Causal influences of testosterone on brain structure change rate: A sex-stratified Mendelian randomization study

Abstract

The impact of testosterone levels on changes in brain structure has been reported. However, it is still unclear which specific brain region could be affected. This study approached Mendelian randomization method to reveal the causal relationship between testosterone levels and the rate of longitudinal structural changes in the brain. The testosterone-related GWAS data were determined from 425,097 European participants. The GWAS data on the rate of longitudinal structural changes in the brain came from the ENIGMA consortium, which included 15,640 all-age participants from 40 longitudinal cohorts. The inverse variance weighted was considered as the main estimate, MR Egger and weighted median methods were used to supplement IVW. A positive correlation was found between total testosterone levels and bioavailable testosterone levels in women and age-independent longitudinal changes in cerebral WM and surface area. The sex hormone-binding globulin levels were found a negative correlation with age-dependent longitudinal structural changes of cortical GM in men. Additionally, we also found that the bioavailable testosterone level in males was negatively associated with the quadratic age-dependent longitudinal change rate in the globus pallidum. We also found estradiol levels and sex hormone-binding globulin levels were negatively associated with the quadratic age-dependent longitudinal change rate of total brain in men. Moreover, we found a positive correlation between total testosterone levels and linear age-dependent longitudinal changes in the hippocampus in both males and females. The testosterone levels in different genders may have varying degrees of causal effects on the structural changes of brain regions. These findings provide evidence for the influence of the brain glandular axis on brain structure, particularly during female brain development.