A[C]2-deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. V. the septal area

The [¹⁴C]2-deoxyglucose (2-DG) metabolic mapping technique has been used to identify the regions responding with an augmented rate of metabolism following focal electrical stimulation of various sites within the lateral septal nucleus and medial septal nucleus/diagonal band (MSN/DB) complex in the rat. Since 2-DG uptake has been correlated with rates of functional activity, it was the intention of this study to suggest the anatomical substrates underlying various physiological and behavioral responses elicited by stimulation of the septal area. The results show that stimulation of any region within the lateral septal nucleus produced a profound bilateral activation of both the lateral septal nucleus, as well as the hippocampal formation. While stimulation of a number of different fiber systems associated with the lateral septum could contribute to the observed pattern of labeling, the data suggest that, functionally, a major consequence of such stimulation is the antidromic activation of CA3 → lateral septum fibers to axonal branch points, beyond which, orthodromic propagation of the impulse produces activation in CA3 target regions, including subfields CA1 and CA3, as well as the lateral septal nucleus, bilaterally. In addition, regions typically manifesting metabolic activation following stimulation of the lateral septal nucleus included the ipsilateral diagonal band of Broca, nucleus accumbens, lateral preoptic area and lateral hypothalamus, posteriorly, and the prelimbic cortex, anteriorly. Occasionally, target regions of the postcommissural fornix, including the medial mammillary nucleus and anterior thalamic nuclei were also activated following stimulation of the lateral septal nucleus. In contrast to the widespread pattern of activation resulting from stimulation of the lateral septal nucleus stimulation of the MSN/DB complex produced activation which was largely confined to the mediall forebrain bundle. In a final phase of the experiment, afterdischarge activity was elicited by sodium penicillin injection into the lateral septal nucleus. Such treatment produced more widespread 2-DG uptake, including more extensive activation within the lateral septal nucleus, hippocampal formation, amygdala, and thalamus. Additionally, the prefrontal cortex and temporal neocortex were activated.     

A [C]2-Deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. III. The hippocampal formation

The regions metabolically activated in the rat brain following focal electrical stimulation of various components of the hippocampal formation were identified with the use of [¹⁴C]2-deoxyglucose (2-DG) autoradiography. The results of these experiments, conducted in the rat, showed that in the absence of elicited afterdischarge activity, stimulation of either the CA1 or CA3 field of the dorsal hippocampus resulted in bilateral metabolic activation of only the dorsal hippocampus as well as of a relatively restricted region within the dorsomedial aspect of the lateral septal nucleus, bilaterally. In contrast, stimulation of either the CA1 or CA3 field of the ventral hippocampus resulted in bilateral activation of the ventral hippocampus and no region of the dorsal hippocampus. Following such stimulation, the lateral septal nucleus was also labeled bilaterally, but the activated regions were situated in a position ventrolateral to those resulting from stimulation at dorsal levels. Stimulation of the subicular cortex, in contrast, resulted in only ipsilateral activation of the hippocampal formation and lateral septal nucleus. Further rostral levels of the lateral septal nucleus were noted to be activated following stimulation of subicular cortex as compared to stimulation of the cornu Ammonis.The hypothalamus was directly activated by two pathways, the postcommissural fornix and the medial corticohypothalamic tract. Following stimulation at dorsal and posterior levels of CA1 and the subiculum, the mammillary bodies were demonstrably labeled by input from the postcommissural fornix. Regions of the medial hypothalamus were activated via the medial corticohypothalamic tract following stimulation of the ventral subiculum.The amygdala, stria terminalis and its bed nucleus were also shown to be demonstrably activated following stimulation of the ventral subiculum, ventral CAl field and posterior prosubiculum. This pathway may represent an additional route by which hippocampal modulation may indirectly modulate hypothalamic function.The presence of elicited afterdischarges resulted in more extensive patterns of metabolic labeling within the hippocampal formation and lateral septal nuclei as compared to experiments in which afterdischarges were not elicited. The extent of the demonstrable labeling, both within, and extrinsic to the hippocampal formation appeared to be a function of the duration and severity of the elicited seizure discharge. Additional structures which were demonstrably labeled following the elicitation of seizure activity include the entorhinal cortex-prepyriform area, amygdala, substantia innominata, putamen, substantia nigra, olfactory and prefrontal cortices and medial thalamic nuclei.     

The topographical organization of the hippocampal projection to the septal area: a comparative neuroanatomical analysis in the gerbil, rat, rabbit, and cat

An experiment was performed to determine the origin of the projection from the hippocampus to the septal area in the subrimate mammalian nervous system. Lesions were made by aspiration or by radio frequency in 4 gerbils, 17 rats, 8 rabbits, and 7 cats. Survival times varied from 2–5 days. Tissues were stained principally with the Fink Heimer I method for identification of degenerating axons and their terminals. Following lesions destroying any one or more of the fields of the dorsal hippocampus of the gerbil, rat, rabbit, or cat, terminal degeneration was observed only in the medial septal area, olfactory tubercle, and adjacent portions of the diagonal band. In addition, lesions producing total destruction of all dorsal hippocampal fields also resulted in the presence of terminal degeneration restricted to the medial septal area. In contrast, superficial lesions of field CA1 of the ventral hippocampus produced terminal degeneration in the lateral septal area, nucleus accumbens, olfactory tubercle, and adjacent portions of the diagonal band. Similar findings were also observed following more widespread lesions of the ventral hippocampus which produced damage to other CA fields as well. Superficial lesions of the posterior crus of the hippocampus (i.e., a position midway between dorsal and ventral hippocampus) resulted in terminal degeneration localized to an intermediolateral region of the septum. Combined lesions of the dorsal hippocampus and fimbria produced widespread terminal degeneration in both the lateral and medial septum indicating that the axons contained within the fimbria arise only from the ventral hippocampus. Finally, lesions of the medial and lateral segments of the fornix of the cat produced terminal degeneration in the medial and lateral regions of the septum, respectively. These findings, collectively, indicate that the origin of the topographical projection to the medial and lateral septum are the dorsal and ventral hippocampus, respectively. This projection is unrelated to cytoarchitectonic fields within the hippocampus and is also invariant among the species considered in this study.     

A [C]2-deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. IV. A pathway from the prefrontal cortical-medial thalamic system to the hypothalamus

The present study utilized the [¹⁴C]2-deoxyglucose (2-DG) cell labeling procedure to characterize a functional pathway from the prefrontal cortex (Pfc) and mediodorsal thalamic nucleus (MD) to the hypothalamus. Rats were injected with 2-DG prior to a 45 min experimental paradigm consisting of alternating 30 s on-off periods of electrical brain stimulation. Standard procedures were utilized for the removal and processing of brain tissue for X-ray autoradiography. In the first phase of this study, stimulation applied to the prefrontal cortex generally yielded a pattern of 2-DG distribution consistent with the findings of classical anatomical studies. Stimulation of the dorsomedial and ventromedial prefrontal cortex or the infralimbic cortex produced the most effective activation of the diencephalon. This activation was primarily limited to MD, with no involvement of any region of the hypothalamus. In the second phase of this study, brain regions activated following stimulation of sites along the rostro-caudal axis of MD were examined. Stimulation of MD resulted in the activation of the nucleus reuniens and other midline and non-specific thalamic nuclei. Stimulation of this nucleus also activated the ventromedial thalamic nucleus, medial aspects of the nucleus accumbens and the medial and sulcal prefrontal cortices. Again, in each of these cases, labeling within any region of the hypothalamus could not be detected. Since MD stimulation activated the midline thalamus, and the nucleus reuniens in particular, the last phase of this experiment involved stimulation of the nucleus reuniens in order to determine the source of medial thalamic inputs to the hypothalamus. Stimulation of the nucleus reuniens activated fibers which were distributed to both the medial and lateral hypothalamus. In addition, stimulation also activated the descending periventricular system, which could be followed to the level of the midbrain central gray and such limbic structures as the hippocampal formation, septal area, amygdala and prefrontal cortex. These findings indicate that Pfc-MD activation of the hypothalamus is achieved indirectly via interneurons within the nucleus reuniens.     

Efferent connections of the hippocampal formation in the rat.

In this investigation the projections of the hippocampal formation to the septal area and hypothalamus were studied in the rat with the combined use of 3H-amino acid radioautography and horseradish peroxidase histochemistry. The results indicate that all of the fibers which project to the hypothalamus and the majority of fibers which project to the septum arise from the subicular cortex and not from hippocampal pyramidal cells. The projection to both of these areas are topographically organized along the longitudinal axis of the hippocampal formation. Specifically, fibers from subicular cortical cells situated at the septal end of the hippocampal formation which project through the medial part of the dorsal fornix terminate in the dorsomedial quadrant of the lateral septal nucleus and in the dorsal portion of the pars posterior of the medial mammillary nucleus. Fibers from progressively more posteroventral levels of the hippocampal formation which project through more lateral portions of the dorsal fornix and fimbria terminate in progressively lateral and ventral quadrants of the lateral septal nucleus and in progressively more ventral portions of the pars posterior. Concerning the specific origin of the fornix system, fibers from only the prosubiculum and subiculum project through both the pre- and postcommissural fornix. Hippocampal pyramidal cells from all CA fields have a restricted projection through the precommissural fornix and terminate in the caudal half of the septum while the presubiculum projects solely through the postcommissural fornix. The medial corticohypothalamic tract (MCHT) was found to arise from cells located in anterior ventral levels of the subicular cortex. Fibers from this tract appeared to be distributed throughout the pericellular region of the entire ventromedial extent of the hypothalamus from the level of the suprachiasmatic nucleus through the level of the medial mammillary nucleus. In this way, the mammillary bodies receive input from the subicular cortex via two routes: the descending column of the fornix and the MCHT.     

Efferent connections of septal nuclei of the domestic chick (Gallus domesticus): an anterograde pathway tracing study with a bearing on functional circuits

Small iontophoretic injections of the anterograde tracer Phaseolus vulgaris leucoagglutinin were placed in different subregions of the septum of domestic chicks. The main targets of septal projections comprised the ipsi- and contralateral septal nuclei, including the nucleus of the diagonal band, basal ganglia, including the ventral paleostriatum, lobus parolfactorius, nucleus accumbens, and olfactory tubercle, archistriatum, piriform cortex, and anterior neostriatum. Further diencephalic and mesencephalic septal projections were observed in the ipsilateral preoptic region, hypothalamus (the main regions of afferentation comprising the lateral hypothalamic nuclei, ventromedial, paraventricular and periventricular nuclei, and the mammillary region), dorsal thalamus, medial habenular and subhabenular nuclei, midbrain central gray, and ventral tegmental area. Contralateral projections were also encountered in the septal nuclei, ventral paleostriatum, periventricular and anteromedial hypothalamic nuclei, suprachiasmatic nucleus, and the lateral hypothalamic area. Avian septal efferents are largely similar to those of mammals, the main differences being a relatively modest hippocampal projection arising mainly from the nucleus of the diagonal band (as confirmed by a specific experiment with the retrograde pathway tracer True blue), the lack of interpeduncular projection, and a greater contingent of amygdalar efferents arising from the lateral septum rather than the nucleus of the diagonal band. This pattern of connectivity is likely to reflect an important role of the avian septal nuclei in the coordination of limbic circuits and the integration of a wide variety of information sources modulating the appropriate behavioral responses: attention and arousal level, memory formation, hormonally mediated behaviors, and their affective components (such as ingestive, reproductive, and parental behaviors), social interaction, locomotor modulation, and circadian rhythm.     

Single unit analysis of the hippocampal projections to the septum in the cat

The purpose of this study was to map the hippocampal efferent projections to the septum and to determine the synaptic organization of the hippocampal-septal system in the cat. Single unit responses were recorded in the septum with tungsten microelectrodes following electrical stimulation of the dorsal or ventral hippocampus in the anesthetized cat. Dorsal hippocampal stimulation produced excitatory unit driving at short latencies in the medial septum while ventral hippocampal stimulation produced short latency excitation in the lateral septum. The excitatory phase was invariably followed by a period of inhibition. Inhibition without a prior excitatory phase was seen in widespread regions of the septum upon either dorsal or ventral hippocampal stimulation. It was concluded that efferents from the dorsal and ventral hippocampus terminate in a topographic manner in the medial and lateral septum, respectively, and have excitatory effects upon these neurons as well. An interneuronal inhibitory network capable of influencing large regions of the septum was suggested from the data.     

An analysis of the efferent connections of the septal area in the cat

The neuroanatomical organization of the efferent connections of the septal area in the cat was analyzed by the use of anterograde ([3H]leucine radioautography) and retrograde (horseradish peroxidase histochemistry) tracing techniques. The results indicate that the lateral septal nucleus projects to the nuclei of the diagonal band, preoptic area, lateral hypothalamus, and supramammillary region. The projections of the septofimbrial nucleus supply the nuclei of the diagonal band and the medial habenular nucleus. Projection targets of the vertical limb of the diagonal band are widespread and include the preoptic area, lateral hypothalamus, anterior limbic cortex, amygdala, medial habenular nucleus, interpeduncular nucleus and hippocampal formation. The projection from the vertical limb to the hippocampal formation is organized in a topographical manner in such a fashion that cells positioned near the midline project to the dorsal hippocampus and adjoining subicular cortex while fibers originating from cells situated more laterally project to more ventral parts of the hippocampal formation. In general, the projections from the horizontal limb were similar to those from the vertical limb, but several differences were noted. Fibers arising from the horizontal limb are distributed to the ventral tegmental area and interpeduncular nucleus but this region seems to lack a projection to either the habenular complex or to the ventral aspect of the hippocampal formation. Fibers arising from the bed nucleus of the anterior commissure are distributed to the preoptic region, lateral hypothalamus, supramammillary region, posterior aspect of the medial mammillary nucleus and lateral habenular nucleus.     

The organization of the hypothalamic pathways mediating affective defense behavior in the cat

The purpose of this study was to describe the hypothalamic pathways which mediate affective defense in the cat utilizing the methods of [14C]2-deoxyglucose (2-DG) and [3H]leucine radioautography in concert with the technique of electrical brain stimulation. The feline affective defense response, characterized by pupillary dilatation, piloerection, ear retraction, hissing, growling and striking with the forepaws, was elicited consistently by stimulation of sites within the ventromedial hypothalamus and anterior aspect of the medial hypothalamus. In one series of experiments, 2-DG autoradiography was employed to describe the brain regions activated following stimulation of sites in the region of the ventromedial hypothalamus from which affective defense had been elicited. Ventromedial hypothalamic stimulation produced activation primarily in forebrain regions situated rostral to the level of the stimulating electrode. These structures included principally the anteromedial hypothalamus and medial preoptic area, as well as the bed nuclei of the stria terminalis and anterior commissure, diagonal band and lateral septal area. The caudal extent of activation included only the dorsal and perifornical hypothalamus at the level of the stimulation site. In a second series of experiments, affective defense sites in the anteromedial hypothalamus were stimulated and the regional distribution of 2-DG label was identified. In contrast to the results obtained from ventromedial hypothalamic stimulation, these experiments revealed a marked descending distribution of label within the posterior hypothalamus, midbrain central gray and ventral tegmental area. Results obtained from studies in which tritiated amino acids were injected into affective defense sites in both the ventromedial nucleus and anteromedial hypothalamus confirmed the general findings observed with 2-DG autoradiography. From these observations, we have concluded that the organization of the pathway mediating affective defense behavior from the ventromedial hypothalamus to the midbrain involves an initial synapse within the region of the anteromedial hypothalamus and a second synapse in the midbrain central gray substance. The significance of the anteromedial hypothalamus for the expression of affective defense behavior was considered in the Discussion.     

Origin and topography of fibers contributing to the fornix in macaque monkeys

The distribution of neurons contributing to the fornix was mapped by placing the retrograde tracer horseradish peroxidase (HRP) in polyacrylamide gels in different medial to lateral locations within the fornix of three rhesus monkeys (Macaca mulatta). The HRP was placed from 3 to 5 mm caudal to the descending columns of the fornix. Additional information came from a series of rhesus and cynomolgus monkeys (Macaca fasciculata) with anterograde tracer injections in the medial temporal lobe. The hippocampal formation, including the subiculum and presubiculum, together with the entorhinal cortex (EC) and perirhinal cortex (area 35) contribute numerous axons to the fornix in a topographical manner. In contrast, the lateral perirhinal cortex (area 36) and parahippocampal cortical areas TF and TH only contained a handful of cells labeled via the fornix. The medial fornix originates from cells in the caudal half of the subiculum, the lamina principalis interna of the caudal half of the presubiculum, and from the perirhinal cortex (area 35). The intermediate portion of the fornix (i.e., that part midway between the midline and most lateral parts of the fornix) originates from cells in the rostral half of the subiculum and prosubiculum, the anterior presubiculum (only from the lamina principalis externa), the caudal presubiculum (primarily from lamina principalis interna), the rostral half of CA3, the EC (primarily 28I and 28M), and the perirhinal cortex (area 35). The lateral parts of the fornix arise from the rostral EC (28L only) and the most rostral portion of CA3. Subcortically, the medial septum, nucleus of the diagonal band, supramammillary nucleus, lateral hypothalamus, dorsal raphe nucleus, and the thalamic nucleus reuniens all send projections through the fornix, which presumably terminate in the hippocampus and adjacent parahippocampal region. These results not only help to define those regions that project via the fornix, but also reveal those subcortical projections to the hippocampal formation most likely to rely entirely on nonfornical pathways.     

Efferent systems of primary visual cortex: A review of structure and function

In an attempt to further characterize the nature of the functional organization of the amygdala, patterns of uptake of [¹⁴C]2-deoxyglucose (2-DG) were assessed following electrical stimulation of various sites within the amygdala and associated structures in the rat. The experimental paradigm consisted of electrical brain stimulation delivered continuously for periods of 30 s on and 30 s off for 45 min following injection of 2-DG. Brains were removed and processed for autoradiography. It was noted that a specificity existed regarding nuclei, which when stimulated, resulted in demonstrable metabolic activation of the hypothalamus. Amygdaloid structures producing such activation included the medial, cortical, and basomedial nuclei, and the amygdalo-hippocampal area. Stimulation of these nuclear regions differentially activated areas within the hypothalamus. Stimulation of the medial and cortical nuclei effectively labeled fibers supplying the ventromedial nucleus. Stimulation of the amygdalo hippocampal area activated fibers which innervated the medial preoptic region. In contrast, activated fibers supplying the ventrolateral hypothalamus were present following stimulation of the basomedial nucleus. Activated fibers were observed in the lateral region of the medial forebrain bundle following stimulation of the central nucleus. Since these fibers could be followed directly into the substantia nigra, it was not possible to determine the extent to which they may have terminated within the lateral hypothalamus. Stimulation of other regions of the amygdala produced no demonstrable activation of any region of hypothalamus.The overall patterns of subcortical activation following stimulation of amygdaloid nuclei revealed the presence of considerable quantities of labeled fibers within the bed nucleus of the stria terminalis and substantia innominata. In addition, stimulation of the bed nucleus activated fibers which predominantly supply the ventromedial hypothalamus while stimulation of the substantia innominata activated fibers which supply the lateral hypothalamus. These data, taken collectively with previous anatomical data, suggest the likelihood that these two regions may serve as important relays through which much of the amygdala can communicate with the hypothalamus.An analysis was also made of the distribution of labeled fibers following the generation of seizure activity induced by focal injections of penicillin placed into various regions of amygdala. The data suggest that the pathways involved in the manifestation of a given induced seizure follow the known routes of axons, either arising from or passing through the structure associated with the seizure focus. While the activation pattern associated with focal seizures induced by the placement of penicillin into the basal nuclear complex appeared to be limited to first-order neurons, the patterns associated with seizures induced from the posterior cortical nucleus reflected the additional activation of second and third-order neurons of the hippocampal-septal system.     

A 14C-2-deoxyglucose analysis of the neural pathways of the limbic forebrain in the rat: II. The hypothalamus

An attempt was made to characterize the nature of the functional organization of the hypothalamus by observing the patterns of uptake of 14C-2-deoxyglucose (2DG) following electrical stimulation of different regions within the preoptico-hypothalamus in the rat. The experimental paradigm consisted of electrical brain stimulation delivered continuously for periods of 30 sec on and 30 sec off for 45 minutes following injection of 2DG. Brains were removed and processed for autoradiography. Activation of the medial forebrain bundle was noted following stimulation of the nucleus accumbens and lateral preoptico-hypothalamus. Activated fibers could be followed only in a caudal direction through the medial forebrain bundle and into the ventral tegmental area as a result of nucleus accumbens stimulation. Stimulation of the lateral preoptic region or of the anterior half of lateral hypothalamus produced activation of the lateral septal nucleus, lateral habenular nucleus, perifornical region, midline thalamus and ventral tegmental area. Since stimulation of the perifornical hypothalamus significantly activated the rostro-caudal extent of the midbrain cental gray, it is suggested that impulses from the lateral hypothalamus reach the lower brainstem via its connections with the perifornical hypothalamus. Ventromedial hypothalamic stimulation activated only the lateral septal nucleus, cortico-medial amygdala and medial preoptico-hypothalamus, while medial preoptico-hypothalamic stimulation resulted in increased 2DG uptake in the midbrain central gray, thus suggesting that medial hypothalamic impulses reach the brainstem by first ascending to the level of the preoptico-hypothalamus. Mammillary body stimulation orthodromically activated fibers in the mammillothalamic and mammillotegmental tracts and antidromically fibers in the fornix for a short distance.     

A C-2-Deoxyglucose analysis of the neural pathways of the limbic forebrain in the rat: II. The hypothalamus

An attempt was made to characterize the nature of the functional organization of the hypothalamus by observing the patterns of uptake of ¹⁴C-2-deoxyglucose (2DG) following electrical stimulation of different regions within the preoptico-hypothalamus in the rat. The experimental paradigm consisted of electrical brain stimulation delivered continuously for periods of 30 sec on and 30 sec off for 45 minutes following injection of 2DG. Brains were removed and processed for autoradiography. Activation of the medial forebrain bundle was noted following stimulation of the nucleus accumbens and lateral preoptico-hypothalamus. Activated fibers could be followed only in a caudal direction through the medial forebrain bundle and into the ventral tegmental area as a result of nucleus accumbens stimulation. Stimulation of the lateral preoptic region or of the anterior half of lateral hypothalamus produced activation of the lateral septal nucleus, lateral habenular nucleus, perifornical region, midline thalamus and ventral tegmental area. Since stimulation of the perifornical hypothalamus significantly activated the rostro-caudal extent of the midbrain central gray, it is suggested that impulses from the lateral hypothalamus reach the lower brainstem via its connections with the perifornical hypothalamus. Ventromedial hypothalamic stimulation activated only the lateral septal nucleus, cortico-medial amygdala and medial preoptico-hypothalamus, while medial preoptico-hypothalamic stimulation resulted in increased 2DG uptake in the midbrain central gray, thus suggesting that medial hypothalamic impulses reach the brainstem by first ascending to the level of the preoptico-hypothalamus. Mammillary body stimulation orthodromically activated fibers in the mammillothalamic and mammillotegmental tracts and antidromically fibers in the fornix for a short distance.     

A PHA-L analysis of ascending projections of the dorsal raphe nucleus in the rat.

Ascending projections from the dorsal raphe nucleus (DR) were examined in the rat by using the anterograde anatomical tracer, Phaseolus vulgaris leucoagglutinin (PHA-L). The majority of labeled fibers from the DR ascended through the forebrain within the medial forebrain bundle. DR fibers were found to terminate heavily in several subcortical as well as cortical sites. The following subcortical nuclei receive dense projections from the DR: ventral regions of the midbrain central gray including the 'supraoculomotor central gray' region, the ventral tegmental area, the substantia nigra-pars compacta, midline and intralaminar nuclei of the thalamus including the posterior paraventricular, the parafascicular, reuniens, rhomboid, intermediodorsal/mediodorsal, and central medial thalamic nuclei, the central, lateral and basolateral nuclei of the amygdala, posteromedial regions of the striatum, the bed nucleus of the stria terminalis, the lateral septal nucleus, the lateral preoptic area, the substantia innominata, the magnocellular preoptic nucleus, the endopiriform nucleus, and the ventral pallidum. The following subcortical nuclei receive moderately dense projections from the DR: the median raphe nucleus, the midbrain reticular formation, the cuneiform/pedunculopontine tegmental area, the retrorubral nucleus, the supramammillary nucleus, the lateral hypothalamus, the paracentral and central lateral intralaminar nuclei of the thalamus, the globus pallidus, the medial preoptic area, the vertical and horizontal limbs of the diagonal band nuclei, the claustrum, the nucleus accumbens, and the olfactory tubercle. The piriform, insular and frontal cortices receive dense projections from the DR; the occipital, entorhinal, perirhinal, frontal orbital, anterior cingulate, and infralimbic cortices, as well as the hippocampal formation, receive moderately dense projections from the DR. Some notable differences were observed in projections from the caudal DR and the rostral DR. For example, the hippocampal formation receives moderately dense projections from the caudal DR and essentially none from the rostral DR. On the other hand, virtually all neocortical regions receive significantly denser projections from the rostral than from the caudal DR. The present results demonstrate that dorsal raphe fibers project significantly throughout widespread regions of the midbrain and forebrain.     

Connections of the rat lateral septal complex

The organization of lateral septal connections has been re-examined with respect to its newly defined subdivisions, using anterograde (PHAL) and retrograde (fluorogold) axonal tracer methods. The results confirm that progressively more ventral transverse bands in the hippocampus (defined by the orientation of the trisynaptic circuit) innervate progressively more ventral, transversely oriented sheets in the lateral septum. In addition, hippocampal field CA3 projects selectively to the caudal part of the lateral septal nucleus, which occupies topologically lateral regions of the transverse sheets, whereas field CA1 and the subiculum project selectively to the rostral and ventral parts of the lateral septal nucleus, which occupy topologically medial regions of the transverse sheets. Finally, the evidence suggests that progressively more ventral hippocampal bands innervate progressively thicker lateral septal sheets. In contrast, ascending inputs to the lateral septum appear to define at least 20 vertically oriented bands or subdivisions arranged orthogonal to the hippocampal input (Risold, P.Y. and Swanson, L.W., Chemoarchitecture of the rat lateral septal nucleus, Brain Res. Rev., 24 (1997) 91–113). Hypothalamic nuclei forming parts of behavior-specific subsystems share bidirectional connections with specific subdivisions of the lateral septal nucleus (especially the rostral part), suggesting that specific domains in the hippocampus may influence specific hypothalamic behavioral systems. In contrast, the caudal part of the lateral septal nuceus projects to the lateral hypothalamus and to the supramammillary nucleus, which projects back to the hippocampus and receives its major inputs from brainstem cell groups thought to regulate behavioral state. The neural system mediating defensive behavior shows these features rather clearly, and what is known about its organization is discussed in some detail.     

Catecholamin innervation of the basal forebrain. I. The septal area.

The septal area of the rat is known to receive a rich innervation by axons of catecholamine (CA) neurons. In the present study this innervatic was studied using biochemical assay of CA content and fluorescence histochemical analysis of the distribution of CA-producing axons to determine the nuclei origin of the septal CA innervation and the effects of lesions on these parameters. The autoradiographic tracing technique and the horseradish peroxidase (HRP)-retrograde transport technique also were used for this purpose. The norepinephrine (NE) content of the normal septal area is 1,162 ± 127 ng/g and the dopamine (DA) content is 522 ± 106 ng/g. Hemisection of the brainstem caudal to the locus coeruleus results in a 47% decrease in septal NE content and a unilateral locus coeruleus lesion produces a 48% decrease in septal NE content. These observations suggest that the NE innervation of the septal area arises approximately equally from the locus coeruleus and nuclei in the caudal brain stem. This is confirmed, at least in part, by the anterograde and retrograde transport studies. The DA innervation of the septal area is shown by all of the techniques employed to arise almost exclusively from cells of the ventral tegmental area. NE axons arising from the locus coeruleus distribute in the septal area to the hippocampal rudiment, the nucleus of the diagonal band, the interstitial nucleus of the stria terminalis, the medial septal nucleus, the lateral septal nucleus and the nucleus septofimbrialis. In each area the innervation is sparse to moderate in density and has the plexiform organization typical of locus coeruleus innervation. The brainstem NE innervation is very dense in the interstitial nucleus of the stria terminalis, moderately dense in the lateral septal nucleus and sparse in the nucleus of the diagonal band. The DA axons innervating the septum terminate in two distinct patterns. The first is identical to that seen in the neostriatum. Preterminal axons are very fine and non-varicose. As they reach a terminal area they branch markedly and give rise to extremely numerous, closely-packed, fine varicosities. This type of DA innervation is found in the medial part of the lateral septal nucleus in a dense band and about some scattered lateral septal nucleus neurons, in the nucleus accumbens and in the interstitial nucleus of the stria terminalis. The second pattern is for non-varicose preterminal axons to branch and terminate in pericellular baskets about lateral septal nucleus neurons or simply terminate in the lateral septal nucleus neuropil. In each case the terminal branches give off distinct varicosities which are larger than those formed in the first pattern of DA innervation. Thus, the septal area has a complex organization of CA innervation with NE axons arising from caudal brainstem nuclei and the locus coeruleus and DA axons arising from the ventral tegmental area.     

PHA-L analysis of projections from the supramammillary nucleus in the rat.

The projections of the supramammillary nucleus (SUM) were examined in the rat by the anterograde anatomical tracer Phaseolus vulgaris leucoagglutinin (PHA-L). The majority of labeled fibers from SUM ascended through the forebrain within the medial forebrain bundle. SUM fibers were found to terminate heavily in the hippocampal formation, specifically within the granule cell layer and immediately adjoining molecular layer of the dentate gyrus. In addition, SUM fibers were shown to distribute densely to several structures with strong connections with the hippocampus, namely, the nucleus reunions of the thalamus, the medial and lateral septum, the entorhinal cortex, and the endopiriform nucleus. SUM fibers were also shown to project significantly to several additional subcortical and cortical sites. The subcortical sites were the dorsal raphe nucleus, the midbrain central gray, the fields of Forel/zona incerta, the dorsomedial hypothalamic area, midline/intralaminar nuclei of the thalamus (posterior paraventricular, rhomboid, central medial, intermediodorsal, and mediodorsal), the medial and lateral preoptic areas, the bed nucleus of the stria terminalis, the substantia innominata, the vertical limb of the diagonal band nucleus, and the claustrum. The cortical sites were the occipital, temporal, parietal, and frontal cortices. Some notable differences were observed in projections from the lateral as compared to the medial SUM. For example, fibers originating from the lateral SUM distributed heavily to the hippocampal formation and parts of the cortex, whereas those from the medial SUM projected sparsely to these two regions. The SUM projections to the hippocampal formation and associated structures may serve as the substrate for a SUM involvement in the generation of the theta rhythm of the hippocampus and the gating of information flow through the hippocampal formation.     

Brain stimulation reward and dopamine terminal fields. II. Septal and cortical projections

The boundaries and relative sensitivities of the substrates of septal and cortical brain stimulation reward were mapped in relation to the dopamine terminal fields in these regions using a dorsal-ventral moveable electrode. Brain stimulation was rewarding at all levels of the posterior lateral septum and not just in the region of dopamine terminal innervation. Reward thresholds, ease of training, maximum response rates and stability of responding were all unrelated to the proximity of the stimulating electrode to the band of dopamine terminals revealed by glyoxylic acid-induced dopamine fluorescence. Stimulation was also rewarding in the interior lateral septum; the best sites were in the ventral portions of this region although dopamine terminal fluorescence was uniform throughout. Thus the anatomy of the brain stimulation reward substrate of the lateral septal nucleus does not bear a special relation to the anatomy of dopamine terminals within this region. Stimulation was also rewarding in each of the dopamine terminal fields of the cerebral cortex. The best self-stimulation was obtained with electrodes in the medial frontal cortex; sulcal frontal cortex was next best, entorhinal cortex was next, and pyriform cortex, though reliably positive, supported the weakest self-stimulation. Variations in self-stimulation threshold were seen as electrodes were moved through homogenous regions of dopamine terminal density in some regions, while stable thresholds were associated with movements through areas of varying dopamine terminal density in others; thus, again, there was no special relation between goodness of self-stimulation and density of dopaminergic innervation. These data suggest that rewarding brain stimulation in these regions is not due to direct activation of either the dopaminergic terminals or the cells that they innervate.     

Cholinergic enzymes in neocortex, hippocampus and basal forebrain of non-neurological and senile dementia of alzheimer-type patients

Choline acetyltransferase (ChAT) activity and acetylcholinesterase (AChE) staining were examined in different cortical regions, hippocampal formation and basal forebrain of non-neurological controls and of patients afflicted with senile dementia of Alzheimer type (SDAT). Both enzymes showed a clear topographical distribution in the various regions studied. In SDAT cases, ChAT activity was reduced by 0–60% in the neocortex and by up to 97% in the hippocampus depending on the area and layer examined. In the nucleus of the diagonal band of Broca (NDB) and the medial septal nucleus (MSN), the activity was decreased by 65% and 55%, respectively; no significant change was found in the lateral septal nucleus (LSN), nucleus basalis of Meynert (NBM), substantia innominata (SI) and globus pallidus (GP). Comparable changes were seen in AChE staining. The results indicate that degeneration or dysfunction of cholinergic neurons in the medial septal area and possibly neocortex is an important characteristic of SDAT.     

The DNA content of cerebral cortex neurons. Determinations by cytophotometry and high performance liquid chromatography

The boundaries and relative sensitivities of the substrates of septal and cortical brain stimulation reward were mapped in relation to the dopamine terminal fields in these regions using a dorsal-ventral moveable electrode. Brain stimulation was rewarding at all levels of the posterior lateral septum and not just in the region of dopamine terminal innervation. Reward thresholds, ease of training, maximum response rates and stability of responding were all unrelated to the proximity of the stimulating electrode to the band of dopamine terminals revealed by glyoxylic acid-induced dopamine fluorescence. Stimulation was also rewarding in the anterior lateral septum; the best sites were in the ventral portions of this region although dopamine terminal fluorescence was uniform throughout. Thus the anatomy of the brain stimulation reward substrate of the lateral septal nucleus does not bear a special relation to the anatomy of dopamine terminals within this region. Stimulation was also rewarding in each of the dopamine terminal fields of the cerebral cortex. The best self-stimulation was obtained with electrodes in the medial frontal cortex; sulcal frontal cortex was next best, entorhinal cortex was next, and pyriform cortex, though reliably positive, supported the weakest self-stimulation. Variations in self-stimulation threshold were seen as electrodes were moved through homogeneous regions of dopamine terminal density in some regions, while stable thresholds were associated with movements through areas of varying dopamine terminal density in others; thus, again, there was no special relation between goodness of self-stimulation and density of dopaminergic innervation. These data suggest that rewarding brain stimulation in these regions is not due to direct activation of either the dopaminergic terminals or the cells that they innervate.