Cells were stained in the dark on ice for 15 min with flow cytometry antibodies. Cells were then washed once with 1× PBS and resuspended in 1× PBS for sorting as described previously11 (link)–13 (link),73 (link). Antibodies used in this study were: PE anti-mouse CD45R/B220 (BD Biosciences, #553089, 1:100), PE anti-mouse TER-119 (Biolegend, #116207, 1:100), PE anti-O4 (R&D Systems, #FAB1326P, 1:100), PE anti-CD105 (eBioscience, #12–1051-82, 1:100), PE anti-CD140a (eBioscience, #12–1401-81, 1:100), PE anti-Ly-6G (Biolegend, #127608, 1:100), PerCP anti-Ly-6C (Biolegend, #128028, 1:100), APC anti-CD45 (eBioscience, #17–0451-83, 1:100), APC-Cy7 anti-CD11c (BD Biosciences, #561241, 1:100), and FITC anti-CD11b (eBioscience, #11–0112-85, 1:100). All cells were gated on the following parameters: CD105negCD140anegO4negTer119negLy-6GnegCD45Rneg. Astrocytes were subsequently gated on: CD11bnegCD45negLy-6CnegCD11cneg. Microglia were subsequently gated on: CD11bhighCD45lowLy-6Clow. Pro-inflammatory monocytes were subsequently gated on: CD11bhighCD45highLy-6Chigh. Compensation was performed on single-stained samples of cells and an unstained control. Cells were sorted on a FACS Aria IIu (BD Biosciences). For sorting of TdTomato+ astrocytes, cells were sorted according to TdTomato fluorescence judged against a wild-type control animal using a yellow-green laser on a FACS Aria IIu.