Episodic demyelination and subsequent remyelination within the murine central nervous system: changes in axonal calibre.

Exposure of young adult C57BL/6 mice to cuprizone in the diet initiated profound and synchronous demyelination of the corpus callosum, which was virtually complete by 4 weeks of exposure. Interestingly, even in the face of a continued exposure to cuprizone, there was spontaneous remyelination 2 weeks later. This remyelination preferentially involved smaller calibre axons. There was a suggestion of yet another cycle of demyelination (at 10 weeks) and remyelination (at 12 weeks), but by 16 weeks of exposure, the regenerative capacity was exhausted and the animals were near death. The relapsing-remitting pattern suggests this may be a useful model for certain human demyelinating disorders. In contrast to the above chronic model, the corpus callosum from mice exposed to cuprizone for only 6 weeks continued to remyelinate, with 67% of the axons being myelinated or remyelinated at 10 weeks. Interestingly, a significant reduction in the mean value for axonal diameter was observed during acute demyelination. Upon remyelination, however, the axonal calibre distribution returned to near-normal. In contrast, when mice were maintained on a cuprizone diet for 16 weeks, the mean value for axonal diameter was reduced to 60% of normal. These results provide further evidence that the interactions between oligodendrocytes and axons alter axonal calibre.

Distribution of enzyme-bearing cells in GM2 gangliosidosis mice: regionally specific pattern of cellular infiltration following bone marrow transplantation.

Tissue distribution of beta-hexosaminidase was investigated using 5-bromo-4-chloro-3-indolyl N-acetyl beta-D-glucosaminide (X-Hex) as substrate in wild-type mice, four GM2 gangliosidosis model mice (Hexa-/-, Hexb-/-, Gm2a-/- and Hexa-/-Hexb-/-) and Hexb-/- mice that received bone marrow transplantation (BMT). In wild-type mice histochemical localization of beta-hexosaminidase was detected in the perikarya of the majority of neurons, small process-bearing microglial cells, perivascular macrophages, and macrophages in the choroid plexus and leptomeninges. X-Hex positivity was also noted in the renal tubular epithelium and macrophages in the liver and spleen. The staining pattern in the Gm2a-/- and Hexa-/- mice was generally similar to those of wild type, but in these mice, X-Hex stain was also noted in some storage neurons with swollen perikarya. No X-Hex-positive cells were detected in Hexb-/- or Hexa-/-Hexb-/- (DKO) mice. In Hexb-/- mice that received wild-type BMT (Hexb-/- +WBMT), many X-Hex-positive cells were detected in the spleen, and to a far lesser extent, in liver and kidney. In the CNS of these mice, X-Hex-positive cells were largely detected in the leptomeninges and choroid plexus. Some positive cells were also detected, mostly in the perivascular regions of the cerebrum, in particular in the regions of the posterior thalamus, brain stem and spinal cord. Some of X-Hex-positive cells were immunoreactive with Mac-1 and F4/80 antibodies and, thus, were cells of microglia/macrophage lineage. X-Hex-positive staining was not detected in neurons in these mice despite clinical improvement following BMT. This is the first time, as far as we know, that the regional distribution of the donor cells in the CNS has been investigated in a model of neuronal storage disease. Our study indicated that donor-derived cells of microglia/macrophage lineage infiltrated the CNS in a regionally specific manner following the BMT.

Mice deficient in all forms of lysosomal beta-hexosaminidase show mucopolysaccharidosis-like pathology.

Lysosomal beta-hexosaminidase consists of 2 subunits, alpha and beta. Mutations in the alpha-subunit gene cause Tay-Sachs disease, while mutations in the beta-subunit gene cause Sandhoff disease. Mice generated by targeted disruption of either the alpha- or beta-subunit genes displayed the pathological features of Tay-Sachs disease or Sandhoff disease, respectively. In this report we describe the pathologic features of mice that carry both disrupted genes and that are deficient in all forms of beta-hexosaminidase activity. These mice displayed physical dysmorphia and extensive neuro-visceral storage. Neurons in the CNS and PNS contained pleomorphic inclusions in addition to membranous cytoplasmic bodies characteristic of gangliosidosis. Diffuse hypomyelination was also apparent in the CNS. Vacuolated cytoplasm was a conspicuous feature of chondrocytes, osteocytes and renal tubular epithelium on routine hematoxylin and eosin (H&E) -stained sections. Numerous vacuolated cells were also noted in the connective tissue, cornea, heart valves, arterial walls, liver, spleen, skin and throughout other visceral organs. These vacuolated cells stained positive with PAS, colloidal iron and alcian blue, indicating an accumulation of glycosaminoglycans. Furthermore, cultured fibroblasts showed a defect in the degradation of glycosaminoglycans, and glycosaminoglycans were excreted in the urine of these mice (1). Thus, morphological and biochemical features in these mice are consistent with those of mucopolysaccharidosis and demonstrate an essential role of beta-hexosaminidase in the degradation of glycosaminoglycans.

Neuropathology of mice with targeted disruption of Hexa gene, a model of Tay-Sachs disease.

A murine model of Tay-Sachs disease, the prototype of the GM2 gangliosidoses, was produced through the targeted disruption of the Hexa gene encoding the subunit of alpha-hexosaminidase A. The mice were completely devoid of beta-hexosaminidase A activity and accumulated GM2 ganglioside in the CNS in an age-dependent manner. Neurons with membranous cytoplasmic bodies (MCBs), identical to those described in Tay-Sachs disease, were identified in the brain of these mice. The neurons with MCBs were periodic acid-Schiff-positive on frozen sections and immunostained with anti-GM2 ganglioside antibody. However, unlike Tay-Sachs disease in which neurons throughout the brain are affected, the localization of storage neurons in these mice appeared to be limited to certain regions, i.e., cerebral cortex, the hippocampus, amygdala, hypothalamus, mammillary nucleus, etc. Storage neurons were absent in the olfactory bulb, cerebellar cortex and spinal anterior horns. The difference in the distribution of storage neurons suggests a difference of ganglioside metabolism between humans and mice. This model is useful for the study of the pathogenic mechanisms of neuronal storage in Tay-Sachs disease and for the evaluation of therapeutic strategies.