The Neuroregulatory Effects of Gonadal Steroids in Humans

点击翻译按钮获取中文摘要

This report includes work arising from the following clinical protocols: NCT00026832, NCT00100360, NCT00001177, and NCT00001322. Previous studies have identified the presence of abnormalities in resting state activity and reward processing in mood disorders. Indeed, altered resting state function represents an intriguing possible explanation for the generation of ovarian hormone-related state changes characterized by predictable alterations in social, cognitive (e.g., autobiographical memory), and affective function in women with PMDD. Results from our resting state H215O PET studies in healthy premenopausal women (n = 40) demonstrate effects of both hypogonadism and ovarian steroids on resting rCBF in several brain regions, including the lateral orbitofrontal cortex and medial prefrontal cortex (PFC), subgenual cingulate cortex (SGCC), inferior and superior temporal gyri, sensorimotor cortex, and supramarginal gyrus. Post-hoc analyses in these regions revealed that resting rCBF differed between the progesterone and estradiol add-back conditions for all of these regions. Additionally, comparing progesterone to Lupron-alone conditions, rCBF differences were observed in all of these regions except the sensorimotor cortex, and the supramarginal gyrus. Finally, only the inferior and superior temporal gyri, the primary motor cortex, and the supramarginal gyrus showed significant differences in resting rCBF between the estradiol and Lupron-alone conditions Evidence also suggests that genomic variation in sex-steroid receptors or in the systems regulated by sex steroids may contribute to the ovarian hormone-sensitivity observed in PMDD. Thus, as a first step, we examined the effects of common genetic variations in women during each of the hormonal conditions established within the GnRH agonist-induced hypogonadism and ovarian steroid addback protocol. We have tested specific, common functional variants (BDNF Val66Met and COMT Val158Met) that are regulated by sex steroids and that are known to alter brain function. In our studies, therefore, we employ these gene variants as probes for the impact of genetic substrates on the effects of ovarian steroids on womens brain function. We previously identified that variations in the BDNF gene, which is modulated by ovarian hormones and contains an estrogen response element (ERE), differentially affect working memory-related functions only in the presence of estradiol and only in the hippocampus in healthy women. Similarly, the gene coding for Catechol-O-Methyltransferase (COMT) contains both EREs and progesterone response elements (PREs) in its promoter region, its expression in brain can be regulated by estradiol (and in some tissues by progesterone) and COMT activity is increased by circulating estradiol in women. Indeed, two studies in women reported the modulatory effects on working memory-related brain regions of both estradiol and COMT genotype (Dumas et al. 2018, Jacobs and D'Esposito 2011). To further understand the interactions between ovarian steroids and BDNF/COMT genotypes on brain circuits related to hippocampal and PFC activity, we used oxygen-15 water regional cerebral blood flow (rCBF) positron emission tomography (PET) to study healthy women who participated the GnRH agonist hormone manipulation study. A significant COMT-genotype-by-ovarian-steroids interaction was observed in the right dorsolateral PFC (DLPFC) that reflected a pattern only observed during estradiol addback of a genotype-related step-wise DLPFC activation pattern in which Met homozygotes had the highest activation, followed by the heterozygotes and then the Val homozygotes. We also showed that the modulatory effects of physiologically-relevant allelic variations in sex steroid-regulated genes, specifically COMT and BDNF, on the actions of estradiol are region specific. In the presence of estradiol, the differences between the three COMT genotype was only seen in the DLPFC regardless of the BDNF genotype, whereas the differences in rCBF between the two BDNF genotype was only seen in the hippocampus regardless of the COMT genotype. These data are important for several reasons. First, methodologically, our data suggest that variations in sex steroid-regulated genes such as BDNF and COMT influence a range of brain-related outcome measures. Thus, failure to consider both the effects of genes (and genotype) and reproductive hormonal state in brain imaging studies of women could introduce several confounds and result in a heterogeneous sample of women. Second, mechanistically, these data demonstrate the importance of tissue-specificity in the effects of estradiol on brain function similar to that demonstrated in the rest of the body. Finally, our findings also have important clinical implications both for our understanding of individual differences in the effects of estradiol on cognitive and affective functions and for the biological substrates of the risk for affective disorders in women. Our functional genomic studies in cell lines from both women with and without PMDD have pursued our initial findings of altered function of the ESC/E(Z) gene complex in PMDD as follows: First, neuroprogenitor cells (NPCs) were successfully differentiated from induced pluripotent stem cells (iPSCs) made from women with PMDD and control women (confirmed by immunofluorescent staining and transcriptome analysis). Overlapping gene expression between lymphoblastoid cell lines (LCLs) and NPCs reveals that several genes and gene pathways potentially important for PMDD pathophysiology are expressed in both. Indeed, both LCLs and NPCs show signatures of differential gene silencing through similar mechanisms at baseline and in response to ovarian steroids. These findings in NPCs reveal neuronal differences between women with PMDD and controls both at baseline and after ovarian steroids, including ESC/E(Z) targets. Second, we employed two stock neuronal cell lines: Luhmes Cells, which are immortalized neuronal precursor cell lines (both undifferentiated/neuronal precursors and differentiated/mature dopaminergic neurons), as well as SH-SY5Y, an immortalized neuroblastoid cell line characterized by an adrenergic and dopaminergic phenotype. Both stock neuronal cell lines contained evidence of ESC/E(Z) gene complex function and responsivity to ovarian steroids. Thus, these data provide strong supportive evidence of the neural relevance of our initial findings in lymphoblastoid cell lines (immune-origin tissue) in several more pathophysiologically-relevant tissues (i.e., human NPCs, two stock neuronal cell lines). Finally, we have evidence that the transcriptomic response to ovarian steroid exposure differs between undifferentiated stock neuronal cells, NPCs and that observed in mature, differentiated neuronal cell lines. These latter findings may provide a model to examine the differential effects of ovarian steroids on neuronal measures during development compared with those during adulthood (and further demonstrates the importance of developmental stage and context on the neural effects of sex steroids).