Dissecting the genetic basis of transcriptional and translational regulation in heart and liver

Abstract

Gene expression regulation is a multi-layered process and genetic variation can modulate expression levels at various stages, for example by changing the activity of regulatory DNA elements or via nonsense mutations. One layer of regulation for which the genome-wide genetic effects have not been fully assessed yet is translational regulation. The main objective of this thesis is to elucidate the role of genetic variants on transcriptional and translational levels of gene expression. We therefore perform a combinatorial RNA-Seq, Ribo-Seq and genotyping approach to liver and heart tissue of a rat genetic disease model system for spontaneous hypertension and metabolic disease. We use the well-established HxB/BxH rat recombinant inbred (RI) panel that consists of 30 lines that have been derived from a reciprocal cross of the spontaneously hypertensive rat (SHR.Ola) and the normotensive Brown Norway rat (BN.Lx/Cub). The RI lines, which have been previously comprehensively described on multiple levels, enable us to perform quantitative trait loci (QTL) mapping in order to link causal genetic variants to quantitative differences in both transcription and translation. We identify local and distant associations on the transcriptional (eQTL) and the translational (riboQTL) level in each tissue. The majority of detected associations occurred between SNPs and genes that are in close proximity to each other and are termed local QTL. However, we also detect genetic variants that affect gene expression levels over longer distances, termed distant associations. In this thesis, we assessed both types of QTL in detail in order to explain the mechanistic and genetic basis of these QTL. Interestingly, we observe 17 genes regulated by a single distant riboQTL on chromosome 3. To follow-up this finding, we performed an independent RNA-Seq and Ribo-Seq experiment in 2 congenic rat strains that differ specifically in this riboQTL region. The differential expression analysis on both layers of regulation reproduces the results and suggests that the chromosome 3 locus is in control of translation of many additional genes. In hypertensive hearts, we observe global translatome changes that appear protein-length dependent. Translation rates of long proteins are downregulated and shorter proteins, including many ribosomal subunits, tend to be upregulated. As this phenotype is absent from the liver, we hypothesize that the chromosome 3 locus hosts a genetic variant responsible for a cardiac-specific state of translational stress in strained hearts. These data provide insight into the genetic effects on multiple layers of gene expression regulation and show that translational regulation is an important mediator of molecular phenotypes in complex diseases.