Ptbp1 depletion does not effectively induce the astrocyte-to-neuron conversion in mouse cortex

(A) Schematic workflow to genetically generate the adult astrocyte specific Ptbp1 cKO mouse model and timeline of tamoxifen administration and sample collection. (B) Representative images of mouse cortex collected 12 weeks after tamoxifen injection. White boxes indicate the location of images shown in Figure 1C (Scale bars are 100 μm). (C) Representative immunostaining images of mouse cortex collected 12 weeks after tamoxifen IP injection. White arrowheads in the control panels indicate the expression of PTBP1 in tdTomato+ (tdT+) astrocytes. Yellow arrowheads indicate PTBP1 was successfully depleted in Ptbp1 cKO astrocytes. The absence of NeuN and tdTomato double-positive (NeuN+ tdT+) cells demonstrates no astrocyte-to-neuron conversion with Ptbp1 depletion. Scale bars are 100 μm. (D) Quantification of tdT+ astrocyte proportion at 4, 8, and 12 weeks following tamoxifen induction. (E-F) Quantification of Ptbp1 knockout efficiency in control and Ptbp1 cKO mouse cortex at 4, 8, and 12 weeks following tamoxifen induction. (G) Quantification of PTBP1+ NeuN+ double positive cells indicating PTBP1 is not expressed in neurons. (H) Quantification of NeuN+ tdT+ cells indicating the absence of astrocyte-to-neuron conversion. (I) Quantification of NeuN+ cells at 4, 8, and 12 weeks following tamoxifen induction showing no changes in the proportion of neurons in control or Ptpb1 cKO cortex. Animal numbers are n = 3 for both control and KO groups at all three time points. For quantification, the individual cortical images taken per brain are N=20-24 for 4 weeks, 7-10 for 8 weeks and 10-20 for 12 weeks. The quantification results represent the average and stdev of biological replicates (n). The significance test was carried out by t-test, *p< 0.05, **p< 0.01, ***p< 0.001 and “ns” means no difference with p>0.05.

tdTomato expression in cortical and striatal astrocytes.

(A) Aldh1l1-Cre induced tdTomato+ (tdT+) cells are mostly S100β+ astrocytes in the cortex. (B) Cropped images of (A) show the cell type details in higher magnification. The image locations in cortex are indicated by the white squares in (A). (C) Representative images of striatum from the same animals of (B). Most of the tdT+ cells are S100β+ (white arrowheads). Although there are a few S100β−/tdT+ (violet arrowheads) and S100β+/tdTomato− (yellow arrowheads) cells, they exist in both controls and KOs and have no difference in both groups. The time point is 4 weeks after tamoxifen injection. Animal numbers are n=2 for each group. Scale bars are 100 μm.

PTBP1 is expressed in both astrocytes and microglia but not in neurons in adult mouse brain.

(A) PTBP1 is not expressed in adult neurons. Neurons are visualized by NeuN antibody staining. (B) PTBP1 expression in astrocytes stained by S100β. PTBP1 expression is not exclusive to S100β+ cells indicating its expression in other cell types. (C) PTBP1 expression in both astrocytes (GFAP) and microglia (Iba1). The representative images in (A) and (B) are all single focal plane images and in (C) are z-stack images acquired with 10 μm z and 1 μm/step. Two months old BL6/C57 mice were used in this experiment and n=3. Scale bars are 100 μm.

PTBP1 depletion for 4 and 8 weeks does not induce astrocyte-to-neuron conversion in the cortex.

(A-B) Representative low-magnification immunostaining images of mouse cortex collected at 4 weeks (A), and 8 weeks (B) after tamoxifen IP injection. Scale bars are 100 μm. (C-D) Representative high-magnification images in panels A-B. White arrowheads in the control panels indicate PTBP1-expressing astrocytes (tdT+ cells). Yellow arrowheads indicate PTBP1 was successfully depleted in astrocytes. Scale bars are 100 μm.

Ptbp1 depletion does not induce the astrocyte-to-neuron transition in striatum.

(A) Representative images of control and Ptbp1 cKO mouse striatum collected 12 weeks after tamoxifen injection. White boxes indicate the location of images shown in Figure 2B. Scale bars are 100 μm. (B) Representative immunostaining images of mouse striatum collected 12 weeks after tamoxifen induction. White arrowheads in the control panels indicate the expression of PTBP1 in astrocytes (tdT+ cells). Yellow arrowheads indicate the efficient PTBP1 depletion in Ptbp1 cKO astrocytes. The absence of NeuN+tdT+ cells in the striatum demonstrates no astrocyte-to-neuron conversion with Ptbp1 depletion. Scale bars are 100 μm. (C) Quantification of striatal tdT+ astrocyte proportion at 4, 8, and 12 weeks following tamoxifen induction. (D-E) Quantification of Ptbp1 knockout efficiency in control and Ptbp1 cKO mouse striatum at 4, 8, and 12 weeks following tamoxifen induction. (F) Quantification of PTBP1+NeuN+ double positive cells indicating PTBP1 is not expressed in striatal neurons. (G) Quantification of NeuN+tdT+ cells indicating absence of astrocyte-to-neuron conversion in the striatum. (H) Quantification of NeuN+ cells at 4, 8, and 12 weeks following tamoxifen induction showing minimal changes in the proportion of neurons in control or Ptbp1 cKO striatum. Animal numbers are n=3 for both control and KO groups at all three time points. For quantification, the individual cortical images taken per brain are N=4-6 for 4 weeks, 3-6 for 8 weeks and 4-6 for 12 weeks. The quantification results represent the average and stdev of biological replicates (n). The significance test was carried out by t-test, *p< 0.05, **p< 0.01, ***p< 0.001 and “ns” means no difference with p>0.05.

4 and 8 weeks of PTBP1 depletion does not induce astrocyte-to-neuron conversion in the striatum.

(A-B) Representative immunostaining images of mouse striatum collected at 4 weeks (A) and 8 weeks (B) after tamoxifen induction. White arrowheads in the control panels indicate the expression of PTBP1 in astrocytes (tdTomato+ cells). Yellow arrowheads reveal the efficient depletion of PTBP1 in Ptbp1 cKO astrocytes. (C-D) Representative low-magnification images of mouse striatum collected 4 (C) and 8 weeks (D) weeks after tamoxifen injection. White boxes indicate the location of images shown in panels A-B. Scale bars are 100 μm.

Ptbp1 depletion does not induce the astrocyte-to-neuron transition in substantia nigra.

(A-C) Representative images of the immunostaining results (substantia nigra) of the mouse brains collected at 4 weeks (A), 8 weeks (B), 12 weeks (C) after tamoxifen induction. White arrowheads indicate the locations of astrocytes (tdT+ cells). However, none of the tdT+ cells express either NeuN or TH. The absence of NeuN or TH and tdTomato double positive cells reveals no astrocyte-to-neuron conversion in Ptbp1 cKO. Scale bars are 100 um. Animal numbers are n=3 for both control and KO groups at all the three time points expect for the control group at 8 weeks (n=2).

PTBP1 depletion in substantia nigra 4 weeks after tamoxifen induction.

Representative immunostaining images of mouse substantia nigra collected at 4 weeks after tamoxifen induction. White arrowheads in the control images reveal the expression of PTBP1 in astrocytes (tdT+ cells). Yellow arrowheads indicate PTBP1-depleted tdT+ cells in the Ptbp1 cKO substantia nigra. The absence of PTBP1 and tdTomato double positive cells reveals efficient depletion of PTBP1 in Ptbp1 cKO astrocytes. Scale bars are 100 um. Animal numbers are n=2 for both control and KO groups.

Widespread splicing changes in Ptbp1 cKO astrocytes.

(A) Schematic of the experimental design of bulk RNA-seq. (B) Volcano plot of differential splicing analysis, highlighting the significant exon 2 skipping in Ptbp1 in the cKO samples. Upregulated events are highlighted in pink and downregulated events in green. (C) Bar plot showing the number of DSEs identified in eight types of alternative splicing events. SE: Skipped exon, FIVE: Alternative 5′ prime splice site, THREE: Alternative 3′ prime splice site, MXE: Mutually exclusive exons, RI: Retained intron, AFE: Alternative first exon, ALE: Alternative last exon, MSE: Multiple skipped exons. (D) Genome browser track of the Ptbp1 gene locus and its exon 2 (E2) with RNA-seq signals of control and Ptbp1 cKO samples. (E) Enriched motifs in 3′ spliced sites of differentially spliced skipped exons by Ptbp1 cKO. (F) Gene ontology enrichment analysis of differentially spliced genes in Ptbp1 cKO astrocytes. The coronal section drawing in (A) was created using BioRender (bioRender.com/jnvnpo8).

Fluorescence activated cell sorting of tdTomato+ cells.

(A) Forward and side scatter profiles of a representative cell suspension isolated from Ptbp1+/+;tdT+/+;Aldh1l1-Cre+/− (control) cortex used to select whole cell population. (B) Side scatter area and height profiles of cells from the previous gate in (A) with an overlaid gate to select single cells and exclude doublets. (C) Side scatter and tdTomato fluorescence intensity profiles of single cells from the previous gate in (B) with fluorescence gates showing the selection of tdTomato+ and tdTomato-cells. (D) Histogram of fluorescence emission of the control single cell population. (E) Forward and side scatter profiles of a representative cell suspension isolated from Ptbp1loxP/loxP;tdT+/−;Aldh1l1-Cre+/− (Ptbp1 cKO) cortex used to select whole cell population. F) Side scatter area and height profiles of cells from the previous gate in (E) with an overlaid gate to select single cells and exclude doublets. (G) Side scatter and tdTomato fluorescence intensity profiles of single cells from the previous gate in (F) with fluorescence gates showing the selection of tdTomato+ and tdTomato-cells. (H) Histogram of fluorescence emission of the Ptbp1 cKO single cell population.

Motif enrichment analysis for PTBP1-regulated exons.

(A) Schematic illustrating the regions analyzed for motif enrichment: 200 bp upstream and downstream intronic sequences flanking skipped exons. (B–E) Results of motif enrichment analysis using XSTREME for CU-rich motifs in upstream regions of exons activated by PTBP1 loss (B), upstream regions of exons repressed by PTBP1 loss (C), downstream regions of exons activated by PTBP1 loss (C), and downstream regions of exons repressed by PTBP1 loss (D). Enrichment ratios and q-values (SEA algorithm) are shown below each motif.

Impact of Ptbp1 loss on astrocyte splicing profiles.

(A) PCA of PSI values across control, Ptbp1 cKO astrocytes, and in vitro differentiated neuron samples at various differentiation stages (DIV-8, DIV-4, DIV0, DIV1, DIV7, DIV16, DIV21, and DIV28). (B) Spearman’s correlation analysis of PSI values between control, Ptbp1 cKO astrocytes, and in vitro differentiated neuron samples (DIV0 and DIV28). (C) Scatter plot of dPSI in Ptbp1 cKO astrocytes (cKO vs. Control) against splicing changes in in vitro differentiated neurons (DIV28 vs. DIV0). Splicing events are categorized into eight functional groups (F1-F8). (D) Bar plot showing the number of alternative splicing events in each functional category (F1-F8).

Comparison of splicing profiles between Ptbp1 cKO astrocytes and developmental cortical tissue.

(A) PCA of PSI values for splicing events in control and Ptbp1 cKO samples as well as developmental mouse cortical tissue samples across various developmental stages. (B) Spearman’s correlation analysis comparing gene expression profiles of control and Ptbp1 cKO astrocytes with embryonic (E10) and postnatal (P0) cortical tissue samples. (C) Scatter plot of dPSI values in Ptbp1 cKO astrocytes (cKO vs. Control) versus developmental splicing changes in the cortical tissue (P0 vs. E10). Splicing events are categorized into eight functional groups (F1-F8). (D) Bar plot showing the number of alternative splicing events in each functional category (F1-F8).

Comparison of splicing profiles between Ptbp1 cKO astrocytes and astrocyte-derived neurons.

(A–B) PCA of PSI values for splicing events in control and Ptbp1 cKO astrocytes, as well as astrocyte-derived neurons generated by overexpression of Ngn2 or mutant PmutNgn2, and GFP controls, plotted on PC1-PC2 axes (A) and PC3-PC4 axes (B). (C) Spearman’s correlation analysis comparing PSI values across control, Ptbp1 cKO, and astrocyte-derived neurons. (D) Scatter plot of dPSI values in Ptbp1 cKO astrocytes (cKO vs. Control) versus astrocyte-derived neurons (PmutNgn2 vs. GFP). Splicing events are categorized into eight functional groups (F1–F8) based on direction and magnitude of change. (E) Bar plot showing the number of alternative splicing events in each functional category (F1–F8).

Minimal gene expression changes in Ptbp1 cKO astrocytes.

(A) Volcano plot showing DEGs between control and Ptbp1 cKO astrocyte samples. Upregulated genes are highlighted in pink and downregulated genes in green. (B) PCA of TPM values across control, Ptbp1 cKO astrocytes, and in vitro differentiated neuron samples at various differentiation stages (DIV-8, DIV-4, DIV0, DIV1, DIV7, DIV16, DIV21, and DIV28). (C) Spearman’s correlation analysis comparing TPM values between control, Ptbp1 cKO astrocytes, and in vitro differentiated neuron samples (DIV0 and DIV28). (D) PCA of TPM values across control, Ptbp1 cKO astrocytes, and cortical tissue samples at various developmental stages (E10, E11, E12, E13, E14, E15, E16, and P0). (E) Spearman’s correlation analysis comparing TPM values between control, Ptbp1 cKO astrocytes, and cortical tissue samples (E10 and P0).

Comparison of gene expression profiles between Ptbp1 cKO astrocytes and astrocyte-derived neurons.

(A–B) PCA of TPM values for genes in control and Ptbp1 cKO astrocytes, as well as astrocyte-derived neurons generated by overexpression of Ngn2 or mutant PmutNgn2, and GFP controls, plotted on PC1-PC2 axes (A) and PC3-PC4 axes (B). (C) Spearman’s correlation analysis comparing TPM values across control, Ptbp1 cKO, and astrocyte-derived neurons.

Single-cell RNA-Seq analysis of Ptbp1 cKO astrocytes shows limited astrocyte-to-neuron conversion.

(A) Schematic of the experimental design of single-cell RNA-seq. (B) UMAP plot of all identified cell types based on gene expression profiles. Cells were classified into ten distinct cell types: astrocytes (Astro), excitatory neurons (Exc), inhibitory neurons (Inh), microglia (Micro), immune cells (Immune), oligodendrocytes (OL), endothelial cells (Endo), pericytes (Peri), vascular leptomeningeal cells (VLMC), and ependymal cells (Ependymal). Two excitatory neuron subpopulations, Exc-1 and Exc-2, are highlighted. (C-D) UMAP plots showing the distribution of cells in control (n = 10,851) (C) and Ptbp1 cKO (n = 8,594) (D) samples. (E) The Cre transgene expression projected on the UMAP plot. (F) Dot plot representing the expression of marker genes across identified cell types. (G) Bar plot showing the proportion of each cell type in control and Ptbp1 cKO samples. (H) Bar plot showing the number of Cre-negative and Cre-positive cells in Exc-1 and Exc-2 clusters for control and Ptbp1 cKO samples. The coronal section drawing in (A) was created using BioRender (bioRender.com/s4174fi).