Collagen VI is a fibrosis-associated signal disrupting muscle regeneration across distinct human myopathies

Figures

• 

  Figure 1. FAPs are abundant in affected muscles of OPMD, IBM, and DMD patients and impair myotubes fusion in vitro. (A) Sirius red coloration of human skeletal muscle biopsies from healthy CTL, OPMD, IBM, and DMD patients. Scale bar: 50 μm - Quantification of the percentage of fibrosis by Sirius red coloration (biological replicates n = 5–8 biopsies per condition). OPMD P < 0.0001, IBM P < 0.0001, DMD P < 0.0001. (B) Immunostaining of laminin (white), Hoechst (blue), and embryonic myosin heavy chain (MyHC-emb (MYH3)) (green). Scale bar: 50 μm. Quantification of the percentage of MyHC-emb positive fibers for each condition (biological replicates n = 3–14 biopsies per condition) OPMD P = 0.0018, IBM P = 0.0003, DMD P = 0.0107. (C) Immunostaining of laminin (white), Hoechst (blue), and CD90 (red) on human biopsies. Quantification of the number of interstitial cells positive for CD90 per 100 fibers (biological replicates n = 3–4 biopsies per condition). Scale bar: 50 μm. OPMD P = 0.0006, IBM P = 0.0148, DMD P = 0.0106. (D) In vitro proliferation rate expressed as the mean population doubling (PDL) /day of FAPs isolated from healthy CTL, OPMD, IBM, and DMD muscle biopsies (biological replicates…

• 

  Figure 2. Proteomic analysis highlights different disease profiles and a shared ECM signature. (A) Venn diagram representing differentially expressed proteins shared between myopathies- Number of differentially expressed proteins identified among pathologies when compared to their age-matched control muscles (increased in red, decreased in blue). (B) Significantly altered proteins (Appendix Tables S1–S4) were subjected to a GO term analysis to identify associated biological processes using gProfiler. Bar charts of the top enriched terms for each comparison. (C) Representation of the repartition of proteins on their predicted secretion potential using the Outcyte prediction tool (UPS unconventional protein secretion). Source data are available online for this figure.

• 

  Figure 3. FAPs are the main actors responsible for matrisome alterations in myopathies. (A) Venn diagram representing matrisome proteins (from the matrisome database matrisomeproject.mit.edu) differentially expressed between pathologies and their age-matched control muscles. In blue, proteins decreased in DMD biopsies. (B) Representation of the repartition of the differentially expressed proteins for each group of ECM-related proteins, in each pathology. (C) Representation of the fold changes in abundance of differentially expressed ECM-associated proteins for each pathology (same color code as (A)) compared to their controls. Color code for proteins corresponds to the matrisome database protein groups shown in (B). (D) Immunostaining of laminin (white), Hoechst (blue), and Collagen alpha-3(VI) chain (red). Scale bar: 50 μm. (E) Heatmap of the expression of ECM proteins in several resident cell types within skeletal muscle on sc-RNA-seq dataset extracted from De Micheli et al 2020 (data ref: GSE143704). MuSCs muscle stem cells, Sk muscle skeletal muscle. (F) Immunostaining of Collagen alpha-3(VI) chain, fibronectin, Collagen IV, and Tenascin X on ECM produced by FAPs in vitro. Scale bar:…

• 

  Figure 4. FAPs-driven ECM modifications impact muscle regeneration and stem cell niche maintenance. (A) Experimental setup of FAPs transplantation. FAPs from healthy CTL, OPMD, or DMD were injected into the TA of immunodeficient mice after injury. TA were retrieved and analyzed 30 days post injections. For each analysis, the portion of the TA containing human cells (FAPs-remodeled regions) was compared to the regenerated mouse region without any human FAPs. (B) Immunostaining of laminin (white), Hoechst (blue), and human-specific Collagen alpha-3(VI) chain (red) or human-specific laminin gamma 1 (red). Scale bar: 50 μm. (C) Quantification of minimal feret fiber diameter in the muscle region remodeled by FAPs normalized to the non-injected muscle area. (biological replicates n = 4–8 per condition) CTL ns; P = 0.1008, OPMD P = 0.0009, DMD P = 0.0115. (D) Immunostaining of laminin (white), Hoechst (blue), Pax7 (red), and human-specific laminin gamma 1 (green). Arrows point to Pax7-positive cells. Scale bar: 50 μm. Quantification of Pax7+ cells per fiber normalized to the non-injected muscle area (biological replicates n = 4–8) CTL ns; P = 0.3960, OPMD P < 0.0001, DMD P = 0.0002. Data are…

• 

  Figure 5. Collagen VI is involved in the impairment of myoblast fusion in vitro. (A) RT-qPCR quantification of ECM genes normalized to the expression of human-specific B2M (biological replicates n = 3–4) OPMD - hTNXB P = 0.0366, hFN1 P = 0.0004; hLAMC1 ns P = 0.0519, hCOL6A1 P = 0.0421 hCOL6A3 P = 0.0357. DMD - hDCN ns P = 0.0551, hFBN1 ns P = 0.240 hLAMC1 ns P = 0.1594 hCOL6A1 P = 0.0051 hCOL6A3 P = 0.0067. Data are presented as mean ± SD, with P values obtained by t test. (B) Expression of COL6A1, COL6A2, and COL6A3 in several resident cell types within skeletal muscle on sc-RNA-seq dataset extracted from De Micheli et al 2020 (Data ref: GSE143704) (biological replicates n = 8). (C) Immunostaining of myotubes (Myosin Heavy Chain, green; Hoechst, blue) after 5 days of differentiation on different concentrations of collagen type VI coatings. Quantification of the fusion index (technical replicates n = 3 per condition). Scale bar: 200 μm. 1.25 vs 5 ns P = 0.3288; 1.25 vs 25 P = 0.0418; 1.25 vs 50 P = 0,0014; 5 vs 50 P = 0.0129. Data are presented as mean ± SD, with P values obtained by ordinary one-way ANOVA test followed by Tukey’s multiple comparisons test (*P < 0.1, **P < 0.01, ***P <…