Unit: Oncogenic Viruses - URA1644 du CNRS

Director: YANIV Moshe

Our research unit studies the interplay between transcription factors, chromatin remodelling complexes and other chromosomal proteins in activating and repressing genes in mammalian cells. Our research examines the importance of several families of transcription factors in controlling cell differentiation and organogenesis, cell growth or apoptosis. We use complete or conditional gene inactivation in mice, micro-array analysis and bioinformatics. Our work has implications for understanding viral and non-viral malignant cell transformation and for deciphering the mechanisms underlying other diseases such as Type II diabetes, kidney polykystic disease or hepatic ductal plates malformation.

HNF1α and HNF1β : development and disease

Group leader : Marco Pontoglio

Other members of the group : Olivier Bluteau, Anna d'Angelo, Antonia Doyen, Evelyne Fischer, Serge Garbay, Lionel Gresh and Andreas Reimann

Organ formation during development is a complex phenomenon whose molecular mechanisms are only partially understood. Hepatocyte Nuclear Factors 1 alpha and beta (HNF1α and HNF1β) are two homologous atypical homeoproteins that appeared during evolution with the first vertebrates. They are expressed in the polarized epithelia of different organs, including liver, kidney, pancreas and intestine, where they control the expression of numerous tissue-specific genes. To gain more insight into the role of these genes during development, we have generated mouse models carrying null mutations in HNF1 genes. Mice lacking HNF1β die in utero at embryonic day 7.5 because of a defect in visceral endoderm differentiation. However, the conditional inactivation of HNF1β has shown that this factor plays a crucial role during organ development. In the liver, HNF1β is essential for bile duct and hepatic artery formation, whereas in the kidney the lack of HNF1β leads to polycystic kidney disease. This renal phenotype is due to the defective expression of PKD2 and PKHD1, two genes whose mutations causes polycystic kidney disease with dominant and recessive inheritance, respectively. Interestingly, in humans, mutations in the HNF1β gene are associated with renal cysts and type 2 diabetes. Inactivation of HNF1α leads to postnatal dysfunctions of liver, kidney and pancreas. HNF1α -deficient mice suffer from phenylketonuria, renal Fanconi syndrome and type 2 diabetes. Fetal liver development requires the presence of, at least, one HNF1 gene. Using DNA oligonucleotide microarray hybridization technology (Affymetrix) we identified several hundred genes whose expression is downregulated in the absence of either of these two transcription factors in the liver. By an in silico approach, we have shown that genes under-expressed in mutant mice have a highly significant enrichment of HNF1 binding sites that are conserved by evolution in different species. Most of these sites are actually bound by HNF1 in vivo. One of our aims is to develop approaches to predict genetic programs controlled by specific transcription factors. Another goal is to understand the exact role of HNF1α and β in the genetic programmes that control liver, kidney and pancreas development and function.

Proto-oncogenes Jun and Fos: transcription factors regulating the cellular stress response

Group leaders: Fatima Mechta-Grigoriou and Jonathan Weitzman

Other members of the group : Damien Gérald, Gaëlle Laurent, Frank Toledo, Aurore Toullec.

The AP-1 transcription factor is central to the cell's ability to integrate multiple extracellular signals and initiate the appropriate genetic programme. AP-1 plays a critical role in regulating the cell cycle and the cellular response to stress. AP-1 is composed of dimers of Jun (c-Jun, JunB or JunD) and Fos (c-Fos, FosB, Fra1 and Fra2) proto-oncoproteins. Our laboratory focuses on unravelling the functions of the different Jun and Fos proteins using genetic and biochemical approaches. We have taken a genomic approach to dissect the complex functional relationship between AP-1 and p53, in order to identify their convergent transcriptional programmes and their roles in tumorigenesis. We discovered a functional cooperation between oncogenic RAS and specific AP-1 heterodimers (c-Jun/Fra-1) in the regulation of the p14/p19ARF tumour suppressor gene, a key regulator of the p53 pathway. Furthermore, we identified novel p53 transcriptional targets that place p53 upstream of the JNK signal transduction cascade, leading to c-Jun phosphorylation and activation. These genomic studies have highlighted the complex convergence of the p53 and AP-1 pathways and their cooperation in determining tumour progression and apoptosis. We have shown that the balance between different AP-1 dimers (Jun/Fos) controls cell cycle progression and the process of oncogenic transformation. In contrast to c-Jun, JunD slows cell proliferation and partially inhibits transformation by Ras. Moreover, JunD reduces tumor angiogenesis by protecting cells from oxidative stress. We have established a new molecular mechanism linking oxidative stress to angiogenesis. junD-deficient mice are viable, but they exhibit premature aging and suffer from age-dependent defects, including cachexia, kyphosis, cataracs, symptoms most probably related to constitutive oxidative stress. They also develop various aggressive cancers that invariably lead to early death. All these symptoms are exacerbated by the deletion of one copy of the c-jun gene, providing the first evidence of redundant functions between c-jun and junD in vivo. Finally, c-Jun and JunD display overlapping functions in heart and vascular development during embryogenesis.

Chromatin, transcription and diseases

Group leader : Christian Muchardt

Other members of the group : Eric Batsché, Yaïr Botbol, Brigitte Bourachot, Marc Lavigne, Bogdan Mateescu.

Opening and closing chromatin is essential for the control of gene expression throughout the life of eukaryotic cell. The SWI/SNF complex is one of the machineries controlling the compaction of chromatin, using the energy of ATP hydrolysis to increase accessibility of histone-associated DNA. During the last year, we have demonstrated that the function of SWI/SNF is not restricted to chromatin remodelling and that this complex also plays a role in the regulation of alternative splicing ot nRNA. This effect on splicing is dependent on the RNA-binding phospho-protein Sam68 and interaction of the SWI/SNF sub-unit Brm with several components of the spliceosome. We have also continued our studies on the HP1 proteins that antagonize SWI/SNF chromatin remodelling. Mainly, we have shown that binding of HP1 to histone H3 tails methylated on lysine 9 can be blocked by two additional modifications, namely phosphorylation of serine 10 and acetylation of lysine 14. These observations provide a better understanding of the mechanisms allowing re-activation of genes repressed by HP1. Currently, we are attempting to get further insights into the mechanisms allowing the SWI/SNF complex to favour inclusion of exons into messenger RNAs. In addition, we are studying the recruitment of HP1 proteins to inducible promoters, using the HIV1 LTR as a model. Finally, we are trying to modify the properties of the HIV1 integrase by coupling it to chromo- and bromo-domains that associate with modified histones.

papillomavirus and cancer

Group leader : Françoise Thierry

Other members of the group : Stéphanie Blachon, Caroline Demeret, Youcef Ben Khalifa, Celia Payen, Sébastien Teissier.

Our research focuses on the mechanisms of malignant conversion of cervical cells infected with Human Papillomavirus (HPV). Cervical cancer is the second cause of mortality in women from cancer worldwide, after breast cancer. It has been shown to be highly associated (more than 90%) with HPV infection and is one of the best examples of virally-induced cancer. The "high risk" HPV types that induce cervical carcinoma, mainly HPV16 or HPV18, specifically infect the genital tract, where they replicate in the upper layers of the epidermis. These viruses encode two oncogenic proteins E6 and E7, which alter the normal proliferative control of the host cell by interfering with two regulatory proteins of the cell cycle, p53 and pRB. Continuous expression of the viral oncogenes is required for maintaining the transformed phenotype. We have shown that transcription of the viral oncogenes is repressed by the viral E2 protein. We used this property of E2 to compare the transcriptome of HeLa cells, a cervical carcinoma cell line associated with HPV18, expressing or not the viral E2 protein in microarray analyses. We were able to define two groups of cellular genes, targets of p53 or E2F, whose expression was specifically modulated in cervical carcinoma cell line. We are now studying these groups of genes in the cervical cancer biopsies by real-time PCR. In addition to its transcriptional function, our recent data indicate that E2 could play a key role in carcinogenic progression through its interactions with cellular proteins. We are deciphering these interactions using tap-tagged E2 proteins.

Legend to Photo :

Anti-angiogenic effect of JunD

(a-b) Representative views of tumors derived from Ras- (Ras) (a) and JunD-overexpressing Ras-transformed fibroblasts (Ras+JunD) (b) after injection into nude mice. Tumors derived from Ras-transformed cells were highly vascularized and hemorrhagic. In contrast, tumors derived from the Ras+JunD cell lines remained pale and poorly vascularized. (c-d) Sections and histological analysis (CD31 immunostaining) of tumors derived from Ras (c) and Ras+JunD (d) cells. Overexpression of JunD reduces the number and the size of blood vessels.

Keywords: oncogens, transcription, chromatin, development, cell cycle, diabetes, renal disease

Activity Reports 2004 - Institut Pasteur

Page Top research Institut Pasteur homepage

If you have problems with this Web page, please write to rescom@pasteur.fr