Cellular taxonomy and spatial organization of the murine ventral posterior hypothalamus.

The ventral posterior hypothalamus (VPH) is an anatomically complex brain region implicated in arousal, reproduction, energy balance, and memory processing. However, neuronal cell type diversity within the VPH is poorly understood, an impediment to deconstructing the roles of distinct VPH circuits in physiology and behavior. To address this question, we employed a droplet-based single-cell RNA sequencing (scRNA-seq) approach to systematically classify molecularly distinct cell populations in the mouse VPH. Analysis of >16,000 single cells revealed 20 neuronal and 18 non-neuronal cell populations, defined by suites of discriminatory markers. We validated differentially expressed genes in selected neuronal populations through fluorescence in situ hybridization (FISH). Focusing on the mammillary bodies (MB), we discovered transcriptionally-distinct clusters that exhibit neuroanatomical parcellation within MB subdivisions and topographic projections to the thalamus. This single-cell transcriptomic atlas of VPH cell types provides a resource for interrogating the circuit-level mechanisms underlying the diverse functions of VPH circuits.

Distribution of the neuronal inputs to the ventral premammillary nucleus of male and female rats.

The ventral premammillary nucleus (PMV) expresses dense collections of sex steroid receptors and receptors for metabolic cues, including leptin, insulin and ghrelin. The PMV responds to opposite sex odor stimulation and projects to areas involved in reproductive control, including direct innervation of gonadotropin releasing hormone neurons. Thus, the PMV is well positioned to integrate metabolic and reproductive cues, and control downstream targets that mediate reproductive function. In fact, lesions of PMV neurons blunt female reproductive function and maternal aggression. However, although the projections of PMV neurons have been well documented, little is known about the neuronal inputs received by PMV neurons. To fill this gap, we performed a systematic evaluation of the brain sites innervating the PMV neurons of male and female rats using the retrograde tracer subunit B of the cholera toxin (CTb). In general, we observed that males and females show a similar pattern of afferents. We also noticed that the PMV is preferentially innervated by neurons located in the forebrain, with very few projections coming from brainstem nuclei. The majority of inputs originated from the medial nucleus of the amygdala, the bed nucleus of the stria terminalis and the medial preoptic nucleus. A moderate to high density of afferents was also observed in the ventral subiculum, the arcuate nucleus and the ventrolateral subdivision of the ventromedial nucleus of the hypothalamus. Our findings strengthen the concept that the PMV is part of the vomeronasal system and integrates the brain circuitry controlling reproductive functions.

[Review of the target for the deep brain stimulation of chronic cluster headache]. / Revision de la diana para la estimulacion cerebral profunda de la cefalea en racimos cronica.

Cluster headache is included in the group of trigeminal autonomic cephalalgias. Although the pathophysiology of cluster headache has not yet been sufficiently established, the theory of a central origin tells us that this headache is produced by hypothalamic dysfunction. More than 50 patients have been treated with deep brain stimulation of the posterior nucleus of the hypothalamus from 2001. The results show clinical improvement in more than 60% of the cases, opening a promising issue for the treatment of the cluster headache persistent after medical treatment. The surgical target that have been used until now is based on the origin of the cluster headache in the hypothalamic dysfunction. Nevertheless, It has still some open questions as the lack of proving the posterior nucleus of the hypothalamus is the real origin of the cluster headache, the lack of consensus about the anatomy of the surgical target and the variability of the structures stimulated with the surgery. The aim of this article is a review of the target used and propose another surgical target based on physiopathological concepts to explain the improvement with the deep brain stimulation in these patients.

A direct neuronal connection between the subparafascicular and ventrolateral arcuate nuclei in non-lactating female rats. Could this pathway play a role in the suckling-induced prolactin release?

The neuronal pathways, through which prolactin secretion is regulated during lactation, have still not been fully explored. Studies indicate that the suckling stimulus travels through the spinal cord, the brain stem, and then reaches the hypothalamus. The focus of this present experiment is to further explore the neuronal connections between the brain stem and the arcuate nucleus that may be involved in suckling-induced prolactin release. Ante- and retrograde tracing techniques were used. To chemically characterize the explored neurons neuropeptide immunohistochemistry was applied. Previous studies have indicated that the peripeduncular nucleus is a relay of the suckling stimulus in the midbrain, conveying the information to the hypothalamus. In our experiments, we have found an additional cell group in the subparafascicular parvocellular nucleus located just behind the posterior thalamus that projects to the arcuate neurons. The injection of the retrograde tracer into the ventrolateral part of the arcuate nucleus labeled cells in the lateral subdivision of the subparafascicular parvocellular nucleus. Anterograde tracing from the subparafascicular parvocellular nucleus resulted in fiber labeling in the arcuate nucleus in close apposition with dynorphin immunopositive neurons. Double labeling revealed that a subpopulations of the subparafascicular parvocellular neurons projecting to the arcuate nucleus contained tuberoinfundibular peptide of 39 residues or calcitonin gene-related peptide. The presented findings suggest that the ascending fibers from the subparafascicular parvocellular nucleus might be in the pathway involved in the suckling-induced prolactin release.

Connectivity of an effective hypothalamic surgical target for cluster headache.

The purpose of this study was to look at the connectivity of the posterior inferior hypothalamus in a patient implanted with a deep brain stimulating electrode using probabilistic tractography in conjunction with postoperative MRI scans. In a patient with chronic cluster headache we implanted a deep brain stimulating electrode into the ipsilateral postero-medial hypothalamus to successfully control his pain. To explore the connectivity, we used the surgical target from the postoperative MRI scan as a seed for probabilistic tractography, which was then linked to diffusion weighted imaging data acquired in a group of healthy control subjects. We found highly consistent connections with the reticular nucleus and cerebellum. In some subjects, connections were also seen with the parietal cortices, and the inferior medial frontal gyrus. Our results illustrate important anatomical connections that may explain the functional changes associated with cluster headaches and elucidate possible mechanisms responsible for triggering attacks.

Posterior hypothalamic GABAergic mediation of hippocampal theta in the cat.

The effect of posterior hypothalamic (PH) microinjections of GABAA and GABAB agonists (muscimol and baclophen, respectively) and antagonists (bicuculline and 2-OH saclophen) on spontaneous, sensory and electrically induced hippocampal formation (HPC) theta EEG activity was investigated in freely behaving cats. Administration of GABAergic agonists abolished the theta rhythm recorded from HPC. This effect was reversible. A substantial difference in the recovery time course between frequency versus amplitude and power of hippocampal theta was observed. While theta frequency exhibited a rapid reappearance with a shallow slope, the power and amplitude showed a gradual recovery with a steeper slope. The PH injection of GABAergic antagonists produced HPC theta with increased power. These results demonstrate that both types of GABAergic receptors localized in PH are engaged in mechanisms responsible for generating hippocampal theta oscillations in freely behaving cats. The present study provides additional support for the essential difference between rats and cats in the programming of HPC theta amplitude and frequency. While the PH in rats is involved in programming the frequency of theta rhythm, the same region in cats mainly determines theta amplitude.

Update on neurosurgical treatment of chronic trigeminal autonomic cephalalgias and atypical facial pain with deep brain stimulation of posterior hypothalamus: results and comments.

The objective of this study is to describe the therapeutic effect and the technical and surgical problems of deep brain stimulation (DBS) of the posterior hypothalamus over seven years, for treatment of chronic trigeminal autonomic cephalalgias and atypical facial pain. We report a surgical series of 20 patients that underwent DBS of the posterior hypothalamus. This series includes 16 patients with chronic cluster headache (CH), one patient with short-lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT) and three patients with atypical facial pain. All patients of this series were resistant to any pharmacological and conservative treatment. The stimulated target was the same in the whole series even though stereotactic coordinates of the target referred to the midcommissural point differ slight in the Y anteroposterior value due to individual anatomical variability. The commissures based reference system was adapted to individual anatomical landmarks of the brainstem adding to the registration a third reference point below the commissural plane. The stimulation parameters of unipolar stimulation were similar in the whole series: 180 Hz, 60 mus, 1-3 V. In the CH series, at five years follow-up the percentage of total number of days free from pain attacks improved from 1%-2% to 71%. Ten patients of this series had a complete and persistent pain-free state at 18 months follow-up and the patient with SUNCT has complete pain relief. In the three patients with atypical facial pain, the neurostimulation procedure was absolutely unsuccessful. DBS of the posterior hypothalamus produced a significant and marked reduction of pain bouts in CH patients and in the SUNCT patient. The attempts to treat atypical facial pain in three patients failed.

Deep brain stimulation in craniofacial pain: seven years' experience.

Cluster headache (CH) is a primary headache with excruciatingly painful attacks that are strictly unilateral. About 10% of cases experience no significant remission, and about 15% of these do not respond to medication, so surgery is considered. Neuroimaging studies show that the posterior inferior hypothalamus is activated during CH attacks and is plausibly the CH generator. We report on 16 chronic CH patients, with headaches refractory to all medication, who received long-term hypothalamic stimulation following electrode implant to the posterior inferior hypothalamus. After a mean follow-up of 23 months, a persistent pain-free to almost pain-free state was achieved in 13/16 patients (15/18 implants; 83.3%) a mean of 42 days (range 1-86 days) after monopolar stimulation initiation. Ten patients (11 implants) are completely pain-free. A common side effect was transient diplopia, which limited stimulation amplitude. In one patient, a small non-symptomatic haemorrhage into the 3rd ventricle occurred following implant, but regressed 24 h later. Persistent side effects are absent except in one patient with bilateral stimulation, in whom stimulation was stopped to resolve vertigo and worsened bradycardia, but was resumed later without further problems. Hypothalamic stimulation is an effective, safe and well tolerated treatment for chronic drug-refractory CH. It appears as a valid alternative to destructive surgical modalities, and has the additional advantage of being reversible.

Conditional CRF receptor 1 knockout mice show altered neuronal activation pattern to mild anxiogenic challenge.

RATIONALE: Regional-specific corticotropin-releasing factor receptor 1 (CRF-R1) knockout mice have been generated recently as a tool to dissociate CNS functions modulated by this receptor. In these mice, CRF-R1 function is postnatally inactivated in the anterior forebrain including limbic brain structures but not in the pituitary leading to normal activity of the hypothalamic-pituitary-adrenocortical (HPA) axis under basal conditions and reduced anxiety-related behavior in the light-dark box and the elevated plus maze (EPM) as compared to wild-type (WT) mice (Müller et al., Nat Neurosci 6:1100-1107, 2003). OBJECTIVE: To identify neurobiological correlates underlying this reduced anxiety-like behavior, the expression of c-Fos, an established marker for neuronal activation, which was examined in response to a mild anxiogenic challenge. MATERIALS AND METHODS: Mice were placed for 10 min on the open arm (OA) of the EPM, and regional c-Fos expression was investigated by immunohistochemistry. RESULTS: OA exposure enhanced c-Fos expression in both conditional CRF-R1 knockout and WT mice in a number of brain areas (39 of 55 quantified), including cortical, limbic, thalamic, hypothalamic, and hindbrain regions. The c-Fos response in conditional CRF-R1 knockout animals was reduced in a restricted subset of activated neurons (4 out of 39 regions) located in the medial amygdala, ventral lateral septum, prelimbic cortex, and dorsomedial hypothalamus. CONCLUSIONS: These results underline the importance of limbic CRF-R1 in modulating anxiety-related behavior and suggest that reduced neuronal activation in the identified limbic and hypothalamic key structures of the anxiety circuitry may mediate or contribute to the anxiolytic-like phenotype observed in mice with region-specific deletion of forebrain CRF-R1.

Microinjection of carbachol in the supramammillary region suppresses CA1 pyramidal cell synaptic excitability.

Previous studies have established that the posterior hypothalamus-supramammillary (SUM) region is involved in the control of the hippocampal theta rhythm and also modulates the synaptic excitation of hippocampal neurons. Particularly, the medial but not lateral SUM region mediates reticular stimulation-induced suppression of CA1 pyramidal cell synaptic excitation to Schaffer collateral stimulation. In the present study using urethane anesthetized rats, we have investigated the effect of direct chemical stimulation of the posterior hypothalamus-SUM region on CA1 pyramidal cell excitability. It was observed that microinjection of the cholinergic muscarinic receptor agonist, carbachol (0.1 microl, concentration of either 0.0052, 0.156, or 0.625 microg/microl), evoked concentration-dependent suppression of CA1 pyramidal cell excitability that was dissociated from theta activation. Further, carbachol microinjection preferentially recruited the lateral SUM region when compared with the medial SUM and the posterior hypothalamic regions. In this context, the shortest latencies to suppression at the lowest concentration of carbachol and the strongest suppression at higher concentrations were observed with lateral microinjections. The carbachol-induced suppression was attenuated by inactivation of the medial septal region by microinjection of procaine (0.5 microl, 20% w/v). These results underscore a possible role for cholinergic mechanisms in the lateral SUM region in modulation of CA1 pyramidal cell synaptic excitation via the medial septal region. Furthermore, the present findings when juxtaposed with the medial SUM mediation of reticularly-elicited suppression suggest a medial-lateral topographic organization of the SUM region in modulation of CA1 excitability.

Dynamic changes in the direction of the theta rhythmic drive between supramammillary nucleus and the septohippocampal system.

Neurons in the supramammillary nucleus (SUM) of urethane-anesthetized rats fire rhythmically in synchrony with hippocampal theta rhythm. As these neurons project to the septum and hippocampus, it is generally assumed that their role is to mediate ascending activation, leading to the hippocampal theta rhythm. However, the connections between SUM and the septohippocampal system are reciprocal; there is strong evidence that theta remains in the hippocampus after SUM lesions and in the SUM after lesioning the medial septum. The present study examines the dynamics of coupling between rhythmic discharge in the SUM and hippocampal field potential oscillations, using the directionality information carried by the two signals. Using directed transfer function analysis, we demonstrate that during sensory-elicited theta rhythm and also during short episodes of theta acceleration of spontaneous oscillations, the spike train of a subpopulation of SUM neurons contains information predicting future variations in rhythmic field potentials in the hippocampus. In contrast, during slow spontaneous theta rhythm, it is the SUM spike signal that can be predicted from the preceding segment of the electrical signal recorded in the hippocampus. These findings indicate that, in the anesthetized rat, SUM neurons effectively drive theta oscillations in the hippocampus during epochs of sensory-elicited theta rhythm and short episodes of theta acceleration, whereas spontaneous slow theta in the SUM is controlled by descending input from the septohippocampal system. Thus, in certain states, rhythmically firing SUM neurons function to accelerate the septal theta oscillator, and in others, they are entrained by a superordinate oscillatory network.

The effect of reversible inactivation of the supramammillary nucleus on passive avoidance learning in rats.

Previous studies have shown that the presence of hippocampal theta activity is important for learning and memory, and that the medial supramammillary nucleus (mSuM) is involved in the control of the frequency of theta rhythm. It has also been shown that the depression of mSuM activity by chlordiazepoxide causes modest impairment of spatial learning. On the other hand, the lateral supramammillary nucleus (lSum) increases long-term potentiation (LTP) of hippocampal population spikes. However, to our knowledge, no reports exist concerning the role of the supramammillary area (SuM) in passive avoidance (PA) learning. In the present study, rats were chronically implanted with a cannula aimed at SuM and were trained on a step-through PA task. They received intra-SuM injection of lidocaine or saline at the following intervals: 5 min before training, 5, 90, and 360 min after the acquisition trial, or 5 min before the retrieval test. When lidocaine was injected 5 min before training there was no effect on acquisition of PA but retrieval was significantly poorer than the control group injected with saline. Lidocaine injection 5 min after the acquisition trial impaired PA retention, but reversible inactivation of SuM at 90 and 360 min after training and 5 min before the retrieval test showed no significant effect on PA retention. It can be concluded that SuM contributes to PA consolidation at least 5 min after the acquisition trial and that this effect may be accomplished through SuM projections to the septal and/or hippocampal areas participating in the PA memorization processes.

Hypothalamic inferior lobe and lateral torus connections in a percomorph teleost, the red cichlid (Hemichromis lifalili).

The neuroanatomic connections of the inferior lobe and the lateral torus of the percomorph Hemichromis lifalili were investigated by 1,1', dioctadecyl-3,3,3',3'-tetramethylindo-carbocyanine perchlorate (DiI) tracing. The inferior lobe and the lateral torus both receive afferents from the secondary gustatory nucleus. Additional afferents reach the inferior lobe from the nucleus glomerulosus, nucleus suprachiasmaticus, dorsal and central posterior thalamic nucleus, nucleus lateralis valvulae, magnocellular part of the magnocellular nucleus of the preoptic region, caudal nucleus of the preglomerular region, posterior tuberal nucleus, area dorsalis of the telencephalon, and a tegmental nucleus (T2). Efferents from the inferior lobe and the lateral torus terminate in the dorsal hypothalamic neuropil and corpus mamillare. Furthermore, the inferior lobe projects to the medial nucleus of the lateral tuberal hypothalamus and perhaps makes axo-axonal synapses in the tractus tectobulbaris rectus. The inferior lobe and the torus lateralis have reciprocal connections with the preglomerular tertiary gustatory nucleus and posterior thalamic nucleus and are also mutually interconnected. The inferior lobe is also reciprocally connected with the medial nucleus of the preglomerular region, reticular formation and sparsely with the anterior dorsal thalamic and the ventromedial thalamic nuclei. Thus, whereas the lateral torus is exclusively connected with the gustatory system, the inferior lobe is of a multisensory nature. In comparison with the goldfish (Carassius auratus), the connectivity pattern of the inferior lobe of Hemichromis lifalili reflects its specialization with respect to the visual system, as it receives qualitative (i.e., dorsal posterior, anterior, and ventromedial thalamic nuclei) as well as quantitative (i.e., nucleus glomerulosus) additional visual input.

The afferent connections of the posterior hypothalamic nucleus in the rat using horseradish peroxidase.

The posterior hypothalamic nucleus has been implicated as an area controlling autonomic activity. The afferent input to the nucleus will provide evidence as to its role in autonomic function. In the present study, we aimed to identify the detailed anatomical projections to the posterior hypothalamic nucleus from cortical, subcortical and brainstem structures, using the horseradish peroxidase (HRP) retrograde axonal transport technique in the rat. Subsequent to the injection of HRP into the posterior hypothalamic nucleus, extensive cell labelling was observed bilaterally in various areas of the cerebral cortex including the cingulate, frontal, parietal and insular cortices. At subcortical levels, labelled cells were observed in the medial and lateral septal nuclei, the bed nucleus of stria terminalis, and various thalamic and amygdaloid nuclei. Also axons of the vertical and horizontal limbs of the diagonal band were labelled and labelled cells were localised at the CA1 and CA3 fields of the hippocampus and the dentate gyrus. The brainstem projections were from the medial, lateral and parasolitary nuclei, the intercalated nucleus of the medulla, the sensory nuclei of the trigeminal nerve, and various reticular, vestibular, raphe and central grey nuclei. The posterior hypothalamic nucleus also received projections from the lateral and medial cerebellar nuclei and from upper cervical spinal levels. The results are discussed in relation to the involvement of the posterior hypothalamic nucleus in autonomic function and allows a better understanding of how the brain controls visceral function.

The posterior hypothalamic area: chemoarchitecture and afferent connections.

This study provides an analysis of the chemoarchitecture of the posterior hypothalamic area (PHA) and a retrograde transport analysis of inputs to the PHA in the rat. The chemoarchitectural analysis reveals that the majority of PHA neurons contain glutamate. Hypocretin, melanin concentrating hormone, tyrosine hydroxylase, neuropeptide Y and gamma-aminobutyric acid are also found in subsets of PHA neurons, and fibers immunoreactive for these substances as well as for serotonin, dopamine-beta-hydroxylase and met-enkephalin are observed in the area and aid in the delineation of its borders. The retrograde tracing study demonstrates that the PHA receives input from multiple, diverse neuron populations. Descending projections to the PHA arise from the limbic forebrain (cingulate cortex and lateral septum) and both the medial and lateral hypothalamus. Subcortical visual nuclei, including the ventral lateral geniculate nucleus and intergeniculate leaflet, pretectal area, and superior colliculus, and the subthalamus (zona incerta, fields of Forel) also project to the PHA. Ascending projections to the PHA arise from brainstem cholinergic nuclei, the reticular formation, midbrain raphe nuclei, periaqueductal gray and parabrachial nucleus. Retrograde transport studies using the psuedorabies virus (PRV) demonstrate that the PHA receives input indirectly from the hippocampus, amygdala and suprachiasmatic nucleus through circuits including nuclei in the limbic forebrain and hypothalamus. These data suggest that the PHA is important in the neural control of behavioral state, modulating aspects of hippocampal, autonomic and cortical function as they relate to the elaboration of adaptive behavior.

Descending projections of the posterior nucleus of the hypothalamus: Phaseolus vulgaris leucoagglutinin analysis in the rat.

No previous report in any species has systematically examined the descending projections of the posterior nucleus of the hypothalamus (PH). The present report describes the descending projections of the PH in the rat by using the anterograde anatomical tracer, Phaseolus vulgaris leucoagglutinin. PH fibers mainly descend to the brainstem through two routes: dorsally, within the central tegmental tract, and ventromedially, within the mammillo-tegmental tract and its caudal extension, ventral reticulo-tegmental tracts. PH fibers were found to distribute densely to several nuclei of the brainstem. They are (from rostral to caudal) 1) lateral/ ventrolateral regions of the diencephalo-mesopontine periaqueductal gray (PAG); 2) the peripeduncular nucleus; 3) discrete nuclei of pontomesencephalic central gray (dorsal raphe nucleus, laterodorsal tegmental nucleus, and Barrington's nucleus); 4) the longitudinal extent of the central core of the mesencephalic through meduallary reticular formation (RF); 5) the ventromedial medulla (nucleus gigantocellularis pars alpha, nucleus raphe magnus, and nucleus raphe pallidus); 6) the ventrolateral medulla (nucleus reticularis parvocellularis and the rostral ventrolateral medullary region); and 7) the inferior olivary nucleus. PH fibers originating from the caudal PH distribute much more heavily than those from the rostral PH to the lower brainstem. The PH has been linked to the control of several important functions, including respiration, cardiovascular activity, locomotion, antinociception, and arousal/wakefulness. It is likely that descending PH projections, particularly those to the PAG, the pontomesencephalic RF, Barrington's nucleus, and parts of the ventromedial and ventrolateral medulla, serve a role in a PH modulation of complex behaviors involving integration of respiratory, visceromotor, and somatomotor activity.

The spontaneous theta rhythm recorded from the hypothalamus posterior in the cat in vivo.

Rhythmical slow activity (theta) was mapped in the hypothalamic region in freely moving cats. We recorded well synchronized and high amplitude theta rhythm in the medial part of the hypothalamus posterior area. The EEG recordings made from lateral part of this hypothalamic region contained only irregular activity. These findings support earlier observations concerning the topography of hippocampal formation desynchrony and synchrony system. The observations of the present study also suggest that the hypothalamus posterior area is actively involved in the mechanisms responsible for generating theta oscillations in the cat.

Medullary projections to the vagus nerve and posterolateral hypothalamus.

BACKGROUND: Vagal visceromotor reflexes are dependent upon reciprocal neural connections existing between the medulla and the hypothalamus. Medullohypothalamic neurons may provide feedback cues to the hypothalamus regarding the activity of vagal motor neurons. As yet, however, studies investigating the spatial relationships between medullohypothalamic neurons and vagal motor neurons have not been performed. METHODS: A variety of retrogradely transported tracers were used for the purpose of mapping the relative locations of medullovagal and medullohypothalamic neurons. Tracers were injected into the cervical vagus nerve and/or the posterolateral hypothalamus, and subsequently the retrogradely labeled medullary neurons were plotted. RESULTS: Labeling of the two neuronal populations was primarily observed within the ventrolateral and dorsomedial medulla. Within the ventrolateral medulla, medullovagal neurons were found within the retrofacial nucleus and nucleus retroambiguus, whereas medullohypothalamic neurons were located subjacent to these nuclei. Within the dorsomedial medulla, labeling of the two neuronal populations was primarily limited to the vagal-solitary complex. At this location medullovagal neurons were found within the dorsal vagal nucleus, whereas medullohypothalamic neurons were largely confined to the caudal aspect of the solitary nucleus. CONCLUSIONS: Because of the spatial proximity existing between medullovagal and medullohypothalamic neurons, it is suggested that functional interrelationships may exist between these two neuronal populations. Specifically, it is suggested that the medullohypothalamic neurons identified in this study may support vagal-related functions by providing feedback cues to the posterolateral hypothalamus.

Injection of muscimol in the posterior hypothalamus reduces the PGE1-hyperthermia in the rat.

The firing rate of the nerves innervating interscapular brown adipose tissue (IBAT), IBAT and colonic temperatures (TIBAT and Tc), heart rate, and oxygen (O2) consumption were monitored in urethane-anesthetized male Sprague-Dawley rats. These variables were measured for 40 min before (baseline values) and 40 min after a 56 ng muscimol injection in the posterior hypothalamus and an intracerebroventricular administration of 500 ng prostaglandin E1 (PGE1). The same variables were monitored in other rats with muscimol injection or PGE1 administration alone. No drug was injected in control rats. The results show that muscimol injection reduces the increases in firing rate, TIBAT, Tc, heart rate, O2 consumption induced by PGE1. These findings suggest that GABAergic tone in the posterior hypothalamus is important in the control of thermogenic changes induced by PGE1.