Integrated lateral hypothalamic neural responses to natural and artificial rewards and cue signals in the rat

Effects of natural and intracranial electrical rewarding stimuli and cue signals were investigated while recording from single neurons in the rat lateral hypothalamus. The rat obtained both rewards using identical behavior, viz. licking. When both rewarding stimuli influenced a neuron, the responses were usually similar, i.e. both excitatory or both inhibitory. Only neurons that responded to either or both rewards acquired responses to tone cues, and these acquired responses were in the same direction as reward responses. The data indicate that the same single neuron in the lateral hypothalamus might be implicated in reward processes and learning.     

Energy balance and hypothalamic self-stimulation

The effects of negative energy balance on self-stimulation are a matter of considerable disagreement. This disagreement undoubtedly reflects the inadequacies of the continuous reinforcement self-stimulation procedures used in this type of experimentation. The present experiment uses a new fixed-interval reinforcement shuttle-box procedure which provides indices of reward and stimulation escape that are free from the numerous performance altering effects that confound continuous reinforcement performance.Whereas 24 h of food deprivation had no effect on stimulation initiation or escape rates, 48 h of food deprivation selectively increased initiation rates. The enhancement of reward was seen over virtually the entire anterior-posterior extent of the lateral hypothalamus and occurred irrespective of the occurrence of any stimulus-bound behaviors. Thus negative energy balance appears to selectively increase the excitability of reward-related neurons in the lateral hypothalamus.The self-stimulating rats became clearly hyperphagic, yet their weight gains were not significantly different from those of controls. The self-stimulation must, therefore, have greatly increased energy expenditure. Thus, not only does energy balance affect self-stimulation, but self-stimulation appears to affect energy balance.     

Two types of medial hypothalamic inhibition of lateral hypothalamic reward

Electrode-implanted rats pressed a lever to self-stimulate a reward mechanism in their lateral hypothalamus (LH) each time with a brief train of pulses. Interdigitation of a medial hypothalamus (MH) train into the ipsilateral LH train inhibited the reward as indicated by a reduced lever pressing rate. Such interdigitation into the contralateral LH train resulted in only a minimal inhibition accounted for by an across-midline current spread, thus indicating only an ipsilateral MH-to-LH route for inhibiting reward. By varying the phase of interdigitation according to a pulse-pair technique¹³ (cf. 20), the time course of this ipsilateral reward inhibition proved bimodal. The peripheral administration of strychnine, a glycine transmitter antagonist, resulted in disinhibition of the fast rising, fast decay (0.1–2.0 ms) intervals, while peripheral administration of picrotoxin, a GABA transmitter antagonist, increased bar pressing only at the slow rising, slow decay (2–20 ms) intervals. The placement of the MH electrodes was then varied dorsoventrally and the strength of the MH current was reduced to minimize dorsal-ventral current spread. From the resulting differential reductions in the two modes it was inferred that the neural elements in the MH responsible for short duration inhibitory effects on LH reward are concentrated more dorsally whereas those responsible for the longer duration inhibitory effects are concentrated more ventrally. These findings demonstrate the relevance of the pulse-pair technique for visualizing congruences between over behavior and underlying neurophysiological, neuroanatomical and molecular events.     

Intrinsic neurons are involved in lateral hypothalamic self-stimulation

The recent technique of using ibotenic acid to lesion selectively local neurons while sparing fibers of passage permitted us to answer a long-standing question: is lateral hypothalamic self-stimulation supported by fibers of passage or are the intrinsic hypothalamic neurons involved? Three groups of adult male Sprague-Dawley rats were used. In a normal group, electrodes were bilaterally implanted in the lateral hypothalamus and self-stimulation (ICSS) was obtained separately with the right and left electrodes, at various current intensities, using a nose-poke response. In the experimental group, the intrinsic neurons of the lateral hypothalamus were destroyed unilaterally by local injection of ibotenic acid (4 or 6 μg in 0.5 μl); the other side served as the sham-lesion control. Ten days later ICSS electrodes were implanted bilaterally, one in the lesioned area, the other in the contralateral hypothalamus. As in the case of the normal animals, the rate of nose-poking (ICSS) was then determined separately for each electrode. In the normal rats, ICSS rates were the same with stimulation on either side and the increase in ICSS rate as a function of the increase in current intensity was the same on each side. In the experimental rats, ICSS of the lesioned side was decreased in all cases; moreover, after lesion with the 6 μg dose, ICSS was totally suppressed. Self-stimulation of the sham-lesioned side was not significantly different from that observed in the normal rats. In 6 rats sampled from the lesioned groups as well as in 3 additional unimplanted animals, biochemical assays compared dopamine and serotonin contents of the two striata and noradrenaline and serotonin contents of the two hippocampi. No difference was observed for these two structures between the side ipsilateral to the lesion and the contralateral side. Moreover, none of these monoamine levels differed from those seen in the unimplanted rats. These results, taken together, suggest that intrinsic lateral hypothalamic neurons are involved in ICSS.     

Brain stimulation reward in the lateral hypothalamic medial forebrain bundle: Mapping of boundaries and homogeneity

The boundaries and relative fiber concentration of the brain stimulation reward (BSR) sustaining system coursing through the lateral hypothalamic medial forebrain bundle (MFB) were mapped using a dorso-ventral moveable electrode. High response rates for BSR were found in a region extending dorso-ventrally from the zona incerta (ZI) to the base of the brain and medio-laterally from the fornix to the medial tip of the internal capsule (IC). Self-stimulation associated with perifornical area and self-stimulation associated with the tip of the internal capsule were mixed with aversion and forced movements, respectively. Current intensity threshold variations suggest: (i) that the reward system has a well-defined dorsal boundary ventral to the ZI, and (ii) that the core of the MFB contains a relatively higher concentration of reward relevant fibers than do its lateral, medial, dorsal and ventral components. No evidence was seen of independent mid-lateral and far-lateral MFB systems, though independent BSR sites in the dorsomedial and ventromedial hypothalamus were seen.     

Evidence for a role of the preoptic area in lateral hypothalamic self-stimulation

We examined the effects of unilateral radiofrequency lesions in the preoptic area on lateral hypothalamic self-stimulation in 15 rats. The animals were tested for self-stimulation in the lateral hypothalamus at 3 different current intensities from electrodes placed in both hemispheres, and then received a unilateral lesion in the preoptic area. Four hours later they were again tested for self-stimulation at the 3 current intensities and then daily over the following 14 days, or until they recovered their presurgical rates of self-stimulation. Rate of self-stimulation decreased in the damaged hemisphere, and recovered to prelesion levels within 2 weeks in 6 of 9 rats. In the intact hemisphere rate of self-stimulation increased above the prelesion level during a period from 1 to 2 weeks after the lesion. These results suggest that the preoptic area is involved in lateral hypothalamic self-stimulation. The effects of d-amphetamine (1 mg/kg) and apomorphine (2 mg/kg) injections on turning behavior were also studied in an open field and in a rotometer. Apomorphine induced contraversive turning to the lesion side in the open field and d-amphetamine induced ipsiversive turning in the rotometer.     

Effects of rewarding hypothalamic stimulation on plasma catecholamine levels in rats

Plasma catecholamines were measured before, during and after exposure to lick-contingent rewarding hypothalamic stimulation, clock-triggered neutral hypothalmaic stimulation, and licking maintained by water. Rewarding hypothalamic stimulation elicited a marked rise in plasma epinephrine levels which returned to baseline levels 3 min after the self-stimulation session. No changes in plasma epinephrine were observed under the latter two conditions. Increases in norepinephrine levels were more variable and seemed to correspond to motor activity being similar in the licking groups but unchanged with clock-triggered stimulation. These results indicate that stimulation of reward pathways may selectively influence adreno-medullary secretion.     

Role of the hypothalamic paraventricular nucleus in neuroendocrine responses to daylength in the golden hamster

Solitary horizontal cells dissociated from goldfish retinas depolarized when exposed to micromolar doses of either L-glutamate or kainic acid. The responses to both of these agonists were antagonized by D-aspartate, and unaffected by L-aspartate, L-glutamic acid diethyl ester and folic acid. the results of the present study thus suggest that L-glutamate and kainic acid may produce depolarizations of horizontal cells by interacting with pharmacologically similar membrane receptors.     

An HRP study of the afferent connections to rat medial hypothalamic region

Afferent connections to the medial hypothalamic region in the rat were studied using horseradish peroxidase (HRP). HRP was injected iontophoretically by a parapharyngeal approach. After HRP injections into the ventromedial hypothalamic nucleus, labeled cells were found mainly in the medial and basolateral amygdaloid nuclei, subiculum, peripeduncular nucleus and the parabrachial area. Labeled cells following HRP injections into the dorsomedial hypothalamic nucleus were found mainly in the lateral septal nucleus, nucleus accumbens, bed nucleus of the stria terminalis, pontine central gray and the parabrachial area. HRP-labeled cells following the medial preoptic area injections were found mainly in the infralimbic cortex, lateral and medial septal nuclei, nucleus accumbens, diagonal band, bed nucleus of the stria terminalis, medial amygdaloid nucleus, subiculum, peripedunclar nucleus and the parabrachial area. The intrahypothalamic connections were also discussed.     

Diffusional barrier around the hypothalamic arcuate nucleus in the rat

The contour lines of horseradish peroxidase injection sites in the ventrobasal hypothalamus were distorted by the border between arcuate and ventromedial nuclei as well as between arcuate nucleus and median eminence. The dense array of tanycyte processes is assumed to isolate the arcuate nucleus from the neighboring territories by establishing a diffusional barrier surface.     

Self-stimulation responses are altered following long-term but not short-term treatment with clorgyline

Clorgyline (a selective monoamine oxidase-inhibiting antidepressant) given chronically facilitated hypothalamic self-stimulation in rats, while acute treatment was without effect. Furthermore, long-term but not short-term clorgyline treatment significantly attenuated the suppressive effect of the selective α₂-adrenergic agonist clonidine on this behavior. These findings suggest that adaptive alterations in the modulation of rewarded behavior by inhibitory presynaptic noradrenergic receptors may be involved in antidepressant efficacy.     

Medullary unit responses to changes in local and hypothalamic temperatures in the cat

Single unit activity was recorded with glass microelectrodes in the reticular formation of the medulla oblongata of cats lightly anesthetized with urethan, while medullary temperature was changed by surface irrigation with warm or cold artificial cerebrospinal fluid. In addition, water-perfusion thermodes were implanted over the preoptic anterior hypothalamus (POAH) region, and the effects of heating and cooling of the POAH on firing rate of medullary units were studied. One hundred and twenty-five temperature-sensitive neurons were studied in the medullary reticular formation. Out of these 125 neurons, 80 were warm-sensitive and 45 were cold-sensitive. Sixteen warm-sensitive medullary neurons were examined in responses to changes of POAH temperature. Ten units (62.5%) responded, and the remaining 6 units were not responsive to changes of the POAH temperature. Of 13 cold-sensitive neurons examined, 10 units (77.0%) responded. On the other hand, only 1 out of 7 (14.3%) temperature-insensitive neurons tested did respond to changes in the POAH temperature. These results suggest that the temperature signals sent out from thermosensitive structures in the hypothalamus might be transmitted to a major portion of temperature-sensitive neurons in the medullary reticular formation.     

Autoradiography of GABA in the rat hypothalamic median eminence

Light microscopy autoradiographs of the rat hypothalamic median eminence were prepared after injection of high specific activity tritiated GABA and GABA structural analogs.Following intracardiac injection of labeled GABA with short (15 min) survival time, a dense accumulation of silver grains was observed over the external layer of the median eminence. The silver grains appeared much less numerous and randomly scattered over the internal layer. No conspicuously labeled cells could be detected in the median eminence.A similar pattern of labeling was observed after 10 min in vitro incubation of the median eminence with a low concentration(2.5 × 10⁻⁷M) of labeled GABA. Clusters of silver grains were also visible over the external layer following intraventicular injection of labeled GABA. In this latter case, however, other sites of labeling were revealed over the internal and ependymal layers.The dense labeling over the external layer with tritiated GABA was partially reduced by a simultaneous intracardiac injection of a 50-fold excess of non-radioactive cis-aminocyclohexane car☐ylic acid — a reported preferential substrate for GABA neuronal uptake — but it was not displaced by a 2000-fold excess of non-radioactive β-alanine — a reported specific substrate for GABA glial uptake.Intracardiac injection of triated β-alanine led to a faint and even labeling over the entire median eminence with no preferential accumulation of silver grains over the various layers. Following intraventricular injection of labeled β-alanine the tanycytes and their processes as well as numerous glial cells appeared heavily labeled.These results suggested that there exist cell elements in the external layer of the hypothalamic median eminence which are capable of accumulating exogenous GABA according to its neuronal uptake characteristics. Although the exact nature of these cells is not readily apparent at this stage of our investigations, these findings led us to speculate that there might be a subpopulation of GABAergic nerve endings in the vicinity of the primary plexus capillaries.     

Increased hypothalamic norepinephrine metabolism after water deprivation in the rat

Rat brain catecholamine metabolism was changed over a period of several days by limited access to water (10 min/day). One or two weeks limited access to water caused an increase in hypothalamic norepinephrine metabolism as measured with alpha-methyl-para-tyrosine. Brain stem and telencephalon norepinephrine was not affected by the limited access to water regimen. Dopamine metabolism in the corpus striatum and the hypothalamus was not altered by limited access to water. If the limited access to water was continued for 3 or more weeks, hypothalamic norepinephrine metabolism then returned to normal. The increase in hypothalamic norepinephrine metabolism was confirmed by a second method measuring in vivo tyrosine hydroxylase activity. Additional experiments demonstrate that this affect is specific for water deficits. Limited access to food had no effect on the metabolism of norepinephrine in the hypothalamus. Water deficits produced by replacing water with a 2% NaCl solution caused a similar increase in hypothalamic norepinephrine metabolism to that observed after one week limited access to water. Furthermore, 10 min access to water stopped the increased hypothalamic metabolism of norepinephrine seen after one week of limited access to water. The regional specificity (effect seen in hypothalamus but not the telencephalon and brain stem), and the stimulus specificity (water and not food deficits) suggest hypothalamic norepinephrine involvement in thirst or hormonal control of water regulation.     

Water deprivation increases anterior hypothalamic norepinephrine metabolism in the rat

Rats were limited to 10 min of access to water per day. After 1 week, concentrations and rate of metabolism of dopamine, norepinephrine and epinephrine were determined in hypothalamic and limbic areas associated with regulation of water homeostasis. Chronic water deprivation caused hypovolemia, hypotension and ingestion of a large volume of water when water became available. Norepinephrine metabolism was consistently increased in samples containing the anterior hypothalamic nucleus, but no other catecholamine in any other brain area was significantly affected by the deprivation schedule. We conclude that the anterior hypothalamic nucleus is involved in the response to chronic disruption of water balance in the rat.     

Lateral hypothalamic stimulation gates nucleus gigantocellularis-induced aversion via a reward-independent process

A monophasic pulse-pair stimulation technique was used behaviorally to infer neurophysiological interaction between ‘reward’ sites in the lateral hypothalamus (LH) and pain-implicated nucleus reticularis gigantocellularis (NGC). Rats lever-pressed for 3 s escapes from an otherwise continuous train of 0.1 ms NGC pulses delivered every 40 ms. When pulses to the LH were interdigitated with those to NGC, inter-response latencies were significantly longer. This occurred whether LH pulses were also interrupted by each lever-press (experiment 1) or not (experiment 2). Further, the same intensities of LH stimulation which inhibited escape supported high rates of lever-pressingfor 3 s LH trains during inescapable NGC stimulation even though these LH trains were not of sufficient intensity to support ‘rewarding’ lever-pressing in the absence of NGC stimulation. This finding indicates that escape-inhibition in the first two experiments was not due to interfering motor effects of LH stimulation but most probably due to LH amelioration of NGC-induced aversion. Indeed, when aversion was ameliorated by morphine, lever-pressing for sub-reward-threshold LH stimulation during inescapable NGC stimulation decreased although classical self-stimulation did not.In the escape experiments, inhibition was greatest when each NGC pulse was preceded by an LH pulse at an interval of 0.1 or 10 ms. During inescapable NGC stimulation, rates of lever-pressing for LH trains were greatest when LH pulses preceded NGC pulses again at intervals of 0.1 or 10 ms. This congruence of the temporal bimodal profiles in the response functions of these two different behavioral experiments strongly suggests that the same integrative mechanism underlies both LH inhibition of NGC escape and lever-pressing for LH trains during inescapable stimulation of NGC. Since gastric loading did not alter lever-pressing for sub-reward-threshold LH trains during inescapable NGC stimulation but did inhibit classical self-stimulation, it was concluded that the inferred common integrative mechanism mediates a supraspinal gating of aversion which is independent of the LH self-stimulation reward process.     

Hypothalamic unit responses to alimentary perfusions in the anesthetised rat

Single unit discharges in the ventromedial hypothalamic nucleus (VMH) and lateral hypothalamic area (LH) were extracellularly recorded in urethane anesthetised female rats, while various solutions were perfused through the stomach or duodenum via implanted polythene tubes. Perfusates, which were maintained at 38°C, were 0.9% (w/v) NaCl, 2.5% NaCl, 5.25% glucose, 30% glucose, 5.0% casein hydrolysate, and liquid food. Only units which did not respond to somatosensory stimuli were tested. One hundred and twenty six units were recorded for periods of up to 3 hr, occasionally longer, in 57 animals. Of these, 74 were recorded during gastric perfusions and 52 during duodenal perfusions. Distension of the stomach elicited changes in firing rate in 16 LH and 8 VMH units. Both increases and decreases in firing rate in response to gastric distension were observed in both the LH and VMH. There was no evidence that nutrient or osmotic properties of the perfusates exerted any modifying influence on hypothalamic unit discharges. Distention of the duodenum by the perfusions elicited changes in firing rate in 12 LH and 6 VMH units. In this case all LH units decreased firing and all VMH units increased firing during distension of the lumen. In addition 2 VMH units appeared to increase their firing rate in association with glucose perfusions. The results are discussed in terms of the role of gastrointestinal feedback to the hypothalamus in food intake regulation.     

Neurotensin release by rat hypothalamic fragments in vitro

The release of neurotensin by hypothalami from male rats was investigated in vitro uding tissue fragments incubated in Krebs-Ringer bicarbonate-glucose buffer at 37°C. Neurotensin was measured by radioimmunoassay using an antibody directed toward the C-terminal portion of the peptide. Neurotensin-like immunoreactivity release into the incubation medium eluted from Sephadex G-25 in a position identical to that of the synthetic peptide and serial dilutions of incubation medium were parallel to those of synthetic neurotensin in the radioimmunoassay. Neurotensin-like immunoreactivity released into the incubation medium was degraded during the incubation period by released hypothalamic peptidases. The addition of bacitracin (0.5 mg/ml) to the medium partially prevented this degradation. Neurotensin release was stimulated by dibutyryl cyclic AMP (10⁻⁴ M) and by depolarizing concentrations of potassium. The latter effect was shown to be Ca²⁺-dependent. Dopamine (10⁻⁴−10⁻⁶ M) stimulated neurotensin release in a dose-dependent manner and this effect was blocked by the dopamine receptor antagonist, haloperidol. Neurotensin release was not stimulated by either norepinephrine (10⁻⁴ M) or serotonin (10⁻⁴ M). The results indicate that neurotensin is released by the hypothalamus in vitro; its release is stimulated by membrane depolarization in a Ca²⁺-dependent manner and may involve an adenylate cyclase mechanism; and dopamine appears to serve as a stimulatory neurotransmitter for neurotensin-containing neurons.     

Organization of cortical, basal forebrain, and hypothalamic afferents to the parabrachial nucleus in the rat.

In a previous study (Herbert et al., J. Comp. Neurol. [1990];293:540-580), we demonstrated that the ascending afferent projections from the medulla to the parabrachial nucleus (PB) mark out functionally specific terminal domains within the PB. In this study, we examine the organization of the forebrain afferents to the PB. The PB was found to receive afferents from the infralimbic, the lateral prefrontal, and the insular cortical areas; the dorsomedial, the ventromedial, the median preoptic, and the paraventricular hypothalamic nuclei; the dorsal, the retrochiasmatic, and the lateral hypothalamic areas; the central nucleus of the amygdala; the substantia innominata; and the bed nucleus of the stria terminalis. In general, forebrain areas tend to innervate the same PB subnuclei from which they receive their input. Three major patterns of afferent termination were noted in the PB; these corresponded to the three primary sources of forebrain input to the PB: the cerebral cortex, the hypothalamus, and the basal forebrain. Hypothalamic afferents innervate predominantly rostral portions of the PB, particularly the central lateral and dorsal lateral subnuclei. The basal forebrain projection to the PB ends densely in the external lateral and waist subnuclei. Cortical afferents terminate most heavily in the caudal half of the PB, particularly in the ventral lateral and medial subnuclei. In addition, considerable topography organization was found within the individual projections. For example, tuberal lateral hypothalamic neurons project heavily to the central lateral subnucleus and lightly to the waist area; in contrast, caudal lateral hypothalamic neurons send a moderately heavy projection to both the central lateral and waist subnuclei. Our results show that the forebrain afferents of the PB are topographically organized. These topographical differences may provide a substrate for the diversity of visceral functions associated with the PB.