Biochemical and pathological changes result from mutated Caveolin-3 in muscle

José Andrés González Coraspe, Joachim Weis, Mary E. Anderson, Ute Münchberg, Kristina Lorenz, Stephan Buchkremer, Stephanie Carr, René Peiman Zahedi, Eva Brauers, Hannah Michels, Yoshihide Sunada, Hanns Lochmüller, Kevin P. Campbell, Erik Freier, Denisa Hathazi, Andreas Roos

Caveolin-3 (CAV3) is a muscle-specific protein localized to the sarcolemma. It was suggested that CAV3 is involved in the connection between the extracellular matrix (ECM) and the cytoskeleton. Caveolinopathies often go along with increased CK levels indicative of sarcolemmal damage. So far, more than 40 dominant pathogenic mutations have been described leading to several phenotypes many of which are associated with a mis-localization of the mutant protein to the Golgi. Golgi retention and endoplasmic reticulum (ER) stress has been demonstrated for the CAV3 p.P104L mutation, but further downstream pathophysiological consequences remained elusive so far. We utilized a transgenic (p.P104L mutant) mouse model and performed proteomic profiling along with immunoprecipitation, immunofluorescence and immunoblot examinations (including examination of α-dystroglycan glycosylation), and morphological studies (electron and coherent anti-Stokes Raman scattering (CARS) microscopy) in a systematic investigation of molecular and subcellular events in p.P104L caveolinopathy. Our electron and CARS microscopic as well as immunological studies revealed Golgi and ER proliferations along with a build-up of protein aggregates further characterized by immunoprecipitation and subsequent mass spectrometry. Molecular characterization these aggregates showed affection of mitochondrial and cytoskeletal proteins which accords with our ultra-structural findings. Additional global proteomic profiling revealed vulnerability of 120 proteins in diseased quadriceps muscle supporting our previous findings and providing more general insights into the underlying pathophysiology. Moreover, our data suggested that further DGC components are altered by the perturbed protein processing machinery but are not prone to form aggregates whereas other sarcolemmal proteins are ubiquitinated or bind to p62. Although the architecture of the ER and Golgi as organelles of protein glycosylation are altered, the glycosylation of α-dystroglycan presented unchanged. Our combined data classify the p.P104 caveolinopathy as an ER-Golgi disorder impairing proper protein processing and leading to aggregate formation pertaining proteins important for mitochondrial function, cytoskeleton, ECM remodeling and sarcolemmal integrity. Glycosylation of sarcolemmal proteins seems to be normal. The new pathophysiological insights might be of relevance for the development of therapeutic strategies for caveolinopathy patients targeting improved protein folding capacity.