Dysferlin is a newly identified binding partner of AβPP and it co-aggregates with amyloid-β42 within sporadic inclusion-body myositis (s-IBM) muscle fibers

Abstract

composed of Aβ42 [1], and in both diseases Aβ42 was proposed to be pathogenically important [1, 7]. Accordingly, we asked whether dysferlin abnormalities, including accumulation within Aβ42-immuno-positive aggregates, might occur in s-IBM myofibers. Immunofluorescence using a well-characterized antidysferlin monoclonal antibody (Hamlet1, Novocastra) revealed, in all 5 s-IBM muscle biopsies studied, that dysferlin was prominently reduced in the muscle fiber sarcolemma, in contrast to the 4 age-matched controls that had normal sarcolemmal distribution of dysferlin (Fig. 1a–d, f). In addition, most of the s-IBM-vacuolated and/or otherwise abnormal muscle fibers had intra-myofiber dysferlin-immunoreactive “plaque-like” cytoplasmic aggregates (Fig. 1b–d, f). None of 8 disease-control biopsies (4 ALs, 4 peripheral neuropathies) had similar dysferlinimmunoreactive aggregates (not shown). Immunoblots performed from those same s-IBM and control biopsies did not reveal any significant differences in the amount of total dysferlin (Fig. 2a, a’). These data suggest that the sarcolemmal absence of dysferlin might relate to its abnormal distribution into the cytoplasm, perhaps secondary to a dysferlin trapping into the visible cytoplasmic protein aggregates, and probably sub-microscopic forms thereof. Double immunofluorescence, using an anti-Aβ42-specific polyclonal antibody (Calbiochem, La Jolla) plus a monoclonal anti-dysferlin antibody, revealed that virtually all dysferlin-immunoreactive aggregates co-localized with Aβ42-immunoreactive aggregates, both being usually “plaque-like” and of various size and distribution (Fig. 1d–g). Very rarely, there were dysferlin only aggregates. To verify whether immunohistochemical co-localization of dysferlin and Aβ42 reflects their physical association, we performed co-immunoprecipitation/immunoblotting using the anti-dysferlin and 6e10 Dysferlin is a 237 kDa type II trans-membrane protein involved in plasmalemmal repair and resealing after injury, and in T-tubule construction and calcium homeostasis [reviewed in 2]. Although in skeletal and cardiac muscle dysferlin is localized mainly to plasmalemma and cytoplasmic vesicles, recently, it was also localized to T-tubules and other myofiber structures [2]. DYSF gene mutations cause three main clinical phenotypes of autosomal recessive myopathies, called dysferlinopathies [2]. Unlike normal human muscle biopsies in which dysferlin is immunohistochemically present in the muscle fiber sarcolemma, in dysferlinopathies, dysferlin is absent from the sarcolemma, and is either absent or prominently reduced per immunoblots of muscle biopsies and monocytes [2, 5]. Beyond muscle diseases, in Alzheimer disease (AD) brain, dysferlin was abnormally accumulated in dystrophic neurites within amyloid-β (Aβ) plaques [6]. s-IBM, a common aging-associated degenerative myopathy, is pathogenically multifaceted [reviewed in 1]. s-IBM biopsied muscle fibers exhibit a unique complex phenotype, characterized by vacuolated muscle fibers containing congophilic multiprotein aggregates [1]. s-IBM muscle fiber degeneration shares several pathomolecular aspects with AD brain, including accumulations of Aβ and phosphorylated tau, the latter in the form of paired-helical filaments [reviewed in 1]. similarly to AD brain, s-IBM muscle fiber-accumulated Aβ is mainly