Murine model of Alexander disease: analysis of GFAP aggregate formation and its pathological significance.

Alexander disease is caused by a coding mutation in the glial fibrillary acidic protein (GFAP) gene. The pathological hallmark is the formation of cytoplasmic inclusions within astrocytes known as Rosenthal fibers (RFs), which primarily consist of GFAP and several heat shock proteins. The presence of mutant GFAP would appear to be involved in RF formation; however, overproduction of wild type human GFAP in mouse brain also results in RF formation. Here, we investigated the in vivo conditions leading to formation of RF-like aggregates. We used transgenic mice (mouse GFAP promoter-human GFAP cDNA with R239H mutation) in which the dosage of the GFAP transgene could be manipulated within the same genetic locus. We found that the presence of mutant GFAP per se was insufficient for aggregate formation. Instead, a 30% increase in GFAP content over that in wild type was also required. GFAP aggregates upregulated endogenous GFAP and nestin gene expression, and intermediate filament structure revealed by immunostaining was fragmented under these conditions. However, overall morphology of astrocytes, including their fine processes, was unaffected. In this transgenic animal model, mice did not show megalencephaly, leukodystrophy, or seizure characteristic of Alexander disease with R239H mutation. Nevertheless, their mortality after kainate challenge was dramatically increased, whereas transgenic mice lacking aggregates exhibited mortality similar to that of wild type mice. These results indicate that the presence of GFAP aggregates containing mutant GFAP is not sufficient to induce a major phenotype of Alexander disease, even though it causes some abnormalities in the mouse.