In our earlier studies of regional cerebral blood flow using PET, we reported for the first time in humans that induced hypogonadism was associated with the elimination of the normal pattern of cortical activation in the dorsolateral prefrontal cortex as well as in the posterior inferior temporal cortices and the inferior parietal lobule;whereas both estradiol and progesterone replacement restored the normal pattern of cortical activation during a working memory task. 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). We employ several different activational paradigms during the neuroimaging procedures that will allow us to both vary the cognitive load of the task (and possibly the level of stress involved in performing the task) and test neural circuits more relevant to the phenomenology of affective disorders (i.e., reward system). For example, using fMRI and parametrically varied working memory load (i.e., N-back test), we are testing the possibility that change in hormonal state causes a shift in the dose-response curve between cognitive load and neurophysiological response. Additionally, using the N-back task with two functional imaging methods (PET with repeated scans and fMRI), we are determining the activation pattern for each woman at each hormonal phase to test for variance in their sites of activation. Finally, using new analytical approaches, we are testing the hypothesis that the hormonally-induced alteration in function includes changes in interregional functional connectivity either in the resting state or during activation. Preliminary results from our O15 PET studies, demonstrate that estradiol significantly increased resting regional cerebral blood flow (rCBF) activity in the subgenual cingulate (BA25), medial prefrontal cortex (mPFC), posterior cingulate, and parahippocampal gyrus compared with the hypogonadal state;whereas, progesterone significantly increased resting rCBF in the mPFC and putamen compared with the hypogonadal state. The results of functional connectivity analyses investigating the effects of hormone condition on the covariance of BA25 with the default resting network are pending. Our findings that estradiol regulates activity within BA25 have implications for understanding the potential impact of estradiol on mood. The subgenual cingulate (BA25) has been implicated in the modulation of emotion and affective adaptation in humans. Neuroimaging studies have documented both structural and functional abnormalities in BA25 in depression, and differences in the activity of BA25 are proposed to underlie some of the therapeutic benefits of deep brain stimulation in depression. Finally, resting-state BA25 connectivity with the default resting network is significantly more robust in patients with mood disorders. Future initiatives will be determined by the results of our current studies examining the effects of gonadal steroids in isolation on regional cerebral blood flow and, by the successful importation of relevant emotion-imaging paradigms (e.g., reward paradigm) for investigating the neurocircuitry of reproductive endocrine-related mood disorders. Finally, recent data has documented the influence of genetic background on cognitive performance and affective adaptation. Thus the imaging studies performed in the GnRH agonist paradigm will provide a unique opportunity to examine the interactions between reproductive hormones and specific genes regulated by ovarian steroids (e.g., catechol-o-methyl transferase and brain-derived neurotrophic factor) on measures of cognitive performance, affective adaptation, and task-activated regional cerebral blood flow. Additionally, in a related project, we demonstrated changes in reward-related neurocircuitry across the normal menstrual cycle with increased activations during the expectancy of the reward during the follicular phase, when estradiol levels are high. These data demonstrate, for the first time in humans, that ovarian steroids modulate reward system function, with increased follicular phase activation of the orbitofrontal cortex and amygdala during reward anticipation and of the midbrain, striatum, and left ventrolateral prefrontal cortex (VLPFC) during reward delivery. We have further investigated these finding across the menstrual cycle by employing the reward paradigm in women who are participating in the GnRH agonist-induced hypogonadism study in which we examine the effects of estradiol and progesterone separately compared with a hypogonadal state. Preliminary findings demonstrate that both estradiol and progesterone modulate reward-related neurocircuitry. During the anticipation of reward, estradiol increased activity in the amygdala and orbitofrontal cortex compared with the hypogonadal state (consistent with our previous data across the menstrual cycle), and progesterone increased activity within the right inferior frontal gyrus compared with both estradiol and hypogonadism. These data suggest a substrate in which changes in gonadal steroids could modulate affective state, and are consistent with preclinical studies that document the modulatory effects of ovarian steroids on reward-related behaviors and brain regions, as well as a growing literature demonstrating the importance of the reward system in depression. We are awaiting completion of the analyses of possible diagnostic differences (i.e., PMD versus controls) in ovarian steroid-related alterations of reward neurocircuitry. In our previous studies of the effects of gonadal steroids on stress responsivity, we identified that progesterone rather than estradiol has a greater impact on HPA axis activation in humans, unlike in rodents, but the response to progesterone appears to differ in women with PMD and controls. We have begun conducting dexamethasone-corticotrophin releasing hormone (dex/CRH) testing in women with PMD and controls in both menstrual cycle and Lupron studies, as this measure is believed to be particularly sensitive to differences in vasopressin and is substantially less difficult to perform. We are awaiting the final results of our dex/CRH studies in women with PMD and asymptomatic controls, who have been tested during each of the three hormone conditions established in the Lupron paradigm. Preliminary data confirm a progesterone-related enhancement of cortisol secretion in both women with and without PMD (albeit largely reflecting an increased secretion in those without PMD), and the failure to suppress CRH-stimulated adrenocorticotrophic hormone (ACTH) secretion in women with PMD (but not controls) during both hypogonadism and estradiol replacement;whereas ACTH secretion is suppressed in both women with PMD and controls during progesterone replaced conditions. First, we intend to evaluate the role of early life trauma in these data, since women with PMD report high rates of early life trauma, and early life trauma accounts for a considerable amount of the variance in the elevated CRH-stimulated ACTH secretion observed in depression. Second, we will compare these data during the Lupron paradigm with dex/CRH tests performed across the menstrual cycle to examine the possible role of neurosteroids in the HPA axis responses in the women with PMD and controls. Finally, our preliminary data suggest an abnormality of HPA axis negative feedback regulation in PMD that could be normalized by progesterone.
Showing the most recent 10 out of 15 publications
