Search
Close this search box.

Denervation alters the secretome of myofibers and thereby affects muscle stem cell lineage progression and functionality – npj Regenerative Medicine

Skeletal muscle denervation alters properties of MuSCs

To identify the changes that occur in MuSCs after denervation, we first examined TA (tibialis anterior) cross sections from C57BL/6 J mice subjected to control (Sham) or denervation (DEN) surgery 7, 21 or 42 days prior to isolation, allowing us to choose the appropriate time point for in-depth analysis of MuSCs (Supplementary Fig. 1A–G). Briefly, at 21 d after denervation the following parameters were very pronounced: muscle weight loss driven by myofiber atrophy without loss of myofibers, as well as MuSC proliferation evidenced by a sharp increase in the number of Pax7/Ki67 positive cells. Therefore, we chose this time point for further analyses (Fig. 1A). Notably, normalizing mononucleated marker positive cells to muscle area—as it is often done—is not applicable here due to the strong size difference between muscle cross sections of surgery groups. Therefore, cell population counts are represented as absolute numbers of mononucleated marker positive cells per cross section throughout the manuscript. In addition to an increase in proliferation (Pax7/Ki67) specifically in mainly fast-twitch muscles (TA and Gastrocnemius, Supplementary Fig. 1 H, I), we identified a commitment to the myogenic lineage (MyoD/Myog) (Fig. 1B, C). Of note, intrinsic differences of slow and fast muscles, such as a higher number of MuSCs on type I compared to type II myofibers, indicated inherent MuSC differences and might explain the differences observed here18,19. This commitment was accompanied by an increase in the expression of eMHC in all muscles after DEN, with a peak at 7 d post-surgery in TA muscles (Supplementary Fig. 1J–N) and a shift in myofiber types (Supplementary Fig. 1O–T), suggesting a deregulation of myosin isoform expression in denervated TA muscles.

Fig. 1: Skeletal muscle denervation alters properties of MuSCs.
figure 1

A Experimental scheme (upper panel). Male C57BL/6 J mice were subjected to either Sham or DEN surgery and TA muscles were harvested for histological analysis 21 days after surgery. Representative images of TA muscles of Sham or DEN operated mice 21 d after surgery (lower panel). Scale bar is 1 cm. B Representative images of immunofluorescent stainings of TA muscle cross sections 21 d after Sham or DEN surgery. White arrowheads indicate Pax7 positive cells (red), MyoD positive cells (purple) or Myog positive cells (yellow). Yellow arrowheads indicate Pax7/Ki67 double positive cells. Nuclei were counterstained with DAPI. Scale bar is 50 µm. Inserts shows individual cells positive for Pax7, Pax7/Ki67, MyoD or Myog. Scale bar for the inserts is 2 µm. C Quantifications of (B). n = 6 – 7 animals, each data point represents one animal. Statistical testing was done by unpaired two-tailed t-test with Welch’s correction. Error bars represent SD, *p < 0.05, **p < 0.01. D Experimental scheme. Male C57BL/6 J mice were subjected to either Sham or DEN surgery. 21 days after surgery, hind limb muscles were harvested for FACS-isolation of MuSCs. E Multidimensional scaling plot of transcriptomics (left) and principal component plot of proteomics (right) data of isolated MuSCs. Ellipses represent 95% confidence intervals. F Volcano plot of DEGs (left, adj. p–value < 0.05) and DAPs (right, adj. p–value < 0.05) in MuSCs of DEN mice. Represented is the log2 fold change relative to expression in Sham samples, with selected downregulated genes in blue and upregulated genes in red. Dotted lines mark log2 fold change of -/+ 0.58 and adj. p–value < 0.05. n = 3–4 animals for transcriptomics and n = 4–5 technical replicates for proteomics. G 20 most significant biological processes in MuSCs after denervation identified as activated via GSEA. GeneRatio represents the fraction of enriched genes within a GO term.

Next, we asked which specific changes occur in MuSCs after denervation, comparing MuSCs 21 days after Sham or DEN surgery (Fig. 1D and Supplementary Fig. 2A, B). MuSCs were clearly distinguishable according to the surgery group on both the mRNA and protein level, suggesting profound alterations after denervation (Fig. 1E). In the transcriptome, 2613 DEGs were detected (adj. p-value < 0.05) while the proteome analysis revealed 1096 differentially abundant proteins (DAPs, q-value < 0.05) between surgery groups (Fig. 1F). As Lgals3 (encoding Galectin-3) was significantly upregulated after DEN both on the transcriptome and proteome level, we used it to exemplarily validate the omics data (Supplementary Fig. 2C). However, another validation of the transcriptome data is given for the expression analysis of Junb (Fig. 5B and Supplementary Fig. 7A). GSEA of the transcriptome identified the activation of “muscle tissue development” and “muscle cell differentiation”, indicating changes in MuSCs reminiscent of regeneration following denervation (Fig. 1G). Notably, we also identified the activation of immune response-linked processes (“cytokine-mediated signaling pathway”) suggesting an inflammation-associated milieu in the MuSC niche. Strikingly, an overrepresentation analysis (ORA) with an additional filter for log2 fold changes of >0.5, translating to a change of +1 (upregulated genes only), indicated that genes upregulated in MuSCs after DEN are involved in processes like “muscle tissue development”, “muscle cell differentiation” or “axon development” (Supplementary Fig. 2D). These results strengthen our notion that denervation induces regeneration-like gene expression changes in MuSCs. With a correlation analysis between DEGs and DAPs that were additionally filtered for log2 fold changes of </> −0.58/ + 0.58 (translating to a change of ± 1.5) we show that a subset of genes, including transcription factors such as Jun and Fos as well as transmembrane receptors like Integrins (Itgb4 and Itga6), is similarly regulated at the mRNA and protein level (Supplementary Fig. 2E, F), suggesting a correlation between changes of the MuSC transcriptome and proteome after denervation. In a study by Machado et al.20, the transcriptomes of in situ fixed isolated MuSCs and MuSCs in early activation were compared. Among the genes that are highly expressed in quiescent MuSCs but downregulated during activation are Abat, Dhcr24, Mapt, Mccc2 and Prodh, which we also found to be downregulated in MuSCs after DEN. Since these genes seem to be important for MuSC quiescence, their downregulation might cause the premature activation and break of quiescence observed in MuSCs after denervation. Of note, the unique factors that did not overlap between the analyses include genes of all classes, such as genes encoding transcription factors, receptors or metabolic enzymes.

Alterations in the myofiber niche are driving MuSC dysfunction in denervated skeletal muscle

To assess whether the observed changes in MuSCs are translated into functional impairments, we investigated the regeneration of skeletal muscle concomitant with denervation, allowing us to analyze whether regeneration is directly depending on innervation (Fig. 2A). CTX + DEN muscles had a constantly lower relative muscle weight compared to CTX + Sham muscles, mainly due to strong myofiber atrophy, thereby mimicking conditions when acute trauma of leg muscles coincide with lesions of the spinal cord (Fig. 2A and Supplementary Fig. 3A, B). Interestingly, 21 d after surgery regenerating myofibers of CTX + DEN muscles contained less nuclei than control muscles (Supplementary Fig. 3C). However, this was not observed at an earlier time point of regeneration (10 d post-surgery) and thus, it is likely that the size reduction of regenerating myofibers in CTX + DEN muscles is not only driven by a smaller number of nuclei fusing into myofibers but additionally by a decline in protein synthesis. Moreover, we identified an increased amount of eMHC positive myofibers in CTX + DEN muscles in the late phase of regeneration (Supplementary Fig. 3D). This suggests that, while the formation of new myofibers by MuSC fusion during early regeneration seemed to be independent of innervation (same percentage of eMHC positive myofibers between surgery groups at 5 and 7 days post injury), the downregulation of eMHC during late regeneration is dependent on the neuronal input. Reinnervation of skeletal muscle typically takes place around day 8 after injury supporting our notion that reinnervation plays an important role in MuSC behavior as well as formation and maturation of myofibers as observed here21,22. This deregulation of regeneration dynamics was accompanied by a higher proportion of fibrotic tissue in CTX + DEN muscles, indicating inadequate tissue regeneration (Supplementary Fig. 3G, H).

Fig. 2: Alterations in the myofiber niche are driving MuSC dysfunction in denervated skeletal muscle.
figure 2

A Experimental scheme (left). Male C57BL/6 J mice were subjected to cardiotoxin (CTX) – induced muscle injury in combination with either Sham or DEN surgery and TA muscles were harvested for histological analysis at the respective time points. Representative images of TA muscle of CTX+Sham or CTX + DEN operated mice at each time point after surgery (middle). Scale bar is 1 cm. TA muscle weight relative to bodyweight at each time point after surgery (right). The dotted line represents the average TA muscle weight relative to body weight of age-matched control mice. n = 16 for control mice and n = 4 for Sham and DEN mice. B Immunofluorescent staining of TA muscle cross sections of CTX+Sham and CTX + DEN operated mice at different time points after surgery. MuSCs are marked by Pax7 (red), proliferating cells are marked by Ki67 (green) and nuclei are counterstained with DAPI (blue). White arrowheads indicate Pax7/Ki67 double positive cells. Scale bar is 50 µm. Panel of inserts shows individual cells positive for Pax7 (1, red), Ki67 (2, green) or Pax7 and Ki67 (3, merge). Scale bar for the inserts is 2 µm. C Quantification of (B). n = 4 animals per surgery group at each time point, each data point represents one animal. Statistical testing was done by Two-way-ANOVA with post-hoc Tukey’s multiple comparisons test. Error bars represent SD. ns = not significant, *p < 0.05, ***p < 0.001, ****p < 0.0001. D Experimental scheme (left). Male and female CAG-GFP mice were subjected to Sham or DEN surgery. 21 days after surgery, hind limb muscles were harvested for FACS-isolation of GFP positive donor MuSCs. Male C57BL/6 J recipient mice obtained CTX injury, implantation of an osmotic pump containing the immunosuppressive FK506 and either Sham or DEN surgery 2 days before intramuscular injection of ten thousand donor MuSCs. After 21 days, TA muscles were harvested for histological analyses. Representative images of TA muscles of each group (middle; donor is indicated in green). Scale bar is 1 cm. TA muscle weight relative to bodyweight for each group at 21 days after MuSC transplantation (right). E Immunofluorescent staining of TA muscle cross sections for GFP 21 d after MuSC transplantation. Scale bar is 50 µm. F Quantification of (E) with GFP positive myofibers as percentage of total myofibers (left) and myofiber diameter of GFP positive myofibers (right). n = 3–4 animals per surgery group, each data point represents one animal. Statistical testing was done by One-way-ANOVA with post-hoc Tukey’s multiple comparisons test. Error bars represent SD. ns not significant, **p < 0.01, ****p < 0.0001.

Surprisingly, the total number of Pax7 positive MuSCs per cross section was comparable between surgery groups throughout the regeneration time course, while proliferation of MuSCs was reduced in CTX + DEN muscles until day 10 after injury (Fig. 2B, C). However, denervated muscles were considerably smaller than their Sham counterparts and therefore display a higher density of Pax7 positive MuSCs. Notably, the expression of the differentiation marker Myogenin showed similar dynamics as Ki67 in MuSCs (Supplementary Fig. 3E, F).

We then asked whether the phenotypic alterations in MuSCs following denervation persist in a fully innervated context or whether changes in the niche triggered by denervation are causing those alterations. Therefore, MuSCs from CAG-GFP mice23, which are characterized by ubiquitous GFP expression, after Sham or DEN surgery were transplanted into CTX-injured TA muscles of Sham or DEN operated C57BL/6 J recipient mice (Fig. 2D; MuSC donors are highlighted in green). Although DEN surgery induced a strong myofiber atrophy in the TA muscles of donor and recipient mice, the innervation state of donor MuSCs did not influence the recipient muscle weight (Fig. 2D–F and Supplementary Fig. 3I, J). Interestingly, the percentage of GFP positive myofibers, which originate from either the fusion of donor-derived MuSCs to each other or with resident MuSCs from the recipient or to existing myofibers, was comparable between the four surgery groups, suggesting that denervation does not affect the ability of MuSCs to fuse (Fig. 2E, F). Even though the increased number of Pax7 positive MuSCs in muscles that received MuSCs from a denervated donor did not lead to an increased proportion of GFP positive myofibers (Supplementary Fig. 3K), the size of GFP positive myofibers was severely decreased in DEN recipient muscles independent of the innervation state of the MuSC donor (Fig. 2F). This indicates that the MuSC niche is strongly affected by denervation and is causing the main effects on MuSCs following denervation reminiscent of the activation and differential gene expression we observed.

Denervation disrupts the immediate MuSC niche

The previous experiments suggest that the myofiber niche is driving MuSC dysfunction after denervation. Therefore, we wondered how MuSCs behave when cultured on their adjacent myofibers isolated from mice after Sham or DEN surgery (Fig. 3A). In line with the observed increase in Pax7 positive and Pax7/Ki67 positive cells in DEN TA cross sections (Fig. 1C), we observed an increase in the numbers of those cell populations on EDL myofibers after denervation (Fig. 3B, C). After 42 and 72 h of culture, myofibers from DEN mice harbored clearly more single cells, doublets and clusters per myofiber (Fig. 3D, E, Supplementary Fig. 4A–C). Although MuSCs on DEN myofibers were in a more proliferative state directly after isolation, this did not lead to an increase in cluster size at any time point (Fig. 3E and Supplementary Fig. 4D), indicating that proliferation was either only slightly or not permanently increased. Interestingly, activation of MuSCs was not affected in this assay (Supplementary Fig. 4E). Together, these results strengthen the notion that changes in MuSCs after denervation are driven by the immediate niche.

Fig. 3: Denervation disrupts the immediate MuSC niche.
figure 3

A Experimental scheme. Male C57BL/6 J mice were subjected to either Sham or DEN surgery and EDL muscles were harvested for myofiber culture 21 days later. B Immunofluorescent staining of isolated myofibers for Pax7 (red) and Ki67 (green) 0 h after isolation. White arrowhead indicates Pax7/Ki67 double positive cell. Nuclei were counterstained with DAPI. n = 12 animals per surgery group, each data point represents one animal. Scale bar is 10 µm. C Quantification of (B). D Immunofluorescent staining of isolated myofibers for Pax7 (red) after 72 h of culture. Nuclei were counterstained with DAPI. Scale bar is 10 µm. E Quantification of (D). n = 7 animals per surgery group, each data point represents one animal. Statistical testing was done by unpaired two-tailed t-test with Welch’s correction. Error bars represent SD. ns = not significant, *p < 0.05, **p < 0.01, ****p < 0.0001. F Experimental scheme. Male C57BL/6 J mice were subjected to either Sham or DEN surgery. 21 days later, EDL muscles were harvested and single myofibers were isolated for transcriptomic analysis. G Multidimensional scaling plot of myofiber transcriptome (left). Volcano plot of DEGs (adj. p–value < 0.05) in myofibers of DEN mice (right). n = 4 animals per surgery group, each data point represents one animal. Represented is the log2 fold change relative to expression in Sham samples, with selected downregulated genes in blue and upregulated genes in red. Dotted lines mark log2 fold change of -/+ 0.58 and adj. p–value < 0.05.

Since myofibers comprise the largest component of the immediate MuSC niche, we investigated changes in the transcriptome of myofibers following denervation (Fig. 3F). We observed a clear clustering of surgery groups, driven by massive changes in the expression landscape after denervation, including upregulation of genes associated with an inflammatory response (Fig. 3G and Supplementary Fig. 5A). Notably, the transcriptome of whole muscles containing different cell types following denervation and myofibers that were isolated by collagenase digestion to remove contaminating cell types shared over 4000 DEGs, with 99% of genes being regulated in the same direction (Supplementary Fig. 5B–F), emphasizing that myofibers display major changes following denervation.

Factors secreted by myofibers drive MuSC fate alterations after denervation

We speculated that alterations in factors secreted by myofibers cause the changes in MuSC functionality following denervation. Therefore, we performed a DAVID-based prediction analysis of the myofiber transcriptome for secreted/extracellular factors. Thereby, we obtained a list of 691 predicted secreted factors, which we filtered for an average RPKM value of at least 2 between Sham and DEN samples. Since the RPKM value indicates the level of gene expression we used it here to identify genes which display a reasonably high expression in both surgery conditions. From the remaining 272 predicted secreted factors (Fig. 4A), we chose Spp1 (secreted phosphoprotein 1, encoding the protein Opn) and Tgfb1 (transforming growth factor beta 1) for further analyses due to their central role in inflammation and cell fate specification. Moreover, S100a8, which was shown before to be involved in the early inflammatory response after peripheral nerve injury, was strongly upregulated in myofibers after denervation, indicating an ongoing immune reaction24. We identified an increased gene expression of Spp1 and Tgfb1 in myofibers after denervation and a strongly increased protein abundance of Opn and Tgfb1 in whole muscle as well as myofibers following denervation (Fig. 4A, B). Strikingly, ELISA analysis of supernatants from cultured myofibers from Sham or DEN muscles corroborated the prediction analysis, as we detected a significant accumulation of both proteins in the supernatant of myofibers from mice that had undergone DEN surgery (Fig. 4C, D). To investigate whether Opn and Tgfb1 are drivers of MuSC activation after DEN, we isolated myofibers with their adjacent MuSCs from C57BL6/J mice and treated them with Opn (for 72 h) or Tgfb1 (for 48 h) recombinant proteins to mimic the high protein abundance in the supernatant of DEN myofibers (Supplementary Fig. 6A–F). Surprisingly, neither Opn nor Tgfb1 alone caused alterations in the number of single cells, clusters or cluster size, indicating that a single factor is not sufficient to induce DEN-like phenotypes in myofiber-associated MuSCs. Therefore, we next examined whether the whole myofiber secretome is sufficient to induce MuSC proliferation, as seen directly after isolation of myofibers from DEN mice (Fig. 3C). In brief, we added culture supernatants from myofibers of Sham or DEN mice to cultures of isolated myofibers with their adjacent MuSCs from healthy C57BL/6 J mice (Fig. 4E). Strikingly, the supernatant (SUP) from DEN myofibers was sufficient to increase the number of Pax7 positive, MyoD positive and Myog positive cells per myofiber as well as the number of clusters and the cluster size (Fig. 4F). Thereby, we demonstrate that changes in the myofiber secretome cause MuSC expansion, further supporting our notion that the altered microenvironment following denervation is driving MuSC alterations. We then investigated whether MuSC activation by the secretome of DEN myofibers could be blocked by neutralizing Tgfb1 from the supernatant (SUP). To this end, isolated myofibers from C57BL6/J mice were incubated with supernatant (SUP) from Sham or DEN myofibers together with a blocking antibody targeting Tgfb1 (Supplementary Fig. 6G–I). Interestingly, blocking of Tgfb1 was not sufficient to prevent the increased cluster formation induced by SUP from DEN myofibers, further strengthening the notion that one factor alone is not sufficient to drive MuSC activation after denervation. Therefore, we next used a cocktail of recombinant proteins (Opn, Chrd and Ostn), all of which were found among the upregulated predicted secreted factors (Supplementary Fig. 6K–M). Surprisingly, incubation of isolated myofibers with their adjacent MuSCs with the recombinant protein cocktail even decreased the number of single cells and clusters per myofiber, opposite to the effects from the DEN SUP. We suggest that the right combination of all or the majority of secreted factors in the supernatant from DEN conditions are required to cause the alterations in MuSCs – at least in the 72 h time frame which was analyzed here.

Fig. 4: Factors secreted by myofibers drive MuSC fate alterations after denervation.
figure 4

A Volcano plot of gene expression changes of predicted secreted factors in the denervated myofiber transcriptome (upper graph). Prediction of secreted factors was performed with the DAVID bioinformatics database. Dotted lines mark log2 fold change of -/+ 0.58 and average RPKM of 2. qRT-PCR for Spp1 and Tgfb1 mRNA in myofibers of Sham or DEN operated mice (lower graphs). n = 3–4 animals per surgery group, each data point represents one animal. B Western blot analysis of Opn and Tgfb1 protein level in muscle lysates (upper panel) and myofiber lysates (lower panel) of Sham or DEN operated mice. C Experimental scheme. Male C57BL/6 J mice were subjected to either Sham or DEN surgery and EDL muscles were harvested for myofiber culture 21 days later. After 4 h of culture, myofiber supernatant was collected for ELISA. D Opn and Tgfb1 concentrations in the myofiber supernatant samples after 4 h of culture. Calculation of standard curve and concentrations was performed using an online ELISA calculator (https://www.arigobio.com/ELISA-calculator). Each sample was analyzed in duplicates. n = 7–8 animals per surgery group, each data point represents one animal. E Experimental scheme (left). Male C57BL/6 J mice were subjected to either Sham or DEN surgery and EDL muscles were harvested for myofiber culture 21 days later. After 42 h of culture, myofiber supernatant was collected and directly used for treatment of freshly isolated myofibers from healthy C57BL/6 J mice. Immunofluorescent staining of isolated myofibers for Pax7 (red) 72 h after isolation and culture in Sham or DEN supernatant (SUP) (right). Nuclei were counterstained with DAPI. Scale bar is 10 µm. F Quantification of (E, right). Plots show data normalized to Sham samples. n = 4–12 animals per surgery/treatment group, each data point represents one animal. Statistical testing was done by unpaired two-tailed t-test with Welch’s correction. Error bars represent SD. ns = not significant, *p < 0.05, **p < 0.01, ****p < 0.0001.

Denervation induces expression of Junb in MuSCs

After demonstrating that alterations in the myofiber secretome by denervation affect MuSC behavior (Fig. 4), we asked whether myofiber-secreted factors directly affect the expression of target genes in MuSCs. Therefore, we first performed a KEGG pathway enrichment analysis of the MuSC transcriptome and identified pathways activated in MuSCs following denervation, like “MAPK signaling pathway” and “Hippo signaling pathway” among the top activated pathways, both being important for cell proliferation (Fig. 5A). To identify potential transcriptional regulators we compared the DEGs of the MuSC data set with two publicly available lists of murine transcription factors25,26, resulting in an overlap of 152 differentially expressed transcription factors (Fig. 5B). Of those, the gene expression of Jun and Fos family members was increased in MuSCs after denervation. Interestingly, only Junb motifs were also found in the top 15 enriched motifs in a HOMER known motif analysis (Fig. 5C). Additionally, we observed a very pronounced upregulation of Junb expression in MuSCs after denervation (Fig. 5B, right and Supplementary Fig. 7A). To investigate whether induction of Junb expression in MuSCs from denervated mice was driven by myofiber-secreted factors, we incubated myoblasts of healthy C57Bl/6 J mice with supernatants from isolated myofibers of Sham or DEN mice. 6 h later, we assessed Junb gene expression and detected a clear trend for increased Junb expression in myoblasts incubated with supernatant from denervated myofibers (Fig. 5D). To further investigate the role of Junb in MuSCs after denervation, we used a publicly available list of Junb target genes27 and compared them to significantly upregulated genes (adj. p–value < 0.05, log2fc > 0.5) from our DEN MuSCs transcriptome (Supplementary Fig. 7B). From the 702 DEGs, 409 are Junb target genes, accounting for ~58% of DEGs. By performing ORA and generating enrichment plots of Junb over non-Junb target genes, we identified processes in DEN MuSCs where Junb target genes are involved (Supplementary Fig. 7C). Interestingly, we observed an enrichment of processes like “response to interferon-alpha” and “response to interferon-beta”, which is in line with our GSEA from DEN MuSCs (Fig. 1G), indicating a response to inflammatory signals.

Fig. 5: Denervation induces expression of Junb in MuSCs.
figure 5

A 20 most significant KEGG pathways in MuSCs after denervation identified as activated via GSEA. GeneRatio represents the fraction of enriched genes within a KEGG pathway. B Overlap analysis of DEGs from MuSC transcriptome (blue circle) and publicly available lists of mouse transcription factors (red and yellow circle) (left) and Volcano plot of the 152 identified transcription factors in the intersection (middle). Represented is the log2 fold change relative to expression in Sham samples, with selected downregulated genes in blue and upregulated genes in red. Dotted lines mark log2 fold change of -/+ 0.58 and adj. p–value < 0.05. qRT-PCR for Junb mRNA in MuSCs of Sham or DEN operated mice (right). C HOMER known motif analysis showing top 15 motif enrichments in denervated MuSCs. D qRT-PCR for Junb mRNA in myoblasts that were treated for 6 h with supernatant (SUP) of myofibers from Sham or DEN operated mice. E Western blot analysis of Junb protein level in lysates of myoblasts cultured under growth conditions. Myoblasts were treated either with 20 ng/ml Tgfb1 recombinant protein or solvent control (0.1% BSA in 10 mM citric acid). F qRT-PCR for Junb mRNA in cultured myoblasts after treatment with Tgfb1 recombinant protein. n = 3–4 animals per surgery/treatment group, each data point represents one biological replicate. Statistical testing was done by unpaired two-tailed t-test with Welch’s correction. Error bars represent SD. (*) p < 0.05.

As Tgfb1 was strongly induced in myofibers after denervation (Fig. 4A and Supplementary Fig. 7D) and accumulated in the supernatant of myofibers from DEN mice (Fig. 4D), we tested whether Tgfb1 alone was able to induce Junb expression in myoblasts (Fig. 5E, F). Although it has been shown before that Tgfb can induce Junb expression, for example during breast cancer invasion28, it is unknown whether Junb in MuSCs is a target gene of myofiber-secreted Tgfb1 in denervated muscle. Strikingly, Junb expression and protein abundance in MuSCs were increased after incubation with Tgfb1, indicating that Tgfb1 is acting as one of the upstream regulators of Junb (Fig. 5E, F). Together, we demonstrate here that denervation alters the myofiber secretome, which then drives gene expression changes in MuSCs, leading to altered functionality (Fig. 6).

Fig. 6: Graphical summary.
figure 6

Skeletal muscle denervation is driving MuSCs activation by myofiber-secreted factors.