Response of septal cholinergic neurons to axotomy.

In the present study we employed quantitative morphometric techniques to assay the response of septal cholinergic neurons following unilateral transection of the fimbria/fornix and supracallosal stria. Analysis of 50-micron-thick tissue sections with a Quantimet 920 image analysis system demonstrated a reduction in ChAT immunoreactivity as early as 1 day following denervation. This decrease was associated with a drop in the number of labeled cells ipsilateral to the lesion and a decrease in the area of cholinergic perikarya on the lesioned and nonlesioned side of the septum. The response at 1 day, however, was transient, and at 4 days the number of labeled neurons was not significantly different from controls. By 8 days we observed a dramatic reduction in the number and size of ChAT-positive cells ipsilateral to the lesion and a reduction in the size of cholinergic perikarya on the contralateral (i.e., nonlesioned) side. These values persisted throughout the remainder of the study. To assess more completely the morphologic response of neurons to axotomy than can be determined in 50-micron-thick tissue sections, we embedded the adjacent immunolabeled tissue section in Epon and then serially sectioned it to a thickness of 0.75-1.0 micron. By using this method, we were able to measure the area, length, and width of the cell, the area of the nucleus and nucleolus, and the position of the nucleus (i.e., eccentricity). Measurements were performed on ChAT-labeled and nonlabeled cells. The results of our studies demonstrate that cholinergic and noncholinergic cells responded to axotomy in a characteristic yet different fashion from each other and that this response could be quantitatively assayed. In general, labeled and nonlabeled cells on the lesioned side of the septum shrink in response to denervation. This shrunken state was reflected in measurements of cellular area, length, width, and nuclear area. Moreover, other measurements of cellular morphology (i.e., area of the nucleolus, position of the nucleus) indicate that none of the neuronal populations examined in the present study displayed morphologic evidence of regeneration. Our results indicate a dramatic loss of cholinergic perikarya ipsilateral to the lesion. Moreover, although a few neurons do persist they do so in a shrunken state. These data provide an essential baseline for the second study in this series, which will evaluate the effect of nerve growth factor on the survival of denervated septal neurons.