RhoA-mediated G12-G13 signaling maintains muscle stem cell quiescence and prevents stem cell loss – Cell Discovery

Functional characterization of different GPCRs in quiescent MuSCs

To systematically determine changes in the expression pattern of GPCRs during the transition of MuSC from quiescence to activation, we analyzed five published bulk RNA sequencing (RNA-seq) datasets^{15,16,17,18,19}. We identified 39 GPCRs, which showed higher expression in quiescent compared to activated MuSCs in at least two datasets (Supplementary Fig. S1a). Subsequent RT-qPCR analysis revealed substantially higher expression of 19 out of these 39 GPCRs in quiescent compared to activated MuSCs (foldchange > 2) (Supplementary Fig. S1b). Immunofluorescence staining of MuSCs localized on freshly isolated single myofibers confirmed the presence of seven GPCRs in the plasma membrane of quiescent MuSC (Supplementary Fig. S1c, d). We corroborated the enrichment of CalcR and S1P₃ in quiescent MuSCs^13,14,20, but also detected GPCRs that have not been described in quiescent MuSC before, such as ET_B and NTS₂ (Supplementary Fig. S1c, d). Localization of the remaining GPCRs could not be validated in the plasma membrane, since suitable antibodies were not available.

Expression of a GPCR does not necessarily prove a physiologically relevant function. Thus, we combined the expression analysis with an unbiased pharmacological screen²¹, utilizing a customized GPCR compound library consisting of 259 compounds, targeting 24 subfamilies of GPCRs, including both natural and chemical agonists and antagonists (Supplementary Fig. S2a). We reasoned that manipulation of GPCRs or ligand‒GPCR pairs, functionally important for inducing or maintaining quiescence, should alter activation of MuSC in vitro. To assess MuSC activation, we performed EdU-incorporation assays and quantified the ratio of quiescent PAX7⁺MYOD^‒ to all PAX7⁺ cells. (Fig. 1a). Cut-offs for potential hits were defined by activities of Oncostatin M (OSM) (Fig. 1b, c), which induces quiescence of MuSCs¹⁸, and FGF2 and IGF1 (Fig. 1b, c), which induce proliferation and myogenic differentiation of MuSCs, respectively^22,23. We identified 15 GPCR subfamilies, whose manipulation changed quiescence or activation of MuSCs, including adrenergic receptors (ARs), Cannabinoid receptors (CB1 and CB2), metabotropic glutamate receptors (mGluRs), and Thyroid-stimulating hormone receptor (TSHR) (Supplementary Fig. S2b). Furthermore, we found that ET-3 and NT prevented expansion and myogenic commitment of MuSC at low dosages (Fig. 1d). Similar results were obtained when MuSCs attached to isolated single myofibers from mouse flexor digitorum brevis (FDB) muscles were treated with ET-3 or NT (Supplementary Fig. S2c, d). The strong activity of ET-3 and NT is consistent with the presence of the corresponding receptors ET_B and NTS₂ in the plasma membrane of quiescent MuSCs (Supplementary Fig. S1c).

a Schematic representation of the GPCR compound screening strategy. b, c Scatter plots of EdU⁺ (b) and PAX7⁺MYOD^‒ (c) MuSC ratios following 5-day treatment with 259 synthetic and natural GPCR compounds (10 μM, GPCR antagonists (red) and agonists (blue)). Green dashed lines: threshold for positive selection (mouse OSM treatment); red dashed lines: negative selection (FGF2 or IGF1 treatment); black lines: DMSO-treated MuSCs. d Inhibition of proliferation (upper panel) and myogenic activation (lower panel) of MuSCs by ET-3 and NT (1 nM–1 µM). e, f Immunofluorescence (e) and quantification of GFP⁺ nuclei and Dystrophin⁺ myofibers (f) in transverse sections of Dmd^mdx-4Cv/Y TA muscles, engrafted with freshly isolated MuSC (FSC) or cultured in the presence of DMSO, ET-3, and NT for 5 days, after 3 consecutive CTX-induced injuries. The cartoon in e represents the schematic outline of MuSC engraftment (n = 3). g Immunofluorescence for CalcR (red), ET-3/NT (green), and DAPI (blue) on transverse sections of TA muscles. h Western blot analysis for ET-3/NT in uninjured (U) TA muscles and injured TA muscles at D2, D7, and D14 after CTX injury. Quantification is shown in the lower panel (n = 3). The data represent means ± SEM, analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test (f, h). Scale bars: 20 µm in e, 5 µm in g.

Full size image

We next tested whether ET-3 or NT is able to retain stem cell properties of MuSCs, which are normally lost during expansion in vitro, reducing the ability of expanded MuSC to contribute to skeletal muscle regeneration in a mouse model of Duchenne muscular dystrophy (Dmd^mdx-4Cv/Y mice)^24,25. MuSCs were isolated from Pax7nGFP^+/+ mice, specifically labeling nuclei of PAX7⁺ cells with GFP, and cultured for 5 days with and without the presence of ET-3 or NT. Afterward, cultured MuSCs were grafted into the tibialis anterior (TA) muscle of 18 Gy-irradiated hindlimbs of Dmd^mdx-4Cv/Y mice, shortly after the first three consecutive cardiotoxin (CTX) injections (Fig. 1e). We also used freshly isolated MuSCs, which typically show a higher regenerative capacity than cultured cells^7,26. Self-renewal and regenerative capacity of transplanted MuSCs were examined by counting the number of GFP-positive nuclei and dystrophin-positive myofibers after completion of regeneration. Treatment with ET-3 or NT enabled transplanted MuSCs to massively increase the numbers of dystrophin-positive myofibers and GFP-positive nuclei within the host tissue compared to untreated controls, although the efficiency of freshly isolated MuSCs was not fully reached (Fig. 1e, f). The results suggested that NT or ET-3 treatment promotes engraftment of cultured MuSC as indicated by higher numbers of Pax7-GFP⁺ cells and also improves the contribution of transplanted MuSC to regenerating muscle fibers in mdx muscles.

ET-3 and NT are niche-derived factors, preventing premature activation of MuSCs in vivo

To investigate the origin of ET-3 and NT from cells in the MuSC niche, we first consulted the “scmuscle” database (scRNAseq.org), a recently constructed omics database that provides information about transcript levels in single cells of skeletal muscles under various conditions²⁷. According to the scmuscle database, expression of Edn3 (the gene encoding for ET-3) is highest in quiescent MuSCs compared to other cells in the skeletal muscle, implying an autoregulatory loop to keep MuSCs in quiescence (Supplementary Fig. S2e). In contrast, Nts (the gene encoding for NT) is mainly found in endothelial cells (Supplementary Fig. S2e), consistent with prior findings²⁸. RT-qPCR and immunofluorescence analyses confirmed that Edn3 is expressed in quiescent but not in activated MuSCs, whereas Nts is expressed in lymphatic and blood endothelial cells, directly adjacent to quiescent MuSCs (Fig. 1g; Supplementary Fig. S2f‒i). Specificity of antibodies against NT was verified by western blot analysis of lymph endothelial cells after siRNA-mediated knockdown of NTS (Supplementary Fig. S2j, k). Both ET-3 and NT are strongly downregulated but gradually restored towards the final stages of muscle regeneration (Fig. 1h). Interestingly, we also observed a substantial decline of Edn3, Ednrb, and Ntsr2 expression in aged (24-month-old) MuSCs (Supplementary Fig. S2l, n). The expression of ET-3 and NT in the MuSC niche as well as their decline following muscle injury, suggest a role of ET-3 and NT in regulating stem cell quiescence. Similarly, the reduced expression of Edn3, Ednrb, and Ntsr2 during aging may contribute to the reduced quiescence of aged MuSCs.

Consistent with the results of our initial GPCR expression profiling, subsequent RT-qPCR and immunofluorescence analyses revealed high expression of the receptors for ET-3 and NT, ET_B and NTS₂ in quiescent but not in activated MuSCs (Fig. 2a‒c). We also found that treatment of cultured MuSC with ET-3 and NT increases expression of Ednrb, and Ntsr2, respectively, indicating the existence of a positive feedback loop, which secures enhanced Ednrb and Ntsr2 expression in MuSCs when regeneration is completed (Supplementary Fig. S2o, p). The second receptor for NT, NTS₁, is neither expressed in quiescent nor in activated MuSCs (Fig. 2a‒c). To study the roles of ET_B and NTS₂ in the regulation of MuSC quiescence, we utilized selective antagonists for ET_B (BQ788) and NTS₂ (SR142948A). Treatment of MuSCs attached to isolated single FDB myofibers with BQ788 or SR142948A did not change proliferation of MuSCs or the ratio of PAX7⁺MYOD⁺ to all PAX7⁺ MuSCs, probably due to low concentration of ET-3 and NT in the experimental set-up (Fig. 2d, e). However, treatment with BQ788 after prior administration of ET-3, abrogated inhibitory effects of ET-3 on MuSC proliferation and MYOD expression (Fig. 2f, g). Similar consequences were observed when SR142948A was added to the growth medium of single FDB myofibers treated with NT (Fig. 2f, g). Combined treatment with ET-3 and NT further increased the ratio of PAX7⁺MYOD^‒ relative to all PAX7⁺ MuSCs compared to single treatments, although no synergistic effects were observed. This might be due to the rather artificial experimental set-up, providing higher concentrations of ET-3 and NT than normally available to MuSCs in vivo (Fig. 2g). To validate the roles of ET_B and NTS₂ for maintaining MuSC quiescence in vivo, we injected BQ788 or SR142948A into TA muscles of WT mice. After 7 days of exposure to either BQ788 or SR142948A, 33% of MuSCs in BQ788-treated mice and 65% in NT-treated mice were located outside the basal lamina in the interstitial space, indicating MuSC activation. Likewise, the ratios of KI67⁺PAX7⁺ and MYOD⁺PAX⁺ to all PAX7⁺ cells increased dramatically, whereas the overall numbers of PAX7⁺ cells did not change after treatment with BQ788 or SR142948A (Fig. 2h‒l). We concluded that activation of ET_B and NTS₂ is essential to keep MuSC arrested in quiescence within the niche.

a RT-qPCR analysis of Ednrb, Ntsr1, and Ntsr2 expression in freshly isolated MuSCs (FSC) and activated stem cells (ASC) (n = 3). b Representative images of immunostaining for ET_B/NTS₁/NTS₂ (red), PAX7 (green) and DAPI (blue) on FDB myofibers of WT mice at 0 h (upper panel) and 24 h (lower panel) of culturing. c The quantification of GPCR staining intensity (n = 3). d, e Quantification of the ratios of PAX7⁺KI67⁺ (d) and PAX7⁺MYOD^‒ (e) MuSCs on FDB myofibers after 24 h culturing in the presence of DMSO, BQ788 or SR142948A (n = 6). f, g Quantification of the ratio of PAX7⁺KI67⁺ (f) and PAX7⁺MYOD⁻ (g) MuSCs on FDB myofibers after 24 h culturing in the presence of DMSO, ET-3, NT, ET-3 + BQ788, NT + SR142948A, or ET-3 + NT (n = 6). h‒l Representative images (h) and quantification of PAX7⁺ (i), the ratio of PAX7⁺KI67⁺ (j), and PAX7⁺MYOD⁺ (k) cells, and quantification of MuSCs outside the basal limina (l) on transverse sections of WT mice TA muscle after intramuscular injection of solvent, BQ788 (1 mg/kg bodyweight), or SR142948A (0.5 mg/kg bodyweight), respectively (n = 3). The cartoon in h depicts the outline of the experimental design. The data represent means ± SEM, analyzed by unpaired t-test (a, c) and one-way ANOVA with Bonferroni’s multiple comparisons test (d‒g and i‒l). ND not detected. Scale bars: 5 µm in b, 10 µm in h.

Full size image

G₁₂-G₁₃ integrate signals from different GPCRs to maintain quiescence of MuSCs

Upon ligand binding, GPCRs couple to different G-proteins, enabling them to activate various intracellular signaling pathways. Signals from distinct GPCRs may also converge on the same G-protein subtype, which can integrate signaling events initiated by multiple ligands. To investigate which G-protein is employed by NTS₂ and ET_B to promote MuSC quiescence, we focused on G₁₂-G₁₃ and G_q-G₁₁, since either of these G-proteins has been reported to interact with NTS₂ and ET_B in different physiological processes^29,30. Because the functions of G₁₂ and G₁₃ as well as of G_q and G₁₁ strongly overlap and single inactivation of G₁₂, G₁₃, G_q, and G₁₁ often does not yield clear effects¹², we focused on G₁₂-G₁₃ and G_q-G₁₁ compound mutants. Inactivation of Gna12–Gna13 but not of Gna11–Gnaq by adenoviral transduction of Cre recombinase into isolated MuSCs from Gna11^‒/‒Gnaq^fl/fl and Gna12^‒/‒Gna13 ^fl/fl mice (Fig. 3a), followed by treatment with ET-3 or NT, strongly increased the ratio of EdU⁺PAX7⁺ to all PAX7⁺ MuSCs (Fig. 3b, d). Accordingly, the number of PAX7⁺MYOD^‒ to all PAX7⁺ MuSCs declined only in Cre adenovirus-transduced MuSCs from Gna12^‒/‒Gna13^fl/fl but not from Gna11^‒/‒Gnaq^fl/fl mice (Fig. 3c, e). We also used the same experimental model to investigate potential quiescence-promoting effects of S1P₃, a GPCR that mediates sphingosine-1-phosphate signaling, and BK₁, a GPCR that mediates bradykinin signaling. Both GPCRs are enriched in quiescent compared to activated MuSCs. Treatment of MuSCs on isolated myofibers with CYM5541 and des-Arg⁹-Bradykinin (Des-Arg⁹-BK), agonists for S1P₃ and BK₁, respectively, reduced the ratios of KI67⁺PAX7⁺ and increased the ratios of PAX7⁺MYOD^‒ to all PAX7⁺ cells to a similar extent as ET-3 and NT (Supplementary Fig. S3a, b). Likewise, inactivation of G₁₂-G₁₃ abrogated effects of CYM5541 and Des-Arg⁹-BK on MuSCs (Fig. 3f, g). Since only inactivation of G₁₂-G₁₃ but not of G_q-G₁₁ abrogated the responsiveness of MuSCs to ET-3 and NT (Fig. 3b‒e), we concluded that G₁₂-G₁₃ is mandatory for enabling signaling by ET_B and NTS₂, although it is possible that inactivation of G₁₂-G₁₃ prevents acquisition of quiescence irrespective of specific ligands.

a Schematic experimental outline of b‒g. b, c Ratio of PAX7⁺EdU⁺ (b) and PAX7⁺MYOD^‒ (c) MuSCs isolated from Gnaq^‒/‒Gna11^fl/fl mice infected with Ad-Null or Ad-Cre virus and cultured in the presence of DMSO, ET-3, or NT for 5 days (n = 5). d, e Ratios of PAX7⁺EdU⁺ (d) or PAX7⁺MYOD^‒ (e) MuSCs isolated from Gna12^‒/‒Gna13^fl/fl mice infected with Ad-Null or Ad-Cre virus and cultured in the presence of DMSO, ET-3, or NT for 5 days in vitro (n = 5). f, g Ratio of PAX7⁺EdU⁺ (f) and PAX7⁺MYOD^‒ (g) MuSCs isolated from Gna12^‒/‒Gna13^fl/fl mice infected with Ad-Null or Ad-Cre virus and cultured in the presence of DMSO, CYM5541, or des-Arg⁹-Bradykinin (des-Arg9-BK) for 5 days (n = 5). h, i Response curves to ET-3 for activation of G₁₂(134)-RLuc8/Gβ3γ9-GFP2 (h) and G₁₃(126)-RLuc8/Gβ3γ9-GFP2 (i) BRET biosensors in control (vector) or EDNRB-overexpressing HEK cells (3 independent experiments, 3 replicates per group). j, k Response curves to NT for activation of G₁₂(134)-RLuc8/Gβ3γ9-GFP2 (j) and G₁₃(126)-RLuc8/Gβ3γ9-GFP2 (k) BRET biosensors in control (vector) or NTRS2-overexpressing HEK cells (3 independent experiments, 3 replicates per group). The data represent means ± SEM, analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test (b‒g).

Full size image

To confirm that activation of ET_B or NTS₂ enhances coupling to G₁₂ or G₁₃, we employed TRUPATH biosensors, which are optimized bioluminescence resonance energy transfer (BRET2) Gαβγ biosensors³¹. Dissociation of the G₁₂ or G₁₃ subunits and the Gβγ heterodimer is monitored by measuring the BRET2 signal upon agonist-induced activation of GPCRs. Reduction of the BRET2 signal indicates receptor-mediated G-protein dissociation. Co-expression of full-length ET_B or NTS₂ together with the BRET2 reporters Gβ3, Gγ9-GFP2, and G₁₂(134)-RLuc8 or G₁₃(126)-RLuc8 in HEK293 cells generated stable BRET2 luminescence signals, which declined in a dose-dependent manner after addition of increasing concentrations of either ET-3 or NT (Fig. 3h‒k). The decline of G₁₃-dependent signals was more pronounced and occurred at lower concentrations of ligands compared to G₁₂-dependent signals but coupling of ET_B and NTS₂ was evident for both G₁₂ and G₁₃. Taken together, the results indicate that binding of either ET-3 to ET_B or NT to NTS₂ directly activates G₁₂-G₁₃ signaling.

G₁₂-G₁₃ signaling is indispensable for maintaining quiescence of MuSCs and preventing depletion of the MuSC pool during aging

To further examine the role of G₁₂-G₁₃ as a signaling hub and integrator of GPCR-dependent signaling for inducing quiescence of MuSC, we generated MuSC-specific conditional compound knock-out mice for Gna12–Gna13 (G12/13^scKO) (Supplementary Fig. S3c, d). Ten days after initiation of Gna12–Gna13 inactivation, we observed a massive increase of MuSCs, of which 80% were localized outside the basal laminar (Fig. 4a, b). In addition, we detected a major increase of PAX7⁺KI67⁺ double-positive cells in the TA muscles of G12/13^scKO mice, indicating increased activation and proliferation of MuSCs (Fig. 4c; Supplementary Fig. S3e). The numbers of PAX7⁺MYOD⁺ and PAX7^‒MYOD⁺ were strongly elevated as well, whereas the number of PAX7⁺MYOD^‒ MuSC declined (Fig. 4d; Supplementary Fig. S3f), indicating reduced self-renewal and enhanced myogenic differentiation. This conclusion was further supported by the presence of numerous MyoG-positive nuclei in G12/13^scKO TA muscles (Fig. 4e, f). Intriguingly, the number of centrally located nuclei surged in G12/13^scKO TA muscles (Fig. 4g, h), associated with increased levels of eMyHC (Fig. 4i), indicating that activation of MuSCs due to depletion of G₁₂-G₁₃ results in fusion of MuSC to adjacent myofibers or formation of new fibers. This hypothesis was confirmed by genetic lineage tracing, revealing that all centrally located myonuclei are derived from Gna12–Gna13-deficient mCherry-labeled MuSCs (Supplementary Fig. S3g, h). Numerous mCherry⁺ nuclei were also detected in the periphery of myofibers, demonstrating continuous addition of Gna12–Gna13-deficient MuSCs and subsequent maturation (Supplementary Fig. S3i). Altogether, these findings demonstrate a critical role of active G₁₂-G₁₃ signaling for maintaining quiescence of MuSCs. To analyze whether the aberrant activation of G12/13^scKO MuSCs compromises skeletal muscle regeneration and prevents return of mutant MuSCs to the stem cell niche, we subjected G12/13^scKO mice to one- and three-times CTX-induced muscle injury. As expected, G12/13^scKO mice showed signs of compromised muscle regeneration 20 days after the injuries, reflected by fiber size heterogeneity and increased numbers of mononuclear cells (Supplementary Fig. S3j). The numbers of PAX7⁺ MuSCs were substantially lower in G12/13^scKO muscles after completion of regeneration. Moreover, only very few PAX7⁺ MuSCs were detected under the basal lamina, indicating reduced self-renewal and the inability of G12/13^scKO MuSCs to return to the stem cell niche (Supplementary Fig. S3k, l).

a, b Immunofluorescence (a) and quantification of MuSCs outside the basal lamina (b) (n = 3). c Ratios of KI67⁺ MuSCs in TA muscles of control and G12/13^scKO mice (n = 3). d Quantification of PAX7⁺MYOD^‒ (green bars), PAX7⁺MYOD⁺ (gray bars), PAX7⁻MYOD⁺ (ivory bars) cells in TA muscles of control and G12/13^scKO mice (n = 3). e, f Immunofluorescence (e) and quantification of MYOG⁺ cells (f) in TA muscles of control and G12/13^scKO mice (n = 3). g H&E staining of TA muscle sections from control and G12/13^scKO mice. h Quantification of centronuclear myofibers (n = 3). i Western blot analysis of eMyHC and GAPDH in TA muscles of control and G12/13^scKO mice (n = 3). j–l Body weight (j), TA (k) and GAS muscle (l) weights of control and G12/13^scKO mice (n = 3). m Distribution of cross-sectional areas (CSA) of myofibers in TA muscles of control (blue) and G12/13^scKO (yellow) mice (n = 3). n Quantification of myofibers on transverse sections of aged control and G12/13^scKO TA muscles (n = 3). o H&E staining of TA muscle sections from control and G12/13^scKO mice. p Quantification of PAX7⁺ cells in TA muscles of aged control and G12/13^scKO mice (n = 3). q Proliferation curve of young (YO, 2-month-old male mice) and old (OD, 80-week-old male mice) control and G12/13^scKO MuSCs (statistical significance pertains to the last time point, n = 3). The data represent means ± SEM, analyzed by unpaired t-test (b‒d, f, h‒l, n and p) and one-way ANOVA with Bonferroni’s multiple comparisons test (m, q). Male 2-month-old mice were used in a‒i and 80-week-old male mice in j‒q. Scale bars: 10 µm in a, e, 20 µm in g, and 50 µm in o.

Full size image

In contrast to G12/13^scKO mice, no activation of MuSC was observed in single G12^KO (Gna12^‒/‒) and G13^scKO (Pax7^CreERT2 Gna13^fl/fl) mice, indicating that G₁₂ and G₁₃ serve overlapping functions in MuSCs (Supplementary Fig. S4a‒e). Similarly, inactivation of Gnaq-Gna11 in MuSCs did not increase the numbers of PAX7⁺ or PAX7⁺MyoD⁺ MuSCs, the fraction of PAX7⁺ MuSCs outside the basal lamina, or KI67⁺PAX7⁺ cells (Supplementary Fig. S4a‒e). Apparently, G_q-G₁₁ does not play a major role for initiating or maintaining quiescence of MuSC.

To analyze long-term consequences of G₁₂-G₁₃ depletion and loss of MuSC quiescence, we inactivated Gna12–Gna13 in MuSCs of 2-month-old mice and then allowed the mice to age. At 20 months of age, G12/13^scKO mice showed decreased body weight and muscle mass (Fig. 4j‒l). Aged, 20-month-old G12/13^scKO mice experienced a 26% reduction of TA and a 23% reduction of gastrocnemius (GAS) muscle mass compared to aged control mice (Fig. 4j‒l). Increased loss of muscle mass was associated by disorganized myofiber structures and variations in size, but the numbers of muscle fibers did not decline in statistically significant manner (Fig. 4m‒o).

We also noted changes in fiber-type composition, characterized by a decrease in oxidative type I fibers (Supplementary Fig. S4f, g), different from the consequences of Gna13 inactivation in myofibers, which increases type 1 and 2a oxidative fibers³². Moreover, the number of MuSCs in TA muscles of aged G12/13^scKO dramatically declined (Fig. 4p; Supplementary Fig. S4h), associated with a much lower proliferation rate of aged MuSCs from G12/13^scKO mice compared to MuSCs from control mice (Fig. 4q). The changes in fiber-type composition of G12/13^scKO mice skeletal muscles did not rely on conversion of myofibers into a G12/13^scKO state, which might have been caused by continuous accretion of a G12/13^scKO MuSCs. We only observed a minor reduction of G13 protein in aged muscle fibers of G12/13^scKO mice, even in 80-weeks-old mice (Supplementary Fig. S4i). We also genotyped individual aged muscle fibers from G12/13^scKO mice, reasoning that replacement of existing nuclei from mutant MuSC should result in accumulation of the mutant allele. We clearly detected the genomic fragment derived from the Gna13 mutant allele in individual myofibers, but the wild-type band was much stronger, indicating that only a subset of myonuclei was replaced or added over time (Supplementary Fig. S4j). These results are well in line with previous reports, demonstrating that myonuclei are rather stable without prior muscle injury³³. Even in conditions of severe atrophy, the number of myonuclei remain constant without major replacement³⁴. Taken together, our findings demonstrate that G₁₂-G₁₃ signaling is instrumental for maintaining the MuSC pool and ensuring proper MuSC function during aging. Abrogation of G₁₂-G₁₃ signaling within MuSCs decreases the number of type I myofibers, which is correlated with enhanced skeletal muscle sarcopenia during aging and disrupts normal muscle morphology without loss of G13 protein in aged muscle fibers.

G₁₂-G₁₃ signaling in response to ET-3 and NT requires RhoA for suppressing MuSC activation

G₁₂-G₁₃ activate different intracellular signaling pathways, including Jun kinase (JNK) and cyclooxygenase-2 (COX-2), but the main downstream event is direct regulation of RH-RhoGEFs, which activates Rho GTPases³⁵. To gain insights into the signaling pathways activated by G₁₂-G₁₃, we investigated changes in the transcriptional activity of fluorescence-activated cell sorting (FACS)-purified MuSCs treated with ET-3 or NT for 5 days. Principal component analysis (PCA) of RNA-seq data revealed strong differences between treated and non-treated sample but a high correlation between ET-3- or NT-treated MuSCs with an R-value of 0.86 (Fig. 5a). The majorities of upregulated and downregulated genes (69.47% and 79.55%, respectively) were identical between ET-3 and NT treatments, relative to treatment with solvent (Supplementary Fig. S5a). Further Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA) indicated that upregulated and downregulated genes were primarily associated with Rho signaling pathways, suggesting that members of Rho family are the major effector molecules mediating G₁₂-G₁₃ signaling after ET-3 or NT stimulation (Fig. 5b; Supplementary Fig. S5b, c). Next, we examined the levels of active-Rho (Rho-GTP) following 5 days of G₁₂-G₁₃ activation by ET-3 and NT treatment, unraveling a substantial increase of active-Rho levels in MuSCs compared to control (Fig. 5c). Consistent with these findings, we observed a similar upregulation of active-Rho levels in MuSCs on isolated single FDB myofibers after treatment for 30 min with ET-3 or NT. The increase of active-Rho induced by ET-3 or NT was comparable to the effects of WNT4, which has been previously reported to activate RhoA in quiescent MuSCs (Fig. 5d, e). However, unlike ET-3 or NT and despite increased levels of active Rho following WNT4 treatment, WNT4 was unable to limit activation and proliferation of MuSC in vitro (Fig. 5f; Supplementary Fig. S5d). We speculate that effects of WNT4 treatment on RhoA activation is dissenting, less durable, or more indirect compared to ET-3 and NT, although we did not analyze such possibilities in detail. Consistent with these findings, the Rho inactivator C3 transferase (C3) blocked ET-3/NT-induced effects on MuSCs ex vivo (Fig. 5f; Supplementary Fig. S5d), confirming that Rho family members are the pivotal downstream mediators of G₁₂-G₁₃ signaling induced by ET-3 or NT. Analysis of active-Rho levels in MuSCs from G12/13^scKO mice confirmed the dependency of Rho activation on G₁₂-G₁₃ signaling. In Gna12–Gna13-deficient MuSCs on single myofibers isolated of G12/13^scKO mice, we observed significantly lower levels of active Rho and phosphorylated-Myosin Light Chain (pMLC), which is regulated by the Rho/Rho-kinase signaling pathway (Supplementary Fig. S5e‒h).

a PCA of RNA-seq data from DMSO-, ET-3- and NT-treated MuSCs and Pearson’s values. Each dot represents the mean of three biological samples. b GSEA of upregulated and downregulated genes, overlapping between ET-3- or NT-treated MuSCs compared to those treated with DMSO. c Rho activity of MuSCs measured by the G-LISA Kit after a 5-day culture with DMSO, ET-3, or NT. Active RhoA-GTP was determined by luminescence at 490 nm (n = 8). d, e Immunofluorescence (d) and quantification (e) of active Rho in MuSCs on isolated FDB myofibers of WT mice, after 30 min exposure to DMSO, ET-3 or NT (n = 5). f Ratios of PAX7⁺MYOD^‒ MuSCs on FDB myofibers from WT mice after 24 h exposure to DMSO, ET-3, or NT, with or without Rho inhibitor (C3) (n = 6). g RT-qPCR analysis of Rhoa expression in fresh isolated MuSCs of control (blue), Rhoa^scKO/+ (purple), and Rhoa^scKO (green) mice (n = 3). h, i Immunofluorescence (h) and quantification (i) of MuSCs outside the basal lamina in TA muscles of control, Rhoa^scKO/+, and Rhoa^scKO mice (n = 4). j Ratios of KI67⁺ MuSCs in TA muscles of control, Rhoa^scKO/+ and Rhoa^scKO mice (n = 3). k Quantification of PAX7⁺ (green bars), PAX7⁺MYOD⁺ (gray bars), and MYOD⁺ (ivory bars) cells in TA muscles of control, Rhoa^scKO/+, and Rhoa^scKO mice (n = 4). l, m Immunofluorescence (l) and quantification of MYOG⁺ cells (m) in TA muscles of Rhoa^scKO mice (n = 4). n, o H&E staining of Rhoa^scKO TA muscles sections (n). Quantification of centronuclear myofibers (o) in TA muscles of control, Rhoa^scKO/+, and Rhoa^scKO mice (n = 4). The data represent means ± SEM, analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test (c, e–g, i, j, m and o). Scale bars: 5 µm in d, 10 µm in h, l and 20 µm in n.

Full size image

Twenty mammalian Rho GTPases have been described, of which RhoA is one of the most prominent members, often activated through G₁₂-G₁₃³⁶. To test whether RhoA is indeed the critical mediator of G₁₂-G₁₃ signaling induced by ET-3 and NT, we conditionally inactivated Rhoa in MuSCs using Rhoa^scKO mice, in which gene inactivation is driven by Pax7-CreERT after tamoxifen administration (Fig. 5g). Homozygous inactivation of Rhoa in MuSCs dramatically increased the number of PAX7⁺ MuSCs outside the stem cell niche (Fig. 5h, i) and the number of proliferating Ki67⁺PAX7⁺ cells 24 days after initiation of gene inactivation (Fig. 5j). Furthermore, Pax7⁺MYOD^‒ MuSCs declined whereas Pax7⁺MYOD⁺ and Pax7^‒MYOD⁺ cells increased (Fig. 5k). Heterozygous inactivation of Rhoa had similar but far less pronounced effects and showed some notable differences to the homozygous mutant state (Fig. 5h‒j). We did not observe a significant increase of Pax7^‒MYOD⁺ cells in heterozygous Rhoa^scKO/+ muscles (Fig. 5k) and no increase of MYOG-positive muscle cells and centrally located myonuclei, which were abundant in homozygous Rhoa^scKO muscles (Fig. 5i‒o). Apparently, full activation of the myogenic program, indicated by expression of MYOG, resulting in the fusion of MuSCs to adjacent fibers and giving rise to centrally located myonuclei, requires nearly complete repression of RhoA activity. Taken together, inactivation of Rhoa in MuSCs fully phenocopies the loss of Gna12–Gna13, indicating that RhoA is necessary and irreplaceable for relaying G₁₂-G₁₃-dependent quiescence signals, probably serving as the primary downstream effector of G₁₂-G₁₃ in MuSCs to secure quiescence.

Activation of RhoA is sufficient to maintain MuSC quiescence in the absence of G₁₂-G₁₃ and has no effects on cytoplasmic projections of quiescent MuSC

To investigate whether RhoA is not only necessary but also sufficient to arrest MuSCs in quiescence, we generated a mouse model enabling conditional, MuSC-specific expression of a constitutively active mutant (Q63L) of human RhoA (Pax7^CreERT2; Tg: R26^LSL-caRhoa). We also fused a GFP-reporter to the N-terminus of caRhoA to facilitate detection (Fig. 6a). The approach was highly effective and specific. Essentially all Pax7⁺ MuSCs showed GFP-fluorescence, without any GFP fluorescence outside Pax7⁺ MuSCs (Supplementary Fig. S6a, b), and active RhoA was strongly increased (Supplementary Fig. S6c). The enhanced RhoA activity strongly suppressed motility of MuSCs, a hallmark of MuSC activation³⁷, in a transwell migration assay (Fig. 6b) and essentially annihilated proliferation of freshly isolated MuSCs (Fig. 6c). Furthermore, enhanced RhoA activity prevented activation of MuSC as indicated by the absence of MYOD expression in MuSCs on isolated myofibers from FDB muscles after 8 h of culture (Fig. 6d; Supplementary Fig. S6d).

a Design of the caRhoa overexpression allele. b Motility measurement of control and caRhoa^scOE MuSCs for 5 days in transwell migration assays (n = 5). c Time-lapse imaging of MuSCs proliferation. Gray dots: control MuSCs; red dots: caRhoa^scOE MuSCs (statistical significance at the last time point, n = 3). d Ratio of MYOD⁺PAX7⁺ MuSCs on FDB myofibers, 8 h after isolation from control and caRhoa^scOE mice (n = 6). e Experimental design of f‒h. f‒h Quantification of MuSCs outside the basal limina (f) and ratios of MYOD⁺ (g) and KI67⁺ (h) MuSCs in control and caRhoa^scOE mice in TA muscles intramuscularly injected with saline, BQ788 (1 mg/kg bodyweight), or SR142948A (0.5 mg/kg bodyweight) (n = 3). i Experimental design of j‒l. j‒l Quantification of MuSCs outside the basal lamina (j), ratio of MYOD⁺ MuSCs (k), and number of centronuclear myofibers (l) in control, caRhoa^scOE, G12/13^scKO, and caRhoa^scOE/G12/13^scKO (G12/13^RSU) TA muscles (n = 3). m Immunofluorescence for PAX7 (green), α-TUBULIN (red), DAPI (blue), and GFP (green) for control and caRhoa^scOE EDL myofibers (upper panel). The proportion of MuSCs with or without QPs is shown in the lower panel (n = 3). n, o Immunofluorescence (n) and quantification of Rac-GTP (o) in MuSCs on EDL myofibers (n = 6). The data represent means ± SEM, analyzed by unpaired t-test (b, d, m, o) and one-way ANOVA with Bonferroni’s multiple comparisons test (c, f‒h, j‒l, n, o). Scale bars: 10 µm in m and 5 µm in n.

Full size image

To analyze whether constitutively active RhoA preserves quiescence even after inhibition of ET-3 and NT signaling, which normally abrogates MuSC quiescence, we administered the ET_B and NTS₂ blockers SR142948A and BQ788, respectively, to TA muscles of caRhoa^scOE mice (Fig. 6e). Expression of caRhoA prevented activation of MuSC induced by SR142948A and BQ788 in control mice, measured by the absence of MuSCs outside the basal lamina, missing increase of MyoD⁺Pax7⁺ MuSCs, and missing increase in proliferating Ki67⁺/Pax7⁺ MuSCs (Fig. 6f‒h; Supplementary Fig. S6g). We also employed the genetic G12/13^scKO mouse model to demonstrate that increased levels of active RhoA are sufficient to arrest MuSC in quiescence, despite the absence of G₁₂-G₁₃-mediated signals. Analysis of caRhoa^scOE/G12/13^scKO triple mutant mice (G12/13^RSU mice) revealed a full rescue of the G12/13^scKO phenotype by expression of caRhoA, completely eliminating premature MuSC activation and normalizing the number of centrally located myonuclei to control levels (Fig. 6i‒l; Supplementary Fig. S6e‒j). We concluded that RhoA is a central signaling hub, in which not only G₁₂-G₁₃-derived signals but also others are integrated.

A recent report described that a Rac1-to-Rho switch is required for exiting quiescence and for retraction of complex cytoplasmic projections of quiescent MuSC, named quiescent projections (QPs). The authors postulated a cross-regulated equilibrium between Rac1 and Rho, in which increased Rho activity breaks quiescence³⁸. The claim that increased Rho/ROCK signaling abrogates quiescence and results in retraction of QPs is clearly in conflict with our findings. We therefore decided to examine the morphology of QPs as a further indicator of MuSC quiescence and analyze RhoA targets including Rac proteins and ROCKs in caRhoa^scOE MuSCs. Immunofluorescence staining of MuSCs for PAX7 and α-Tubulin on isolated single myofibers from control and caRhoa^scOE mice revealed no effects of increased RhoA activity on the formation of QPs, which is consistent with the arrest of caRhoa^scOE MuSCs in quiescence. Neither did the length of QPs change nor the number of MuSCs with visible QPs (Fig. 6m). Moreover, increased expression of active RhoA did not alter the levels of active Rac-GTP, indicating that RhoA does not suppress Rac in MuSCs in vivo (Fig. 6n, o).

After establishing that RhoA does not exert its effects by altering Rac activity, we turned to two well-known targets of RhoA, the Rho-Kinase 1/2 (ROCK1/2) and Diaphanous-related formins (DRFs), which rearrange the cytoskeleton by polymerizing actin³⁹. Proliferation of MuSC was measured by EdU incorporation on isolated myofibers and activation of myogenic differentiation by determining the number of MYOD⁺PAX7⁺ cells. Pharmacological inhibition of ROCK1-ROCK2 by the ROCK inhibitor Y-27632 (Y27) released the proliferation block of caRhoa^scOE MuSCs (Fig. 7a, b) but did not release the inhibition of myogenic differentiation, imposed by expression of caRhoA (Fig. 7c). Vice versa, inhibition of DRFs by the general formin inhibitor SMIFH2 induced expression of MYOD in caRhoa^scOE MuSCs on isolated myofibers but had no effects on proliferation (Fig. 7d, e). The differential effects of Y27 and SMIFH2 on MuSCs expressing caRhoA indicate that RhoA controls two different pathways to enforce quiescence, ROCKs for preventing cell cycle entry and formins to suppress myogenic differentiation. Notably, mDia1, a member of the DRF family, has been reported to inhibit expression of MYOD and MYOG, which confirms our findings^40,41. Since RhoA induces ROCK-mediated phosphorylation of MLC, which suppresses activation and nuclear translocation of YAP in some cell types⁸, we analyzed the presence of YAP in nuclei of control, G12/13^scKO, Rhoa^scKO/+, and Rhoa^scKO MuSCs. Only very few YAP⁺ nuclei were detected in control MuSCs, whereas deletion of Gna12-Gna13, homozygous inactivation of Rhoa, and in particular heterozygous inactivation of Rhoa raised the number of YAP⁺ nuclei (Fig. 7f). However, we would like to emphasize that a large fraction of MuSCs in Rhoa^scKO and G12/13^scKO mice did not show nuclear translocation of YAP (Fig. 7f), suggesting that RhoA does not solely rely on inhibition of YAP to induce quiescence of MuSCs. To confirm the critical role of YAP downstream of RhoA for stem cell activation, we inhibited ROCK with Y27 and formins with SMIFH2 in caRhoa^scOE MuSCs on isolated myofibers, with or without concomitant treatment with the YAP inhibitor verteporfin (VP) (Fig. 7g). Strikingly, administration of VP abrogated stimulation of EdU incorporation instigated by blockage of ROCK with Y27 but had no effect on MyoD expression, induced by SMIFH2 (Fig. 7h, i). Taken together, these data corroborated the hypothesis that RhoA controls two different pathways to synergistically enforce quiescence, a formin-dependent pathway that inhibits myogenic differentiation and a ROCK-dependent pathway, which inhibits proliferation by suppressing YAP activation (Fig. 7j).

a Experimental design of b‒e. b, c Ratios of EdU⁺ (b) and MYOD⁺ (c) MuSCs in control and caRhoa^scOE EDL myofibers, following 8-h exposure to DMSO and 10 µM Y27 (n = 4). d, e Ratios of MYOD⁺ (d) and EdU⁺ (e) MuSCs in control and caRhoa^scOE EDL myofibers following 8-h exposure to DMSO and 20 µM SMIFH2 (n = 4). f Proportions of YAP⁺ MuSCs on control, G12/13^scKO, Rhoa^scKO/+ and Rhoa^scKO TA muscle sections (n = 3). g Schematic representation of the experimental design of h, i. h, i Quantifications of the ratio of EdU⁺ (h) and MYOD⁺ (i) MuSCs in isolated myofibers from FDB muscles of caRhoa^scOE mice after 8 h treatment with DMSO, 10 µM Y27, or 20 µM SMIFH2 alone or in combination with 5 mΜ VP (n = 4). j Model depicting ROCK-YAP and Formin signaling pathways downstream of RhoA, preventing cell cycle entry and myogenic commitment in quiescent MuSC. The data represent means ± SEM, analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test (b‒e, h, i). Drawings displayed in a, g and j were created with BioRender.

Full size image

• SEO Powered Content & PR Distribution. Get Amplified Today.
• PlatoData.Network Vertical Generative Ai. Empower Yourself. Access Here.
• PlatoAiStream. Web3 Intelligence. Knowledge Amplified. Access Here.
• PlatoESG. Carbon, CleanTech, Energy, Environment, Solar, Waste Management. Access Here.
• PlatoHealth. Biotech and Clinical Trials Intelligence. Access Here.
• Source: https://www.nature.com/articles/s41421-024-00696-7