Slides were evaluated APO866 cell line using a Leica DMR upright microscope equipped for epifluorescence microscopy. ST sections were stained for TH immunoreactivity (IR) using nickel enhancement and slides were scanned as 600 dpi, 8 bit grayscale tiff images

using a CanoScan8400F flatbed scanner with automatic settings disabled. One section at the level of the anterior commissure (−0.26 mm from Bregma) was chosen from each brain for fiber density measurement. Using Fiji software (http://imageJ.nih.gov/ij), each ST was outlined using the freehand tool, processed to correct for background and brightness, converted to a binary image and the number of pixels in the ST determined as a measure of TH-IR fiber density. ST fiber density data are expressed as a percentage of injected vs control ST for each treatment group. Sections of SN were stained for TH-IR without nickel enhancement. For the dose response study, the number of TH-IR neurons was counted in both injected and control SNs in one section from each brain at the level of the medial terminal nucleus accessory optic tract (−5.3 mm from Bregma) using Neurolucida software (Micro Bright Field Biosciences). AZD4547 molecular weight For the efficacy study, unbiased stereology was used to determine the total number of TH-IR neurons in each SN. Seven sections at similar anterior to posterior levels were chosen throughout

the anterior-posterior extent of the SN in every brain. The number of TH-IR cells was counted in injected and control SN using StereoInvestigator™ software (Micro Bright Field Biosciences). Parameters were as follows: grid size (80×80), frame size (175×175), guard zones (3 μm), optical dissector height (10 μm). The Gundersen coefficient of error was ≤0.07 for TH neuronal counts in the

control SN and ≤0.11 for Branched chain aminotransferase TH neuronal counts in experimental SN. For both counting methods, only large TH-IR neurons (greater than ~15 μm in diameter) were counted to avoid counting interneurons or dying neurons. Histology images were collected using a Retiga 4000R digital camera on a Leica DMR upright microscope. Adobe Photoshop CS5 was used to compile multi-photo plate figures. Data were analyzed using Prism software. Kruskal–Wallis one-way ANOVAs followed by Dunn’s post hoc tests were used to compare treatment groups for forelimb behavior analysis. All other data were analyzed using a parametric one-way ANOVA followed by Tukey’s post hoc tests. Statistically significant differences were set at p≤0.05. Data are expressed as mean±SEM. This study was supported by the Department of Defense Neurotoxicology Program (NO06079001) and NIH grants (NS31957 and NS054989 to M.C.B. and T32 NS041234 to C.E.K.), the Harry F. and Elaine M. Chaddick Foundation and the Medical Research Institute Council of Ann and Robert H. Lurie Children’s Hospital of Chicago. The University of Pennsylvania Viral Vector Core is acknowledged for AAV production. The technical assistance of Jianping Xie and Brian Corstange (Ann and Robert H.