S1Supplemental Data Glia Engulf Degenerating Axons during Developmental Axon Pruning

Supplemental Experimental Procedures labeling session before attenuation of LysoTracker labeling (in 1 hr). Only images from a single session were compared to avoid variability introduced by the labeling or imaging protocol. ConfocalFly Strains Genotypes of flies used in this study are shown in the figure legends Z stacks were flattened by using maximum intensity in Image J. Histograms were obtained from each image, which calculates thein the main text. UAS-HRP::CD2/CyO flies were kindly provided by C.L. and Jean-Paul Vincent [S3]. Other flies used in this study are mean intensity of LysoTracker staining. These calculations likely represent an underestimate of the total intensity as overlappingas previously described [S4, S5]. labeling in the Z plane is obscured when projecting three-dimen-sional images. Because a relative number is derived when compar-Electron Microscopy ing intensity of pixels versus the total number of pixels, we did notStaging of pupae was performed as previously described [S5] and introduce more variability by quantifying an arbitrarily defined MBprocessing for EM was similar to other HRP fusion protein protocols region. We therefore could not avoid LysoTracker staining associ-previously described [S6]. Brains were dissected at appropriate time ated with other events independent of MB axon pruning. However,points in 0.1 % gluteraldehyde, fixed for an additional 15 min at room our estimates suggest that at least 80 % of the staining is associatedtemperature, and washed in 0.1 M PB overnight at 4C, followed with MB pruning in any given image. Mean intensity was then calcu-by application of TSATM Biotin system (NEN) and ABC kit (Vector