, using long-term treatment the first time that p53 negatively regulates oxaliplatin- schedules, this agent was shown to induce apoptosis mediated TRAIL-induced apoptotic activity through through the mitochondrial intrinsic pathway (Shankar DcR1 upregulation. Our findingscould have important and Srivastava, 2004). implications for future therapeutic strategies, and suggest Unlike oxaliplatin however, tumour necrosis factor- that the association oxaliplatin/TRAIL should be related apoptosis-inducing ligand (TRAIL) activates restricted to patients harbouring a non-functional p53 the extrinsic apoptosis pathway in most tumour cell protein. types, but is unable to trigger cell deathin normal cells Oncogene (2008) 27, 4161–4171; doi:10.1038/onc.2008.52; (Ashkenazi, 2002). TRAIL binding to its cognate published online 17 March 2008 agonistic receptors, DR4 and DR5, induces the forma- tion of the death-inducing signaling complex, a multi- Keywords: oxaliplatin; TRAIL; antagonism; p53; molecular platform that allows the activation of the DcR1; colon cancer initiator procaspases-8 and -10, which triggers the proteolytic caspase cascade leading to apoptosis (Thorburn, 2004). Cross talks exist between the extrinsic and the intrinsic pathway. In some tumour cells, the mitochondrial amplification loop is required to engage the apoptotic machinery upon TRAIL stimulation (Scaffidi et al., 1999). On the other hand, some chemotherapeutic drugs involve, at least partially, the Correspondence: Dr J Abello, INSERM Unite´ 865, Faculte´ de engagement of deathdomain-containing receptors to Me´ decine Lae¨ nnec, 7 rue G. Paradin, F-69372 Lyon Cedex 08, France. trigger cell death(Micheau et al., 1999). Recent evidence E-mail: [email protected] 3These authors contributed equally to this work. indicates that both tumour and normal cells can acquire Received 16 July 2007; revised 20 December 2007; accepted 28 January resistance to TRAIL-induced killing by upregulating 2008; published online 17 March 2008 either of the two antagonistic receptors, DcR1 and Oxaliplatin/TRAIL combination in colon cancer F Toscano et al 4162 DcR2, which do not transfer any apoptotic signal contrast, there was no correlation between CIs and (Davidovich et al., 2004; Clancy et al., 2005; Merino sensitivities to oxaliplatin (Figure 1c, right graph). et al., 2006), indicating that the four TRAIL receptors Used alone at 5 mM, a sub-toxic concentration, are involved for regulating TRAIL-induced apoptosis oxaliplatin failed to induce detectable poly-ADP ribose (Kelley and Ashkenazi, 2004; Bouralexis et al., 2005; polymerase (PARP) cleavage in the six colon cancer cell Merino et al., 2007). lines tested. In contrast, TRAIL used at 10 ng mlÀ1 Combination of compounds that induce genotoxic resulted in PARP cleavage in all but two (HT29 stress to ligands of the TNF family, that activate death and V9P, which are resistant to TRAIL) cell lines receptors, was demonstrated to generate synergistic anti- (Figure 2a). Combining oxaliplatin to TRAIL however tumour responses in several human tumour types and to sensitized LS1034, SW480 and V9P cells to cell death overcome tumour resistance (Ashkenazi, 2002; Wajant through a mechanism that involved the apoptotic et al., 2005). Modification of TRAIL agonistic receptor machinery as evidenced by PARP cleavage expression by chemotherapeutic agents was shown to (Figure 2a), but failed to sensitize HT29 cells at that account for TRAIL-induced sensitization in some cells concentration. Considering two cell lines, from our cell (Nagane et al., 2000; Bouralexis et al., 2003). Although panel, exhibiting different response to the TRAIL/ all four receptors have been described as p53 target oxaliplatin combination, the HCT116 and the HT29, genes, the importance of p53 in chemotherapy-mediated at concentrations that inhibit growth by 50% (IC50), we sensitization to TRAIL remains unclear (Jang et al., could demonstrate that in contrast to TRAIL, oxalipla- 2004). The goal of our study was thus to clarify the role tin never triggered caspase-3 activation (Figure 2b). of p53 in controlling TRAIL receptor expression in However, when cells were treated with both drugs, colon cancer cells treated withoxaliplatin, and to caspase-3 activation increased as compared to TRAIL determine its impact on oxaliplatin-induced sensitization treatment in HT29 cell line but remained undetectable in to TRAIL. HCT116 cells. In line withthisfinding, a synergy between oxaliplatin and TRAIL was observed on PARP cleavage only in HT29 cells (Figure 2c). Moreover, the percentage of sub-G1 cells was significantly higher in Results HT29 cells treated withtheoxaliplatin/TRAIL combi- nation as compared to any single drug (Figure 2d), while Oxaliplatin and TRAIL combination exhibits synergistic the opposite effect was observed when HCT116 cells toxic effects on resistant colon cancer cell lines were treated withbothdrugs, as compared to TRAIL. Sensitivities to oxaliplatin- or TRAIL-induced cell death According to our previous findings, oxaliplatin-treated in the six colon cancer cell lines tested, as measured p53 wild-type HCT116 cells exhibited no DNA accu- using (3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyl-tetrazolium mulation in the sub-G0/G1 phase, no caspase-3 activa- bromide) assays (MTTs), varied from one cell line to tion nor PARP cleavage but a G0/G1 arrest that another (Table 1). However, as shown in Figure 1a, impaired cell growth(Toscano et al., 2007). These oxaliplatin and TRAIL exerted important synergistic results indicate that the synergistic apoptotic activity effects in SW480, HT29 and V9P cells (combination induced by associating oxaliplatin and TRAIL could be index (CI), 0.76±0.06, 0.56±0.11 and 0.45±0.15, controlled by p53, since only p53-mutated cell lines respectively), but nearly additive effects in LS1034 and responded in a synergistic manner to that combination, Isreco1 cells (CI, 0.91±0.10 and 1.02±0.04, respec- whereas the p53 wild-type cell line HCT116 exhibited tively), and antagonistic effects in the HCT116 cell line antagonism to the combined treatment. (CI, 1.28±0.02). Synergy was also evidenced using cytotoxicity assays as illustrated in HT29 cells Antagonism between oxaliplatin and TRAIL in the (Figure 1b). Interestingly, the more the cell lines were HCT116cell line is tightly dependent on p53 status resistant to TRAIL, the more oxaliplatin and TRAIL We therefore hypothesized that the p53 status could play showed synergistic toxicity (Figure 1c, left graph). In an important role in controlling the synergistic apoptotic activity in colon cancer cells. To address this question, Table 1 Sensitivities to oxaliplatin- and TRAIL-induced cell deathof