The thalamic nuclei at the medial border of the medial geniculate body (i.e. the suprageniculate nucleus, the medial division of the medial geniculate nucleus, the posterior intralaminar nucleus and the peripeduncular nucleus) which relay sensory information to the amygdala are thought to receive convergent input from multiple sites. In order to delineate the organization of these multimodal thalamic nuclei, the locations of superior and inferior collicular neurons projecting to these nuclei were studied by means of retrograde transport methods. Small injections of the tracer Miniruby were made into single paralaminar thalamic nuclei. Injections of Miniruby into the suprageniculate nucleus labelled predominantly neurons in the stratum opticum of the superior colliculus, whereas injections into the medial division of the medial geniculate body, the posterior intralaminar nucleus and the peripeduncular nucleus labelled predominantly neurons in the deep layers of the superior colliculus. These injections also labelled neurons in the inferior colliculus. The majority of retrogradely labelled neurons were found in the external nucleus of the inferior colliculus and here predominantly in layer 2. Injections focused onto the medial division of the medial geniculate body additionally labelled magnocellular neurons in layer 3 of the external nucleus and a few neurons in the central nucleus. More ventrally located injections, focused onto the posterior intralaminar and peripeduncular nucleus, almost exclusively labelled neurons in layer 1 of the external nucleus and the dorsal part of the dorsal nucleus. After injections into the suprageniculate nucleus, only neurons in layer 2 were found. Neurons in the central nucleus of the inferior colliculus were only found after injections that involved the medial division of the medial geniculate body. The present results suggest that, despite a considerable degree of convergence in this thalamic region, each of these thalamic nuclei receives a unique pattern of projections from the superior and inferior colliculi. It appears that the thalamic nuclei may be concerned mainly, but not exclusively, with a single sensory modality, and give rise to parallel multimodal and unimodal pathways to the amygdala.

Injections of calcitonin gene-related peptide (CGRP) into the amygdala evoke fear-related behaviors and antinociceptive effects. In the present study we therefore characterized CGRP-containing amygdaloid afferents by injecting the retrograde tracer FluoroGold (FG) into subnuclei of the amygdala and adjacent divisions of the extended amygdala, namely, the lateral (LA) and central (CE) amygdaloid nuclei, interstitial nucleus of the posterior limb of the anterior commissure (IPAC), and the amygdalostriatal area (AStr). The distribution of retrogradely FG-labeled neurons and colocalization of CGRP-immunoreactivity with FG-labeling were mapped in the posterior paralaminar thalamic complex and parabrachial nuclei. The analysis of the posterior thalamus revealed that about 50% of CGRP-containing neurons projected to the AStr, the projections originating in the medial part of the medial geniculate body, posterior intralaminar nucleus, parvicellular subparafascicular nucleus, and peripeduncular nucleus. However, the percentage of CGRP-containing thalamic neurons projecting to the adjacent LA, medial part of the CE, and ventrocaudal part of the caudatoputamen rapidly dropped to 3-9%. There were no double-labeled cells after injections into the lateral and capsular parts of the CE and the IPAC. Thus, the AStr received the heaviest CGRP-containing projection from the posterior thalamus. CGRP-containing parabrachial neurons projected to the AStr and lateral, capsular, and medial parts of the CE, the projections originating in the external, crescent, and central parts of the lateral parabrachial nucleus and external part of the medial parabrachial nucleus. The results demonstrate a distinct projection pattern of CGRP-containing thalamic and parabrachial neurons to subnuclei of the amygdala and extended amygdala.

No abstract

The factors determining the development of specific fiber tracts in the central nervous system as well as the interactions of growth cones with the surrounding micromilieu are largely unknown. Here we investigated the ontogenetic development of the septohippocampal projection in the rat with the lipophilic carbocyanine dye DiI which is transported anterogradely and retrogradely in neurons and can be applied to fixed embryonic tissue. Photoconversion of anterogradely labeled fibers allowed us to study individual growth cones by electron microscopy. The first axons originating from the septal complex were found in the hippocampus as early as on embryonic day (ED) 19, reaching the fimbrial pole of the hippocampus on ED 18. However, on ED 17 we consistently found retrogradely labeled cells in the hippocampus, indicating that the development of the hippocamposeptal projection precedes that of the septohippocampal projection. On ED 19, the majority of the axons directed toward the hippocampal formation passed the hippocampus and grew further into the subicular complex and entorhinal cortex. These axons gave off collaterals that invaded the hippocampus proper. A fairly adult pattern of the septohippocampal projection was reached on postnatal day 10, although may growth cones were still found. A comparative analysis of individual growth cones found in the fimbria and the hippocampus proper revealed no striking differences in their morphology. Electron microscopic analysis showed that growth cones in the fimbria were mainly contacted by other axons, whereas growth cones in the hippocampus had contact with all available elements. This may indicate that growing septohippocampal fibers are guided by axons of the earlier formed hippocamposeptal projection. In the hippocampus proper, other cues, probably derived from the target itself, may guide the septohippocampal axons to their appropriate target cells.

Thalamic nuclei adjacent to the medial geniculate body play a pivotal role in processing of sensory stimuli during emotional situations. These nuclei, which include the suprageniculate nucleus (SG), the posterior intralaminar nucleus (PIN), the peripeduncular nucleus (PP) and the medial division of the medial geniculate body (MGm), project to both cortex and amygdala, but target areas and the extent of the projection of individual nuclei are not known yet. The aim of the present study was to analyze the contribution of individual nuclei to the cortical projection with modern sensitive tracing techniques. Small injections of Miniruby or PHA-L were made into single thalamic nuclei. All thalamic nuclei have in common a projection into the upper portion of layer I of the temporal aspect of the cortical mantle. Furthermore, SG, PIN, MGm and PP each demonstrated a convergent projection to lower layer III and to layer IV of the ectorhinal and visceral cortex. Only MGm projects to layer VI of primary auditory and temporal association cortices. Within the perirhinal cortex zones of convergence and divergence exist. The present results demonstrate a differential thalamocortical projection of single thalamic nuclei to those cortical areas which are involved in the transmission of sensory signals to the amygdala via the thalamocortico-cortical pathway and to the hippocampus via the entorhinal cortex. The thalamic nuclei are thus in a position to activate the amygdala and to modulate the information flow of the thalamocortico-cortical pathway to both amygdala and hippocampus.