Larger amygdala volumes after childhood trauma associated with depression and cortisol response to psychosocial stress in adulthood

Autor:	Buss, Claudia, Pruessner, Jens C., Mayberg, Helen, Mletzko, Tanja, Nemeroff, Charles, Heim, Christine, Li, Hui, Jia, Ning, Su, Qian, Zhu, Zhongliang, Liu, Xiang, Sungkur, Monesh Kumar, Padari, Samjida Majeed, Zhang, Yanyan, Yang, Li, Tang, Guokui, Ouellet-Morin, Isabelle, Wong, Chloe, Danese, Andrea, Pariante, Carmine M., Papadopoulos, Andrew S., Mill, Jonathan, Arseneault, Louise, Michels, Nathalie, Clays, Els, de Buyzere, Marc, Huybrechts, Inge, Vanaelst, Barbara, de Henauw, Stefaan, Sioen, Isabelle, Bublitz, Margaret H., Stroud, Laura R., Walker, Adam, Smith, Robin, Beenders, Bret, Kelley, Keith, Dantzer, Robert, Füchsl, Andrea, Neumann, Inga D., Reber, Stefan O., Bartlang, Manuela S., Slattery, David A., Uschold-Schmidt, Nicole, Kraus, Dominik, Helfrich-Förster, Charlotte, Reberb, Stefan O., Brambilla, Francesca, Serra, Mariarosa, Perini, Giulia, Biggio, Giovanni, Lehrner, Amy, Bierer, Linda M., Bader, Heather N., Makotkine, Iouri, Passarelli, Vincent, Flory, Janine D., Yehuda, Rachel, Sindi, Shireen, Fiocco, Alexandra, Juster, Robert-Paul, Marin, Marie France, Lord, Catherine, Lupien, Sonia J., Lopez-Duran, Nestor L., Giese, Maria, Beck, J., Brand, Serge, Muheim, F., Hatzinger, Martin, Holsboer-Trachsler, Edith, Eckert, Anne, Lischke, Alexander, Gamer, Matthias, Berger, Christoph, Grossmann, Annette, Hauenstein, Karlheinz, Heinrich, Markus, Herpertz, Sabine C., Domes, Gregor, Lange, Claudia, Bermpohl, Felix, Ising, Marcus, Uhr, Manfred, Adli, Mazda, Cave, Sinai, Hirvikoski, Tatja, Nordström, Anna-Lena, Nordström, Peter, Åsberg, Marie, Jokinen, Jussi, Roepke, Stefan, Wingenfeld, Katja, Kuehl, Linn K., Hinkelmann, Kim, Otte, Christian, Kuhlman, Kate Ryan
Jazyk:	angličtina
Rok vydání:	2012
Předmět:	biochemical stress affective go/nogo task daily stress neuroactive steroids pituitary indoleamine 2 3 dioxygenase light-phase immunology TSST stress time of day growth curve modeling social defeat glucocorticoid receptor adolescents panic disorder threat childhood trauma symptoms risk fludrocortisone DNA methylation childhood victimization therapy response lipopolysaccharide heart rate variability adult trauma symptoms PTSD amygdala abuse diurnal pattern salivary cortisol female trauma intergenerational transmission HPA-axis depression head circumference fear visuospatial memory pregnancy psychosocial stress executive functioning young adults acute stress ketamine remitted depression early life stress NBNA hippocampal volume cortisol eye tracking CSC dark-phase children interpersonal violence oxytocin Supplement 1 2012 mouse cognitive impairment mineralocorticoid receptor thyroid hormones Holocaust SERT HPA axis corticosterone aging birth weight prenatal psychosocial stress infant sleep deprivation functional magnetic resonance imaging kynurenine ACTH behaviour traumatic stress BDNF child sexual abuse maternal stress adolescent bullying physiology major depression borderline personality disorder
Zdroj:	European Journal of Psychotraumatology
ISSN:	2000-8066 2000-8198
Popis:	Background Childhood trauma is a major risk factor for the development of affective disorders later in life. We sought to determine whether this risk is linked to neurostructural changes in limbic brain regions after childhood trauma. Methods We recruited 49 medically healthy adult women (28.2±7.1 years of age) from the Atlanta area to include women with/without childhood trauma and with/without major depression (MDD). Childhood trauma exposure was quantified using the Childhood Trauma Questionnaire (CTQ). Lifetime and current diagnoses of MDD and posttraumatic stress disorder (PTSD) were assessed using the Structured Clinical Interview for DSM-IV (SCID). Current depressive symptoms were assessed using the Hamilton Rating Scale for Depression (HAM-D). Magnetic resonance images were acquired, preprocessed, and registered into stereotactic space. Volume analyses of the left and right amygdala were performed using the interactive software package DISPLAY developed at the Brain Imaging Center of the Montreal Neurological Institute, and a standardized segmentation protocol was applied to outline the anatomical boundaries of the amygdala. Total plasma cortisol responses to the Trier Social Stress Test (TSST) were measured. Results When stratifying groups by childhood trauma exposure and MDD, women with both childhood trauma and MDD had largest right amygdala volumes compared to all other groups (interaction effect: F=6.172, p= 0.017). Correlational analyses revealed that higher CTQ scores were associated with larger left (r=0.31, p Statement of the problem Previous animal experiments suggest that prenatal stress affects pregnancy outcomes and impairs cognitive functions of offspring. Our goal was to investigate how prenatal exposure to stressful life events influence pregnancy outcome and infant's physical and neurobehavioral development. Methods A clinical trial was performed. One thousand eight hundred and fifty-six pregnant women were willingly assessed using the Life Events Scale for Pregnant Women (LESPW) before delivery. Those whose score were more than or equal to 375 on LESPW were assigned to higher levels of psychological stress during pregnancy. One hundred and forty-two cases were selected from 1856 pregnant women controlling for variables such as gestational age, maternal age, obstetric complications, socioeconomic status, and trait anxiety. The prenatal stress (PNS) group and the control (CON) group were composed of 71 full-term infants each (1:1 pair matched). Infants’ birth weight (BW) and head circumference (HC) from both groups were assessed at birth and the neonatal neurobehavioral development was evaluated at 72 hours using the neonatal behavioral neurological assessment (NBNA). Results Three hundred and twenty-seven cases from 1856 scored more than 375 on LESPW, incidence of stress was 17.62%. The proportion of undesirable pregnancy outcomes from 327 cases were 147 cases (44.95%), with threatened abortion 38 (11.62%), premature delivery 31 (9.48%), pregnancy complications 73 (22.32%), stillbirth 5 (1.53%), and low birth weight infants 120 (36.7%). The pregnancy outcomes of the non-stressed 1529 cases were undesirable in 579 cases (37.87%), with threatened abortion 123 (8.04%), premature delivery without cause 208 (13.6%), pregnancy complications 240 (15.70%), stillbirth 8 (0.52%), and low birth weight infants 159 (10.4%). BW, HC and the score of NBNA of full-term infants in the PNS were lower than those of the CON (P < 0.05). The score of NBNA of boys was significantly lower with no change in BW and HC in the PNS, and BW and HC of the girls were lower compared to boys in the PNS (P Childhood adverse experiences are known to induce persistent changes in the hypothalamic–pituitary–adrenal (HPA) axis reactivity to stress. However, the mechanisms by which these experiences shape neuroendocrine response to stress remain unclear. We tested whether bullying victimization influenced SERT DNA methylation using a discordant monozygotic (MZ) twin design. A sub-sample of 28 MZ twin pairs discordant for bullying victimization, with data on cortisol and DNA methylation, was identified in the Environmental Risk (E-Risk) Longitudinal Twin Study, a nationally representative 1994–1995 cohort of families with twins. Bullied twins had higher SERTDNA methylation at age 10 compared to their nonbullied MZ co-twins. This group difference cannot be attributed to the children's genetic makeup or their shared familial environments because of the study design. Bullied twins also showed increasing methylation levels between age 5, prior to bullying victimization, and age 10 whereas no such increase was detected in nonbullied twins across time. Moreover, children with higher SERTmethylation levels had lower cortisol responses to stress. Our study extends findings drawn from animal models, supports the hypothesis that early-life stress modifies DNA methylation at a specific CpG site in the SERTpromoter and HPA functioning, and suggests that these two systems may be functionally associated. Rationale/statement of the problem Stress is a complex phenomenon coordinated by several neural systems and has consequently been measured by several biomarkers. Salivary cortisol is the classical used stress biomarker representing the hypothalamic-pituitary-adrenal system. Heart rate variability (HRV), defined as the distance variability between consecutive R peaks, is increasingly used as marker of the autonomic nervous system and as a result also as a stress marker (defined as sympathetic over parasympathetic dominance). Associations between children's salivary cortisol and HRV will be examined. Methods In 190 children (5–10 year) of the Belgian ChiBS study salivary cortisol and HRV were sampled. Salivary cortisol samples were collected when waking up, 30 minutes and 60 minutes after wake-up and in the evening on two weekdays. HRV measurements in supine position were undertaken with Polar chest belts during 5 minutes. Apart from HRV time-domain analysis, also frequency-domain analysis was performed in the low-frequency (LF) and high-frequency (HF) bands. Multilevel growth curve modelling with adjustments for age, sex, physical activity and wake-up time was used to analyse the HRV associations with overall cortisol, cortisol awakening response (CAR) and cortisol diurnal decline. Results Higher overall cortisol levels were negatively associated with mean RR, root mean square of successive differences (RMSSD), percentage of RR intervals differing more than 5 0ms (pNN50) and HF. A steeper diurnal decline was positively associated with normalised LF and the LF/HF ratio and negatively with HF. The CAR was positively associated with normalised LF and the LF/HF ratio and negatively with normalised HF. Conclusion Higher salivary cortisol levels were associated with lower parasympathetic activity. A larger CAR and steeper diurnal decline were associated with a sympathetic over parasympathetic dominance. Consequently, the two main neural stress systems (represented by cortisol and HRV) show good agreement in reflecting children's stress status, although not all parameters were significantly related. Measuring both pathways stays recommended as the pathways might be stimulated differently depending on the stressor. Introduction Maternal hypothalamic–pituitary–adrenal (HPA)-axis functioning is a key mechanism linking stress in pregnancy to adverse neonatal outcomes. Past research has failed to consider whether a woman's history of child maltreatment impacts her stress biology in pregnancy. In this study we assessed whether association between daily stress and diurnal cortisol was moderated by maternal history of child sexual abuse (CSA). Methods Participants were 30 pregnant women (M age=26, SD = 5) who completed a larger study of effects of maternal mood on fetal and infant development. At baseline, women completed a modified version of the Adverse Childhood Experiences Scale. Women were categorized into those who had a history of CSA, non-sexual child abuse (CA), or no child abuse (NA). Women reported the severity of daily stress for 3 days at 20 (SD = 2), 28 (SD = 1), and 35 (SD = 1) weeks gestation by completing a modified version of the Pregnancy Experiences Scale. Finally, women provided salivary cortisol samples at wake-up, +30 minutes after waking, and bedtime on each day of diary collection. Results Twenty-three percentage of women in this pilot study reported CSA (N=7), 47% (N=14) reported CA, and 30% (N=9) reported NA. Hierarchical linear modeling (HLM) analyses were computed to assess whether prior day or same day stress predicted daily cortisol values (adjusting for gestational age at sampling and time of awakening). We found that maternal history of abuse moderated the association between prior daily stress and cortisol at awakening (p=0.07), and 30 minutes post awakening (p=0.006), but not bedtime (p=0.18). As stress the previous day increased, morning cortisol values in women with CSA history decreased, cortisol in NA women increased, and cortisol in CA women showed no change. Maternal history of child abuse did not moderate the association between daily stress and maternal cortisol on the same day (p's > 0.10). Conclusions Results show a dynamic association between daily stressful experiences and diurnal cortisol in pregnancy and suggest that patterns of maternal cortisol following stress differ according to maternal abuse history. These findings have important implications for understanding links between maternal CSA history and adverse neonatal outcomes. Background Inflammation associated with cancer and induced by cancer therapy is associated with clinical signs of sickness behavior that can culminate in the development of symptoms of depression. Intraperitoneal administration of lipopolysaccharide to mice induces depressive-like behavior. Lipopolysaccharide-induced depressive-like behavior is mediated by activation of indoleamine 2,3-dioxygenase (IDO) that degrades tryptophan along the kynurenine pathway and produces kynurenine metabolites such as quinolinic acid that acts as a N-Methyl-D-aspartate (NMDA) receptor agonist. The present study was carried out to determine whether the NMDA receptor antagonist ketamine alleviates lipopolysaccharide-induced depressive-like behavior. Methods Mice were injected intraperitoneally with ketamine or saline immediately before administration of the cytokine inducer lipopolysaccharide or saline via the same route. Their behavior and bodyweight were monitored up to 28 h post-injection. Depressive-like behavior was measured by increased immobility in the forced swim test and decreased sucrose preference for a sucrose solution. Brain, liver and plasma were collected 6 h and 28 h after treatment to measure biomarkers of inflammation. Results Lipopolysaccharide induced the expression of cytokine, IDO, tryptophan 2,3-dioxygenase (TDO) and heme oxygenase-1 at the periphery and in the brain. This effect was not altered by ketamine pretreatment. Ketamine blocked the development of depressive-like behavior but had no effect on sickness behavior measured by body weight loss, reduced food intake and decreased motor activity. Conclusions These data indicate that ketamine is able to abrogate inflammation-induced depressive-like behavior by antagonising the activating effects of kynurenine metabolites on NMDA receptors. Background Chronic subordinate colony housing (CSC, 19d) is an established mouse model for chronic psychosocial stress and causes glucocorticoid (GC) resistance in splenocytes and IL-4 producing peripheral lymph node cells. Here we tested the hypothesis that CSC further causes development of GC resistance at the level of the pituitary gland, which is, in turn, causally involved in the increased plasma ACTH response to prolonged heterotypic stressor exposure (4h of restraint/ shaking) following CSC exposure. Methods Male mice were either exposed to the CSC model or single-housed for control (SHC), in order to investigate changes at the level of the pituitary gland, measured by molecular (Western Blotting, qPCR) and in vitro techniques. Results To exclude that the increased plasma ACTH secretion in response to acute heterotypic stressors is mediated by an increased responsiveness of the pituitary to hypothalamic corticotrophin releasing hormone (CRH) and/or arginine vasopressin (AVP) we first employed Western Blotting to reveal possible changes in pituitary CRH receptor 1 (CRH-R1) and AVP receptor 1b (AVP-R1b) expression. However, CRH-R1 expression was significantly lower, while AVP-R1b expression was unaffected in CSC compared with single-housed control (SHC) mice, arguing against an increased pituitary responsiveness. In order to exclude altered receptor sensitivity we are currently investigating if the in vitroACTH release from whole pituitaries in response to CRH, AVP and CRH+AVP are similar between CSC and SHC mice. Conclusions In line with the hypothesis of a reduced negative feedback inhibition, CSC compared with SHC mice showed a down-regulation of glucocorticoid receptor (GR), mRNA (qPCR) and protein (Western Blotting) expression in pituitary tissue. Although a comparable corticosterone-mediated in vitroinhibition of CRH-induced ACTH release from whole pituitaries between SHC and CSC mice argues against a CSC-induced reduction of negative feedback inhibition we are currently testing this under in vivoconditions employing the dexamethasone suppression test. Recent findings in rats employing repeated restraint stress indicated that the physiological consequences of stressor exposure are strongly dependent on the time of day of stressor exposure. To investigate whether this is also true for clinically more relevant chronic/repeated psychosocial stressors and whether repeated stressor exposure during light- or dark-phase is more detrimental for an organism, we exposed male C57BL/6 mice to social defeat (SD; 2 h) for 19 consecutive days (except day 7 and 14) either in the light-phase between Zeitgeber-time (ZT)1 and ZT3 (SDL mice) or in the dark-phase between ZT13 and ZT15 (SDD mice) and compared them with single-housed control mice in four different experiments. While SDL mice showed a more prolonged increase in adrenal weight and an attenuated adrenal responsiveness to ACTH in vitroafter stressor termination, SDD mice showed reduced dark-phase home-cage activity on observation days 7, 14 and 20, flattening of the diurnal corticosterone rhythm, lack of social preference towards an unfamiliar male conspecific and higher in vitroIFN-γ secretion from mesenteric lymph node cells on day 20/21. In addition, the colitis-aggravating effect of stressor exposure was more pronounced in SDD than SDL mice, indicated by increased body weight loss and inflammatory shortening of the colon following 8 days of dextran sulphate sodium treatment. In conclusion, the present findings demonstrate that chronic/intermittent SD effects on behaviour, physiology and immunology strongly depend on the time of day of stressor exposure. While physiological parameters were more affected by SD during the light-phase, that is, the inactive phase of mice, behavioural and immunological parameters were more affected by SD during the dark-phase. Our results imply that repeated daily psychosocial stressor exposure has a more negative outcome when applied during the dark/active phase. In contrast, the minor physiological changes seen in SDL mice might represent beneficial adaptations preventing the formation of those maladaptive consequences. Grant Support:This study was supported by the Deutsche Forschungsgemeinschaft (DFG FO-207/13-1). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Centrally secreted neurosteroids and peripherally secreted centrally active neurosteroids increase after stressful stimuli modulating anxiety in experimental animals and possibly in humans. Their altered secretions have been reported in panic disorders (PDs) and suggested to be a possible cause or an effect of the panic symptomatology. We have studied in two successive experiments related to PDs the secretions of neuroactive steroids. In experiment 1, we measured the neuroactive steroids in 25 women and 13 men with PD during an interictal phase to see whether their secretion is related to panic symptomatology and whether it changes after successful pharmacotherapy. In experiment 2, neuroactive steroids were examined in normal probands, 42 women and 17 men who reacted with acute panic symptomatology to inhalation of 7% CO2to see whether neurosteroid alterations precede as a cause or follow as an effect of the acute panic attacks, possibly clarifying their pathogenetic mechanism in PDs. In experiment 1, we found that compared to controls, women with PDs had significantly higher levels of progesterone in the midluteal phase of the cycle, of pregnenolone in the premenstrual phase, of 3α,5α-tetrahydroprogesterone in the follicular phase, and of 3α,5α-tetrahydrodeoxycorticosterone in the premenstrual phase, whereas in men with PDs, the plasma concentrations of progesterone and dehydroepiandrosterone were greater than in controls both in a drug-free month and during paroxetine therapy, in both cases correlating significantly with the panic scores. In experiment 2, we found a trend toward a decrease of neurosteroid concentrations during and after CO2inhalation, with no correlations between panic symptomatology and neurosteroid secretion. This would exclude that their possible alterations could be either a cause or an effect of panic attacks. Rationale Holocaust offspring are at greater risk for the development of Posttraumatic stress disorder (PTSD) than comparison subjects, and the increased vulnerability appears to be associated with maternal PTSD. Paternal PTSD is associated with increased vulnerability to depression in Holocaust offspring. In prior studies, neuroendocrine alterations have been observed in Holocaust offspring that resemble those described in association with PTSD and PTSD risk. Holocaust offspring were more likely to show cortisol hypersuppression on the dexamethasone suppression test (DST) in association with parental PTSD even if offspring had not developed lifetime PTSD. These data imply an enhanced glucocorticoid receptor responsiveness that might be associated with PTSD risk, as has been demonstrated in other populations at risk for PTSD, such as soldiers preparing to deploy for combat. More recently, it has been of interest to consider the differential contributions of maternal vs. paternal PTSD on offspring neuroendocrinology. Such work may identify putative epigenetic changes underlying neuroendocrine alterations associated with PTSD risk. Methods The current study examined glucocorticoid responsiveness as reflected by the 0.50 mg dexamethasone suppression test (DST) and the lysozyme stimulation test (LST) in 81 Holocaust offspring, the majority of whom (73%) had two Holocaust exposed parents. The offspring were subdivided based on maternal and paternal PTSD. The LST is an in vitro test of glucocorticoid responsiveness carried out in live cultured lymphocytes exposed to varying doses of dexamethasone (DEX) for which a lower IC50-DEXfor lysozyme inhibition indicates increased GR responsiveness. Results In this sample, there was a significant correlation between the cortisol response to DEX (expressed as cortisol decline from pre- to post-DEX), and the lysozyme IC50-DEX(r= − 0.521, df = 43, p≤0.0005; controlling for age, gender, BMI and DEX levels), reflecting that the cortisol response following ingestion of an oral dose of 0.5 mg DEX resulted in a similar response to that associated with exposing cultured lymphocytes from the same individual to DEX in vitro. There was a main effect [F(1,42) = 4.48, p=0.04] of paternal PTSD on the DST, controlling for age, gender, BMI, DEX levels, and parental Holocaust exposure (i.e., to control for those cases in which a non-PTSD parent was also not a Holocaust survivor). This main effect demonstrated that offspring with only paternal PTSD had evidence of diminished cortisol suppression (non-suppression) compared to the three other groups (offspring without paternal PTSD with or without maternal PTSD, and offspring with both maternal and paternal PTSD). With respect to the IC50-DEXthere was a similar main effect indicating that offspring with paternal PTSD had higher IC50-DEXvalues (M±SD, 6.9 8± 4.04 nM) reflecting diminished glucocorticoid sensitivity [F(1,53) = 4.40, p=0.04], whereas offspring with maternal PTSD (5.40±4.16 nM) had lower IC50-DEXvalues [F(1,53) = 6.81, p=0.01]. In a logistic regression analysis, paternal PTSD predicted offspring lifetime depression (β = 1.40, p=0.032, OR = 4.07) and lifetime MDD (β = 1.25, p=0.048, OR = 3.50), controlling for maternal and paternal Holocaust exposure and maternal PTSD. Interestingly, maternal and paternal PTSD were associated with different self-reported offspring impressions of the effects of their parents’ Holocaust exposure. Maternal PTSD was associated with believing one has psychological scars as a result of being raised by survivor parents, whereas paternal PTSD was associated with believing one has a greater sensitivity to violence and injustice because of parents’ Holocaust experiences. Conclusions The data indicate that maternal and paternal PTSD may contribute uniquely to neuroendocrine alterations and to psychiatric and psychological outcomes in Holocaust survivor offspring. Background Smaller hippocampal volume (HV) and a history of childhood trauma are important risk factors for developing post-traumatic stress disorder (PTSD) in adulthood. However, many individuals experience traumatic events throughout lifespan but do not develop a PTSD diagnosis. Few studies have assessed how these vulnerabilities might correspond with adult trauma symptoms among healthy populations. The goal of this study was to, therefore, assess how childhood trauma and HV are associated with adult trauma symptoms among young and older adults without PTSD. Methods Thirty-two healthy older adults and 28 young adults completed the Trauma Symptoms Checklist and the Childhood Trauma Questionnaire. Magnetic resonance imaging scans were performed and HV measurements were obtained through manual segmentation using a well-validated protocol. Results Multiple regressions computed for each age group showed that in both young and older adults, childhood trauma was not associated with HV. However, in young adults, HV was significantly associated with magnitude of adult trauma symptoms such that young adults with small HV reported more adult trauma symptoms. No significant associations between HV and adult trauma symptoms were observed in older adults. Conclusions These novel results reveal that smaller HV is associated with adult trauma symptoms among healthy young adults, but not older adults. It is possible that small HV supersedes childhood trauma in explaining adulthood trauma symptoms among healthy young adults. Based on these findings, it would be interesting for future studies to investigate how hippocampal-dependent processes might contextualize concomitant trauma symptoms. Background There is accumulating evidence of hypothalamic–pituitary–adrenal (HPA)-axis dysregulation in depressed children and adolescents. For example, depressed children tend to show a pattern of nonsuppression to the Dexamethasone Suppression Test, suggesting atypical feedback sensitivity of the axis. However, evidence linking HPA-axis stress reactivity during laboratory tasks and pediatric depression is more limited and contradictory, with studies showing both blunted and hyperactive cortisol responses to stress in depressed youth. These conflicting findings may been partially attributed to key group differences across studies, such as the age of the sample, depression severity, or history of maltreatment. In this talk, I discuss another key source of variability: individual differences in peak timing, as a factor that can obscure the nature of the depression-HPA-axis relation. I then compare two common modeling approaches (GLM repeated measures and Growth Curve Modeling) with a Functional Data Analysis approach in their ability to account for individual differences in peak times. Methods We examined depressive symptoms as predictive of HPA-axis reactivity in two different studies. The first study involved 65 children ages 6–7 participating in a large longitudinal study of the development of internalizing and externalizing symptoms. These children completed two standard laboratory stress tasks. Saliva cortisol was sampled at 5-minute intervals for 60 minutes. Internalizing symptoms were measured via the parent-completed Child Behavior Checklist. The second study involved 60 children ages 9–16 participating in a study of neuroendocrine functioning in depressed youth. These children also completed a standard laboratory stress task after which saliva cortisol was measured at 10 minute intervals. Depressive symptoms were measured via the Children's Depression Inventory. We then compared three approaches to model cortisol stress reactivity as predicted by internalizing (study 1) depressive symptoms (study 2): (1) a standard repeated measures (RM) approach, (2) a standard growth curve (GC) approach modeling change from the time of the stressor (using pre-stress time as the intercept), and (3) a functional data analysis (FDA) approach using individual peak times as a common anchor. This FDA approach involves the shifting of individual curves horizontally so that all individual peaks are anchored on a common time point. This allows for the estimation of true peaks (intercept) and acceleration towards this peak (slopes) while controlling for individual variability in peak times. Results In both of our studies, the RM and the CG models failed to find a link between internalizing/depressive symptoms and atypical HPA-axis reactivity (pfor internalizing and depressive symptoms in both studies and both approaches >0.20). Specifically, the RM approach did not reveal an impact of depressive or internalizing symptoms on any specific time point. The GC modeling approach did not reveal an impact of symptoms on baseline levels (intercept) or the acceleration from baseline (slopes). However, using a functional data analysis framework we found that youth depression was associated with higher peak levels (anchored intercept) but not with reactivity slopes (pfor internalizing and depressive symptoms impact on the intercept in both studies Background Depression is one of the most prevalent forms of mood disorders. Compelling evidence suggests that mood disorders are characterized by reduced neuronal plasticity, which can be brought about by exposure to stress. Furthermore, there is good agreement in considering key proteins such as the brain-derived neurotrophic factor (BDNF), as a central player for the effects of stress on brain function and plasticity and psychopathological implications. Still, there is a high non-responder rate in antidepressant therapy, which explains the need to find reliable predictors for adequate treatment. Previous studies revealed that plasma and serum BDNF levels in depressed patients were significantly lower than in healthy controls. Since the protein can cross the blood brain-barrier serum content correspondingly correlates with cortical BDNF concentrations suggesting BDNF levels as a promising candidate biomarker for depression and antidepressant treatment response. Methods To investigate the association between serum BDNF levels and treatment outcome, blood was drawn from 28 patients with a major depressive episode (DMS-IV, ICD-10) that participated in a double-blind placebo controlled treatment study. All patients were treated with a stable mirtazapine monotherapy. Partial sleep deprivation (PSD) was performed after one week. Placebo controlled additional morning treatment with the stimulant modafinil to reduce microsleep throughout the day was started during PSD and maintained over two weeks. Serum concentrations of BDNF and cortisol were assessed by an enzyme-linked immunosorbent assay (ELISA) from day 1 (“before PSD”) at 8 am, 2 pm, 8 pm and day 2 (“after PSD”) at 8 am, 2 pm and 8 pm. Samples were appropriately diluted and detection of soluble BDNF or cortisol was carried out in an antibody sandwich format in duplicates and means were calculated for the corresponding group. Moreover, sleep EEG and microsleep episodes were recorded with a portable EEG. Depression severity using the Hamilton Depression Rating Scale and mood, tiredness and relaxation were assessed with visual analog scales (VASs) for psychological functioning at days 1, 2 and 3 (“after recovery night”) as well as after one and two weeks of ongoing treatment. Results Notably, depressive patients who showed an acute HDRS-6 improvement after PSD exhibited a prominent diurnal pattern of serum BDNF levels during the day before PSD whereas acute non-responders did not show such a pattern and BDNF levels were rather constantly expressed. Serum BDNF levels were significantly elevated in acute responders compared to non-responders in the morning at 8.00 am before PSD corrected for Bonferroni (p>0.01). Also responders after two weeks (FU2) exhibited a prominent diurnal serum BDNF pattern before and after PSD on day one and two, while it was more pronounced after PSD. There was no diurnal pattern for non-responders after two weeks before; however, after PSD on day two an even modest diurnal change was visible in this group but less pronounced compared to FU2-responders. We found no association between treatment condition placebo vs. modafinil and response for acute neither response after two weeks. When we linked daily peak BDNF levels from day two at 2 pm with overall HDRS-6 improvement, responders were associated with elevated BDNF levels compared to non-responders on day three after recovery night already. Even after one (FU1) and two (FU2) weeks increased BDNF levels of day two at 2 pm were more prominent in the responder group. This difference between responders and non-responders of peak serum BDNF levels from 2 pm after PSD was statistically significant after two weeks. In addition, HDRS-6 improvement after two weeks of on-going treatment was significantly correlated with elevated serum BDNF levels in all patients. Moreover, peak levels of serum BDNF after PSD on day 2 at 2 pm were correlated with increased relaxation and improved mood in all depressive patients. In addition, placebo treated patients during PSD exhibited a significant increase of serum cortisol levels during PSD when compared to morning peak levels at 8 am between day one and two. There was no serum cortisol increase in the modafinil treated group during PSD intervention. Conclusions Altogether, it seems that a diurnal pattern of serum BDNF during the day is necessarily associated with acute and response after two weeks in terms of partial sleep deprivation independently from additional treatment (modafinil vs. placebo). BDNF levels peaking in the morning and declining during the day seem to be favourable for an antidepressant response. Therefore, BDNF expression profile in serum at baseline could be used as possible predictor for therapy outcome. Background Over the last decade, the neuropeptide oxytocin has attracted considerable attention for its crucial role in social behavior. Motivated by animal studies showing that oxytocin attenuates stress and anxiety in rodents, numerous studies have been conducted to investigate whether oxytocin has similar effects on stress and anxiety in humans. Most of these studies have revealed that oxytocin also attenuates stress and anxiety in humans, presumably by decreasing amygdala activity during the processing of threatening stimuli. However, these studies have almost exclusively been conducted in males, leaving open whether the observed effects can be generalized to females. Methods To investigate how oxytocin affects amygdala-dependent threat-processing in females, we used functional magnetic resonance imaging (fMRI) to measure females’ amygdala reactivity to threatening and nonthreatening faces (study one: n=16) and scenes (study two: n=14) after intranasal application of oxytocin or placebo. We also recorded females’ eye movements during stimulus processing to investigate whether oxytocin-induced changes in amygdala activity are accompanied by corresponding changes in gaze behavior. Results There were no differences in females’ gaze behavior during stimulus processing after oxytocin as compared to placebo administration. However, after oxytocin administration females showed more amygdala activity to threatening faces (study one) and scenes (study two) than after placebo administration. Conclusions Taken together, the present findings suggest that oxytocin enhances amygdala-dependent threat-processing in females, which is in sharp contrast with previous findings showing that oxytocin attenuates amygdala-dependent threat-processing in males. Although these findings point to a possible sexual dimorphism in oxytocin-mediated threat-processing, future studies are warranted to further address this issue, preferably by directly comparing oxytocin effects on threat-processing between males and females. Rationale/statement of the problem Despite numerous studies on the influence of psychosocial stress on hypothalamic-pituitary-adrenal axis (HPA) system responsivity, heterogeneous results have been found with regard to depression in remission. In addition, knowledge concerning cognitive functioning in the remitted state is also narrow showing thus far inconsistent results. The present study investigated the effect of psychosocial stress on the cortisol response and cognitive performance in patients recovered from depression in comparison to healthy controls. Methods Eighty patients who have recovered from depression for at least 6 months (average: 31 months) and 80 healthy matched controls were investigated on the effects of psychosocial stress (TSST) on the performance in an affective go/nogo task. Cortisol responses, behavioral inhibition, reaction time performance and emotional-cognitive functioning were analyzed. We hypothesized that stress vulnerability of cognitive performance is positively correlated to HPA system responsiveness (measured by salivary cortisol) in both healthy subjects and remitted patients but larger in remitted patients compared to healthy controls. Results Thus far, preliminary analyses reveal no abnormal stress-associated HPA system response in patients recovered from depression in comparison to healthy controls. However, remitted patients showed impaired attentional set shifting in the go/nogo task. This impairment was positively correlated with the duration of illness. Conclusion Our study is the first to investigate affective go/nogo task performance and effects of a stress challenge test in patients recovered from depression. Our data demonstrate that attentional set shifting deficits are not only present during acute episodes but also in remission. These deficits seem to be correlated with the duration of illness. Nonetheless, restored stress-associated HPA system function suggests recovery of the HPA system reactivity to psychosocial stress in patients remitted from depression. This in turn suggests that the observed cognitive impairment is not mediated by abnormal HPA responses. Cognitive impairment in the area of executive functioning may be considered a specific trait marker that persists after clinical and neuroendocrinological remission. Background A relationship between exposure to sexual violence and thyroid hormone alterations have been observed among women with premenstrual dysphoric disorder (PMDD), as well as women with posttraumatic stress disorder (PTSD). Women with Borderline Personality Disorder (BPD) report a high estimate of childhood trauma. The aim of the present study was to assess relationships between thyroid hormone measures and exposure to violence in childhood in females with BPD. Methods Ninety-two clinically euthyroid women with BPD diagnosis and at least two prior serious suicide attempts in their history were assessed with the Karolinska Interpersonal Violence Scales (KIVS). The KIVS is a new structured interview, containing four subscales with concrete examples of exposure to violence and expressed violent behavior in childhood (between 6 and 14 years of age) and during adult life (15 years or older). In addition to serum cortisol, baseline thyroid functioning was evaluated by measuring plasma thyroid stimulating hormone (TSH), free and bound Triiodothyronine (T3) and Thyroxine (T4) levels, as well as the FT3/FT4 (free T3/ free T4) ratio, by immunoassays. Results The FT3/FT4 ratio showed a significant negative correlation with exposure to violence as a child. Conclusions Altered thyroid activity, especially FT3/FT4 levels, was associated with exposure to violence in childhood in suicide attempters. Severe childhood trauma-related stress may promote lasting altered thyroid levels and/or contribute to the development of psychopathology associated with BPD traits, coming to the notice of psychiatric care. Background Cortisol is closely associated with memory function via mineralocorticoid and glucocorticoid receptors (MR/GR) in the brain. While GR are expressed throughout the brain, MR is predominantly expressed in the hippocampus. Patients with borderline personality disorder (BPD) often show impaired memory function as well as alterations in cortisol secretion. We have previously shown that hydrocortisone, an unspecific GR/MR agonist, enhances memory function in BPD. To disentangle GR and MR effects on memory processes in BPD, we tested hippocampus-dependent visuospatial memory performance after MR stimulation with fludrocortisone in the current study. Methods In a placebo-controlled, within-subject crossover study, patients with BPD received placebo or 0.4 mg of fludrocortisone orally before memory testing. We used the Rey-Osterrieth Complex Figure Test to measure visuoconstruction, immediate visuospatial memory, and delayed visuospatial memory. Results After fludrocortisone intake, BPD patients (n=17) showed significantly impaired visuospatial memory function compared to the placebo condition (effect of treatment, p=0.01). Conclusions In contrast to the mixed GR/MR agonist hydrocortisone, the MR agonist fludrocortisone impairs hippocampus-dependent visuospatial memory function in patients with BPD. This impairing effect might be mediated by MR-induced negative feedback inhibition of cortisol leading to decreased GR signaling. Background To date there are inconsistent empirical findings regarding the nature of hypothalamic–pituitary–adrenal (HPA)-axis dysregulation in depressed youth. Some of these inconsistencies may be explained in part by exposure to different types of trauma. The purpose of this study is to clarify the interplay between trauma exposure and neuroendocrine reactivity in depressed youth. We hypothesized that exposure to specific subtypes of trauma will moderate the effect of depression on dysregulation of the stress response system. Methods Participants were 51 depressed and non-depressed youth (22 males; mean age = 12.9, SD = 2.8). Participants completed a semi-structured clinical interview, an Early Trauma Inventory (ETI), and a 90-minute stress task which included saliva samples upon arrival, after a 30 minute baseline, 25, 35, 45, 55, and 65 minutes following the SE-Current Procedural Terminology (CPT). Results There were no differences between depressed and non-depressed participants in reported exposure to trauma of any type. However, depressed participants had higher cortisol levels during the regulation phase, 45 [F(15,35) = 2.65, p < 0.05], 55 [F(15,35) = 2.64, p < 0.05], and 65 [F(15,35) = 3.11, p < 0.01] minutes after the stressor. We also found that there was a main effect of depression symptoms [F(11,35) = 296.18, p
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=pmc_________::26abe1c28094ec23da1c0bf7a51c389e http://europepmc.org/articles/PMC3442736 Zobrazit plný text záznamu