Proteasomal inhibition restores biological function of mis-sense mutated dysferlin in patient-derived muscle cells

Abstract

Dysferlin is a transmembrane protein implicated in surface membrane repair of muscle cells. Mutations in dysferlin cause the progressive muscular dystrophies Miyoshi myopathy, limb girdle muscular dystrophy 2B, and distal anterior compartment myopathy. Dysferlinopathies are inherited in an autosomal recessive manner, and many patients with this disease harbor mis-sense mutations in at least one of their two pathogenic DYSF alleles. These patients have significantly reduced or absent dysferlin levels in skeletal muscle, suggesting that dysferlin encoded by mis-sense alleles is rapidly degraded by the cellular quality control system. We reasoned that mis-sense mutated dysferlin, if salvaged from degradation, might be biologically functional. We used a dysferlin-deficient human myoblast culture harboring the common R555W mis-sense allele and a DYSF-null allele, as well as control human myoblast cultures harboring either two wild-type or two null alleles. We measured dysferlin protein and mRNA levels, resealing kinetics of laser-induced plasmalemmal wounds, myotube formation, and cellular viability after treatment of the human myoblast cultures with the proteasome inhibitors lactacystin or bortezomib (Velcade). We show that endogenous R555W mis-sense mutated dysferlin is degraded by the proteasomal system. Inhibition of the proteasome by lactacystin or Velcade increases the levels of R555W mis-sense mutated dysferlin. This salvaged protein is functional as it restores plasma membrane resealing in patient-derived myoblasts and reverses their deficit in myotube formation. Bortezomib and lactacystin did not cause cellular toxicity at the regimen used. Our results raise the possibility that inhibition of the degradation pathway of mis-sense mutated dysferlin could be used as a therapeutic strategy for patients harboring certain dysferlin mis-sense mutations.

Figures

FIGURE 1.

Characterization of human myoblast cultures.A, light microscopy image showing capability of myotube formation in 134/04 and impairment thereof in ULM1/01 and in 180/06 human myoblasts after 5 days in fusion medium. Scale bar, 250 μm. B, Western blot for dysferlin in 134/04 myoblasts and myotubes and ULM1/01 and 180/06 myoblasts. The two null alleles in ULM1/01 cells introduce stop codons in exons 44 and 46, respectively. Truncated dysferlin proteins, if generated at all from either of these two null alleles, would lack the C2 domains F and G as well as the transmembrane domain. The anti-DYSF antibody used in this study recognizes polypeptides encoded by exon 54 (amino acids 2020–2037) and would thus not be able to recognize such potentially truncated proteins. 180/06 cells carry one null allele and the R555W mis-sense allele. Lower, level of α-tubulin as loading control of identical samples run on a parallel gel. IB, immunoblot. C, Western blot for desmin in 134/04 myoblasts and myotubes and ULM1/01 and 180/06 myoblasts. Lower, level of α-tubulin as loading control of identical samples run on a parallel gel. D, quantitative data of relative fluorescence intensity over time after laser-induced injury of 134/04 (green triangles, n = 10), ULM1/01 (blue diamonds, n = 10), and 180/06 (red squares, n = 10) myoblasts, indicating defective plasma membrane resealing in ULM1/01 and 180/06 myoblasts and effective resealing in 134/04 myoblasts. Data are presented as means ± 1 S.D. (error bars). E, membrane repair assay performed on human myoblast cultures 134/04, ULM1/01, and 180/06 in the presence of Ca2+. The panel shows fluorescence accumulation of the FM1-43 dye over time. The lack of fluorescence intensity increase at the wound site in 134/04 myoblasts indicates that the injured plasma membrane has been resealed, whereas increased fluorescence intensity at the wound site of ULM1/01 and 180/06 myoblasts indicates impaired membrane resealing.

FIGURE 2.

Proteasomal inhibitors, but not lysosomal inhibitors, significantly increase protein levels of the dysferlin mis-sense mutant R555W in cultured human myoblasts.A, confluent cultures of 134/04, ULM1/01, and 180/06 myoblasts were treated with increasing concentrations of the lysosomal inhibitor chloroquine. Anti-cathepsin D and anti-LC3 antibodies were used for Western blotting of protein extracts to demonstrate successful lysosomal inhibition (lower panels) or anti-dysferlin antibody to detect the expression of full-length dysferlin (top panel). α-Tubulin was used as a loading control (middle panel). B and D, confluent cultures of 134/04, ULM1/01, and 180/06 myoblasts were treated with increasing concentrations of the proteasomal inhibitors lactacystin (B) or Velcade (D). Western blots of protein extracts were stained with anti-ubiquitin antibodies to demonstrate successful proteasomal inhibition (lower panel), with anti-α-tubulin antibodies as a loading control (middle panel), and with anti-dysferlin antibodies to detect the expression of full-length dysferlin (upper panel) (n = 4). Dysferlin levels increase significantly in the myoblast culture 180/06 harboring the R555W DYSF mis-sense allele and to a lesser degree in the wild-type myoblast culture 134/04. Western blot of ULM1/01 myoblasts, which harbor two null alleles, not able to generate full-length dysferlin, demonstrates that the C-terminally directed anti-dysferlin antibody used in this study is specific to dysferlin. C and E, y axis depicts the ratio between dysferlin and α-tubulin in 180/06 cells at each lactacystin (C) or Velcade (E) concentration normalized to the ratio between dysferlin and α-tubulin in 134/04 cells in absence of inhibitors. Stars indicate that differences were statistically significant (***, p < 0.001; ****, p < 0.0001). F, salvaged mis-sense mutated dysferlin is ubiquitylated. Confluent cultures of 180/06 myoblasts were treated with proteasomal inhibitor lactacystin or Velcade. Dysferlin was immunoprecipitated with rabbit polyclonal anti-dysferlin antibody and blotted with anti-ubiquitin antibody (see “Experimental Procedures”). IP, immunoprecipitation; IB, immunoblot; SM, standard material, 5% of total protein loaded. Error bars, S.D.

FIGURE 3.

Velcade treatment leads to localization of mis-sense mutated dysferlin to the plasma membrane and increases dysferlin mRNA.A, immunostaining against an extracellular dysferlin epitope in 180/06 myoblasts, carrying the R555W dysferlin mis-sense allele, incubated for 24 h with Velcade (50 nm) demonstrates membrane localization of dysferlin (green), DAPI (blue). An inverted image in black and white is represented on the right to better visualize the plasma membrane staining with the anti-dysferlin antibody in the Velcade-treated cells. Scale bar, 50 μm. B, levels of dysferlin mRNA (left panel) increase 4-fold after 24-h incubation of 180/06 human myoblasts with 10 nm Velcade. Levels of mis-sense mutated dysferlin protein (right panel) increase 23-fold after 24-h incubation of 180/06 human myoblasts with 10 nm Velcade. Protein levels were measured by densitometric analysis using the Western blot shown. Stars indicate that differences were statistically significant (****, p < 0.0001). Error bars, S.D.

FIGURE 4.

Mis-sense mutated dysferlin can rescue defective membrane resealing. Plasma membrane repair assay was performed on myoblast culture 180/06 which harbors a dysferlin mis-sense allele R555W and a DYSF-null allele. The laser-induced injury was performed after incubating the myoblasts for 24 h with increasing concentrations of lactacystin (A and B) or Velcade (C and D). Quantitative data of relative fluorescence intensity over time after laser-induced injury of 180/06 myoblasts treated with increasing concentrations of lactacystin (A) or Velcade (C) are presented as means ± 1 S.D. Numbers of individual measurements are as follows: for lactacystin 0 μm (n = 7), 2 μm (n = 10), 4 μm (n = 10), 8 μm (n = 12), 12 μm (n = 17); for Velcade 0 nm (n = 7), 5 nm (n = 10), 10 nm (n = 12), 25 nm (n = 20), 50 nm (n = 20). B and D, show the fluorescence accumulation of the FM1-43 dye over time at the plasma membrane injury site after incubating the myoblasts for 24 h with increasing concentrations of lactacystin (B) or Velcade (D). Scale bars, 1 μm.

FIGURE 5.

Treatment with proteasome inhibitors induces myotube formation in myoblasts harboring the dysferlin mis-sense allele R555W. Light microscopy images of human myoblasts 180/06 (A) and (G) and ULM1/01 (D) and (J) were treated with increasing concentrations of lactacystin (A–F) or Velcade (G–L) for 5 days in fusion medium to induce myotube formation. Scale bars, 50 μm. Desmin expression levels are shown as a marker for fusion, and α-tubulin levels represent loading controls of identical samples run on parallel gels (B, E, H, and K). Desmin levels were quantified in three independent experiments using ImageJ and were normalized to the levels of α-tubulin (C, F, I, and L). Data are presented as means ± 1 S.D. (error bars). 180/06 myoblasts can fuse when treated with concentrations as low as 8 μm lactacystin or 10 nm Velcade. ULM1/01 myoblasts remain unable to fuse irrespective of the concentrations of inhibitors used.

FIGURE 6.

Concentrations of lactacystin and Velcade used to achieve the biological effects are not toxic to the cultured human myoblasts. Cytotoxicity was measured in the human cultured myoblasts treated with increasing concentrations of lactacystin (A–C) or Velcade (D–F) after 24 h (black bars), 48 h (gray bars), and 5 days (white bars). The y axis represents the percentage of surviving cells compared with control cells without proteasomal inhibitor treatment. Data are presented as means ± 1 S.D. (error bars).