Morphological evidence for direct interaction between gonadotrophin-releasing hormone neurones and astroglial cells in the human hypothalamus

In rodents, there is compelling evidence indicating that dynamic cell-to-cell communications involving cross talk between astroglial cells (such as astrocytes and specialised ependymoglial cells known as tanycytes) and neurones are important in regulating the secretion of gonadotrophin-releasing hormone (GnRH), the neurohormone that controls both sexual maturation and adult reproductive function. However, whether such astroglial cell-GnRH neurone interactions occur in the human brain is not known. In the present study, we used immunofluorescence to examine the anatomical relationship between GnRH neurones and glial cells within the hypothalamus of five women. Double-staining experiments demonstrated the ensheathment of GnRH neurone perikarya by glial fibrillary acidic protein (GFAP)-immunoreactive astrocyte processes in the periventricular zone of the tuberal region of the hypothalamus. GFAP immunoreactivity did not overlap that of GnRH at the GnRH neurone's projection site (i.e. the median eminence of the hypothalamus). Rather, human GnRH neuroendocrine fibres were found to be closely associated with vimentin or nestin-immunopositive radial glial processes likely belonging to tanycytes. In line with these light microscopy data, ultrastructural examination of GnRH-immunoreactive neurones showed numerous glial cells in direct apposition to pre-embedding-labelled GnRH cell bodies and/or dendrites in the infundibular nucleus, whereas postembedding immunogold-labelled GnRH nerve terminals were often seen to be enwrapped by glial cell processes in the median eminence. GnRH nerve button were sometimes visualised in close proximity to fenestrated pituitary portal blood capillaries and/or evaginations of the basal lamina that delineate the pericapillary space. In summary, these data demonstrate that GnRH neurones morphologically interact with astrocytes and tanycytes in the human brain and provide evidence that glial cells may contribute physiologically to the process by which the neuroendocrine brain controls the function of GnRH neurones in humans.     

Glial-gonadotrophin hormone (GnRH) neurone interactions in the median eminence and the control of GnRH secretion

A wealth of information now exists showing that glial cells are actively involved in the cell–cell communication process generating and disseminating information within the central nervous system. In the hypothalamus, two types of glial cells, astrocytes and ependymal cells lining the latero-ventral portion of the third ventricle (known as tanycytes), regulate the secretory activity of neuroendocrine neurones. This function, initially described for astrocytes apposing magnocellular neurones, has been more recently characterised for neurones secreting gonadotrophin hormone-releasing hormone (GnRH). The available evidence suggests that glial cells of the median eminence regulate GnRH secretion via two related mechanisms. One involves the production of growth factors acting via receptors with tyrosine kinase activity. The other involves plastic rearrangements of glia–GnRH neurone adhesiveness. GnRH axons reach the median eminence, at least in part, directed by basic fibroblast growth factor. Their secretory activity is facilitated by insulin-like growth factor 1 and members of the epidermal growth factor family. A structural complement to these soluble molecules is provided by at least three cell–cell adhesion systems endowed with signalling capabilities. One of them uses the neuronal cell adhesion molecule (NCAM), another employs the synaptic cell adhesion molecule (SynCAM), and the third one consists of neuronal contactin interacting with glial receptor-like protein tyrosine phosphatase-β. It is envisioned that, within the median eminence, soluble factors and adhesion molecules work coordinately to control delivery of GnRH to the portal vasculature.     

Distribution of galanin receptor-2 immunoreactive neurones in the ovine hypothalamus: no evidence for involvement in the control of gonadotrophin-releasing hormone secretion

Galanin is a small neuropeptide that mediates its effects via three receptor isoforms: galanin receptor-1, galanin receptor-2 and galanin receptor-3 (Gal-R1, Gal-R2 and Gal-R3). Galanin is thought to be an important intermediate in signalling in the hypothalamic-pituitary-gonadal axis and has been widely detected in the ovine hypothalamus. The expression of galanin and Gal-R1 has been reported to fluctuate during the reproductive cycle. Although the distribution of Gal-R1 has been determined in the ovine hypothalamus, the distribution of Gal-R2 was hitherto unknown. Using immunohistological and immunofluorescence techniques, we have mapped the distribution of Gal-R2 in the ovine hypothalamus, collected during the follicular phase of the oestrous cycle and examined colocalisation of Gal-R2 with oestrogen receptor alpha (ERalpha) and gonadotrophin-releasing hormone (GnRH). Gal-R2 was expressed in several regions of the hypothalamus (supraoptic nucleus, paraventricular nucleus, ventromedial nucleus, arcuate nucleus) but not as widely expressed as Gal-R1. Areas of Gal-R2 expression overlapped with those reported for Gal-R1. We observed that, in certain defined regions of the hypothalamus, up to 50% of neurones that express Gal-R2 also express ERalpha. No neurones coexpressed Gal-R2 and GnRH. Thus, we conclude that, in follicular phase animals, this receptor plays little or no role in direct intermediary signal transmission in GnRH-mediated control of the reproductive cycle.     

Temporal and spatial regulation of CRE recombinase expression in gonadotrophin-releasing hormone neurones in the mouse

Gonadotrophin-releasing hormone (GnRH) neurones located within the brain are the final neuroendocrine output regulating the reproductive hormone axis. Their small number and scattered distribution in the hypothalamus make them particularly difficult to study in vivo . The Cre/loxP system is a valuable tool to delete genes in specific cells and tissues. We report the production of two mouse lines that express the CRE bacteriophage recombinase in a GnRH-specific manner. The first line, the GnRH-CRE mouse, contains a transgene in which CRE is under the control of the murine GnRH promoter and targets CRE expression specifically to GnRH neurones in the hypothalamus. The second line, the GnRH-CRETeR mouse, uses the same murine GnRH promoter to target CRE expression to GnRH neurones, but is modified to be constitutively repressed by a tetracycline repressor (TetR) expressed from a downstream tetracycline repressor gene engineered within the transgene. GnRH neurone-specific CRE expression can therefore be induced by treatment with doxycycline which relieves repression by TetR. These GnRH-CRE and GnRH-CRETeR mice can be used to study the function of genes expressed specifically in GnRH neurones. The GnRH-CRETeR mouse can be used to study genes that may have distinct roles in reproductive physiology during the various developmental stages.     

Aromatase knockout mice show normal steroid-induced activation of gonadotrophin-releasing hormone neurones and luteinising hormone surges with a reduced population of kisspeptin neurones in the rostral hypothalamus

We recently reported that female aromatase knockout (ArKO) mice show deficits in sexual behaviour and a decreased population of kisspeptin‐immunoreactive neurones in the rostral periventricular area of the third ventricle (RP3V), resurrecting the question of whether oestradiol actively contributes to female‐typical sexual differentiation. To further address this question, we assessed the capacity of ArKO mice to generate a steroid‐induced luteinising hormone (LH) surge. Adult, gonadectomised wild‐type (WT) and ArKO mice were given silastic oestradiol implants s.c. and, 1 week later, received s.c. injections of either oestradiol benzoate (EB) followed by progesterone, EB alone, or no additional steroids to activate gonadotrophin‐releasing hormone (GnRH) neurones and generate an LH surge. Treatment with EB and progesterone induced significant Fos/GnRH double‐labelling and, consequently, an LH surge in female WT and in ArKO mice of both sexes but not in male WT mice. ArKO mice of both sexes had fewer cells expressing Kiss‐1 mRNA in the RP3V compared to female WT mice but had more Kiss‐1 mRNA‐expressing cells compared to WT males, reflecting an incomplete sexual differentiation of this system. To determine the number of cells expressing kisspeptin, the same experimental design was repeated in Experiment 2 with the addition of groups of WT and ArKO mice that were given EB + progesterone and sacrificed 2 h before the expected LH surge. No differences were observed in the number of kisspeptin‐immunoreactive cells 2 h before and at the time of the LH surge. The finding that ArKO mice of both sexes have a competent LH surge system suggests that oestradiol has predominantly defeminising actions on the GnRH/LH surge system in males and that the steroid‐induced LH surge can occur in females even with a greatly reduced population of kisspeptin neurones in the RP3V.     

Early development of forebrain gonadotrophin-releasing hormone (GnRH) neurones and the role of GnRH as an autocrine migration factor

Normal migration of the gonadotrophin-releasing hormone (GnRH) neurones during early development, from the olfactory region to the hypothalamus, is crucial for reproductive development in all vertebrates. The establishment of the GnRH system includes tangential migration of GnRH perikarya as well as extension of GnRH fibres to various areas of the central nervous system (CNS). The exact spatio-temporal nature of this process, as well as the factors governing it, are not fully understood. We studied the development of the GnRH system and the effects of GnRH knockdown using a newly developed GnRH3:EGFP transgenic zebrafish line. We found that enhanced green fluorescent protein is specifically and robustly expressed in GnRH3 neurones and fibres. GnRH3 fibres in zebrafish began to extend as early as 26 h post-fertilisation and by 4–5 days post-fertilisation had developed into an extensive network reaching the optic tract, telencephalon, hypothalamus, midbrain tegmentum and hindbrain. GnRH3 fibres also innervated the retina and projected into the trunk via the spinal cord. GnRH3 perikarya were observed migrating along their own fibres from the olfactory region to the preoptic area (POA) via the terminal nerve ganglion and the ventral telencephalon. GnRH3 cells were also observed in the trigeminal ganglion. The establishment of the GnRH3 fibre network was disrupted by morpholino-modified antisense oligonucleotides directed against GnRH3 causing abnormal fibre development and pathfinding, as well as anomalous GnRH3 perikarya localisation. These findings support the hypothesis that GnRH3 neurones migrate from the olfactory region to the POA and caudal hypothalamus. Novel data regarding the early development of the GnRH3 fibre network in the CNS and beyond are described. Moreover we show, in vivo , that GnRH3 is an important factor regulating GnRH3 fibre pathfinding and neurone localisation in an autocrine fashion.     

Receptor protein tyrosine phosphatase from stem cells to mature glial cells of the central nervous system

The cornerstone of cell signaling is largely based on the phosphorylation state that is defined by the equili-brium of the activity of protein kinases and protein phosphatases. The role of protein tyrosine kinases in brain development, brain tumors, and neurodegenerative diseases was studied extensively, yet, the importance of protein tyrosine phosphatases (PTPs) in the development of glial cells was somewhat neglected. In this review, we have summarized recent findings of PTP expression during development of the central nervous system and the different cell types of the brain, from stem cells to mature glial cells, and highlighted the potential role of these enzymes in neuronal stem cell development, glioblastomas, and myelination.     

Effects of chronic stress on hypothalamic lnterleukin-1beta, interleukin-2, and gonadotrophin-releasing hormone gene expression in ovariectomized rats

The influence of chronic stress on the expression of interleukin (IL)-1beta and IL-2 mRNAs in ovariectomized rat brains, and the physiological consequences of the expression of these cytokines on hypothalamic-pituitary-gonadal (HPG) activity were investigated. Using polymerase chain reaction (PCR)-assisted semiquantitative analysis, we demonstrated alterated expression of IL-1beta and IL-2 mRNA during repeated cold stress; the expression of both IL-beta and IL-2 mRNA increased in the medial preoptic area and ventromedial hypothalamus, and decreased in the lateral hypothalamic area. In the arcuate nucleus/median eminence, IL-2 mRNA expression was dramatically decreased, in contrast to the increase in IL-1beta mRNA expression. Concomitant analysis of GnRH mRNA expression indicated significant suppression of GnRH synthesis in the chronic phase, and a strong negative correlation with cytokine expression in the medial preoptic area. Similar results were obtained in intact females exposed to this stress. These results, together with previous pharmacological studies, suggest that chronic stress may induce reproductive dysfunction through the effects of stress-induced expression of endogenous cytokines.     

Expression and induction of intercellular adhesion molecules (ICAMs) and major histocompatibility complex (MHC) antigens on cultured murine oligodendrocytes and astrocytes

Expression of intercellular adhesion molecule-1 (ICAM-1), ICAM-2-like molecule (Lgp55), and class I/II major histocompatibility complex (MHC) antigens (H-2 and Ia) was investigated in cultures of murine oligodendrocytes and astrocytes. Under unstimulated conditions, low levels of ICAM-1 expression were observed on astrocytes (less than 20%), but not on oligodendrocytes. Lgp55 was expressed intensely on oligodendrocytes (greater than 90%) and to a lesser degree on astrocytes (greater than 70%). A weak class I MHC (H-2) immunoreactivity was identified on both oligodendrocytes and astrocytes (50-70%). Class II MHC (Ia) antigen was undetectable on both cell types. After 48-hr exposure to immune mediators that include interferon-gamma (IFN-gamma), 500 U/ml, and supernatant from concanavalin A (Con A)-activated spleen cells, ICAM-1 expression was markedly increased on astrocytes (greater than 80%), but not on oligodendrocytes. Lgp55 expression on both cell types was not altered. Induction of H-2 antigen expression by immune mediators was quite high on both cell types (greater than 95%), while Ia antigen induction was low on astrocytes (less than 50%) and did not occur on oligodendrocytes. Cell type-specific expression and induction of ICAMs and MHC antigens by immune mediators may play roles in lymphocyte-glial cell interactions at the sites of inflammation in the central nervous system (CNS).     

Depolarising and hyperpolarising actions of GABA(A) receptor activation on gonadotrophin-releasing hormone neurones: towards an emerging consensus

The gonadotrophin-releasing hormone (GnRH) neurones represent the final output neurones of a complex neuronal network that controls fertility. It is now appreciated that GABAergic neurones within this network provide an important regulatory influence on GnRH neurones. However, the consequences of direct GABA(A) receptor activation on adult GnRH neurones have been controversial for nearly a decade now, with both hyperpolarising and depolarising effects being reported. This review provides: (i) an overview of GABA(A) receptor function and its investigation using electrophysiological approaches and (ii) re-examines the past and present results relating to GABAergic regulation of the GnRH neurone, with a focus on mouse brain slice data. Although it remains difficult to reconcile the results of the early studies, there is a growing consensus that GABA can act through the GABA(A) receptor to exert both depolarising and hyperpolarising effects on GnRH neurones. The most recent studies examining the effects of endogenous GABA release on GnRH neurones indicate that the predominant action is that of excitation. However, we are still far from a complete understanding of the effects of GABA(A) receptor activation upon GnRH neurones. We argue that this will require not only a better understanding of chloride ion homeostasis in individual GnRH neurones, and within subcellular compartments of the GnRH neurone, but also a more integrative view of how multiple neurotransmitters, neuromodulators and intrinsic conductances act together to regulate the activity of these important cells.     

The interaction between F3 immunoglobulin domains and protein tyrosine phosphatases zeta/beta triggers bidirectional signalling between neurons and glial cells

F3, a mouse glycosyl-phosphatidylinositol anchored molecule of the immunoglobulin superfamily, is known to influence axonal growth and fasciculation via multiple interactions of its modular immunoglobulin-like domains. We prepared an Fc chimeric molecule (F3IgFc) to identify molecules interacting with these domains and characterize the functional impact of the interactions. We affinity-isolated tenascin-C and isoforms of the proteoglycan-type protein tyrosine phosphatases zeta/beta (PTPzeta/RPTPbeta) from extracts of developing mouse brain. We showed that both PTPzeta/RPTPbeta and tenascin-C can bind directly to F3, possibly in an exclusive manner, with the highest affinity for the F3-PTPzeta/RPTPbeta interaction. We observed a strong binding of F3IgFc-coated fluorospheres to astrocytes in neural primary cultures and to C6 astrocytoma cells, and demonstrated, in antibody perturbation experiments, that F3-Ig binding on astrocytes depends on its interaction with PTPzeta/RPTPbeta. We also found by confocal analysis that tenascin-C and PTPzeta/RPTPbeta were colocalized on astrocytes which suggests a complex interplay of interactions between PTPzeta/RPTPbeta, tenascin-C and F3. We showed that the interaction between PTPzeta/RPTPbeta and F3-Ig-like domains can trigger bidirectional signalling. C6 glia-expressed PTPzeta/RPTPbeta stimulated neurite outgrowth by cortical and cerebellar neurons, whereas preclustered F3IgFc specifically modified the distribution of phosphotyrosine labelling in these glial cells. Both effects could be prevented and/or mimicked by anti-F3 and anti-6B4PG antibodies. These results identify F3 and PTPzeta/RPTPbeta as potential mediators of a reciprocal exchange of information between glia and neurons.     

Inverse regulation of inducible nitric oxide synthase (iNOS) and arginase I by the protein tyrosine phosphatase SHP-1 in CNS glia

We have previously shown that the SH2 domain-containing protein tyrosine phosphatase SHP-1 plays a critical role in controlling virus infection in CNS glia in vivo and in vitro. The present study addressed whether increased virus replication in SHP-1-deficient glia in vitro may be a result of altered expression of inducible nitric oxide synthase (iNOS/NOS2). First, we observed a profound reduction in iNOS protein expression and production of nitric oxide (NO) in response to the viral mimic double-stranded RNA (dsRNA), despite the induction of high levels of iNOS mRNA, in SHP-1-deficient motheaten mouse compared to wild type littermate mouse glia. Because both iNOS expression and NO production are suppressed by multiple pathways involving arginase I activity, it was important that we observed abnormally high constitutive expression of arginase I in cultured glia of SHP-1-deficient compared to wild type mice. Further, both constitutive and IL-4/IL-10-induced expression of arginase I correlated with elevated STAT6 nuclear binding activity, decreased NO production, and increased virus replication in motheaten compared to wild type astrocytes. These findings provide the first evidence of an inverse relationship between NO and arginase I activity regulated by SHP-1 in CNS glia that is relevant to modulation of innate anti-viral responses. Thus, we propose that SHP-1 is a critical regulator of innate immunity to virus infections in CNS cells.     

Facilitation or inhibition of the oestradiol-induced gonadotrophin surge in the immature female rat by progesterone: effects on pituitary responsiveness to gonadotrophin-releasing hormone (GnRH), GnRH self-priming and pituitary mRNAs for the progesterone receptor A and B isoforms

Progesterone can either facilitate or inhibit the oestradiol (E(2))-induced gonadotrophin surge. We have previously developed immature female rat models to characterise and investigate the mechanisms of progesterone inhibition or facilitation. The aim of the present study was to determine the role of pituitary responsiveness to gonadotrophin-releasing hormone (GnRH) and GnRH self-priming under conditions of progesterone-facilitation and progesterone-inhibition, and whether the underlying mechanisms reflect changes in mRNAs encoding the A and B isoforms of the progesterone receptor (PR) in the pituitary gland. Pituitary responsiveness to GnRH, determined by measuring the luteinising hormone (LH) response to one i.v. injection of GnRH, was decreased by 60-80% (P < 0.001) in the progesterone-inhibition model. GnRH self-priming, estimated as the increment in the LH response to the second of two injections of GnRH separated by 60 min, was also significantly reduced (P < 0.05) in this model. In the progesterone-facilitation model, the LH response to GnRH injection was increased 2.5-3-fold (P < 0.05), an effect suppressed by the progesterone receptor antagonist, mifepristone. Progesterone-facilitation of LH release and increased pituitary responsiveness to GnRH were blocked by sheep anti-GnRH serum injected i.v. immediately after insertion of progesterone implants. The PR-B mRNA isoform, measured by solution hybridisation/RNase protection assay, was the predominant form in the pituitary of the immature female rat. PR-B was increased by E(2) and decreased by progesterone in both models. Thus, in immature female rats, progesterone-inhibition of the E(2)-induced LH surge is due to significant reduction in pituitary responsiveness to GnRH as well as in the magnitude of GnRH self-priming. Progesterone-facilitation of the E(2)-induced LH surge is due to increased pituitary responsiveness to GnRH, which is mediated by PR, and depends on endogenous GnRH release. The differences between progesterone-facilitation and progesterone-inhibition are not due to differences in regulation of pituitary PR-B mRNA.     

A novel neurotransmitter-independent communication pathway between axons and glial cells

Recent studies have provided evidence that transmitters released by neurons can activate glial receptors and stimulate calcium signalling in glial cells. Glial calcium signalling, in turn, may affect neuronal performance such as long-term changes in synaptic efficacy. Olfactory ensheathing cells (OECs) are a special glial cell type in vertebrates and insects and promote axon growth in the developing and mature nervous system. Physiological properties of OECs, however, have not been studied so far in detail. We measured changes in the calcium concentration in OECs of the moth Manduca sexta, in situ and in vivo. Electrical stimulation of olfactory receptor neurons in pupae or odour stimulation of receptor neurons in adults resulted in calcium transients in OECs. Olfactory receptor axons release acetylcholine; however, application of acetylcholine or other transmitters such as glutamate, GABA or nitric oxide did not induce calcium transients in OECs. Upon nerve stimulation, extracellular potassium rose by several millimolar as measured with potassium-sensitive microelectrodes. When potassium in the perfusion saline was increased from 4 to 10 mM or higher, voltage-dependent calcium transients in OECs that resembled stimulation-induced calcium transients were evoked. Blocking neuronal potassium channels with TEA reduced both the stimulation-induced increases in extracellular potassium and the calcium transients in OECs, whereas calcium transients in receptor axons were augmented. Our results show for the first time that accumulation of potassium, released by electrically active axons, is sufficient to evoke voltage-dependent calcium influx into glial cells, whereas neurotransmitters appear not to be involved in this neuron-glia communication in Manduca.