Epigenetic silencing of Bim transcription by Spi-1/PU.1 promotes apoptosis resistance in leukaemia.

Deregulation of transcriptional networks contributes to haematopoietic malignancies. The transcription factor Spi-1/PU.1 is a master regulator of haematopoiesis and its alteration leads to leukaemia. Spi-1 overexpression inhibits differentiation and promotes resistance to apoptosis in erythroleukaemia. Here, we show that Spi-1 inhibits mitochondrial apoptosis in vitro and in vivo through the transcriptional repression of Bim, a proapoptotic factor. BIM interacts with MCL-1 that behaves as a major player in the survival of the preleukaemic cells. The repression of BIM expression reduces the amount of BIM-MCL-1 complexes, thus increasing the fraction of potentially active antiapoptotic MCL-1. We then demonstrate that Spi-1 represses Bim transcription by binding to the Bim promoter and by promoting the trimethylation of histone 3 on lysine 27 (H3K27me3, a repressive histone mark) on the Bim promoter. The PRC2 repressive complex of Polycomb is directly responsible for the deposit of H3K27me3 mark at the Bim promoter. SUZ12 and the histone methyltransferase EZH2, two PRC2 subunits bind to the Bim promoter at the same location than H3K27me3, distinct of the Spi-1 DNA binding site. As Spi-1 interacts with SUZ12 and EZH2, these results indicate that Spi-1 modulates the activity of PRC2 without directly recruiting the complex to the site of its activity on the chromatin. Our results identify a new mechanism whereby Spi-1 represses transcription and provide mechanistic insights on the antiapoptotic function of a transcription factor mediated by the epigenetic control of gene expression.

Futile caspase-8 activation during the apoptotic cell death induced by DNA damaging agents in human B-lymphoblasts.

Caspase-8 plays an essential role in apoptosis induced by Fas activation. Moreover, caspase-8 can be processed also in response to exposure to genotoxic agents. To decipher the role of caspase-8 in DNA damaging agent (DDA)-induced apoptosis as well as the pathway(s) leading to its activation in response to genotoxic stress, we investigated caspase-8 processing induced by ionizing radiation (IR) or mitomycin C (MMC) treatment in human B-lymphoblasts. Altogether, our observations establish that caspase-8 is actively processed in both receptor-mediated and DDA-induced cell death. However, while Fas-dependent apoptosis absolutely required caspase-8 activity, it is not necessary for completion of the apoptotic program induced by IR and MMC. Experiments performed to understand the molecular pathway(s) of the caspase-8 activation after DDA demonstrated that for both IR and MMC, the Fas/Fas-L interaction is dispensable. Data obtained from caspase inhibitors and from lymphoblasts carrying mutations in ATM and FANCC proteins, involved in DDA response, clearly showed that distinct mechanisms are responsible for caspase-8 activation by IR and MMC in B-lymphoblasts. IR-dependent processing of caspase-8 involves ATM, mitochondrial collapse, FANCC, and caspase-3 activation. Caspase-8 activation by MMC evokes the mitochondrial pathways involving FANCC but not ATM. Collectively, our data indicate that caspase-8 activation is essentially a bystander effect and not a major determinant of the behavior of DDA-exposed cells.

Loss of the Fanconi anemia group C protein activity results in an inability to activate caspase-3 after ionizing radiation.

Fanconi anemia (FA) is a human genetic disease featuring cancer predisposition, genetic instability and DNA damage hypersensitivity. Although abnormalities in DNA repair and cell cycle checkpoint have been proposed as the underlying defect in this syndrome, these hypotheses did not provide full explanations of the complex phenotype. Although not exclusive of such possibilities, alterations in the control of apoptosis might account for the pleiotropic phenotype of this syndrome. We and others have previously reported a deregulation of the apoptotic response to mitomycin C, suggesting that the products of the Fanconi anemia group C protein (FANCC) contribute to the regulation of apoptosis. To explore the functional importance of the apoptotic alterations in FA we analyzed biochemical steps of the execution phase of apoptosis stimulated by another DNA damaging agent, the gamma-ray using FA cell lines derived from complementation group C (FA-C) independent patients. It is shown that the poly(ADP-ribose) polymerase, a target of caspase-3, is not cleaved in FA-C after ionizing radiation (IR). Moreover, caspase-3 is not processed in its active form and, its activity is not increased by IR in FA-C cells compared to normal cells. Altogether, these results demonstrate that loss of the FANCC activity results in a deficiency of the IR-induced apoptosis which is due to an inability to activate caspase-3. Our work suggests that apoptosis signaling induced by mitomycin C and IR is subject to common regulation involving the FANCC protein.

Fanconi anemia C protein acts at a switch between apoptosis and necrosis in mitomycin C-induced cell death.

Deregulation of apoptosis seems to be a hallmark of the Fanconi anemia (FA) syndrome. In order to further define the role of the FA protein from complementation group C (FAC) in apoptosis, we characterized parameters modified during the mitomycin-C (MMC)-induced apoptotic program. It is shown that despite a higher level of cell death for FA compared to normal lymphoblasts after MMC treatment, FA cells do not display a marked DNA fragmentation. Furthermore, while playing a central role in MMC apoptosis of normal lymphoblasts, the activity of caspase-3-like proteases is altered in FA cells. Interestingly, the disruption of the mitochondrial transmembrane potential (Deltapsi), an early event that can lead to apoptotic or to necrotic death, is accomplished similarly in FA and in normal cells. Finally, it is shown that the overexpressed FAC protein inhibited the apoptotic steps, with the exception of the decrease of the Deltapsi. Altogether, our results indicate that the FAC protein acts at a step preceding the activation of the caspases and after the modification of the Deltapsi, a decision point at which cells can be pushed toward either apoptosis or necrosis and which, consequently, regulates the balance between the two modes of cell death.

Role of a mutant p53 protein in apoptosis: characterization of a function independent of transcriptional trans-activation.

Wild-type (wt) tumor suppressor p53 has been implicated in cellular radiosensitivity, mediated by its role in apoptosis and growth arrest. Intriguingly, it was observed that the temperature sensitive (ts) mutant p53val135 protein functions as a positive modulator of cellular radiosensitivity, as evident from acceleration of irradiation-induced apoptosis of M1p53ts (p53val135) cells at the non-permissive temperature; this effect was correlated with acceleration of exit from the G2 checkpoint of the cell cycle. In this work it is shown that the ability of mutant p53val135 to accelerate irradiation-induced apoptosis, at the non-permissive temperature, was devoid of transcriptional trans-activation of p53 target genes. In contrast, the apoptotic function of wt p53val135 was observed to include components which are both dependent and independent of transcriptional trans-activation. Taken together, these observations suggest that mutant p53val135 protein retains the apoptotic component of wt p53 that is devoid of transcriptional trans-activation, and that, although this activity is insufficient to induce apoptosis on its own, it can cooperate to accelerate DNA damage-induced cell death. The results of this work contribute to a better understanding of the complexity of the apoptotic response elicited by wt p53, and highlight the potential role of mutant p53 proteins, as well as trans-activation independent apoptosis, in tumor suppression by irradiation therapy.

Deregulated apoptosis is a hallmark of the Fanconi anemia syndrome.

Fanconi anemia (FA) is a genetic human disorder associated with bone marrow failure and predisposition to cancer. FA cells show poor growth capacity and spontaneous chromosomal anomalies as well as cellular and chromosomal hypersensitivity to DNA cross-linking agents such as mitomycin C (MMC). Because it is likely that disruption of the apoptotic control would lead to such a phenotype, we investigated the implication of apoptosis in the FA syndrome. It is shown that, although demonstrating a high frequency of spontaneous apoptosis, FA cells from four genetic complementation groups are deficient in gamma-ray-induced apoptosis and their MMC hypersensitivity is not due to apoptosis. Fas is a cell surface receptor belonging to the tumor necrosis factor receptor family and is involved in apoptosis. We show that, independently of DNA damage, the alteration in the control of apoptosis in FA concerns also the pathway initiated by Fas activation. Finally, ectopic expression of the wild-type FAC gene corrects the MMC hypersensitivity and anomalies in apoptosis and cell cycle response in FA cells. Altogether, these findings strongly implicate the FA genes as playing a major role in the control of apoptosis. Thus, further studies with FA syndrome will be instrumental toward molecularly dissecting the apoptotic pathways.

The mutagenic processing of psoralen photolesions leaves a highly specific signature at an endogenous human locus.

To assess the role of a given genotoxic agent in the etiology of human cancers, it is useful to establish the mutational specificity of this agent. The aim of this study was to investigate whether the processing of psoralen photolesions, interstrand cross-links (CL) and monoadducts (MA), leaves a specific molecular signature in the mutational events produced at an endogenous locus, HPRT. Human lymphoblasts were treated by 4,5',8-trimethylpsoralen (Me3Pso) in association with a double irradiation protocol (365 plus 365 nm) which allows us to increase the proportion of CL for a given constant number of total photoadducts. The molecular spectrum of mutations at the HPRT locus induced in these conditions was compared to the previously reported spectra of mutations induced by the same psoralen in combination with a single irradiation of either 365 nm (induction of MA and a low proportion of CL) or 405 nm (producing almost exclusively MA). In all treatment conditions, base substitutions constitute the major type of Me3Pso photoinduced mutations. The majority of base substitutions involve a T residue preferably within a 5'-TpA sequence which corresponds to the favoured sites of psoralen photoadducts. In other words, the Me3Pso photolesions induce at the endogenous HPRT locus a high specific signature. Moreover, base substitutions have been essentially found in the non-transcribed strand of the HPRT gene suggesting that the psoralen photolesions are preferentially removed from the transcribed strand. In spite of the considerable difference between the proportion of lesions of both types (CL or MA) induced in different treatment conditions, the kind of mutations and their sequence distribution are similar suggesting that the mutagenic processing of psoralen CL and MA is similar at least for the steps resulting in base substitutions.

p53 involvement in control of G2 exit of the cell cycle: role in DNA damage-induced apoptosis.

DNA damage in proliferating mammalian cells induces a complex cellular response comprising perturbation of the cell cycle and programmed cell death. The relationship between p53-dependent and p53-independent apoptotic cell death, as well as the cell cycle checkpoints induced by DNA damaging agents were explored in hematopoietic cells, using M1 myeloblastic leukemia cells, which are null for p53 expression, genetically engineered M1 variants, expressing p53ts and bcl-2 transgenes, as well as myeloblast enriched bone-marrow cells obtained from wild type p53 (wt p53) and p53-deficient mice. It is shown that gamma-irradiation of M1p53ts cells activated a function of the temperature sensitive mutant transgene p53 (p53ts), promoting increased apoptosis relative to parental, null p53 M1 cells. It is also shown that the kinetics of apoptotic cell death induced by gamma-irradiation correlated with the rapidity of exit from gamma-ray-induced G2 arrest for all the different hematopoietic cell types indicated above. Finally, data has been obtained to demonstrate that, in addition to a role in apoptosis and G1 arrest, wild-type p53 positively modulated the exit from the gamma-ray-induced G2 checkpoint. Taken together, these findings indicate that this new function for p53 is a component of the physiological pathway by which p53 exerts its role in apoptosis.

Dissection of the genetic programs of p53-mediated G1 growth arrest and apoptosis: blocking p53-induced apoptosis unmasks G1 arrest.

Employing the myeloblastic leukemia M1 cell line, which does not express endogenous p53, and genetically engineered variants, it was recently shown that activation of p53, using a p53 temperature-sensitive mutant transgene (p53ts), resulted in rapid apoptosis that was delayed by high level ectopic expression of bcl-2. In this report, advantage has been taken of these M1 variants to investigate the relationship between p53-mediated G1 arrest and apoptosis. Flow cytometric cell cycle analysis has provided evidence that activation of wild-type (wt) p53 function in M1 cells resulted in the induction of G1 growth arrest; this was clearly seen in the M1p53/bcl-2 cells because of the delay in apoptosis that unmasked p53-induced G1 growth arrest. This finding was further corroborated at the molecular level by analysis of the expression and function of key cell cycle regulatory genes in M1p53 versus M1p53/bcl-2 cells after the activation of wt p53 function; events that take place at early times during the p53-induced G1 arrest occur in both the M1p53 and the M1p53/bcl-2 cells, whereas later events occur only in the M1p53/bcl-2 cells, which undergo delayed apoptosis, thereby allowing the cells to complete G1 arrest. Finally, it was observed that a spectrum of p53 target genes implicated in p53-induced growth suppression and apoptosis were similarly regulated, either induced (gadd45, waf1, mdm2, and bax) or suppressed (c-myc and bcl-2), after activation of wt p53 function in M1p53 and M1p53/bcl-2 cells. Taken together, these findings show that wt p53 can simultaneously induce the genetic programs of both G1 growth arrest and apoptosis within the same cell type, in which the genetic program of cell death can proceed in either G1-arrested (M1p53/bcl-2) or cycling (M1p53) cells. These findings increase our understanding of the functions of p53 as a tumor suppressor and how alterations in these functions could contribute to malignancy.

Induction of bax by genotoxic stress in human cells correlates with normal p53 status and apoptosis.

DNA-damaging agents such as ionizing radiation (IR) activate the tumor suppressor p53 and in some cases can cause apoptosis. M1 cells, which do not express the endogenous tumor suppressor gene p53, undergo apoptosis following activation of a temperature sensitive p53 transgene, where it has been shown that bax, an important mediator of apoptosis, is a p53 target gene (Selvakumaran et al, Oncogene 9, 1791-8, 1994). Since p53 can function as a transcription factor after activation by IR, the genetic response to this stress was examined in a panel of human cells with defined p53 status. Like the p53-regulated gene gadd45, bax was rapidly induced, as measured by increased mRNA levels, in the p53 wt (wild type) human myeloid line ML-1, and it was not induced in cells lacking functional p53. However, unlike other p53-regulated genes, bax was only induced in p53 wt cells in which IR also triggered apoptosis. In the case of bcl2, which opposes bax function, mRNA levels were reduced in ML-1 cells after IR. Thus, bax appears to be an unique p53-regulated gene in that its induction by IR not only requires functional p53 but also requires that the cells be apoptosis "proficient."

Induction of p21 (WAF-1/CIP1) during differentiation.

The recently cloned protein, p21 (WAF1/CIP1) is a downstream effector of p53, and mediates growth arrest by inhibiting the action of G1 cyclin-dependent kinases. Since cellular differentiation is frequently characterized by G1 arrest, we examined whether p21 upregulation occurs in differentiation. We show that p21 expression is triggered by multiple differentiation-inducing agents in hematopoietic and hepatoma cells through a p53-independent pathway. The dramatic rise in p21 levels occurs as an immediate early response to differentiation inducers. The induction of p21 is coupled to the expression of early differentiation markers, and is uncoupled from apoptosis. Finally, evidence is presented that p21 expression is uncoupled from G1 arrest in the presence of deregulated c-myc.

Molecular spectrum of mutations induced at the HPRT locus by a cross-linking agent in human cell lines with different repair capacities.

Molecular characterization of mutations photoinduced by a cross-linking agent, 4,5',8-trimethylpsoralen (Me3Pso), in normal human lymphoblasts was conducted in parallel with lymphoblasts derived from Fanconi anemia patients. Such cells have been previously described to be impaired in repair of psoralen photolesions. The endogenous HPRT locus was used as a target gene. The treatment of cells with Me3Pso in combination with 365 nm irradiation leads to the formation of interstrand cross-links, and specific monoadducts. Our analysis revealed that the mutagenic processing of Me3Pso photoadducts in normal human cells results essentially in base substitutions (84%). These are localized to sequences shown previously to be favored for the formation of Me3Pso monoadducts. The mutagenic processing of the same lesions in Fanconi anemia cells results in fewer base substitutions (22%), with deletions (66%) being the predominant class of mutation. In contrast to prokaryotic systems, frameshifts are poorly represented among Me3Pso induced mutations in human cells. In spite of important differences between the kinds of mutations observed in the two cell lines, our analysis reveals similarities in the type of base substitutions and their sequence distribution. In both normal and Fanconi anemia cell lines mutations, mostly targeted on thymine residues, are preferentially located on the non-transcribed strand.

Mutagenic processing of psoralen monoadducts differ in normal and Fanconi anemia cells.

The molecular spectra of mutations photoinduced (405 nm) by 4,5',8-trimethylpsoralen monoadducts (MA), at an endogenous locus, hypoxanthine-guanine phosphoribosyl-transferase (HPRT) in normal and in a Fanconi anemia (FA) lymphoblast cell line, complementation group D, are presented. We show that, in normal cells, MA induce only base substitutions. In contrast, in FA cells which are partially deficient in the incision of MA, deletions are preferentially induced over point mutations (62% of the total). Although the proportion of base substitutions is lower in FA cells, their type and sequence distribution are similar in FA and normal cell lines. The majority of base substitutions are located at sites of psoralen MA which suggest that 4,5',8-trimethylpsoralen photoinduced mutations are targeted and preferentially formed in the non-transcribed strand. Moreover, point mutations induced by MA in normal and FA cells are not homogeneously distributed, they preferentially occur in exon 8 of the HPRT gene. This heterogeneous distribution of mutations is ascribed to processing of MA. Great similarities were found between normal and FA cells with respect to the nature and location of point mutation at the HPRT gene; the high proneness to deletions remains one of the major instability features of FA.

HPRT gene expression differs in mutants derived from normal and Fanconi anemia cells: analysis of spontaneous and psoralen-photoinduced mutants.

Fanconi anemia (FA) is an autosomal recessive disorder characterized by chromosomal instability and abnormalities in the processing of DNA lesions induced by cross-linking agents. We previously reported that after photoaddition of psoralen derivatives the frequency of HPRT- mutants was significantly lower in FA than in normal human lymphoblasts. The hypomutability in FA cells was shown to be associated with an increased deletion frequency at the HPRT gene level. Further characterization of 70 unrearranged mutants (without detectable changes in restriction enzyme fragment length) according to the HPRT gene expression is reported here. Northern blot hybridization analysis demonstrates considerable differences in mRNA phenotyping between normal and FA cells. In normal cells, the minority of spontaneous (31%) and psoralen-induced mutants (0% and 14% according to treatment) arise from mutations that alter the HPRT gene transcription. In contrast to normal cells, in the majority of mutants isolated from FA cells, HPRT gene expression is found to be affected. Indeed a large proportion of either spontaneous (67%) or psoralen-induced (56% and 46%) mutants did not produce detectable amounts of mRNA. These results suggest that the mutagenic processing of spontaneous and psoralen-photoinduced lesions differs in normal and FA cells.

Hypomutability in Fanconi anemia cells is associated with increased deletion frequency at the HPRT locus.

Fanconi anemia (FA) is an inherited human disorder associated with a predisposition to cancer and characterized by anomalies in the processing of DNA cross-links and certain monoadducts. We reported previously that the frequency of psoralen-photoinduced mutations at the HPRT locus is lower in FA cells than in normal cells. This hypomutability is shown here to be associated with an increased frequency of deletions in the HPRT gene when either a mixture of cross-links and monoadducts or monoadducts alone are induced. Molecular analysis of mutants in the HPRT gene was carried out. In normal cells the majority of spontaneous and induced mutants are point mutations whereas in FA deletion mutations predominate. In that case a majority of mutants were found to lack individual exons or small clusters of exons whereas in normal cells large (complete or major gene loss) and small deletions are almost equally represented. Thus we propose that the FA defect lies in a mutagenic pathway that, in normal cells, involves bypassing lesions and subsequent gap filling by a recombinational process during replication.

Fanconi's anemia: genetic and molecular aspects of the defect.

Several features of Fanconi's anemia (FA) are reported: relative inefficiency in the processing of DNA cross-links and monoadducts, hypomutability at the two loci analysed, complementation of the cytogenetic defect by cocultivation with mouse cells and homology of the FA group A with the mouse cellular mutant MCS attempts to clone and characterise a DNA fragment which complements the defect. The relation between observed features and predisposition to leukemia are discussed.