Our team is interested in the process of cellular DNA replication and how it relates to genome instability. The stability of genetic information relies on proper coordination of replication initiation events, ensuring that all regions of the chromosomes are replicated once and only once per cell cycle.
However, some regions of the genome, notably common fragile sites (CFSs), raise specific problems. CFSs are loci that recurrently exhibit breaks on mitotic chromosomes following perturbation of DNA replication, and are now increasingly regarded as a driving force of oncogenesis. CFSs are indeed involved in the formation of chromosome rearrangements found in tumor cells (deletions, gene amplifications). In addition, some CFSs overlap tumor suppressor genes, whose inactivation further fuels tumor progression. CFSs now appear as preferential targets for oncogene-induced replication stress in precancerous lesions, which suggests that their instability promotes oncogenesis from early stages of the process. Understanding the molecular mechanisms responsible for the instability of these major actors of cancer-genome remodeling is therefore of prime importance.
Nowadays it is admitted that replication stress delays completion of CFS replication more than the rest of the genome, and that breaks occur at under-replicated sequences upon chromosome condensation at mitosis. My team has recently demonstrated that this delay is due to a specific replication program combining late replication with failure to activate origins along the fragile region. Of note, a major consequence of our findings is that commitment to fragility depends on the tissue-specificity of replication programs.
The epigenetic setting of CFSsi lead us to map them, in collaboration with B. Dutrillaux, in a broad variety of human cell types. We found that the map of fragile regions differs markedly between cell types although a limited reservoir of 49 loci accounts for all CFSs we observed. Only a subset of these loci becomes fragile in a given cell type, and their combination varies from one cell type to the other.
It has long been observed that some CFSs co-localize with large genes. We have found that 43 out of the 49 CFSs constituting the reservoir localize to chromosome bands containing genes at least 300 kb-long. Interestingly, two recent reports that have catalogued recurrent focal deletions in large cohorts of human tumors and cancer cell lines. Indeed, re-analysis of the results show that large genes host 51% of recurrent cancer deletions, and that many of these genes are associated with CFSs in at least one of the cell types in which we mapped them. Therefore, a majority of recurrent focal deletions found in human cancers originate from CFSs instable in the cell types from which the cancers derive. Deciphering how large genes impact CFS instability is therefore crucial to our understanding of genome remodeling in tumor cells.
Our future work aims to uncover the molecular mechanisms linking transcription status of large genes to the organization of chromatin loops by their regulatory sequences, and how these features impact the setting of replication origins and CFS instability.