Demyelination in experimental beta, beta'-iminodipropionitrile and hexacarbon neuropathies. Evidence for an axonal...

beta, beta'-Iminodipropionitrile and 2,5-hexanediol are best known for their ability to induce axonal pathology, including formation of giant axonal swellings. During studies of the pathology of rats exposed to these agents for long periods, we found extensive recurrent demyelination in the spinal roots. To determine whether the demyelination occurred in response to the axonal disease or whether it represented a direct toxic effect on Schwann cells, we examined the time course and distribution of axonal changes and demyelination, asking whether demyelination correlated with, or was independent of, axonal pathology. Experimental animals were continuously intoxicated with one of these agents, and groups were taken for pathologic examination at intervals of up to 2 years; in both models, the relationship between axonal pathology and demyelination was systematically studied in multiple regions of the L5 spinal roots. In control rats, mild demyelination was present by 14 months and increased with age. By 24 months, untreated animals showed widespread demyelination in the spinal roots; in these animals, there was no predilection for proximal or distal regions of the roots, nor was their evidence of recurrent demyelination. Administration of beta, beta'-iminodipropionitrile produced giant axonal swellings located primarily in the proximal 10 mm. of the ventral root and the distal 10 mm. of the dorsal root. By 12 months of exposure, intramyelinic vacuoles (myelin bubbles) and demyelinated segments were numerous in the same regions. By 24 months, the affected regions contained elaborate onion bulbs. The regions without axonal swellings showed only mild demyelination. In contrast, in the 2,5-hexanediol group, giant axonal swellings and axonal degeneration began distally and progressed more proximally with time. By 15 months, when axonal swellings were present primarily in the distal ventral root, there were numerous myelin bubbles. By 24 months, onion bulbs, predominantly involving the distal ventral roots, had developed. Semiquantitative analysis of the time course and distribution of demyelination in these toxic models showed a relationship between axonal abnormalities and subsequent development of demyelination. We concluded that changes in the axons contribute to the development of demyelination in these models and determine the distribution of the lesions. These experimental neuropathies provide models for studies of the stimulus and mechanisms of secondary demyelination.