Assessment Committee
Prof. Bjørn H. Ebdrup, Centre for Neuropsychiatric Schizophrenia Research, Copenhagen University (chair)
Associate Prof. Charlotte Cecil, Erasmus MC University
Associate Prof. Christiane Gasse, Aarhus University Hospital, Psychiatry

Academic Advisors
Principal Supervisor: Prof. Vibe Gedsø Frøkjær, NRU and Copenhagen University
Primary Co-supervisor: Associate Prof. Patrick MacDonald Fisher, NRU and Copenhagen University
Co-supervisors: Prof. Gitte Moos Knudsen, NRU and Copenhagen University; Prof. Klaus-Peter Lesch, University Hospital Würzburg

Thesis Summary
Serotonin transmission crucially regulates a variety of brain functions, including neurodevelopment, cognition, mood and stress responses. The serotonin system is also the main target of
antidepressant medications and alterations in serotonin function have been reported in relation to psychopathology e.g. depression, although the link between serotonin (dys)function and depression is still unclear.
Genetic and epigenetic variation (e.g. DNA methylation) within the serotonin system has been suggested to be relevant for
depression risk. However, it is unknown to what extent this variation affects human serotonin neurotransmission, in the healthy and in the pathological (e.g. depressed) state.
Specifically, DNA methylation of serotonin-relevant genes (e.g. serotonin transporter [SLC6A4], tryptophan hydroxylase 2 [TPH2]) has been proposed as a promising biomarker of gene-environment interactions, that could be used to reflect adversities experienced in early life, predict individual risk for depression and likelihood to respond to antidepressant treatment as well as to inform on underlying mechanisms. However, findings are mixed and replication of previous observations is strongly needed to fully uncover the biomarker potential and clinical implications of this epigenetic modification.
In addition, DNA methylation, which is a tissue-specific modification, is generally measured in DNA from peripheral blood cells. Nonetheless, the link between peripheral DNA methylation of serotonergic genes and brain proxies for serotonin neurotransmission measured in-vivo (e.g. serotonin transporter, 5-HTT; serotonin 4 receptor, 5-HT4) is unexplored.
The main goals of this thesis were to characterize how genetic and epigenetic variation within genes relevant for serotonin function can shape in-vivo serotonergic neurotransmission, both in the healthy and in the depressed state, and to gain a better understanding of how DNA methylation of serotonin-related genes can be used as a biomarker in the context of depression and antidepressant treatment.
In Study I, we 1) examined the association between a set of genetic variants within five serotonin-relevant genes and brain 5-HTT levels in healthy adult participants and 2) evaluated whether genetic variation per se could predict brain 5-HTT levels. In Study II, we evaluated whether peripheral epigenetic variation within the SLC6A4 and TPH2 genes was associated with 1) 5-HTT or 5-HT4 brain levels in healthy adults or 5-HT4 in patients with depression and/or 2) with measures of early life and recent stress, depressive and anxiety state traits in healthy participants and patients with
depression. In Study III, we evaluated whether SLC6A4 and/or TPH2 methylation 1) predicted clinical outcomes following antidepressant treatment and/or 2) changed following antidepressant treatment. In Study IV, we tried to replicate the link between SLC6A4 and/or TPH2 and depression status and childhood trauma that was described in literature using data from four large datasets (three based on blood and one on postmortem brain samples). Next, we moved beyond these two genes by examining whether DNA methylation at 27 centrally-relevant genes for serotonin function is enriched in depression or childhood trauma and, in one of the cohorts, if it is associated with: 1) depressive symptoms, 2) childhood trauma or 3) depression chronicity.
In Study I, we found that individuals carrying the T-allele of the rs1137070 variant in the monoamine oxidase A gene (MAOA) had increased 5-HTT binding but, despite this association, genetic
information was not sufficient to predict 5-HTT brain levels. In Study II, we found no link between SLC6A4/TPH2 methylation measured in blood and brain 5-HTT or 5-HT4 levels, nor with measures of environmental stress, depressive or anxiety state symptoms. In Study III, we found that patients with higher baseline TPH2 methylation levels were more likely to respond to treatment after 8 weeks of treatment with SSRI. However, neither SLC6A4 nor TPH2 methylation could predict clinical outcomes following antidepressant treatment and only marginal changes in their methylation levels were observed over 12 weeks of treatment. In Study IV, we found no evidence for an association between DNA methylation of neither gene and depression status, childhood trauma or depression chronicity in four independent datasets.
The findings from these studies suggest that genetic and epigenetic variation within the serotonin system as captured in peripheral blood might have limited impact on in-vivo serotonin
neurotransmission, both in healthy participants and in patients with depression. In addition, peripheral DNA methylation of serotonin-relevant genes is unlikely to be used as a biomarker for
neither depression risk nor antidepressant treatment outcomes, and their DNA methylation levels are not associated with depression status or childhood trauma when measured peripherally nor with depression status in postmortem brain tissue.
Taken together, this thesis shows valuable insights into the interpretation of genetic variation and peripheral DNA methylation in serotonin-relevant genes in relation to serotonin neurotransmission and suggests that DNA methylation of serotonin-relevant genes unlikely provides critical insights into
mechanisms underlying depression or clinical outcomes after antidepressant treatment or represents a clinically useful biomarker of depression status or early life adversities.
​