Functional mapping of the circuits involved in the expression of contextual fear responses in socially defeated animals.

In this study, we have aimed at outlining the neural systems underlying the expression of contextual fear to social defeat. First, we have developed an experimental procedure, where defeated animals could express, without the presence of a dominant aggressive male, robust and reliable conditioned fear responses to the context associated with social defeat. Next, by examining the pattern of Fos expression, we have been able to outline a brain circuit comprising septal and amygdalar sites, as well as downstream hypothalamic paths, putatively involved in the expression of contextual fear to social threat. Of particular relevance, we have found that exposure to a defeat-associated context results in a striking Fos up-regulation in the dorsomedial part of the dorsal premammillary nucleus (PMDdm). To further understand the role of the PMDdm in the circuit organizing conditioned fear to social threats, we have been able to observe that pharmacological blockade of the PMDdm reduced fear responses to a social defeat-associated context. Next, we observed that pharmacological blockade of the dorsomedial part of the periaqueductal gray, one of the main targets of the PMDdm, produced an even higher reduction of conditioned fear in defeated intruders, and appears as an important node for the expression of contextual defensive responses to social threats. The present results help to elucidate the basic organization of the neural circuits underlying contextual conditioned responses to social defeat, and reveal that they share at least part of the same circuit involved in innate responses to social defeat to an aggressive conspecific.

Lesions of the ventral premammillary nucleus disrupt the dynamic changes in Kiss1 and GnRH expression characteristic of the proestrus-estrus transition.

We have recently demonstrated that the ventral premammillary nucleus (PMV) plays a key role in the metabolic control of the female reproductive axis. However, whether PMV neurons modulate the reproductive neural circuitry and/or the expression of sexual behaviors has not been determined. Here, we showed that the expression of estrogen and progesterone receptors in the PMV is modulated by changing levels of sex steroids across the estrous cycle. We also showed that sexual behavior, not the high physiologic levels of sex steroids, induces Fos in PMV neurons. Bilateral lesions of the PMV caused no significant changes in proceptive behavior but a high percentage of PMV-lesioned rats failed to exhibit lordosis behavior when exposed to a sexually experienced male rat (50% vs. 18% in the control group). Notably, lesions of the PMV disrupted the physiologic fluctuations of Kiss1 and GnRH mRNA expression characteristic of the proestrus-to-estrus transition. This neurochemical imbalance may ultimately alter female reproductive behavior. Our findings suggest that the PMV is a component of the neural circuitry that modulates the physiologic fluctuations of key neuroendocrine players (i.e., Kiss1 and GnRH) in the control of the female reproductive physiology.

Hypothalamic sites responding to predator threats--the role of the dorsal premammillary nucleus in unconditioned and conditioned antipredatory defensive behavior.

In this study we provide a comprehensive analysis of the hypothalamic activation pattern during exposure to a live predator or an environment previously associated with a predator. Our results support the view that hypothalamic processing of the actual and the contextual predatory threats share the same circuit, in which the dorsal premammillary nucleus (PMd) plays a pivotal role in amplifying this processing. To further understand the role of the PMd in the circuit organizing antipredatory defensive behaviors, we studied rats with cytotoxic PMd lesions during cat exposure and examined the pattern of behavioral responses as well as how PMd lesions affect the neuronal activation of the systems engaged in predator detection, in contextual memory formation and in defensive behavioral responses. Next, we investigated how pharmacological blockade of the PMd interferes with the conditioned behavioral responses to a context previously associated with a predator, and how this blockade affects the activation pattern of periaqueductal gray (PAG) sites likely to organize the conditioned behavioral responses to the predatory context. Behavioral observations indicate that the PMd interferes with both unconditioned and conditioned antipredatory defensive behavior. Moreover, we have shown that the PMd influences the activation of its major projecting targets, i.e. the ventral part of the anteromedial thalamic nucleus which is likely to influence mnemonic processing, and PAG sites involved in the expression of antipredatory unconditioned and conditioned behavioral responses. Of particular relevance, this work provides evidence to elucidate the basic organization of the neural circuits integrating unconditioned and contextual conditioned responses to predatory threats.

Functional mapping of the prosencephalic systems involved in organizing predatory behavior in rats.

The study of the neural basis of predatory behavior has been largely neglected over the recent years. Using an ethologically based approach, we presently delineate the prosencephalic systems mobilized during predation by examining Fos immunoreactivity in rats performing insect hunting. These results were further compared with those obtained from animals killed after the early nocturnal surge of food ingestion. First, predatory behavior was associated with a distinct Fos up-regulation in the ventrolateral caudoputamen at intermediate rostro-caudal levels, suggesting a possible candidate to organize the stereotyped sequence of actions seen during insect hunting. Insect predation also presented conspicuous mobilization of a neural network formed by a distinct amygdalar circuit (i.e. the postpiriform-transition area, the anterior part of cortical nucleus, anterior part of basomedial nucleus, posterior part of basolateral nucleus, and medial part of central nucleus) and affiliated sites in the bed nuclei of the stria terminalis (i.e. the rhomboid nucleus) and in the hypothalamus (i.e. the parasubthalamic nucleus). Accordingly, this network is likely to encode prey-related motivational values, such as prey's odor and taste, and to influence autonomic and motor control accompanying predatory eating. Notably, regular food intake was also associated with a relatively weak Fos up-regulation in this network. However, during regular surge of food intake, we observed a much larger mobilization in hypothalamic sites related to the homeostatic control of eating, namely, the arcuate nucleus and autonomic parts of the paraventricular nucleus. Overall, the present findings suggest potential neural systems involved in integrating prey-related motivational values and in organizing the stereotyped sequences of action seen during predation. Moreover, the comparison with regular food intake contrasts putative neural mechanisms controlling predatory related eating vs. regular food intake.

c-Fos expression induced by electroacupuncture at the Zusanli point in rats submitted to repeated immobilization.

In laboratory animals, acupuncture needs to be performed on either anesthetized or, if unanesthetized, restrained subjects. Both procedures up-regulate c-Fos expression in several areas of the central nervous system, representing therefore a major pitfall for the assessment of c-Fos expression induced by electroacupuncture. Thus, in order to reduce the effect of acute restraint we used a protocol of repeated restraint for the assessment of the brain areas activated by electroacupuncture in adult male Wistar rats weighing 180-230 g. Repeated immobilization protocols (6 days, 1 h/day and 13 days, 2 h/day) were used to reduce the effect of acute immobilization stress on the c-Fos expression induced by electroacupuncture at the Zusanli point (EA36S). Animals submitted to immobilization alone or to electroacupuncture (100 Hz, 2-4 V, faradic wave) in a non-point region were compared to animals submitted to electroacupuncture at EA36S (4 animals/subgroup). c-Fos expression was measured in 41 brain areas by simple counting of cells and the results are reported as number of c-Fos-immunoreactive cells/10,000 m . The protocols of repeated immobilization significantly reduced the immobilization-induced c-Fos expression in most of the brain areas analyzed (P < 0.05). Animals of the EA36S groups had significantly higher levels of c-Fos expression in the dorsal raphe nucleus, locus coeruleus, posterior hypothalamus and central medial nucleus of the thalamus. Furthermore, the repeated immobilization protocols intensified the differences between the effects of 36S and non-point stimulation in the dorsal raphe nucleus (P < 0.05). These data suggest that high levels of stress can interact with and mask the evaluation of specific effects of acupuncture in unanesthetized animals.

c-Fos expression induced by electroacupuncture at the Zusanli point in rats submitted to repeated immobilization

In laboratory animals, acupuncture needs to be performed on either anesthetized or, if unanesthetized, restrained subjects. Both procedures up-regulate c-Fos expression in several areas of the central nervous system, representing therefore a major pitfall for the assessment of c-Fos expression induced by electroacupuncture. Thus, in order to reduce the effect of acute restraint we used a protocol of repeated restraint for the assessment of the brain areas activated by electroacupuncture in adult male Wistar rats weighing 180-230 g. Repeated immobilization protocols (6 days, 1 h/day and 13 days, 2 h/day) were used to reduce the effect of acute immobilization stress on the c-Fos expression induced by electroacupuncture at the Zusanli point (EA36S). Animals submitted to immobilization alone or to electroacupuncture (100 Hz, 2-4 V, faradic wave) in a non-point region were compared to animals submitted to electroacupuncture at EA36S (4 animals/subgroup). c-Fos expression was measured in 41 brain areas by simple counting of cells and the results are reported as number of c-Fos-immunoreactive cells/10,000 æm². The protocols of repeated immobilization significantly reduced the immobilization-induced c-Fos expression in most of the brain areas analyzed (P < 0.05). Animals of the EA36S groups had significantly higher levels of c-Fos expression in the dorsal raphe nucleus, locus coeruleus, posterior hypothalamus and central medial nucleus of the thalamus. Furthermore, the repeated immobilization protocols intensified the differences between the effects of 36S and non-point stimulation in the dorsal raphe nucleus (P < 0.05). These data suggest that high levels of stress can interact with and mask the evaluation of specific effects of acupuncture in unanesthetized animals.

Combinatorial amygdalar inputs to hippocampal domains and hypothalamic behavior systems.

The expression of innate reproductive, defensive, and ingestive behaviors appears to be controlled by three sets of medial hypothalamic nuclei, which are modulated by cognitive influences from the cerebral hemispheres, including especially the amygdala and hippocampal formation. PHAL analysis of the rat amygdala indicates that a majority of its cell groups project topographically (a) to hypothalamic behavior systems via direct inputs, and (b) to partly overlapping sets of hypothalamic behavior control systems through inputs to ventral hippocampal functional domains that in turn project to the medial hypothalamus directly, and by way of the lateral septal nucleus. Amygdalar cell groups are in a position to help bias or prioritize the temporal order of instinctive behavior expression controlled by the medial hypothalamus, and the memory of associated events that include an emotional or affective component.

Lesions of the dorsomedial hypothalamic nucleus do not influence the daily profile of pineal metabolism in rats.

The present study attempted to characterize the effects of electrolytic lesions of the hypothalamic dorsomedial nucleus on the daily profile of pineal metabolism as well as on the inhibition of pineal melatonin synthesis induced by acute light exposure during the night. Adult male Wistar rats (n = 107, 12:12 h light-dark cycle) were left intact (n = 47) or lesioned (n = 60). Lesioned rats and their respective controls were killed at six time points distributed throughout the light-dark cycle. At ZT (zeitgeber time) 18 the animals were killed either in the dark or after 15 min of light stimulation. Pineal glands were assayed using high-performance liquid chromatography with electrochemical detection (HPLC-ED). There was no difference in the amounts of pineal indoles between lesioned and control rats under any of the experimental situations tested. These results suggest that in rats, the hypothalamic dorsomedial nucleus does not participate in either the neural control of daily pineal metabolism or the nocturnal light-induced inhibition of the pineal metabolism.

Afferent connections of the dorsal premammillary nucleus.

The dorsal premammillary nucleus (PMd) is thought to play a critical role for the expression of fear responses to environmental threats. We have reported previously that during an encounter with a predator the PMd presents an impressive increase in Fos levels and cell body-specific chemical lesions therein virtually eliminate the expression of escape and freezing responses. In the present study, we carried out a systematic analysis of PMd afferent connections combining anterograde and retrograde tracing methods in the rat. We show that the nucleus receives inputs from several widely distributed areas in the forebrain and, to a much lesser extent, from the brainstem as well. From this information, it seems that the major telencephalic source of input to the PMd is the interfascicular nucleus of the bed nuclei of the stria terminalis. In addition, substantial telencephalic inputs to the nucleus seem to arise from the infralimbic and prelimbic areas, and the lateral septal nucleus. In the diencephalon, massive inputs to the PMd arise from the anterior hypothalamic nucleus, specific parts of the perifornical region, the retinoceptive region of the lateral hypothalamic area, and the anterior and dorsomedial parts of the ventromedial hypothalamic nucleus. In contrast, the ventral tegmental nucleus seems to be the only brainstem site that provides substantial inputs to the PMd. Overall, the present analysis helps to delineate prosencephalic circuits seemingly critical for the organization of innate fear responses to environmental threats, where the PMd presents a major associative role. Furthermore, by means of massive inputs from the ventral tegmental nucleus, the PMd is in a position to integrate information from a neural system involved in spatial working memory, which may be of particular relevance for an effect of attentional mechanisms on the selection of appropriate escape strategies.

Organization of projections from the medial nucleus of the amygdala: a PHAL study in the rat.

The organization of axonal projections from the four recognized parts of the medial amygdalar nucleus (MEA) were characterized with the Phaesolus vulgaris leucoagglutinin (PHAL) method in male rats. The results indicate that the MEA consists of two major divisions, ventral and dorsal, and that the former may also consist of rostral and caudal regions. As a whole, the MEA generates centrifugal projections to several parts of the accessory and main olfactory sensory pathways, and projections to a) several parts of the intrahippocampal circuit (ventrally); b) the ventral striatum, ventral pallidum, and bed nuclei of the stria terminalis (BST) in the basal telencephaon; c) many parts of the hypothalamus; d) midline and medial parts of the thalamus; and e) the periaqueductal gray, ventral tegmental area, and midbrain raphé. The dorsal division of the MEA (the posterodorsal part) is characterized by projections to the principal nucleus of the BST, and to the anteroventral periventricular, medial, and central parts of the medial preoptic, and ventral premammillary hypothalamic nuclei. These hypothalamic nuclei project heavily to neuroendocrine and autonomic-related parts of the hypothalamic periventricular zone. The ventral division of the MEA (the anterodorsal, anteroventral, and posteroventral parts) is characterized by dense projections to the transverse and interfascicular nuclei of the BST, and to the lateral part of the medial preoptic, anterior hypothalamic, and ventromedial hypothalamic nuclei. However, dorsal regions of the ventral division provide rather dense inputs to the medial preoptic region and capsule of the ventromedial nucleus, whereas ventral regions of the ventral division preferentially innervate the anterior hypothalamic, dorsomedial, and ventral parts of the ventromedial nuclei. Functional evidence suggests that circuits associated with dorsal regions of the ventral division may deal with reproductive behavior, whereas circuits associated with ventral regions of the ventral division may deal preferentially with agonistic behavior.

Connections of the posterior nucleus of the amygdala.

The connections of a relatively homogeneous band of neurons in the caudal amygdala have been examined with anterograde and retrograde axonal tracing methods in the rat. This region, called here the posterior nucleus of the amygdala (PA), corresponds in part to an area that has been referred to as the cortico-amygdaloid transition area, posterior part of the medial nucleus of the amygdala, amygdalo-hippocampal transition area, and posteromedial basal nucleus. Experiments with fluorogold and phaseolus vulgaris leucoagglutinin (PHAL) indicate that the major neuronal input to the PA arises in the ventral premammillary nucleus, and that substantial projections also arise in olfactory-related areas such as the medial nucleus of the amygdala, bed nucleus of the accessory olfactory tract, and posterior cortical nucleus of the amygdala, as well as in the ventral subiculum and adjacent parts of hippocampal field CA1. Other seemingly minor inputs, including cholinergic fibers from the substantia innominata, dopaminergic fibers from the ventral tegmental area, and serotoninergic fibers from the dorsal nucleus of the raphe, were also identified. The efferent projections of the PA as determined with the PHAL method appear to follow five major routes: 1) a relatively small group of laterally directed fibers innervates the dorsal endopiriform nucleus, and a few of these fibers reach cortical area TR and the lateral entorhinal area; 2) another small group of fibers courses medially to innervate the ventral subiculum and adjacent parts of field CA1; 3) many fibers course ventrally to innervate the outer molecular layer of the medial part of the posterior cortical nucleus of the amygdala; 4) a moderate group of fibers courses rostrally to innervate primarily the posterodorsal part of the medial nucleus of the amygdala, although some fibers continue on to end less densely in rostral parts of the medial nucleus of the amygdala before leaving the amygdala through the ansa peduncularis; and 5) the major output of the PA courses through the stria terminalis. One branch of this pathway massively innervates the principal nucleus of the bed nuclei of the stria terminalis before entering the medial hypothalamus, where it ends massively in the anteroventral periventricular and medial preoptic nuclei, ventrolateral part of the ventromedial nucleus and adjacent parts of the basal lateral hypothalamic area, and ventral premammillary nucleus. The other branch sends fibers to the ventral lateral septal nucleus, nucleus accumbens, olfactory tubercle, and infralimbic area of the prefrontal cortex.(ABSTRACT TRUNCATED AT 400 WORDS)

Projections of the ventral premammillary nucleus.

The projections of the ventral premammillary nucleus (PMv) have been examined with the Phaseolus vulgaris leucoagglutinin (PHAL) method in adult male rats. The results indicate that the nucleus gives rise to two major ascending pathways and a smaller descending pathway. One large ascending pathway terminates densely in most regions of the periventricular zone of the hypothalamus, with the notable exception of the suprachiasmatic, suprachiasmatic preoptic, and median preoptic nuclei. This pathway is in a position to influence directly many cell groups known to regulate anterior pituitary function. The second large pathway ascends through the medial zone of the hypothalamus and densely innervates the ventrolateral part of the ventromedial nucleus and adjacent basal parts of the lateral hypothalamic area, medial preoptic nucleus, principal nucleus of the bed nuclei of the stria terminalis, ventral lateral septal nucleus, posterodorsal part of the medial nucleus of the amygdala, posterior nucleus, and immediately adjacent regions of the posterior cortical nucleus of the amygdala. It is already known that these regions are major components of the sexually dimorphic circuit, and, interestingly, that they provide the major neural inputs to the PMv. The smaller descending projection from the PMv seems to innervate preferentially the posterior hypothalamic nucleus, although a small number of fibers appear to end in the tuberomammillary nucleus, supramammillary nucleus, specific regions of the medial mammillary nucleus, interfascicular nucleus, interpeduncular nucleus, periaqueductal gray, dorsal nucleus of the raphe, laterodorsal tegmental nucleus, Barrington's nucleus, and locus coeruleus. Relatively sparse terminal fields associated with ascending fibers were also observed in the dorsomedial nucleus of the hypothalamus; in the nucleus reuniens, parataenial nucleus, paraventricular nucleus of the thalamus, and mediodorsal nucleus; in the central nucleus of the amygdala, anterodorsal part of the medial nucleus of the amygdala, posterior part of the basomedial nucleus of the amygdala; and in the ventral subiculum and adjacent parts of hippocampal field CA1, and the infralimbic and prelimbic areas of the medial prefrontal cortex. Taken as a whole, the evidence suggests that the PMv receives two major inputs--one from the sexually dimorphic circuit, and the other from the blood in the form of gonadal steroid hormones--and gives rise to two major outputs: one (perhaps feed-forward) to the neuroendocrine (periventricular) zone of the hypothalamus, and the other (perhaps feed-back) to the sexually dimorphic circuit.

Projections of the ventral subiculum to the amygdala, septum, and hypothalamus: a PHAL anterograde tract-tracing study in the rat.

The projections of the ventral subiculum are organized differentially along the dorsoventral (or septotemporal) axis of this cortical field, with more ventral regions playing a particularly important role in hippocampal communication with the amygdala, bed nuclei of the stria terminalis (BST), and rostral hypothalamus. In the present study we re-examined the projection of the ventral subiculum to these regions with the Phaseolus vulgaris leucoagglutinin (PHAL) method in the rat. The results confirm and extend earlier conclusions based primarily on the autoradiographic method. Projections from the ventral subiculum course either obliquely through the angular bundle to innervate the amygdala and adjacent parts of the temporal lobe, or follow the alveus and fimbria to the precommissural fornix and medial corticohypothalamic tract. The major amygdalar terminal field is centered in the posterior basomedial nucleus, while other structures that appear to be innervated include the piriformamygdaloid area, the posterior basolateral, posterior cortical, posterior, central, medial, and intercalated nuclei, and the nucleus of the lateral olfactory tract. Projections from the ventral subiculum reach the BST mainly by way of the precommissural fornix, and provide rather dense inputs to the anterodorsal area as well as the transverse and interfascicular nuclei. The medial corticohypothalamic tract is the main route taken by fibers from the ventral subiculum to the hypothalamus, where they innervate the medial preoptic area, "shell" of the ventromedial nucleus, dorsomedial nucleus, ventral premammillary nucleus, and cell-poor zone around the medial mammillary nucleus. We also observed a rather dense terminal field just dorsal to the suprachiasmatic nucleus that extends dorsally and caudally to fill the subparaventricular zone along the medial border of the anterior hypothalamic nucleus and ventrolateral border of the paraventricular nucleus. The general pattern of outputs to the hypothalamus and septum is strikingly similar for the ventral subiculum and suprachiasmatic nucleus, the endogenous circadian rhythm generator.