This report includes work arising from the following clinical protocol: NCT01434368. Recruitment efforts for this longitudinal study continue and we have completed approximately 90% of our planned enrollment of for our first cohort of 12-13 year old children and approximately 60% of our planned enrollment of prepubertal children (8 year olds). This year several children completed their sixth seventh repeated clinic visits. We have observed several important methodologic variables that will inform future studies: First, age alone is not sufficient to determine prepubertal status since we have identified approximately 10 - 20% of girls who show advanced Tanner breast stage at age 8. Breast development was unilateral and reflected early estradiol secretion in these girls and, therefore they were not prepubertal. Second, neither chronologic age nor Tanner stage are sufficient markers of sex steroid (i.e., estradiol) exposure in boys and girls, since we have identified abnormally advanced bone age relative to their chronologic age and controlled for sex (as measured by left wrist x-rays) in approximately 10% of children screened for enrollment in this study (who otherwise are within normal range for body mass index BMI). We also have documented the reversal of pubertal stage in both boys and girls during early and late puberty consistent with the original reports by Tanner (but rarely reported in the subsequent clinical literature). We are currently investigating the potential endocrine accompaniments of this observed reversal of pubertal stage. Finally, we have identified abnormally increased or decreased BMI (defined by normal BMI percentiles between 15th and 85th percentiles) which is accompanied by alterations in sex steroid secretion in approximately 17% of otherwise normal healthy children screened for this study thus emphasizing the importance of controlling for BMI in studies of the relationship between sex steroid secretion and brain development. We are currently investigating the effects of these early changes in gonadal activity (i.e., advanced bone age) on brain structure and function. These data emphasize the importance of our strict criteria for defining prepuberty which will allow more meaningful evaluations of the potential impact of puberty-related physiologic events on our outcome measures. We have identified several important sex-differences that are present in prepubertal children in the absence of differences in either gonadal (by definition and confirmed with blood hormone measures) and adrenal sex steroids (related to adrenarche). For example, preliminary results of HPA axis measures in these children in which we employ the cortisol awakening response (CAR), suggest that sex differences in CAR exist, with girls having greater HPA axis activity (i.e., higher peak cortisol and more prolonged secretion) compared with boys. However, we observed no significant main or interactive effects of pubertal stage on salivary cortisol (CAR) measures. These data are consistent with studies in animals that demonstrate increased HPA axis responsivity to a variety of stressors in females compared with males; however, our preliminary findings document that the sex-differences in humans emerge prior to puberty and therefore do not reflect the onset of gonadarche for their expression. Preliminary results of multi-modal neuroimaging studies in this project suggest that pubertal stage contributes to several measures of adolescent brain development and accounts for differences between adolescent and adult brain development that are both brain region- and sex-specific. First, resting state functional connectivity studies have identified sex-differences in pubertal development of amygdalar resting functional connectivity. Our data suggest that biological sex influences amygdalar functional connectivity across puberty, with boys showing higher coupling between the amygdala and ventral striatum, temporal gyrus, and lateral prefrontal cortex compared with girls. However, our data also show similar amygdala medial prefrontal functional connectivity in boys and girls, suggesting that developmental alterations in amygdala connectivity are region specific and influenced by age and sex. Second, cognition-activated fMRI studies document differential working memory-related prefrontal-hippocampal functional connectivity in children (early puberty) and adolescents (late puberty). We found differential negative prefrontal-hippocampal functional coupling during working memory in children but not adolescents, which could indicate altered reciprocal coupling between an immature prefrontal cortex and the hippocampus, along with changes in the level of cooperativity between these brain regions accompanying prefrontal cortical development during adolescence. Third,in our investigations of the development of neural inhibition in which we use fMRI and the well-validated Stop Signal Task (a reliable metric of neural inhibition), in which a pre-potent go response must be inhibited, we have been unable to confirm previous reports of significant sex- or age-related differences in the brain that subserve this key cognitive function. We did demonstrate a main effect of task, consistent with the current literature in both adults and children, with activation in the inferior frontal gyrus bilaterally, the supplementary motor area, and the subthalamic nucleus. However, there were no significant effects of sex or pubertal status in these task-related brain regions, despite both sex and pubertal differences between behavioral measures, such as speed of go response and stop accuracy. We did observe a pubertal status-by-sex interaction in the right medial prefrontal cortex (BA10). Post-hoc testing showed there was a significant between-sex difference in the pre-pubertal cohort. This puberty group by sex interaction was found in the default mode network in a region where activity is reduced during non-referential goal related tasks; activation of this region also has shown to be related to social and emotional processes. Although preliminary, the sex difference in mPFC activation in the pre-pubertal children did not appear to be reflective of (or mediated by) differences in adrenal steroid levels either. Nonetheless, it may be possible that either adrenal steroid levels or organizational sex-differences in mPFC that emerge early in childhood are systematically altered during puberty. Indeed, we found that serum testosterone levels negatively correlated with mPFC activation in boys, possibly suggesting that when testosterone levels rise during puberty, they act to decrease mPFC activation, at least within the context of behavioral inhibition. Further investigation is needed to explore the correlates of the robust deactivation in pre-pubertal females. Finally, we also have investigated the effects of sex and puberty in the developmental trajectory of emotion processing. Previous research has shown both structural and functional sex differences in brain regions associated with socioemotional processing. We examined sex and pubertal stage effects on neural processing while children viewed emotional faces by fMRI. In both prepubertal and postpubertal groups, there was a main effect of sex when viewing aversive faces such that girls showed greater activation than boys in the right intraparietal sulcus, left orbitofrontal cortex (OFC), left dorsolateral prefrontal cortex (DLPFC), right superior temporal sulcus, left anterior cingulate cortex (ACC), and right posterior cingulate cortex. Albeit preliminary, our findings in studies of emotional face-processing suggest that sex differences in socioemotional processing also are established prior to puberty and develop independent of gonadal sex-steroid secretion.
