SOXF factors regulate murine satellite cell self-renewal and function through inhibition of β-catenin activity

Abstract

Muscle satellite cells are the primary source of stem cells for postnatal skeletal muscle growth and regeneration. Understanding genetic control of satellite cell formation, maintenance, and acquisition of their stem cell properties is on-going, and we have identified SOXF (SOX7, SOX17, SOX18) transcriptional factors as being induced during satellite cell specification. We demonstrate that SOXF factors regulate satellite cell quiescence, self-renewal and differentiation. Moreover, ablation of Sox17 in the muscle lineage impairs postnatal muscle growth and regeneration. We further determine that activities of SOX7, SOX17 and SOX18 overlap during muscle regeneration, with SOXF transcriptional activity requisite. Finally, we show that SOXF factors also control satellite cell expansion and renewal by directly inhibiting the output of β-catenin activity, including inhibition of Ccnd1 and Axin2. Together, our findings identify a key regulatory function of SoxF genes in muscle stem cells via direct transcriptional control and interaction with canonical Wnt/β-catenin signaling.

Keywords: SoxF; adult stem cells; developmental biology; mouse; regenerative medicine; satellite cells; self-renewal; skeletal muscle regeneration; stem cells; ß-catenin.

Figure 1.. SoxF genes are induced at onset of satellite cell emergence and regulate adult myogenesis.

(A,B) Expression levels of SoxF genes (Sox7, Sox17, Sox18) in FACS-isolated Pax3^GFP/+ cells from Affymetrix expression analysis (A) and RT-qPCR (B). E, Embryonic day; P, Postnatal day; MO, age in months. (C) Representative immunolabeling of a satellite cell (PAX7+) co-expressing SOX17 on a freshly isolated adult myofiber (T0). Scale bar, 10 μm. Nuclei are counterstained with DAPI. (D) Expression profile of fresh FACS-sorted and cultured satellite cells for quiescence (Pax7), activation/commitment (Myod, Myog), proliferation (Ki67), terminal differentiation (Myh1), and for SoxF (Sox7, Sox17, Sox18) transcripts. Quiesc., quiescence; Prolif., proliferation; Diff., differentiation conditions. n = 3 mice (each quantified in triplicate) for all experiments. Data expressed as mean ± s.e.m.

Figure 1—figure supplement 1.. Minimal CD31+ cell contamination in FACS-isolated skeletal muscle stem cells.

Pax3^GFP/+ trunk muscles from adult mice were digested in a solution of collagenase/dispase, filtered, and immunolabeled for the endothelial cell marker CD31-PE (Phycoerythrin fluorochrome) before FACS. (A) Gating for CD31-PE/GFP. (B) Gating for single cell (SSC-side scatter)/GFP. (C) Histograms and gating for cell number/CD31+ cells in GFP- and GFP+ cell fractions. (D) Graphic illustrating the proportion of CD31+ cells. n = 3. Data expressed as mean ± s.e.m.

Figure 2.. SOXF factors modulate satellite cell behavior.

(A–E–I) Immunofluorescence of satellite cells transduced with SOXF-encoding retroviruses after 72 hr in culture on isolated adult wild type EDL myofibers. SOXF-FL, construct overexpressing SOXF; SOXFΔCt, altered construct lacking the C-terminus (preserving the HMG DNA binding domain); CTRL, encoding just eGFP. GFP marks transduced cells. Nuclei are counterstained with DAPI (blue). Scale bars, 20 μm. (B–D, F–H, J–L) Quantification of the transduced satellite cells illustrated in (A–E–I) for quiescence (PAX7), activation (MYOD), and proliferation (KI67), compared to CTRL. n ≥ 50 fibers/EDL per condition; ≥1000 satellite cells/EDL. Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test: *, p<0.05; **, p<0.01; ***, p<0.001, compared to CTRL.

Figure 2—figure supplement 1.. SoxF gene function in satellite cell homeostasis.

(A) Protein structure of SOXF constructs. SOXF-FL refers to full-length protein; SOXFΔCt to C-terminal deletion of the protein retaining only the DNA binding site (HMG domain); and SOXFΔBCAT to deletion of the β-catenin binding site, conserving the HMG and transactivation (TA) domains of the protein. aa, amino acids. (B) Quantification of transduced satellite cells on myofibers with SOXF-FL and SOXFΔCt-encoding retroviruses, after 48 hr in culture, treated with EdU for 72 hr. (C) Representative images of satellite cells on myofibers overexpressing SOX17 (SOX17FL) or the mutant SOX17ΔCt, after 72 hr in culture (T72). Scale bar, 20 μm. CTRL, retrovirus econding just eGFP. GFP indicates transduced cells. Nuclei are counterstained with DAPI (blue). (D–F) Quantification of the transduced satellite cells illustrated in (C) showing the effects on differentiation (MYOG; myogenin). n ≥ 50 fibers/EDL per condition; ≥1000 satellite cells/EDL. Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test: ***, p<0.001, compared to CTRL.

Figure 3.. Sox17-knockout during prenatal establishment of satellite cells modifies adult myofiber content and morphology.

(A) Representative Soleus muscle cryosection images of adult control and Sox17 mutant mice. Immunofluorescence was performed with LAMININ to identify the myofibers. Higher magnification is shown in the boxed area. Scale bar, 200 µm. (B–C) Quantification of myofiber number (B) and cross-sectional area in µm² (C). (D) Distribution of the cross-sectional myofiber area in µm². ‘Poly.’, polynomial curve fitting the distribution of myofiber size. (E) Quantification of myonuclei number per 100 fibers in adult Soleus cross-sections from control and Sox17-knockout mice. CTRL, Sox17^GFP/fl; KO, Pax3^Cre/+;Sox17^GFP/fl. n ≥ 4 mice (each quantified in triplicate) for all experiments. Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test: ***, p<0.001, compared to CTRL.

Figure 3—figure supplement 1.. Muscle characterization in control and Sox17-knockout mice.

(A–C) Relative weight of different muscles (muscle weight/total body weight) in two-week-old (P14) (A) and two-month-old adult (B) male mice, and total body weight in grams (C). TA, Tibialis anterior, EDL; Extensor digitorum longus. CTRL, Sox17^GFP/fl; KO, Pax3^Cre/+;Sox17^GFP/fl. n ≥ 4 mice (each quantified in triplicate) for all experiments. Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test: *, p<0.05; **, p<0.01; ***, p<0.001, compared to CTRL.

Figure 3—figure supplement 2.. Muscle characterization in control and Sox17-conditional knockout mice.

(A) Schematic outline of the experimental procedure for tamoxifen (TMX) injection (i.p., intraperitoneal). d, days. (B) Representative images of the histological characterization from adult resting Soleus muscles at d21 after TMX treatment: Hematoxylin and eosin (left panel), Oil red O (middle panel), and Sirius red (left panel) staining. Scale bars, 100 μm. CTRL, Sox17^fl/fl; cKO, Pax7^CreERT2/+;Sox17^fl/fl. n ≥ 3 mice.

Figure 4.. SOX17 is necessary to maintain satellite cell quiescence in adult muscles.

(A,F) Representative Soleus cryosection images showing immunofluorescence for satellite cells (PAX7+, arrows) in Pax3^Cre/+;Sox17^GFP/fl and Pax7^CreERT2/+;Sox17^fl/fl mice, with appropriate controls. Scale bars, 25 μm. Fibers are identified by LAMININ and nuclei are counterstained with DAPI. (B,G) Quantification of satellite cell number during postnatal growth (P14) and in adult. (C) Quantification of the ratio PAX7/MYOD+ satellite cells in P14 Soleus cryosections. (D) RT-qPCR analysis on adult TA muscles for Pax7 and SoxF genes in fresh FACS-isolated satellite cells from control and Sox17-knockout mice. (A–D) CTRL, Sox17^GFP/fl; KO, Pax3^Cre/+;Sox17^GFP/fl. (E) Schematic outline of the experimental procedure for tamoxifen (TMX) injection (i.p., intraperitoneal) in Sox17^fl/fl (CTRL) and Pax7^CreERT2/+;Sox17^fl/fl (cKO) mice. d, days. (E–G) CTRL, Sox17^fl/fl; cKO, Pax7^CreERT2/+;Sox17^fl/fl. Quantification was performed in whole cross-sections. n ≥ 4 mice (each quantified in triplicate) for all experiments. Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test: *, p<0.05; **, p<0.01, compared to CTRL.

Figure 4—figure supplement 1.. Satellite cells characterization of control and Sox17-knockout mice.

(A) Immunofluorescence of satellite cells (MCAD; M-cadherin) in adult Soleus cryosections from control and Sox17 mutant mice. Scale bar, 25 μm. (B) Quantification of satellite cell number illustrated in (A). CTRL, Sox17^GFP/fl; KO, Pax3^Cre/+;Sox17^GFP/fl. n ≥ 4 mice (each quantified in triplicate) for all experiments. Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test: *, p<0.05, compared to CTRL.

Figure 5.. SOX17 regulates adult muscle regeneration after injury in Pax3^Cre/+;Sox17^GFP/fl mutant mice.

(A) RT-qPCR analysis of Pax7 and SoxF genes in satellite cells isolated during CTX-induced regeneration in adult wild type TA muscles. d; days post-injury. (B) Representative images of TA muscles 10 days after CTX injection. Scale bar, 5 mm. (C) RT-qPCR of muscle markers 10 days after CTX injection. (D) Representative images of cryosections from regenerating adult TA muscles seven days after injury showing immunofluorescence for PAX7+ cells (quiescent; arrows) and PH3+PAX7+ cells (proliferating, arrowheads). Scale bar, 25 μm. (E) Quantification of satellite cells as illustrated in (D). (F) Quantification of satellite cells (PAX7+) by the end of the regeneration process (d28-CTX). (G) Representative images of the histological characterization of adult TA muscles seven days after injury with Hematoxylin and eosin (cell infiltration; upper panel), Oil red O (fat infiltration; middle panel), and Sirius red (fibrosis; bottom panel) staining. Insets: enlargement of the indicated regions. Scale bars, 100 μm. CTRL, Sox17^GFP/fl; KO, Pax3^Cre/+;Sox17^GFP/fl. n ≥ 3 mice (each quantified in triplicate) for all experiments. Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test: *, p<0.05, compared to CTRL.

Figure 5—figure supplement 1.. Gene expression profile during CTX-induced regeneration in adult wild type TA muscles and satellite cells.

(A) SoxF transcripts, (B) specific transcripts of satellite cells, and (C) transcripts marking activation and early differentiation profiling of this stem cell population. Total RNA was extracted on days (d) 0–7, 10, 15, 21, and 28 covering initial, intermediate, and final steps of muscle regeneration from whole muscle (A–C) or FACS-isolated satellite cells (d0–d2–d5–d7) (D). n ≥ 3 mice (each quantified in triplicate) for all experiments. Data expressed as mean ± s.e.m.

Figure 5—figure supplement 2.. Impaired clonogenic and regenerative potential of Sox17-knockout muscle stem cells.

(A) Quantification of cells per colony in a clonal assay of FACS-isolated satellite cells, from adult control and Sox17-knockout hindlimb muscles, after four days in proliferation conditions. (B) Distribution of the number of colonies of FACS-isolated satellite cells, from adult control and Sox17-knock hindlimb muscles, as in (A). Poly., polynomial curve fitting the distribution of cell colonies. (C) Schematic outline of the experimental procedure. (D–E) Representative images of the Hematoxylin and eosin (cell infiltration) staining of adult TA muscles 28 days after injury (D), and seven days after second injury (E). Scale bars, 100 μm. CTRL, Sox17^GFP/fl; KO, Pax3^Cre/+;Sox17^GFP/fl. n ≥ 3 mice (each in triplicate) for all experiments. Data expressed as mean ± s.e.m., statistically analyzed with Student’s t-test: **, p<0.01, compared to CTRL.

Figure 6.. SOX17 regulates adult muscle regeneration after injury in Pax7^CreERT2/+;Sox17^fl/fl mutant mice.

(A) Schematic outline of the experimental procedure for tamoxifen (TMX) injection (i.p., intraperitoneal). CTX, cardiotoxin injection; d, days. (B) Representative images of cryosections from regenerating adult TA muscles d7 after injury, showing immunofluorescence for PAX7+ (quiescent, arrows) and PH3+PAX7+ (proliferating, arrowheads) cells. Scale bar, 25 μm. (C–D) Quantification of satellite cells as illustrated in (B). (E) Schematic outline of the experimental procedure for TMX diet. CTX, cardiotoxin injection; d, days. (F) Representative images of cryosections from regenerating adult TA muscles d28 after injury, showing immunofluorescence for PAX7+ (quiescent, arrows) cells. Scale bar, 25 µm. (G) Quantification of satellite cells as illustrated in (F). (H–I) Quantification of the cross-sectional area in µm² (H) and myofiber number per mm² (I). (J–K) Quantification of fat infiltration (Oil red O) (J) and fibrosis (Sirius red) (K) indicated as proportion of the stained section (average of five sections per muscle). (L) Representative images of the histological characterization of adult TA muscles 28 days after injury with Hematoxylin and eosin (cell infiltration; upper panel), Oil red O (fat infiltration; middle panel), and Sirius red (fibrosis; bottom panel) staining. Scale bars, 100 µm. CTRL, Sox17^fl/fl; cKO, Pax7^CreERT2/+;Sox17^fl/fl. n ≥ 3 mice (each quantified at least in triplicate) for all experiments. Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test (C,D,G) and Mann-Whitney ranking test (H–K): n.s., not significant; *, p<0.05; **, p<0.01; ***, p<0.001, compared to CTRL.

Figure 7.. Compensatory effect of SOXF factors in satellite cells on ex vivo culture and in vivo injury-induced regeneration.

(A–C) Quantification of transduced satellite cells with SOXF-encoding retroviruses after 72 hr in culture on EDL isolated myofibers. Adult control satellite cells were transduced with the eGFP-encoding retrovirus (CTRL-RV) and Sox17-knockout cells with CTRL-RV or SOXF-FL. Quiescence (A; PAX7), activation (B; MYOD), and differentiation (C; MYOG) were measured. In red, CTRL vs. KO comparison; in black, KO transduced with CTRL-RV vs. KO transduced with SOXF-FL. n ≥ 30 fibers/EDL per condition; ≥1000 satellite cells/EDL. CTRL, Sox17^GFP/fl; KO, Pax3^Cre/+;Sox17^GFP/fl. (D) Schematic outline of the experimental procedure for electroporation into regenerating TA muscle of wild type mice. CTX, cardiotoxin; d, days. (E) Histology characterization by Hematoxylin and eosin (cell infiltration, top panel), Oil red O (fat infiltration, middle panel), and Sirius red (fibrosis, bottom panel) staining of cryosections from electroporated wild type adult TA muscles five days after injury. TA muscles were electroporated with control (CTRL, left) or dominant negative SOX17 construct (SOX17ΔCt, right). Insets show enlarged images of the indicated regions. Quantification of fat infiltration (Oil red O) and fibrosis (Sirius red) are indicated as proportion of stained area. Scale bars, 100 μm. (F) RT-qPCR analysis seven days after CTX injection. n ≥ 3 mice (≥ 5 different areas). Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test: ns, not significant; *, p<0.05; **, p<0.01; ***, p<0.001, compared to CTRL-RV in CTRL (red asterisks in A-C), CTRL-RV in KO (black asterisks in A-C) or CTRL (E).

Figure 7—figure supplement 1.. Muscle electroporation during injury-induced muscle regeneration.

(A) Schematic outline of the experimental procedure. (B–C) Representative images of electroporated wild type adult TA muscles five (B) and ten (C) days (d) after cardiotoxin (CTX) injection. TA muscles were injected with either pCIG (CTRL; green) or pCIG-expressing a dominant negative SOXF construct (SOX17ΔCt; green) (Figure 7), together with TdTomato (red) to identify the electroporated area. Scale bars, 5 mm. (D) Representative images of electroporated muscle cryosections five (left) and ten (right) days after CTX injection. BF, brightfield. n = 3 mice.

Figure 8.. SoxF genes inhibit β-catenin transcriptional activity to regulate satellite cell behavior.

(A–B) Transactivation of SoxF-B-TKnLacZ (A) and pTOP-TKnLacZ (B) reporters by SOXF and β-catenin in LiCl-treated C2C12 myoblasts. Quantification is expressed as mean of the amount (nmoles) of hydrolyzed ONPG normalized to control (first bar). Comparison of activity with or without β-catenin (A) or with and without SOXF co-expression (B). Relative amounts of transfected DNA are listed below the chart (ng). n ≥ 4 (A); n ≥ 6 (B). (C) Expression profile of β-catenin target genes in adult control and Sox17 mutant TA muscles. Ccnd1, Cyclin-D1. n ≥ 4 mice (each in triplicate). (D) Immunolabeling for β-catenin (β-cat, red) in quiescent (T0, PAX7+, green) and activated (T24, MYOD+, green) satellite cells from adult wild type EDL isolated myofibers. Nuclei are counterstained with DAPI (blue). Scale bar, 50 μm. (E–G) Immunofluorescence of satellite cells transduced with SOXFΔBCAT constructs after 72 hr in culture (T72) in adult wild type EDL isolated myofibers. SOXFΔBCAT, SOXF-encoding retroviruses lacking the binding site for β-catenin; CTRL, encoding just eGFP. GFP indicates transduced cells. Nuclei are counterstained with DAPI (blue). Scale bars, 20 μm. (H–J) Quantification of the transduced satellite cells illustrated in (E–G) for quiescence (PAX7), activation (MYOD), and differentiation (MYOG; myogenin). n ≥ 50 fibers/EDL; ≥1000 satellite cells/EDL. Data expressed as mean ± s.e.m., statistically analyzed with Mann-Whitney ranking test (A–B) or Student’s unpaired t-test (H-J): *, p<0.05; **, p<0.01; ***, p<0.001; ****, p<0.0001, compared to absence of β-catenin (A), presence of β-catenin (B) or CTRL retrovirus (H-J).

Figure 8—figure supplement 1.. Validation of SOXF constructs.

(A–B) Transactivation of SoxF-B-TKnLacZ reporter by SOXFΔBCAT constructs (A; n ≥ 5) and pTOP-TKnLacZ reporter by SOXF constructs (B; n ≥ 4) in LiCl-treated C2C12 myoblasts. Quantification is expressed as mean of the amount (nmoles) of hydrolyzed ONPG normalized to control (first bar). Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test to the respective FL form: ns, not significant; *, p<0.05; **, p<0.01.

Figure 9.. SOXF factors inhibit β-catenin target genes.

(A) Effect of β-catenin stabilizer LiCl in adult wild type EDL myofiber cultures, to analyze satellite cell proliferation rate upon transduction with the indicated retroviral constructs. SOXF-FL, construct overexpressing SOXF; SOXFΔCt, SOXF proteins C-terminal deletions preserving the HMG DNA binding domain. n ≥ 50 fibers/EDL; ≥1000 satellite cells/EDL. (B) RT-qPCR of adult quiescent satellite cells. Pax7 is the marker of this stem cell population. SoxF transcripts were detected but not Ccnd1 (Cyclin-D1). n = 3. (C–D) Fold transactivation of Ccnd1 (Ccnd1-nLacZ) (C; n = 3) or Axin2 (Axin2-nLacZ) (D; n = 4) proximal promoters by β-catenin in C2C12 myoblasts co-transfected with SOX7 constructs in presence versus absence of LiCl. Quantification is expressed as mean of the amount (nmoles) of hydrolyzed ONPG normalized to control (first bar). Comparison is related to β-catenin only transfection. Relative amounts of transfected DNA are listed below the chart (ng). (E) Expression levels of Axin2 in FACS-isolated Pax3^GFP/+ cells from Affymetrix expression analysis. E, Embryonic day; P, Postnatal day; MO, age in months. Data expressed as mean ± s.e.m., statistically analyzed with Student’s unpaired t-test (A) or Mann-Whitney ranking test (C–D): ns, not significant; *, p<0.05; ***, p<0.001, compared to absence of LiCl (A, CTRL), SOX17FL (A, LiCL treated CTRL and SOXFΔCt) or β-catenin only transfection (C-D).

Author response image 1.. Evaluation of cycling and self-renewal status.

Quantification of the cycling (PH3+) satellite cells (PAX7+) at P14. Data expressed as mean ± s.e.m.

Author response image 2.. Effect of Sox17 deletion on myofiber type distribution.

Quantification of the slow-type MyHCI+ myofibers expressed as percentage of all fibers in whole adult Soleus (A), TA (B), and EDL (C) cross-sections from control and Sox17 mutant mice. CTRL, Sox17^GFP/fl; KO, Pax3^Cre/+;Sox17^GFP/fl. n≥4 mice (each in triplicate) for all experiments. Data expressed as mean ± s.e.m, statistically analyzed with Student’s unpaired t-test: ***, p<0.001.

Author response image 3.. Overexpression levels of SOX-FL proteins.

C2C12 and HEK293 cells were transfected with GFP-tagged SOXF constructs for 48h. After lysis, 10 µg of proteins were loaded on a 4-12% gradient acrylamide gel. Overexpressed proteins were probed with anti-GFP antibody (Abcam). Loading is controlled using an anti-TBP antibody (Cell Signaling).