Molecular and computational approaches to identification of genes underlying complex traits

Understanding the genetic architecture of complex traits is of great interest to the biomedical community. HXB/BXH recombinant inbred (RI) strains, derived from the spontaneously hypertensive rat (SHR) and normotensive Brown Norway (BN.Lx), are an important genomic resource for complex trait analysis by means of genetic linkage mapping. The power and accuracy of quantitative trait locus (QTL) analysis critically depends on the quality of the genetic map. To maximize the potential of the HXB/BXH RI strains for complex trait mapping, the latest available genotype information was used to construct a new genetic linkage map. Further, gene expression profiling and biochemical phenotyping in the adrenal glands of the HXB/ BXH rats was performed to address the possible link between the dysregulated catecholamine biosynthesis in the SHR and the development of hypertension. Expression levels and enzyme activities of the two main catecholamine biosynthetic enzymes, Dbh and Pnmt, were found to be regulated from their genic regions (i.e., in cis). Pnmt re -sequencing revealed promoter polymorphisms, which resulted in a decreased response of the transfected SHR promoter to glucocorticoid stimulation. Dbh activity was negatively correlated with systolic blood pressure in RI strains, and Pnmt activity was negatively correlated with heart rate. These heritable changes in enzyme expression suggest primary genetic mechanisms for regulation of catecholamine action and blood pressure control in the SHR. In a separate analysis, genetic determinants of gene expression in the adrenal gland were explored. The adrenal transcriptome assayed via microarrays was subjected to expression quantitative trait locus (eQTL) mapping. Significant clustering of trans-eQTLs was observed, implying that groups of genes are jointly regulated from a single locus. A novel multivariate distance-matrix regression analysis (MDMR) method was applied to identify cis-eQTL genes whose expression profiles strongly correlate with those of the trans-eQTL cluster genes. The resulting genes, Rbm16 and Prp4b, are involved in pre-mRNA processing and as such present leading candidates for further studies aimed at better understanding of the quantitative genetics of gene expression. In conclusion, an important genomic resource was enhanced and then utilized to identify genetic loci controlling key aspects of catecholamine physiology, and differences in global gene expressio