Cytotoxic Aggregation and Amyloid Formation by the Myostatin Precursor Protein

Citation: Starck CS, Sutherland-Smith AJ ( Cytotoxic Aggregation and Amyloid Formation by the Myostatin Precursor Protein Carlene S. Starck 0 Andrew J. Sutherland-Smith 0 Ashley M. Buckle, Monash University, Australia 0 Institute of Molecular BioSciences, Massey University , Palmerston North , New Zealand Myostatin, a negative regulator of muscle growth, has been implicated in sporadic inclusion body myositis (sIBM). sIBM is the most common age-related muscle-wastage disease with a pathogenesis similar to that of amyloid disorders such as Alzheimer's and Parkinson's diseases. Myostatin precursor protein (MstnPP) has been shown to associate with large molecular weight filamentous inclusions containing the Alzheimer's amyloid beta peptide in sIBM tissue, and MstnPP is upregulated following ER stress. The mechanism for how MstnPP contributes to disease pathogenesis is unknown. Here, we show for the first time that MstnPP is capable of forming amyloid fibrils in vitro. When MstnPP-containing Escherichia coli inclusion bodies are refolded and purified, a proportion of MstnPP spontaneously misfolds into amyloid-like aggregates as characterised by electron microscopy and binding of the amyloid-specific dye thioflavin T. When subjected to a slightly acidic pH and elevated temperature, the aggregates form straight and unbranched amyloid fibrils 15 nm in diameter and also exhibit higher order amyloid structures. Circular dichroism spectroscopy reveals that the amyloid fibrils are dominated by b-sheet and that their formation occurs via a conformational change that occurs at a physiologically relevant temperature. Importantly, MstnPP aggregates and protofibrils have a negative effect on the viability of myoblasts. These novel results show that the myostatin precursor protein is capable of forming amyloid structures in vitro with implications for a role in sIBM pathogenesis. - Funding: This work was funded by Tertiary Education Commission Top Achiever Doctoral and New Zealand Neuromuscular Alliance Henry Kelsey Scholarships to C.S.S. and grants from the Massey University Research Fund and the Palmerston North Medical Research Foundation to A.J.S.-S.. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Myostatin is a member of the transforming growth factor-beta (TGF-b) superfamily of growth and differentiation factors and is a primary regulator of muscle growth both pre- and postnatally, primarily via inhibition of myoblast proliferation and differentiation [14]. Like other members of the family, myostatin is translated as a precursor protein (MstnPP) that consists of an Nterminal signal sequence, a regulatory propeptide domain (residues 21266) and a growth factor domain (residues 267374) which dimerises at the C-terminus via an inter-molecular disulfide bond [57]. The mature growth factor dimer is cleaved from the propeptide region by furin convertase proteolysis in the endoplasmic reticulum (ER) at a conserved RSRR sequence [2,7]. The propeptide region plays at least two important functions. First, as a chaperone in the ER to assist in the folding of the growth factor region [8,9] that contains the intricate cysteine-knot motif characteristic of the family [10,11]. Second, the N-terminal propeptide plays a regulatory role after cleavage, remaining noncovalently associated with the mature dimer to form a latent complex which is exported from the cell [7]. Myostatin remains latent until a second cleavage event immediately N-terminal to aspartate 76 of the propeptide region, most probably by metalloproteinases, that disrupts the association [12,13]. It is possible that furin cleavage of MstnPP also occurs post-secretion with a pool of extracellular MstnPP identified in skeletal muscle [14]. The mature growth factor dimer is structurally similar to other members of the TGF-b family [11]; however the structural characteristics of MstnPP or the propeptide region remain undetermined. Signalling by the myostatin growth factor via activin and TGFb receptors [2] ultimately results in cell-cycle arrest through the upregulation of genes involved in cell-cycle withdrawal such as p21 and p53 and the downregulation of myogenic regulatory factors such as MyoD and myogenin [1518]. Although postnatally this action maintains the quiescence of muscle satellite cells, the prenatal role is more complex and depends on the environmental context during development, with signalling by myostatin ensuring that a balance between proliferation and differentiation is maintained [19]. Myostatin-null mutations have been identified in dogs, cattle and sheep, resulting in a double-muscled phenotype [1,20,21] and an exceptionally muscular and strong human lacking functional myostatin protein was also recently identified [22,23]. Myostatin overexpression in mice induces profound muscle and fat loss analogous to that seen in human cachexia syndromes [4] and ectopically expressed myostatin rapidly lowers muscle mass in rats [24]. Myostatin signalling can have negative consequences in a diseased background such as the muscular dystrophies [25] and may contribute to cachexia associated with many chronic disease states [4] including HIV [26] and cancer [27]. For these reasons, since its discovery in 1997 [3], the processed myostatin growth factor dimer has been suggested to hold exciting potential for inhibitory targeting in a wide range of muscle wastage diseases [28,29]. Less focus had been placed on the involvement of MstnPP or the propeptide region in disease until a role for MstnPP in the pathogenesis of sporadic inclusion body myositis (sIBM) was proposed [30,31]. sIBM is the most common progressive muscle wastage disease associated with aging where progressive muscle loss leads to severe atrophy and weakness. Although the pathogenesis is unknown, it is likely that oxidative damage contributes to aging of the muscle fibers [30]. Endoplasmic reticulum (ER) stress and the unfolded protein response (UPR) have been demonstrated in sIBM muscle fibers [30,32]. Inflammation and amyloid formation appear to be predominant features but whether these are causally related and which is the primary cause of sIBM, remain matters of debate [30,33]. The presence of fibrillar inclusions in some diseased tissue suggests that sIBM may be an amyloid disease, where a prominent feature is protein aggregation and the subsequent formation and deposition of large amyloid fibrils analogous to those observed for neurodegenerative disorders such as Alzheimers, Parkinsons and Huntingtons diseases, the spongiform encephalopathies, the systemic amyloidoses and type II diabetes [34,35]. Despite extensive research, the mechanisms behind amyloid formation and how they contribute to disease remain poorly understood. It is well-documented that fibril formation occurs via one or more oli (...truncated)