This report includes work arising from the following clinical protocols: NCT00005011, NCT00056901, NCT00059228, NCT00082043, NCT00100360, NCT00001177, NCT00001259, and NCT00001481 Overlapping co-morbidities between premenstrual dysphoric disorder (PMDD) and postpartum depression (PPD) indicate that these disorders could represent a continuum of vulnerability with shared pathophysiology. Indeed, the results of some studies suggest that women with PMDD are at increased risk for developing PPD. These observations have important implications for both the clinical management of women with PMDD (specifically regarding risk counseling) and for research directions. Nonetheless, comorbidity does not necessarily indicate similar pathophysiology, and the comorbidity shared between PMDD and PPD could be uninformative about pathophysiology, analogous to the comorbidity shared between PMDD and non-reproductive depression. We examined the past histories of PPD in a clinic-based sample of women meeting criteria for PMDD. Two hundred and fifteen women who attended the NIMH mood disorders clinic seeking treatment for PMDD and in whom we confirmed the diagnosis of PMDD, were administered the Structured Clinical Interview for DSM. The frequency of PPD (major or minor) was established in the subgroup of women (n=137) who had delivered at least one child. Ninety-three women (43.3%) had a past history of a mood disorder (i.e. either major (MDE) or minor depression). Of these women only 16 (i.e., 11.7%) of the 137 childbearing women with prospectively confirmed PMDD met criteria for a past PPD, not substantially different from the reported incidence of PPD in the general population of women. Our data demonstrate that PMDD and PPD do not frequently co-occur, and do not suggest that PMDD and PPD share similar pathophysiology beyond both being ovarian-steroid triggered mood disorders. In our studies we have developed experimental models for the triggering of symptoms in PMDD and PPD that we employ in our efforts to identify the underlying biology of these conditions. In our gonadotropin releasing hormone (GnRH) agonist-induced ovarian suppression studies (henceforth referred to as the Lupron paradigm) we have demonstrated that the change in ovarian steroid level not the absolute level itself is the trigger for affective destabilization in these women. In addition to our studies on the behavioral effects of changes in ovarian steroids, we employ methodologies to investigate the underlying biological mechanisms of these conditions including neuroimaging studies, as well as both metabolomics platforms for evaluating neurosteroid synthesis, and functional genomics studies examining the effects of a change in ovarian steroid on cellular function. First, our neuroimaging studies have identified both a neural substrate of risk in PMDD as well as a brain-region specific response that could underlie the differential behavioral response to ovarian steroids in this condition. In a previous study, we observed that women with PMDD show abnormal prefrontal recruitment, specifically greater activation than controls throughout the dorsolateral prefrontal cortex (DLPFC) bilaterally. DLPFC activations in PMDD were independent of hormone condition, and correlated negatively with measures of PMDD functional impairment, age of onset of PMDD, and pre-treatment symptom severity measures. These findings suggest an enduring, trait-like predisposition to this hormonally-triggered disorder. In a second completed study, in women with PMDD and controls who underwent resting state PET studies during each of the three hormone conditions in the Lupron paradigm, we observed differences in resting rCBF in women with PMDD compared with controls in the subgenual cingulate (BA25), and medial orbital frontal cortex (mOFC). Higher resting state rCBF in PMDD was present during hypogonadism (when PMDD symptoms are in remission) compared with estradiol or progesterone replacement (when PMDD symptoms are at risk of recurring), whereas no hormone-related changes were observed in controls in either brain region. These data demonstrate for the first time a differential pattern of rCBF within neuroanatomical loci implicated in the process of affective adaptation. Additionally, the fact that the changes in rCBF correspond in time to the hormone state-related changes in symptoms raises the possibility that the changes in rCBF may reflect the altered mood state rather than a direct effect of ovarian steroid hormonal exposure. One possible trigger for the abnormal ovarian steroid-triggered mood state in PMDD is an altered profile of neurosteroid metabolites during exposure to estradiol or progesterone. This possibility is also suggested by our demonstration of the therapeutic efficacy of dutasteride to mitigate symptoms of PMDD since dutasteride inhibits the rate limiting enzymatic step in neurosteroid synthesis. Thus, we next examined the profile of steroid and neurosteroid metabolites in women with PMDD using a metabolomics platform. Metabolomic studies (employing liquid chromatography-tandem mass spectroscopy) are performed in women with PMDD who respond to Lupron with elimination of symptoms and who experience return of symptoms during progesterone or estradiol replacement, as well as controls who experience no change in mood during the identical experimental paradigm. Preliminary data suggest that women with PMDD exhibit differences from control women in a key enzyme involved with steroid (both estrogen and androgen) metabolism, steroid sulphatase. However, in this sample of women, we observed no differences in progesterone-derived neurosteroid levels in plasma. Thus, we plan to expand this study in a larger group of women with PMDD and controls to permit confirmation that no differences in neurosteroid levels exist between women with PMDD and controls (consistent with previous naturalistic studies across the menstrual cycle), and to allow us to examine steroid metabolic pathways in more detail. Our metabolomic and neuroimaging studies are complemented by functional genomic studies, performed in collaboration with David Goldman's laboratory, in which we employ lymphoblastoid and induced pluripotent cells obtained from women with PMDD and controls who participate in our Lupron studies. These experiments allow us to explore the nature of the differential behavioral response by examining gene expression and changes in cellular behaviors associated with the exposure to physiologic levels of either estradiol or progesterone across the two different behavioral phenotypes (i.e., women with PMDD and controls). Results show that women with PMDD express more estrogen receptor (ER) alpha (but not beta) compared with controls. Additionally, RNA sequencing studies show statistically significant, quantitative differences in the PMDD transcriptome compared to control women. We identified significant differences in the expression of the chromatin modifying ESC/E(Z) complex genes, which is of particular interest as a potential candidate since it is regulated by ovarian steroids and leads to genomic regions of transcriptional repression. The mechanism by which these pathways converge and possibly alter CNS function sufficient to manifest in PMDD will be the focus of our future studies in induced neuronal and glial cells.
Showing the most recent 10 out of 24 publications
