(A) Western blotting of mouse muscle lysates under reducing and denaturing conditions shows an increase in TGF-β1 and its downstream effectors in Col6a2^–/– muscle lysates. Corresponding bands were quantified and normalized to GAPDH as an internal control and graphed as fold change in comparison with WT. Error bars represent SEM. For statistical comparisons, a parametric, 2-tailed, unpaired t test with Welch’s correction was used (n = 4). (B) Immunofluorescence staining of muscle tissue shows increased p-SMAD3 staining in myonuclei, including internal nuclei of Col6a2^–/– muscle. Scale bars: 25 μm. (C) HEK293-Luc TGF-β reporter assay shows that Col6a2^–/– muscle fibroblast cultures deposit 2- to 3-fold higher levels of active TGF-β compared with WT controls in conditioned medium. Each data point is from an independent preparation (biological replicates) normalized to the average of WT samples for each run (WT, n = 10; Col6a2^–/–, n = 8). (D) Blue native gel electrophoresis (not denatured, not reduced), followed by Western blotting of muscle lysates using anti–collagen VI or anti–TGF-β1 antibodies (as indicated). As expected, collagen VI tetramers are absent from Col6a2^–/– muscle. TGF-β1 migrates in 2 different protein complexes (arrows), with the lower complex highly increased in Col6a2^–/– muscle lysates compared with WT controls. (E) Collagen VI coimmunoprecipitation (using anti-Col6a1 antibody) of muscle lysates from WT and Col6a2^–/– mice followed by quantitative data-independent acquisition (DIA) mass spectrometry identifies putative collagen VI binding partners in skeletal muscle (n = 5). (F) TGF-β coimmunoprecipitation (using anti-1D11 antibody) of muscle lysates followed by quantitative DIA mass spectrometry identifies different TGF-β binding partners in WT and Col6a2^–/– skeletal muscle, corresponding to the different protein complexes identified in D (n = 4). Only proteins with an extracellular or plasma membrane localization are shown.