Identifying gene expression networks in hepatic fibrosis

Since liver fibrosis is a complex trait influenced by susceptibility variants as well as age, sex and environmental covariates, the underlying networks of genetic factors have not yet been defined. Here, we report the first systematic transcriptome analysis of liver fibrosis in a 'genetic reference population' of recombinant inbred lines differing in fibrosis susceptibility. The aim of our study was to identify genetic networks and regulatory mechanisms of gene expression during fibrosis progression. Therefore we generated 96 hepatic expression profiles of fibrotic livers, using Mouse Gene 1.0 ST microarrays (Affymetrix) and hepatic RNA from 32 recombinant inbred (BXD) lines after fibrosis induction by CCl4 challenge. Transcript measures were used as traits and implemented into quantitative trait linkage analysis in order to map loci regulating gene expression during fibrogenesis (eQTL) in the whole genome. We searched for locally regulated genes (cisQTL) within previously mapped regions associated with hepatic collagen contents and histological fibrosis stages. In addition, we refined our search by comparing the regulation of cisQTL in CCl4-treated fibrotic animals to expression data of healthy mice. Overall, we found several differenzially regulated cisQTL, harboring known fibrosis-associated genes such as tenascin C and nuclear receptor Nr1h2 (LXR), providing proof-of-principle of our mapping strategy. The significant correlation of the fibrosis-specific cisQTL to liver phenotypes and the presence of non-synonymous single nucleotide polymorphisms within coding regions of the cis-regulated genes provided further evidence for the relevance of our candidates during fibrogenesis. Our findings demonstrate that eQTL mapping in murine genetic reference populations represents a powerful experimental framework for modeling of networks that regulate gene expression during liver fibrogenesis.