Excitatory projections from the anterior hypothalamus to periaqueductal gray neurons that project to the medulla: a functional anatomical study

Modulation of social investigation by anterior hypothalamic nucleus rhythmic neural activity

Social interactions involve both approach and avoidance toward specific individuals. Currently, the brain regions subserving these behaviors are not fully recognized. The anterior hypothalamic nucleus (AHN) is a poorly defined brain area, and recent studies have yielded contradicting conclusions regarding its behavioral role. Here we explored the role of AHN neuronal activity in regulating approach and avoidance actions during social interactions. Using electrophysiological recordings from behaving mice, we revealed that theta rhythmicity in the AHN is enhanced during affiliative interactions, but decreases during aversive ones. Moreover, the spiking activity of AHN neurons increased during the investigation of social stimuli, as compared to objects, and was modulated by theta rhythmicity. Finally, AHN optogenetic stimulation during social interactions augmented the approach toward stimuli associated with the stimulation. These results suggest the role for AHN neural activity in regulating approach behavior during social interactions, and for theta rhythmicity in mediating the valence of social stimuli.

Midcingulate somatomotor and autonomic functions

The midcingulate cortex (MCC) is viewed as a central node within a large-scale system devoted to adjusting behavior in the face of changing environments. Whereas the role of the MCC in interfacing action and cognition is well established, its role in regulating the autonomic nervous system is poorly understood. Yet, adaptive reactions to novel or threatening situations induce coordinated changes in the sympathetic and the parasympathetic systems. The somatomotor maps in the MCC are organized dorsoventrally. A meta-analysis of the literature reveals that the dorsoventral organization might also concern connections with the autonomic nervous system. Activation of the dorsal and ventral parts of the MCC correlate with recruitments of the sympathetic and the parasympathetic systems, respectively. Data also suggest that, in the MCC, projections toward the sympathetic system are mapped along the sensory-motor system following the same cervico-sacral organization as projections on the spinal cord for skeletal motor control.

Role played by periaqueductal gray neurons in parasympathetically mediated fear bradycardia in conscious rats

Freezing, a characteristic pattern of defensive behavior elicited by fear, is associated with a decrease in the heart rate. Central mechanisms underlying fear bradycardia are poorly understood. The periaqueductal gray (PAG) in the midbrain is known to contribute to autonomic cardiovascular adjustments associated with various emotional behaviors observed during active or passive defense reactions. The purpose of this study was to elucidate the role played by PAG neurons in eliciting fear bradycardia. White noise sound (WNS) exposure at 90 dB induced freezing behavior and elicited bradycardia in conscious rats. The WNS exposure-elicited bradycardia was mediated parasympathetically because intravenous administration of atropine abolished the bradycardia (P < 0.05). Moreover, WNS exposure-elicited bradycardia was mediated by neuronal activation of the lateral/ventrolateral PAG (l/vlPAG) because bilateral microinjection of muscimol, a GABAA agonist, into the l/vlPAG significantly suppressed the bradycardia. It is noted that muscimol microinjected bilaterally into the dorsolateral PAG had no effect on WNS exposure-elicited bradycardia. Furthermore, retrograde neuronal tracing experiments combined with immunohistochemistry demonstrated that a number of l/vlPAG neurons that send direct projections to the nucleus ambiguus (NA) in the medulla, a major origin of parasympathetic preganglionic neurons to the heart, were activated by WNS exposure. Based on these findings, we propose that the l/vlPAG-NA monosynaptic pathway transmits fear-driven central signals, which elicit bradycardia through parasympathetic outflow.

Hypovolemic hemorrhage induces Fos expression in the rat hypothalamus: Evidence for involvement of the lateral hypothalamus in the decompensatory phase of hemorrhage

This study tested the hypothesis that the hypothalamus participates in the decompensatory phase of hemorrhage by measuring Fos immunoreactivity and by inhibiting neuronal activity in selected hypothalamic nuclei with lidocaine or cobalt chloride. Previously, we reported that inactivation of the arcuate nucleus inhibited, but did not fully prevent, the fall in arterial pressure evoked by hypotensive hemorrhage. Here, we report that hemorrhage (2.2 ml/100g body weight over 20 min) induced Fos expression in a high percentage of cells in the paraventricular, supraoptic and arcuate nuclei of the hypothalamus as shown previously. Lower densities of Fos immunoreactive cells were also found in the medial preoptic area (mPOA), anterior hypothalamus, lateral hypothalamus (LH), dorsomedial hypothalamus, ventromedial hypothalamus (VMH) and posterior hypothalamus. Bilateral injection of lidocaine (2%; 0.1 μl or 0.3 μl) or cobalt chloride (5mM; 0.3 μl) into the tuberal portion of the LH immediately before hemorrhage was initiated reduced the magnitude of hemorrhagic hypotension and bradycardia significantly. Lidocaine injection into the VMH also attenuated the fall in arterial pressure and heart rate evoked by hemorrhage although inactivation of the mPOA or rostral LH was ineffective. These findings indicate that hemorrhage activates neurons throughout much of the hypothalamus and that a relatively broad area of the hypothalamus, extending from the arcuate nucleus laterally through the caudal VMH and tuberal LH, plays an important role in the decompensatory phase of hemorrhage.

Reduction in 50-kHz call-numbers and suppression of tickling-associated positive affective behaviour after lesioning of the lateral hypothalamic parvafox nucleus in rats

The parvafox nucleus is located ventrolaterally in the lateral hypothalamic area (LHA). Its core and shell are composed of neurons expressing the calcium-binding protein parvalbumin (PV) and the transcription factor Foxb1, respectively. Given the known functions of the LHA and that the parvafox nucleus receives afferents from the lateral orbitofrontal cortex and projects to the periaqueductal gray matter, a functional role of this entity in the expression of positive emotions has been postulated. The purpose of the present study was to ascertain whether the deletion of neurons in the parvafox nucleus influenced the tickling-induced 50-kHz calls, which are thought to reflect positive affective states, in rats. To this end, tickling of the animals (heterospecific play) was combined with intracerebral injections of the excitotoxin kainic acid into the parvafox nucleus. The most pronounced surgery-associated reduction in 50-kHz call-numbers was observed in the group of rats in which, on the basis of PV-immunoreactive-cell counts in the parvafox nucleus, bilateral lesions had been successfully produced. Two other parameters that were implemented to quantify positive affective behaviour, namely, an approach towards and a following of the hand of the tickling experimenter, were likewise most markedly suppressed in the group of rats with bilaterally successful lesions. Furthermore, positive correlations were found between each of the investigated parameters. Our data afford evidence that the parvafox nucleus plays a role in the production of 50-kHz calls in rats, and, more generally, in the expression of positive emotions.

Role of thalamic nuclei in the modulation of Fos expression within the cerebral cortex during hypertonic saline-induced muscle nociception

It has been proposed that thalamic mediodorsal (MD) and ventromedial (VM) nuclei form thalamic 'nociceptive discriminators' in discrimination of nociceptive afferents, and specifically govern endogenous descending facilitation and inhibition. The present study conducted in rats was to explore the role of thalamic MD and VM nuclei in modulation of cerebral neuronal activities by means of detection of spatiotemporal variations of Fos expression within the cerebral cortex. Following a unilateral intramuscular injection of 5.8% saline into the gastrocnemius muscle, Fos expression within the bilateral, different areas of the cerebral cortex except S2 was significantly increased (P<0.05). Particularly, the increases in Fos expression within the cingulate cortex and the insular cortex occurred at 0.5h, 4h and reached the peak level at 4h, 16h, respectively. Electrolytic lesion of the contralateral thalamic MD and VM nuclei significantly blocked the 5.8% saline intramuscularly induced increases in Fos expression within the bilateral cingulate and insular cortices, respectively. Additionally, the 5.8% saline-induced Fos expression in the cingulate cortex and the insular cortex were dose-dependently attenuated by microinjection of μ-opioid antagonist β-funaltrexamine hydrochloride into the thalamic MD and VM nuclei. It is suggested that (1) the neural circuits of 'thalamic MD nucleus - cingulate cortex' and 'thalamic VM nucleus - insular cortex' form two distinct pathways in the endogenous control of nociception, (2) mirror or contralateral pain is hypothesized to be related to cross-talk of neuronal activities within the bilateral cerebral cortices modulated by μ-opioid receptors within the thalamic MD and VM nuclei.

Selective inhibition from the anterior hypothalamus of C- versus A-fibre mediated spinal nociception

Modulation of spinal nociception from the anterior hypothalamus/preoptic area (AH/POA), and consequent alterations in the pain experience may contribute to integrated responses brought into play during fear or stress and as part of the sickness response. This study was designed to compare the effects of descending control from AH/POA on A- versus C-fibre-evoked spinal nociception, since any differential control is of behavioural and clinical importance given that A-fibre and C-fibre nociceptors convey different qualities of the pain signal (first and second pain, respectively), and play different roles in the development and maintenance of chronic pain states. In anaesthetised rats, electromyographic responses were recorded to monitor thresholds of withdrawal to slow (2.5 degrees Cs(-1)) or fast (7.5 degrees Cs(-1)) rates of skin heating of the hindpaw, to preferentially activate C- or A-nociceptors, respectively. Neuronal activation by microinjection of dl-homocysteic acid at sites within a specific region of AH/POA, lateral area of the anterior hypothalamus (LAAH), significantly increased response thresholds to slow heating rates (p<0.02, n=11), but not those to fast rates of heating (p=0.48, n=10). Injection of DLH adjacent to LAAH (n=9) had no significant effect on responses to slow (n=8) or fast (n=9) rates of skin heating. The functional significance of differential descending control of spinal processing of C- and A-nociceptive inputs is discussed with respect to roles both of the LAAH in pain processing, and of C- and A-nociceptive inputs in acute and chronic pain.

[Mechanisms of hypertension in the central nervous system]

This article reviews studies by the author on central mechanisms of hypertension. Spontaneously hypertensive rats (SHR) have been developed as a rat model of genetic hypertension, and central acetylcholine has been implicated in hypertension in SHR. The rostral ventrolateral medulla (RVL), a major source of efferent sympathetic activity, has cholinergic pressor systems. The release of acetylcholine is enhanced in the RVL of SHR, leading to hypertension. The alteration of the RVL cholinergic system in SHR results from enhanced angiotensin systems in the anterior hypothalamic area (AHA). Angiotensin II-sensitive neurons are present in the AHA and they are tonically activated by endogenous angiotensins. The basal activity of AHA angiotensin II-sensitive neurons is enhanced in SHR, mainly due to enhanced sensitivity of AHA neurons to angiotensin II. The AHA angiotensin system is also responsible for hypertension induced by emotional stress and central Na(+) increases. These findings suggest that the AHA angiotensin system may play a critical role in the development of hypertension.

Inhibitory modulation of chemoreflex bradycardia by stimulation of the nucleus raphe obscurus is mediated by 5-HT3 receptors in the NTS of awake rats

Several studies demonstrated the involvement of 5-hydroxytryptamine (5-HT) and its different receptor subtypes in the modulation of neurotransmission of cardiovascular reflexes in the nucleus tractus solitarii (NTS). Moreover, anatomic evidence suggests that nucleus raphe obscurus (ROb) is a source of 5-HT-containing terminals within the NTS. In the present study we investigated the possible changes in the cardiovascular responses to peripheral chemoreceptor activation by potassium cyanide (KCN, i.v.) following ROb stimulation with L-glutamate (10 nmol/50 nL) and also whether 5-HT3 receptors in the caudal commissural NTS are involved in this neuromodulation. The results showed that stimulation of the ROb with L-glutamate in awake rats (n=15) produced a significant reduction in the bradycardic response 30 s after the microinjection (-182+/-19 vs -236+/-10 bpm; Wilcoxon test) but no changes in the pressor response to peripheral chemoreceptor activation (43+/-4 vs 51+/-3 mmHg; two-way ANOVA) in relation to the control. Microinjection of 5--HT3 receptors antagonist granisetron (500 pmol/50 nL), but not the vehicle, into the caudal commissural NTS bilaterally prevented the reduction of chemoreflex bradycardia in response to microinjection of L-glutamate into ROb. These data indicate that 5-HT-containing projections from ROb to the NTS play an inhibitory neuromodulatory role in the chemoreflex evoked bradycardia by releasing 5-HT and activating 5-HT3 receptors in the caudal NTS.

Muscimol injection into the lateral hypothalamus inhibits the hypotension and bradycardia caused by somato-visceral nociception

This study investigated whether the lateral hypothalamus (LH) contributes to the depressor response evoked by somato-visceral nociception. Lidocaine (2%; 0.1, 0.3 or 1.0 microl) or muscimol (0.34 nmol; 0.5 microl) was microinjected into the rostral LH of halothane-anesthetized rats bilaterally and somato-visceral nociception was induced 2 min later by injecting 5% acetic acid (0.5 ml) intraperitoneally. Lidocaine and muscimol inhibited the hypotension and bradycardia caused by somato-visceral nociception significantly without affecting cardiovascular function in normotensive animals.

Prostaglandin E2 in the medial preoptic area produces hyperalgesia and activates pain-modulating circuitry in the rostral ventromedial medulla

Prostaglandin E2 (PGE2) produced in the medial preoptic region (MPO) in response to immune signals is generally accepted to play a major role in triggering the illness response, a complex of physiological and behavioral changes induced by infection or injury. Hyperalgesia is now thought to be an important component of the illness response, yet the specific mechanisms through which the MPO acts to facilitate nociception have not been established. However, the MPO does project to the rostral ventromedial medulla (RVM), a region with a well-documented role in pain modulation, both directly and indirectly via the periaqueductal gray. To test whether PGE2 in the MPO produces thermal hyperalgesia by recruiting nociceptive modulating neurons in the RVM, we recorded the effects of focal application of PGE2 in the MPO on paw withdrawal latency and activity of identified nociceptive modulating neurons in the RVM of lightly anesthetized rats. Microinjection of a sub-pyrogenic dose of PGE2 (50 fg in 200 nl) into the MPO produced thermal hyperalgesia, as measured by a significant decrease in paw withdrawal latency. In animals displaying behavioral hyperalgesia, the PGE2 microinjection activated on-cells, RVM neurons thought to facilitate nociception, and suppressed the firing of off-cells, RVM neurons believed to have an inhibitory effect on nociception. A large body of evidence has implicated prostaglandins in the MPO in generation of the illness response, especially fever. The present study indicates that the MPO also contributes to the hyperalgesic component of the illness response, most likely by recruiting the nociceptive modulating circuitry of the RVM.

Hypothalamic and midbrain circuitry that distinguishes between escapable and inescapable pain

Characteristics of emotional, behavioral, and physiological responses to pain are determined to a large extent by the behavioral significance of the pain, in particular to the degree to which the pain can be escaped. This review presents evidence that these different patterns of response depend on the activation of distinct pathways within the brain.

Transneuronal labelling of nerve cells in the CNS of female rat from the mammary gland by viral tracing technique

Using the viral transneuronal tracing technique, the cell groups in the CNS transneuronally connected with the female mammary gland were detected. Lactating and non-lactating female rats were infected with pseudorabies virus injected into the mammary gland. The other group of animals was subjected to virus injection into the skin of the back. Four days after virus injection, infected neurons detected by immunocytochemistry, were present in the dorsal root ganglia ipsilateral to inoculation and in the intermediolateral cell column of the spinal cord. In addition, a few labelled cells could be detected in the dorsal horn and in the central autonomic nucleus (lamina X) of the spinal cord. At this survival time several brain stem nuclei including the A5 noradrenergic cell group, the caudal raphe nuclei (raphe obscurus, raphe pallidus, raphe magnus), the A1/C1 noradrenergic and adrenergic cell group, the nucleus of the solitary tract, the area postrema, the gigantocellular reticular nucleus, and the locus coeruleus contained virus-infected neurons. In some animals, additional cell groups, among others the periaqueductal gray and the red nucleus displayed labelling. In the diencephalon, a significant number of virus-infected neurons could be detected in the hypothalamic paraventricular nucleus. In most cases, virus-labelled neurons were present also in the lateral hypothalamus, in the retrochiasmatic area, and in the anterior hypothalamus. In the telencephalon, in some animals a few virus-infected neurons could be found in the preoptic area, in the bed nucleus of the stria terminalis, in the central amygdala, and in the somatosensory cortex. At the longer (5 days) survival time each cell group mentioned displayed immunopositive neurons, and the number of infected cells increased. The pattern of labelling was similar in animals subjected to virus inoculation into the mammary gland and into the skin. The distribution and density of labelling was similar in lactating and non-lactating rats. The present findings provide the first morphological data on the localization of CNS structures connected with the preganglionic neurons of the sympathetic motor system innervating the mammary gland. It may be assumed that the structures found virus-infected belong to the neuronal circuitry involved in the control of the sympathetic motor innervation of the mammary gland.

Evidence that hemorrhagic hypotension is mediated by the ventrolateral periaqueductal gray region

Severe hemorrhage lowers arterial pressure by suppressing sympathetic activity. This study tested the hypothesis that the decompensatory phase of hemorrhage is mediated by the ventrolateral periaqueductal gray (vlPAG), a region importantly involved in the autonomic and behavioral responses to stress and trauma. Neuronal activity in the vlPAG was inhibited with either lidocaine or cobalt chloride 5 min before hemorrhage (2.5 ml/100 g body wt) was initiated in conscious, unrestrained rats. Bilateral injection of lidocaine (0.5 microl of a 2% or 1 microl of a 5% solution) into the caudal vlPAG delayed the onset and reduced the magnitude of the hypotension produced by hemorrhage significantly. In contrast, inactivation of the dorsolateral PAG with lidocaine was ineffective. Cobalt chloride (5 mM; 0.5 microl), which inhibits synaptic transmission but not axonal conductance, also attenuated hemorrhagic hypotension significantly. Microinjection of lidocaine or cobalt chloride into the vlPAG of normotensive, nonhemorrhaged rats did not influence cardiovascular function. These data indicate that the vlPAG plays an important role in the response to hemorrhage.

The chemical architecture of the rat's periaqueductal gray based on acetylcholinesterase histochemistry: a quantitative and qualitative study

The chemoarchitecture of the periaqueductal gray has been extensively studied, based on acetylcholinesterase reaction and comparing it to other chemical markers. We have divided the periaqueductal gray into four main longitudinal columns, namely dorsomedial, dorsolateral, lateral and ventrolateral. We also identified the dorsal midline column, the supraoculomotor cap and the juxta-aqueductal ring. The acetylcholinesterase gave rise to a strong reaction in the outer half of the lateral column, the outer half of the dorsomedial column, the supraoculomotor cap and the ventral half of the juxta-aqueductal ring. This labeling was in part complementary to that of the NADPH diaphorase and allowed the lateral column to be differentiated from the ventrolateral column. However, the inner half of both lateral and ventrolateral columns displayed the same chemical properties including acetylcholinesterase, tyrosine hydroxilase and serotonin. Thus, from the chemical view, these inner halves should be considered as one different region. Finally, the juxta-aqueductal ring was composed of two clearly different halves, i.e. dorsal and ventral. The dorsal half did not show any clear differences from the above columns and was negative for acetylcholinesterase, NADPH diaphorase and tyrosine hydroxilase, while the ventral half was clearly different from the lateral and ventrolateral columns and displayed a positive reaction to all those chemical markers. From these results, we strongly suggest the use of acetylcholinesterase histochemistry as a tool for accurate parcellation of the periaqueductal gray.

Activation of the ventrolateral periaqueductal gray reduces locomotion but not mean arterial pressure in awake, freely moving rats

Activation of the ventrolateral periaqueductal gray produces immobility and antinociception. It has been argued that these behaviors are part of either a defensive fear response to threat or a recuperative quiescence response to deep tissue injury. Data collected in anesthetized animals showing that activation of the ventrolateral periaqueductal gray has a hypotensive effect supports the quiescence hypothesis. Our objective was to determine whether activation of the ventrolateral periaqueductal gray in awake, freely moving rats results in a decrease in blood pressure as it does in anesthetized animals. Changes in blood pressure produced by microinjection of the neuroexcitant D,L-homocysteic acid were measured using radio telemetry while rats were awake and while anesthetized with pentobarbital. Consistent with earlier reports, microinjection of D,L-homocysteic acid into the ventrolateral periaqueductal gray caused a decrease in blood pressure in anesthetized rats. In contrast, microinjection at the same ventrolateral periaqueductal gray sites while rats were awake had no effect on blood pressure, even though the animals became immobile and heart rate decreased. Thus, the immobility evoked from ventrolateral periaqueductal gray is not associated with a fall in mean arterial pressure. Two conclusions can be drawn from these data. (1) Caution must be used in generalizing from data collected in anesthetized animals. (2) The ventrolateral periaqueductal gray is as likely to contribute to defensive fear as to recuperative quiescence.

Inhibitory effects evoked from the rostral ventrolateral medulla are selective for the nociceptive responses of spinal dorsal horn neurons

The aim of the present study was to determine whether or not descending control of spinal dorsal horn neuronal responsiveness following neuronal activation at pressor sites in the rostral ventrolateral medulla is selective for nociceptive information. Extracellular single-unit activity was recorded from 49 dorsal horn neurons in the lower lumbar spinal cord of anaesthetized rats. The 30 Class 2 neurons selected for investigation responded to noxious (pinch and radiant heat) and non-noxious (prod, stroke and/or brush) stimulation within their cutaneous receptive fields on the ipsilateral hindpaw. The excitatory amino acid, DL-homocysteic acid, was microinjected into either the rostral or the caudal rostral ventrolateral medulla at sites that evoked increases in arterial blood pressure. Effects of neuronal activation at these sites were then tested on the responses of Class 2 neurons to noxious and non-noxious stimulation within their excitatory receptive fields. The noxious pinch and radiant heat responses of Class 2 neurons were depressed, respectively to 13+/-3.8% (n=23) and to 16+/-3.7% (n=18) of control, following stimulation at sites in the rostral rostral ventrolateral medulla. In contrast, the low-threshold (prod) responses of eight Class 2 neurons tested were not depressed following neuronal activation at the same sites. When tested, control injections of the inhibitory amino acid, GABA, at the same sites in the rostral rostral ventrolateral medulla had no significant effects on neuronal activity. Neither intravenous administration of noradrenaline (to mimic the pressor responses evoked by DL-homocysteic acid microinjections in the rostral ventrolateral medulla) nor activation at pressor sites in the caudal rostral ventrolateral medulla had any significant effect on neuronal responsiveness. With regard to sensory processing in the spinal cord, these data suggest that descending inhibitory control that originates from neurons in pressor regions of the rostral rostral ventrolateral medulla is highly selective for nociceptive inputs to Class 2 neurons. These data are discussed in relation to the role of the rostral ventrolateral medulla in executing the changes in autonomic and sensory functions that are co-ordinated by higher centres in the CNS.

Analysis of mouse plus-maze behavior modulated by ovarian steroids

Ovarian hormones, acting through membrane and nuclear receptors, can evoke different changes in psychophysiological status of the subject. Such effects are most crucial in female organism because of the natural fluctuations of ovarian hormones due to the estrous cycle. In this study, we have analyzed how estrogen and progesterone (P) affect the anxiety level in ovariectomized (OVX) mice. Using plus-maze paradigm, we have found that mice receiving injection of progesterone in 6 h revealed the lowest level of open arm activity and the highest grooming activity, as compared with other experimental groups. Time spent on the open arms and locomotor activity of OVX animals treated with estradiol benzoate (EB) did not differ from the vehicle-treated animals. In addition, the hormonal treatments modified the PR-immunoreactivity in many brain areas, such as medial preoptic area of hypothalamus, ventromedial hypothalamus, mammilar body of hypothalamus, septo-hippocampal region, and periaqueductal gray. The injections of EB and P induced sequential changes of PR-ir in comparison with vehicle-treated mice, whereas injection of vehicle induced the significant decrease of PR-ir in all brain areas examined. Our data suggest that sex steroid hormones play an important role in induction of anxiety. Furthermore, this action might be partially mediated through the classical steroid receptors.

Visceral inputs to neurons in the anterior hypothalamus including those that project to the periaqueductal gray: a functional anatomical and electrophysiological study

The present study was designed to examine peripheral, in particular noxious visceral, inputs to neurons in the hypothalamus that project to the midbrain periaqueductal gray. The induction of Fos protein was used to localize hypothalamic neurons that were activated by noxious visceral stimulation. This was combined with retrograde transport of fluorescent latex microspheres from identified "pressor" and "depressor" sites in the dorsolateral/lateral or ventrolateral columns of the periaqueductal gray. A second series of electrophysiological experiments examined the receptive field characteristics, including the incidence of viscerosomatic convergence, of neurons in the ventral part of the anterior hypothalamus. Noxious visceral stimulation (intraperitoneal acetic acid) induced Fos-like immunoreactivity in significantly more neurons in the hypothalamus than control stimuli (intraperitoneal saline and intravenous phenylephrine). Particularly high numbers of Fos-positive neurons were found in the paraventricular nucleus, the supraoptic nucleus and ventral regions of the anterior hypothalamus. When combined with retrograde tracing from "depressor" sites in the ventrolateral periaqueductal gray, the highest numbers of double-labelled neurons were localized in the paraventricular nucleus and the lateral area of the anterior hypothalamus. However, the regions that contained the greatest proportions of Fos-positive neurons that projected to "depressor" sites in the ventrolateral periaqueductal gray were the lateral area of the anterior hypothalamus and its rostral extension, the lateral preoptic area. Fewer double-labelled neurons were localized in the hypothalamus after retrograde transport from sites in the dorsolateral/lateral periaqueductal gray compared to the results obtained from injections of tracer in the ventrolateral periaqueductal gray. Furthermore, the numbers of Fos-positive hypothalamic neurons that projected to the dorsolateral/lateral periaqueductal gray were very similar in experimental and control animals. The electrophysiological study confirmed that a large proportion of neurons in and around the lateral area of the anterior hypothalamus can be driven by noxious visceral stimulation and demonstrated a high incidence of viscerosomatic convergence in these cells (66% of cells driven from somatic structures were also driven by electrical stimulation of the splanchnic nerve). Somatic receptive fields of these neurons were generally large, often including all four limbs and the face. The results of the functional anatomical and electrophysiological studies have identified neurons in an area of the ventral anterior hypothalamus that are a focus of nociceptive visceral input and which project to the midbrain periaqueductal gray, in particular to its ventrolateral column. These results are discussed in relation to the roles of the anterior hypothalamus and the different longitudinal columns of the periaqueductal gray in co-ordinating autonomic and sensory functions in response to visceral pain.