Greti Aguilera, MD, Head, Section on Endocrine Physiology
Ying Liu, MD, Research Associate
Jun Chen, PhD, Postdoctoral Fellow
Sivan Subburaju, PhD, Postdoctoral Fellow
Anna Kamitakahara, BS, Predoctoral Fellow

The goal of the laboratory is to understand the neuroendocrine mechanisms underlying the stress response, with emphasis onn the regulation of the hypothalamic pituitary adrenal (HPA) axis. The ability of an organism to adapt to acute and chronic stress situations is determined by genetic constitution and previous experiences. Studies in our laboratory have shown that exposure to a repeated somatosensory stress causes hyperresponsiveness of the HPA axis to a novel stress. Given that hyperactivity of the HPA axis has been implicated in the pathogenesis of a number of psychiatric and metabolic disorders, self-limitation of the stress response is critical to avoid deleterious effects of glucocorticoid excess. The laboratory studies the mechanisms by which positive and negative regulation of expression of the hypothalamic hormones corticotropin releasing hormone (CRH) and vasopressin (VP) and their pituitary receptors are achieved under different stress situations and the consequences of such regulation for ACTH secretion and adrenal steroidogenesis.

Regulation of hypothalamic CRH expression

Liu, Aguilera, Kamitakahara

Our studies have illuminated the role of the interaction between CRH and VP in the regulation of pituitary ACTH and in the regulation of the expression of these peptides in the paraventricular nucleus (PVN) during stress and other alterations of the HPA axis. Previous studies showed that CRH and VP co-expressed in the same parvocellular neuron of the PVN are differentially regulated during stress or exposure to glucocorticoids. VP becomes the predominant peptide expressed in parvocellular neurons of the PVN during chronic stress. However, our results suggest that, despite the prevalence of VP, ACTH secretion depends primarily on rapid but limited increases in CRH secretion. During the past year, we continued to examine the regulation of CRH expression, using as models the hypothalamic cell line 4B, which exhibits characteristics of the parvocellular neuron, and primary cultures of hypothalamic neurons.

While CRH is essential for full ACTH and corticosterone responses to stress, excessive CRH production leads to developmental, psychiatric, metabolic, immune, and reproductive disorders. CRH secretion in response to stress is accompanied by rapid but transient increases in CRH transcription. Our previous work showed that termination of CRH transcription is independent of the stress-induced glucocorticoid surge and that termination of CRH transcription is associated with increased expression of inducible cAMP early repressor (ICER), a repressor isoform of cAMP--responsive element modulator (CREM). In keeping with a role of ICER in regulating CRH transcription, CREM mRNA induced during stress was localized in CRH cells of the PVN. The electromobility gel shift assay (EMSA) and chromatin-immunoprecipitation assays showed late association of ICER with the CRH promoter during stress, concomitant with increases in ICER protein and a return of CRH transcription to basal values.

An exciting new finding demonstrated that blockade of ICER production, using anti-ICER siRNA oligonucleotides, prevented the decline of CRH transcription during sustained forskolin stimulation of 4B cells or in primary cultures of hypothalamic neurons in the presence of forskolin. In these experiments, transfection of cells with ICER siRNA oligonucleotides significantly reduced ICER formation following a three-hour incubation with forskolin. Co-transfection of ICER siRNA oligonucleotides with a CRH promoter--driven luciferase reporter (CRHp-Luc) potentiated the stimulatory effect of forskolin on CRH promoter activity. In addition, measurement of endogenous CRH transcription using intronic quantitative PCR in hypothalamic cell cultures revealed transient activation of transcription following incubation with forskolin. Pretreatment of the cultures for 24 hours with ICER siRNA oligonucleotides prevented the decline in transcription, reinforcing the view that ICER is required for termination of CRH transcription. These data support the hypothesis that ICER formation is responsible for the inhibitory phase of CRH transcription and suggest that induction of ICER is part of an intracellular feedback mechanism limiting CRH transcription during stress.

While cAMP appears to be required for positive and negative regulation of CRH transcription, the major regulators of CRH neurons (norepinephrine and glutamate) are not mediated by cAMP, suggesting that other messenger systems activate CRH transcription by potentiating the effects of low cAMP levels. We examined this hypothesis in neuronal cell cultures treated with the phorbol ester PMA, the cAMP stimulator forskolin, or with a combination of both. In the hypothalamic cell line 4B, transfected with a CRH promoter-driven luciferase reporter, incubation with forskolin markedly increased CRH promoter activity. PMA alone had no effect but potentiated the stimulatory effect of forskolin. Co-transfection of the dominant negative form of CREB (A-CREB) reduced forskolin-stimulated CRH promoter activity in the absence or presence of PMA, suggesting that pCREB is required for CRH transcriptional activation. However, the lack of effect of PMA on CRH promoter activity, in spite of marked CREB phosphorylation, suggested that pCREB must interact with other transcription factors to induce CRH transcription. PMA also potentiated the stimulatory effect of forskolin on CRH transcription (measured by intronic qRT-PCR) in primary cultures of hypothalamic cells. These data point to a mechanism by which potentiation of cAMP-induced transcription by phospholipid-dependent pathways could mediate positive and negative regulation of CRH transcription in response to non--cAMP dependent regulators during stress.

Central actions of prolactin

Liu, Aguilera; in collaboration with Blume

In addition to its role in lactation, prolactin (PRL), which is produced in the brain, can act as a neurotransmitter/neuromodulator, attenuating behavioral and HPA axis responses to stress. Given that expression of PRL and PRL receptors in the PVN of the hypothalamus increases during lactation, PRL has been implicated in the HPA axis adaptation observed during late pregnancy and lactation. Studies aimed at understanding the actions of PRL in the brain showed marked phosphorylation of ERK1/2 following intracerebroventricular injection of PRL or incubation of neuronal cell lines expressing PRL receptors with PRL. Immunohistochemical studies revealed strong co-localization of pERK and VP in the supraoptic nucleus while, in the PVN, only a minor population of pERK-expressing cells were VP-positive. Western blot of protein extracts from 4B cells with a PRL receptor antibody showed a 40kDa band consistent with the short form of the PRL receptor. Incubation of the cells with PRL for 5 to 60 minutes raised pERK and markedly potentiated the effect of serum on ERK1/2 phosphorylation. In 4B cells transfected with a CRH promoter-driven luciferase reporter gene, incubation with PRL for six hours potentiated the stimulatory effect of forskolin on CRH promoter activity, an effect that ERK1/2 inhibitors prevented. In quantitative RT-PCR experiments, PRL induced the expression of early growth-response genes 1 and 2 (egr-1 and egr-2) in an ERK1/2--dependent manner. The data show that PRL activates the ERK1/2 kinase cascade in selected hypothalamic nuclei. The ability of PRL to induce ERK phosphorylation in the PVN and to increase CRH promoter activity in a hypothalamic cell line suggests that PRL can directly modulate CRH neuron function. Furthermore, it is likely that induction of Egr1/2 contributes to neuronal plasticity changes associated with high PRL production during lactation.

Pituitary actions of vasopressin

Subburaju, Chen, Aguilera

By acting through plasma membrane receptors of the V1b subtype (V1bR), vasopressin produced by parvocellular neurons of the PVN potentiates the stimulatory effect of CRH on pituitary ACTH secretion. The expression of parvocellular VP and pituitary V1bR but not that of hypothalamic CRH and pituitary CRHR1 increases during chronic stimulation of the hypothalamic pituitary adrenal axis. This suggests that VP becomes the main regulator of ACTH secretion during long-term adaptation to stress and that the peptide mediates the characteristic hyperresponsiveness of the hypothalamic pituitary adrenal (HPA) to a novel stress. To test this hypothesis, we measured plasma ACTH responses to intraperitoneal hypertonic saline injection (ipHS) in rats subjected to daily handling or repeated restraint stress (1 hour per day for 14 days) while receiving a minipump infusion of the V1 receptor antagonist dGly[Phaa1,D-tyr(et),Lys, Arg]VP (V1R-ant) or vehicle. V1R-ant infusion for 14 days blunted ACTH responses to intravenous injection of 100 ng VP without affecting changes in blood pressure and heart rate during acute restraint stress as compared with responses in vehicle-infused rats. Basal plasma ACTH levels showed a tendency to decrease in both handled and repeatedly restrained rats receiving the V1R-ant. In handled rats, exposure to the novel stress of ipHS caused marked increases in plasma ACTH levels, an effect that was significantly reduced in the group receiving V1R-ant. As expected, ACTH responses to ipHS were higher in repeatedly restrained rats than in handled controls but, in contrast to handled rats, ACTH responses to ipHS in repeatedly restrained rats were unaffected by the V1R-ant. The data show that VP contributes to the ACTH responses to an acute stressor but that increased vasopressinergic activity is not responsible for the enhanced ACTH responses to the novel stress.

The low impact of vasopressinergic blockade on HPA axis activity during chronic stress suggests that VP plays additional roles, such as controlling the number of pituitary corticotrophs. We studied the role of VP in mediating pituitary corticotroph mitogenesis in adrenalectomized rats (ADX) by examining the effect of the peptide VP receptor antagonist V1R-ant on the number of cells incorporating bromouridine (BrdU). Long-term ADX increased the number of both BrdU-labeled cells and ACTH-stained cells. Infusion of V1-ant for 28 days prevented ADX-induced increases in BrdU incorporation but not changes in the number of ACTH-stained cells. Unexpectedly, co-localization of BrdU uptake in ACTH-positive cells was minor and unaffected by ADX or V1-antagonist infusion. We observed no BrdU-stained nuclei in LH, TSH, PRL, GH, folliculo-stellate cells, or nestin-labeled progenitor cells. The pituitary corticotroph--exclusive transcription factor Tpit co-localized in over 80 percent of ACTH-containing cells but in only 5 percent of BrdU-labeled nuclei in controls and 10 percent of those in ADX rats. In V1bR knockout mice, the number of cells incorporating BrdU following ADX was lower than in wild-type mice, and we detected no major co-localization of BrdU and ACTH. The data demonstrate that VP mediates mitogenic activity in the pituitary during long-term ADX. The lack of co-localization of ACTH and BrdU suggests that recruitment of corticotrophs during adrenalectomy occurs from undifferentiated cells.

Interaction between CRH and V1b receptors

Young,¹ Aguilera

Increasing evidence suggests that, although originally thought to function as monomers, G protein--coupled receptors (GPCRs) can form oligomers. Regulation of ACTH secretion involves strong interaction between activation of the GPCR V1b receptor and corticotropin releasing hormone receptor type 1 (CRHR1), but it is not known whether a physical interaction occurs between the receptors. We first addressed this issue by using bioluminescence resonance energy transfer (BRET) to study dimerization in living cells. The V1bR and CRHR1 fused to either Renilla luciferase (rluc) or yellow fluorescent protein (YFP) at the carboxy-terminus of the receptor were fully bioactive when transiently transfected into CHO cells. We determined dimerization by measuring the excitation of receptor-YFP by energy transferred from receptor-rluc, using a multiplate reader. Energy transfer from V1bRrluc to V1bYFP, CRHR1YFP, or the unrelated bradykinin 2 receptor-YFP (B2RYFP) was very low, and levels were unchanged by co-transfection with wild-type V1bR or following incubation of the cells with the ligands CRH or VP. However, the OTRrluc/OTRYFP pair, used as a positive control, displayed the expected BRET levels and lack of interaction with B2RYFP. In contrast to the BRET data, co-immunoprecipitation using receptors tagged with the c-myc and Flag epitopes demonstrated specific homodimerization of the V1b receptor and heterodimerization of the V1b receptor with both the OTR and CRHR1 receptors, suggesting that the position of the rluc and YFP tags impaired the BRET signal. Our current studies focus on the functional significance of V1b and CRHR1 receptor dimerization.

¹ Sharla Young, PhD, former Postdoctoral Fellow

COLLABORATOR

Annegret Blume, PhD, Universität Regensburg, Regensburg, Germany

For further information, contact aguilerg@cc1.nichd.nih.gov.