Segmental and propriospinal projection systems of frog lumbar interneurons

Spinal interneuronal networks have been implicated in the coordination of reflex behaviors and limb postures in the spinal frog. As a first step in defining these networks, retrograde transport of horseradish peroxidase (HRP) was used to examine the anatomical organization of interneuronal circuitry in the lumbar spinal cord of the frog. Following neuronal degeneration induced by spinal transection and section of the dorsal and ventral roots, HRP was placed at different locations in the spinal cord and the positions of labeled neuronal cell bodies plotted using a Eutectics Neuron Tracing System. We describe four spinal interneuronal systems, three with cell bodies located in the lumbar cord and one with descending projections to the lumbar cord. Interneurons with cell bodies located in the lumbar cord include: (1) Lumbar neurons projecting rostrally. Those projecting to thoracic segments tended to be located in the lateral and ventrolateral gray and in the lower two-thirds of the dorsal horn, with projections that were predominantly uncrossed. Those projecting to the brachial plexus and beyond were located in the dorsal part of the dorsal horn (uncrossed) and in the lateral, ventrolateral, and ventromedial gray (crossed). (2) Lumbar neurons with segmental projections within the lumbar cord. These neurons, which were by far the most numerous, had both uncrossed and crossed projections and were distributed throughout the dorsal, lateral, ventrolateral, and ventromedial gray matter. (3) Lumbar neurons projecting to the sacral cord. This population, which arose mainly from the dorsal horn and lateral or ventrolateral gray, was much smaller than in the other systems. Neuronal density of some of these populations of lumbar interneurons appeared to vary with rostrocaudal level. Finally, a population of neurons with cell bodies in the brachial and thoracic segments that projects to the lumbar cord is described. The most rostral of these neurons were multipolar cells with uncrossed projections, while those with crossed projections were confined almost exclusively to the ventral half of the cord. The distribution of spinal interneurons reported here will provide guidance for future studies of the role of interneuronal networks in the control of movements using the spinal frog as a model system.