From GWAS to gene : Zebrafish as a model system for large-scale studies of obesity and comorbidities

Sammanfattning: Obesity increases the risk of developing comorbidities such as cardiovascular and metabolic diseases. However, not all obese individuals develop comorbidities, and those that do not are referred to as “metabolically healthy obesity” individuals. These contrast with lean individuals that display the metabolic profile of unhealthy obese individuals and are classified as “metabolically obese normal weight” individuals. The mechanisms underlying the variability in susceptibility to metabolic diseases are not fully understood and suggest the presence of biological and genetic components. This project aims to identify and characterize causal genes in loci identified by genome-wide association studies (GWAS) that are associated with higher adiposity and lower risk of comorbidities, or vice versa.In Study I, I integrated results from several bioinformatic approaches to prioritize candidate genes in loci identified in a genome-wide cross-phenotype meta-analysis of adiposity and cardiometabolic trait pairs. I prioritized 61 candidate genes in 42 of the 62 identifies loci, of which 39 genes were located in 25 novel loci.In Study II, I developed an experimental pipeline and validated 10-day old zebrafish larvae as a model system for image and CRISPR/Cas9-based characterization of candidate genes for obesity. I examined the effect of overfeeding and the effect of CRISPR/Cas9-induced mutations in 15 zebrafish orthologues of 12 established human obesity genes and of 16 human genes with an anticipated role in food intake. I show that 10 days is too early to see an effect of the genetic perturbation on lipid accumulation in adipocytes, but that such experiments can be used to see an effect on other cardiometabolic traits.In Study III, I described a framework to functionally characterize candidate genes in CRISPR/Cas9 founders by targeting the housekeeping kita gene in both mutagenized larvae for a candidate gene and sibling controls. By targeting both cases and controls at kita, the framework ensures that both groups undergo micro-injections, DNA editing, and DNA repair, and that any differences in phenotype can be attributed to mutations in the candidate gene.In Study IV, I applied the approach developed in Study III to examine the effect of mutations in five genes prioritized in Study I and in five additional genes for their effect on adiposity and cardiometabolic traits in crispants. I show that while 10 days is too early to see an effect on adiposity, effects on the cardiometabolic traits the genes were anticipated to affect can be observed.