Axotomy affects the retrograde labeling of cervical and lumbar-cord-projecting rubrospinal neurons differently.

The effect of axotomy at cervical and lumbar spinal levels upon the ability of rubrospinal neurons to retrogradely transport tracer was compared. Unilateral rubrospinal tractotomy was performed first at C5 and, after a few days, at C2 vertebral levels. Different retrograde tracers were applied at the lesioned sites right after tractotomy. Tracer applied at C5 labeled both cervical and lumbar-cord-projecting neurons. Tracer applied at C2 also labeled both groups of neurons if performed 2 days after that at C5; however, only cervical-cord-projecting neurons were labeled when it was performed 3 or 5 days after that at C5. In another set of experiments, a T10 tractotomy without tracer application was performed 2 or 5 days prior to the C5/C2, series of tract lesions. When preceded by a T10 lesion 2 days in advance, tracer applied at C5 labeled both cervical and lumbar-cord-projecting neurons. However, a T10 lesion 5 days in advance resulted in the labeling of only cervical-cord-projecting neurons by the tracer applied at C5. In either case, tracer applied at C2 consistently labeled only cervical-cord-projecting neurons, irrespective of the intervals-2, 3, or 5 days-allowed between C5 and C2 lesions. Most neurons labeled from C2 were also double-labeled by the tracer applied at C5. Thus, unlike lumbar-cord-projecting counterparts, cervical-cord-projecting rubrospinal neurons retain the ability to uptake and/or transport retrograde tracer several days following axotomy. This implies that cervical-cord-projecting rubrospinal neurons survive in a different functional state from their lumbar-cord-projecting counterparts following axonal injury.

Axotomy induces retraction of the dendritic arbor of adult rat rubrospinal neurons.

The effect of distal axonal injury on the soma-dendritic morphology of intrinsic central neurons was examined using adult rat lumbar spinal cord-projecting rubrospinal neurons as a model. The soma-dendritic morphology was revealed using an improved Golgi-aldehyde method. Impregnated neurons were reconstructed in the two-dimensional plane for analysis. Four weeks after axotomy, neurons had reduced soma sizes and remained multipolar in shape. Some dendrites were found to end not far from their cell bodies. In addition, no long dendrite was identified following axotomy. Sholl's analysis [The Organization of the Cerebral Cortex. London, Methuen, [1956] revealed that axotomized neurons had fewer dendritic branches than control neurons. Total dendritic length was also reduced. Subsequent analyses showed that the average number of dendritic trunks was not altered however the mean number of terminal branches per dendritic trunk was reduced. The dendritic membrane of the normal neurons was usually smooth with occasional short protuberances on the proximal dendrites and spines on the distal dendrites, which did not change after axotomy. In control neurons, we identified an elaborate type of dendritic structure named dendritic appendage aggregates. These aggregates were located preferentially on terminal dendrites and were classified into three categories according to their complexity. The incidence of occurrence for these aggregates decreased following distal axotomy. These phenomena indicate that rat lumbar spinal cord-projecting rubrospinal neurons retract their distal dendrites in response to distal axotomy. The observed anatomic restructuring following axonal injury is likely to be accompanied by an alteration of afferents which normally synapse on distal dendrites.