DCLRE1B/Apollo germline mutations associated with renal cell carcinoma impair telomere protection.

Hereditary renal cell carcinoma (RCC) is caused by germline mutations in a subset of genes, including VHL, MET, FLCN, and FH. However, many familial RCC cases do not harbor mutations in the known predisposition genes. Using Whole Exome Sequencing, we identified two germline missense variants in the DCLRE1B/Apollo gene (ApolloN246I and ApolloY273H) in two unrelated families with several RCC cases. Apollo encodes an exonuclease involved in DNA Damage Response and Repair (DDRR) and telomere integrity. We characterized these two functions in the human renal epithelial cell line HKC8. The decrease or inhibition of Apollo expression sensitizes these cells to DNA interstrand crosslink damage (ICLs). HKC8 Apollo-/- cells appear defective in the DDRR and present an accumulation of telomere damage. Wild-type and mutated Apollo forms could interact with TRF2, a shelterin protein involved in telomere protection. However, only ApolloWT can rescue the telomere damage in HKC8 Apollo-/- cells. Our results strongly suggest that ApolloN246I and ApolloY273H are loss-of-function mutants that cause impaired telomere integrity and could lead to genomic instability. Altogether, our results suggest that mutations in Apollo could induce renal oncogenesis.

NEIL3-mediated proteasomal degradation facilitates the repair of cisplatin-induced DNA damage in human cells.

Anti-neoplastic effect of DNA cross-linking agents such as cisplatin, mitomycin C, and psoralen is attributed to their ability to induce DNA interstrand cross-links (ICLs), which block replication, transcription, and linear repair pathways by preventing DNA strand separation and trigger apoptosis. It is generally agreed that the Fanconi anemia (FA) pathway orchestrates the removal of ICLs by the combined actions of various DNA repair pathways. Recently, attention has been focused on the ability of the NEIL3-initiated base excision repair pathway to resolve psoralen- and abasic site-induced ICLs in an FA-independent manner. Intriguingly, overexpression of NEIL3 is associated with chemo-resistance and poor prognosis in many solid tumors. Here, using loss- and gain-of-function approaches, we demonstrate that NEIL3 confers resistance to cisplatin and participates in the removal of cisplatin-DNA adducts. Proteomic studies reveal that the NEIL3 protein interacts with the 26S proteasome in a cisplatin-dependent manner. NEIL3 mediates proteasomal degradation of WRNIP1, a protein involved in the early step of ICL repair. We propose that NEIL3 participates in the repair of ICL-stalled replication fork by recruitment of the proteasome to ensure a timely transition from lesion recognition to repair via the degradation of early-step vanguard proteins.

NPTX1 mutations trigger endoplasmic reticulum stress and cause autosomal dominant cerebellar ataxia.

With more than 40 causative genes identified so far, autosomal dominant cerebellar ataxias exhibit a remarkable genetic heterogeneity. Yet, half the patients are lacking a molecular diagnosis. In a large family with nine sampled affected members, we performed exome sequencing combined with whole-genome linkage analysis. We identified a missense variant in NPTX1, NM_002522.3:c.1165G>A: p.G389R, segregating with the phenotype. Further investigations with whole-exome sequencing and an amplicon-based panel identified four additional unrelated families segregating the same variant, for whom a common founder effect could be excluded. A second missense variant, NM_002522.3:c.980A>G: p.E327G, was identified in a fifth familial case. The NPTX1-associated phenotype consists of a late-onset, slowly progressive, cerebellar ataxia, with downbeat nystagmus, cognitive impairment reminiscent of cerebellar cognitive affective syndrome, myoclonic tremor and mild cerebellar vermian atrophy on brain imaging. NPTX1 encodes the neuronal pentraxin 1, a secreted protein with various cellular and synaptic functions. Both variants affect conserved amino acid residues and are extremely rare or absent from public databases. In COS7 cells, overexpression of both neuronal pentraxin 1 variants altered endoplasmic reticulum morphology and induced ATF6-mediated endoplasmic reticulum stress, associated with cytotoxicity. In addition, the p.E327G variant abolished neuronal pentraxin 1 secretion, as well as its capacity to form a high molecular weight complex with the wild-type protein. Co-immunoprecipitation experiments coupled with mass spectrometry analysis demonstrated abnormal interactions of this variant with the cytoskeleton. In agreement with these observations, in silico modelling of the neuronal pentraxin 1 complex evidenced a destabilizing effect for the p.E327G substitution, located at the interface between monomers. On the contrary, the p.G389 residue, located at the protein surface, had no predictable effect on the complex stability. Our results establish NPTX1 as a new causative gene in autosomal dominant cerebellar ataxias. We suggest that variants in NPTX1 can lead to cerebellar ataxia due to endoplasmic reticulum stress, mediated by ATF6, and associated to a destabilization of NP1 polymers in a dominant-negative manner for one of the variants.

Involvement of PBRM1 in VHL disease-associated clear cell renal cell carcinoma and its putative relationship with the HIF pathway.

Von Hippel-Lindau (VHL) disease is the main cause of inherited clear-cell renal cell carcinoma (ccRCC) and is caused by germline mutations in the VHL tumor suppressor gene. Bi-allelic VHL alterations lead to inactivation of pVHL, which plays a major role by downstream activation of the hypoxia inducible factor (HIF) pathway. Somatic VHL mutations occur in 80% of sporadic ccRCC cases and the second most frequently mutated gene is polybromo 1 (PBRM1). As there is currently no data regarding PBRM1 involvement in VHL disease-associated ccRCC, the aim of the present study was to assess the PBRM1 mutational status, and PBRM1 and HIF expression in VHL disease-associated ccRCC series compared with a sporadic series. PBRM1 gene was screened by Sanger sequencing for 23 VHL-disease-associated ccRCC and 22 sporadic ccRCC cases. Immunohistochemical studies were performed to detect the expression of PBRM1, HIF1 and HIF2 for all cases. In VHL-associated tumors, 13.0% (n=3/23) had PBRM1 somatic mutations and 17.4% (n=4/23) had a loss of PBRM1 nuclear expression. In sporadic cases, 27.3% (n=6/22) showed PBRM1 somatic mutations and 45.5% (n=10/22) had a loss of PBRM1 nuclear expression. Loss of PBRM1 was associated with an advanced tumor stage. HIF1-positive tumors were observed more frequently in the VHL-associated ccRCC than in the sporadic series. Furthermore, in the VHL cohort, PBRM1 expression appeared to be associated more with HIF1 than with HIF2. Given that hereditary tumors tend to be less aggressive, these results would suggest that co-expression of PBRM1 and HIF1 may have a less oncogenic role in VHL-associated ccRCC.

Germline mutation in the NBR1 gene involved in autophagy detected in a family with renal tumors.

Hereditary Renal Cell Carcinomas (RCC) are caused by mutations in predisposing genes, the major ones including VHL, FLCN, FH and MET. However, many families with inherited RCC have no germline mutation in these genes. Using Whole Exome Sequencing on germline DNA from a family presenting three different histological renal tumors (an angiomyolipoma, a clear-cell RCC and an oncocytic papillary RCC), we identified a frameshift mutation in the Neighbor of BRCA1 gene 1 (NBR1), segregating with the tumors. NBR1 encodes a cargo receptor protein involved in autophagy. Genetic and functional analyses suggested a pathogenic impact of the mutation. Indeed, functional study performed in renal cell lines showed that the mutation alters NBR1 interactions with some of its partners (such as p62/SQSTM1), leading to a dominant negative effect. This results in an altered autophagic process and an increased proliferative capacity in renal cell lines. Our study suggests that NBR1 may be a new predisposing gene for RCC, however its characterization needs to be further investigated in order to confirm its role in renal carcinogenesis.

Differentiation of neural-type cells on multi-scale ordered collagen-silica bionanocomposites.

Cells respond to biophysical and biochemical signals. We developed a composite filament from collagen and silica particles modified to interact with collagen and/or present a laminin epitope (IKVAV) crucial for cell-matrix adhesion and signal transduction. This combines scaffolding and signaling and shows that local tuning of collagen organization enhances cell differentiation.

FGF1 induces resistance to chemotherapy in ovarian granulosa tumor cells through regulation of p53 mitochondrial localization.

Ovarian cancer remains associated with a high mortality rate and relapse is too frequently seen after chemotherapeutic treatment of granulosa cell tumors (GCTs) or epithelial ovarian cancers (EOCs). It is thus of major importance to progress in the knowledge of the molecular mechanisms underlying chemoresistance of ovarian tumors. Overexpression of Fibroblast Growth Factor 1 (FGF1) is observed in various cancers, correlates with poor survival and could be responsible for resistance to platinum-based chemotherapy of serous ovarian cancers. How FGF1 promotes escape to chemotherapy remains unknown. In previous studies, we showed that FGF1 inhibits p53 transcriptional activities, leading to increased cell survival of neuronal or fibroblast cell lines. In this study, we show that FGF1 favors survival of COV434 cells upon treatment with etoposide and cisplatin, two common chemotherapeutic molecules used for ovarian cancer. Etoposide and cisplatin induced mitochondrial depolarization, cytochrome c release and caspase activation in COV434 cells. Overexpression of FGF1 counteracts these events and thus allows increased survival of ovarian cells. In this study, FGF1 had little effect on p53 stability and transcriptional activities. Etoposide induced p21 expression as expected, but p21 protein levels were even increased in the presence of FGF1. Using RNA interference, we showed that p21 exerts an anti-apoptotic activity in COV434 cells. However abrogating this activity was not sufficient to restore cell death of FGF1-overexpressing cells. We also show for the first time that p53 mitochondrial pathway is involved in the cell death of COV434 cells. Indeed, p53 accumulates at mitochondria upon etoposide treatment and inhibition of p53 mitochondrial localization using pifithrin-µ inhibits apoptosis of COV434 cells. FGF1 also decreases mitochondrial accumulation of p53 induced by etoposide. This constitutes a novel mechanism of action for FGF1 to promote cell survival in response to chemotherapy.

FGF1 protects neuroblastoma SH-SY5Y cells from p53-dependent apoptosis through an intracrine pathway regulated by FGF1 phosphorylation.

Neuroblastoma, a sympathetic nervous system tumor, accounts for 15% of cancer deaths in children. In contrast to most human tumors, p53 is rarely mutated in human primary neuroblastoma, suggesting impaired p53 activation in neuroblastoma. Various studies have shown correlations between fgf1 expression levels and both prognosis severity and tumor chemoresistance. As we previously showed that fibroblast growth factor 1 (FGF1) inhibited p53-dependent apoptosis in neuron-like PC12 cells, we initiated the study of the interaction between the FGF1 and p53 pathways in neuroblastoma. We focused on the activity of either extracellular FGF1 by adding recombinant rFGF1 in media, or of intracellular FGF1 by overexpression in human SH-SY5Y and mouse N2a neuroblastoma cell lines. In both cell lines, the genotoxic drug etoposide induced a classical mitochondrial p53-dependent apoptosis. FGF1 was able to inhibit p53-dependent apoptosis upstream of mitochondrial events in SH-SY5Y cells by both extracellular and intracellular pathways. Both rFGF1 addition and etoposide treatment increased fgf1 expression in SH-SY5Y cells. Conversely, rFGF1 or overexpressed FGF1 had no effect on p53-dependent apoptosis and fgf1 expression in neuroblastoma N2a cells. Using different FGF1 mutants (that is, FGF1K132E, FGF1S130A and FGF1S130D), we further showed that the C-terminal domain and phosphorylation of FGF1 regulate its intracrine anti-apoptotic activity in neuroblastoma SH-SY5Y cells. This study provides the first evidence for a role of an intracrine growth factor pathway on p53-dependent apoptosis in neuroblastoma, and could lead to the identification of key regulators involved in neuroblastoma tumor progression and chemoresistance.

zVAD-fmk upregulates caspase-9 cleavage and activity in etoposide-induced cell death of mouse embryonic fibroblasts.

Caspases are key effectors of programmed cell death. Down- and up-regulation of their activity are involved in different pathologies. In most cells, zVAD-fmk prevents apoptosis. However, unexpected effects of zVAD-fmk have been characterized in different laboratories, cell models and cell death processes. We have previously shown that zVAD-fmk accelerates p53-dependent apoptosis in rat embryonic fibroblasts. In this study, we pursued our investigations on zVAD-fmk effects and focused our study at the mitochondrial level in mouse embryonic fibroblasts (MEFs). In both primary and immortalized (by AgT or 3T9 protocol) MEFs, zVAD-fmk increased etoposide-induced loss of ΔΨm. This increase correlated with an increase of the number of apoptotic cells in primary and 3T9 MEFs, but did not in AgT MEFs. In both types of immortalized MEFs, zVAD-fmk regulated neither p53 levels nor transcriptional activities, suggesting that zVAD-fmk acts downstream of p53. In MEFs, zVAD-fmk increased p53-dependent loss of ΔΨm, cytochrome c release and caspase-9 activity. Indeed, zVAD-fmk inhibited effector caspases (caspases-3, -6, -7) as expected but increased caspase-9 cleavage and activity in etoposide-treated MEFs. Q-VD-OPh, another caspase inhibitor, also increased both loss of ΔΨm and caspase-9 cleavage in etoposide-treated MEFs. Invalidation of bax and bak suppressed p53-dependent cell death and zVAD-fmk regulation of this process. Invalidation of caspase-9 did not inhibit mitochondrial membrane depolarization but suppressed zVAD-fmk amplification of this process. Altogether, our data suggest that caspase-9 activity is up-regulated by zVAD-fmk and is involved in an amplification loop of etoposide-induced cell death at the mitochondrial level in MEFs.

The p76(Rb) and p100(Rb) truncated forms of the Rb protein exert antagonistic roles on cell death regulation in human cell lines.

Several caspase-cleaved forms of the retinoblastoma protein have been described. Here, we compared the effect of full-length Rb versus the truncated p76(Rb) and p100(Rb) proteins on cell death regulation in five human cell lines. Interestingly, we observed that p76(Rb) triggers cell death in all tested cell lines and that p100(Rb) protects two cell lines against etoposide or TNF-alpha-induced cell death, whereas full-length Rb has no apoptotic effect. These results show that truncated forms of Rb can have specific activities in the regulation of cell death. They also suggest that caspase cleavage of Rb should not be simply assimilated to a degradation process. Finally, we show that cell death induced by p76(Rb) is Bax-dependent and is diminished by Bcl-2 overexpression or by caspase inhibition and that p100(Rb) could inhibit cell death by decreasing both p53 stability and caspase activity.

FGF1 nuclear translocation is required for both its neurotrophic activity and its p53-dependent apoptosis protection.

Fibroblast growth factor 1 (FGF1) is a differentiation and survival factor for neuronal cells both in vitro and in vivo. FGF1 activities can be mediated not only by paracrine and autocrine pathways involving FGF receptors but also by an intracrine pathway, which is an underestimated mode of action. Indeed, FGF1 lacks a secretion signal peptide and contains a nuclear localization sequence (NLS), which is consistent with its usual intracellular and nuclear localization. To progress in the comprehension of the FGF1 intracrine pathway in neuronal cells, we examined the role of the nuclear translocation of FGF1 for its neurotrophic activity as well as for its protective activity against p53-dependent apoptosis. Thus, we have transfected PC12 cells with different FGF1 expression vectors encoding wild type or mutant (Delta NLS) FGF1. This deletion inhibited both FGF1 nuclear translocation and FGF1 neurotrophic activity (including differentiation and serum-free cell survival). We also show that endogenous FGF1 protection of PC12 cells against p53-dependent cell death requires FGF1 nuclear translocation. Strikingly, wild type FGF1 is found interacting with p53, in contrast to the mutant FGF1 deleted of its NLS, suggesting the presence of direct and/or indirect interactions between FGF1 and p53 pathways. Thus, we present evidences that FGF1 may act by a nuclear pathway to induce neuronal differentiation and to protect the cells from apoptosis whether cell death is induced by serum depletion or p53 activation.

Mitochondrial localization of the low level p53 protein in proliferative cells.

p53 protein plays a central role in suppressing tumorigenesis by inducing cell cycle arrest or apoptosis through transcription-dependent and -independent mechanisms. Emerging publications suggest that following stress, a fraction of p53 translocates to mitochondria to induce cytochrome c release and apoptosis. However, the localization of p53 under unstressed conditions remains largely unexplored. Here we show that p53 is localized at mitochondria in absence of apoptotic stimuli, when cells are proliferating, localization observed in various cell types (rodent and human). This is also supported by acellular assays in which p53 bind strongly to mitochondria isolated from rat liver. Furthermore, the mitochondria subfractionation study and the alkaline treatment of the mitochondrial p53 revealed that the majority of mitochondrial p53 is present in the membranous compartments. Finally, we identified VDAC, a protein of the mitochondrial outer-membrane, as a putative partner of p53 in unstressed/proliferative cells.

Toxicities induced in cultured human hepatocarcinoma cells exposed to ochratoxin A: oxidative stress and apoptosis status.

Ochratoxin A (OTA) is a mycotoxin currently detected in stored animal and human food supplies as well as in human sera worldwide. OTA has diverse toxicological effects; however, the most prominent one is the nephrotoxicity. The present investigation was conducted to determine the molecular aspects of OTA toxicity in cultured human hepatocellular carcinoma cells. With this aim, we have monitored the effects of OTA on (i) cell viability, (ii) heat shock protein expressions as a parameter of protective and adaptive response, (iii) oxidative damage, and (iv) cell death signaling pathway. Our results clearly showed that OTA treatment inhibits cell proliferation, downregulates Hsp 70 and Hsp 27 protein and mRNA levels, and did not induce a significant reactive oxygen species generation. We have also demonstrated a decrease in mitochondrial membrane potential, a cytochrome c release, and an activation of caspase 9 and caspase 3 in response to OTA exposure. Moreover, OTA activates p53 expression, while some of its transcriptional target genes (Bax, Bak, PUMA, and p21) were found to downregulate. According to these data, we concluded that OTA may exert an inhibitory action on the transcriptional process. Besides, oxidative damage is not a major contributor to OTA toxicity. This mycotoxin induces a mitochondrial and caspase-dependent apoptotic cell death, which seems to be mediated by p53 transcriptional independent activities.

Comparative mechanisms of zearalenone and ochratoxin A toxicities on cultured HepG2 cells: is oxidative stress a common process?

Zearalenone (ZEN) and Ochratoxin A (OTA) are structurally diverse fungal metabolites that can contaminate feed and foodstuff and can cause serious health problems for animals as well as for humans. In this study, we get further insight of the molecular aspects of ZEN and OTA toxicities in cultured human HepG2 hepatocytes. In this context, we have monitored the effects of ZEN and OTA on (i) cell viability, (ii) heat shock protein (Hsp) 70 and Hsp 27 gene expressions as a parameter of protective and adaptive response, (iii) oxidative damage, and (iv) cell death pathways. Our results clearly showed that both ZEN and OTA inhibit cell proliferation. For ZEN, a significant induction of Hsp 70 and Hsp 27 was observed. In the same conditions, ZEN generated an important amount of reactive oxygen species (ROS). Antioxidant supplements restored the major part of cell mortality induced by ZEN. However, OTA treatment downregulated Hsp 70 and Hsp 27 protein and mRNA levels and did not induce ROS generation. Antioxidant supplements did not have a significant effect on OTA-induced cell mortality. Using another cell system (Vero monkey kidney cells), we demonstrated that OTA downregulates three members of HSP 70 family: Hsp 70, Hsp 75, and Hsp 78. Our findings showed that oxidative damage seemed to be the predominant toxic effect for ZEN, while OTA toxicity seemed to be rather because of the absence of Hsps protective response. Furthermore, the two mycotoxins induced an apoptotic cell death.

Fibroblast Growth Factor 1 inhibits p53-dependent apoptosis in PC12 cells.

The survival activity of FGF1 and the pro-apoptotic activity of p53 were characterized in vitro and/or in vivo for different types of neurons after different stresses and in different neurodegenerative pathologies. To investigate whether or not FGF1 and p53 pathways interact in neuronal cells, we studied the effect of FGF1 on p53-dependent apoptosis in PC12 cells. We first characterized p53-dependent PC12 cell death induced by etoposide (a DNA damaging agent). We showed that etoposide increased p53 stabilization, phosphorylation (Ser-15), nuclear translocation and transcriptional activity. In particular, p53 promoted mdm2, p21, puma and noxa expression in PC12 cells. The activation of p53 initiated a classical mitochondrial apoptosis process associated with caspases activation and nuclear degradation. We demonstrated that FGF1 protected PC12 cells from p53-dependent apoptosis upstream from mitochondrial and nuclear events. FGF1 inhibited etoposide-induced p53 phosphorylation, stabilization, nuclear translocation and transcriptional activity. This study presents the first evidence that FGF1 and p53 pathways interact in neuronal cells, and that FGF1 protects neuronal cells from p53-dependent apoptosis, suggesting that alterations of FGF1/p53 crosstalk could be involved in a large range of neurons and in neurological disorders.

FGF1 inhibits p53-dependent apoptosis and cell cycle arrest via an intracrine pathway.

We analysed the relationships between p53-induced apoptosis and the acidic fibroblast growth factor 1 (FGF1) survival pathway. We found that p53 activation in rat embryonic fibroblasts induced the downregulation of FGF1 expression. These data suggest that the fgf1 gene is a repressed target of p53. Unlike extracellular FGF1, which has no effect on p53-dependent pathways, intracellular FGF1 inhibits both p53-dependent apoptosis and cell growth arrest via an intracrine pathway. FGF1 increases MDM2 expression at both mRNA and protein levels. This increase is associated with an acceleration of p53 degradation, which may partly account for the ability of endogenous FGF1 to counteract p53 pathways. In the presence of FGF1, p53 was unable to transactivate bax, but no modification of p21 gene transactivation was observed. As Bax is an essential component of the p53-dependent apoptosis pathway, this suggests that intracellular FGF1 inhibits p53 pathways not only by decreasing the stability of p53, but also by modifying some of its transactivation properties. In conclusion, we showed that p53 and FGF1 pathways may interact in the cell to determine cell fate. Deregulation of one of these pathways modifies the balance between cell proliferation and cell death and may lead to tumor progression.

Caspase-9 can antagonize p53-induced apoptosis by generating a p76(Rb) truncated form of Rb.

The tumor suppressor Rb (retinoblastoma protein) is known to regulate p53-dependent apoptosis, but the mechanisms involved are unclear. In a rat fibroblast model, we previously observed that caspase inhibition potentiates p53-dependent apoptosis and prevents the Rb cleavage associated with p53 activation. These results suggested that a caspase(s) can antagonize p53-mediated apoptosis via the production of a protective Rb truncated form. Here, we identify caspase-9 as the caspase that interferes, upstream of the mitochondrion, with p53-induced apoptosis in both immortalized and primary fibroblasts. This caspase can be detected as a p38 processed form in living cells, in the absence of apoptosome formation and apoptotic signal. We also provide evidence that the involvement of caspase-9 in a pre-mitochondrial protective pathway results from the previously undescribed cleavage of Rb, at a LExD site, into a p76(Rb) form, which antagonizes p53-induced apoptosis. These results establish that a truncated form of Rb can display an antiapoptotic activity, rather than just being a by-product of Rb degradation.

Early loss of E-cadherin from cell-cell contacts is involved in the onset of Anoikis in enterocytes.

Anoikis, i.e. apoptosis induced by detachment from the extracellular matrix, is thought to be involved in the shedding of enterocytes at the tip of intestinal villi. Mechanisms controlling enterocyte survival are poorly understood. We investigated the role of E-cadherin, a key protein of cell-cell adhesion, in the control of anoikis of normal intestinal epithelial cells, by detaching murine villus epithelial cells from the underlying basement membrane while preserving cell-cell interactions. We show that upon the loss of anchorage, normal enterocytes execute a program of apoptosis within minutes, via a Bcl-2-regulated and caspase-9-dependent pathway. E-cadherin is lost early from cell-cell contacts. This process precedes the execution phase of detachment-induced apoptosis as it is only weakly modulated by Bcl-2 overexpression or caspase inhibition. E-cadherin loss, however, is efficiently prevented by lysosome and proteasome inhibitors. We also found that a blocking anti-E-cadherin antibody increases the rate of anoikis, whereas the activation of E-cadherin using E-cadherin-Fc chimera proteins reduces anoikis. In conclusion, our results stress the striking sensitivity of normal enterocytes to the loss of anchorage and the contribution of E-cadherin to the control of their survival/apoptosis balance. They open new perspectives on the key role of this protein, which is dysregulated in the intestinal epithelium in both inflammatory bowel disease and cancer.

Transcriptional repression by p53 promotes a Bcl-2-insensitive and mitochondria-independent pathway of apoptosis.

p53 can induce apoptosis in various ways including transactivation, transrepression and transcription-independent mechanisms. What determines the choice between them is poorly understood. In a rat embryo fibroblast model, caspase inhibition changed the outcome of p53 activation from standard Bcl-2-regulated apoptosis to caspase-independent and Bcl-2-insensitive cell death, a phenomenon not described previously. Here, we show that caspase inhibition affects cell death commitment decisions by modulating the apoptotic functions of p53. Indeed, in the Bcl-2-sensitive pathway, transactivation-dependent signalling is activated leading to a rapid MDM2-mediated degradation of p53. In contrast, in the Bcl-2-insensitive pathway, p53 is stable and this is associated with transrepression-dependent signalling. A study with microarrays identified these genes regulated by p53 in the absence of active caspases.

Bcl-2 can promote p53-dependent senescence versus apoptosis without affecting the G1/S transition.

With the aim to identify events involved in the determination of p53-dependent apoptosis versus growth arrest, we used rat embryo fibroblasts expressing a temperature-sensitive mutant (tsA58) of the SV40 large tumour antigen (LT). Heat-inactivation of LT leads to p53 activation and commitment to a senescent-like state (REtsA15 cell line) or apoptosis (REtsAF cell line). We report that senescence is associated with high levels of the anti-apoptotic Bcl-2 protein and a cell cycle arrest in G1 phase, whereas apoptosis is associated with low levels of Bcl-2 and a cell cycle arrest in G2 phase. Here we show that Bcl-2, which can inhibit apoptosis and proliferation, turns the apoptotic phenotype into a senescent-like phenotype in G2 phase. This result suggests that Bcl-2-dependent inhibition of apoptosis could be crucial for the commitment to replicative senescence, whereas its ability to inhibit G1 progression would not be required.