Glial cell lineage expression of mutant ataxin-1 and huntingtin induces developmental and late-onset neuronal pathologies in Drosophila models.

Background: In several neurodegenerative disorders, toxic effects of glial cells on neurons are implicated. However the generality of the non-cell autonomous pathologies derived from glial cells has not been established, and the specificity among different neurodegenerative disorders remains unknown.

Methodology/principal Findings: We newly generated Drosophila models expressing human mutant huntingtin (hHtt103Q) or ataxin-1 (hAtx1-82Q) in the glial cell lineage at different stages of differentiation, and analyzed their morphological and behavioral phenotypes. To express hHtt103Q and hAtx1-82Q, we used 2 different Gal4 drivers, gcm-Gal4 and repo-Gal4. Gcm-Gal4 is known to be a neuroglioblast/glioblast-specific driver whose effect is limited to development. Repo-Gal4 is known to be a pan-glial driver and the expression starts at glioblasts and continues after terminal differentiation. Gcm-Gal4-induced hHtt103Q was more toxic than repo-Gal4-induced hHtt103Q from the aspects of development, locomotive activity and survival of flies. When hAtx1-82Q was expressed by gcm- or repo-Gal4 driver, no fly became adult. Interestingly, the head and brain sizes were markedly reduced in a part of pupae expressing hAtx1-82Q under the control of gcm-Gal4, and these pupae showed extreme destruction of the brain structure. The other pupae expressing hAtx1-82Q also showed brain shrinkage and abnormal connections of neurons. These results suggested that expression of polyQ proteins in neuroglioblasts provided a remarkable effect on the developmental and adult brains, and that glial cell lineage expression of hAtx1-82Q was more toxic than that of hHtt103Q in our assays.

Conclusion/significance: All these studies suggested that the non-cell autonomous effect of glial cells might be a common pathology shared by multiple neurodegenerative disorders. In addition, the fly models would be available for analyzing molecular pathologies and developing novel therapeutics against the non-cell autonomous polyQ pathology. In conclusion, our novel fly models have extended the non-cell autonomous pathology hypothesis as well as the developmental effect hypothesis to multiple polyQ diseases. The two pathologies might be generally shared in neurodegeneration.