Afferent projections to the hypothalamic area controlling emotional responses (HACER).

The Mesulam technique for horseradish peroxidase was used to study the subcortical afferent projections to a location in the hypothalamus that has been shown to control the complete cardiovascular (CV) response accompanying a specific emotional behavior. Major projections common to all baboons injected included the lateral septal nucleus; medial, cortical and basal amygdala; the anteroventral third ventricle area; the preoptic areas; the subiculum; the paraventricular nucleus of the thalamus; periventricular gray and the central gray of the midbrain; the midbrain tegmentum; locus ceruleus, parabrachial and raphe cells in the pons; and in the medulla, raphe nuclei, the nucleus of the solitary tract, in and around the dorsal motor nucleus of the vagus, and in the region of the nucleus ambiguus. Other projections in some but not all baboons included the subfornical organ and the midline and dorsomedial nuclei of the thalamus. The nucleus of the diagonal band of Broca was labeled to some degree with all injections but was most heavily labeled with the injection extending more laterally in the hypothalamus. These results fit well with physiological and behavioral studies dealing with neural control of emotional and CV responses and support the concept of an integrative area in the hypothalamus concerned specifically with the control of CV response accompanying emotion.

Origin of the pyramidal tract determined with horseradish peroxidase.

The origin of the axons contained in the pyramidal tract (PT) of the cat was established using retrograde transport of horseradish peroxidase (HRP). A complete section was made through a PT at the level of the medulla oblongata and HRP was applied to the sectioned axons. Cat brains were cut in frontal and sagittal planes and HRP-labeled cells were plotted in outlines of the brain sections. The entire cortical region containing PT cells was divided into 8 subregions and the percent of PT cells was determined in each. Surface cortex, subregions 1, 3 and 8, contained only 30--40% of PT cells; the majority resided in deep sulcal cortex, in subregions 2, 4, 5, 6 and 7. Subregion 1 (containing 6--12% of PT cells) extends rostral to the cruciate sulcus; subregion 3 (15--22%) extends from the cruciate sulcus caudally to the ansate sulcus; subregion 8 (7--8%) covers cortex laterally adjacent to subregion 3. The hidden banks of the cruciate sulcus contained the greatest concentration of PT cells, 28--34% in the dorsal bank (subregion 5) and 15--20% in the ventral bank (subregion 4). The coronal sulcus contained only 2--5% of PT cells in its dorsal bank (subregion 6) and 1--4% in its ventral bank (subregion 7). The presylvian sulcus contained 8--12% of all PT cells in its lateral bank (subregion 2). This new cortical area is not yet considered part of 'PT cortex'. Qualitative limitations of this study are discussed.

Unmyelinated axons in the pyramidal tract of the cat.

Ultrastructural preparations revealed the presence of unmyelinated axons in the pyramidal tract (PT) of the adult cat. At the level of the medulla oblongata, unmyelinated axons constituted 8-15% of the total PT population. Axon diameters ranged from 0.05 to 0.06 micron with a mean of 0.18 micron. Although axons were distributed throughout the PT, their density was highest in the medial part.

Subcortical projections to the hippocampal formation in squirrel monkey (Saimira sciureus).

Subcortical afferents to the hippocampal formation in squirrel monkey were investigated using horseradish peroxidase as a retrograde marker. Labeled cells were found in the medial septal area, the diagonal band of Broca, anterior and laterodorsal thalamic nuclei, reuniens and periventricular thalamic nuclei, lateral hypothalamus, supramamillary nucleus, and the dorsal and superior central midbrain raphe nuclei. These results in a primate confirm previous findings in rats and cats with the exception of the noradrenergic cell groups, where the interpretation of retrograde label was hampered by high levels of endogenous pigment.

A horseradish peroxidase-autoradiographic study of parietopulvinar connections in Saimiri sciureus.

Descending connections from parietal cortex to pulvinar in squirrel monkey were investigated with the autoradiographic method. Somatosensory areas I (SI) and II (SII) were found to project to the oral (PuO) and medial (PuM) subdivisions of the pulvinar. Projections from the posterior parietal region were recorded in circumscribed areas of PuM and the lateral (PuL) and inferior (PuI) subdivisions of pulvinar. Retrograde labeling with horseradish peroxidase (HRP) demonstrated that rostral parietal cortex including the lateral cortex of SI and the rostral part of area 5 received reciprocal projections from PuO and rostral PuM. Injections of HRP into medial and lateral regions of SI also resulted in labeled cells in PuO and PuM. Within the limitations of the HRP technique, the latter results suggest a direct pathway from pulvinar to primary sensory cortex. The experimental results confirm the accepted view of projections from parieto-temporo-occipital "association" cortex to PuM, PuL and PuI. In addition, reciprocal connections of rostral parietal cortex with PuO and PuM were demonstrated.

An analysis of potentially converging inputs to the rostral ventral thalamic nuclei of the cat.

Potentially convergent inputs to cerebellar-receiving and basal ganglia-receiving areas of the thalamus were identified using horseradish peroxidase (HRP) retrograde tracing techniques. HRP was deposited iontophoretically into the ventroanterior (VA), ventromedial (VM), and ventrolateral (VL) thalamic nuclei in the cat. The relative numbers of labeled neurons in the basal ganglia and the cerebellar nuclei were used to assess the extent to which the injection was in cerebellar-receiving or basal ganglia-receiving portions of thalamus. The rostral pole of VA showed reciprocal connections with prefrontal portions of the cerebral cortex. Only the basal ganglia and the hypothalamus provided non-thalamic input to modulate these cortico-thalamo-cortical loops. In VM, there were reciprocal connections with prefrontal, premotor, and insular areas of the cerebral cortex. The basal ganglia (especially the substantia nigra), and to a lesser extent, the posterior and ventral portions of the deep cerebellar nuclei, provided input to VM and may modulate these cortico-thalamo-cortical loops. The premotor cortical areas connected to VM include those associated with eye movements, and afferents from the superior colliculus, a region of documented importance in oculomotor control, also were labeled by injections into VM. The dorsolateral portion of the VA-VL complex primarily showed reciprocal connections with the medial premotor (area 6) cortex. Basal ganglia and cerebellar afferents both may modulate this cortico-thalamo-cortical loop, although they do not necessarily converge on the same thalamic neurons. The cerebellar input to dorsolateral VA-VL was from posterior and ventral portions of the cerebellar nuclei, and the major potential brainstem afferents to this region of thalamus were from the pretectum. Mid- and caudo-lateral portions of VL had reciprocal connections with primary motor cortex (area 4). The dorsal and anterior portions of the cerebellar nuclei had a dominant input to this cortico-thalamo-cortical loop. Potentially converging brainstem afferents to this portion of VL were from the pretectum, especially pretectal areas to which somatosensory afferents project.

An autoradiographic study of efferent connections of the globus pallidus in Macaca mulatta.

Radioactive amino acids were injected into restricted regions of the globus pallidus of rhesus macaques to allow identification of the organization and courses of efferent pallidal projections. The previously identified projection of the internal pallidal segment (GPi) to ventral thalamic nuclei showed a topographic organization, with the predominant projection from ventral GPi being to medial and caudal ventralis anterior (VA) and lateralis (VL) and from dorsal GPi to lateral and rostral VA and VL. Pallidal efferent fibers also extended caudally and dorsally into pars caudalis of VL, but they spared the portion of pars oralis of VL shown by others to receive input from the cerebellum. In addition to centromedian labeling in all animals, the parafascicular nucleus was also labeled when isotope was injected into dorsal GPi. The medial route from GPi to the midbrain tegmentum was more substantial than has been shown before, and along this route there was an indication that some fibers terminated in the prerubral region. The projection to the pedunculopontine nucleus was extensive, and fibers continued caudally into the parabrachial nuclei. Pallidal projections to the thalamus seem to be topographically organized but spare thalamic regions that interact with area 4. Caudally directed efferent fibers follow multiple routes and extend more caudally than to the pedunculopontine nuclei.

A horseradish peroxidase study of afferent connections of the globus pallidus in Macaca mulatta.

The high tonic discharge rates of globus pallidus neurons in awake monkeys suggest that these neurons may receive some potent excitatory input. Because most current electrophysiological evidence suggests that the major described pallidal afferent systems from the neostriatum are primarily inhibitory, we used retrograde transport of horseradish peroxidase (HRP) to identify possible additional sources of pallidal afferent fibers. The appropriate location was determined before HRP injection by mapping the characteristic high frequency discharge of single pallidal units in awake animals. In animals with injections confined to the internal pallidal segment, retrograde label was seen in neurons of the pedunculopontine nucleus, dorsal raphe nucleus, substantia nigra, caudate, putamen, subthalamic nucleus, parafascicular nucleus, zona incerta, medial and lateral subthalamic tegmentum, parabrachial nuclei, and locus coeruleus. An injection involving the external pallidal segment and the putamen as well resulted in additional labeling of cells in centromedian nucleus, pulvinar, and the ventromedial thalamus.

Organization of central nervous system pathways influencing blood pressure responses during emotional behavior.

A series of studies has demonstrated that the perifornical area of the hypothalamus ("acro-named" HACER, for Hypothalamic Area Controlling Emotional Responses) is responsible for producing the elevated blood pressure and other cardiovascular responses that accompany emotional behavior. The central neural structures providing afferents to the HACER are detailed, and the efferent outflow is analyzed to demonstrate how the HACER produces cardiovascular responses.

Neurons controlling cardiovascular responses to emotion are located in lateral hypothalamus-perifornical region.

We did four experiments to determine whether the lateral hypothalamus-perifornical (LH/PF) region is the source of neuronal cell bodies responsible for producing the cardiovascular (CV) responses associated with emotion or the defense reaction. Of particular concern was whether the paraventricular nucleus (PVN) plays a role in the generation of these CV responses. Mapping the hypothalamus with electrical stimulation showed that the CV pattern of responses was never produced by stimulating the PVN and was invariably produced by stimulating the LH/PF region. Complete electrolytic destruction of the PVN and subsequent axonal degeneration did not change the CV pattern of responses elicited by LH/PF stimulation, whereas any encroachment of the lesion on the LH/PF region decreased the magnitude of the CV responses. Injection of the neuroexcitotoxin ibotenic acid (Ibo) into the PVN did not affect responses to LH/PF stimulation, whereas Ibo injection into the LH/PF region eliminated or severely attenuated the CV responses. Retrograde labeling of cells from the thoracic cord and the ventrolateral reticular formation revealed a scattered group of cells in the LH/PF region that may be the cells controlling the CV responses. These results point directly to the LH/PF region as the source of the cell bodies responsible for the autonomic responses associated with emotion or defense reactions.

Volumetric growth of the major brain divisions in fetal Macaca nemestrina.

Brains from 22 age-dated fetal Macaca nemestrina were embedded in celloidin and prepared for histological serial sections. A computer-based morphometric system was used to digitize contours of neural structures and to calculate their areas and volumes. Shrinkage of the brain sections was corrected by a multiplication factor relating pre-processed brain volume to the computer-calculated volume of the processed brain. Volume growth of the telencephalon, mesencephalon and pons-medulla was linear over the fetal period of 60 days postconception to near-term at 166 days postconception. Volume growth of the total brain, diencephalon and cerebellum, was curvilinear with respect to age, with slower growth initially and faster growth in the later stages of gestation. Total brain and body grew with an almost 1:1 relation during the fetal period. The proportionate growth of the brain was largely accounted for by telencephalic growth. The other brain divisions all showed different growth rates in relation to growth of the body.

Functional analysis of hypothalamic control of the cardiovascular responses accompanying emotional behavior.

The cardiovascular (CV) responses to an acute emotional situation in unanesthetized, chair-restrained baboons include elevations in heart rate, blood pressure, and terminal aortic flow and a complex biphasic reduction in renal flow. The same CV responses can be produced by stimulating an area in the hypothalamus. Furthermore, bilateral ablation of the hypothalamic area eliminates CV responses to the emotional behavior while responses to exercise, free feed, and lever press remain unaltered. This effect is not due to memory loss, loss of emotionality, or a general loss of CV regulatory capacity. Efferent projections of the hypothalamic site were traced by means of autoradiography and afferent sources were traced by horseradish peroxidase injections. Efferents include projections to amygdala, central gray, zona incerta, midline thalamic nuclei, dorsal midbrain tegmentum, the parabrachial region. Afferents were widely distributed and included inputs from the subiculum, amygdala, septal area, central gray, locus ceruleus, interpeduncular nucleus, and bilateral labeling in and around the dorsal motor nucleus of X and the nucleus ambiguus.