The glutamate decarboxylase-, leucine enkephalin-, methionine enkephalin- and substance P-immunoreactive neurons in the neostriatum of the rat and cat: evidence for partial population overlap.

We have examined the populations of neurons in the neostriatum of both rat and cat that are immunoreactive for glutamate decarboxylase, [Leu]enkephalin, [Met]enkephalin and substance P. Neurons that were immunoreactive for glutamate decarboxylase made up 47% of the neurons in our samples from the rat and ranged from 39 to 49% of the neurons in the cat. Those immunoreactive for [Leu]enkephalin made up 44-49% of the neurons in rat neostriatum, and 38-47% in the cat, and those immunoreactive for [Met]enkephalin made up 36-41% of the neurons in rat and 43-49% of the neurons in the cat. Substance P-immunoreactive neurons made up 30-38% of neurons in rat and 32-39% in cat. Most substance P neurons (particularly the most darkly staining ones) were, however, clustered such that they were most numerous in the patch compartment of neostriatum; within the patches the substance P neurons comprised 59% of neurons in the rat and 55% in cat, but in the matrix substance P neurons comprised only 32% of neurons in the rat and 25% in the cat. Samples taken from sections processed for two-color double labeling immunocytochemistry revealed that 12% of neurons label for both glutamate decarboxylase and [Leu]enkephalin, 12% for both glutamate decarboxylase and [Met]enkephalin, 11-12% for both glutamate decarboxylase and substance P, and 17% for both [Met]enkephalin and substance P. These results provide evidence for chemical heterogeneity within the medium-sized neostriatal neurons, and provide the first evidence for coexistence of glutamate decarboxylase and substance P within a single neuron, and the first evidence for the coexistence for substance P and [Met]enkephalin within single neurons of the central nervous system.

Substance P-immunoreactive neurons in the neocortex of the rat: a subset of the glutamic acid decarboxylase-immunoreactive neurons.

Substance P-immunoreactive (SP-IR) neurons are found throughout the layers of the rat neocortex. Within the somatosensory (SI) cortex these cells make up about 2% of all neurons and within area 17 about 3%. Colocalization studies reveal that the SP-IR neurons are a subset of the population immunoreactive for glutamic acid decarboxylase.

Topographical projections of the cerebral cortex to the subthalamic nucleus.

Corticosubthalamic projections in the rat were investigated using the autoradiographic anterograde axonal tracing technique. After unilateral injections of tritiated amino acids in the cerebral cortex, projections to the ipsilateral subthalamic nucleus (STH) could be found arising only from the frontal agranular cortex and the zone of MI-SI overlap. Injections into granular areas of the cortex (e.g., somatosensory and visual areas) did not result in labeling in STH. Following injections in the frontal agranular cortex, labeling was present in the ipsilateral but not the contralateral STH. In general, injections that involved the lateral agranular field of frontal cortex, as defined by Donoghue and Wise ('82), resulted in a greater amount of labeling in STH than injections within the medial agranular area or the zone of MI-SI overlap. The projection from the frontal agranular areas to STH is topographically organized. The rostral part of the lateral agranular cortex projects to the lateral portion of the rostral two-thirds of STH, and the caudal part of this field projects to the ventral aspect of the middle third of STH. Injections in the rostral part of the medial agranular cortex resulted in labeling throughout the ventral two-thirds of the medial half of STH. The caudal part of the medial agranular cortex projects to the dorsolateral part of the caudal two-thirds of STH. The present results reveal projections from only the frontal agranular cortex and the zone of MI-SI overlap to STH in the rat. The cortico-STH projection is ipsilateral and terminates in a topographical manner in all parts of STH.

Light microscopic analysis of Golgi-impregnated rat subthalamic neurons.

The neuronal morphology of the rat subthalamic nucleus (STH) was studied using Golgi techniques and Nissl stain. The results show that the somatic shapes of STH neurons vary from fusiform to oval or polygonal. Somatic cross-sectional areas vary between 140 microns2 and 440 microns2. Some of the cells have a few somatic spines. Two to six primary dendrites gave rise to tapering daughter dendrites which extend up to 500 microns. These dendrites are sparsely covered with spines. Some distal dendrites and primary dendrites of the STH also bear filiform appendages. Neurons located in the deep portion of the STH have oval dendritic fields whose long axis is parallel to the long axis of the nucleus in frontal or sagittal planes. Some of these neurons have one or two dendrites which cross the borders of the STH into the zona incerta, the lateral hypothalamus, or the cerebral peduncle. Generally, neurons located at the borders of the STH have their dendritic fields extending parallel to the borders and are confined to the nucleus. However, some neurons adjacent to the ventrolateral border of the nucleus have some dendrites extending into the cerebral peduncle. Quantitative analysis of the STH neurons showed a unimodal distribution of somatic sizes as well as the number of primary dendrites. No neurons with obvious Golgi type II characteristics were found. Two types of afferent fibers were observed entering the STH. One type consists of axon collaterals arising from the cerebral peduncle ventrolaterally, or the internal capsule rostrally, while the other enters the nucleus after crossing the internal capsule rostrally. These results suggest that the rat STH is an open nucleus in contrast to other species such as man, monkey, and cat, where it is closed, and that the rat STH may contain only one type of neuron.