Toxic effects of human and rodent variants of alpha-synuclein in vivo.

In Parkinson's disease, abnormal alpha-synuclein (asyn) accumulation leads to the formation of soluble oligomeric species thought to be toxic to cells as well as intraneuronal inclusions. To date, the precise mechanisms leading to aggregation of asyn in the brain is not well-understood. Previous studies in yeast, drosophila, and transgenic mice suggested that a non-A beta component depleted version of human asyn [h-asyn(D70-83)] or human beta-synuclein (h-bsyn), naturally lacking this centrally located hydrophobic region, are less prone to form aggregates in vitro and are expected to be less toxic compared to h-asyn in vivo, although not all experimental studies unequivocally support the latter view. To address this outstanding issue, we directly compared the neurotoxicity of human asyn against that of h-asyn(D70-83), h-bsyn as well as rat asyn using an adeno-associated viral vector to express these proteins in a dose-response study where the vector load was varied over two orders of magnitude. By quantifying the neurodegeneration of rat substantia nigra dopamine neurons here we show that h-asyn, h-bsyn, and h-asyn(D70-83) display comparable neurotoxicity across the vector doses tested. On the other hand, rat asyn and GFP control vectors displayed a different profile, where no detectable neurodegeneration was seen except at the highest vector titer. Thus, the two main conclusions of our study are that (i) deletion of the central hydrophobic region in h-asyn is not sufficient to alter its neurotoxic properties and (ii) expression of the widely used GFP control protein can cause measurable neurodegeneration at high titers.