Bridging the link between trauma, brain development and depression: epigenetic mechanisms

Despite its large impact on the individual and society, we currently have only a rudimentary understanding of the biological basis of Major Depressive Disorder, even less so in adolescent populations. This thesis focuses on two research questions. First, how do adolescents with depression differ from adolescents who have never been depressed on (1a) brain morphology and (1b) DNA methylation? We studied differences in the fronto-limbic system (a collection of areas responsible for emotion regulation) and methylation at the serotonin transporter (SLC6A4) and FK506 binding protein gene (FKBP5) genes (two genes strongly linked to stress regulation and depression). Second, how does childhood trauma, which is known to increase risk for depression, affect (2a) brain development and (2b) SLC6A4 and FKBP5 methylation? Further, (2c) how might DNA methylation explain how trauma affects brain development in depression? We studied these questions in 24 adolescent depressed patients and 21 controls. We found that (1a) depressed adolescents had decreased left precuneus volume and greater volume of the left precentral gyrus compared to controls; however, no differences in fronto-limbic morphology were identified. Moreover, (1b) individuals with depression had lower levels of FKBP5 methylation than controls. In line with our second hypothesis (2a) greater levels of trauma were associated with decreased volume of a number of fronto-limbic regions. Further, we found that (2b) greater trauma was associated with decreased SLC6A4, but not FKBP5, methylation. Finally, (2c) greater FKBP5, but not SLC6A4, methylation was associated with decreased volume of a number of fronto-limbic regions. The results of this study suggest an association among trauma, DNA methylation and brain development in youth, but the direction of these relationships appears to be inconsistent. Future studies using a longitudinal design will be necessary to clarify these results and help us understand how the brain and epigenome change over time in depressed youth.

Glycemic State Regulates Brain-Derived Neurotrophic Factor Responsiveness of Neurons in the Paraventricular Nucleus of the Hypothalamus

Brain derived neurotrophic factor (BDNF) is a member of the family of neurotrophins and binds to the tropomyosin-related kinase B (TrkB) receptor. Like other neurotrophic factors, BDNF is involved in the development and differentiation of neurons. Recently, studies have suggested important roles for BDNF in the regulation of energy homeostasis. The paraventricular nucleus (PVN) is critical for normal energy balance contains high levels of both BDNF and TrkB mRNA. Studies have shown that microinjections of BDNF into the PVN increase energy expenditure, suggesting BDNF plays a role in energy homeostasis through direct actions in this hypothalamic nucleus. We used male Sprague-Dawley rats to perform whole-cell current-clamp experiments from PVN neurons in slice preparation. BDNF was bath applied at a concentration of 2nM and caused depolarizations in 54% of neurons (n = 25; mean change in membrane potential: 8.9 ± 1.2 mV), hyperpolarizations in 23% (n = 11; mean change in membrane potential: -6.7 ± 1.4 mV), while the remaining cells tested were unaffected. Previous studies showing effects of BDNF on γ-aminobutyric acid type A (GABAA) mediated neurotransmission in PVN led us to examine if these BDNF-mediated changes in membrane potential were maintained in the presence of tetrodotoxin (TTX) sodium channel blocker (N = 9; 56% depolarized, 22% hyperpolarized, 22% non-responders) and bicuculline (GABAA antagonist) (N = 12; 42% depolarized, 17% hyperpolarized, 41% non-responders), supporting the conclusion that these effects on membrane potential were postsynaptic. We also evaluated the effects of BDNF on these neurons across varying physiologically relevant extracellular glucose concentrations. At 10 mM 23% (n = 11; mean: -6.7 ± 1.4 mV) of PVN neurons hyperpolarized in response to BDNF treatment, whereas at 0.2 mM glucose, 71% showed hyperpolarizing effects (n = 12; mean: -6.3 ± 2.8 mV). Our findings reveal that BDNF has direct impacts on PVN neurons and that these neurons are capable of integrating multiple sources of metabolically relevant input. Our analysis regarding glucose concentrations and their effects on these neurons’ response to other metabolic signals emphasizes the importance of using physiologically relevant conditions for study of central pathways involved in the regulation of energy homeostasis.