Different targets of different neurons in nucleus gracilis of the cat.

Efferent projections of neurons in the cat nucleus gracilis to the ventrobasal complex (VB) of the thalamus and the dorsal accessory portion of the inferior olive (IOd) were studied using tracing techniques that exploit neuronal orthograde and retrograde intra-axonal transport processes. These projections were studied in order to determine if the heterogeneity of the morphology, physiology and afferent input of the nucleus gracilis is paralleled by a similar heterogeneity in its efferent projections. In an orthograde study, 3H-leucine and/or 3H-proline were injected into the nucleus gracilis of different subjects in an attempt to label different proportions of large (GREATER THAN 18 MU) AND SMALL CELLS. This procedure permitted study of the efferent targets of the two cell types. The amount of labeling in VB was a constant ratio of the number of large cells in the nucleus gracilis that had incorporated the isotope. This relationship was not observed for the small cells. The amount of labeling in IOd was dependent neither on the number of large nor the number of small cells that had incorporated the isotope. In a retrograde study following extensive injections of horseradish peroxidase (HRP) into the thalamus, only large cells (greater than 18 mu) in the dorsal column nuclei were filled with HRP reaction product. These cells were located predominantly in clusters in the middle zone of the nucleus gracilis as well as rostrally. After injections including (but not confined to) the inferior olive, small cells located in the rostral and middle zones were filled with reaction procduct. A few large cells located ventrally in the middle zone of the nucleus gracilis and rostrally were also filled. Some of these ventrally located large cells may project in a collateral fashion both to the thalamus and to the inferior alive. The results of both the orthograde and retrograde studies are consistent with other evidence on the heterogeneity of the nucleus gracilis. These data strongly support the conclusion that the population of cells in the nucleus gracilis that projects to the thalamus overlaps with but is not identical to the population of cells that projects to the inferior olive.