HEPREG · Regulation of gene expression during liver development

Final Activity Report Summary - HEPREG (Regulation of gene expression during liver development)

The main objective of this program was to study the chromatin structure of genes activated during cellular differentiation. Focus was on genes expressed in the liver, including genes involved in metabolic pathways and gene regulation. We studied several genes throughout embryonic and postnatal liver development, in order to test the concept of "poised, or committed" state of active chromatin at the individual gene level early during liver development. In addition, animal models which are deficient for a key hepatic transcription factor, or a key histone modifying enzyme was examined to evaluate the step(s) at which their hepatic targets are de-regulated. Gene expression profiling revealed that 254 genes were up-regulated more than 4-fold during liver development. Detailed analysis of the mechanism of activation of these genes revealed that certain histone modifications serve as early "bookmarks" of genes to be activated during development, which provides support for our original hypothesis. Further evidence for this came from experiments using primary hepatocyte cultures arrested at different stages of the cell cycle. The histone modifications once generated remained stable, suggesting that the epigenetic landscape of promoters confer important marks on genes that persist through cell divisions. Analysis of transcription factor recruitment, including those of DNA-binding factors, general transcription factors and co-regulators, not only confirmed our original "bookmarking hypothesis" but also provided novel insights into the mechanisms of gene activation during development and during various physiological conditions. FISH analysis revealed that bookmarked genes are located at euchromatic nuclear territories, which provide evidence that bookmarking prevents heterochromatinization of genes. Furthermore the data collected in embryonic liver revealed that in mitotic hepatocytes HNF-4a remains associated with the chromatin. This result confirmed our previous ChIP data obtained in cultured hepatoma cells. Genome-wide mapping of nucleosome positions in mouse liver revealed that nucleosomes are only partially lost from the promoters of transcriptionally active genes. The promoters of bookmarked genes are occupied by nucleosomes at the transcriptional start site until they get activated. Upon activation nucleosome loss is observed in less than 40% of genes. Our studies with HNF-4 KO mice revealed that several inactive genes, which are activated by different signalling pathways were found to be bound by HNF-4a. The most interesting ones are metabolic genes, which are induced during fasting conditions. We found that a number of key genes in the fatty acid metabolism pathway, whose expression is elevated in HNF-4a-deficient cells are repressed by Hes6, which is recruited to promoters through interaction with HNF-4a. Of particular interest is the finding that Hes6 is a bona fide HNF-4a regulated gene and that in turn, feedback regulates the expression of its own activator, HNF-4a. The results unravel an intricate, previously unanticipated feed-forward repression mechanism, in which an activator (HNF-4a) induces the expression of a repressor (Hes6) and subsequently recruits the repressor to other target genes of the activator, including its own regulatory region. Our data also suggest that HNF-4a is regulating of other transcription factors involved in metabolic adaptation of the liver to fasting-feeding conditions. Our studies on Set9 KO mice revealed a role of Set9 in the pathway of extracellular matrix biogenesis. This was an unexpected finding that will be further pursued in the future. ChIP analysis revealed that although Set9 can act on specific gene promoters as a histone methylase, its major role is exerted through transcription factor modification instead of chromatin modification. The project also led to the development of novel bioinformatics tools for the computational analysis of ChIP-seq and RNA-seq data.

Data: CORDIS, © European Union