Light and electron microscopic immunocytochemistry of beta-endorphin/beta-LPH-like immunoreactive neurons in the arcuate nucleus and surrounding areas of the rat hypothalamus

Cellular and circuit diversity determines the impact of endogenous opioids in the descending pain modulatory pathway

The descending pain modulatory pathway exerts important bidirectional control of nociceptive inputs to dampen and/or facilitate the perception of pain. The ventrolateral periaqueductal gray (vlPAG) integrates inputs from many regions associated with the processing of nociceptive, cognitive, and affective components of pain perception, and is a key brain area for opioid action. Opioid receptors are expressed on a subset of vlPAG neurons, as well as on both GABAergic and glutamatergic presynaptic terminals that impinge on vlPAG neurons. Microinjection of opioids into the vlPAG produces analgesia and microinjection of the opioid receptor antagonist naloxone blocks stimulation-mediated analgesia, highlighting the role of endogenous opioid release within this region in the modulation of nociception. Endogenous opioid effects within the vlPAG are complex and likely dependent on specific neuronal circuits activated by acute and chronic pain stimuli. This review is focused on the cellular heterogeneity within vlPAG circuits and highlights gaps in our understanding of endogenous opioid regulation of the descending pain modulatory circuits.

Endogenous opioid peptides in the descending pain modulatory circuit

The opioid epidemic has led to a serious examination of the use of opioids for the treatment of pain. Opioid drugs are effective due to the expression of opioid receptors throughout the body. These receptors respond to endogenous opioid peptides that are expressed as polypeptide hormones that are processed by proteolytic cleavage. Endogenous opioids are expressed throughout the peripheral and central nervous system and regulate many different neuronal circuits and functions. One of the key functions of endogenous opioid peptides is to modulate our responses to pain. This review will focus on the descending pain modulatory circuit which consists of the ventrolateral periaqueductal gray (PAG) projections to the rostral ventromedial medulla (RVM). RVM projections modulate incoming nociceptive afferents at the level of the spinal cord. Stimulation within either the PAG or RVM results in analgesia and this circuit has been studied in detail in terms of the actions of exogenous opioids, such as morphine and fentanyl. Further emphasis on understanding the complex regulation of endogenous opioids will help to make rational decisions with regard to the use of opioids for pain. We also include a discussion of the actions of endogenous opioids in the amygdala, an upstream brain structure that has reciprocal connections to the PAG that contribute to the brain's response to pain.

Plasma and Cerebrospinal Fluid Beta-Endorphin Levels Show a Strong Association in Children with Cerebral Malaria

BACKGROUND

Beta (β)-endorphins are endogenous neuropeptides found in the plasma and cerebrospinal fluid (CSF) of humans but there have been reports of the relationship between the plasma and CSF β-endorphin levels in different clinical conditions. However, the relationship between β-endorphin levels in the plasma and CSF of children with cerebral malaria (CM) has not been reported.

AIM

To determine the relationship between β-endorphin levels in the CSF and plasma of children with CM.

SETTINGS AND DESIGN

This cross-sectional study involved 40 children, aged between 6 months and 14 years, admitted with a diagnosis of CM at the Obafemi Awolowo University Teaching Hospitals Complex (OAUTHC), Ile-Ife, Nigeria.

MATERIALS AND METHODS

One milliliter (mL) of venous blood and 1mL of CSF obtained from each subject at admission were used to determine the β-endorphin levels using enzyme-linked immunosorbent assay (ELISA) method.

STATISTICAL ANALYSIS

Bivariate linear regression was used to determine the association between plasma and CSF β-endorphin levels using the correlation coefficient (r), coefficient of determination (R 2), and P values.

RESULTS

The plasma β-endorphin levels significantly positively correlated with CSF β-endorphin (r = 0.568, P = 0.001) such that for every unit rise in plasma β-endorphin, CSF β-endorphin rose by 0.252 pmol/L (confidence interval: 0.132-0.371 pmol/L).

CONCLUSION

The finding of positive correlation between plasma and CSF β-endorphin levels in this study suggests a possible direct link between plasma and CSF in CM, probably from the disruption of the blood-brain barrier that has been reported in CM.

Mechanisms of cardiovascular actions of urocortins in the hypothalamic arcuate nucleus of the rat

The presence of urocortins (UCNs) and corticotropin-releasing factor (CRF) receptors has been reported in the hypothalamic arcuate nucleus (ARCN). We have previously reported that UCNs are involved in central cardiovascular regulation. Based on this information, we hypothesized that the ARCN may be one of the sites where UCNs exert their central cardiovascular actions. Experiments were done in artificially ventilated, adult male Wistar rats anesthetized with urethane. Unilateral microinjections (30 nl) of UCN1 (0.12-2 mM) elicited decreases in mean arterial pressure (MAP) and heart rate (HR). Maximum cardiovascular responses were elicited by a 1 mM concentration of UCN1. Microinjections of UCN2 and UCN3 (1 mM each) into the ARCN elicited similar decreases in MAP and HR. UCN1 was used as a prototype for the other experiments described below. HR responses elicited by UCN1 were significantly attenuated by bilateral vagotomy. Prior microinjections of NBI-27914 (CRF-1 receptor antagonist) and astressin (CRF-1 receptor and CRF-2 receptor antagonist) (1 mM each) into the ARCN significantly attenuated the cardiovascular responses elicited by UCN1 microinjections at the same site. Microinjections of UCN1 into the ARCN decreased efferent renal sympathetic nerve activity. It was concluded that microinjections of UCN1, UCN2, and UCN3 into the ARCN elicited decreases in MAP and HR. Decreases in MAP, HR, and renal sympathetic nerve activity elicited by UCN1 microinjections into the ARCN were mediated via CRF receptors. Bradycardic responses to UCN1 were mediated via the activation of vagus nerves, and decreases in MAP may be mediated via decreases in sympathetic nerve activity.

Effect of barodenervation on cardiovascular responses elicited from the hypothalamic arcuate nucleus of the rat

We have previously reported that chemical stimulation of the hypothalamic arcuate nucleus (ARCN) in the rat elicited increases as well as decreases in blood pressure (BP) and sympathetic nerve activity (SNA). The type of response elicited from the ARCN (i.e., increase or decrease in BP and SNA) depended on the level of baroreceptor activity which, in turn, was determined by baseline BP in rats with intact baroreceptors. Based on this information, it was hypothesized that baroreceptor unloading may play a role in the type of response elicited from the ARCN. Therefore, the effect of barodenervation on the ARCN-induced cardiovascular and sympathetic responses and the neurotransmitters in the hypothalamic paraventricular nucleus (PVN) mediating the excitatory responses elicited from the ARCN were investigated in urethane-anesthetized adult male Wistar rats. Bilateral barodenervation converted decreases in mean arterial pressure (MAP) and greater splanchnic nerve activity (GSNA) elicited by chemical stimulation of the ARCN with microinjections of N-methyl-D-aspartic acid to increases in MAP and GSNA and exaggerated the increases in heart rate (HR). Combined microinjections of NBQX and D-AP7 (ionotropic glutamate receptor antagonists) into the PVN in barodenervated rats converted increases in MAP and GSNA elicited by the ARCN stimulation to decreases in MAP and GSNA and attenuated increases in HR. Microinjections of SHU9119 (a melanocortin 3/4 receptor antagonist) into the PVN in barodenervated rats attenuated increases in MAP, GSNA and HR elicited by the ARCN stimulation. ARCN neurons projecting to the PVN were immunoreactive for proopiomelanocortin, alpha-melanocyte stimulating hormone (alpha-MSH) and adrenocorticotropic hormone (ACTH). It was concluded that increases in MAP and GSNA and exaggeration of tachycardia elicited by the ARCN stimulation in barodenervated rats may be mediated via release of alpha-MSH and/or ACTH and glutamate from the ARCN neurons projecting to the PVN.

Role of the hypothalamic arcuate nucleus in cardiovascular regulation

Recently the hypothalamic arcuate nucleus (Arc) has been implicated in cardiovascular regulation. Both pressor and depressor responses can be elicited by the chemical stimulation of the Arc. The direction of cardiovascular responses (increase or decrease) elicited from the Arc depends on the baseline blood pressure. The pressor responses are mediated via increase in sympathetic nerve activity and involve activation of the spinal ionotropic glutamate receptors. Arc-stimulation elicits tachycardic responses which are mediated via inhibition of vagal input and excitation of sympathetic input to the heart. The pathways within the brain mediating the pressor and tachycardic responses elicited from the Arc have not been delineated. The depressor responses to the Arc-stimulation are mediated via the hypothalamic paraventricular nucleus (PVN). Gamma aminobutyric acid type A receptors, neuropeptide Y1 receptors, and opiate receptors in the PVN mediate the depressor responses elicited from the Arc. Some circulating hormones (e.g., leptin and insulin) may reach the Arc via the leaky blood-brain barrier and elicit their cardiovascular effects. Although the Arc is involved in mediating the cardiovascular responses to intravenously injected angiotensin II and angiotensin-(1-12), these effects may not be due to leakage of these peptides across the blood-brain barrier in the Arc; instead, circulating angiotensins may act on neurons in the SFO and mediate cardiovascular actions via the projections of SFO neurons to the Arc. Cardiovascular responses elicited by acupuncture have been reported to be mediated by direct and indirect projections of the Arc to the RVLM.

Cardiovascular responses to chemical stimulation of the hypothalamic arcuate nucleus in the rat: role of the hypothalamic paraventricular nucleus

The mechanism of cardiovascular responses to chemical stimulation of the hypothalamic arcuate nucleus (ARCN) was studied in urethane-anesthetized adult male Wistar rats. At the baseline mean arterial pressure (BLMAP) close to normal, ARCN stimulation elicited decreases in MAP and sympathetic nerve activity (SNA). The decreases in MAP elicited by ARCN stimulation were attenuated by either gamma-aminobutyric acid (GABA), neuropeptide Y (NPY), or beta-endorphin receptor blockade in the ipsilateral hypothalamic paraventricular nucleus (PVN). Combined blockade of GABA-A, NPY1 and opioid receptors in the ipsilateral PVN converted the decreases in MAP and SNA to increases in these variables. Conversion of inhibitory effects on the MAP and SNA to excitatory effects following ARCN stimulation was also observed when the BLMAP was decreased to below normal levels by an infusion of sodium nitroprusside. The pressor and tachycardic responses to ARCN stimulation at below normal BLMAP were attenuated by blockade of melanocortin 3/4 (MC3/4) receptors in the ipsilateral PVN. Unilateral blockade of GABA-A receptors in the ARCN increased the BLMAP and heart rate (HR) revealing tonic inhibition of the excitatory neurons in the ARCN. ARCN stimulation elicited tachycardia regardless of the level of BLMAP. ARCN neurons projecting to the PVN were immunoreactive for glutamic acid decarboxylase 67 (GAD67), NPY, and beta-endorphin. These results indicated that: 1) at normal BLMAP, decreases in MAP and SNA induced by ARCN stimulation were mediated via GABA-A, NPY1 and opioid receptors in the PVN, 2) lowering of BLMAP converted decreases in MAP following ARCN stimulation to increases in MAP, and 3) at below normal BLMAP, increases in MAP and HR induced by ARCN stimulation were mediated via MC3/4 receptors in the PVN. These results provide a base for future studies to explore the role of ARCN in cardiovascular diseases.

Volume transmission of beta-endorphin via the cerebrospinal fluid; a review

There is increasing evidence that non-synaptic communication by volume transmission in the flowing CSF plays an important role in neural mechanisms, especially for extending the duration of behavioral effects. In the present review, we explore the mechanisms involved in the behavioral and physiological effects of β-endorphin (β-END), especially those involving the cerebrospinal fluid (CSF), as a message transport system to reach distant brain areas. The major source of β-END are the pro-opio-melano-cortin (POMC) neurons, located in the arcuate hypothalamic nucleus (ARH), bordering the 3^rd ventricle. In addition, numerous varicose β-END-immunoreactive fibers are situated close to the ventricular surfaces. In the present paper we surveyed the evidence that volume transmission via the CSF can be considered as an option for messages to reach remote brain areas. Some of the points discussed in the present review are: release mechanisms of β-END, independence of peripheral versus central levels, central β-END migration over considerable distances, behavioral effects of β-END depend on location of ventricular administration, and abundance of mu and delta opioid receptors in the periventricular regions of the brain.

Moderate hypercapnia-induced anesthetic effects and endogenous opioids

The purpose of this report is to explore the mechanisms of hypercapnia-induced antinociception. We carried out three experiments, the first to confirm whether moderate hypercapnia induces anesthetic effects, the second to determine whether naloxone reverses the anesthetic effects, and the third to evaluate whether beta-endorphin is related to the anesthetic effects. In a pre-test, we determined the optimal CO(2) concentration in a chamber which would cause moderate hypercapnia in rats. Eighteen rats were divided into control, hypercapnia, and hypercapnia plus naloxone groups in experiment 1. The naloxone group rats were injected with naloxone (10 mg/kg) intraperitoneally before gas inhalation. After 60 min gas inhalation, 10% formalin was injected into the left rear paw of all rats, and nociceptive behaviors were observed for 1 h. In experiment 2, 11 rats were divided into control and hypercapnia groups. The brain was removed and fixed under pentobarbital anesthesia. Sections were immunostained for c-Fos and beta-endorphin (ACTH) with the ABC method. All neurons double-labeled for c-Fos and beta-endorphin (ACTH) in the arcuate nucleus were counted by blinded investigators. Moderate hypercapnia (PaCO(2) 83+/-7 mmHg) reduced nociceptive behavior in the formalin test and naloxone pre-treatment attenuated this phenomenon. However, beta-endorphin-producing neurons were not activated by CO(2) inhalation. Endogenous opioids are related to moderate, hypercapnia-induced anesthetic effects, but, beta-endorphin-producing neurons in the hypothalamus were not activated by the CO(2) inhalation stress.

Functional and ultrastructural neuroanatomy of interactive intratectal/tectonigral mesencephalic opioid inhibitory links and nigrotectal GABAergic pathways: Involvement of GABAA and μ1-opioid receptors in the modulation of panic-like reactions elicited by electrical stimulation of the dorsal midbrain

In the present study, the functional neuroanatomy of nigrotectal-tectonigral pathways as well as the effects of central administration of opioid antagonists on aversive stimuli-induced responses elicited by electrical stimulation of the midbrain tectum were determined. Central microinjections of naloxonazine, a selective mu(1)-opiod receptor antagonist, in the mesencephalic tectum (MT) caused a significant increase in the escape thresholds elicited by local electrical stimulation. Furthermore, either naltrexone or naloxonazine microinjected in the substantia nigra, pars reticulata (SNpr), caused a significant increase in the defensive thresholds elicited by electrical stimulation of the continuum comprised by dorsolateral aspects of the periaqueductal gray matter (dlPAG) and deep layers of the superior colliculus (dlSC), as compared with controls. These findings suggest an opioid modulation of GABAergic inhibitory inputs controlling the defensive behavior elicited by MT stimulation, in cranial aspects. In fact, iontophoretic microinjections of the neurotracer biodextran into the SNpr, a mesencephalic structure rich in GABA-containing neurons, show outputs to neural substrate of the dlSC/dlPAG involved with the generation and organization of fear- and panic-like reactions. Neurochemical lesion of the nigrotectal pathways increased the sensitivity of the MT to electrical (at alertness, freezing and escape thresholds) and chemical (blockade of GABA(A) receptors) stimulation, suggesting a tonic modulatory effect of the nigrotectal GABAergic outputs on the neural networks of the MT involved with the organization of the defensive behavior and panic-like reactions. Labeled neurons of the midbrain tectum send inputs with varicosities to ipsi and contralateral dlSC/dlPAG and ipsilateral substantia nigra, pars reticulata and compacta, in which the anterograde and retrograde tracing from a single injection indicates that the substantia nigra has reciprocal connections with the dlSC/dlPAG featuring close axo-somatic and axo-dendritic appositions in both locations. In addition, ultrastructural approaches show inhibitory axo-axonic synapses in MT and inhibitory axo-somatic/axo-axonic synapses in the SNpr. These findings, in addition to the psychopharmacological evidence for the interaction between opioid and GABAergic mechanisms in the cranial aspects of the MT as well as in the mesencephalic tegmentum, offer a neuroanatomical basis of a pre-synaptic opioid inhibition of GABAergic nigrotectal neurons modulating fear in defensive behavior-related structures of the cranial mesencephalon, in a short link, and through a major neural circuit, also in GABA-containing perikarya and axons of nigrotectal neurons.

Detection of beta-endorphin in the cerebrospinal fluid after intrastriatal microinjection into the rat brain

We have investigated to what extent microinjected beta-endorphin could migrate from the rat brain parenchyma into the CSF compartment. Exogenous rat beta-endorphin (0.1 nmol) was microinjected into the left striatum 1 mm from the lateral ventricle in anesthetized male rats. CSF samples were collected at different time points up to 2 h post-injection from a catheter affixed to the atlanto-occipital membrane of the cisterna magna. Radioimmunoassay and mass spectrometry were performed on the CSF samples, and brain sections were immunostained for beta-endorphin and mu-opioid receptors. The beta-endorphin injected rats showed a marked increase in beta-endorphin immunoreactive (IR) material in the CSF, with a peak at 30-45 min post-injection, and this beta-endorphin-IR material existed mainly as the intact beta-endorphin peptide. The immunohistochemistry results revealed the appearance of distinct beta-endorphin-IR cell bodies in the globus pallidus and the bed nucleus of stria terminalis supracapsular part, regions distant from the injection site, at 2 h post-injection of exogenous beta-endorphin. The beta-endorphin-IR in several of the globus pallidus cell bodies colocalized with the mu-opioid receptor-IR at the cell surface. These findings show that upon delivery of synthetic beta-endorphin, there is a significant intracerebral spread of the injected peptide, reaching regions far from the site of injection via diffusion in the extracellular space and flow in the cerebrospinal fluid. This may be of relevance when interpreting studies based on intracerebral injections of peptides, and advances our knowledge regarding the migration of compounds within the brain.

Neuroanatomical and psychopharmacological evidence for interaction between opioid and GABAergic neural pathways in the modulation of fear and defense elicited by electrical and chemical stimulation of the deep layers of the superior colliculus and dorsal periaqueductal gray matter

The effects of central administration of opioid antagonists on the aversive responses elicited by electrical (at the freezing and escape thresholds) or chemical stimulation (crossings, rearings, turnings and jumps, induced by microinjections of bicuculline) of the midbrain tectum were determined. Central microinjections of naloxone and naltrexone in the mesencephalic tectum caused a significant increase in the freezing and escape thresholds elicited by electrical midbrain tectum stimulation. Furthermore, both opioid antagonists caused a significant decrease in the mean incidence of aversive behavioral responses induced by microinjections of bicuculline in the deep layers of the superior colliculus (DLSC) and in dorsal aspects of the periaqueductal gray matter (DPAG), as compared with controls. These findings suggest an opioid modulation of the GABAergic inhibitory inputs controlling the aversive behavior elicited by midbrain tectum stimulation. In fact, immunohistochemical evidence suggests that the dorsal mesencephalon is rich in beta-endorphin-containing neurons and fibers with varicosities. Iontophoretical microinjections of the neurotracer biodextran in the substantia nigra, pars reticulata (SNpr), show nigro-tectal pathways connecting SNpr with the same neural substrate of the DPAG rich in neuronal cells immunoreactive for opioid peptides. Labeled neurons of the DLSC and periaqueductal gray matter send inputs with varsicosities to ipsi- and contralateral DPAG and ipsilateral SNpr. These findings, in addition to the psychopharmacological evidence for the interaction between opioid and GABAergic mechanisms, offer a neuroanatomical basis of a possible presynaptic opioid inhibition of GABAergic nigro-tectal neurons modulating the fear in aversive structures of the cranial mesencephalon, in a short link, and maybe through a major neural circuit, also in GABA-containing perikarya of nigro-tectal neurons.

Proopiomelanocortin (POMC) mRNA expression: distribution and region-specific down-regulation by chronic morphine in female guinea pig hypothalamus

There is compelling evidence that endogenous opioid peptides are regulated by exogenous opiates. Our previous studies have shown that the mu-opioid receptor protein and mRNA are down-regulated in the mediobasal hypothalamus of the female guinea pig following chronic morphine treatment. In addition, electrophysiological studies have shown that hypothalamic beta-endorphin (beta-EP) neurons express mu-opioid receptors that are uncoupled and down-regulated following chronic morphine treatment. Currently, we tested the hypothesis that chronic morphine, which produces down-regulation of mu-opioid receptors, causes a down-regulation of pro-opiomelanocortin (POMC, the precursor of beta-EP) mRNA expression in female guinea pig hypothalamus. Female guinea pigs were ovariectomized and implanted subcutaneously (s.c.) with 4 x 75 mg pellets for 2 days plus six more pellets of either morphine (n = 6) or placebo (n = 6) for another 5 days. Animals were sacrificed between 1000 and 1100 h on day 7. The expression of POMC mRNA were investigated using in situ hybridization histochemistry with a guinea pig specific 35S-labeled cRNA probe in hypothalamic tissue sections. POMC mRNA was localized to the arcuate nucleus (Arc) and median eminence (ME) of the medial basal hypothalamus. The distribution pattern was the same in both morphine and placebo control animals. However, the density of silver grains was less in morphine treated animals versus placebo control animals. Overall, the level of POMC mRNA was decreased by 22% in the Arc of morphine-treated guinea pigs as compared with the placebo controls (p < 0.05). This decrease in POMC mRNA expression was even greater in the caudal Arc (28%, p < 0.01) in morphine-treated animals. These results suggested that the biosynthetic activity of POMC neurons is down-regulated with chronic exposure to morphine.

Beta-endorphin in the brain. A role in nociception

We have known the endogenous opioid peptide beta-endorphin for 20 years. Surprisingly, our knowledge of the physiological role of this peptide and its receptors in modulation of pain perception is still fragmentary. Whereas most studies have tried to elucidate the physiological role of beta-endorphin by reversing evoked responses by the opioid antagonist naloxone, this review focuses on quantification of release of beta-endorphin in the brain as the approach to define physiological and pathophysiological roles of beta-endorphin in relation to nociception. Using a lateral ventricle-cisterna magna perfusion model in the anesthetized rat, it was shown that depolarization of neurons in the arcuate nucleus of the hypothalamus, where beta-endorphin in produced, was followed by release of beta-endorphin to the cerebrospinal fluid compartment. Intense activation of spinal nociceptive pathways by intrathecal capsaicin injections also led to beta-endorphin release. It is concluded that there may still be good reason to quantify beta-endorphin in human cerebrospinal fluid to elucidate the role of beta-endorphin in pain perception.

Distribution of catecholamine-synthesizing enzymes and some neuropeptides in the median eminence-arcuate nucleus complex (MEARC) of the immature female pig

The presence of the catecholamine-synthesizing enzymes tyrosine hydroxylase (TH) and dopamine-beta-hydroxylase (D beta H) and some neuropeptides, including neuropeptide Y (NPY), Leu5-enkephalin (LENK), vasoactive intestinal polypeptide (VIP), calcitonin gene-related peptide (CGRP), substance P (SP), galanin (GAL) and somatostatin (SOM) was investigated in nerve fibres and perikarya of the median eminence-arcuate nucleus complex (MEARC) of the sexually immature female pigs by means of the immunohistochemical avidin-biotin complex method. Although immunoreactivities to all the studied substances were found in nerve fibres of the porcine MEARC, there were differences in the distribution and density of particular subsets of nerve fibres within the complex. While loose D beta H-immunoreactive (D beta H-IR) and dense TH-, NPY- and VIP-IR nerve meshworks occurred predominantly in the internal layer of the MEARC, nerve fibres immunoreactive to TH, CGRP, SOM, SP and LENK were more numerous in the external than in the internal layer of the median eminence (ME). Numerous TH-, D beta H-, NPY-, VIP-, SP- and CGRP-IR perivascular nerve fibres were also observed within both layers of the median eminence. There were also differences in the distribution of a particular subset of neurons within the porcine MEARC: NPY-, VIP-, GAL-, SP- and TH-IR (but not D beta H-IR) perikarya were found in the arcuate nucleus, while in the median eminence only subpopulations of NPY-, VIP and GAL-IR neurons were observed.

Release of beta-endorphin immunoreactivity into ventriculo-cisternal perfusate by lumbar intrathecal capsaicin in the rat

A model employing perfusion of artificial cerebrospinal fluid from the lateral ventricle to the cisterna magna in the halothane anesthetized rat was used to study beta-endorphin release in the brain. Injection of 75 micrograms capsaicin into the lumbar intrathecal space released beta-endorphin immunoreactivity into perfusate. The release was blocked by intrathecal pretreatment with 1.25 mg lidocaine and the capsaicin receptor antagonist capsazepine (92 micrograms), showing that the release is caused by binding of capsaicin to a spinal receptor. The release was also blocked by intrathecal pretreatment with the NMDA antagonist MK-801 (3 micrograms) and the NK-1 receptor antagonist CP96,345 (200 micrograms), whereas the AMPA receptor antagonist NBQX (6 micrograms) yielded no significant inhibition. Surprisingly, morphine (30 micrograms) and sufentanil (1.5 micrograms) did not prevent release of beta-endorphin immunoreactivity, although blocking the cardiovascular responses to a noxious heat stimulus. High performance liquid chromatography characterization of perfusates collected after capsaicin injection showed that all beta-endorphin immunoreactivity coeluted with authentic beta-endorphin1-31. beta-Endorphin immunoreactivity in plasma was increased 10 min, but not 25 min, after capsaicin injection. Capsaicin injection abolished the motor and cardiovascular responses to tail immersion in 52.5 degrees C water. Addition of MK-801 (10(-4) mol/l) to the lateral ventricle-cisterna magna perfusate blocked the capsaicin-induced beta-endorphin release, showing that our previous demonstration of an NMDA receptor regulating arcuate nucleus beta-endorphin neuron activity has functional significance. We conclude that in this in vivo, anesthetized preparation including three hot water tail immersions, beta-endorphin can be released into a ventriculo-cisternal perfusate, by activation of the central axons of small primary afferent neurons by capsaicin. These data support the idea that central beta-endorphin may be released in response to prolonged, intense noxious stimulation.

Estrogen modulation of two subpopulations of β-endorphin neurons in ovariectomized guinea pigs distinguished by peripherally injected fluorogold

β-endorphin released by neurons in the arcuate nucleus affects the output of several neuroendocrine systems and estrogen levels modulate the production and secretion of β-endorphin. We used intraperitoneal injection of fluorogold to retrogradely label the cell bodies of neurons that project outside the blood-brain-barrier in conjunction with immunohistochemistry for β-endorphin to dual label the subpopulation of β-endorphin neurons that project to the median eminence or other sites of access to the peripheral circulation. We found that some identified β-endorphin neurons in the arcuate nucleus of ovariectomized guinea pigs sequestered fluorogold. Approximately 7% of β-endorphin-containing cells co-localized with fluorogold. The effect of estrogen on the number of identified β-endorphin cells was examined. A single estradiol benzoate injection to ovariectomized guinea pigs 24 h prior to sacrifice dramatically decreased the total number of β-endorphin cells identified in the rostral, medial and the caudal portions of the arcuate nucleus. Also, a significantly smaller percentage of fluorogold-filled cells was found to contain β-endorphin immunoreactivity in the estrogen-treated group. These data suggest that a subpopulation of β-endorphin neurons has access to the peripheral circulation and may alter the output of neurosecretory terminals at the level of the median eminence. Furthermore, estrogen affects this subpopulation and the general population of β-endorphin neurons in the arcuate nucleus in a similar manner.

Cerebrospinal fluid beta-endorphin in models of hyperalgesia in the rat

Cerebrospinal fluid (CSF) obtained by acute percutaneous puncture of the cisternal membrane of the halothane anesthetized rat has low but measurable concentrations of beta-endorphin-like immunoreactivity (beta-EPir: 32.8 +/- 3.0 pmol/l). Chromatographic separation of beta-EPir showed that authentic beta-endorphin1-31 was the main component of beta-EPir in cisternal CSF. Subcutaneous injection of 5% formalin in the hind paws did not increase beta-EPir in cisternal CSF. Rats with tactile paw hyperalgesia evoked by unilateral ligation of the L5/6 nerve roots 2 weeks earlier had beta-EPir concentrations that did not differ from sham operated or unoperated control animals. In contrast, capsaicin injected in the hindpaws increased the mean beta-EPir concentration compared to saline injections (P = 0.006) 45 min after emerging from anesthesia following injection. These results show that acute activation of C fibers (by capsaicin) will evoke the release of beta-endorphin into the CSF, suggesting activation of the beta-endorphin terminal systems in the brain/midbrain. The failure of formalin injections to release beta-EPir to CSF may be due to specificity of the afferent stimulus evoking beta-EPir release, a lower stimulus intensity, and/or the duration of the stimulus generated by formalin. The normal concentrations of beta-EPir found in the hyperalgesic state following nerve injury suggest that the supraspinal beta-endorphin system does not display tonic changes under such conditions.

Release into ventriculo-cisternal perfusate of beta-endorphin- and Met-enkephalin-immunoreactivity: effects of electrical stimulation in the arcuate nucleus and periaqueductal gray of the rat

To examine the resting and evoked release of the endogenous opioid peptides beta-endorphin and Met-enkephalin from brain, we examined the levels of the respective immunoreactivities in the lateral ventricle-cisterna magna perfusate of the halothane-anesthetized rat. Ten Hz but not 100 Hz stimulation in the arcuate nucleus (ARC) of the hypothalamus released beta-endorphin immunoreactivity (beta-EPir) to the perfusate, whereas 100 Hz but not 10 Hz stimulation in the periaqueductal gray (PAG) of the mid brain released Met-enkephalin immunoreactivity (MEir). MEir was not released by stimulation in ARC and beta-EPir was not released by stimulation in PAG. Characterization of the released beta-EPir and MEir by high performance liquid chromatography showed that authentic beta-endorphin and Met-enkephalin were the major constituents of beta-EPir and MEir, respectively. Systemic administration of the dopaminergic antagonist haloperidol increased plasma, but not perfusate levels of beta-EPir. Both the opioid antagonist naloxone and the NMDA antagonist MK-801 failed to affect beta-EPir or MEir release. ARC and PAG stimulated inhibited a nociceptive reflex (tail-dip in 52.5 degrees C water), and naloxone did not reliably reverse this inhibition. These data support the previously suggested possibility of opioid mediation of stimulation induced analgesia, although we were unable to confirm the theory by naloxone reversibility in this study. Furthermore, the data support the assumption that measurement of opioid peptides in cerebrospinal fluid is a relevant approach in research aimed at elucidating the physiological and pathophysiological roles of endogenous opioid peptides.

Release of beta-endorphin immunoreactivity from brain by activation of a hypothalamic N-methyl-D-aspartate receptor

Lateral ventricle-cisterna magna perfusion in the halothane-anesthetized rat was used as a model to study beta-endorphin release in the brain. Microinjection of N-methyl-D-aspartate into the arcuate nucleus of the hypothalamus released beta-endorphin immunoreactivity into perfusate and the release was blocked by systemic pretreatment with the N-methyl-D-aspartate antagonist dizocilpine (MK-801). N-methyl-D-aspartate microinjections did not increase beta-endorphin immunoreactivity in plasma, and pretreatment with dexamethasone did not prevent release of beta-endorphin immunoreactivity into perfusate, emphasizing that the released beta-endorphin immunoreactivity did not come from plasma. The non-N-methyl-D-aspartate glutamate receptor agonist alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid hydrobromide did not release beta-endorphin immunoreactivity. High-performance liquid chromatography characterization of perfusates collected after N-methyl-D-aspartate microinjection showed that a major part, but not all, of the beta-endorphin immunoreactivity co-eluted with authentic beta-endorphin. Microinjection of N-methyl-D-aspartate provoked an algogenic response in the anesthetized rat, and inhibited the motor and cardiovascular responses to tail immersion in 52.5 degrees C water. This block was reversed by pretreatment with MK-801, but not naloxone. Injection of alpha-amino-3-hydroxy-5-methyl-isoxazole-4-propionic acid hydrobromide elicited the same behavioral response and blocked the nociceptive tail-dip reaction, but did not release beta-endorphin immunoreactivity.(ABSTRACT TRUNCATED AT 250 WORDS)

Immunohistochemical mapping of neuropeptides in the premamillary region of the hypothalamus in rats

The topographical distribution of neuropeptide-containing cell bodies, fibers and terminals was studied in the premamillary region of the rat hypothalamus using light microscopic immunohistochemistry. Alternate coronal sections through the posterior third of the hypothalamus of normal and colchicine-treated male rats were immunostained for 19 different neuropeptides and their distributions were mapped throughout the following structures: the ventral and dorsal premamillary, the supramamillary, the tuberomamillary and the posterior hypothalamic nuclei, as well as the premamillary portion of the arcuate nucleus and the postinfundibular median eminence. Seventeen of the investigated neuropeptides were present in neuronal perikarya, nerve fibers and terminals while the gonadotropin associated peptide and vasopressin occurred only in fibers and terminals. Growth hormone-releasing hormone-, somatostatin-, alpha-melanocyte stimulating hormone-, adrenocorticotropin-, beta-endorphin- and neuropeptide Y-immunoreactive neurons were seen exclusively in the premamillary portion of the arcuate nucleus. Thyrotropin-releasing hormone-, dynorphin A- and galanin-containing neurons were distributed mainly in the arcuate and the tuberomamillary nuclei. A high number of methionine- and leucine-enkephalin-immunoreactive cells were detected in the arcuate and dorsal premamillary nuclei, as well as in the area ventrolateral to the fornix. Substance P-immunoreactive perikarya were present in very high number within the entire region, in particular in the ventral and dorsal premamillary nuclei. Cell bodies labelled with cholecystokinin- and calcitonin gene-related peptide antisera were found predominantly in the supramamillary and the terete nuclei, respectively. Corticotropin-releasing hormone-, vasoactive intestinal polypeptide- and neurotensin-immunoreactive neurons were scattered randomly in low number, mostly in the arcuate and the ventral and dorsal premamillary nuclei. Peptidergic fibers were distributed unevenly throughout the whole region, with each peptide showing an individual distribution pattern. The highest density of immunoreactive fibers was presented in the ventral half of the region including the arcuate, the ventral premamillary and the tuberomamillary nuclei. The supramamillary nucleus showed moderately dense fiber networks, while the dorsal premamillary and the posterior hypothalamic nuclei were poor in peptidergic fibers.

Effects of an acute immobilization stress upon proopiomelanocortin (POMC) mRNA levels in the mediobasal hypothalamus: a quantitative in situ hybridization study

The aim of this study was to examine by quantitative in situ hybridization the effects of an acute stress on the expression of the POMC gene in the mediobasal hypothalamus (MBH) of the rat. In control animals, the highest levels of POMC mRNA were observed in the posterior periventricular region of the MBH. Lower levels were found in the anterior and posterior arcuate nucleus. At the end of a one hour immobilization, a small decrease (-8%) was observed in the periventricular region only. Four hours after the end of immobilization, increases in POMC mRNA levels were detected in the anterior part (7%), in the posterior part (25%) and in the periventricular region (13%) of the MBH. These results suggest that MBH POMC-derived peptides might be an important component in the central response to stress.

Release of beta-endorphin and methionine-enkephalin into cerebrospinal fluid during deep brain stimulation for chronic pain. Effects of stimulation locus and site of sampling

The authors systematically studied the release of the endogenous opioid peptides beta-endorphin and methionine (met)-enkephalin into the cerebrospinal fluid (CSF) during deep brain stimulation in patients suffering from otherwise intractable chronic pain. Nine patients were included in the study; six had stimulation electrodes placed in both the periventricular gray matter (PVG) and the thalamic nucleus ventralis posterolateralis (VLP) and three in the PVG only. Immunoreactivity of beta-endorphin and met-enkephalin (beta-EPir and MEir, respectively) was measured by radioimmunoassays in ventricular and lumbar CSF samples obtained before, during, and after stimulation. Prestimulation concentrations of beta-EPir and MEir were lower in ventricular than in lumbar CSF (6.6 +/- 0.5 vs. 13.7 +/- 1.0 pmol/liter, p = 0.0001, for beta-EPir; 33.6 +/- 5.1 vs. 48.3 +/- 3.2 pmol/liter, p < 0.05, for MEir). Ventricular CSF concentrations of both beta-EPir and MEir increased significantly during PVG stimulation, whereas VPL stimulation was without effect. No changes were seen in lumbar CSF levels of the peptides during stimulation in either site. A significant inverse relationship was found between the "during:before stimulation" ratios of visual analog scale ratings and beta-EPir levels during PVG stimulation. The beta-EPir and MEir concentration during:before stimulation ratios were positively correlated, whereas no correlation was present in prestimulation samples from ventricular or lumbar CSF. High-performance liquid chromatography of ventricular CSF pools obtained during PVG stimulation revealed that major portions of beta-EPir and MEir eluted as synthetic beta-endorphin and met-enkephalin, respectively, thus documenting the release of beta-endorphin and met-enkephalin into ventricular CSF during PVG stimulation. The finding of a direct relationship between beta-EPir release and pain alleviation may suggest a role for beta-endorphin in the analgesic mechanism of PVG stimulation.

Neuronal projections to the medial preoptic area of the sheep, with special reference to monoaminergic afferents: immunohistochemical and retrograde tract tracing studies

The preoptic area contains most of the luteinizing hormone releasing hormone immunoreactive neurons and numerous monoaminergic afferents whose cell origins are unknown in sheep. Using tract tracing methods with a specific retrograde fluorescent tracer, fluorogold, we examined the cells of origin of afferents to the medial preoptic area in sheep. Among the retrogradely labeled neurons, immunohistochemistry for tyrosine hydroxylase, dopamine-beta-hydroxylase, phenylethanolamine N-methyltransferase, and serotonin was used to characterize catecholamine and serotonin fluorogold labeled neurons. Most of the afferents came from the ipsilateral side to the injection site. It was observed that the medial preoptic area received major inputs from the diagonal band of Broca, the lateral septum, the thalamic paraventricular nucleus, the lateral hypothalamus, the area dorsolateral to the third ventricle, the perimamillary area, the amygdala, and the ventral part of the hippocampus. Other numerous, scattered, retrogradely labeled neurons were observed in the ventral part of the preoptic area, the vascular organ of the lamina terminalis, the ventromedial part of the hypothalamus, the periventricular area, the area lateral to the interpeduncular nucleus, and the dorsal vagal complex. Noradrenergic afferents came from the complex of the locus coeruleus (A6/A7 groups) and from the ventro-lateral medulla (group A1). However, dopaminergic and adrenergic neuronal groups retrogradely labeled with fluorogold were not observed. Serotoninergic fluorogold labeled neurons belonged to the medial raphe nucleus (B8, B5) and to the serotoninergic group situated lateral to the interpeduncular nucleus (S4). In the light of these anatomical data we hypothesize that these afferents have a role in the regulation of several functions of the preoptic area, particularly those related to reproduction. Accordingly these afferents could be involved in the control of luteinizing hormone releasing hormone (LHRH) pulsatility or of preovulatory LHRH surge.

Aging changes in the beta-endorphin neuronal system in the preoptic area of the C57BL/6J mouse: ultrastructural analysis

In hypothalami of aging rodents, beta-endorphin (beta-EP) neuron number and content are reduced. The objectives of this study were: first, to analyze ultrastructurally the population of neuronal elements in a selected region of the preoptic area (POA) in young and old mice; second, to study the beta-EP neuronal system in the same region to determine whether or not this population remains stable with age. Vibratome sections from the most caudal POA through the diagonal band of Broca were examined by light microscopy and immunocytochemistry in mature, cycling (5-6 months old) and old, acyclic, disease-free (24-26 months old) mice. A subset of beta-EP-like perikarya and associated structures was observed in the periventricular POA. When this subregion was examined at the ultrastructural level, there was a significant decrease in the number of recipient dendrites [3.78 +/- 0.04 SEM/micron 2 young vs. 0.82 +/- 0.03/micron 2 old; p < 0.007, analysis of variance (ANOVA)], but a significant increase in the number of nonmyelinated axons (20.0 +/- 2.6/micron 2 young vs. 26.8 +/- 0.7/micron 2 old; p < 0.05). Immunolabeled terminals that contained a synapse comprised 2.56 +/- 0.08% of all terminals with synapses in young mice but only 0.34 +/- 0.04% in old ones when corrected for surface area examined (p < 0.03). A significant age-related loss was also observed in the nonmyelinated beta-EP-labeled axon population (1.50 +/- 0.10% young vs. 0.40 +/- 0.01% old; p < 0.009, ANOVA). We conclude that there are critical changes in the microenvironment of the POA in old, noncycling female mice that are likely to affect neuron function.

Loss during aging of beta-endorphinergic neurons in the hypothalamus of female C57BL/6J mice

Beta-endorphin (B-EP) content is often reduced in hypothalami of aging rodents. The objective of this study was to determine whether reduced B-EP content is associated with a reduced number of B-EP immunoreactive neurons. Serial coronal sections extending from the caudal hypothalamus through the retrochiasmatic area were examined by quantitative light microscopy in mature (5-6 month) and senescent (24-28 month) mice that had been ovariectomized 1 week earlier and injected with colchicine 24-48 h before sacrifice. Old mice were acyclic. As expected, B-EP immunoreactive cell bodies were restricted to the region of the arcuate nucleus. There was a 35% loss of B-EP immunopositive neurons in old, macroscopically disease-free animals. By contrast, some old animals with pituitary tumors had no loss of B-EP neurons. These results suggest that a subpopulation of B-EP neurons either die or stop synthesizing detectable concentrations of B-EP in aged mice. The basis for the absence of reduced B-EP neurons in some mice with pituitary tumors is unclear, but this observation underscores the importance of distinguishing age-related changes associated with diseases of aging from those that are independent of such diseases.

Intrahypothalamic Mu-, not Delta- or Kappa-Opioid Receptor Activation Causes Growth Hormone Secretion

Abstract The possible effects of opioid receptor agonists on growth hormone (GH)-releasing factor or somatostatin neurons were examined by measuring the effects of localized intracerebral injections of mu-, delta- and kappa-selective agonists on GH secretion. Serial GH concentrations were measured in plasma in unanaesthetized male rats chronically prepared with venous and intracerebral cannulae, before and after treatment with bilateral intracerebral injections of opioid agonists in the preoptic anterior hypothalamic area and medial basal hypothalamus. In the medial basal hypothalamus, injections of the mu-agonist DAGO (Tyr-D-Ala-Gly-(Me)Phe-Gly-ol) caused dose-responsive increases in GH, the maximally effective dose being 0.001 nmoles. Injection of 10,000-fold higher doses of the delta-agonist DPDPE ([D-Pen, D-Pen]enkephalin) and the kappa-agonist U50,488H were also effective in stimulating GH secretion. In the preoptic anterior hypothalamic area, DAGO caused dose-responsive increases in GH, the maximally effective dose being 0.01 nmoles. U50.488H was ineffective at 1,000-fold higher doses while DPDPE was effective at 100- to 1,000-fold higher doses. We conclude that hypothalamic mu-opioid receptor activation on or near somatostatin or GH-releasing factor neurons causes GH secretion. Opioids capable of acting on other opioid receptors may also stimulate GH secretion, though only at doses that seem likely to affect mu-receptors.

Activation of opioid receptors in the mediobasal hypothalamus stimulates prolactin secretion in the conscious rat

Abstract In an attempt to localize the opioid receptor(s) (mu, delta and kappa) involved in opioid-stimulated prolactin release in the conscious male rat, opioid agonists were microinjected into the mediobasal hypothalamus and prolactin levels measured before and after injection. The specific mu agonist, DAGO ((D-Ala(2), NMe-Phe(4), Gly-ol(5))-enkephalin) was the most effective in eliciting prolactin release, the smallest effective dose being 0.01 nmoles. The specific delta agonist, DPDPE ((D-Pen(2), D-Pen(5))-enkephalin) had no significant effect even at the highest dose of 10 nmoles. The specific kappa agonist, U50,488H ((trans-3,4-dichloro-N-methyl-N-(2-(1-pyrrolidinyl)-cyclohexyl)-benzene acetamide) was effective at the doses 1.0 and 10 nmoles. We conclude that mu and kappa opioid receptors in the mediobasal hypothalamus are involved in opioid-stimulated prolactin release and that delta receptors are not.

beta-Endorphin and gonadotropin-releasing hormone synaptic input to gonadotropin-releasing hormone neurosecretory cells in the male rat

Physiological and pharmacological evidence has suggested that both endogenous opiates and gonadotropin-releasing hormone (GnRH) itself can act centrally to exert a tonic inhibition on gonadotropin secretion via an inhibition of the neurosecretion of GnRH. To determine if the effects of these two peptides might be mediated via a direct synaptic input to the GnRH neuron, we undertook a double label ultrastructural study. We were able to localize in the same tissue section beta-endorphin and GnRH. Analysis of serial sections through GnRH perikarya and dendrites in the male rat diagonal band/preoptic area revealed that almost 10% of the synapses impinging on the GnRH neuron contained beta-endorphin; an additional 10% of the terminals contained GnRH. These data provide anatomical evidence in support of both a direct modulation of GnRH release by opiates and of the presence of an ultrashort feedback loop.

Differential colocalization of neuropeptide Y- and methionine-enkephalin-Arg6-Gly7-Leu8-like immunoreactivity in catecholaminergic neurons in the rat brain stem

The present study, using a combination of catecholamine (CA) histofluorescence and peptide immunocytochemistry in the same tissue sections, investigated the coexistence of neuropeptide Y (NPY) and methionine-enkephalin-Arg6-Gly7-Leu8 (MEAGL)-like immunoreactivity (LI) in catecholaminergic neurons of colchicine-treated rat brain stems. Of the total number of catecholaminergic neurons in the A1/C1, A2/C2, A3, A4, and A6 regions approximately 83, 28, 98, 76, and 36%, respectively, contained both NPY-LI and CA. Of the total number of catecholaminergic neurons in A1/C1, A2/C2, A3, and A5 regions, approximately 47, 4, 8, and 17%, respectively, contained both MEAGL-LI and CA. Moreover, about 24% of the catecholaminergic neurons in the A1/C1 region contained both NPY- and MEAGL-LI. Neither the noradrenergic neurons (A7) in the pons nor any of the dopaminergic neurons in the midbrain (A8, A9, A10) contained NPY- or MEAGL-LI. Neurons containing both NPY- and MEAGL-like immunoreactive peptides without CA were not found in the rat brain stem. These findings indicate that catecholaminergic neurons in the brain stem of the rat can be subdivided into distinct subgroups on the basis of the coexistence of specific peptides.

The effects of nitrous oxide on the secretory activity of pro-opiomelanocortin peptides from basal hypothalamic cells attached to cytodex beads in a superfusion in vitro system

Dispersed cells from adult rat basal hypothalami, attached to Cytodex-3 microcarrier beads, were placed in a column and superfused with aerated high glucose media or media enriched with variable concentrations of nitrous oxide with oxygen. beta-Endorphin and alpha-MSH content was measured in the effluent collected during superfusion and demonstrated a near constant baseline release. Nitrous oxide, 60% (P less than 0.025) and 80% (P less than 0.02), caused significant increases in release of beta-endorphin. Potassium chloride (50 mM) caused a significant increase in release (P less than 0.007) of beta-endorphin whereas saline and 30% nitrous oxide did not. Neither nitrous oxide-enriched media nor potassium chloride had any statistically significant effect on alpha-MSH release. The increase in beta-endorphin secretory activity during exposure to nitrous oxide demonstrates that nitrous oxide may have a stimulatory effect on central pro-opiomelanocortin neurons.

Immunocytochemical distribution of met-enkephalin-Arg6-Gly7-Leu8 in the rat lower brainstem

The distribution of methionine-enkephalin-Arg6-Gly7-Leu8, a unique peptide derived from proenkephalin A in the rat brainstem, was studied immunocytochemically by using a highly specific antiserum to this octapeptide sequence. Immunoreactive perikarya with various shapes and sizes were detected in many regions of the rat brainstem. Dense accumulation of immunoreactive perikarya and fibers was seen in the nuclei associated with special sensory and visceral functions, such as the interpeduncular nucleus, the parabrachial nucleus, the nucleus of the solitary tract, and the nucleus of the spinal tract of the trigeminal nerve. Clusters of methionine-enkephalin-Arg6-Gly7-Leu8-like immunoreactive perikarya and fibers were observed in certain areas considered to play a role in nociception and analgesia, such as the central gray of the midbrain central gray and the raphe magnus nucleus. Some methionine-enkephalin-Arg6-Gly7-Leu8-like immunoreactive perikarya were distributed in the lateral reticular nucleus, the nucleus of the solitary tract, and the raphe magnus nucleus, where monoaminergic neurons were also detected. In addition to the previously reported enkephalinergic cells, we found many methionine-enkephalin-Arg6-Gly7-Leu8 containing neurons; the rostral and caudal linear nucleus of raphe, the median raphe nucleus, entire length of the raphe magnus nucleus, the medial longitudinal fasciculus, the cuneate nucleus, the external cuneate nucleus, the gracile nucleus, and the area postrema. The wide distribution of this octapeptide-like immunoreactivity reflected neurons expressing the preproenkephalin A gene distributed more widely than previously reported and that innervated many regions.

Light and electron microscopic immunocytochemistry of GRF-like immunoreactive neurons and terminals in the rat hypothalamic arcuate nucleus and median eminence

Growth hormone-releasing factor (GRF) synthesizing neuronal perikarya and terminals were investigated by light and electron microscopic immunocytochemistry using rat hypothalamus. Immunoreactive neuronal perikarya were located mainly in the ventrolateral part of the arcuate nucleus. They contained well developed cell organella such as mitochondria and rough surfaced endoplasmic reticulum with some expansion. They also contained immunoreactive dense granules (80-120 nm in diameter). On the surface of the immunoreactive neuronal perikarya were frequently found non-immunoreactive axo-somatic synapses. Therefore, the GRF-like immunoreactive neurons were assumed to receive neuronal inputs from other neurons on their neuronal soma. In the external layer of the median eminence large numbers of immunoreactive terminals were distributed particularly around the capillaries of the portal vessel. Electron microscopic immunocytochemistry revealed large numbers of immunoreactive terminals containing immunoreactive dense granules, synaptic vesicles and mitochondria in the vicinity of the basement membrane of the pericapillary space of the portal vessel. Therefore, we concluded that GRF-like immunoreactive substances are released into the portal capillaries from the nerve terminals, which originate from the neuronal perikarya in the ventrolateral part of the arcuate nucleus, and act on growth hormone release in the anterior pituitary. We also suggest that GRF-like immunoreactive neurons have abundant terminal arborization in the external layer of the median eminence.