Variations in serum amino acid (AA’s) levels are associated with several diseases. However, the evidence from epidemiological studies remains conflicting and inconclusive. In this thesis, I aimed to identify genetic variants associated with serum AA’s and investigate genetic and lifestyle-related determinants of circulatory AA’s and their effect on the risk of complex disorders.
I conducted genome-wide association (GWAS) of circulating AA’s including a sample of ~ 117 000 participants of the UK-Biobank (UKB) study, and the data were incorporated in a Mendelian Randomization- Phenome-wide association (MR-PheWAS) framework, using genotypic and phenotypic data of the UKB (n=488 377), to investigate the involvement of circulating AA’s on a wide range of the clinical traits (n=617 with > 200 cases). To further validate the results, I performed MR analysis using the largest available GWAS studies on common complex disorders. I also performed an observational and MR analysis, exploring the effect of modifiable lifestyle risk factors on circulating AA’s levels.
GWAS identified significant SNPs associated with serum AA’s ranging from 48 to 138 SNPs for serum phenylalanine and glutamine, respectively. The total proportion of variance explained for serum AA’s, ranged from 1% to 6%. MR analysis revealed that genetically elevated dietary protein and fat (% of energy) were associated with higher levels of BCAAs, while the carbohydrate showed an inverse association. Higher genetically determined accelerometer-measured physical activity was associated with higher valine levels. Higher genetically determined body fat (%) was nominally related to lower levels of glycine and glutamine. MR-PheWAS analysis and subsequent two-sample MR analysis provided evidence supporting potential causal associations between genetically elevated levels of serum valine and increased risk of type 2 diabetes, and elevated levels of serum alanine and BCAAs with higher levels of total cholesterol, triglyceride, and LDL, and lower HDL levels. These observed associations were successfully replicated using the largest available GWAS studies.
6

This study provides insight into the genomic and phenotypic profile of serum AA’s concentrations, as well as novel findings that can be investigated further to understand the underlying mechanism of abnormal levels of serum AA’s and their role in risk of complex disorders.