Neurotoxic effects of colchicine: differential susceptibility of CNS neuronal populations.

Colchicine injected into the dentate gyrus of the hippocampus in adult rats preferentially destroys dentate granule cells. In the present study, we examine the light- and electron-microscopic correlates of the degeneration and evaluate whether the selectivity is preserved across the range of doses between 0.18 and 25 micrograms. Colchicine in a similar dose range was also injected into the cerebellum, olfactory bulb, striatum and cerebral cortex to examine local and regional differences in susceptibility to colchicine. The morphological changes accompanying degeneration in the dentate gyrus include fragmentation of the granule cell layer, appearance of small dark staining bodies in the cell layer, massive microglial invasion and profound disruption of granule cell axons and dendrites. Electron-microscopic observations suggest that the small dark bodies are probably condensed nuclei. The preferential vulnerability of dentate granule cells following intrahippocampal injection was observed at all doses. At doses between 0.18 and 2.5 micrograms there was little evidence of damage to neurons other than dentate granule cells. At the highest dose tested (25 micrograms) some pyramidal cells of regio superior near the injection site were destroyed, while granule cell destruction extended several mm from the injection site. Injection of 0.5-25 micrograms into the cerebellum resulted in the destruction of both granule cells and Purkinje cells, while cells which appeared to be neurons in the molecular layer were less affected. Following injection of 0.5 microgram into the olfactory bulb, granule cells were extensively destroyed and there appeared to be some loss of mitral cells and an overall shrinkage of the injected bulb. Neuronal destruction in the striatum was observed with colchicine injections ranging from 2.5 to 25 micrograms, but at a given dose, the destruction was less extensive than for any other region tested except cerebral cortex. A possible application of this method and the implications of these results for other investigators using colchicine in the brain are discussed.