SRSF1 Is Crucial for Maintaining Satellite Cell Homeostasis During Skeletal Muscle Growth and Regeneration

Abstract

Background: The splicing factor SRSF1 emerges as a mater regulator of cell proliferation, displaying high expression in actively proliferative satellite cells (SCs). In SRSF1 knockout mice (KO) generated via MyoD-Cre, early mortality and muscle atrophy are observed during postnatal muscle growth. Despite these findings, the precise mechanisms through which SRSF1 loss influences SCs' functions and its role in muscle regeneration remain to be elucidated.

Methods: To unravel the exact mechanisms underlying the impact of SRSF1 deficiency SC functions, we employed single-cell RNA sequencing (scRNA-seq) on a mononuclear cell suspension isolated from the newborn diaphragm of KO and control mice. Concurrently, we subjected diaphragm muscles to RNA-seq analysis to identify dysregulated splicing events associated with SRSF1 deletion. For the analysis of the effect of SRSF1 deletion on muscle regeneration, we generated mice with inducible SC-specific Srsf1 ablation through Pax7-CreER. SRSF1 ablation was induced by intraperitoneal injection of tamoxifen. Using cardiotoxin-induced muscle injury, we examined the consequences of SRSF1 depletion on SC function through HE staining, immunostaining and EdU incorporation assay. C2C12 myoblasts and isolated myoblasts were employed to assess stem cell function and senescence.

Results: Utilizing scRNA-seq analysis, we observed a noteworthy increase in activated and proliferating myoblasts when SRSF1 was absent. This increase was substantial, with the proportion rising from 28.68% in the control group to 77.06% in the knockout group. However, these myoblasts experienced mitotic abnormalities in the absence of SRSF1, resulting in cell cycle arrest and the onset of cellular senescence. In the knockout mice, the proportion of Pax7+ cells within improper niche positioning increased significantly to 25% compared to 12% in the control cells (n ≥ 10, p < 0.001). Furthermore, there was an observation of persistent cell cycle exit specifically in the Pax7+ cells deficient in SRSF1 (n = 6, p < 0.001). SRSF1 plays a pivotal role in regulating the splicing of Fgfr1op2, favouring the full-length isoform crucial for mitotic spindle organization. Disrupting SRSF1 in C2C12 and primary myoblasts results in multipolar spindle formation (p < 0.001) and dysregulated splicing of Fgfr1op2 and triggers cellular senescence. Consequently, adult SCs lacking SRSF1 initially activate upon injury but face substantial challenge in proliferation (n = 4, p < 0.001), leading to a failure in muscle regeneration.

Conclusions: SRSF1 plays a critical role in SCs by ensuring proper splicing, maintaining mitotic progression and preventing premature senescence. These findings underscore the significant role of SRSF1 in controlling SC proliferation during skeletal muscle growth and regeneration.

Keywords: SRSF1; cellular senescence; dysregulated splicing; muscle regeneration; satellite cells; scRNA‐seq.

FIGURE 1

scRNA‐seq showed a reduced number of skeletal muscle cells with abnormal distribution in the KO diaphragm. (A) The process involved diaphragm preparation from Srsf1 ^flox/flox (WT) and Srsf1 ^flox/flox ; MyoD‐Cre (KO) mice immediately after birth, single‐cell isolation and the construction of chromium 10× Genomics library for scRNA‐seq analysis. (B) The t‐SNE plot was employed to depict eight distinct cell types within diaphragm muscles, colour‐coded for identification. (C) A heatmap was generated to display the top 10 marker genes for each identified cell type. (D) Specific cell marker gene expression was visualized through t‐SNE plots to distinguish the eight cell types. (E) Individual t‐SNE plots illustrated the distribution of SKM and other cell types in both WT and KO groups. (F) Comparisons were made between the two groups to determine the proportion of each cell type present.

FIGURE 2

The absence of SRSF1 led to an accumulation of activated and cycling myoblasts while impairing myogenic differentiation. (A) A t‐SNE plot displayed six distinct subpopulations of combined WT and KO skeletal muscle (SKM) cells. (B) A violin plot illustrated marker gene expression in the different subclusters. (C) The trajectory plot of SKM cells was colour‐coded by subclusters on the left and pseudotime order on the right. The trajectory was divided into two branches, indicating two distinct cell fates. (D) t‐SNE plots showed WT SKM subclusters in the top panel and KO SKM subclusters in the bottom panel. Different colours represented different subclusters as described in (A). (E) The proportion of non‐cycling MB (sC1), cycling MB (sC2), activated MB (sC3) and differentiated myocytes (sC4 and sC6) was compared between the two groups. (F‐a‐b) Violin plots showed the expression levels of specific genes between WT and KO subclusters.

FIGURE 3

Deletion of SRSF1 induces mitotic abnormalities in cycling myoblasts, leading to cell cycle arrest and cellular senescence. (A) Violin plots illustrate the expression of cell cycle–related genes and SRSF1 in WT subclusters. (B) Violin plots display the expression levels of cell cycle–related genes in activated MB (sC3) between the WT and KO groups. (C) GO enrichment analysis was conducted to identify upregulated and downregulated terms in cycling MB (sC2) of the KO group compared to the control. (D) Violin plots show the expression levels of genes involved in mitotic spindle organization and cell division in cycling MB (sC2) between the WT and KO groups. (E) Violin plots displayed the expression levels of indicated genes involved in cell death and stress in activated MB (sC3) and cycling MB (sC2) between the WT and KO groups.

FIGURE 4

SRSF1‐deficient SCs display improper niche positioning, withdrawal from the cell cycle and elevated apoptosis during the perinatal stage. (A) Hindlimb sections from mice on the first day after birth (P1) were prepared and stained for Pax7 (red), laminin (green) and DAPI (blue). Red arrowheads indicate the satellite cells in the niche location, whereas white arrowheads indicate the interstitial satellite cells. Scale bars, 50 μm. The histograms on the right display the quantification of Pax7 + cells per 0.18 mm² (n = 12 per group) and the ratio of interstitial Pax7 + cells to total Pax7 + cells (n = 10 per group), respectively. (B) Diaphragm sections from P1 mice were prepared and stained for p21 (green), Pax7 (red) and DAPI (blue) (n = 6 per group). The histograms on the right display the quantification of p21 + cells and double Pax7+/p21 + cells, respectively. Scale bars, 25 μm. (C) Immunostaining for TUNEL (green), Pax7 (red), laminin (dark grey) and DAPI (blue) were performed on the hindlimb sections from P1 mice (n = 9 per group). Please note that the merging of red, green and blue results in white. Scale bars, 50 μm. The histograms on the right display the quantification of double TUNEL+/Pax7 + cells per 0.18 mm². (D) qRT‐PCR analysis was conducted to examine the expression of stress‐related and inflammation‐related genes in the WT and KO muscles (n = 6 per group). (E) qRT‐PCR analysis was conducted to examine the expression of mitochondria‐related genes in the WT and KO muscles (n = 6 per group). Results are mean ± SD, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (unpaired Student's t‐test).

FIGURE 5

Knockdown of SRSF1 leads to multipolar spindle formation, cell cycle arrest and the onset of cellular senescence in C2C12 myoblasts. (A) Immunostaining for Ki67 (red) and DAPI (blue) was conducted on C2C12 myoblasts after transient transfection with siRNAs against SRSF1 (siSRSF1‐#1 and siSRSF2‐#2) and control siRNA (siNC). Scale bars, 50 μm. The percentage of Ki67+ cells is shown in the right histogram (n = 5). (B) Immunostaining of phalloidin (red) and DAPI (white) was performed on C2C12 cells transfected with the indicated siRNA. Binucleated cells are indicated by green arrowheads. Scale bars, 50 μm. The quantification of cells containing more than two nuclei is presented on the right (n = 5). (C) Immunostaining of pericentrin (green) and α‐tubulin (red) was carried out on C2C12 cells after transient transfection with the indicated siRNA for 48 h. Scale bars, 10 μm. The quantification of cells with mitotic defects is displayed on the right (n = 4). (D) qRT‐PCR analysis was conducted to assess the expression of genes associated with mitotic spindle assembly in C2C12 cells transfected with the indicated siRNA (n = 6). (E) Cell cycle distribution was monitored by using flow cytometry. The cells were transfected with the indicated siRNAs and synchronized using a double thymidine block. They were then released into thymidine‐free media for 0 h (orange), 4 h (blue) or 10 h (pink). (F‐a) Immunostaining of p21 (green) and DAPI (blue) was conducted in C2C12 cells after transient transfection for 48 h (top panel). Scale bars, 50 μm. The percentage of p21+ cells is shown on the right (n = 9). (F‐b) SA‐β‐gal staining was performed on C2C12 cells (bottom panel). The cells were transfected with the indicated siRNAs, cultured for 5 days and then stained with SA‐β‐gal. Quantification of SA‐β‐gal‐positive cells is displayed on the right (n = 3). Scale bars, 200 μm. Results are mean ± SD, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (one‐way ANOVA followed by Dunnett's multiple comparison test).

FIGURE 6

SRSF1‐mediated Fgfr1op2 splicing plays a pivotal role in governing mitotic spindle formation and cell cycle progression. (A) Immunostaining was conducted on C2C12 cells for α‐tubulin (red), Fgfr1op2 (green) and DAPI (blue). The top images depict interphase, whereas the bottom images show metaphase. Scale bars, 10 μm. (B‐a) Schematic diagrams illustrate the two splice variants of Fgfr1op2: the long isoform (L) containing exon 4 and the short isoform (S) lacking exon 4. (B‐b) RT‐PCR results of two Fgfr1op2 isoforms in control and KO diaphragm muscles or control and SRSF1‐KD cells. The percent Spliced In (PSI) values were displayed at the bottom. PSI is calculated as Inclusion/(Inclusion + Exclusion) %. (C) Immunostaining of pericentrin (green) and α‐tubulin (red) was performed on C2C12 cells after transient transfection with siRNAs targeting Fgfr1op2‐L (Fgfr1op2‐#1 and Fgfr1op2‐#2) and siNC. Scale bars, 10 μm. Quantification of cells with mitotic defects is shown on the right (n = 3). (D) Whole‐cell lysates were prepared from C2C12 cells transiently transfected with indicated siRNAs and subjected to WB analysis with specified antibodies, along with the molecular weight of each protein in the WB analysis. The protein quantification analyses are presented in the Figure S6A. (E‐a‐b‐c) C2C12 cells were co‐transfected with siRNAs and plasmids, followed by immunostaining and WB analysis. The siRNAs and plasmids used were siNC + pcDNA3.0‐HA‐vector (1), siSRSF1‐#1 + pcDNA3.0‐HA (2) and siSRSF1‐#1 + pcDNA3.0‐HA‐Fgfr1op2‐L. Representative confocal images of EdU (red) and DAPI (blue) staining were presented at the top (a). Scale bars, 50 μm. The histogram below (b) illustrates the percentage of EdU + cells (n = 10). Results of WB analysis are depicted in panel (c). The protein quantification analysis of panel (c) is shown in Figure S6B. (F) Immunostaining was conducted on C2C12 cells as described in (E), and quantification of cells with mitotic defects is presented (n = 4). Results are mean ± SD; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (one‐way ANOVA for C, E‐b and F, Dunnett's or Tukey's multiple comparison test).

FIGURE 7

The absence of SRSF1 in adult QSCs impedes their ability to proliferate, leading to a failure of muscle regeneration following injury. (A) Generation of Srsf1 ^flox/flox ; Pax7‐CreER mice, and the experimental strategy involved multiple timelines of interventions, including Tmx injection (as a control with coil oil), CTX injection, EdU injection and collection times of tibialis anterior (TA) muscle samples, both in the absence of injury and after CTX‐induced injury. (B) Representative confocal images of Pax7 (red) and laminin (white) staining in TA muscle sections from control and Tmx‐treated mice on Day 3 after injury. EdU was detected using Alexa‐488 labelling (green) and nuclei were stained with DAPI (blue). Dotted boxes indicated merged EdU+ and Pax7+cells. Scale bars, 50 μm. Quantification of Edu+/Pax7+ cells per area is shown on the right bar graph (n = 4 per group). (C) Representative HE images of TA muscles harvested in control and mutant mice at different time points (absence of injury, 3, 5, 7, 14 and 60 days after CTX injury). The new fibres are indicated by the dark arrowheads. Scale bars, 50 μm. (D) Representative confocal images of Myh3 (red) staining with TA muscle sections from control and Tmx‐treated mice on Day 7 after CTX injury. Scale bars, 50 μm. The bar graph in the lower left shows the quantification of Myh3+ myofibres per field (n = 5 per group). Results are mean ± SD; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (unpaired Student's t‐test).

FIGURE 8

SRSF1 deficiency induces multipolar spindle formation and dysregulated splicing of Fgfr1op2 in primary myoblasts. (A) RT‐PCR showed mRNA levels of SRSF1 and two Fgfr1op2 isoforms in both DMSO‐treated and 4‐OHT‐treated myoblasts. PSI values were presented at the bottom. (B) Whole‐cell lysates from myoblasts treated with DMSO or 4‐OHT were analysed via WB with specific antibodies. The protein quantification analysis is shown at the bottom panel (n = 3). (C) qRT‐PCR evaluated gene expression associated with mitotic spindle assembly in DMSO and 4‐OHT‐treated myoblasts (n = 6). (D) Immunostaining of EdU (red) and DAPI (blue) in DMSO and 4‐OHT‐treated myoblasts. Scale bar, 50 μm. The percentage of EdU+ cells is presented in the histogram on the right (n = 4). (E‐a) Immunostaining of phalloidin (red) and DAPI (white), showing cells with more than two nuclei. Scale bar, 20 μm. The percentage of ≥ 2 nuclei cells is presented in the histogram on the right (n = 4). (E‐b) Immunostaining of pericentrin and α‐tubulin in DMSO and 4‐OHT‐treated myoblasts. Scale bar, 20 μm. Quantification of cells with mitotic defects is shown on the right (n = 3). (F) A working model illustrates the role of SRSF1‐mediated Fgfr1op2 splicing in regulating satellite cell mitosis, which is essential for satellite cell proliferation and expansion during skeletal muscle growth and regeneration. The absence of SRSF1 caused exon 4 skipping of Fgfr1op2, thereby decreasing the generation of functional proteins, leading to disrupted mitosis, cell cycle arrest and premature senescence. Dashed arrows indicate other splicing targets of SRSF1, as well as additional mechanisms beyond splicing regulation that might be involved in the regulation of mitosis progression in SCs. Results are mean ± SD; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (unpaired Student's t‐test).