Methylation of HPA axis related genes in men with hypersexual disorder (2016)

Printer-friendly version

Jussi Jokinen, Adrian E. Boström, Andreas Chatzittofis, Diana M. Ciuculete, Katarina Görts Öberg, John N. Flanagan, Stefan Arver, Helgi B. Schiöth



  • •Patients with hypersexual disorder had reduced levels of methylation in a locus of the CRH gene.
  • •Patients with hypersexual disorder had higher (TNF)-α levels compared to healthy volunteers.


Hypersexual Disorder (HD) defined as non-paraphilic sexual desire disorder with components of compulsivity, impulsivity and behavioral addiction, and proposed as a diagnosis in the DSM 5, shares some overlapping features with substance use disorder including common neurotransmitter systems and dysregulated hypothalamic-pituitary-adrenal (HPA) axis function. In this study, comprising 67 HD male patients and 39 male healthy volunteers, we aimed to identify HPA-axis coupled CpG-sites, in which modifications of the epigenetic profile are associated with hypersexuality.

The genome-wide methylation pattern was measured in whole blood using the Illumina Infinium Methylation EPIC BeadChip, measuring the methylation state of over 850 K CpG sites. Prior to analysis, the global DNA methylation pattern was pre-processed according to standard protocols and adjusted for white blood cell type heterogeneity. We included CpG sites located within 2000 bp of the transcriptional start site of the following HPA-axis coupled genes: Corticotropin releasing hormone (CRH), corticotropin releasing hormone binding protein (CRHBP), corticotropin releasing hormone receptor 1 (CRHR1), corticotropin releasing hormone receptor 2 (CRHR2), FKBP5 and the glucocorticoid receptor (NR3C1). We performed multiple linear regression models of methylation M-values to a categorical variable of hypersexuality, adjusting for depression, dexamethasone non-suppression status, Childhood Trauma Questionnaire total score and plasma levels of TNF-alpha and IL-6.

Of 76 tested individual CpG sites, four were nominally significant (p < 0.05), associated with the genes CRH, CRHR2 and NR3C1. Cg23409074–located 48 bp upstream of the transcription start site of the CRH gene – was significantly hypomethylated in hypersexual patients after corrections for multiple testing using the FDR-method. Methylation levels of cg23409074 were positively correlated with gene expression of the CRH gene in an independent cohort of 11 healthy male subjects. The methylation levels at the identified CRH site, cg23409074, were significantly correlated between blood and four different brain regions.

CRH is an important integrator of neuroendocrine stress responses in the brain, with a key role in the addiction processes. Our results show epigenetic changes in the CRH gene related to hypersexual disorder in men.


In this study, we found that male patients with hypersexual disorder had reduced levels of methylation in a methylation locus (cg23409074) site located 48 bp upstream of the transcription start site of the CRH gene. Furthermore, this methylation locus was significantly positively correlated with CRH gene expression in an independent cohort of healthy male subjects. To our knowledge, this is the first report on epigenetic changes related to hypersexual disorder. We used genome-wide methylation chips with over 850K CpG sites, however, based on our earlier findings on HPA dysregulation in men with hypersexual disorder (Chatzittofis et al., 2016), we applied a targeted approach on candidate genes of the HPA axis.

CRH is an important integrator of neuroendocrine stress responses in the brain, modulating behavior and the autonomic nervous system (Arborelius et al., 1999), as well as in neuroplasticity (Regev&Baram, 2014). Considering hypersexual disorder in the frame of addiction neurobiology, it is well established that CRH has a key role in the addiction process (Zorrilla et al., 2014). In rodent models, the CRF system drive addiction via actions in the central extended amygdala, producing anxiety-like behavior, reward deficits, compulsivelike drug self-administration and stress-induced drug seeking behavior (Zorrilla et al., 2014). Furthermore, activation of CRF neurons in the medial prefrontal cortex may contribute to the loss of control seen in HD subjects. It has been shown that the chronic drug use leads to a hyperactive HPA-axis with increased ACTH levels while the CRH plays a central role in mediating negative affective responses to stress during drug withdrawal (Kakko et al., 2008; Koob et al., 2014). Similarly, a hyperactive HPA-axis with higher ACTH levels and epigenetic changes in the CRH gene in male patients with hypersexual disorder may lead to a circle of craving and relapse, with a new negative emotional allostatic state, maintaining hypersexual behavior in a futile effort to compensate for a dysphoric emotional state. To repetitively engage in sexual fantasies, urges or behaviors in response to dysphoric mood states and/or in response to stressful life events are key symptoms in the proposed diagnostic criteria of hypersexual disorder (Kafka, 2010). Our findings of hypomethylation of a CRH gene associated methylation locus couple which was associated with gene expression in an independent cohort, adds to the previous findings of HPA axis dysregulation in male patients with hypersexual disorder on a molecular level. Heroin self-administration behavior was associated with differential CRH signaling gene expression partly regulated by methylation shifts in an animal model (McFalls et al., 2016) and promoter methylation has been reported to impact on the expression pattern of CRH (Chen et al., 2012). However, the magnitude of methylation difference in CRH gene locus (cg23409074) was quite low (mean difference approximately 1.60 %), and physiological relevance of subtle methylation changes is not fully elucidated. There is though, a growing body of literature on specific genes, suggesting wide ranging transcriptional and translational consequences of subtle methylation changes (1-5%), especially in complex multifactorial syndromes like depression or schizophrenia (Leenen et al., 2016).

In this study, we took the most relevant confounders, such as depression, DST non-suppression status, CTQ total score and plasma levels of TNF-alpha, into consideration, on the association analyses between methylation of HPA-axis related genes and hypersexual disorder. Interestingly, patients with hypersexual disorder had significantly higher (TNF)-α levels compared to healthy volunteers (Jokinen et al., 2016). Due to the interplay between glucocorticoids and inflammation and the group differences in TNF-alpha and IL-6 levels between patients and healthy controls, we used inflammatory markers as covariates to take into account potential confounding of low grade neuroinflammation. Immune dysregulation is of importance in the pathophysiology underlying several psychiatric disorders including major depression, bipolar disorder and schizophrenia (Danzer et al., 2008). Low grade neuroinflammation is often seen in patients with HPA axis dysregulation (Horowitz et al., 2013) and the inflammatory hypothesis emphasizes the role of psycho-neuroimmunological dysfunctions (Zunszain et al., 2013). It is possible that inflammation and glucocorticoid signaling may act independently on the same structures and processes without direct interaction results in an additive damage effect; in this cohort male patients with HD had higher TNFα levels compared to healthy male volunteers irrespective of HPA-axis dysregulation (Jokinen et al., 2016). As earlier reported (Chatzittofis et al., 2016), antidepressant medication or depression severity were not significantly associated with HPA function measures in this study population.

Further in this study, due to the fact that patients reported more early life adversity compared to healthy controls and the well-known effects of childhood trauma on epigenome, we used early life adversity in the regression models to take into account possible confounding effect of childhood trauma on methylation patterns. HPA-axis dysregulation related to early life adversity reflects vulnerability and the effort for compensation of the effects of childhood adversity (Heim et al. 2008) and early life adversity is related to epigenetic changes of HPA-axis related genes (Turecki&Meaney, 2016).

The conceptualization of the hypersexual disorder has been intensively debated and even though the diagnosis was not included in the DSM-5, the field of study has shown a high degree of reliability and validity for the proposed diagnostic criteria for hypersexual disorder (Reid et al., 2012).

The strengths of the study are a relatively homogenous patient population with thorough diagnostics of the hypersexual disorder, the age matched control group of healthy volunteers, without present or past psychiatric disorders as well as without family history of major psychiatric disorders and severe traumatic experiences. Moreover, the consideration of possible confounders such as childhood adversity, depression, neuroinflammatory markers and dexamethasone test results can be seen as strength.

Some limitations: self-report of early life adversity and the cross sectional design of the study, that does not allow any conclusions about causality. Furthermore, since this is the first study investigating epigenomics in men with hypersexual disorder, it would be of value to replicate our findings in an independent cohort of HD subjects. In addition, while cg23409074 was demonstrated to correlate with gene expression of the CRH gene in healthy controls, it is still not demonstrated to what extent this could reflect modifications occurring in HD subjects and a measure of CRF would have been of value for the study. Further studies are needed to investigate the potential differential expression pattern of CRH in men with HD. An important question is if the whole blood CRH component methylation reflects the effects on the brain. Using a reliable tool to compare methylation between whole blood and brain, the methylation levels at the identified CRH site, cg23409074, were significantly correlated between blood and four different brain regions, with the strongest correlation for prefrontal cortex, a key regulator of stress response. This provides some support that the differential methylation status observed in whole blood can reflect modifications occurring in certain brain regions. Furthermore, the association analysis of methylation and expression was performed in a relatively small group of healthy volunteers and were significant in the robust models, but not by Pearson correlations. This conflicting result could be explained in that robust linear models are recommended to be used in case of a small sample size, to account for any outliers or heteroscedasticity in the data which could bias results (Joubert et al., 2012). In addition, by performing correlation analyses intra-individually, we greatly reduce the likelihood of confounding due to interindividual variance, Other unaccounted potential confounding factors could also induce changes in methylation patterns, e.g. dietary patterns or prandial states (Rask-Andersen et al., 2016) and not controlling for the dexamethasone plasma concentrations during the DST (Menke et al., 2016).

In conclusion our finding of epigenetic state in CRH gene, connecting to the literature on addiction neurobiology, in men with hypersexual disorder, may contribute to elucidating the pathophysiological biological mechanisms of hypersexual disorder.