Unit: Nuclear Organization and Oncogenesis - INSERM U.579
Director: Anne DEJEAN
The work in our lab is dedicated to the study of the molecular and cellular mechanisms involved in the development of human cancers with a particular emphasis on the role of certain post-translational modifications. Apart from the potential clinical applications, the study of the events underlying oncogenic transformation is also aimed at revealing novel fundamental biological processes in normal cells. Two distinct but complementary approaches are taken: a first, or 'global' approach seeks to establish a general profile of genetic, epigenetic or post-transcriptional alterations associated with hepatocellular carcinoma (HCC), whereas the second, more 'mechanistic' approach centers on the role of the SUMO pathway in both development and oncogenesis.
Genetics and epigenentics of primary liver cancer (leading scientist: Pascal Pineau)
One of our main research interest focuses on the genetic and epigenetic basis of hepatocellular carcinoma (HCC). To obtain a global overview of chromosmal alterations in HCC, our previous work has employed two complementary approaches: allelotyping and comparative genomic hybridization (CGH). These methods have allowed us to identify the principal chromosomal regions deleted in HCC and thus hypothesized to harbor onco-suppressive activities. Among these are found the chromosome arms 8p and 4q, in which, however, the search for specific tumor suppressors has, as yet, been fruitless. Our subsequent work has nonetheless shown these regions also to be deletion targets in preneoplastic lesions. In order to define specific signaling pathways involved in HCC, we have used homozygous deletion (HD) mapping in cultured tumor cell lines, an approach employed successfully in the search for tumor suppressor genes in recent years. Our study was conducted on 258 human genomic loci in 166 cell lines (with 60 of HCC origin), with the chosen loci corresponding to those of known or suspected tumor suppressors. In total, more than hundred HD affecting 24 distinct loci have been identified. Among these, the genes for INK4A (26% of cases) and for p53 (4%) were expected. However, six other candidates, previously not shown to display HD in HCC, were identified. Among others, these include PTEN, BAX, STK11 and NF2, comprising tumor suppressors involved in as diverse functions as the transmission of oncogenic stress emanating from the AKT or PI3K (PTEN) kinases, programmed cell death (BAX, STK11) or the control of cell growth via membrane-cytoskeletal interactions (NF2). This approach has similarly allowed us to characterize mutations in various other tumor types such as lung, ovarian, head-and-neck cancers. In order to validate this methodology, these studies have been extended to 32 selected HCC samples for deletion of at least one allele of four chosen loci. Mutations in PTEN, for example, were found in 21% (5/24) of European cases in contrast to 0% (0/8) of Asian patients, indicating the value of the HD approach for the identification of candidate tumor suppressors.
In addition, an extensive transcriptome analysis of hepatocellular carcinoma was undertook with the aim of defining the major expression profiles of this heterogenous class of tumors. This should hopefully lead to the establishment of a tumor classification system of significant prognostic and therapeutic value for primary liver cancer. A newly developed approach seeks to establish general and specific profiles of chromatin modifications, both at the histone and at the DNA level. These studies will be complemented by analysis of the anti-tumor potential of histone deacetylase inhibitors (HDACi) both in vitro and in murine liver cancer model systems. In parallel, we carry out a comprehensive characterization of the principal actors of epigenetic regulation (modifying/de-modifying enzymes, modifiers themselves, histone variants) in HCC. Finally, we seek to establish the role of micro-RNAs in HCC tumorigenesis by a global micro-transcriptome' approach as well as by a more mechanistic, candidate-based approach.
Functional compartmentalization of the nucleus and SUMO modification (leading scientists : Jacob Seeler, Oliver Bischof)
A major aspect of acute promyelocytic leukemia pathogenesis involves, at the cellular level, the expression of the PML-RARα hybrid protein and, at the cellular level, the retinoic acid (RA)-reversible disaggregation of the PML nuclear bodies (NBs). This latter observation, that provides a striking physical parallel to the therapeutic effect of RA in this leukemia, has also implicated, for the first time, a novel subnuclear organelle' in a human disease. Study of the signals that may regulate PML NB dynamics led us to identify a novel ubiquitin-like post-translational modification system, the SUMO pathway, as playing a major role in NB dynamics. Unlike ubiquitination, this modification does not lead to the degradation of its target proteins but raher plays important roles in their subcellular localization. Our laboratory and others have identified several novel SUMO susbtrates, a number of which are associated with the PML NBs (see Figure). The SUMO pathway uses an E1 activation enzyme (Uba2/Aos1 heterodimer), an E2 conjugation enzyme (Ubc9) and three families of E3 ligases (RanBP2, PIAS and Pc2) that are believed to confer substrate specificity. Demodification is achived by SUMO hydrolases (SENPs), of which seven isoforms exist in mammals. Given the intimate link between sumoylation and NB dynamics, a significant effort is dedicated to the study of this pathway and its role in the general organization of the nucleus both in normal and pathological cells. A parallel research path seeks to elucidate the role of the PML protein, notably in the cellular senescence process.
To dissect the mechanisms implicated in PML-induced senescence, we used the HPV oncoproteins E6 and E7 that target respectively the p53 and Rb tumor suppressor pathways and found that both pathways are activated and necessary. More interestingly we found that E7 inhibits PML function through physical interaction thus identifying PML as a novel target for E7. To identify novel mediators involved in PML-induced senescence, we established a collaboration with R. Bernard's lab (NKI, Amsterdam) aimed at characterizing new genes whose silencing is able to bypass senescence driven by PML. In addition, we identified the SUMO E3 ligase PIASy as a potent inducer of senescence, making PIASy a good candidate to be an upstream regulator of oncogene-driven senescence in this PML pathway. Consequently, we decided to identify the mediators of both PML- and PIASy- induced senescence using the senescence-bypass shRNA screen strategy. These two complementary screens should hopefully increase our understanding on the still poorly explored role of the PML/PIASy/SUMO network in oncogene-driven senescence and may open the way to the development of novel anti-tumor strategies based on the reactivation of an efficient cellular senescence program.
To clarify the general role of sumoylation in vivo, we have established mouse lines constitutively inactivated for the gene encoding the unique E2 enzyme, Ubc9. Inactivation of the pathway results in early embryonic lethality associated with severe defects in chromosome segregation, nuclear architecture (nuclear envelope dysmorphism) and nucleolar and PML body disorganization. Moreover, the cross-envelope Ran gradient is strongly perturbed, suggesting major defects in nucleo-cytoplasmic trafficking. These observations reveal a major role of Ubc9, and hence sumoylation, in nuclear architecture and function, in chromosome structure and segregation and thus viability in mammals. In collaboration with B. Arcangioli (Institut Pasteur), we have shown that in the fission yeast S. pombe the PIAS-like protein Pli1p acts as SUMO E3 ligase and that cells deleted for pli1 exhibit enhanced mini-chromosome loss, centromeric instability, sensitivity to the spindle poison thiabendazol (TBZ) as well as elongated telomeres. These results, and those of genetic analyses suggest that Pli1p, and by extension sumoylation, plays a major role in protecting heterochromatic repeated sequences (e.g. centromeres and telomeres) from illegitimate recombination. These data open the way to identifying the SUMO targets involved as well as the mechanisms by which sumoylation controls their activity.
Photo legends: PML Nuclear Bodies and SUMO Modification
Keywords: Oncogenesis, PML nuclear bodies, SUMO, cellular senescence, chromatin, epigenetic, leukemia, liver cancer