Endocrine and Neurobiologic Events Accompanying Puberty

National Institute Of Mental Health

Investigators

Linked publications & trials

Abstract

This report includes work arising from the following clinical protocol: NCT01434368. A prepubertal cohort enters the study at age 8, and clinicians determine that each child meets Pubertal stage (PS) 1 criteria (i.e., no evidence of secondary sexual characteristics in breast, testes, or pubic hair), is medically well, is between the 15th and 85th percentile BMI, and has radiologic evidence of age-appropriate bone development. The absence of psychiatric illness in each child is confirmed (at each visit) by structured interview (i.e., K-SADS), and we confirm the absence of Axis I psychiatric illness in any first-degree relatives. A second cohort enters the study at 12-13 years (PS 2-3) and meets the same selection criteria as those in the prepubertal cohort. Behavioral measures include ratings of mood, social experiences within family and amongst peers, sleep, as well as measures of early life trauma. Reproductive endocrine (i.e., gonadal and adrenal steroids measured by LC-MS/MS), bone age (a reflection of cumulative tissue exposure to estradiol in both boys and girls), metabolic (i.e., MRI and dual-energy X-ray absorptiometry DEXA measures of visceral fat and body composition), and physical measures (i.e., anthropomorphic indices, MRI measures of gonadal volume) are employed to fully characterize the stages, duration and tempo of pubertal development. Brain outcome measures derived via multimodal neuroimaging techniques include the following: resting-state fMRI, structural MRI, DTI, and four fMRI tasks (i.e., emotional processing, reward, impulse inhibition and working memory). These latter fMRI tasks were selected because they target neural systems reported to undergo both structural and functional transformation during adolescence and are relevant for neuropsychiatric disorders. Finally, all participants provide blood samples for genotyping, methylome measures and for the formation of cell-lines (LCLs, h-IPCs) in which functional genomics studies could be performed investigating the effects of puberty-related endocrine events on the transcriptome. This past year we have initiated a collaborative relationship with the pediatric gynecology service at the NIH CC who will employ the MRI measures of gonadal volumes to examine the effects of puberty on these measures as well as employing these measures in comparison to adolescent girls with several forms of ovarian pathology including Turners syndrome, polycystic ovaries and those with premature ovarian insufficiency. We evaluate the neurobiological impact of the onset of adrenarche, the activation of the adrenal androgen secretion, a uniquely human phenomenon that plays an important role in puberty. The impact of adrenal androgens, including dehydroepiandrosterone sulfate (DHEAS), on neurodevelopment has been difficult to disambiguate from that of gonadarche. Adrenarche-specific changes in prepubertal brain function were examined in a sample of typically developing children who were carefully documented to be prepubertal as ascertained by clinicians, and thus had no puberty-related increases in gonadal hormones. Approximately half of this pre-pubertal sample met biochemical criteria for pre-adrenarche (i.e., serum DHEAS levels DHEAS > 40µg/dL. The almost uniquely human experience of adrenarche is thought to serve an evolutionary role to protect the brain during development and to facilitate the slower maturation of brain systems needed in humans compared with other species. Developmental neuroimaging studies support a maturational role of the adrenal androgen DHEAS on brain regions relevant for both reward processing and cognitive control. However, due to the co-occurrence of both adrenarche and gonadarche during the pubertal transition, few studies in humans have been able to examine the effects of DHEAS on brain development independent of those due to gonadal steroids. We investigated 98 prepubertal and pre-gonadarchal 8-year-olds (n=44 girls) who were divided into pre-adrenarchal children and those who were post adrenarchal. Preliminary results suggest that DHEAS impacts brain regions underlying processes of inhibitory control, including the inferior frontal gyrus (IFG). We initially observed between-group differences (i.e., pre- vs. post adrenarche) in the activation of bilateral IFG in a stop-signal task (p<0.005, uncorrected). Post-hoc analyses revealed that plasma levels of DHEAS were positively correlated with activations in these regions (p<0.001, uncorrected). These preliminary results suggest that DHEAS secretion at adrenarche plays a role in brain maturation in humans. Future studies will further examine the role of DHEAS in brain maturation and, specifically, whether DHEAS secretion impacts the effects of gonadarche on typical brain development in boys and girls. The observations of adrenarche-associated neural changes in typically developing children who are otherwise pre-gonadarchal suggest that the onset of adrenarche and adrenal androgen secretion can significantly impact brain development. This work provides a unique opportunity to examine the effects of the onset of adrenarche on brain development and isolate the neural impacts of adrenal hormones from those of other puberty-related hormones. We also are conducting analyses characterizing longitudinal changes in the relationships between pubertal stage and the maturation of large-scale brain network function. We longitudinally tracked 157 healthy children (ages 8-18) every 8-10 months (792 cumulative visits) from pre- to post-puberty and into young adulthood with high-frequency resting-state fMRI, clinician-determined pubertal stage, and plasma estradiol and testosterone measurements. We used generalized-additive-mixed models to test for sex-specific effects of pubertal stage and sex steroids, and their interactions, on large-scale, resting-state functional connectivity. Overall, mean within-network connectivity increased with puberty, whereas mean between-network connectivity remained unchanged. Males showed higher mean within- and between-network connectivity than females throughout pubertal stages. Regarding individual networks, puberty most strongly affected intra-limbic among within-networks and limbic-default mode network (DMN) connectivity among between-networks Males exhibited stronger and more robust functional connectivity in intra-limbic, limbic-DMN, somatomotor, and attention networks than females, with testosterone and estradiol showing distinct modulatory patterns. This work offers novel insights about sex-specific pubertal effects on neurodevelopmental trajectories of intrinsic large-scale functional brain reorganization and identifies associated neurohormonal. Menarche is a critical developmental window accompanying the emergence of the two-fold increased prevalence of depressive disorders in women compared to men. In a preliminary analysis of a prospectively acquired sample of girls with concurrent documentation of menarche (mean age=12.2 years), for whom structural MRIs were collected within one year before and one year after menarche. Several cortical areas with decreases in gray matter volume (GMV) post-menarche were identified, whereas GMV increased in the parahippocampus; similar changes were absent in a sample of age-matched girls who did not enter menarche during a comparable time period. Although additional factors including BMI, pubertal trajectory, and environment can influence both the timing of menarche and GMV, these preliminary data may point to region-specific, menarche-related GMV changes. Future plans include the exploration of the effects of menarche on additional neuroimaging measures (i.e., network-scale connectivity, reward system function), as well as the impact on these neuroimaging measures of the instantiation of regular menstrual cyclicity (i.e., versus menarche), presence of perimenstrual affective symptoms, and the commencement of cyclic E2 (and progesterone) secretion. In addition to investigating neuroimaging data prior to and after pubertal milestones, the repeated measures obtained within each child permits the examination of measures of cumulative hormone exposure within individual children. Our initial analyses of plasma testosterone (T) levels obtained at each clinic visit across the pubertal transition failed to show any effects on hippocampal volume (a finding reported previously in other animals)163. However, a calculated measure of cumulative exposure to T (i.e., area under the T secretory curve that reflects both absolute blood levels and duration of exposure) in each boy across puberty demonstrated a significant effect (p=0.002) of cumulative T exposure on hippocampal volume in accordance with animal studies. We will further examine these effects of cumulative hormone exposure (both calculated and measures of bone age [reflecting exposure to E2 in both boys and girls]) on brain development and attempt to distinguish the direct actions of T compared with those of E2.