Genome-wide association study (GWAS) for depression in an Andalusian epidemiological sample.Relationship between body mass index, physical activity and depression

Abstract

Depression is a highly prevalent mental disorder with devastating consequences in the general population, posing a public health concern worldwide. The increased mortality associated with depression is largely related to its frequent comorbidity with physical illnesses and other mental disorders, aggravating patients' prognosis and complicating their treatment. Furthermore, depression is a genetically complex disorder, with multiple common genetic variants involved in its aetiology, and it is the aggregation of risk alleles that confers a certain genetic predisposition. The general aim of this Doctoral Thesis is to investigate the genetic differences between individuals with depression and controls in an adult epidemiological sample, representative of the general Andalusian population (the PISMA-ep study), and to identify potential interactions between the genetic background of an individual and environmental factors related to physical health. Chapter I aims to conduct a genome-wide association study (GWAS) for depression in an subsample of an epidemiological cohort representative of the general adult population of Andalusia from the PISMA-ep study. Although no variable was found to reach statistical significance at the genome-wide level, 9 genetic risk variants were found in the "grey zone" and were analysed. The construction of a polygenic risk score (PRS) allowed us to establish an association between depression prevalence in the PISMA-ep subsample and a weighted set of genetic variants that were identified in the largest depression GWAS meta-analysis to date. In Chapters II and III, the focus shifts to the relationship between depression and physical health, focusing on the role of body mass index (BMI) and physical activity, respectively. Chapter II aims to investigate the relationship between depression and BMI as an indicator of physical health, because of its relationship with obesity. To this end, firstly, a systematic review of the most studied polymorphism of the FTO gene (rs9939609, classically associated with an increase in BMI) was carried out to elucidate the potential relationship between this genetic variant, depression and BMI, although we highlighted the need for more methodologically homogeneous studies to obtain conclusive results. Subsequently, a new unweighted genetic risk score (GRS) was constructed from the sum of risk alleles from variants of 30 candidate genes for depression in a Spanish epidemiological cohort (PredictD-CCRT). In addition to being associated with depression, we found an improvement in the predictive ability of the model when nongenetic risk factors were included, and in particular an interaction between BMI and the GRS was observed. Chapter III aims to explore the relationship between physical activity and depression. To begin, we conducted a systematic review of BDNF and its relationship with depression and physical activity, both of its most studied polymorphism (rs6265, Val66Met) and of the levels of the protein. Despite the need for consensus that was highlighted, the results suggested a transient increase in the protein as an acute effect of physical activity, as well as a greater antidepressant effect due to exercise reported in carriers of the risk allele (Met) of the variant studied. In the following study, we sought to analyse whether this BDNF polymorphism was related to depression and physical activity in the PISMA-ep study. In both the total sample and in women, we observed the effect described in the systematic review, i.e. a decrease in the prevalence of depression in people carrying the risk allele, which became more pronounced as the reported weekly hours of physical activity increased. Chapter IV integrates the insights from the previous chapters by constructing a PRS for depression in a phenotypically characterised PISMA-ep subsample comprising BMI and physical activity information. Using the summary statistics of the largest GWAS meta-analysis to date, it was observed that a higher polygenic risk was associated with a higher prevalence of depression in the subsample analysed. Furthermore, the addition of non-genetic variables, such as not being physically active or higher BMI, improved the predictive ability of the model and its association with the prevalence of depression. The results described in this Doctoral Thesis contribute to the knowledge about the genetic architecture of depression, by means of a GWAS study and PRS constructions. Furthermore, they provide evidence that, for a better prediction of depression risk, incorporating variables related to physical health such as BMI and physical exercise would be particularly valuable.