The Paraventricular Nucleus of the Hypothalamus: Development, Function, and Human Diseases

The paraventricular nucleus of the hypothalamus (PVH), located in the ventral diencephalon adjacent to the third ventricle, is a highly conserved brain region present in species from zebrafish to humans. The PVH is composed of three main types of neurons, magnocellular, parvocellular, and long-projecting neurons, which play imperative roles in the regulation of energy balance and various endocrinological activities. In this review, we focus mainly on recent findings about the early development of the hypothalamus and the PVH, the functions of the PVH in the modulation of energy homeostasis and in the hypothalamus-pituitary system, and human diseases associated with the PVH, such as obesity, short stature, hypertension, and diabetes insipidus. Thus, the investigations of the PVH will benefit not only understanding of the development of the central nervous system but also the etiology of and therapy for human diseases.

Increased Thyroid Hormone Activation Accompanies the Formation of Thyroid Hormone-Dependent Negative Feedback in Developing Chicken Hypothalamus

The hypothalamic-pituitary-thyroid axis is governed by hypophysiotropic TRH-synthesizing neurons located in the hypothalamic paraventricular nucleus under control of the negative feedback of thyroid hormones. The mechanisms underlying the ontogeny of this phenomenon are poorly understood. We aimed to determine the onset of thyroid hormone-mediated hypothalamic-negative feedback and studied how local hypothalamic metabolism of thyroid hormones could contribute to this process in developing chicken. In situ hybridization revealed that whereas exogenous T4 did not induce a statistically significant inhibition of TRH expression in the paraventricular nucleus at embryonic day (E)19, T4 treatment was effective at 2 days after hatching (P2). In contrast, TRH expression responded to T3 treatment in both age groups. TSHβ mRNA expression in the pituitary responded to T4 in a similar age-dependent manner. Type 2 deiodinase (D2) was expressed from E13 in tanycytes of the mediobasal hypothalamus, and its activity increased between E15 and P2 both in the mediobasal hypothalamus and in tanycyte-lacking hypothalamic regions. Nkx2.1 was coexpressed with D2 in E13 and P2 tanycytes and transcription of the cdio2 gene responded to Nkx2.1 in U87 glioma cells, indicating its potential role in the developmental regulation of D2 activity. The T3-degrading D3 enzyme was also detected in tanycytes, but its level was not markedly changed before and after the period of negative feedback acquisition. These findings suggest that increasing the D2-mediated T3 generation during E18-P2 could provide the sufficient local T3 concentration required for the onset of T3-dependent negative feedback in the developing chicken hypothalamus.

Embryonic GABA(B) receptor blockade alters cell migration, adult hypothalamic structure, and anxiety- and depression-like behaviors sex specifically in mice

Neurons of the paraventricular nucleus of the hypothalamus (PVN) regulate the hypothalamic- pituitary-adrenal (HPA) axis and the autonomic nervous system. Females lacking functional GABA_B receptors because of a genetic disruption of the R1 subunit have altered cellular characteristics in and around the PVN at birth. The genetic disruption precluded appropriate assessments of physiology or behavior in adulthood. The current study was conducted to test the long term impact of a temporally restricting pharmacological blockade of the GABA_B receptor to a 7-day critical period (E11–E17) during embryonic development. Experiments tested the role of GABA_B receptor signaling in fetal development of the PVN and later adult capacities for adult stress related behaviors and physiology. In organotypic slices containing fetal PVN, there was a female specific, 52% increase in cell movement speeds with GABA_B receptor antagonist treatment that was consistent with a sex-dependent lateral displacement of cells in vivo following 7 days of fetal exposure to GABA_B receptor antagonist. Anxiety-like and depression-like behaviors, open-field activity, and HPA mediated responses to restraint stress were measured in adult offspring of mothers treated with GABA_B receptor antagonist. Embryonic exposure to GABA_B receptor antagonist resulted in reduced HPA axis activation following restraint stress and reduced depression-like behaviors. There was also increased anxiety-like behavior selectively in females and hyperactivity in males. A sex dependent response to disruptions of GABA_B receptor signaling was identified for PVN formation and key aspects of physiology and behavior. These changes correspond to sex specific prevalence in similar human disorders, namely anxiety disorders and hyperactivity.

Maternal obesity and IL-6 lead to aberrant developmental gene expression and deregulated neurite growth in the fetal arcuate nucleus

Maternal obesity during pregnancy increases the risk of obesity in the offspring. Several observations have pointed to a causative role for the proinflammatory cytokine IL-6, but whether it is present in the fetal circulation and how it acts on the developing fetus are unclear. We first observed that postnatal day 0 offspring from obese mothers had significantly reduced neuropeptide Y (NPY) innervation of the paraventricular nucleus (PVN) compared with that for offspring of normal-weight controls. Thus, the growth of NPY neurites from the arcuate nucleus (ARC) was impaired in the fetal brain by maternal obesity. The neurite growth regulator, Netrin-1, was expressed in the ARC and PVN and along the pathway between the two at gestational day (GD) 17.5 in normal animals, making it likely to be involved in the development of NPY ARC-PVN projections. In addition, the expression of Dcc and Unc5d, receptors for Netrin-1, were altered in the GD17.5 ARC in obese but not normal weight pregnancies. Thus, this important developmental pathway is perturbed by maternal obesity and may explain the defect in NPY innervation of the PVN that occurs in fetuses developing in obese mothers. To investigate whether IL-6 may play a role in these developmental changes, we found first that IL-6 was significantly elevated in the fetal and maternal circulation in pregnancies of obese mice compared with those of normal-weight mice. In addition, treatment of GD17.5 ARC tissue with IL-6 in vitro significantly reduced ARC neurite outgrowth and altered developmental gene expression similar to maternal obesity in vivo. These findings demonstrate that maternal obesity may alter the way in which fetal ARC NPY neurons respond to key developmental signals that regulate normal prenatal neural connectivity and suggest a causative role for elevated IL-6 in these changes.

Expression of Robo/Slit and Semaphorin/Plexin/Neuropilin family members in the developing hypothalamic paraventricular and supraoptic nuclei

The hypothalamic paraventricular nucleus (PVN) and supraoptic nucleus (SON) contain neuroendocrine cells that modulate pituitary secretion to maintain homeostasis. These two nuclei have a common developmental origin but they eventually form at locations distant from each other. Little is known about the molecular cues that direct the segregation of PVN and SON. As a means to identify potential factors, we have documented expression patterns of genes with known guidance roles in neural migration. Here, we focus on two groups of ligand/receptor families classified to mediate chemo-repulsion of neurons and their axons: the Slit/Robo and the Semaphorin/Plexin/Neuropilin families. Their dynamic expression patterns within and around the common PVN/SON progenitor as well as the mature PVN and SON may provide a framework for understanding the formation of these two important nuclei.

BDNF up-regulates pre-pro-TRH mRNA expression in the fetal/neonatal paraventricular nucleus of the hypothalamus. Properties of the transduction pathway

Brain derived neurotrophic factor (BDNF) increases the levels of pre-pro-thyrotropin releasing hormone (TRH) mRNA in fetal rodent hypothalamic neurons that express TrkB receptors. The present studies aimed at better understanding the role of BDNF in establishing and maintaining the TRH phenotype in hypothalamic neurons during early development. To determine where BDNF regulates the expression of pre-pro-TRH mRNA in vivo, we compared the hypothalamic distribution of pre-pro-TRH mRNA to that of TrkB mRNA. Full-length TrkB (FL-TrkB) mRNA was detected earlier in development than pre-pro-TRH mRNA in the region that gives rise to the paraventricular nucleus of the hypothalamus (PVN). We also evaluated the effects of BDNF on the expression of pre-pro-TRH mRNA in vitro. BDNF up-regulated the levels of pre-pro-TRH mRNA in primary cell cultures obtained from the hypothalamus or the PVN of 17 days old fetuses or newborn rats. This effect was abolished by PD98059, an inhibitor of the mitogen-activated protein kinase kinase (MEK) 1/2 or 5. The effect of BDNF on pre-pro-TRH mRNA levels was reversible. The continuous application of BDNF led to a desensitization of the response at day 10 in vitro, an effect that correlated with a drop in the levels of FL-TrkB protein. In conclusion, BDNF enhances the expression of pre-pro-TRH mRNA in PVN neurons. This effect is reversible, decreases with time, and requires an active MEK. BDNF may contribute to the enhancement of pre-pro-TRH mRNA expression in the hypothalamic PVN during development.

Prenatal development of hypothalamic neuropeptide systems in the nonhuman primate

In the rodent, arcuate nucleus of the hypothalamus (ARH)-derived neuropeptide Y (NPY) and proopiomelanocortin (POMC) neurons have efferent projections throughout the hypothalamus that do not fully mature until the second and third postnatal weeks. Since this process is likely completed by birth in primates we characterized the ontogeny of NPY and melanocortin systems in the fetal Japanese macaque during the late second (G100), early third (G130) and late third trimesters (G170). NPY mRNA was expressed in the ARH, paraventricular nucleus (PVH), and dorsomedial nucleus of the hypothalamus (DMH) as early as G100. ARH-derived NPY projections to the PVH were initiated at G100 but were limited and variable; however, there was a modest increase in density and number by G130. ARH-NPY/agouti-related peptide (AgRP) fiber projections to efferent target sites were completely developed by G170, but the density continued to increase in the postnatal period. In contrast to NPY/AgRP projections, alphaMSH fibers were minimal at G100 and G130 but were moderate at G170. This study also revealed several significant species differences between rodent and the nonhuman primate (NHP). There were few NPY/catecholamine projections to the PVH and ARH prior to birth, while projections were increased in the adult. A substantial proportion of the catecholamine fibers did not coexpress NPY. In addition, cocaine and amphetamine-related transcript (CART) and alpha-melanocyte stimulating hormone (alphaMSH) were not colocalized in fibers or cell bodies. As a consequence of the prenatal development of these neuropeptide systems in the NHP, the maternal environment may critically influence these circuits. Additionally, because differences exist in the neuroanatomy of NPY and melanocortin circuitry the regulation of these systems may be different in primates than in rodents.

Evidence of altered brain sexual differentiation in mice exposed perinatally to low, environmentally relevant levels of bisphenol A

Humans are routinely exposed to bisphenol A (BPA), an estrogenic chemical present in food and beverage containers, dental composites, and many products in the home and workplace. BPA binds both classical nuclear estrogen receptors and facilitates membrane-initiated estrogenic effects. Here we explore the ability of environmentally relevant exposure to BPA to affect anatomical and functional measures of brain development and sexual differentiation. Anatomical evidence of alterations in brain sexual differentiation were examined in male and female offspring born to mouse dams exposed to 0, 25, or 250 ng BPA/kg body weight per day from the evening of d 8 of gestation through d 16 of lactation. These studies examined the sexually dimorphic population of tyrosine hydroxylase (TH) neurons in the rostral periventricular preoptic area, an important brain region for estrous cyclicity and estrogen-positive feedback. The significant sex differences in TH neuron number observed in control offspring were diminished or obliterated in offspring exposed to BPA primarily because of a decline in TH neuron number in BPA-exposed females. As a functional endpoint of BPA action on brain sexual differentiation, we examined the effects of perinatal BPA exposure on sexually dimorphic behaviors in the open field. Data from these studies revealed significant sex differences in the vehicle-exposed offspring that were not observed in the BPA-exposed offspring. These data indicate that BPA may be capable of altering important events during critical periods of brain development.

Ontogeny of the projections from the anteroventral periventricular nucleus of the hypothalamus in the female rat

Neurons in the anteroventral periventricular nucleus of the hypothalamus (AVPV) mediate a variety of autonomic functions. In adults they primarily innervate neuroendocrine nuclei in the periventricular zone of the hypothalamus, including the paraventricular and arcuate nuclei (PVH, ARH). Ascending projections from the AVPV also provide inputs to the ventrolateral septum (LSv) and the principal division of the bed nuclei of the stria terminalis (BSTp). Consistent with a role in regulating preovulatory luteinizing hormone secretion, rostral projections from the AVPV contact gonadotropin-releasing hormone (GnRH) neurons surrounding the vascular organ of the lamina terminalis (OVLT). To study the development of these pathways, we placed implants of the lipophilic tracers DiI and CMDiI into the AVPV of female rats ranging in age from embryonic day 19 (E19) through adulthood. The earliest projections targeted a population of GnRH neurons, with apparent contacts from labeled fibers observed as early as E19. These connections appeared to be fully developed before birth, as similar numbers of appositions from AVPV projections onto the GnRH-immunoreactive cells were observed at all ages examined. Caudal projections were delayed relative to projections to the OVLT. Labeled AVPV fibers reached the PVH during the first postnatal week, and fibers targeting the BSTp and LSv were not observed until the second and third postnatal weeks, respectively. Labeled AVPV fibers were not seen in the ARH of animals at any age. Our results demonstrate that projections from the AVPV develop with both spatial and temporal specificity, innervating each target with a unique developmental profile.

Proopiomelanocortin processing in the anterior pituitary of the ovine fetus after lesion of the hypothalamic paraventricular nucleus

The hypothalamic-pituitary-adrenocortical axis plays an essential role in the maturation of fetal organs and, in sheep, birth. Lesioning the paraventricular nucleus (PVN) in fetal sheep prevents adrenocortical maturation and parturition without altering plasma immunoreactive ACTH concentrations. The purpose of this study was to determine the effect of PVN lesion on anterior pituitary processing of proopiomelanocortin (POMC) to ACTH, plasma concentrations of ACTH and ACTH precursors (POMC; 22-kDa proACTH), and expression of subtilisin-like prohormone convertase 3 (SPC3) in corticotropes in fetal sheep. PVN lesion did not affect anterior pituitary POMC and 22-kDa proACTH levels, whereas ACTH was significantly affected. The ACTH precursor (POMC plus 22-kDa proACTH) to ACTH ratio in the anterior pituitary was significantly increased after PVN lesion. Post-PVN lesion, fetal plasma ACTH(1-39), was below the limit of detection, whereas ACTH precursors (POMC plus 22-kDa proACTH) were not affected. In the inferior region of the anterior pituitary, 40-50% of corticotropes had detectable SPC3 hybridization signal, and PVN lesion did not change the extent of colocalization of POMC and SPC3, or SPC3 mRNA levels within corticotropes. Neither the percent of corticotropes in the superior region containing SPC3 hybridization (7-12%) or hybridization signal strength was altered in response to PVN lesion. In conclusion, the fetal PVN is necessary for sustaining adequate anterior pituitary processing of POMC to ACTH and ACTH release needed for maturing the adrenal cortex in the sheep fetus.

Central regulation of the hypothalamic-pituitary-adrenal axis during fetal development in the Guinea-pig

We have previously shown that the foetal guinea-pig hypothalamic-pituitary-adrenal (HPA) axis is activated near the time of parturition and that this is associated with changes in limbic glucocorticoid receptors (GR) and mineralocorticoid receptors. In the present study, we hypothesized that the foetal hypothalamic paraventricular nucleus (PVN) and pituitary contribute significantly to foetal HPA drive but that these areas remain sensitive to negative feedback by circulating glucocorticoids in late gestation. However, we observed decreased corticotrophin-releasing hormone mRNA expression in the PVN and decreased pro-opiomelanocortin (POMC) mRNA levels in the anterior pituitary with advanced gestational age. The reduction in POMC mRNA expression was likely the result of negative feedback via circulating glucocorticoids because GR mRNA was unchanged during development in the foetal pituitary. Furthermore, we found that maternally administered glucocorticoids significantly decreased foetal pituitary POMC mRNA expression in a dose-dependent manner at gestational day (gd) 62 with male foetuses being more sensitive to these effects. These findings show that the foetal HPA axis remains highly sensitive to glucocorticoid feedback even as plasma adrenocorticotropic hormone and cortisol levels are elevated at the end of gestation.

Angiotensin-induced vasopressin release and activation of hypothalamic neuron in pre-term fetuses

Our previous studies have shown that central administration of angiotensin II (ANG II) causes vasopressin release in the near-term fetus in utero as evidence that the hypothalamic-neurohypophysial system has relatively matured before birth. However, it is still unknown whether the vasopressin controlling centers have been functionally developed in younger fetuses. This study determined fetal plasma vasopressin levels and hypothalamic vasopressin neuron activity in the chronically instrumented pre-term ovine fetuses. Introcerebroventricular (i.c.v.) administration of ANG II did not affect fetal plasma osmolality and sodium concentrations. However, fetal plasma vasopressin levels were significantly increased ( approximately 3-fold) in response to central injection of ANG II. Central ANG II also induced vasopressin-neuron activity marked with c-fos expression in the fetal hypothalamus at pre-term. In addition, the fetal organum vasculosum of the lamina terminalis and the subfornical organ were activated. The results suggest that hypothalamic-neurohypophysial system has been relatively intact and functional at 70% gestational age, and that central angiotensin is important in inducing fetal vasopressin release in utero.

Effects of repeated maternal stress on FOS expression in the hypothalamic paraventricular nucleus of fetal rats

The effects of repeated prenatal stress with different severity (restraint and immobilization) on Fos expression in the maternal and fetal hypothalamic paraventricular nucleus (PVN) were examined in rats. Acute stress treatment was performed for 30 min on gestational day 21, and repeated stress treatment for 30 min daily for 5 days from gestational days 17-21. In the parvocellular region of the maternal PVN, the stress-induced increases in the number of Fos-immunoreactive neurons were smaller in the repeated stress groups than the acute stress groups, indicating an adaptation of Fos expression to repeated stress. The attenuated Fos expression observed in the maternal PVN following repeated mild stress did not occur in the fetal PVN. In contrast, repeated immobilization stress caused a much smaller increase in Fos expression in the fetal PVN than did acute immobilization stress. The reduced Fos expression in the fetal PVN following repeated severe stress was thought to be due to cell death, since the fetal PVN in the chronic immobilization group revealed a reduction in the total number of cells and an increase in the number of apoptotic cells. In the female but not male fetuses, repeated restraint stress induced a significant increase in the number of apoptotic cells in the PVN. These findings suggest that the fetal PVN shows no adaptation of Fos expression to repeated maternal stress, but great vulnerability to cell death, including apoptosis. In addition, stress-induced apoptosis may more easily occur in the fetal PVN in females than males.

Vasopressin mechanism-mediated pressor responses caused by central angiotensin II in the ovine fetus

AVP not only influences renal water excretion but also has profound cardiovascular effects in adults. Our recent studies have demonstrated that central angiotensin induced fetal pressor responses accompanied with AVP release. However, little is known of hormonal mechanisms in angiotensin-mediated fetal blood pressure (BP) changes. The present study determined AVP mechanisms in central angiotensin-mediated fetal pressor responses. The V1-receptor antagonist or V2-receptor antagonist was infused intravenously into the ovine fetus at 90% gestation. Angiotensin II (Ang II; 1.5 microg/kg) was then injected intracerebroventricularly into the chronically instrumented fetus. Ang II produced a significant increase in fetal systolic, diastolic, and mean arterial pressure adjusted to amniotic pressure (A-MAP). The enhanced fetal A-MAP was associated with intense c-fos expression in the central putative cardiovascular area: the paraventricular nuclei (PVN). Double labeling demonstrated that a number of the AVP-containing neurons in the PVN were expressing c-fos in response to central Ang II. Consistent with the activation of AVP neurons in the PVN, fetal plasma AVP was markedly enhanced. Fetal i.v. V1-receptor antagonist or V2-receptor antagonist had no effect on either fetal or maternal baseline BP. However, intracerebroventricular Ang II-increased BP was partially inhibited, although not completely abolished, by the V1-receptor blockade. In contrast, fetal i.v. infusion of V2-receptor antagonist had no effect on the pressor responses induced by central Ang II. The results suggest that the central Ang II-mediated pressor responses at the last third of gestation is mediated partially by the AVP mechanism via V1 not V2 receptors.

Development of paraventricular nucleus (PVN) associated with immunoreactivity to tyrosine hydroxylase (TH) in the second half of gestation

The aim of this study was to shed more light on the developmental characteristics of human paraventricular nucleus (PVN) and hypothalamus in general, using modern immunohistochemical techniques to detect the activity of tyrosine hydroxylase (TH) in the synthesis of catecholamine (CA). Fetal brains were examined at 12, 16, 20 and 23 weeks gestation. Immunohistochemical staining used for sections is a qualitative method for detection and distribution of the chosen protein. The amplification of positive signals was carried out using AVIDIN/BIOTIN technique. The first positive results were obtained at 16th week of gestation and immunoreactivity) grew with the advencement of gestation and that there was no homogenisation of immunoreactivity in some parts of the nucleus. The PVN showed to be an oval formation. nearly parallel with the lateral wall of the 3rd cerebral ventricle, where the upper apex of the nucleus is situated closer to the ependymal ventricular layer. There was an ascendant dorsal movement of the nucleus in the embryonic phase of hypothalamus development.

Glucocorticoids and sex-dependent development of brain glucocorticoid and mineralocorticoid receptors

We have previously shown that repeated antenatal synthetic glucocorticoid exposure has sex-specific effects on hypothalamic-pituitary-adrenal development in the fetal and adult guinea pig. However, little is known about the mechanisms that underlie these sex-specific outcomes. In the current study we demonstrated that glucocorticoid receptors (GR) and mineralocorticoid receptors (MR) exhibit sex differences in their temporal and spatial expression during fetal and early postnatal life. During development, we observed decreased GR mRNA in the paraventricular nucleus, decreased MR mRNA and MR protein in the hippocampus, and increased GR mRNA and GR protein in the hippocampus. We have also shown that on gestational d 50, maternally administered betamethasone (BETA) reduces fetal plasma ACTH and cortisol concentrations. BETA significantly affected hippocampal MR protein expression, and this effect was greatest in males. BETA was unable to autoregulate GR protein during fetal life, indicating that regulation of brain corticosteroid receptors is fundamentally different in fetal compared with adult life. The sex differences in the pattern of GR and MR expression during development may indicate different windows of vulnerability to prenatal glucocorticoid exposure in fetal life.

Materno-fetal coordination of stress-induced Fos expression in the hypothalamic paraventricular nucleus during pregnancy

This study investigates whether maternal stress during pregnancy induces maternal and fetal hypothalamic paraventricular nucleus (PVN) neuronal activation and the effects of maternal stress on fetal hypothalamic and PVN brain-derived neurotrophic factor (BDNF) expression. Pregnant rats were exposed to three types of maternal stress with varying severity (restraint, forced walking and immobilization) for 30 min on gestational day 21. Severity of stress was assessed by measurement of maternal plasma corticosterone 30 min following the stimulus. Maternal plasma corticosterone increased in each stress response group (immobilization>forced walking>restraint). Further, the expression of Fos protein, a marker of neuronal activation, increased in the fetal and maternal PVN in direct relation to the severity of stress treatments. Forced walking and immobilized stress, but not restraint stress, significantly increased BDNF expression in the fetal hypothalamus.These findings suggest that the fetal hypothalamic-pituitary-adrenal (HPA) response following maternal stress mirrors maternal HPA activation. In addition, BDNF may play a role in protecting fetal brain neurons from damage caused by severe stress.

The hypothalamic magnocellular neurosecretory system in developing rats

Studies concerning the development of the magnocellular system are scarce and discordant in literature. We carried out an immunohistochemical study on supraotic and paraventricular hypothalamic nuclei using antivasopressin and antioxytocin antibodies in developing rats between the 15th day of intrauterine life and the 6th day of postnatal life. In addition, we performed RT-PCR experiments to establish the stage at which these hormones appear and neurosecretory activity commences. The results showed that supraoptic and paraventricular nuclei appear, respectively, on the 16th and the 18th day of intrauterine life and both immediately synthetize vasopressin neurohormone. By contrast, synthesis of oxytocin takes place from the 2nd day after birth. Probably, these nuclei synthetize oxytocin in conjunction with the decline of placental maternal oxytocin.

Intact osmoregulatory centers in the preterm ovine fetus: Fos induction after an osmotic challenge

We previously demonstrated a functional systemic dipsogenic response in the near-term fetal sheep (128-130 days; 145 days = full-term) with swallowing activity stimulated in response to central and systemic hypertonic saline. Preterm fetal sheep (110-115 days) do not consistently demonstrate swallowing in response to hypertonic stimuli, and it is unclear whether this is due to immaturity of osmoreceptor mechanisms or neuronal pathways activating swallowing motor neurons. To determine whether osmoreceptive regions in the preterm fetus are activated by changes in plasma tonicity, we examined Fos expression with immunostaining in these neurons in response to an osmotic challenge. Nine preterm fetal sheep [five hypertonic saline-treated fetuses (Hyp) and four isotonic saline-treated fetuses (Iso)] were prepared with vascular and intraperitoneal catheters. Seventy-five minutes before tissue collection, hypertonic (1.5 M) or isotonic saline was infused (12 ml/kg) via an intraperitoneal catheter to fetuses. Brains were examined for patterns of neuronal activation (demonstrated by Fos protein expression). Hyp demonstrated increases in plasma osmolality (~10 mosmol/kg H(2)O) and Na concentrations (5 meq/l). Increased Fos expression was detected in Hyp in the organum vasculosum of the lamina terminalis (OVLT), subfornical organ (SFO), median preoptic nucleus (MnPO), supraoptic (SON), and paraventricular nuclei (PVN) compared with Iso animals. Neuronal activation within the OVLT, SFO, and MnPO indicates intact osmoregulatory mechanisms, whereas activation of the SON and PVN suggests intact fetal neural pathways to arginine vasopressin neurons. These results suggest that preterm fetal swallowing insensitivity to osmotic stimuli may be due to immaturity of integrated motor neuron pathways.

The developmental program of the hypothalamus and its disorders

The hypothalamus integrates physiological processes essential for survival and reproduction. Recent studies have shown that developmental events can affect these processes. Pathways required for the induction of the ventral midline of the hypothalamus or for the differentiation of specific hypothalamic lineages have the potential of causing endocrine and metabolic disorders, including obesity. Also, some genes with paternal monoallelic expression are involved in the development of hypothalamic centers that are critical physiological regulators. Developmental defects affecting the hypothalamus might represent a more frequent cause of clinical disorders than previously suspected.

Defective development of secretory neurones in the hypothalamus of Arnt2-knockout mice

BACKGROUND

Within the basic region-helix-loop-helix (bHLH)-PAS family of transcription factors, Arnt and Arnt2 play unique roles; these two factors not only heterodimerize with themselves, but also with other members of this family and they act as transcription regulators which bind to specific DNA elements. Whereas Arnt is broadly expressed in various tissues, the expression of Arnt2 is known to be limited to the neural tissues.

RESULTS

To elucidate the function of Arnt2 in detail, we cloned the mouse Arnt2 gene and its gene structure was determined. We subsequently generated germ line Arnt2 mutant mice by gene targeting technology. Heterozygous Arnt2 mice were viable, but homozygous Arnt2 gene knockout mice died shortly after birth. Histological and immunological analyses revealed that the supraoptic nuclei (SON) and the paraventricular nuclei (PVN) are hypocellular. Moreover, secretory neurones identified by the expression of neurosecretory hormone such as arginine vasopressin, oxytocin, corticotrophin-releasing hormone and somatostatin are completely absent in SON and PVN in the mutant Arnt2 mice. Consistent with these observations, prospective SON and PVN neurones which express Brn2 appeared around E13.5 in the mantle zone, but no neurones which expressed the neurosecretory hormones were found in the SON and PVN regions.

CONCLUSIONS

These data show that the transcription factor Arnt2 controls the development of the secretory neurones at the later or final stages of differentiation rather than at the beginning stage. Strikingly similar observations have been reported with the Sim1 deficient mice. Taken together, our results demonstrate that Arnt2 is an indispensable transcription factor for the development of the hypothalamus, and suggest that Arnt2 is an obligatory partner molecule of Sim1 in the developmental process of the neuroendocrinological cell lineages.

Osmotic threshold and sensitivity for vasopressin release and fos expression by hypertonic NaCl in ovine fetus

In adults, hyperosmolality stimulates central osmoreceptors, resulting in arginine vasopressin (AVP) secretion. Near-term fetal sheep have also developed mechanisms to respond to intravascular hypertonicity with stimulation of in utero AVP release. However, prior studies demonstrating fetal AVP secretion have utilized plasma tonicity changes greater than those required for adult osmotically induced AVP stimulation. We sought to examine near-term fetal plasma osmolality threshold and sensitivity for stimulation of AVP secretion and to correlate plasma hormone levels with central neuronal responsiveness. Chronically instrumented ovine fetuses (130 +/- 2 days) and maternal ewes simultaneously received either isotonic or hypertonic intravascular NaCl infusions. Maternal and fetal plasma AVP and angiotensin II (ANG II) levels were examined at progressively increasing levels of plasma hypertonicity. Intravenous hypertonic NaCl gradually elevated plasma osmolality and sodium levels. Both maternal and fetal plasma AVP increased during hypertonicity, whereas ANG II levels were not changed. Maternal AVP levels significantly increased with a 3% increase in plasma osmolality, whereas fetal plasma AVP significantly increased only at higher plasma osmolality levels (over 6%). Thus the slope of the regression of AVP vs. osmolality was greater for ewes than for fetuses (0.232 vs. 0.064), despite similar maternal and fetal plasma osmolality thresholds for AVP secretion (302 vs. 304 mosmol/kg). Hyperosmolality induced Fos immunoreactivity (FOS-ir) in the circumventricular organs of the fetal brain. FOS-ir was also demonstrated in the fetal supraoptic and paraventricular nuclei (SON and PVN), and double labeling demonstrated that AVP-containing neurons in the SON and PVN expressed Fos in response to intravenous NaCl. These results demonstrate that, in the ovine fetus at 130 days of gestation, neuroendocrine responses to cellular dehydration are functional, although they evidence a relatively reduced sensitivity for AVP secretion compared with the adult.

Brainstem catecholaminergic neurons activated by hypoxemia express GR and are coordinately activated with fetal sheep hypothalamic paraventricular CRH neurons

In late gestation, challenges to fetal homeostasis are accompanied by increases in adrenocorticotropin (ACTH) concentrations in fetal peripheral plasma and Fos (c-fos protein) activation in corticotropin-releasing hormone (CRH) neurons of the fetal hypothalamic paraventricular nucleus (PVN). In adults, ventrolateral brainstem catecholaminergic (CA) neurons (A1/C1, A2/C2) project to the parvocellular neurons of the PVN, possess glucocorticoid receptors (GR) and are Fos activated in parallel with CRH neurons of the PVN during hypoxia. Such observations suggest a role for the aforementioned medullary neurons in the function of the hypothalamo-pituitary-adrenal axis. The present study utilized late gestation fetal sheep, stereotaxic methodology and retrograde axon tracing and immunocytochemical techniques to investigate the relationship between activation of fetal brainstem CA neurons and activation of fetal PVN CRH immunopositive neurons in response to hypoxemia. Results indicated that: (1) the largest brainstem CA projection to PVN CRH neurons is from A1/C1 neurons, (2) brainstem neurons exhibit GR immunostaining and (3) brainstem CA neurons show a strong correlation (A1/C1 - r(2)=0.894, P<0.005; A2/C2 - r(2)=0. 848; P<0.002) of Fos activation with Fos activation in PVN CRH cells. We conclude that in late gestation the brainstem A1/C1 and A2/C2 areas are in position to influence the function of the hypothalamo-pituitary-adrenal axis during hypoxemic challenges to homeostasis in a fashion similar to that which has been demonstrated in the adult rat.

The murine Otp homeobox gene plays an essential role in the specification of neuronal cell lineages in the developing hypothalamus

Hypothalamic nuclei, including the anterior periventricular (aPV), paraventricular (PVN), and supraoptic (SON) nuclei strongly express the homeobox gene Orthopedia (Otp) during embryogenesis. Targeted inactivation of Otp in the mouse results in the loss of these nuclei in the homozygous null neonates. The Otp null hypothalamus fails to secrete neuropeptides somatostatin, arginine vasopressin, oxytocin, corticotropin-releasing hormone, and thyrotropin-releasing hormone in an appropriate spatial and temporal fashion, and leads to the death of Otp null pups shortly after birth. Failure to produce these neuropeptide hormones is evident prior to E15.5, indicating a failure in terminal differentiation of the aPV/PVN/SON neurons. Absence of elevated apoptotic activity, but reduced cell proliferation together with the ectopic activation of Six3 expression in the presumptive PVN, indicates a critical role for Otp in terminal differentiation and maturation of these neuroendocrine cell lineages. Otp employs distinct regulatory mechanisms to modulate the expression of specific molecular markers in the developing hypothalamus. At early embryonic stages, expression of Sim2 is immediately downregulated as a result of the absence of Otp, indicating a potential role for Otp as an upstream regulator of Sim2. In contrast, the regulation of Brn4 which is also expressed in the SON and PVN is independent of Otp function. Hence no strong evidence links Otp and Brn4 in the same regulatory pathway. The involvement of Otp and Sim1 in specifying specific hypothalamic neurosecretory cell lineages is shown to operate via distinct signaling pathways that partially overlap with Brn2.

Regulation of glucocorticoid receptor mRNA and heat shock protein 70 mRNA in the developing sheep brain

Fetal hypothalamo-pituitary-adrenal (HPA) activity increases dramatically at term in sheep, however, little is known about the regulation of glucocorticoid feedback in the developing brain. Heat shock protein 70 (hsp70) is closely associated with glucocorticoid actions within the cell. We hypothesized that there is a decrease in glucocorticoid negative feedback in the brain, near term, resulting from changes in the expression of glucocorticoid receptors (GR) and hsp70. Brains were removed at various stages of development. GR mRNA levels in the paraventricular nucleus (PVN) and cortex, and hsp70 mRNA in the PVN were determined by in situ hybridization. In the hippocampus, GR mRNA levels were measured by Northern analysis. In the PVN, GR mRNA was present by d60. GR mRNA levels reached a peak at d100-110, but then decreased significantly with progression of gestation, and were lowest at term. Hippocampal GR mRNA levels were highest on day 130 of gestation, decreasing to low levels at term. In the cerebral cortex, GR mRNA levels were expressed at high levels in all layers of the cortex by day 110 of gestation with levels decreasing to term. Hsp70 mRNA was present in both parvocellular and magnocellular regions of the PVN, and there was no significant change in late gestation. In conclusion, (1) The high levels of GR mRNA present in the PVN, hippocampus and cerebral cortex during fetal life are likely important in development of these structures at a time when circulating glucocorticoids are low. (2) Changes in GR mRNA levels in the PVN are not associated with alterations in the expression of hsp70. (3) The decrease in GR mRNA in the hippocampus and PVN in late gestation, at a time when fetal plasma cortisol is increasing, likely facilitates maintained hypothalamic drive to the pituitary corticotroph.

Somatostatin modulation of excitatory synaptic transmission between periventricular and arcuate hypothalamic nuclei in vitro

Hypophysiotropic somatostatin (SRIF) and growth hormone-releasing hormone (GHRH) neurons are primarily involved in the neurohormonal control of growth hormone (GH) secretion. They are located in periventricular (PEV) and arcuate (ARC) hypothalamic nuclei, respectively, but their connectivity is not well defined. To better understand the neuronal network involved in the control of GH secretion, connections from PEV to ARC neurons were reconstructed in vitro and neuronal phenotypes assessed by single-cell multiplex RT-PCR. Of 814 stimulated PEV neurons, monosynaptic responses were detected in only 45 ARC neurons. Monosynaptic excitatory currents were detected in 29 ARC neurons and inhibitory currents in 16, indicating a 2/1 ratio for excitatory versus inhibitory connections. Galanin (GAL), NPY, pro-opiomelanocortin (POMC), and SRIF mRNAs were detected in neurons from both nuclei but GHRH mRNA almost exclusively in ARC. Among the five SRIF receptors, only sst1 and sst2 were expressed, in 94% of ARC and 59% of PEV neurons, respectively. Of 128 theoritical combinations between neuropeptides and sst receptors, only 22 were represented in PEV and 25 in ARC. For PEV neurons, neuropeptide phenotypes did not influence excitatory connections. However, the occurrence of presynaptic sst receptors on GAL and SRIF PEV neurons significantly increased their probability of connection to ARC neurons. GHRH ARC neurons expressing sst2, but not sst1, receptors were always connected with PEV neurons. Physiological responses to sst1 (CH-275) or sst2 (Octreotide) agonists were always correlated with the detection of respective sst mRNAs. In conclusion, 1) SRIF-modulated excitatory transmission develops in vitro from PEV to ARC neurons, 2) ARC GHRH neurons bearing sst2 receptors appears directly controlled by fast glutamatergic transmission from PEV neurons simultaneously expressing one to four neuropeptides, 3) GHRH neurons bearing sst1 receptors lack this control, and 4) these results suggest that fast excitatory neurotransmission and neuropeptide modulation can derive from a small subset of PEV hypothalamic neurons targeted at ARC neuronal subpopulations.

Ovine fetal estrogen sulfotransferase in brain regions important for hypothalamus-pituitary-adrenal axis control

Ovine parturition is initiated by increases in fetal hypothalamus-pituitary-adrenal (HPA) axis activity which, in turn, increase placental estrogen biosynthesis and ultimately increase uterine contractility. In addition to their action in the uterus, estrogens in fetal plasma augment fetal corticotropin (ACTH) secretion. In late gestation, estrone sulfate is more abundant in fetal plasma than unconjugated estrone and it is possible that there is interconversion of sulfoconjugated and unconjugated steroids within the fetal brain. We studied hypothalamus and brainstem tissue from fetal, neonatal, and adult sheep to test the hypothesis that the ovine brain contains estrogen sulfotransferase. Although no significant ontogenic pattern was revealed, the presence of estrogen sulfotransferase within the hypothalamus and brainstem was detectable. Immunohistochemistry revealed the presence of estrogen sulfotransferase in the paraventricular nucleus of the hypothalamus, the nucleus of the tractus solatarius, and the rostral ventral lateral medulla. We conclude that ovine fetal hypothalamus and brainstem contain estrogen sulfotransferase in brain regions important for HPA axis control.

Corticotropin-releasing factor receptor expression in the pituitary of fetal sheep after lesion of the hypothalamic paraventricular nucleus

Both the capacity of CRF to release ACTH and the number of binding sites for CRF in the anterior pituitary decline during the final weeks of gestation in fetal sheep. The present study examined regulation of pituitary CRF receptor expression by the hypothalamic paraventricular nucleus (PVN) during late gestation in fetal sheep. Bilateral radiofrequency lesions of the PVN (PVN-Lx; n = 4) or sham lesions (SHAM; n = 5) were performed in fetal sheep at 118-122 days of gestational age (dGA). Pituitary glands from PVN-Lx and SHAM fetuses were collected at 139-142 dGA (term, approximately 148 dGA). Dual-label in situ hybridization was performed using a digoxigenin-labeled ovine POMC complementary RNA, together with a 35S-labeled ovine CRF type I (CRF1) receptor complementary RNA, to localize and quantify CRF1 receptor mRNA in POMC-hybridizing cells. Binding of [125I]-ovine CRF was also examined in the fetal pituitary of both PVN-Lx and SHAM fetuses using in situ autoradiography. The hybridization signal for the CRF1 receptor mRNA was primarily restricted to POMC-expressing cells in the anterior pituitary of both PVN-Lx and SHAM fetuses; no hybridization signal for the CRF1 receptor was observed in the neurointermediate lobe (NIL) in either group. The hybridization signal for CRF1 receptor mRNA in anterior pituitary corticotropes of PVN-Lx fetuses was significantly lower in both the inferior and superior regions of the anterior pituitary, compared with SHAM fetuses (P < 0.05). In the inferior region of the anterior pituitary, the percentage of POMC-hybridizing cells containing CRF1 receptor hybridization signal was significantly greater in PVN-Lx (90+/-7%; mean +/- SEM), compared with SHAM (67+/-6%; P < 0.05) fetuses. No differences in the percentage of POMC cells containing CRF1 receptor hybridization signal were observed in the superior region of the anterior pituitary between PVN-Lx (89+/-8%) and SHAM (87+/-9%). Binding of [125I]-ovine CRF (oCRF) was significantly greater in anterior pituitaries of PVN-Lx (140+/-19 mean arbitrary densitometry U +/- SEM), compared with SHAM (73+/-23; P < 0.05) fetuses. For both PVN-Lx and SHAM fetuses, there were no differences within group in [125I]-oCRF binding between the inferior and superior regions of the anterior pituitary. A weak, but significant (P < 0.05), autoradiographic signal for [125I]-oCRF binding was observed in the NIL of both SHAM and PVN-Lx fetal sheep. The level of [125I]-oCRF binding was significantly lower in the NIL, compared with anterior pituitary, for both SHAM (P < 0.01) and PVN-Lx fetuses. There were no differences in [125I]-oCRF binding in the NIL between SHAM and PVN-Lx fetal sheep. Our findings support a role for the PVN in regulating anterior pituitary CRF1 receptor expression in the late-gestation sheep fetus.

Distinct populations of hypothalamic dopaminergic neurons exhibit differential responses to brain-derived neurotrophic factor (BDNF) and neurotrophin-3 (NT3)

We have previously demonstrated that differentiation of hypothalamic dopaminergic (DA) neurons can be induced in culture by their pituitary intermediate lobe target cells, through both membrane and diffusible factors. We also showed that subpopulations of DA neurons from the arcuate nucleus only, not the periventricular area, can respond to the target. Here we investigated the possibility that both neuronal subsets could also respond differentially to brain-derived neurotrophic factor (BDNF) or neurotrophin-3 (NT3). Addition of NT3, but not BDNF, enhanced growth and branching of neurites, tyrosine hydroxylase (TH) as well as increasing levels of cultured arcuate DA neurons. Conversely, BDNF, but not NT3, affected the same parameters in cultured periventricular DA neurons. The neurotrophins thus affect DA neurons in a structure and neuronal type-selective manner, since general neuronal markers were not affected by either neurotrophin. Neurotrophin effects were reversed by addition of specific antibodies directed against them or their respective receptors, TrkB or TrkC. By themselves, the antibodies inhibited development of DA neurons below that of control cultures, suggesting involvement of endogenous neurotrophins. BDNF and NT3 were indeed found in both arcuate and periventricular neurons and in the intermediate lobe. BDNF was always present as the mature peptide. The mature form of NT3 was only detected in the periventricular area; a precursor-like heavier form was present in all tissues studied. The present data suggest that NT3, but not BDNF, could participate in the differentiating action of intermediate lobe cells on arcuate DA neurons.

Fos response of fetal sheep anterior circumventricular organs to osmotic challenge in late gestation

We hypothesized that the anterior circumventricular organs (ACVO) and the supraoptic (SON) and hypothalamic paraventricular nuclei (PVN), among other structures that play a role in sensing extracellular body fluid volume and composition in postnatal animals (as demonstrated by Fos protein production by the immediate-early gene c-fos), would show similar activation in fetal sheep during an osmotic challenge. The brains of 10 fetal sheep [6 treated, 4 controls; 129-131 days of gestational age (dGA) = 0.87 gestation] were immunostained for Fos. Seventy-five minutes before tissue collection the dams were given intravenous 20% mannitol (1 ml . min-1 . kg-1 for 10 min). Subsequently, the ACVO, SON, and PVN were scored for the amount of neuronal Fos immunostaining. The subfornical organ (SFO; 24.5 +/- 9.0 vs. 1.7 +/- 1.2), the organum vasculosum of the lamina terminalis (OVLT; 26.8 +/- 5.6 vs. 7.0 +/- 2.0), the SON (39.8 +/- 3.0 vs. 0.15 +/- 0.1), and the PVN (59.8 +/- 7.9 vs. 0.7 +/- 0.7) had increases (P < 0.05) in the average number of Fos-positive cells per field compared with controls, whereas the median preoptic nucleus did not. Double immunostaining for Fos and arginine vasopressin (AVP) or oxytocin (OT) indicated that AVP- but not OT-immunopositive neurons in SON and PVN respond to osmotic challenge. These results demonstrate that the SFO, OVLT, SON, and PVN are activated by osmotic challenge in fetal sheep at 130 dGA.

Ontogeny of the vasopressin and oxytocin RNAs in the mouse hypothalamus

The single-copy genes encoding the vasopressin and oxytocin prepropeptides are closely linked in mouse genome, being separated by an intergenic region of only 3 kbp. These genes are expressed in anatomically defined hypothalamic neurons--in the adult rodent, vasopressin is synthesised in the paraventricular nucleus and the supraoptic nucleus, and in the dorsomedial region of the suprachiasmatic nucleus, whilst oxytocin is expressed in the supraoptic nucleus and paraventricular nucleus, but not in the suprachiasmatic nucleus. The molecular mechanisms that mediate the cell-specific and developmental expression patterns of the two transcription units within the vasopressin-oxytocin locus remain to be elucidated. As a first step in this process, we have used in situ hybridisation to study the expression of the RNAs encoded by the linked vasopressin and oxytocin genes during the development of the mouse hypothalamus. We have revealed a hierarchy of gene activation events, with vasopressin first being observed in presumptive supraoptic nucleus at day 13.5, and in the paraventricular at day 14.5. Oxytocin is seen first in the paraventricular at day 15.5; expression in the supraoptic nucleus is clearly seen at day 18.5. As early as day 15.5, the vasopressin and oxytocin RNAs are expressed in different groups of neurons.

The effects of fetal adrenalectomy at 110 days gestational age on AVP and CRH mRNA expression in the hypothalamic paraventricular nucleus of the ovine fetus

AVP and CRH produced in the parvocellular neurons of the paraventricular nucleus (PVN) have both been implicated in the regulation of anterior pituitary ACTH synthesis and secretion. In sheep, fetal ACTH secretion increases around 120 days gestational age (dGA). Little is known about adrenal regulation of AVP and CRH immediately prior to this critical period. We investigated the effects of adrenalectomy and subsequent cortisol (F) administration on PVN AVP and CRH mRNA in the fetal sheep PVN at 109-125 dGA. At 109-113 dGA, fetal sheep adrenals were removed (ADX)(n = 8); or sham surgery performed (CONT)(n = 4). From day 6 post ADX, maternal plasma cortisol and fetal plasma ACTH and cortisol levels were determined daily by radioimmunoassays. From day 7 post ADX, cortisol (4 micrograms/min) was continuously infused intravenously to four ADX fetuses (ADX + F). Fetal hypothalami were collected at 123-125 dGA, and studied by in-situ hybridization and quantitative autoradiography for AVP and CRH mRNA. Plasma cortisol levels remained low in CONT and ADX fetuses (< 4.9 ng/ml), while during cortisol infusion to ADX + F fetuses, plasma F increased to 16.4 +/- 2.2 and 22.3 +/- 3.2 ng/ml (mean +/- S.E.M.) on day 10 and 13, respectively. Plasma ACTH levels increased significantly in ADX compared with CONT fetuses. This ACTH increase was completely suppressed in ADX + F fetuses. AVP mRNA abundance in the whole PVN was the same in all three groups, however, a separate analysis of AVP mRNA abundance in parvocellular and magnocellular regions of the PVN revealed that AVP mRNA in the parvocellular PVN showed a significant increase in ADX and suppression in ADX + F fetuses when compared to CONT. AVP mRNA in the magnocellular PVN remained unchanged. PVN CRH mRNA expression was augmented in ADX and suppressed in ADX + F when compared to CONT fetuses. We conclude that in fetal sheep at 109-125 dGA: AVP and CRH mRNA abundance in the parvocellular region of the PVN are increased by adrenalectomy and that cortisol inhibits this increase.

Fetal sheep pituitary proopiomelanocortin in late gestation: effect of bilateral lesions of the paraventricular nucleus on regional and cellular messenger ribonucleic acid levels

Previous experiments have clearly indicated that the successful completion of ovine gestation is dependent upon fetal adrenocortical maturation and the associated preterm rise in fetal plasma cortisol. The purposes of this study were to: 1) examine pituitary POMC messenger RNA (mRNA) levels during normal fetal development; and 2) examine the effects of bilateral lesion of the fetal paraventricular nucleus (PVN) on levels and spatial distribution of pituitary POMC mRNA. Pituitary glands were collected from intact fetal sheep of four gestational ages [100-107 days gestational age (dga), n = 8; 117-121 dga, n = 9; 126-130 dga, n = 9; 144-147 dga, n = 8]. Lesions of the PVN (PVN Lx; n = 4) or sham lesions (Sham; n = 5) were performed at 118-122 dga. Pituitary glands from PVN Lx and Sham fetuses were collected at 139-142 dga (term approximately 147 dga). POMC mRNA levels were determined by in situ hybridization. POMC transcript levels were determined by both regional analysis (20x magnification) and analysis of individual corticotropes (400x magnification). There was no difference among gestational age groups in superior anterior pituitary (AP) POMC mRNA levels determined by regional or cellular analysis. POMC mRNA levels were significantly greater in the inferior AP at 144-147 dga, compared with other gestational ages, using regional analysis (P = 0.003) or analysis of individual corticotropes (P < 0.01). POMC mRNA levels in the neurointermediate lobe in 126- to 130-dga fetuses were significantly greater than those in younger fetuses (P = 0.005) but not those in 144- to 147-dga fetuses. There was no difference in POMC mRNA levels in the superior AP between PVN Lx and Sham, using regional analysis or analysis of individual corticotropes. In the inferior AP, there was a significant decrease in POMC mRNA levels in PVN Lx, compared with Sham, using both regional analysis (P < 0.01) and cellular analysis (P < 0.01). There was no difference in POMC mRNA levels in the neurointermediate lobe as the result of bilateral PVN Lx. Our findings support that basal AP POMC mRNA levels are heterogenously distributed in the ovine fetal AP, with POMC mRNA levels in the inferior AP being significantly greater than in superior AP, by 144-147 dga. We further found that the higher POMC mRNA levels in the inferior AP reflect significantly higher corticotrope POMC transcripts and not simply a greater density of corticotropes in this AP region. The increase in POMC mRNA levels at 144-147 dga in the inferior AP seems unrelated to the onset of adrenocortical maturation (at approximately 125-130 dga). Finally, we report that increase in corticotrope POMC transcripts during late gestation in the inferior AP requires an intact PVN.

Regulation of CRH and AVP mRNA in the developing ovine hypothalamus: effects of stress and glucocorticoids

Developmental changes in the abundance, localization, and distribution of corticotropin-releasing hormone (CRH) mRNA and arginine vasopressin (AVP) mRNA in the ovine hypothalamus were examined by in situ hybridization. The effects of fetal hypoxemia in the presence or absence of concomitant cortisol in late gestation (day 135) were also investigated. CRH and AVP mRNA were present at low levels within the paraventricular nucleus (PVN) and AVP mRNA was present in the supraoptic nucleus (SON) by day 60 (full term = 147 days). During late gestation, there were increases (P < 0.05, days 140-143 vs. days 100-120) in CRH mRNA, a further increase (P < 0.05, full term vs. days 140-143) at full term (fetuses delivered in active labor), and a subsequent decline postpartum (compared with full term). AVP mRNA in the magnocellular PVN increased (P < 0.05) in late gestation, levels did not change in parvocellular fields compared with full term fetuses, but magnocellular and parvocellular AVP mRNA increased in the newborn (P < 0.05, newborn vs. full term). AVP mRNA in the SON showed a developmental profile similar to that of the PVN, although there was an increase earlier in gestation (P < 0.05, days 100-120 vs. days 60-80). Hypoxemia caused increases (P < 0.05) in CRH mRNA, plasma adrenocorticotropic hormone, and cortisol concentrations, and although magnocellular and parvocellular AVP mRNA appeared elevated, changes just failed to attain significance. Cortisol infusion attenuated the hypoxemia-induced increase in CRH mRNA and adrenocorticotropic hormone but was without effect on basal CRH mRNA levels.

Developmental regulation of preproenkephalin mRNA in the ovine paraventricular nucleus: effects of stress and glucocorticoids

The opioid peptides have profound effects at several levels within the hypothalamo-pituitary-adrenal (HPA) axis. Activation of fetal HPA function occurs during late gestation, and as part of the fetal adaptive response to stress. Changes in the relative levels, localization and distribution of hypothalamic preproenkephalin (PENK) mRNA in the ovine hypothalamic paraventricular nucleus (PVN) during development were examined by in situ hybridization histochemistry. The effects of fetal hypoxemia applied in the presence or absence of concomitant cortisol, to establish negative feedback potential in late gestation were also investigated. PENK mRNA was present at low levels within the PVN, by d60 (term d147). During mid to late gestation, there was an increase in PENK mRNA levels from d60-80 to d100-120, then reaching a peak at d130-140. Levels then decreased dramatically during the last 5-7 days prior to parturition, but increased again in the newborn lamb. Throughout gestation, PENK mRNA was confined exclusively to the parvocellular region of the PVN. Cortisol infusion induced significant decreases (P < 0.05) in PENK mRNA, in normoxemic fetuses at d135 of gestation. The hypoxemic insult, which is known to stimulate plasma ACTH and cortisol, in these fetuses, did not significantly affect PENK mRNA. There was no significant difference in hypoxemia significantly decreased PENK mRNA compared to the saline-infused normoxemic fetuses. Together, these results indicate that the elevation of endogenous fetal cortisol, that occurs at the end of gestation, may act to inhibit expression of the PENK gene in the hypothalamic PVN of the developing ovine fetus.

Development of the human hypothalamus

The hypothalamus has been claimed to be involved in a great number of physiological functions in development, such as sexual differentiation (gender, sexual orientation) and birth, as well as in various developmental disorders including mental retardation, sudden infant death syndrome (SIDS), Kallman's syndrome and Prader-Willi syndrome. In this review a number of hypothalamic nuclei have therefore been discussed with respect to their development in health and disease. The suprachiasmatic nucleus (SCN) is the clock of the brain and shows circadian and seasonal fluctuations in vasopressin-expressing cell numbers. The SCN also seems to be involved in reproduction, adding interest to the sex differences in shape of the vasopressin-containing SCN subnucleus and in its VIP cell number. In addition, differences in relation to sexual orientation can be seen in this perspective. The vasopressin and VIP neurons of the SCN develop mainly postnatally, but as premature children may have circadian temperature rhythms, a different SCN cell type is probably more mature at birth. The sexually dimorphic nucleus (SDN, intermediate nucleus, INAH-1) is twice as large in young male adults as in young females. At the moment of birth only 20% of the SDN cell number is present. From birth until two to four years of age cell numbers increase equally rapidly in both sexes. After this age cell numbers start to decrease in girls, creating the sex difference. The size of the SDN does not show any relationship to sexual orientation in men. The large neurosecretory cells of the supraoptic (SON) and paraventricular nucleus (PVN) project to the neurohypophysis, where they release vasopressin and oxytocin into the blood circulation. In the fetus these hormones play an active role in the birth process. Fetal oxytocin may initiate or accelerate the course of labor. Fetal vasopressin plays a role in the adaptation to stress--caused by the birth process--by redistribution of the fetal blood flow. Corticotropin-releasing hormone (CRH) neurons of the PVN play a central role in stress response. Thus fetal CRH neurons may play a role in the timing of the moment of birth. Recently, alterations have been described in peptidergic, aminergic and cholinergic transmitters in the hypothalamus in SIDS. Future research will have to establish whether these changes are part of the course of SIDS. A large proportion of the SON and PVN neurons also produce tyrosine hydroxylase (TH). In neonates the majority of TH-immunoreactive neurons colocalizes vasopressin, while in the adult the majority of TH-positive neurons colocalizes oxytocin.(ABSTRACT TRUNCATED AT 400 WORDS)

Early neurogenesis in the anterior hypothalamus of the rhesus monkey

The time of formation of neurons of the various nuclei of the anterior hypothalamus of the rhesus monkey has been determined using [3H]thymidine autoradiography. Plotting of heavily radiolabeled nuclei indicate that in the macaque monkey neurons of the hypothalamus are generated during three weeks, between the 27th and 48th day of the 165 gestational period of this species. The individual nuclei generally display two waves of neuronal generation. Furthermore, a caudolateral to rostromedial gradient of neurogenesis has been observed in the supraoptic nucleus and a dorsal to ventral gradient in the paraventricular nucleus. Thus, neurons of the primate hypothalamus are generated at a much earlier stage of gestation and their generation lasts about three times longer than in rodents. However, most of the basic neurogenetic gradients are remarkably similar, in spite of species specific differences in cytoarchitecture of specific nuclei.

Ontogeny of arginine vasopressin-immunoreactive neurons in the hypothalamus of fetal and newborn sheep

Arginine vasopressin (AVP) is a peptide hormone which is found in neurons within the paraventricular (PVN) and the supraoptic (SON) nuclei of the hypothalamus. In fetal sheep, this neuropeptide is involved in maturational processes and adaptive responses to 'stress'. This study examined the effect of age on the total number and distribution of AVP-containing neurons in the PVN and SON of fetal sheep and newborn lambs by quantitative light-microscopic immunocytochemistry. Serial coronal sections of hypothalami from three groups of animals were studied: fetuses at 104-109 days of gestation (n = 6) comprising the early group, fetuses at 130-139 days of gestation (n = 5) comprising the late group and newborn lambs at 12-20 postnatal days (n = 5) comprising the neonatal group. This period of development was chosen since adaptive mechanisms to stress are operative at or near the time of birth. Hypothalamic dimensions were measured to determine if maturation had an effect on the size of the AVP-containing subregions of the hypothalamus during this period of development. Dimensions included: ventricle height, optic chiasm width, distances from the dorsal margin of the ventricle to the lateral and medial margins of the optic tract, and distance between the medial margins of the optic tracts. As expected, with increase in maturational age, overall dimensions of the AVP-containing subregions increased significantly (p < 0.05). When early- and late-gestation fetuses were compared to newborn lambs, there was a significant increase in the total number of immunoreactive neurons in both the PVN (p < 0.01, Anova) and SON (p < 0.001, Anova) with age. With advancing age, we also observed an increase in the density of AVP neurons in the middle subregion of the PVN and in the midrostral subregion of the SON. These data suggest that, during the late gestational and early postnatal period, de novo synthesis of AVP genes occurs in these hypothalamic nuclei. This study provides a baseline for further investigation to study the effects of stress on these neurons in the developing ovine fetus and newborn lamb.

Formation of neurons in the sexually dimorphic anteroventral periventricular nucleus of the preoptic area of the rat: effects of prenatal treatment with testosterone propionate

An examination was made of neurogenesis in the anteroventral periventricular nucleus (AVPv) of the preoptic area of the rat using bromodeoxyuridine (BrdU), a thymidine analog, and a BrdU-specific antibody. Cells in the AVPv of adult rats were labeled with the antibody when BrdU was injected into pregnant rats once during day 13 to 18 of gestation, but not during day 10 to 12 nor 19 to 20 of gestation nor on postnatal day 1, indicating the neurogenesis of the AVPv occurs during a limited period from day 13 to 18 of gestation. Next, to examine the effects of androgen on neurogenesis, BrdU was injected once on day 15 into pregnant rats that also received injections of testosterone propionate (TP). The number of BrdU-labeled cells in the AVPv was similar in control female and male fetuses and female fetuses from pregnant rats that received daily injections of TP during days 14 to 16, when fetuses were examined on day 17 of gestation. These results suggest that the neurogenesis that was recognized by labeling with BrdU was not affected by the treatment with TP. On day 21 of gestation, BrdU-labeled cells in the AVPv of control male fetuses and female fetuses that received TP during days 14 to 18 were fewer in number than those in female fetuses of the control group, whereas treatments with TP during days 14 to 16 and during days 17 to 18 did not cause any significant decrease in number of BrdU-labeled cells. These findings can support the hypothesis that elimination of a population of cells, for example, by cell death as described previously, is enhanced in male fetuses and in female fetuses treated with TP repetitively.

Ontogeny of prodynorphin gene expression in the rat hypothalamus

Opioid peptides, deriving from prodynorphin, proenkephalin and proopiomelanocortin genes, have been shown to modulate brain development. Prodynorphin gene expression was studied here by in situ hybridization in the developing rat hypothalamus using oligodeoxynucleotide probes. Prodynorphin mRNA-synthetizing cells were observed in the ventromedial hypothalamic nucleus, the supraoptic and the paraventricular nuclei from embryonic days 16, 18 and 21, respectively. We detected no transient expression of prodynorphin gene in the rat hypothalamus. Prodynorphin mRNA-containing cells were also observed prenatally in the striatum, the cortex, the hippocampus and the amygdala. When compared with data from the literature, our results suggest that translation may immediately follow transcription of prodynorphin gene in the supraoptic nucleus. The presence of prodynorphin mRNA in the developing rat hypothalamus also raises the possibility of an involvement of prodynorphin-derived peptides in developmental processes and/or in the maturation of adult neural regulations.

[The survival of anlagen of the human hypothalamic supraoptic and paraventricular nuclei in the cavity of the third ventricle of the rat brain]

We studied survival, integration into surrounding tissues, and morphogenesis of embryonic neurosecretory cells of human supraoptic and paraventricular hypothalamic nuclei transplanted into the third ventricle of the rat brain. It was found that embryonic brain cells survive after xenotransplantation and acquire features characteristic of corresponding cells during normal ontogenesis. These results suggest that the genetic program plays a key role in development of nervous cells.

Levels of corticotropin-releasing hormone messenger ribonucleic acid (mRNA) in the hypothalamic paraventricular nucleus and proopiomelanocortin mRNA in the anterior pituitary during late gestation in fetal sheep

CRH regulates POMC gene expression and subsequent ACTH biosynthesis and release. In sheep, the preterm rise in fetal plasma ACTH commences at approximately 125 days gestation (dGA; 147 dGA = term), preceding the initiation of adrenocortical steroidogenesis. We hypothesized that an increase in CRH expression in the hypothalamic paraventricular nucleus (PVN) and POMC expression in the anterior pituitary in the late gestation sheep fetus may precede adrenal cortex maturation. Fetal sheep were obtained at 105-107 (n = 4), 128-130 (n = 5), and 138-140 (n = 4) dGA. Hypothalami were cryosectioned and subjected to in situ hybridization for ovine CRH mRNA. In all dGA groups, expression of CRH mRNA was observed throughout the rostrocaudal extent of the fetal PVN. The midrostral region of the fetal PVN where the dorsal and ventral divisions of the rostral PVN merge to form a single structure was selected for quantification. The number of copies of CRH probe hybridized per micron 3 were determined to estimate the quantity of hybridized CRH mRNA; the mean estimated CRH mRNA copy number per micron 3 midrostral PVN were 0.064 +/- 0.012 (105-107 dGA), 0.237 +/- 0.048 (128-130 dGA), and 0.108 +/- 0.034 (138-140 dGA; mean +/- SEM copies per micron 3 PVN). CRH mRNA signal significantly increased between 105-107 and 128-130 dGA (P < or = 0.05); 138-140 dGA levels of mRNA were not different from either 105-107 or 128-140 dGA levels. Regional variation in CRH mRNA levels were observed within the midrostral PVN between groups; at 138-140 dGA, a population of lateral midrostral PVN neurons maintain CRH mRNA levels greater than 105-107 dGA (P < 0.05), similar to those at 128-130 dGA. Fetal anterior pituitary RNA was subjected to Northern analysis for POMC mRNA. POMC mRNA levels in fetal anterior pituitaries were 14.1 +/- 2.2 (105-107 dGA), 28.9 +/- 10.9 (128-130 dGA), and 43.2 +/- 6 (138-140 dGA; mean +/- SEM arbitrary units). A significant increase (P < or = 0.05) was observed at 138-140 dGA compared to levels at 105-107 dGA. We conclude CRH mRNA levels in the fetal PVN increase coincident with increased POMC gene expression and the late gestation rise in fetal plasma ACTH. We speculate that a neuroendocrine stimulus at the fetal PVN may precipitate increased levels of CRH mRNA, initiating the maturation of the fetal hypothalamic-hypophyseal-adrenal axis, thus inducing the events of labor and delivery in sheep.

Failure of bilateral paraventricular nuclear lesions to cause hypothalamic hypothyroidism in fetal sheep

Concentrations of T4 indicative of a hyperthyroid state are present in fetal sheep plasma between 100 days gestational age (dGA) and parturition. Fetal pituitary stalk section studies indicate that, as in adults, these high fetal plasma T4 concentrations during pregnancy are controlled by the hypothalamus. We compared peripheral plasma T4 concentrations in fetal sheep with bilateral hypothalamic paraventricular nuclear (PVN) lesions (lesion group; n = 5) to fetal sheep with sham-PVN lesions (sham group; n = 4) at 131 and 146 dGA in the lesion group or at term (mean +/- SEM, 146 +/- 0.9 dGA) in the sham group. Bilateral hypothalamic PVN lesions or sham lesions were placed at 118-122 dGA. Baseline blood samples were taken between 1100-1500 h at 131 dGA in both groups, at term in the sham group, and at 146 dGA in the lesion group. In control sheep, TRH cells were found in the PVN and in a number of extra-PVN sites, and the median eminence received abundant TRH axons. In the lesion group, complete destruction of the PVN bilaterally was confirmed by histology. Extra-PVN TRH neurons remained intact in the lesioned sheep, and axons to the median eminence were reduced, but not eliminated. T4 concentrations in fetal plasma were not different in the lesion group and the sham group at 131 dGA (81 +/- 7 vs. 92 +/- 19 ng/ml) or at term (112 +/- 35 vs. 79 +/- 15 ng/ml), respectively. In contrast, fetal plasma concentrations of cortisol, which were not different in lesion and sham group fetuses at 131 dGA (1.7 +/- 0.3 vs. 1.0 +/- 0.2 ng/ml, respectively), were greatly reduced (P < 0.05) at 146 dGA in the lesion group compared to those in the sham group at term (2.0 +/- 0.5 vs. 58.8 +/- 11.5 ng/ml). We conclude that unlike in adult rats, the ovine fetal PVN is not required to maintain normal plasma T4 concentrations. The many TRH-positive cells that lie outside of the PVN in the fetal sheep appear to enable PVN-lesioned fetuses to remain euthyroid fetuses.

Hypothalamic paraventricular nuclear lesions delay corticotroph maturation in the fetal sheep anterior pituitary

The purpose of this study was to determine whether normal morphological development occurs in pituitary corticotrophs deprived of products of the hypothalamic paraventricular nucleus (PVN), e.g. corticotropin releasing hormone and arginine vasopressin (AVP), after PVN lesions. In addition, we have attempted to ascertain if the neurophysin/AVP-positive fibers innervating the fetal sheep anterior pituitary are affected by PVN lesions. The experimental groups consisted of fetal sheep in which 1) hypothalamic PVN lesions were placed at 118-122 days gestation (dGA) and the fetuses subsequently harvested while still in utero at 157 dGA or more (PVNX; n = 5); 2) sham PVN lesions were placed at 118-122 dGA and subsequently harvested as newborn lambs immediately after birth at 146.5 +/- 0.9 (mean +/- SEM) dGA combined with two uninstrumented fetuses harvested at 144 dGA or more but not in labor (perinatal; n = 6); and 3) no instrumentation was placed, and the fetuses were harvested at 120 dGA (control; n = 4). Two ACTH-immunoreactive cell types were seen in the anterior pituitary: 1) fetal cells: large and variably stained, often columnar, occurring in clusters and arranged in palisades; and 2) adult cells: smaller, darkly staining, and angular, occurring singly or in small groups. Quantification of the distribution of the two ACTH cell types was performed by scanning sections from a one in six series from each pituitary and estimating the percent area of each section in the well that showed adult type staining only. The observer was blind to the treatment group assignment of the sections. The estimated percentages of the portion of the pituitaries of each group that contained adult-type cells only were as follows: PVNX, 42.8 +/- 10.0%; perinatal, 90.9 +/- 2.1%; and control, 3.7 +/- 1.1% (mean +/- SEM; P less than 0.05 for all comparisons). There were no qualitative differences between all groups in the appearance of neurophysin-positive fibers innervating the anterior pituitary. AVP staining was strong in the internal zone of the median eminence in all groups, but was absent in the external zone of PVNX fetuses only. The intermediate pituitary lobes stained darkly in all groups. We conclude that lesions of the PVN at 120 dGA delay development of fetal pituitary corticotrophs, but have no effect on the presence of neurophysin-positive nerve fibers in the anterior pituitary.

Effect of placement of dexamethasone adjacent to the ovine fetal paraventricular nucleus on adrenocortical steroid hydroxylase messenger ribonucleic acid

The preterm rise in the concentration of ovine fetal plasma cortisol that initiates the events of parturition in sheep commences at approximately 125 days of gestational age (dGA; term = approximately 147 dGA). Concurrent with the rise in fetal plasma cortisol, adrenocortical steroid hydroxylase enzyme activity increases. The purpose of this study was 1) to quantitate changes in levels of mRNA for the steroid hydroxylases, 17-hydroxylase cytochrome P450 (P450(17)alpha), side-chain cleavage cytochrome P450 (P450scc), 11 beta-hydroxylase cytochrome P450 (P450(11)beta), and C21-hydroxylase cytochrome P450 (P450C21); and 2) examine the role of the fetal paraventricular nucleus (PVN) in the onset of adrenocortical steroid hydroxylase mRNA expression. Unperturbed fetuses were collected by cesarian section under halothane anesthesia at 105 (n = 4), 120 (n = 4), 126-128 (n = 4), and 136 dGA (n = 3); neonatal animals were collected within 2 h after birth. To examine the role of the fetal PVN in regulation of adrenocortical steroid hydroxylase mRNA expression, ovine fetuses were stereotaxically implanted bilaterally 2 mm lateral to the fetal PVN at 105-107 dGA with either cholesterol (n = 4) or dexamethasone (DEX; n = 3). Implanted fetuses were collected by cesarian section under halothane anesthesia at 126-128 dGA. mRNA for both cytochrome P450(17)alpha and P450scc declined 3-fold from 105 to 120 dGA (P less than or equal to 0.05), and then increased by 126-128 dGA compared to that on 120 dGA (P less than or equal to 0.05) and continued to increase through term. Cytochrome P450C21 increased at 126-128 dGA compared to that on 105 and 120 dGA (P less than or equal to 0.05) and remained elevated through term. Three distinct transcripts [approximately 6.2, 4.2, and 2.5 kilobases (kb)] were observed for cytochrome P450(11)beta; the 4.2-kb transcript was predominant. While total message for P450(11)beta declined over increasing gestational age, no differences were noted for the 4.2-kb transcript until after birth, when levels significantly declined (P less than or equal to 0.025). Placement of DEX adjacent to the fetal PVN prevented reemergence of expression of mRNA for P450(17)alpha and P450scc at 126-128 dGA, but had no effect on mRNA for P450C21 or P45011 beta. We conclude that mRNA for P450(17)alpha and P450scc undergoes a decline in expression concurrent with the previously described period of adrenal hyporesponsiveness from 105-126 dGA, followed by an increase in mRNA that accompanies the preterm rise in fetal plasma cortisol.(ABSTRACT TRUNCATED AT 400 WORDS)

Effect of bilateral lesions of the ovine fetal hypothalamic paraventricular nuclei at 118-122 days of gestation on subsequent adrenocortical steroidogenic enzyme gene expression

Fetal adrenal steroid hydroxylase activity and messenger RNA (mRNA) expression increases concurrent with the preterm rise in fetal plasma cortisol during late gestation in sheep. By placing bilateral lesions of the fetal paraventricular nuclei (PVN) we have previously demonstrated that the fetal PVN is necessary for the initiation of parturition, the late gestation preparturient increase in fetal plasma cortisol and ACTH, and ACTH secretion in response to fetal hypoxemia and hypotension. The purpose of this study was to determine the role of the fetal PVN in the late gestation increase in expression of mRNA for 17 alpha-hydroxylase (P-450(17)alpha), side-chain cleavage (P-450SCC), 11 beta-hydroxylase (P-450(11)beta), 21 hydroxylase (P-450C21), and 3 beta-hydroxysteroid dehydrogenase (3 beta-HSD) in the fetal adrenal. Ovine fetuses were subjected to bilateral lesions of the PVN (Lx; n = 4) or sham lesions (Sh; n = 4) at 118-122 days gestational age (dGA). Lx fetuses were recovered by cesarean section at greater than or equal to 157 dGA; Sh fetuses were recovered immediately postbirth at normal term (146.5 +/- 0.9 dGA). In addition, uninstrumented fetuses were obtained at 145-147 dGA by cesarean section (n = 3). RNA obtained from individual fetal adrenals was subjected to Northern analysis. Lx of the fetal PVN decreased (P less than or equal to 0.05) mRNA for P-450(17)alpha and P-450SCC but did not affect adrenocortical mRNA for P-450C21, P-450(11)beta, or 3 beta-HSD compared to Sh. To determine if the differences observed between Lx and Sh for P-450(17)alpha and P-450SCC mRNA were due to the process of labor, we compared uninstrumented 145-147 dGA to Sh. No differences in adrenal mRNA content were observed for P-450(17)alpha or P-450SCC between these groups. We conclude that in late gestation fetal sheep an intact fetal PVN is necessary for normal gene expression of adrenocortical P-450(17)alpha and P-450SCC while P-450(11)beta, P-450C21, and 3 beta-HSD may be primarily regulated by factors not dependent upon a functional PVN.

[Proliferation and differentiation of nerve cells of the human NN. supraopticus et paraventricularis transplanted to the brain of adult rats]

Human fetal hypothalamic neurosecretory cells (NSC) were studied after transplantation into the third ventricle of the adult rat brain. 3H-thymidine uptake data have shown that the time of origin and proliferation of NSC of grafted tissue and of supraoptic and paraventricular nuclei in normal embryogenesis was similar. Specific staining revealed that the start of neurosecretory function in grafted tissue corresponded to the date when NSC began functioning in normal ontogenesis.

Activation of cFos in ovine fetal corticotropin-releasing hormone neurons at the time of parturition

The ovine fetal pituitary adrenal axis plays an important role in parturition. While lesions of the paraventricular nucleus (PVN) in the fetal sheep delay parturition, suggesting that the PVN is necessary for the processes that lead up to parturition, evidence for stimulation of PVN neurons at time of delivery/labor is lacking. The present study tested the hypothesis that activation (evidenced by expression of the oncogene product cFos) of a specific population of PVN neurons containing CRH accompanies labor in sheep. Monitoring of uterine electromyogram activity determined the onset of labor. The brains of nine fetuses (removed by ceasarian section under anesthesia at gestational ages of 125-145 days) and four newborn sheep were perfused and stained for cFos and CRH. Before labor, less than 5% of fetal paraventricular CRH neurons expressed cFos. Verification that the CRH neurons could express cFos when adequately stimulated was made by exposing an additional group of four preterm animals (125 days gestational age) to hypoxemia; the six untreated fetuses served as controls. Activation of the CRH neurons by hypoxemia produced a rapid induction of cFos in CRH neurons, with approximately 50% of the cells strongly expressing cFos protein 1 h after exposure to hypoxia. At the time uterine contractions were first detected, 70% of CRH neurons expressed cFos, and cFos immunoreactivity persisted until just after birth. cFos staining declined rapidly, reaching prelabor levels in some animals by 2-3 h after birth. These data are consistent with the hypothesis that in addition to their potential role in the processes that initiate parturition over several days before birth, fetal CRH neurons are stimulated during labor, and termination of stimulation probably occurs rapidly after delivery.