The caudal neurosecretory system: control and function of a novel neuroendocrine system in fish.

The caudal neurosecretory system (CNSS) of fish was first defined over 70 years ago yet despite much investigation, a clear physiological role has yet to be elucidated. Although the CNSS structure is as yet thought to be confined to piscine species, the secreted peptides, urotensins I and II (UI and UII), have been detected in a number of vertebrate species, most recently illustrated by the isolation of UII in humans. The apparent importance of these peptides, suggested by their relative phylogenetic conservation, is further supported by the complex control mechanisms associated with their secretion. The CNSS in teleosts is known to receive extensive and diverse innervation from the higher central nervous system, with evidence for the presence of cholinergic, noradrenergic, serotonergic, and peptidergic descending inputs. Recent observations also suggest the presence of glucocorticoid receptors in the flounder CNSS, supporting previous evidence for a possible role as a pituitary-independent mechanism controlling cortisol secretion. The most convincing evidence as to a physiological role for the CNSS in fish has stemmed from the direct and indirect influence of the urotensins on osmoregulatory function. Recent advances allowing the measurement of circulating levels of UII in the flounder have supported this. In addition, there is evidence to suggest some seasonal variation in peptide levels supporting the notion that the CNSS may have an integrative role in the control of coordinated changes in the reproductive, osmoregulatory and nutritional systems of migratory euryhaline species.     

Immunohistochemical demonstration of urotensins I and II in the caudal neurosecretory system of the Japanese charr,Salvelinus leucomaenis,retained in sea water

This paper is concerned with part of the role and function of the caudal neurosecretory system of the charr,Salvelinus leucomaenis, studied by immunohistochemistry. In order to elucidate the different histologic changes, we examined the immunoreactivities of urotenisn I (UI) and urotensin II (UII) in 3 experimental groups: the feral (river) fish, the fresh-water aquarium-, and sea water aquarium-retained fish. Coexistence of UI and UII was demonstrated in most of the smaller and larger neurons distributed in and near the urophyseal system of all 3 groups. However, some of the larger neurons were immunoreactive only to a single hormone, UI or UII. Merely a few neurons indicated no reactivity for either UI or UII. No such clearcut differences were encountered immunohistochemically in the 3 groups. Neuronal and urophysial immuno-reactivity to UI of feral and fresh-water-retained fish was slightly stronger than that of sea water-retained fish. Moreover, in sea water-retained fish, the intensity of immunoreactivity for UI was variable, and the number of neurons positive for UII only was somewhat larger than that in feral and fresh-water-retained fish. A series of UII-positive cerebrospinal fluid (CSF)-contacting neurons were seen in the ependymal and subependymal layers ventral to the central canal of the spinal cord in every group. These CSF-contacting neurons might constitute another neurosecretory system aside from the ordinary caudal neurosecretory system equipped with urophysis. In contrast to the hypothalamohypophysial neurosecretory system, the caudal neurosecretory system did not show any significant changes among the 3 groups. This suggests that urotensins I and II have no essential role in osmoregulation of the charr.     

Immunohistochemical investigation of urotensins in the caudal spinal cord of four species of elasmobranchs and the lamprey,Lampetra japonica

Summary In the four species of elasmobranchs examined (Triakis scyllia, Heterodontus japonicus, Scyliorhinus torazame, Dasyatis akajei), all identifiable caudal neurosecretory cells and their corresponding neurohemal areas showed urotensin II (UII)-immunoreactivity with varied intensity. To localize urotensin I (UI) in the caudal neurosecretory system of the dogfish, Triakis scyllia, h-CRF (1–20) antiserum that cross-reacts with UI was used in place of UI antiserum. CRF/UI-immunoreactivity was demonstrated in the neurosecretory cells and neurohemal areas. A considerable number of neurons showed both UII- and CRF/UI-immunoreactivities, suggesting that UII and UI are produced in the same neurosecretory cells. However, some neurons exhibited UII-immunoreactivity, but no CRF/UI-immunoreactivity. Cells immunoreactive only to CRF antiserum were not detected. At least two populations of neurons exist in the dogfish caudal neurosecretory system: (i) cells immunoreactive for both CRF/UI and UII, and (ii) cells immunoreactive for UII. The dorsal cells of the lamprey, Lampetra japonica, did not react with either UII or CRF antiserum.     

A double sequential immunofluorescence method demonstrating the co-localization of urotensins I and II in the caudal neurosecretory system of the teleost,Gillichthys mirabilis

Summary A double immunofluorescence method was devised to localize simultaneously urotensin-I (UI) and -II (UII) immunoreactivities in the caudal neurosecretory system of the goby, Gillichthys mirabilis. In a sequential fashion, sections of the posterior spinal cord and urophysis were treated with antiserum to corticotropin-releasing factor (CRF) that cross-reacts with UI, fluorescein-conjugated sheep anti-rabbit IgG, biotinylated anti-UII and rhodamine-conjugated avidin. UI and UII immunoreactivities appeared to coexist in some neurons and in most fibers and urophysial tissue; the remainder of the fibers and urophysis and the majority of neurons were immunoreactive for CRF/ UI only. No convincing evidence of immunoreactivity for UII only was found. A few nonreactive cells were seen, but these may not be neurosecretory neurons. The two immunoreactive cell types were not segregated topographically, and the intensity of perikaryal immunofluorescence for CRF/UI was variable. To explain these results a hypothesis that all caudal neurosecretory cells may synthesize both UI and UII and that immunoreactive differences may reflect different states of cellular activity, is suggested. This sequential double immunofluorescence method offers several advantages over other techniques and is especially useful for co-localization studies when primary antisera from different species are not available.     

Development of the caudal neurosecretory system of the chum salmon,Oncorhynchus keta,as revealed by immunohistochemistry for urotensins I and II

In order to make an immunohistochemical analysis of the development of the caudal neurosecretory system of the chum salmon, Oncorhynchus keta, we employed the peroxidase-anti-peroxidase technique using antisera specific for urotensins (U) I and II on artificially reared embryos, larvae, and juveniles of this species. Immunoreactivities for UI and UII were first demonstrated in the embryo immediately before hatching, showing labeled perikarya and fibers in the most caudal region of the spinal cord where the presumptive caudal neurosecretory system is located. However, distinct differentiation of the histological neurohemal organ had not yet begun in the embryo. Immunoreactive perikarya and fibers gradually increased in number, and an elaborate urophysis comparable to that of adults was demonstrated in the larvae about 5 months after hatching. At this stage, weak immunoreactivity against UI was detected in the neurohypophysis.     

Liberation of urotensin II from the teleost urophysis: an historical overview

During the past 20 years, urotensin II (UII) has progressed from being a peptide synthesized only in the urophysis of the caudal neurosecretory system of teleost fish to being considered an important physiological regulator in mammals with implications for the pathogenesis of a range of human cardiovascular and renal diseases. The "liberation" of UII from the urophysis was a gradual process and involved the sequential realization that (a) UII is present not only in the urophysis but also in the central nervous systems (CNS) of teleosts, (b) UII peptides, similar in structure to the urophysial peptides, are present in the diffuse caudal neurosecretory systems and/or CNS of species less evolutionarily advanced than teleosts, including Agnatha, thereby showing that UII is a phylogenetically ancient peptide, (c) UII is present in the brain and spinal cord of a tetrapod, the green frog Rana ridibunda, and (d) the UII gene and its specific receptor (GPR14/UT) are expressed in the CNS and certain peripheral tissues of mammals, including the human. The discovery that the genomes of mammals contain an additional gene encoding a UII-related peptide (URP) and the availability of highly effective peptide and non-peptide antagonists to investigate the role of UII in human physiology and pathophysiology ensure that the peptide will remain "center stage" for several years to come.     

Caudal neurosecretory system of the zebrafish: ultrastructural organization and immunocytochemical detection of urotensins

The caudal neurosecretory system is described here for the first time in the zebrafish, one of the most important models used to study biological processes. Light- and electron-microscopical approaches have been employed to describe the structural organization of Dahlgren cells and the urophysis, together with the immunohistochemical localization of urotensin I and II (UI and UII) peptides. Two latero-ventral bands of neuronal perikarya in the caudal spinal cord project axons to the urophysis. The largest secretory neurons (~20 μm) are located rostrally. UII-immunoreactive perikarya are much more numerous than those immunoreactive for UI. A few neurons are immunopositive for both peptides. Axons contain 75-nm to 180-nm dense-core vesicles comprising two populations distributed in two axonal types (A and B). Large dense vesicles predominate in type A axons and smaller ones in type B. Immunogold double-labelling has revealed that some fibres contain both UI and UII, sometimes even within the same neurosecretory granule. UII is apparently the major peptide present and predominates in type A axons, with UI predominating in type B. A surprising finding, not previously reported in other fish, is the presence of dense-core vesicles, similar to those in neurons, in astrocytes including their end-feet around capillaries. Secretory type vesicles are also evident in ependymocytes and cerebrospinal-fluid-contacting neurons in the terminal spinal cord. Thus, in addition to the urophysis, this region may possess further secretory systems whose products and associated targets remain to be established. These results provide the basis for further experimental, genetic and developmental studies of the urophysial system in the zebrafish.     

Development of the caudal neurosecretory system of the nile tilapia Oreochromis niloticus: an immunohistochemical and electron microscopic study

The development of the caudal neurosecretory system (CNSS) of the Nile tilapia, Oreochromis niloticus, has been investigated by means of UI/oCRF (urotensin I/ovine corticotropin-releasing factor) immunohistochemistry and transmission electron microscopy. UI-like immunoreactive perikarya and fibers are first detected in the caudal spinal cord of larval fish about 4 days after hatching (stage 21). In the region of the future urophysis two bundles of strongly immunoreactive neurosecretory fibers are observed. At this stage, neurosecretory axons terminate on the meninx sheath of the spinal cord with immature neurosecretory terminals. The histogenesis of the urophysis begins at stage 24. The future neurohemal organ consists of a small ventral swelling of the spinal cord, which is associated with dilated vessels. At this stage, neurosecretory axons terminate on the basal lamina of the ingrowing blood vessels. Further development occurs by means of progressive branching of vessels and the concomitant increase in the number of neurosecretory terminals. In the caudal spinal cord, immunoreactive neurons also increase in number and progressively differentiate morphologically. Typical features of the mature CNSS are recognizable in 4-month-old juveniles. Data suggest that in tilapia both the synthesis and the release of urophysial hormones begin before morphogenesis of the neurohemal organ takes place.     

Terminal processes of serotonin neurons in the caudal spinal cord of the molly,Poecilia latipinna,project to the leptomeninges and urophysis

Summary The caudal neurosecretory complex of poeciliids has previously been shown to be innervated by extranuclear and intrinsic serotonergic projections. In the present study, immunohistochemical techniques were used to characterize fibers originating from serotonin neurons intrinsic to the caudal spinal cord. Bipolar and multipolar neurons were oriented ventromedially, and contained numerous large granular vesicles. Three types of serotonergic fibers were distinguished based on their distribution and morphology. Intrinsic Type-A fibers branched into varicose segments near the ventrolateral surface of the spinal cord and contacted the basal lamina beneath the leptomeninges. Type-B fibers coursed longitudinally to enter the urophysis, where they diverged and terminated around fenestrated capillaries. Labelled vesicles in Type-A and Type-B terminals were the same size as those in labelled cells and in unlabelled neurosecretory terminals in the urophysis. Type-C small varicose fibers branched within the neuropil of the caudal neurosecretory complex. Serotonin may be secreted into the submeningeal cerebrospinal fluid, the urophysis, and the caudal vein by Type-A and Type-B fibers, whereas, Type-C fibers may be processes of serotonergic interneurons in the neuroendocrine nucleus. The possibility that urotensins I and II or arginine vasotocin were colocalized in the processes of the intrinsic serotonin neurons was investigated immunohistochemically. The negative results of these experiments suggest that serotonin-containing neurons may represent a neurochemically distinct subpopulation in the caudal neurosecretory complex.     

Comparative distribution of the mRNAs encoding urotensin I and urotensin II in zebrafish

The neural neurosecretory system of fishes produces two biologically active neuropeptides, i.e. the corticotropin-releasing hormone paralog urotensin I (UI) and the somatostatin-related peptide urotensin II (UII). In zebrafish, we have recently characterized two UII variants termed UIIalpha and UIIbeta. In the present study, we have investigated the distribution of UI, UIIalpha and UIIbeta mRNAs in different organs by quantitative RT-PCR analysis and the cellular localization of the three mRNAs in the spinal cord by in situ hybridization (ISH) histochemistry. The data show that the UI gene is mainly expressed in the caudal portion of the spinal cord and, to a lesser extent, in the brain, while the UIIalpha and the UIIbeta genes are exclusively expressed throughout the spinal cord. Single-ISH labeling revealed that UI, UIIalpha and UIIbeta mRNAs occur in large cells, called Dahlgren cells, located in the ventral part of the caudal spinal cord. Double-ISH staining showed that UI, UIIalpha and UIIbeta mRNAs occur mainly in distinct cells, even though a few cells were found to co-express the UI and UII genes. The differential expression of UI, UIIalpha and UIIbeta genes may contribute to the adaptation of Dahlgren cell activity during development and/or in various physiological conditions.     

Interplay of KNDy and nNOS neurons: A new possible mechanism of GnRH secretion in the adult brain

Reproduction in mammals is favoured when there is sufficient energy available to permit the survival of offspring. Neuronal nitric oxide synthase expressing neurons produce nitric oxide in the proximity of the gonadotropin-releasing hormone neurons in the preoptic region. nNOS neurons are an integral part of the neuronal network controlling ovarian cyclicity and ovulation. Nitric oxide can directly regulate the activity of the GnRH neurons and play a vital role neuroendocrine axis. Kisspeptin neurons are essential for the GnRH pulse and surge generation. The anteroventral periventricular nucleus (AVPV), kisspeptin neurons are essential for GnRH surge generation. KNDy neurons are present in the hypothalamus's arcuate nucleus (ARC), co-express NKB and dynorphin, essential for GnRH pulse generation. Kisspeptin-neurokinin B-dynorphin (KNDy) neuroendocrine molecules of the hypothalamus are key components in the central control of GnRH secretion. The hypothalamic neurons kisspeptin, KNDy, nitric oxide synthase (NOS), and other mediators such as leptin, adiponectin, and ghrelin, play an active role in attaining puberty. Kisspeptin signalling is mediated by NOS, which further results in the secretion of GnRH. Neuronal nitric oxide is critical for attaining puberty, but its direct role in adult GnRH secretion is poorly understood. This review mainly focuses on the role of nNOS and its interplay with KNDy neurons in the hormonal regulation of reproduction.     

Immunohistochemical localization of urotensin I and other neuropeptides in the caudal neurosecretory system of three species of teleosts and two species of elasmobranchs

Summary In three species of teleosts — carp Cyprinus carpio; grass carp Ctenopharyngodon idella; and crucian carp Carassius auratus — the caudal neurosecretory system displays small, medium-sized and large neurons. Urotensin I (UI)-immunoreactive and UI-nonreactive neurons were found in all three groups; in general, the number of the latter neurons exceeded that of the former. Noteworthy are: (i) UI-immunoreactive fibers in the caudal spinal cord and (ii) dense accumulations of UI-immunoreactive product around the capillaries of the urophysis. In two species of elasmobranchs — cat shark Heterodontus japonicas and swell shark Cephaloscyllium umbratile — neurosecretory neurons decreased in size in rostro-caudal direction. Most of the neurosecretory perikarya, their axons and the corresponding neurohemal areas were UI-immunoreactive, but a small number of secretory neurons was devoid of immunoreaction. Oxytocin, arginine vasopressin, substance P, somatostatin, neurotensin, vasoactive intestinal polypeptide and gastrin-releasing peptide were not detected in the caudal neurosecretory system of the carp.     

Central and peripheral cardiovascular, ventilatory, and motor effects of trout urotensin-II in the trout

Urotensin-II (U-II) was originally considered to be exclusively the product of the caudal neurosecretory system (CNSS) of teleost fish, but it has now been demonstrated that U-II is widely expressed in peripheral tissues and nervous structures of species from lampreys to mammals. However, very little is known regarding the physiological effects of this peptide in its species of origin. In the present review, we summarize the most significant results relating to the cardiovascular, ventilatory, and motor effects of centrally and peripherally administered synthetic trout U-II in our experimental animal model, the unanesthetized trout Oncorhynchus mykiss. In addition, we compare the actions of U-II with those of other neurohormonal peptides, particularly with the actions of urotensin-I, a 41-amino acid residue peptide paralogous to corticotropin-releasing hormone that is co-localized with U-II within neurons of the CNSS.     

Identification and distribution of nitric oxide synthase in the brain of adult Antarctic teleosts

We investigated the presence of nitric oxide synthase (NOS) in brain of adult Antarctic teleosts by indirect immunofluorescence technique using a synthetic rat neuronal NOS (nNOS) antibody. The following species were examined: Trematomus bernacchii, Gymnodraco acuticeps, Histiodraco velifer, Cygnodraco mawsoni (haemoglobin-rich), Chionodraco hamatus and Pagetopsis macropterus (haemoglobin-free). Immunoreactive cell bodies were localized in dorsal telencephalon, in hypothalamus, in optic tectum of the mesencephalon as well as in Purkinje cells of the cerebellum. No differences were observed in the localization of the nNOS immunopositivity in the Antarctic teleosts brains examined and NOS distribution was similar to that described in other teleosts, suggesting that nitric oxide (NO) may also function as a neurotransmitter in the brain of Antarctic teleosts. A strong immunopositivity was observed in the cerebral blood vessels of the icefishes suggesting that NO may play a pivotal role in the regulation of the cerebral blood flow especially in these haemoglobin-free species.     

Immunoreactive methionine-enkephalin in the caudal neurosecretory system of the carp,Cyprinus carpio

Summary Methionine-enkephalin (Met-enk) was detected by immunocytochemistry and radioimmunoassay in the caudal neuro-secretory system of the carp Cyprinus carpio. Some cells showing urotensin I (UI)-immunoreactivity reacted to Met-enk antiserum, but others did not. Neurons with urotensin II (UII)-immunoreactivity did not react to Met-enk antiserum; neurons with both UI and UII immunoreactivities also displayed a negative Met-enk reaction. Met-enk was detected by radioimmunoassay in the urophysis, although the content was relatively small compared with that found in other parts of the central nervous system and in the pituitary.     

Effect of subchronic acrylamide exposure on the expression of neuronal and inducible nitric oxide synthase in rat brain

Acrylamide (ACR) is a known industrial neurotoxic chemical. Evidence suggests that ACR neurotoxic effect is related to brain neurotransmission disturbances. Since nitric oxide (NO) acts as a neurotransmission modulator and is produced by nitric oxide synthase (NOS), the neuronal NOS (nNOS) and inducible NOS (iNOS) expression pattern were determined in rat cerebral cortex and striatum after subchronic exposure to ACR. Using immunocytochemistry, the neuronal count of nNOS or optical density of iNOS from sections at three coronal levels, bregma 1.0, -0.4, and -2.3 mm, were compared between ACR-treated and control rats. At all three levels, nNOS expressions were uniformly decreased in most of the neocortical subregions following the treatment of ACR. At bregma level 1.0 mm, total numbers of nNOS expressing neurons were significantly decreased to 58.7% and 64.7% of the control in the cortex and striatum of ACR-treated rats, respectively. However, at the bregma level -2.3 mm, ACR treatment did not produce a significant difference in the numbers of nNOS expressing neurons both in the cortex and striatum. Contrary to nNOS, iNOS expressions were consistently increased to approximately 32% in the neocortex and 25% in the striatum, following the subchronic ACR treatment. These data suggest that subchronic ACR exposure involves compensatory mechanism on nNOS and iNOS expression to maintain the homeostasis of NO at the rostral part of the neocortex and the striatum. However, in the caudal brain, increased iNOS expression did not suppress nNOS expression. Therefore, the present study is consistent with the hypothesis that ACR toxicity is mediated through the disturbance to the NO signaling pathway and exhibits a rostrocaudal difference through the differential expressions of nNOS and iNOS in the neocortex and the striatum.     

Cortisol and prolactin modulation of caudal neurosecretory system activity in the euryhaline flounder Platichthys flesus

Previous studies have shown roles for cortisol and prolactin in osmoregulatory adaptation to seawater and freshwater, respectively, in euryhaline fish. This study of the European flounder investigated the potential for these hormones to modulate activity of the caudal neurosecretory system (CNSS), which is thought to be involved in physiological adaptation to changing external salinity. Superfusion of isolated CNSS with either cortisol or prolactin (10 microM; 15 min) led to changes in firing activity in neuroendocrine Dahlgren cells, recorded extracellularly. Cortisol evoked a modest increase in overall firing activity, with the response delayed by 4 h after treatment. The response to prolactin was short latency, continued to build up over the subsequent 4-h wash period, and comprised increased firing activity together with recruitment of previously silent Dahlgren cells. Immunoreactivity for glucocorticoid and prolactin receptors was localised to Dahlgren cells. The CNSS expression level for glucocorticoid-2 receptor mRNA, measured by Q-PCR, was significantly lower in fish fully acclimated to freshwater, compared to seawater. No differences were seen between these two states for prolactin receptor mRNA expression. These results provide evidence for a modulatory action of both hormones on the neurosecretory function of the CNSS.     

Neuronal nitric oxide synthase: expression in rat parietal cells

Nitric oxide synthases (NOS) are enzymes that catalyze the generation of nitric oxide (NO) from L-arginine and require nicotinamide adenine dinucleotide phosphate (NADPH) as a cofactor. At least three isoforms of NOS have been identified: neuronal NOS (nNOS or NOS I), inducible NOS (iNOS or NOS II), and endothelial NOS (eNOS or NOS II). Recent studies implicate NO in the regulation of gastric acid secretion. The aim of the present study was to localize the cellular distribution and characterize the isoform of NOS present in oxyntic mucosa. Oxyntic mucosal segments from rat stomach were stained by the NADPH-diaphorase reaction and with isoform-specific NOS antibodies. The expression of NOS in isolated, highly enriched (>98%) rat parietal cells was examined by immunohistochemistry, Western blot analysis, and RT-PCR. In oxyntic mucosa, histochemical staining revealed NADPH-diaphorase and nNOS immunoreactivity in cells in the midportion of the glands, which were identified as parietal cells in hematoxylin and eosin-stained step sections. In isolated parietal cells, decisive evidence for nNOS expression was obtained by specific immunohistochemistry, Western blotting, and RT-PCR. Cloning and sequence analysis of the PCR product confirmed it to be nNOS (100% identity). Expression of nNOS in parietal cells suggests that endogenous NO, acting as an intracellular signaling molecule, may participate in the regulation of gastric acid secretion.     

Vasoactive properties of urotensins I and II in a columbiform and three galliform birds, and a bioassay for urotensin I

Summary Crude extracts ofGillichthys urophyses and chromatographically purified urotensin I (UI) and urotensin II (UII) (fromCatostomus urophyses) were injected intravenously intoCoturnix coturnix japonica, Colinus virginianus, Alectoris graeca (Galliformes) andColumba livia (Columbiformes). Changes in arterial blood pressure were monitored. UI elicited dose-dependent vasodepressor responses in all birds. Thioglycollate treatment abolished the depressor action of arginine vasotocin but not that of UI (in all birds). UII was a pressor agent inCoturnix andColinus, both members of the Galliformes. However, the hormone had no pressor activity inColumba, a member of the Columbiformes, and in another member of the Galliformes,Alectoris. The pressor effect of UII was also dose-dependent. Injection of crude urophysial extract intoColinus andCoturnix, therefore, elicited biphasic responses. UII effects can be abolished by prior incubation with carboxypeptidase A. Intravenous injection of the -adrenoceptor blocker, phenoxybenzamine, prior to urotensin injections had no effect on the responses to urotensins. As the chukar,Alectoris graeca, is sensitive to UI and resistant to anesthesia and surgery, and does not readily develop tachyphylaxis to repeated UI injections, its use is recommended as a routine bioassay animal for UI.