Impact de l'haploinsuffisance du gène Sim1 sur le développement et la fonction du noyau paraventriculaire de l'hypothalamus

L’obésité provient d’un déséquilibre de l’homéostasie énergétique, c’est-à-dire une augmentation des apports caloriques et/ou une diminution des dépenses énergétiques. Plusieurs données, autant anatomiques que physiologiques, démontrent que l’hypothalamus est un régulateur critique de l’appétit et des dépenses énergétiques. En particulier, le noyau paraventriculaire (noyau PV) de l’hypothalamus intègre plusieurs signaux provenant du système nerveux central (SNC) et/ou de la périphérie, afin de contrôler l’homéostasie énergétique via des projections axonales sur les neurones pré-ganglionnaires du système autonome situé dans le troc cérébral et la moelle épinière. Plusieurs facteurs de transcription, impliqués dans le développement du noyau PV, ont été identifiés. Le facteur de transcription SIM1, qui est produit par virtuellement tous les neurones du noyau PV, est requis pour le développement du noyau PV. En effet, lors d’une étude antérieure, nous avons montré que le noyau PV ne se développe pas chez les souris homozygotes pour un allèle nul de Sim1. Ces souris meurent à la naissance, probablement à cause des anomalies du noyau PV. Par contre, les souris hétérozygotes survivent, mais développent une obésité précoce. De façon intéressante, le noyau PV des souris Sim1+/- est hypodéveloppé, contenant 24% moins de cellules. Ces données suggèrent fortement que ces anomalies du développement pourraient perturber le fonctionnement du noyau PV et contribuer au développement du phénotype d’obésité. Dans ce contexte, nous avons entrepris des travaux expérimentaux ayant pour but d’étudier l’impact de l’haploinsuffisance de Sim1 sur : 1) le développement du noyau PV et de ses projections neuronales efférentes; 2) l’homéostasie énergétique; et 3) les voies neuronales physiologiques contrôlant l’homéostasie énergétique chez les souris Sim1+/-. A cette fin, nous avons utilisé : 1) des injections stéréotaxiques combinées à des techniques d’immunohistochimie afin de déterminer l’impact de l’haploinsuffisance de Sim1 sur le développement du noyau PV et de ses projections neuronales efférentes; 2) le paradigme des apports caloriques pairés, afin de déterminer l’impact de l’haploinsuffisance de Sim1 sur l’homéostasie énergétique; et 3) une approche pharmacologique, c’est-à-dire l’administration intra- cérébroventriculaire (i.c.v.) et/ou intra-péritonéale (i.p.) de peptides anorexigènes, la mélanotane II (MTII), la leptine et la cholécystokinine (CCK), afin de déterminer l’impact de l’haploinsuffisance de Sim1 sur les voies neuronales contrôlant l’homéostasie énergétique. Dans un premier temps, nous avons constaté une diminution de 61% et de 65% de l’expression de l’ARN messager (ARNm) de l’ocytocine (Ot) et de l’arginine-vasopressine (Vp), respectivement, chez les embryons Sim1+/- de 18.5 jours (E18.5). De plus, le nombre de cellules produisant l’OT et la VP est apparu diminué de 84% et 41%, respectivement, chez les souris Sim1+/- adultes. L’analyse du marquage axonal rétrograde des efférences du noyau PV vers le tronc cérébral, en particulier ses projections sur le noyau tractus solitaire (NTS) aussi que le noyau dorsal moteur du nerf vague (X) (DMV), a permis de démontrer une diminution de 74% de ces efférences. Cependant, la composition moléculaire de ces projections neuronales reste inconnue. Nos résultats indiquent que l’haploinsuffisance de Sim1 : i) perturbe spécifiquement le développement des cellules produisant l’OT et la VP; et ii) abolit le développement d’une portion importante des projections du noyau PV sur le tronc cérébral, et notamment ses projections sur le NTS et le DMV. Ces observations soulèvent donc la possibilité que ces anomalies du développement du noyau PV contribuent au phénotype d’hyperphagie des souris Sim1+/-. En second lieu, nous avons observé que la croissance pondérale des souris Sim1+/- et des souris Sim1+/+ n’était pas significativement différente lorsque la quantité de calories présentée aux souris Sim1+/- était la même que celle consommée par les souris Sim1+/+. De plus, l’analyse qualitative et quantitative des tissus adipeux blancs et des tissus adipeux bruns n’a démontré aucune différence significative en ce qui a trait à la taille et à la masse de ces tissus chez les deux groupes. Finalement, au terme de ces expériences, les souris Sim1+/--pairées n’étaient pas différentes des souris Sim1+/+ en ce qui a trait à leur insulinémie et leur contenu en triglycérides du foie et des masses adipeuses, alors que tous ces paramètres étaient augmentés chez les souris Sim1+/- nourries ad libitum. Ces résultats laissent croire que l’hyperphagie, et non une diminution des dépenses énergétiques, est la cause principale de l’obésité des souris Sim1+/-. Par conséquent, ces résultats suggèrent que : i) l’haploinsuffisance de Sim1 est associée à une augmentation de l’apport calorique sans toutefois moduler les dépenses énergétiques; ii) l’existence d’au moins deux voies neuronales issues du noyau PV : l’une qui régule la prise alimentaire et l’autre la thermogénèse; et iii) l’haploinsuffisance de Sim1 affecte spécifiquement la voie neuronale qui régule la prise alimentaire. En dernier lieu, nous avons montré que l’injection de MTII, de leptine ainsi que de CCK induit une diminution significative de la consommation calorique des souris des deux génotypes, Sim1+/+ et Sim1+/-. De fait, la consommation calorique cumulative des souris Sim1+/- et Sim1+/+ est diminuée de 37% et de 51%, respectivement, durant les 4 heures suivant l’administration i.p. de MTII comparativement à l’administration d’une solution saline. Lors de l’administration i.c.v. de la leptine, la consommation calorique cumulative des souris Sim1+/- et Sim1+/+ est diminuée de 47% et de 32%, respectivement. Finalement, l’injection i.p. de CCK diminue la consommation calorique des souris Sim1+/- et Sim1+/+ de 52% et de 36%, respectivement. L’ensemble des résultats suggère ici que l’haploinsuffisance de Sim1 diminue l’activité de certaines voies neuronales régulant l’homéostasie énergétique, et particulièrement de celles qui contrôlent la prise alimentaire. En résumé, ces travaux ont montré que l’haploinsuffisance de Sim1 affecte plusieurs processus du développement au sein du noyau PV. Ces anomalies du développement peuvent conduire à des dysfonctions de certains processus physiologiques distincts régulés par le noyau PV, et notamment de la prise alimentaire, et contribuer ainsi au phénotype d’obésité. Les souris hétérozygotes pour le gène Sim1 représentent donc un modèle animal unique, où l’hyperphagie, et non les dépenses énergétiques, est la principale cause de l’obésité. En conséquence, ces souris pourraient représenter un modèle expérimental intéressant pour l’étude des mécanismes cellulaires et moléculaires en contrôle de la prise alimentaire.

mRNA expression of corticotropin-releasing factor and urocortin 1 after restraint and foot shock together with alprazolam administration

Corticotropin-releasing factor (CRF) is expressed in the paraventricular nucleus of the hypothalamus (PVN), and act centrally to provoke stress-like autonomic and behavioral responses. Urocortins 1-3 are additional ligands to the CRF receptors 1 and 2. Ucn 1 neurons are primarily concentrated in the Edinger-Westphal (EW) nucleus and also have been associated with stress responses. It is also known that UCN 1 respond in different ways depending on the stressor presented. Benzodiazepines can act via the CRF peptidergic system and chronic administration of alprazolam does not interfere with CRF mRNA expression in the PVN, but significantly increase Ucn 1 mRNA expression in the EW. The aim of our study was to investigate the relationship between different stressor stimuli, foot shock (FS) and restraint (R), and the mRNA expression of CRF and Ucn 1 in the PVN and EW using alprazolam (A). We employed fos activation and in situ hybridization. Restraint group presented increased fos-ir and CRF mRNA expression in the PVN compared to FS group. The stress responses of R group were prevented by A. In the EW,fos-ir was higher in the FS group than in the R group, whereas Ucn 1 mRNA expression was higher in the R group than in the FS group. Alprazolam significantly increased fos-ir and Ucn 1 mRNA expression in both groups. Our results show that PVN and EW respond in different ways to the same stressors. Furthermore, EW of stressed animals replies in a complementary way comparing to PVN with the use of Alprazolam. (C) 2010 Elsevier Inc. All rights reserved.

Chemical identity and connections of medial preoptic area neurons expressing melanin-concentrating hormone during lactation

Lactation is an energy-demanding process characterized by massive food and water consumption, cessation of the reproductive cycle and induction of maternal behavior. During lactation, melanin-concentrating hormone (MCH) mRNA and peptide expression are increased in the medial preoptic area (MPO) and in the anterior paraventricular nucleus of the hypothalamus. Here we show that MCH neurons in the MPO coexpress the GABA synthesizing enzyme GAD-67 mRNA. We also show that MCH neurons in the MPO of female rats are innervated by neuropeptides that control energy homeostasis including agouti-related protein (AgRP), alpha-melanocyte stimulating hormone (alpha MSH) and cocaine- and amphetamine-regulated transcript (CART). Most of these inputs originate from the arcuate nucleus neurons. Additionally, using injections of retrograde tracers we found that CART neurons in the ventral premammillary nucleus also innervate the MPO. We then assessed the projections of the female MPO using injections of anterograde tracers. The MPO densely innervates hypothalamic nuclei related to reproductive control including the anteroventral periventricular nucleus, the ventrolateral subdivision of the ventromedial nucleus (VMHvl) and the ventral premammillary nucleus (PMV). We found that the density of MCH-ir fibers is increased in the VMHvl and PMV during lactation. Our findings suggest that the expression of MCH in the MPO may be induced by changing levels of neuropeptides involved in metabolic control. These MCH/GABA neurons may, in turn, participate in the suppression of cyclic reproductive function and/or sexual behavior during lactation through projections to reproductive control sites. (C) 2009 Elsevier B.V. All rights reserved.

Interaction between paraventricular nucleus and medial septal area on the renal effects induced by adrenaline

The aim of the present study was to analyze the role of alpha(1),alpha(2)-adrenoceptors, and the effects of losartan and PD123319 (selective ligands of the AT(1) and AT(2) angiotensin receptors, respectively) injected into the paraventricular nucleus (PVN) on the diuresis, natriuresis, and kaliuresis induced by administration of adrenaline into the medial septal area (MSA). Male Holtzman rats with a stainless steel cannula implanted into the MSA and bilaterally into the PVN were used. The administration of adrenaline into the MSA increased in a dose-dependent manner the urine, sodium, and potassium excretions. The previous administration of prazosin (an alpha(1)-adrenoceptor antagonist) injected into the PVN abolished the above effects of adrenaline, whereas yohimbine (an a-adrenoceptor antagonist) doesn't affect the diuresis, natriuresis, and kaliuresis induced by adrenaline. Pretreatment with losartan into the PVN decreased in a dose-dependent manner the urine, sodium, and potassium excretions induced by MSA administration of adrenaline (50 ng), while PVN PD123319 was without effect. These results indicate that urinary and electrolyte excretion effects induced by adrenaline into the MSA are mediated primarily by PVN AT, receptors. However, the doses of losartan were more effective when combined with the doses of PD123319 than given alone, suggesting that the urinary, natriuretic, and kaliuretic effects of MSA adrenaline may involve activation of multiple angiotensin II receptors subtypes into the PVN. (C) 2004 Elsevier B.V All rights reserved.

Endogenous hydrogen peroxide in the hypothalamic paraventricular nucleus regulates sympathetic nerve activity responses to L-glutamate

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Urocortin in the central nervous system of a primate (Cebus apella): Sequencing, immunohistochemical, and hybridization histochemical characterization

The urocortin (UCN)-like immunoreactivity and UCN mRNA distribution in various regions of the nonprimate mammalian brain have been reported. However, the Edinger-Westphal nucleus (EW) appears to be the only brain site where UCN expression is conserved across species. Although UCN peptides are present throughout vertebrate phylogeny, the functional roles of both UCN and EW remain poorly understood. Therefore, a study focused on UCN system organization in the primate brain is warranted. By using immunohistochemistry (single and double labeling) and in situ hybridization, we have characterized the organization of UCN-expressing cells and fibers in the central nervous system and pituitary of the capuchin monkey (Cebus apella). In addition, the sequence of the prepro-UCN was determined to establish the level of structural conservation relative to the human sequence. To understand the relationship of acetylcholine cells in the EW, a colocalization study comparing choline acetyltransferase (ChAT) and UCN was also performed. The cloned monkey prepro-UCN is 95% identical to the human preprohormone across the matched sequences. By using an antiserum raised against rat UCN and a probe generated from human cDNA, we found that the EW is the dominant site for UCN expression, although UCN mRNA is also expressed in spinal cord lamina IX. Labeled axons and terminals were distributed diffusely throughout many brain regions and along the length of the spinal cord. of particular interest were UCN-immunoreactive inputs to the medial preoptic area, the paraventricular nucleus of the hypothalamus, the oral part of the spinal trigeminal nucleus, the flocculus of the cerebellum, and the spinal cord laminae VII and X. We found no UCN hybridization signal in the pituitary. In addition, we observed no colocalization between ChAT and UCN in EW neurons. Our results support the hypothesis that the UCN system might participate in the control of autonomic, endocrine, and sensorimotor functions in primates.

Haemodynamic effects of hypothalamic disconnection in anaesthetized rats

The aim of the present study was to analyse the haemodynamic effects induced by the hypothalamic disconnection (HD) caudal or rostral to the paraventricular nucleus of the hypothalamus (PVN). Mean arterial pressure (MAP), hindlimb, renal and mesenteric blood flow and vascular conductance (HVC, RVC and MVC, respectively) were measured in urethane (1.2 g/kg, i.v.) anesthetized rats for 60 min after disconnection. HD caudal to the PVN was performed with a double-edged microknife of bayonet shape (R=1 mm, H=2 mm) stereotaxically placed, lowered 2.8 mm caudal to the bregma along the midline. The cut was achieved by rotating the microknife 90° right and 90° left. HD rostral to the PVN was performed with the knife placed 0.8 mm caudal to the bregma. Thirty minutes after the hypothalamic disconnection caudal (HD-C), a decrease in MAP was observed (-14±3 mm Hg), reaching a 60-min decrease of 30±3 mm Hg. Hindlimb conductance increased 10 min after HD (156±14%) and remained elevated throughout the experimental period. On the contrary, we observed a transitory renal vasoconstriction (82±9%, ≤20 min) and a late mesenteric vasodilation, starting at 30 min (108±4%) and reaching 138±6% at 60 min. In rats with HD rostral to the PVN, we only observed minor changes in the cardiovascular parameters. In the MAP, there was a slight decrease 60 min after the hypothalamic disconnection rostral (HD-R) (-9±4 mm Hg). There were no significant changes in HVC. RVC and MVC were increased 60 min after the HD-R (116±12% and 124±11%, respectively). These results suggest that vasodilation in the hindlimb and in the mesenteric bed could contribute to the observed decrease in MAP in HD caudal to PVN rats. © 2002 Elsevier Science B.V. All rights reserved.

Mechanisms in the bed nucleus of the stria terminalis involved in control of autonomic and neuroendocrine functions: A review

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Inhibitory mechanism of the nucleus of the solitary tract involved in the control of cardiovascular, dipsogenic, hormonal, and renal responses to hyperosmolality

The nucleus of the solitary tract (NTS) is the primary site of visceral afferents to the central nervous system. In the present study, we investigated the effects of lesions in the commissural portion of the NTS (commNTS) on the activity of vasopressinergic neurons in the hypothalamic paraventricular (PVN) and supraoptic (SON) nuclei, plasma vasopressin, arterial pressure, water intake, and sodium excretion in rats with plasma hyperosmolality produced by intragastric 2 M NaCl (2 ml/rat). Male Holtzman rats with 15-20 days of sham or electrolytic lesion (1 mA; 10 s) of the commNTS were used. CommNTS lesions enhanced a 2 M NaCl intragastrically induced increase in the number of vasopressinergic neurons expressing c-Fos in the PVN (28 ± 1, vs. sham: 22 ± 2 c-Fos/AVP cells) and SON (26 ± 4, vs. sham: 11 ± 1 c-Fos/AVP cells), plasma vasopressin levels (21 ± 8, vs. sham: 6.6 ± 1.3 pg/ml), pressor responses (25 ± 7 mmHg, vs. sham: 7 ± 2 mmHg), water intake (17.5 ± 0.8, vs. sham: 11.2 ± 1.8 ml/2 h), and natriuresis (4.9 ± 0.8, vs. sham: 1.4 ± 0.3 meq/1 h). The pretreatment with vasopressin antagonist abolished the pressor response to intragastric 2 M NaCl in commNTS-lesioned rats (8 ± 2.4 mmHg at 10 min), suggesting that this response is dependent on vasopressin secretion. The results suggest that inhibitory mechanisms dependent on commNTS act to limit or counterbalance behavioral, hormonal, cardiovascular, and renal responses to an acute increase in plasma osmolality. © 2013 the American Physiological Society.

Catecholaminergic neurons in the comissural region of the nucleus of the solitary tract modulate hyperosmolality-induced responses

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Pontomedullary and hypothalamic distribution of Fos-like immunoreactive neurons after acute exercise in rats

During exercise, intense brain activity orchestrates an increase in muscle tension. Additionally, there is an increase in cardiac output and ventilation to compensate the increased metabolic demand of muscle activity and to facilitate the removal of CO2 from and the delivery of O-2 to tissues. Here we tested the hypothesis that a subset of pontomedullary and hypothalamic neurons could be activated during dynamic acute exercise. Male Wistar rats (250-350 g) were divided into an exercise group (n = 12) that ran on a treadmill and a no-exercise group (n = 7). Immunohistochemistry of pontomedullary and hypothalamic sections to identify activation (c-Fos expression) of cardiorespiratory areas showed that the no-exercise rats exhibited minimal Fos expression. In contrast, there was intense activation of the nucleus of the solitary tract, the ventrolateral medulla (including the presumed central chemoreceptor neurons in the retrotrapezoid/parafacial region), the lateral parabrachial nucleus, the Kolliker-Fuse region, the perifornical region, which includes the perifornical area and the lateral hypothalamus, the dorsal medial hypothalamus, and the paraventricular nucleus of the hypothalamus after running exercise. Additionally, we observed Fos immunoreactivity in catecholaminergic neurons within the ventrolateral medulla (C1 region) without Fos expression in the A2, A5 and A7 neurons. In summary, we show for the first time that after acute exercise there is an intense activation of brain areas crucial for cardiorespiratory control. Possible involvement of the central command mechanism should be considered. Our results suggest whole brain-specific mobilization to correct and compensate the homeostatic changes produced by acute exercise. (c) 2012 IBRO. Published by Elsevier Ltd. All rights reserved.

Bed nucleus of the stria terminalis and the cardiovascular responses to chemoreflex activation

Several studies from our group have indicated that the BNST play an important role in baroreflex modulation. However, the involvement of the BNST in the chemoreflex activity is unknown. Thus, in the present study, we investigated the effect of the local bed nucleus of stria terminalis (BNST) neurotransmission inhibition by bilateral microinjections of the non-selective synaptic blocker cobalt chloride (CoCl2) on the cardiovascular responses to chemoreflex activation in rats. For this purpose, chemoreflex was activated with KCN (i.v.) before and after microinjections of CoCl2 into the BNST. Reversible BNST inactivation produced no significant changes in the magnitude and durations of both pressor and bradycardic responses to intravenous KCN infusion. These findings suggesting that BNST neurotransmission have not influence on both sympathoexcitatory and parasympathoexcitatory components of the peripheral chemoreflex activation. (C) 2012 Elsevier B.V. All rights reserved.

Hepatic-portal vein infusions of glucagon-like peptide-1 reduce meal size and increase c-Fos expression in the nucleus tractus solitarii, area postrema and central nucleus of the amygdala in rats

We recently reported that brief, remotely controlled intrameal hepatic-portal vein infusions of glucagon-like peptide-1 (GLP-1) reduced spontaneous meal size in rats. To investigate the neurobehavioural correlates of this effect, we equipped male Sprague-Dawley rats with hepatic-portal vein catheters and assessed (i) the effect on eating of remotely triggered infusions of GLP-1 (1 nmol/kg, 5 min) or vehicle during the first nocturnal meal after 3 h of food deprivation and (ii) the effect of identical infusions performed at dark onset on c-Fos expression in several brain areas involved in the control of eating. GLP-1 reduced (P < 0.05) the size of the first nocturnal meal and increased its satiety ratio. Also, GLP-1 increased (P < 0.05) the number of c-Fos-expressing cells in the nucleus tractus solitarii, the area postrema and the central nucleus of the amygdala, but not in the arcuate or paraventricular hypothalamic nuclei. These data suggest that the nucleus tractus solitarii, the area postrema and the central nucleus of the amygdala play a role in the eating-inhibitory actions of GLP-1 infused into the hepatic-portal vein; it remains to be established whether activation of these brain nuclei reflect satiation, aversion, or both.

Assessment of Alcohol Reward in Prairie Voles (Microtus ochrogaster)

The purpose of our study was to assess whether prairie voles find alcohol rewarding. Prairie voles have recently become a species of interest for alcohol studies, which have traditionally used other rodent model species including several different strains of mice and rats. The prairie vole is one of only two known rodent species that readily administers high levels of unsweetened alcohol, implicating it as a potentially effective animal model for studying alcohol abuse. However, voluntary consumption does not necessarily imply that prairie voles find it rewarding. Therefore the purpose of our study was to investigate if alcohol has rewarding properties for prairie voles using three different approaches: place conditioning, flavor conditioning, and immunohistochemistry. Furthermore, we sought to characterize their reward profile and compare it to other commonly used rodent models ¿ C57BL/6 mice, DBA/2J mice, and Sprague-Dawley rats. Place and flavor conditioning are behavioral methods that rely on the learned association between a stimulus and the effects of a drug; the drug of interest in these studies is alcohol. To assess whether prairie voles will demonstrate a conditioned preference for alcohol-paired stimuli, seven place conditioning studies were run that investigated a range of different doses, individual conditioning session durations, and trial durations. Video analysis revealed no difference in the amount of time spent on the alcohol-paired floor, suggesting no conditioned place preference for alcohol. Two flavor conditioning tests were conducted to assess whether voles would demonstrate a preference for an alcohol-paired flavored saccharin solution. Voles demonstrated reduced consumption of the alcohol-paired flavored saccharin solution, regardless of dose or flavor, when alcohol administration occurred after conditioning sessions (p=<0.001). When alcohol was administered before conditioning sessions, no difference in consumption of the alcohol-paired and saline-paired flavored saccharin solutions was seen (p=0.545). Previous studies that have documented similar behavior have hypothesized that this is an example of an anticipatory contrast effect. This theory proposes that prairie voles reduce their intake of a hedonic solution (flavored saccharin solution) in anticipation of later drug administration (alcohol). However, conditioning-based behavioral methods of studying alcohol reward are highly sensitive to the parameters of the conditioned stimulus, thus it is possible that voles will not show preference for alcohol-related stimuli, even if they do find alcohol rewarding. Immunohistochemical analysis supplemented this behavioral data by allowing us to identify specific neural regions that were directly activated in response to the acute administration of alcohol. No difference in the number of activated c-Fos neurons in the Nucleus Accumbens (NAc) core or shell was seen (p=0.3364; p=0.6698) in animals that received an acute injection of alcohol or saline. There was a significant increase in the number of activated c-Fos neurons in the Paraventricular Nucleus of the Hypothalamus (PVN) in alcohol-treated animals compared to saline-treated animals (p=0.0034). There was no difference in the pixel count of activated c-Fos neurons or in the % area activated in the Arcuate Nucleus between alcohol and saline-treated animals (p=0.4523; p=0.3304). In conclusion, the place conditioning studies that were conducted in this thesis suggest that prairie voles do not demonstrate preference or aversion towards alcohol-paired stimuli. The flavor conditioning studies suggest that prairie voles do not demonstrate aversion but rather avoidance of the alcohol-paired flavor in anticipation of future alcohol administration. The preliminary immunohistochemical data collected is inconclusive but cannot rule out the possibility of neuronal activation patterns indicative of reward. Taken together, our data indicate that prairie voles hav

The role of the bed nucleus of the stria terminalis in stress: A behavioral, neurophysiological and neuropharmacological study

During the fifty-five years since the origin of the modern concept of stress, a variety of neurochemical, physiological, behavioral and pathological data have been collected in order to define stress and catalogue the components of the stress response. Over the last twenty-five years, as interest in the neural mechanisms underlying the stress response grew, most of the studies have focused on the hypothalamus and major limbic structures such as the amygdala or on nuclei involved in neurochemical changes observed during stress. There are other CNS sites, such as the bed nucleus of the stria terminalis (BNST), that neuroanatomical and neurochemical studies suggest may be involved in stress, but these sites have rarely been studied. Four experiments were performed for this dissertation, the goal of which was to examine the BNST to determine its role in the regulation of the stress response. The first experiment demonstrated that electrical stimulation of BNST was sufficient to produce stress-like behaviors. The second experiment demonstrated that single BNST neurons altered their firing rate in response to both a noxious somatosensory stimulus such as tail pinch and electrical stimulation of the amygdala (AmygS). The third experiment showed that the opioid, cholinergic, and noradrenergic systems, three neurotransmitter systems implicated in the control of the stress response, were effective in altering the firing rate of BNST neurons. The fourth experiment demonstrated that the cholinergic effects were mediated via muscarinic receptors and showed that the effects of AmygS were not mediated via cholinergic pathways. Collectively, these findings provide a possible explanation for the nonspecificity in causation of stress and the invariability of the stress response and suggest a neurochemical basis for its pharmacological control. ^