Files in this Data Supplement:
Fig. S1. Aldh1L1-GFP can be used for flow cytometric sorting of acutely dissociated spinal cord astrocyte populations. (A) 17.6% of P6 spinal cord cells are Aldh1L1-positive. (B) No co-labelling of GFP-positive cells with the pan-haematopoetic and microglial marker CD45. (C) Substantial overlap of the Aldh1L1-positive population with GFAP expression is observed (upper right corner of the plot indicates co-expression of both markers), with varying degrees of staining. Little to no GFAP positivity is seen in the GFP-negative population.
Fig. S2. Developmental epoch of astrocyte proliferation versus oligodendrocyte proliferation profile. (A) Timetable of thymidine analogue injection at five different periods is shown. The animals were then analysed at P15-18 to reveal the percentage of thymidine analogue-positive Aldh1L1-GFP cells versus thymidine analogue-positive OLIG2+ cells. (B,C) Representative pictures of BrdU+ (B) or EdU+ (C) Aldh1L1-GFP cells.
Fig. S3. The activity of PI3-K signalling pathway does not correlate with proliferative astrocyte precursors at embryonic stage. Spinal cord sections from E16 embryos are shown. Two activity readouts of PI3-K pathway, p-AKT and p-S6, were stained and largely absent in Aldh1L1-GFP+ cells.
Fig. S4. Conditional activation of BRAFV600E results in overproduction of Aldh1L1-GFP+ cells at late embryonic/early neonatal stages. (A) Activation of Braflox/lox by Blbp-lacZ-cre results in specific downregulation of BRAF in spinal cord but not in the dorsal root ganglion. (B) Activation of BRAFV600E does not increase apoptotic rate assessed by activated Caspase-3 staining. (C) Astrocyte culture from postnatal day 1 pups of hGFAP-cre; BrafCA shows increased proliferative rate compared with control pups. (D) In hGFAP-cre; BrafCA spinal cords, the activity of pERK1/2 is diminished at later postnatal stage (P4). Error bars represent s.d.
Fig. S5. EGFR transcript expression in Aldh1l1-GFP-positive cells. qPCR showing relative EGFR expression in flow-cytometrically sorted Aldh1l1-GFP positive populations at E13.5, E14.5, E17.5 and P6 (data represent three biological replicates per time point). The result was normalized to E13 expression levels and β-actin was used as internal control. Error bars represent s.d.