tent of neurons expressing QBP1 in the R6/2 mouse brains was probably insufficient to exert a detectable effect on the phenotypes. On the other hand, in the case of AAV5-Hsp40, inhibition of polyQ protein secretion which should lead to an increase in the number of rescued neurons, likely contributed to its improvement of the neurological phenotypes. Other possibilities may also contribute to their varying XAV-939 effects, for example Hsp40 is more effective than QBP1 in inhibiting polyQ protein misfolding/aggregation, and Hsp40 can also support the degradation of misfolded proteins, while QBP1 cannot. In this study we demonstrate a therapeutic strategy against the polyQ diseases using AAV5-Hsp40, which has great potential for clinical application, since AAVs are safe and are widely utilized in clinical trials. We further suggest a novel therapeutic mode of action of Hsp40, namely suppression of pathogenic polyQ protein secretion from cells, which may consequently suppress its cell-cell transmission. Since the transmission of aggregation-prone proteins is thought to be involved also in other neurodegenerative diseases, Hsp40 may exert a non-cell autonomous therapeutic effect on these other diseases. Elucidation of how Hsp40 inhibits polyQ protein secretion should reveal new therapeutic targets and Non-Cell Autonomous Effect of Hsp40 on polyQ strategies for neurodegenerative diseases caused by aggregationprone proteins. Materials and Methods Viral Vectors Adeno-associated virus type 5 vector plasmids contained an expression cassette with a human cytomegalovirus enhancer/chicken b actin promoter followed by the first intron of human growth hormone, target cDNA , or GFP), and a simian virus 40 polyadenylation signal sequence, all positioned between the inverted terminal repeats of the AAV5 genome. AAV5 vectors were produced using the AAV5 plasmid, the AAV5 helper plasmid containing the rep and cap sequences from AAV5, as well as the pHelper plasmid from the AAV HelperFree System containing the E2A, E4, and VA RNA genes of the adenovirus genome. HEK293 cells were co-transfected with the AAV5 plasmid and two helper plasmids by the calcium phosphate method. Seventy-two h later, the cells were harvested and subjected to three rounds of freeze-thaw lysis. AAV5 vectors were then purified by two rounds of cesium chloride density gradient centrifugation. Vector titers were estimated by quantitative DNA dot-blot hybridization to be,0.21.661013 genome copies/ml. Animals All animal experiments were performed in accordance with the guidelines of the Animal Ethics Committee of the National Institute of Neuroscience, National Center of Neurology and Psychiatry, Japan, and performed in accordance with the guidelines. Mice transgenic for human huntingtin exon 1 with approximately 150 CAG repeats were obtained from the Jackson Laboratory, and maintained on a B6CBAF1 background. Genotypes were analyzed and CAG repeat numbers of the transgenic mice were confirmed to be Non-Cell Autonomous Effect of Hsp40 on polyQ similar by PCR as previously described. Mice were housed on a 12-hour light/dark cycle, with food and water provided ad libitum. At least nine male R6/2 mice per group and wild-type littermate controls were used for the phenotype analyses, and two R6/2 mice were used for the inclusion body analyses. AAV Injections P7 old R6/2 mice were stereotaxically injected with 1 ml of virus solution into the striatum at a rate of 0.1 ml/min using a 10 ml Hamilton syring