In the present study, we evaluated the DNA methylation of the FKBP5 gene in Chinese adolescents using a nested case-control study. We also examined whether there were sex differences in such associations. Our study indicated that there were significant differences between adolescents with and without DS in family structure, family relationships, household socioeconomic status, academic performance, and alcohol consumption. The confounding effects of these variables were considered when exploring the association between DNA methylation and DS. The main findings of the current research included the following. (1) Lower methylation value of FKBP5-12 CpG 1 in adolescents with DS was observed after adjusting social-environmental covariates. (2) Given the social environment influences, hypomethylation of FKBP5 CpG sites was not an independent risk factor for DS. (3) No significant difference between sexes was found in these associations. Besides, these results may help us to better understand the role of FKBP5 DNA methylation in adolescents with DS.
In this study, a global reduction of FKBP5 methylation in adolescents with DS was observed, which is consistent with previous studies examining human whole-blood DNA methylation of the FKBP5 gene in relation to depression [20, 21, 26]. A large-scale study reported an inverse association between the severity of DS and methylation levels in FKBP5, though the relations between methylation and lifetime MDD were not found . In addition, some studies have investigated the correlation between epigenetic modifications in FKBP5 and changes in structure and function in the brain, suggesting influence by childhood adversity, demethylation of FKBP5 altered the structure of relevant brain areas and their functions predisposing MDD [21, 26].
We also included depression-related covariates in our analysis, and the results indicate that the percentage of DNA methylation at one site within FKBP5 remained significantly lower in the DS group than in the healthy control group. The univariate analyses showed that a lower methylation level of the FKBP5 gene might be associated with depressive symptoms. This finding may be explained by the fact that FKBP5 affects the GR function, thereby setting off a short negative feedback loop . It has been demonstrated that the function of the GR is impaired during the development of depression due to the high expression level of FKBP5 that leads to destabilizing the negative response and an increasing level of glucocorticoids called “glucocorticoid resistance” . Glucocorticoids overexpression reduces neurogenesis and synaptogenesis, and causes higher emotional lability directly, which are sensitive to the onset of mental disorders .
The univariable logistic regression analyses also found that adolescents living with a single parent or other relatives, coming from poor relationships and economic status families, were more likely to have depressive symptoms. In addition, those who had poor academic performance or drinking were also at higher risk of developing more depressive symptoms. These also accorded with our earlier observations . Further multivariate logistic regression analysis, after adjustment for the above risk factors, showed that hypomethylation was not an independent risk factor for DS, which indicated that epigenetic factors do not exert influence in isolation with respect to the onset of DS.
Depression is a complex disease involving multiple genetic, epigenetic, and environmental alterations. Our results were consistent with previous reports [23–25]. Weder et al.  conducted genome-wide methylation research among 190 children and also discovered that children’s depression scores were significantly predicted by methylation in CpG sites of FKBP5, but did not reach statistical significance after correcting for whole genome testing. A separate longitudinal study on this issue by Humphreys et al  also suggested that DNA methylation levels of CpG sites within FKBP5 did not predict the onset of MDD in 77 adolescent girls. Comparison of the research, another possible explanation for this might pertain to the characteristics of the study population. Similar to the studies reporting no associations, our participants were adolescents ages 12 to 17, drawing from a school-based, population-representative cohort, while other studies were mostly adult participants and MDD patients that suggested more severe DS compared to school samples.
In contrast to our findings, methylation levels have been reported to be elevated in MDD in other studies [27, 28]. Nevertheless, the research subjects in these studies were MDD patients with serious suicidal ideation or remitted MDD patients, which could lead to substantially different results, and the potential impact of social environment covariates was not taken into account. Although Roy et al.  reported a significant increase in DNA methylation of the FKBP5 promoter region in MDD patients, effects might be only carried by the patients accompanied by serious suicidal ideation. In another study, Höhne et al.  showed a non-significant trend of increased methylation in remitted patients with MDD compared to healthy controls. Knowledge about epigenetic modifications of the gene is constantly updated, however, the clinical significance of hypermethylation remains unknown.
As part of an exploratory analysis, we analyzed the sex differences in the association of methylation status with DS. The results of the present study showed that girls had more depressive tendencies than boys, which is in line with the findings of many previous studies [5, 6]. The ABC (affective, biological, cognitive) model suggests that multiple factors contribute to the sex differences in depression, and any single pathway can only partly explain the variation in depression . The univariate analyses in our study suggested that significant sex differences were detected in the association between environmental risk factors and DS, but not in the association between hypomethylation of FKBP5 and DS in the currently studied sample size. The influence of sex on methylation in depression was also reported by Xia et al. . They recently demonstrated that sex-differential DNA methylation with its regulatory networks had a contribution to the risks of psychiatric disorders, such as autism spectrum disorder, schizophrenia, and MDD. They found hypomethylation in the female with MDD at four CpG sites on DUSP6 genes, however, similar results were not observed in the FKBP5 gene. Nevertheless, the sex differences are intriguing, and more research is still needed in the future to explore the specific mechanisms at play.
To the best of our knowledge, relatively few studies have explored the association between FKBP5 methylation levels and DS in Chinese adolescents. A strength of our study is the epidemiological perspective that highlights these associations observed herein. To better identify the role of DNA methylation, our study considered not only sociodemographic variables but also health-related behaviors at baseline as covariates. Meanwhile, we conducted a preliminary exploration of sex differences in epigenetic alterations, while most studies did not consider such differences.
The current study also had several limitations. First, the study was limited by the modest sample size, which might not be large enough to detect small differences. Second, we only explored the methylation level of the FKBP5 promoter region and did not contain other regions of the gene, such as introns 2 and 7, though the function of the promoter region is also closely related to depression . Third, the relationships between gene expression levels and polymorphisms with DS has historically been lacking with respect to available evidence. Finally, the current investigation reports the DNA methylation levels in the blood, which may differ from methylation patterns in other tissues like the brain. There is emerging research, however, suggesting that DNA methylation in the blood biospecimens may serve as a systemic marker and substitute for methylation in brain tissue [46, 47]. In future research, multi-gene effects on the HPA axis signal pathway and environmental stressors, especially adverse early experiences, should be considered towards a better understanding of gene-environment interactions.