The protocols involved in this project are as follows: 05-M-0059, 81-M-0126, and 92-M-0174.This report includes work arising from the following protocols: NCT00026832, NCT00100360, NCT00001177, and NCT00001322. Behavioral observations show that clinically significant depressive symptoms are rare accompaniments of induced hypogonadism in these healthy premenopausal women. Additionally, neither night-time hot flushes nor disturbed sleep are sufficient to cause depressive symptoms in hypogonadal young women (reported in MH002537-23 DIRP: Psychobiology and Treatment of Perimenopausal Mood Disorders). Thus this paradigm serves as an excellent comparison group for the women with reproductive endocrine-related mood disorders who undergo identical hormone manipulations. We are pursuing our original findings that ovarian steroids modulate prefrontal cortical activity in women by augmenting older gold-standard imaging techniques (i.e., O15 PET), in which the technology is relatively stable over time and the activation task paradigms are kept relatively constant over the long-term course of these studies, with newer hypothesis-driven, cutting edge task paradigms and analytic approaches (i.e., fMRI). During N-back activation, ovarian steroids (or their absence) modulate functional connectivity between the DLPFC and hippocampus. Specifically, hormone condition had significant effects on the direction of hippocampal functional connectivity with both left and right DLPFC: the expected working memory-related negative correlation was observed during estradiol;however, during progesterone there was a positive hippocampal-right DLPFC correlation but there was no significant correlation during ovarian suppression with Lupron alone (i.e., hypogonadism). These data demonstrate that ovarian steroids modulate hippocampal-PFC functional connectivity, consistent with a role for these hormones in modulating the hippocampus at the cellular level. Interestingly, our findings indicate that progesterone and estradiol have opposite effects on this in vivo measure of hippocampal-DLPFC cooperativity, substantiating the complex influence of gonadal steroids in brain circuitry and calling for further clinical and preclinical investigation. In collaboration with Karen Berman, we examine the effects of the presence of common allelic variations in the BDNF gene and in the COMT gene in women undergoing multimodal neuroimaging procedures during each of the hormonal conditions established within the GnRH agonist-induced hypogonadism and ovarian steroid add-back protocol. We examined the impact of the interaction of BDNF Val66Met polymorphism and ovarian hormones on PET measurements of rCBF while women performed the 2-back working memory task and a 0-back sensorimotor control. We observed that BDNF genotype interacts with estradiol to impact hippocampal function during working memory. We found a significant interaction between BDNF genotype and hormone condition in right hippocampal activation (2-back vs. 0-back), and a similar trend in the left hippocampus. Post-hoc analyses revealed Met carriers showed robust estradiol-specific changes, whereas Val homozygous women showed no change in activation across hormone conditions. In Met carriers the hippocampus was abnormally activated (not deactivated) during estradiol but not during the hypogonadal state or progesterone replacement. Analyses of the 0-back and 2-back conditions done separately (an approach made possible by using PET), clearly indicate that these changes were due to neural activity during the working memory (2-back) condition and not the sensorimotor (0-back) control task. This study is the first in humans to demonstrate a BDNF genotype by hormone interaction on a cognitively-related neurophysiological response. These data suggest that the Met allele of the BDNF gene conveys an abnormal sensitivity to the presence of estradiol on hippocampal function, similar to that reported in the Met knock-in mice compared with wild type mice. Results of similar studies in women with PMDD are close to completion pending the recruitment of a sufficient number of Met carriers to repeat this analysis in these women whose affective symptoms are menstrual cycle dependent. Finally, a recent collaboration with Dr. McEwen at the Rockefeller University will explore hippocampal function on both behavioral and molecular levels in humanized met knock-in female mice. In a second analysis, we employed similar methods (i.e., the N-back task) to examine the impact of the COMT Val158Met genotype on PET measured rCBF in the PFC. COMT plays an important role in the regulation of intrasynaptic dopamine levels in the PFC and also has a high affinity for the hydroxylated metabolites of estradiol/estrone (i.e., catecholestrogens). Moreover, the activity of COMT is reported to be both sexually dimorphic and modulated by estradiol (by an estrogen response element ERE in some human tissues). In contrast to the predominance of hippocampal findings with BDNF genotype, results of the COMT by hormone interaction analysis were focused on the DLPFC, a finding that reflected estradiol-related changes in both Val and Met homozygotes. Specifically, the Met homozygotes showed a significant increase in DLPFC activation during estradiol compared with hypogonadism, whereas Val homozygotes showed the opposite pattern with significantly lower activation during estradiol but not hypogonadism or progesterone-replaced conditions). DLPFC activation in heterozygotes was intermediate between the two homozygous groups. These data are consistent with observations that both COMT genotype and stage of the menstrual cycle influence working memory-related activations in the PFC and clearly define a COMT genotype by hormone interaction. Thus, the effects of genotype on PFC function in women cannot be inferred in the absence of knowledge about reproductive state. The mechanisms underlying this observation remain to be defined but are consistent with estradiol impacting PFC dopamine tuning through effects on DA synthesis or metabolism with attendant alterations in DLPFC efficiency. Investigations of the relevance of this genotype by hormone interaction on PFC function to women with PMDD are currently underway. Finally, despite well-established sex differences in several cognitive domains (e.g., visuospatial ability), few studies in humans have distinguished between those sex differences evident in the presence of circulating sex hormones and those sex differences present in the relative absence of circulating levels of estradiol and testosterone, respectively (i.e., true sex differences reflecting organizational differences in brain function occurring independent of differences in current exposure to estradiol or testosterone). We evaluated cognitive performance in healthy men and women before (eugonadal) and during GnRH agonist-induced hypogonadism. The well-documented male advantage in visuospatial performance, which we observed during eugonadal conditions, was maintained despite the short-term suppression of gonadal function in both men and women. Significant main effects of sex (but not of either hormone condition or sex by hormone condition) were observed: men performed better than women on visuospatial tasks (mental rotation, line orientation, Money Road Map, complex figure drawing, embedded figures, and maze completion) during both eugonadism and hypogonadism. No significant main or interactive effects of sex or hormone condition were observed in any other cognitive test domain. These findings suggest that, in humans, sex differences in visuospatial performance do not require the presence of circulating sex steroids and, therefore, are not simply reflections of sex differences in the relative exposures to estradiol and testosterone.
Showing the most recent 10 out of 15 publications
