Fig 1.
Fractionation of PAs from the cerebral cortices of adult Mlc1-EGFP mice.
(A–C); Sagittal brain sections through the cerebral cortices of P30 mice were double immunolabeled with anti-GFP (A) and anti-GFAP (B) antibodies. Note the co-expression of GFP and GFAP in astrocytes (C). Scale bars, 50 μm (A) and 10 μm (inset, C). (D); Flow cytometric analyses of dissociated cerebral cortical cells from Mlc1-EGFP mice reveals that a low percentage of cells express EGFP (EGFP+). Note that the vast majority of cells do not express EGFP (Negative).
Fig 2.
Analysis of gene expression in PAs isolated from the cerebral cortices of adult Mlc1-EGFP mice.
(A); Schematic summarizing the experimental strategies for isolating cells from cerebral cortices of adult (P30) Mlc1-EGFP knock-in mice. Brain tissue was dissociated and live PAs were fractionated based on EGFP expression using FACS-based approaches. Single cells were subsequently analyzed using quantitative RNA sequencing. (B); Unsupervised clustering of single cell RNA sequencing data using EGFP-expressing PA cells fractionated from mouse cerebral cortices. Cells are clustered based on shared similarities in mRNA expression, with each of the two main clusters representing a subpopulation of PAs showing differential, but somewhat overlapping gene expression. (C); Feature plots showing the expression of select genes with established roles in mouse brain astrocytes including Mlc1 Aldh1l1, and Slc1a3. Each dot indicates a distinct mRNA sequencing result from a single cell. Red indicates higher gene expression and grey indicates lower gene expression.
Fig 3.
Analyzing PA and non-PA cell distribution in adult Mlc1-EGFP;GLAST-DsRed mice based on EGFP and DsRed reporter expression.
(A-E); Coronal brain sections through the cerebral cortices of an adult Mlc1-EGFP; GLAST-DsRed mouse brain were labeled with DAPI (A), anti-GFP to detect PAs (B), anti-RFP to detect DsRed-expressing PAs and non-PAs (C), and anti-CD31 to detect vascular ECs that comprise blood vessels (D). Note that cells adjacent to cortical blood vessels (PAs) co-express GFP and DsRed (arrows in E), whereas DsRed+ cells (non-PAs) that are more distal to CD31+ ECs lack GFP expression. Scale bars, 100 μm (A) and 20 μm (inset, E). (F-J); Coronal sections through the cerebellum of an adult Mlc1-EGFP;GLAST-DsRed mouse brain stained with DAPI (F), anti-GFP (G), anti-RFP (H), and anti-CD31 (I). Note that most if not all cerebellar Bergmann glial cells, which are a radial glial-like astrocyte sub-population, co-express GFP and DsRed (arrows in J). Scale bars, 100 μm (E) and 20 μm (inset, J). (K-O); Coronal sections through the hippocampus of an adult Mlc1-EGFP;GLAST-DsRed mouse brain labeled with DAPI (K), anti-GFP (L), anti-RFP (M), and anti-GFAP (N). Note that many hippocampal astrocytes co-express GFP, DsRed, and GFAP (arrows in O). Scale bars, 100 μm (K) and 20 μm (inset, O).
Fig 4.
Quantitation of PAs and non-PAs within the cerebral cortices of Mlc1-EGFP;GLAST-DsRed adult mice.
(A-C); Coronal brain sections through the cerebral cortices of adult Mlc1-EGFP;GLAST-DsRed mice were immunolabeled with anti-GFP antibodies (A) to visualize Mlc1-expressing cells (PAs) and anti-RFP antibodies (B) to visualize cells expressing DsRed via the Slc1a3 promoter (PAs and non-PAs). In the merged confocal image (C), note that some cells express both EGFP+ and DsRed+, whereas other cells are single positive for each fluorescent marker. Scale bar, 50 μm (A) and 10 μm (inset, C). (D); Quantification of the EGFP+ and RFP+ single positive cells as well as EGFP+/RFP+ double positive cells in the cortices of Mlc1-EGFP;GLAST-DsRed mice. Numbers of fluorescence positive cells are plotted as a percentage of total numbers of cells based on DAPI labeling, although DAPI images have not been included in this figure. Note that more than 20% of cells within the adult cerebral cortex are EGFP+ or EGFP+/RFP+, whereas non-PAs constitute approximately 20% or more of total cells. The red bars indicate DsRed+/EGFP- cell fractions.
Fig 5.
Quantitative RNA sequencing to identify differentially expressed genes in PAs versus non-PAs isolated from Mlc1-EGFP;GLAST-DsRed adult mice.
(A); Strategy for dissection and analysis of cells from cerebral cortices of Mlc1-EGFP;GLAST-DsRed adult mice. Non-PAs (DsRed+ cells) and PAs (EGFP+/DsRed+ cells) were fractionated using FAC-based sorting approaches followed by comparative RNA sequencing. (B); Flow cytometry plot showing the profile of sorted DsRed+ single positive cells (non-PAs) and EGFP+/DsRed+ double positive cells (PAs) from the cerebral cortices. Note that the majority of cells are DsRed+ and nearly all EGFP+ cells are double positive for DsRed. The few cells in the lower box are EGFP+ single positive PAs that lack DsRed expression. (C, D); Quantitative RNA sequencing comparisons reveal differentially expressed genes in PAs versus non-PAs as revealed by a color-coded heat map (C) and a volcano plot (D). Red indicates higher mRNA expression levels and blue indicates lower expression in PAs versus non-PAs. Differentially expressed genes were identified using the EdgeR package with adjusted p-value cutoffs <0.05 and log2 fold changes > 2. (E) Independent validation of differential gene expression using mRNA isolated from PAs (EGFP+/DsRed+) and non-PAs (EGFP-/DsRed+). Note that the select genes analyzed show enrichment in PAs as compared to non-PAs, as revealed by qRT-PCR. Error bars indicate SE of the mean, n = 3. The red bars indicate EGFP-/DsRed+ cell fractions. Expression differences between sorted cells were determined using two-way analysis of variance with Bonferroni post-hoc test, *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001.
Fig 6.
Correlation analyses of bulk RNAseq and scRNAseq data.
Expression of genes in PAs (EGFP+) fractionated from Mlc1-EGFP mice and analyzed by single cell RNAseq (cell clusters 1 and 2, see Fig 2) were compared to gene expression patterns of PAs (EGFP+/DsRed+) and non-PAs (DsRed+) isolated from Mlc1-EGFP;GLAST-DsRed mice (n = 4). Note that gene expression patterns in purified PAs based on EGFP+ or EGFP+/DsRed+ expression correlate with each other, but do not correlate with non-PA expression patterns. Single cells from the scRNAseq analyses are shown along the y-axis. Heatmaps were colored using Pearson correlation coefficients between the gene expression of single cells in rows (y-axis) and expression of bulk samples (PA and non-PA samples) in columns (x-axis).