Fig. 3: Maturation-associated chromatin remodelling at putative regulatory elements.

a Approach used for MACS enrichment (magnetic bead assisted cell sorting) and ATAC-Seq (assay for transposase-accessible chromatin using sequencing) analysis of immature and mature astrocytes from P4 and 2-month-old cortices, respectively (Astr_P4, Astr_2m), in order to identify stage-specific accessible chromatin regions, i.e. putative regulatory elements. b Strategy for mapping ATAC-Seq peaks to putative target genes. c Genome tracks show chromatin accessibility around the transcription start sites (TSS) of the immature astrocyte-specific gene Sparc and of the mature astrocyte-specific gene Bmp2k. Reduced size of ATAC-Seq peaks in adult vs P4 astrocytes indicates closing chromatin, while increased peak size indicates opening chromatin. d, e Genes regulated during astrocyte maturation (from Fig. 2c) are globally associated with chromatin regions that change in their accessibility between P4 and adult stages (d). This suggests a model whereby changes in enhancer accessibility contribute to changes in gene expression during astrocyte maturation (e). f Transcription factor binding motifs enriched in closing and opening chromatin, identified by de novo motif enrichment analysis. The most enriched motifs are similar to ETS, homeodomain (HOX) and retinoic acid receptor-related orphan receptor (ROR) transcription factor binding motifs. g Expression in cortical astrocytes, based on bulk RNA-Seq analysis in Fig. 2, of selected transcription factors that bind to motifs highlighted in (f). Displayed are all ETS factors enriched in immature astrocytes, and all ROR and HOX factors enriched in mature astrocytes, either in the cortical (bulk) or the striatal (single cell) RNA-Seq dataset. Expression of the factors regulated along pseudotime in striatal astrocytes is shown in Supplementary Fig. 5c. Statistical analysis and data presentation: c Tracks represent merged reads of 3 or 4 replicates (for P4 and 2 m respectively); c, d Differential peaks: DESeq2 analysis, n = 3 (Astr_P4), n = 4 (Astr_2m); two-sided Wald test with Benjamini–Hochberg correction; significance threshold: adjusted p-value ≤ 0.05, absolute log2(fold change) ≥ 1; d Two-sided Wilcoxon Rank Sum test (n = 341 immature vs 268 mature genes with associated differential peaks). Boxplots show: centre line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; points, outliers. g Differential genes: DESeq2 analysis, each n = 3; two-sided Wald test with Benjamini–Hochberg correction; significance threshold: adjusted p-value ≤ 0.05, absolute log2(fold change) ≥1; Shown are log2-transformed mean centred, normalized expression values. See also Supplementary Figs. 4 and 5, Supplementary Data 3–5.