P.2.b.004 Longitudinal associations between BDNF promoter methylation and late-life depression

Abstract

Introduction: Late-life depression is associated with an elevated risk of suicide as well as functional and cognitive impairments, which confer a risk of dementia [1]. Considering these negative consequences of geriatric depression, understanding the etiology of depression is an important step toward early detection and effective treatment of depression in the elderly population. Reduced brain-derived neurotrophic factor (BDNF) function has been suggested as a risk factor for late-life depression. BDNF secretion is influenced by epigenetic (DNA promoter methylation) [2] and genetic (val66met polymorphism) profiles [3]. We investigated the independent and interactive effects of BDNF methylation and val66met polymorphism on late-life depression. Methods: In total, 732 Korean community residents aged 65 years were evaluated, and 521 of them without depression at baseline were followed up 2 years later. The prevalence and incidence of depression were determined using the Geriatric Mental State Schedule, and depression severity was evaluated with the Geriatric Depression Scale. Covariates included age, sex, education, chronic physical disorders, cognitive function, and disability. The effects of BDNF methylation and polymorphism on the diagnosis of depression were investigated using a multivariate logistic regression model, and the relationships between BDNF methylation and depression severity were evaluated using partial correlation tests. Results: Of the 631 participants without depression at baseline, 521 (82.6%) were followed up. Depression was identified 2 years later in 86 (16.5%) of the 521 participants who were followed up. After applying Bonferroni correction, the prevalence of depression at baseline was significantly associated with higher BDNF promoter methylation at CpG site 9 and a higher average methylation percentage. Furthermore, the incidence of depression at the 2-year follow up was significantly associated with higher BDNF promoter methylation at CpG site 9 and a higher average methylation percentage. The results of multivariate logistic regression analysis after adjustment for covariates were that the prevalence and incidence of depression were independently associated with a higher methylation percentage at CpG site 9 and a higher average methylation percentage after applying Bonferroni correction. Partial correlations of BDNF methylation percentages with the GDS scores were summarized as follows. After applying Bonferroni correction, higher baseline GDS scores were significantly correlated with a higher methylation percentage at CpG site 9 and higher average methylation percentage after adjustment for sex, number of chronic medical illnesses, cognitive function, and disability. Additionally, higher GDS scores at follow-up were significantly correlated with a higher average methylation percentage in the same adjusted model. However, there were no significant associations between BDNF promoter methylation percentages and genotype. Furthermore, there were no significant interactive effects of BDNF methylation percentages and genotype on the prevalence or incidence of depression in the multivariate logistic regression model. Conclusions: Higher BDNF methylation was independently associated with prevalence and incidence of depression as well as severe depressive symptoms. No significant methylation-genotype interactions were found. BDNF promoter methylation could be a proxy biomarker for depression late in life.