The kisspeptin system of the human hypothalamus: sexual dimorphism and relationship with gonadotropin-releasing hormone and neurokinin B neurons.

Kisspeptin signaling via the kisspeptin receptor G-protein-coupled receptor-54 plays a fundamental role in the onset of puberty and the regulation of mammalian reproduction. In this immunocytochemical study we addressed the (i) topography, (ii) sexual dimorphism, (iii) relationship to gonadotropin-releasing hormone (GnRH) neurons and (iv) neurokinin B content of kisspeptin-immunoreactive hypothalamic neurons in human autopsy samples. In females, kisspeptin-immunoreactive axons formed a dense periventricular plexus and profusely innervated capillary vessels in the infundibular stalk. Most immunolabeled somata occurred in the infundibular nucleus. Many cells were also embedded in the periventricular fiber plexus. Rostrally, they formed a prominent periventricular cell mass (magnocellular paraventricular nucleus). Robust sex differences were noticed in that fibers and somata were significantly less numerous in male individuals. In dual-immunolabeled specimens, fine kisspeptin-immunoreactive axon varicosities formed axo-somatic, axo-dendritic and axo-axonal contacts with GnRH neurons. Dual-immunofluorescent studies established that 77% of kisspeptin-immunoreactive cells in the infundibular nucleus synthesize the tachykinin peptide neurokinin B, which is known to play crucial role in human fertility; 56 and 17% of kisspeptin fibers in the infundibular and periventricular nuclei, respectively, contained neurokinin B immunoreactivity. Site-specific co-localization patterns implied that kisspeptin neurons in the infundibular nucleus and elsewhere contributed differentially to these plexuses. This study describes the distribution and robust sexual dimorphism of kisspeptin-immunoreactive elements in human hypothalami, reveals neuronal contacts between kisspeptin-immunoreactive fibers and GnRH cells, and demonstrates co-synthesis of kisspeptins and neurokinin B in the infundibular nucleus. The neuroanatomical information will contribute to our understanding of central mechanisms whereby kisspeptins regulate human fertility.

Neonatal stimulation of 5-HT(2) receptors reduces androgen receptor expression in the rat anteroventral periventricular nucleus and sexually dimorphic preoptic area.

Masculinization of the brain is dependent upon a perinatal surge in testosterone. It also requires a transient decrease in hypothalamic 5-HT concentration and turnover and an increase in androgen receptor (AR) expression during the second postnatal week. We have previously shown that increasing 5-HT activity over this period in male or androgenized female rats feminizes their adult behaviour and also feminizes the size of their anteroventral periventricular nucleus (AVPV) and sexually dimorphic nucleus of the preoptic area (SDN-POA). To investigate the role of 5-HT in sexual differentiation of the brain, 5-HT activity was raised over postnatal days 8-16 in male, female and androgenized female rats by daily administration of the 5-HT(2) receptor agonist (-)[2,5 dimethoxy-4-iodophenyl]-2-amino propane hydrochloride [(-)DOI]. By postnatal day 18, the size of the AVPV and SDN-POA was sexually dimorphic; their sizes were feminized by (-)DOI treatment. In the absence of (-)DOI treatment, there were significantly more AR-immunoreactive cells in the AVPV of males, and in the SDN-POA of males and androgenized females, than in those of females on postnatal day 18. (-)DOI treatment reduced the number of AR-immunoreactive cells in the AVPV and SDN-POA of males and androgenized females, but not of females, by postnatal day 18. These results suggest that 5-HT(2) receptor activation can influence sexual differentiation of the brain by controlling AR expression.

Gonadotropin-releasing hormone neurons express estrogen receptor-beta.

CONTEXT: Recent identification of the second estrogen receptor (ER) isoform (ER-beta) within GnRH neurons of the rodent brain has generated much enthusiasm in the field of neuroendocrine research by questioning the dogma that GnRH cells do not directly sense changes in circulating estrogens. OBJECTIVE: To address the issue of whether GnRH neurons of the human hypothalamus also contain ER-beta, we have performed dual-label immunocytochemical studies. DESIGN: Tissue sections were prepared from autopsy samples of male human individuals (n = 8; age < 50 yr), with sudden causes of death. Technical efforts were made to minimize postmortem interval (<24 h), optimize tissue fixation (use of a mixture of 2% paraformaldehyde and 4% acrolein for four tissue samples), and sensitize the immunocytochemical detection of ER-beta (application of silver-intensified nickel-diaminobenzidine chromogen). MAIN OUTCOME MEASURE: Distribution and percent ratio of GnRH neurons that also contained ER-beta immunoreactivity were analyzed under the light microscope. RESULTS: With acrolein in tissue fixative, nuclear ER-beta immunoreactivity was observed in 10.8-28.0% of GnRH neurons of the four different individuals. ER-beta-containing GnRH neurons were widely distributed in the hypothalamus, without showing a noticeable preference in regional location. CONCLUSIONS: The demonstration of ER-beta and the previous lack of detection of ER-alpha in human GnRH cells indicate that estrogens may exert direct actions upon GnRH neurons exclusively through ER-beta. In the light of differing ligand-binding characteristics of ER-beta from those of ER-alpha, this discovery offers a potential new approach to influence estrogen feedback to GnRH neurons through ER-beta-selective receptor ligands.

Glutamatergic innervation of the hypothalamic median eminence and posterior pituitary of the rat.

Recent studies have localized the glutamatergic cell marker type-2 vesicular glutamate transporter (VGLUT2) to distinct peptidergic neurosecretory systems that regulate hypophysial functions in rats. The present studies were aimed to map the neuronal sources of VGLUT2 in the median eminence and the posterior pituitary, the main terminal fields of hypothalamic neurosecretory neurons. Neurons innervating these regions were identified by the uptake of the retrograde tract-tracer Fluoro-Gold (FG) from the systemic circulation, whereas glutamatergic perikarya of the hypothalamus were visualized via the radioisotopic in situ hybridization detection of VGLUT2 mRNA. The results of dual-labeling studies established that the majority of neurons accumulating FG and also expressing VGLUT2 mRNA were located within the paraventricular, periventricular and supraoptic nuclei and around the organum vasculosum of the lamina terminalis and the preoptic area. In contrast, only few FG-accumulating cells exhibited VGLUT2 mRNA signal in the arcuate nucleus. Dual-label immunofluorescent studies of the median eminence and posterior pituitary to determine the subcellular location of VGLUT2, revealed the association of VGLUT2 immunoreactivity with SV2 protein, a marker for small clear vesicles in neurosecretory endings. Electron microscopic studies using pre-embedding colloidal gold labeling confirmed the localization of VGLUT2 in small clear synaptic vesicles. These data suggest that neurosecretory neurons located mainly within the paraventricular, anterior periventricular and supraoptic nuclei and around the organum vasculosum of the lamina terminalis and the preoptic area secrete glutamate into the fenestrated vessels of the median eminence and posterior pituitary. The functional aspects of the putative neuropeptide/glutamate co-release from neuroendocrine terminals remain to be elucidated.

Localization and osmotic regulation of vesicular glutamate transporter-2 in magnocellular neurons of the rat hypothalamus.

In this report we present immunocytochemical and in situ hybridization evidence that magnocellular vasopressin and oxytocin neurons in the hypothalamic supraoptic and paraventricular nuclei express type-2 vesicular glutamate transporter, a marker for their glutamatergic neuronal phenotype. To address the issue of whether an increase in magnocellular neuron activity coincides with the altered synthesis of the endogenous glutamate marker, we have introduced a new dual-label in situ hybridization method which combines fluorescent and autoradiographic signal detection components for vasopressin and vesicular glutamate transporter-2 mRNAs, respectively. Application of this technique provided evidence that 2% sodium chloride in the drinking water for 7 days produced a robust and significant increase of vesicular glutamate transporter-2 mRNA in vasopressin neurons of the supraoptic nucleus. The immunocytochemical labeling of pituitary sections, followed by the densitometric analysis of vesicular glutamate transporter-2 immunoreactivity in the posterior pituitary, revealed a concomitant increase in vesicular glutamate transporter-2 protein levels at the major termination site of the magnocellular axons. These data demonstrate that magnocellular oxytocin as well as vasopressin cells contain the glutamatergic marker vesicular glutamate transporter-2, similarly to most of the parvicellular neurosecretory neurons examined so far. The robust increase in vesicular glutamate transporter-2 mRNA and immunoreactivity after salt loading suggests that the cellular levels of vesicular glutamate transporter-2 in vasopressin neurons are regulated by alterations in water-electrolyte balance. In addition to the known synaptic actions of excitatory amino acids in magnocellular nuclei, the new observations suggest novel mechanisms whereby glutamate of endogenous sources can regulate magnocellular neuronal functions.

Characterization of gonadotrophin-releasing hormone precursor cDNA in the Old World mole-rat Cryptomys hottentotus pretoriae: high degree of identity with the New World guinea pig sequence.

Regulation of pituitary gonadotrophins by the decapeptide gonadotrophin-releasing hormone 1 (GnRH1) is crucial for the development and maintenance of reproductive functions. A common amino acid sequence for this decapeptide, designated as 'mammalian' GnRH, has been identified in all mammals thus far investigated with the exception of the guinea pig, in which there are two amino acid substitutions. Among hystricognath rodents, the members of the family Bathyergidae regulate reproduction in response to diverse cues. Thus, highveld mole-rats (Cryptomys hottentotus pretoriae) are social bathyergids in which breeding is restricted to a particular season in the dominant female, but continuously suppressed in subordinate colony members. Elucidation of reproductive control in these animals will be facilitated by characterization of their GnRH1 gene. A partial sequence of GnRH1 precursor cDNA was isolated and characterized. Comparative analysis revealed the highest degree of identity (86%) to guinea pig GnRH1 precursor mRNA. Nevertheless, the deduced amino acid sequence of the mole-rat decapeptide is identical to the 'mammalian' sequence rather than that of guinea pigs. Successful detection of GnRH1-synthesizing neurones using either a guinea pig GnRH1 riboprobe or an antibody against the 'mammalian' decapeptide is consistent with the guinea pig-like sequence for the precursor and the classic 'mammalian' form for the decapeptide. The high degree of identity in the GnRH1 precursor sequence between this Old World mole-rat and the New World guinea pig is consistent with the theory that caviomorphs and phiomorphs originated from a common ancestral line in the Palaeocene to mid Eocene, some 63-45 million years ago.

Expression of estrogen receptor-beta messenger ribonucleic acid in oxytocin and vasopressin neurons of the rat supraoptic and paraventricular nuclei.

The regulatory actions of estrogen on magnocellular oxytocin (OT) and vasopressin (VP) neurons of the paraventricular (PVN) and supraoptic (SON) nuclei are well documented. To date it is still debated whether the effect of estrogens is exerted directly or mediated by estrogen-sensitive interneurons. Previous immunocytochemical (ICC) and in situ hybridization (ISH) studies detected either low levels or absence of the classical estrogen receptor (ER-alpha) in the PVN and the SON of the rat. The present experiments using a combined ICC and ISH method were undertaken to examine the expression of the recently cloned beta form of ER (ER-beta) in OT- and VP-immunoreactive (IR) neuronal systems of the rat hypothalamus. The results demonstrate that the highest cellular levels of ER-beta messenger RNA (mRNA) in OT-IR neurons can be visualized in the caudal portion of the PVN and in an area ventro-medial to the central core of VP-IR cells. These neurons were previously shown to project caudally to the brain stem and the spinal cord to regulate autonomic functions. In addition, the whole rostro-caudal extent of the PVN and the SON contained OT-IR neurons that coexpressed variable levels of ER-beta mRNA. Similarly, the presence of ER-beta mRNA was seen in a large population of VP-IR paraventricular and supraoptic neurons. In the SON, somewhat stronger hybridization signal was detected in VP-IR neurons as compared with OT-IR neurons. Together, these findings provide strong support for the concept that the functions of OT- and VP-IR neurons in the PVN and the SON are regulated directly by estrogen and that the genomic effects of estrogens are mediated by ER-beta.

Estrogen receptor immunoreactivity is present in the majority of central histaminergic neurons: evidence for a new neuroendocrine pathway associated with luteinizing hormone-releasing hormone-synthesizing neurons in rats and humans.

The central regulation of the preovulatory LH surge requires a complex sequence of interactions between neuronal systems that impinge on LH-releasing hormone (LHRH)-synthesizing neurons. The reported absence of estrogen receptors (ERs) in LHRH neurons indicates that estrogen-receptive neurons that are afferent to LHRH neurons are involved in mediating the effects of this steroid. We now present evidence indicating that central histaminergic neurons, exclusively located in the tuberomammillary complex of the caudal diencephalon, serve as an important relay in this system. Evaluation of this system revealed that 76% of histamine-synthesising neurons display ERalpha-immunoreactivity in their nucleus; furthermore histaminergic axons exhibit axo-dendritic and axo-somatic appositions onto LHRH neurons in both the rodent and the human brain. Our in vivo studies show that the intracerebroventricular administration of the histamine-1 (H1) receptor antagonist, mepyramine, but not the H2 receptor antagonist, ranitidine, can block the LH surge in ovariectomized estrogen-treated rats. These data are consistent with the hypothesis that the positive feedback effect of estrogen in the induction of the LH surge involves estrogen-receptive histamine-containing neurons in the tuberomammillary nucleus that relay the steroid signal to LHRH neurons via H1 receptors.

Immunocytochemical characterization of afferents to estrogen receptor-containing neurons in the medial preoptic area of the rat.

Double-label immunocytochemistry has been employed to elucidate the chemical nature of the afferent neuronal projections to the estrogen receptor-containing neurons located in the medial preoptic area of the rat brain. To ensure a clear separation of the immunolabelled afferent profiles from the estrogen receptors, the former were visualized first and the diaminobenzidine reaction product was silver-gold intensified. Using a monoclonal antibody raised against purified human estrogen receptors, we observed an intense nuclear immunoreactivity in Vibratome, semithin and ultrathin sections. Neuropeptide-Y, serotonin-, phenylethanolamine N-methyltransferase- and adrenocorticotrophin-immunoreactive axons and varicosities were observed in close apposition to the estrogen receptor-positive cells. At the ultrastructural level, neuropeptide-Y-immunoreactive boutons were seen in synaptic contact with cells showing estrogen receptor immunoreactivity in their nucleus. These results indicate that neurons located in the medial preoptic area, one of the principal sites for the control of female reproductive function, may be influenced by both estrogen and neurotransmitters/neuropeptides via, respectively, nuclear receptors and synaptic contacts.

Ageing and the diurnal expression of the mRNAs for vasopressin and for the V1a and V1b vasopressin receptors in the suprachiasmatic nucleus of male rats.

Changes in the function of neuropeptide synthesizing cells within the suprachiasmatic nucleus (SCN), the site of the predominant circadian pacemaker, may underlie the disturbance of rhythms observed during ageing. Arginine vasopressin (AVP) is synthesized by nearly one-third of SCN neurones in the rat. This peptide has predominantly excitatory actions within the SCN mediated by V(1)-type receptors; the extent to which the V(1a) and/or V(1b) receptor subtypes are involved in SCN functions remains to be determined. The present study used isotopic in situ hybridization histochemistry to examine the effects of ageing on expression of mRNAs for AVP and V(1a) in the SCN and for V(1b) in the SCN and supraoptic nucleus (SON) of male rats kept under a 12 : 12 h light/dark cycle. Analysis of film autoradiographs from young adult (2-3-month-old; n = 40) or aged (19-20-month-old; n = 40) animals, at eight time points across the light/dark cycle, revealed an equivalent pattern and amplitude for the diurnal rhythm of AVP mRNA in the SCN of the young adult and aged groups. Both groups also displayed a significant diurnal rhythm in the expression of V(1a) receptor mRNA; however, the amplitude of this rhythm was reduced in the aged group, due to increased levels during the light phase and early part of night. Although the expression of V(1b) mRNA did not display a significant diurnal rhythm within the SCN or SON, persistently elevated levels for V(1b) mRNA were observed in the aged group at both sites.

Ageing and the diurnal expression of mRNAs for vasoactive intestinal peptide and for the VPAC2 and PAC1 receptors in the suprachiasmatic nucleus of male rats.

Ageing alters fundamental aspects of circadian rhythmicity in mammals; the effects include reduced rhythm amplitude and alterations in period length and in entrainment to the light/dark cycle. Such changes may reflect disruptions in cellular function within the suprachiasmatic nucleus (SCN), the site of the predominant circadian pacemaker. In the SCN, vasoactive intestinal peptide (VIP)-synthesizing neurones receive various inputs, including retinohypothalamic projections containing pituitary adenylate cyclase activating peptide (PACAP). SCN VIP cells establish connections with local neurones and send efferents beyond the nucleus. Considerable evidence implicates VIP and PACAP in circadian rhythm maintenance and/or entrainment to photic Zeitgebers. These actions involve members of a distinct family of receptors; mRNAs for two such receptors, VPAC2 and PAC1, are present in the SCN. This study used isotopic in situ hybridization to examine the effects of ageing on expression of mRNAs for VIP, VPAC2 and PAC1 in the SCN of male rats under a 12 : 12 h light/dark cycle. Analysis of film autoradiographs from young adult (2-3 months) or aged (19-20 months) rats, at eight time points across the light/dark cycle, showed loss of diurnal rhythmicity and reduced levels for VIP mRNA in the aged group. A diurnal rhythm of VPAC2 receptor mRNA was present in both groups, but its levels were reduced in the aged rats. There were no differences between the two groups for PAC1 receptor mRNA expression. The present results indicate that ageing reduces VIP and VPAC2 receptor mRNA and eliminates diurnal expression of VIP mRNA within the SCN of aged male rats.

Cellular expression of V1a vasopressin receptor mRNA in the female rat preoptic area: effects of oestrogen.

The preovulatory luteinizing hormone (LH) surge in female rats is dependent upon signals from the suprachiasmatic nucleus (SCN), the site of a dominant circadian pacemaker. Various lines of evidence indicate that arginine-vasopressin (AVP)-containing projections from the SCN to the preoptic area (POA) contribute to the production of the surge of LH-releasing hormone (LHRH). These actions may be mediated by V(1a) because the transcript for this AVP receptor subtype is predominant within the POA of the female rat. In this study, in situ hybridization histochemistry was used to examine V(1a) mRNA expression, either by itself or together with LHRH or glutamic acid decarboxylase 65 (GAD(65)) mRNA, within the POA of ovariectomized rats in the presence or absence of oestrogen. V(1a) mRNA was found in cells across the rostro-caudal axis of the POA; some were in close proximity to cells expressing LHRH mRNA. Coexpression of V(1a) and LHRH mRNAs was detected only very rarely. By contrast, cells with V(1a) mRNA commonly displayed GAD(65) mRNA. The density of V(1a) mRNA-expressing cells was particularly high within the anteroventral periventricular nucleus; at this site, V(1a) mRNA expression was elevated following oestrogen treatment. The present results indicate that V(1a)-mediated AVP actions may influence LHRH release via cells in the immediate vicinity of LHRH neurones and/or via oestrogen-regulated cells in the anteroventral periventricular nucleus, which is a site that lacks LHRH neurones but plays an essential role in initiating the preovulatory LH surge.

Evidence for extensive distribution of oestrogen receptor alpha-immunoreactivity in the cerebral cortex of adult rats.

The presence of oestrogen receptor alpha (ERalpha)-immunoreactivity (IR) in the cerebral cortex of adult rats has been investigated. Previous studies have reported a negligible presence of ERalpha or ERalpha mRNA in this region of the adult rat brain. We have used immunoprecipitation and immunohistochemistry, with various antibodies and fixatives, to detect this protein in the cingulate cortex. When the tissue was fixed using paraformaldehyde alone only faint ERalpha-IR was observed at this site. In contrast, following fixation either with acrolein (with or without paraformaldehyde) or with a mixture of paraformaldehyde and glutaraldehyde there was extensive ERalpha-IR throughout layers II to VI; this was absent when the antibodies were preincubated with the peptide fragment used in their production. The presence of ERalpha-IR in the nonfixed cingulate cortex of adult rats was confirmed by immunoprecipitation.

Oestrogen receptor beta-immunoreactivity in gonadotropin releasing hormone-expressing neurones: regulation by oestrogen.

Double-label immunohistochemistry was employed to establish whether immunoreactivity for the beta subtype of the oestrogen receptor (ER beta-IR) is present in gonadotropin releasing hormone (GnRH)-containing cells. In the immortalized GnRH cell line, GT1-7, almost all nuclei were immunoreactive for ER beta. In the preoptic area of ovariectomized rats, more than one-half of the GnRH neurones (52.0-63.5%) contained ER beta-IR within the nucleus; a smaller proportion of these neurones (5-10%) displayed a particularly intense nuclear signal for ER beta. The presence of ER beta-IR in the nuclei of GT1-7 cells and GnRH neurones is consistent with recent reports of ER beta mRNA in these cells. Oestrogen treatment reduced the percentage of GnRH neurones with detectable ER beta-IR. The range of signal intensity for ER beta and the incidence of the ER beta signal in GnRH neurones were comparable following double-label immunohistochemistry involving either bright field or fluorescent techniques. These findings raise the possibility that ER beta receptors mediate direct effects of oestrogen on GnRH neurones.

The basal forebrain corticopetal system revisited.

The medial septum, diagonal bands, ventral pallidum, substantia innominata, globus pallidus, and internal capsule contain a heterogeneous population of neurons, including cholinergic and noncholinergic (mostly GABA containing), corticopetal projection neurons, and interneurons. This highly complex brain region, which constitutes a significant part of the basal forebrain has been implicated in attention, motivation, learning, as well as in a number of neuropsychiatric disorders, such as Alzheimer's disease, Parkinson's disease, and schizophrenia. Part of the difficulty in understanding the functions of the basal forebrain, as well as the aberrant information-processing characteristics of these disease states lies in the fact that the organizational principles of this brain area remained largely elusive. On the basis of new anatomical data, it is proposed that a large part of the basal forebrain corticopetal system be organized into longitudinal bands. Considering the topographic organization of cortical afferents to different divisions of the prefrontal cortex and a similar topographic projection of these prefrontal areas to basal forebrain regions, it is suggested that several functionally segregated cortico-prefronto-basal forebrain-cortical circuits exist. It is envisaged that such specific "triangular" circuits could amplify selective attentional processing in posterior sensory cortical areas.

The hypothalamo-infundibular growth hormone-releasing hormone (GH-RH) system of the rat.

Two to 10 days after complete unilateral surgical isolation of the medial basal hypothalamus (MBH) or 3 months following neonatal monosodium glutamate (MSG) treatment, the presence of growth hormone-releasing hormone (GH-RH) immunoreactive neuronal structures was studied in rats using vibratome sections and GH-RH immunocytochemistry. Neonatal MSG treatment resulted in a dramatic decrease of GH-RH immunoreactivity in the median eminence (ME), but not complete disappearance as reported earlier. Unilateral complete deafferentation of the MBH caused only a slight decrease in GH-RH immunostaining in the posterior regions of the ipsilateral median eminence (ME). At this level GH-RH accumulation was observed in scattered transected fibers lateral to the cut, outside of the MBH. Our findings indicate that the arcuate nucleus is the major source of GH-RH immunoreactive structures in the ME. Although, however, in very small numbers, the existence of other sources of GH-RH terminals cannot be excluded.

Synaptic connections between substance P-containing axons and estrogen receptor-synthesizing neurons in the medial preoptic area of the rat brain.

Dual-label immunocytochemical procedures were employed to provide ultrastructural evidence for the presence of substance P (SP) in afferents to estrogen-receptive neurons in the medial preoptic area (MPO) of the female rat. SP-immunoreactive axon terminals were observed to innervate the periventricular (PvPO) and medial (MPN) preoptic nuclei of the MPO densely, and to form synaptic connections at these sites with neurons which contain estrogen receptors in their nucleus. These results indicate that estrogen-receptive preoptic neurons may be regulated by SP-containing neuronal pathways via synaptic mechanisms.

Oestrogen receptor-beta and neurohypophysial hormones: functional interaction and neuroanatomical localisation.

Oestrogens affect fluid balance, influencing both ingestive behaviour and renal excretion. The renal effects are partly due to altered release of vasopressin and oxytocin. This study was designed to explore the role of oestrogen receptor-beta (ERbeta) in neurohypophysial hormonal function. Following dietary administration, soya isoflavones reach the brain in sufficient concentration to activate ERbeta, but not oestrogen receptor-alpha (ERalpha). ERbeta function was therefore manipulated by feeding rat diets differing in soya isoflavone content. Fluid balance and neurohypophysial hormone release were measured in male rats maintained for 14 days on a soya isoflavone-free diet or one containing 150 microg/g genistein+daidzein. Food and water intake, body weight, urine flow, osmolality and sodium concentrations were determined daily. After 14 days, plasma and urine osmolality and sodium, vasopressin and oxytocin concentrations were determined. There was no significant difference in weight gain between the two groups or in their excretion of sodium and water or plasma sodium and plasma oxytocin. However, plasma vasopressin was significantly lower in the iso-free group. Double-label immunocytochemistry was used to assess colocalisation of ERbeta with the neurohypophysial hormones in male rats. Cell nuclei showing ERbeta immunoreactivity were abundant in the posterior magnocellular paraventricular nucleus (PVNpm) and in the supraoptic nucleus (SON). Vasopressin-immunoreactive neurones were similarly distributed, forming the core of the PVNpm and the ventral portion of the SON; majority were positive for ERbeta. Cells with oxytocin immunoreactivity were located mainly at the periphery of the PVNpm and in the dorsal SON; only approximately a quarter of these cells showed ERbeta immunoreactivity. Thus, the difference in the effects of the soya diet on vasopressin and oxytocin release may be related to the ERbeta-activating properties of this diet and to the preponderance of this receptor in vasopressin as opposed to oxytocin cells.

Ultrastructural pathology of degenerating "dark" granule cells in the hippocampal dentate gyrus of adrenalectomized rats.

Adrenalectomy-evoked delayed degeneration and death of granule cells in the hippocampal dentate gyrus (DG) of the rat brain were studied by means of electron microscopy and a recently elaborated silver method that selectively stains the "dark", collapsed neurons in a Golgi-like manner. At the light microscopic level, the silver technique revealed degenerating granule cells located exclusively in the dentate gyrus; other glucocorticoid receptor-containing regions of the brain were not affected. The silver-stained cell bodies were shrunken, most of the dendrites had a beaded appearance, and the stained axons could be traced along their route to the CA3 pyramidal neurons of the hippocampus. The analysis of 2.5 microns thick Epon-embedded sections stained with toluidine blue revealed hyperchromatic, dark granule neurons and their remains and a heavy glial activity in the vicinity of collapsing neuronal profiles. At the ultrastructural level, early and late stages of neuronal degeneration were observed. The early phase was characterized by markedly increased electron density, a massive shrinkage of the whole somato-dendritic domain, vacuolization of mitochondria, swelling of the nucleolus and condensation of the nuclear chromatin. In the late stage, subcellular organelles were hardly recognizable due to the extremely high electron density and dramatic shrinkage of the cytoplasm. These profiles exhibited disintegration of the cellular organelles and loss of their afferents. Concomitantly, disintegration of granule cell dendrites (clasmatodendrosis) and lifting of "dark" mossy fibers from cell bodies and dendrites of CA3 pyramidal neurons were observed. In the latter cells, this partial denervation caused no apparent signs of ultrastructural alterations. Proliferation of astrocytes and microglial cells was also obvious as they engulfed and eliminated the degenerating neuronal elements. Degenerating neurons frequently occurred adjacent neurons with normal morphology. These morphological features indicate that the delayed degeneration of hippocampal granule cells following adrenalectomy might proceed through a cytoskeletal collapse terminating in cell death.

Co-localisation of kisspeptin with galanin or neurokinin B in afferents to mouse GnRH neurones.

The gonadotrophin-releasing hormone (GnRH) secreting neurones, which form the final common pathway for the central regulation of reproduction, are directly targeted by kisspeptin (KP) via the G protein-coupled receptor, GPR54. In these multiple labelling studies, we used ovariectomised mice treated with 17ß-oestradiol (OVX + E(2)) or vehicle (OVX + oil) to determine: (i) the ultrastructural characteristics of KP-immunoreactive (IR) afferents to GnRH neurones; (ii) their galanin or neurokinin B (NKB) content; and (iii) the co-expression of galanin or NKB with KP in the two major subpopulations of KP neurones located in the rostral periventricular area of the third ventricle (RP3V) and the arcuate nucleus (Arc). Electron microscopic investigation of the neuronal juxtapositions revealed axosomatic and axodendritic synapses; these showed symmetrical or asymmetrical characteristics, suggesting a phenotypic diversity of KP afferents. Heterogeneity of afferents was also demonstrated by differential co-expression of neuropeptides; in OVX + E(2) mice, KP afferents to GnRH neurones showed galanin-immunoreactivity with an incidence of 22.50 ± 2.41% and NKB-immunoreactivity with an incidence of 5.61 ± 2.57%. In OVX + oil animals, galanin-immunoreactivity in the KP afferents showed a major reduction, appearing in only 5.78 ± 1.57%. Analysis for co-localisation of galanin or NKB with KP was extended to the perikaryal level in animal models, which showed the highest KP incidence; these were OVX + E(2) females for the RP3V and OVX + oil females for the ARC. In the RP3V of colchicine-treated OVX + E(2) animals, 87.84 ± 2.65% of KP-IR neurones were galanin positive. In the Arc of the colchicine-treated OVX + oil animals, galanin immunoreactivity was detected in only 12.50 ± 1.92% of the KP expressing neurones. By contrast, the incidence of co-localisation with NKB in the Arc of those animals was 98.09 ± 1.30%. In situ hybridisation histochemistry of sections from OVX + E(2) animals identified galanin message in more than a third of the KP neurones in the RP3V (38.67 ± 11.57%) and in the Arc (42.50 ± 12.52%). These data suggest that GnRH neurones are innervated by chemically heterogeneous KP cell populations, with a small proportion deriving from the Arc group. The presence of galanin within KP axons innervating GnRH neurones and the oestrogen-dependent regulation of that presence add a new dimension to the roles played by galanin in the central regulation of reproduction.