No abstract

The sarcomeres of skeletal and cardiac muscle are highly structured protein arrays, consisting of thick and thin filaments aligned precisely to one another and to their surrounding matrix. The contractile mechanisms of sarcomeres are generally well understood, but how the patterning of sarcomeres is initiated during early skeletal muscle and cardiac development remains uncertain. Two of the most widely accepted hypotheses for this process include the "molecular ruler" model, in which the massive protein titin defines the length of the sarcomere and provides a scaffold along which the myosin thick filament is assembled, and the "premyofibril" model, which proposes that thick filament formation does not require titin, but that a "premyofibril" consisting of non-muscle myosin, α-actinin and cytoskeletal actin is used as a template. Each model posits a different order of necessity of the various components, but these have been difficult to test in vivo. Zebrafish motility mutants with developmental defects in sarcomere patterning are useful for the elucidation of such mechanisms, and here we report the analysis of the herzschlag mutant, which shows deficits in both cardiac and skeletal muscle. The herzschlag mutant produces a truncated titin protein, lacking the C-terminal rod domain that is proposed to act as a thick filament scaffold, yet muscle patterning is still initiated, with grossly normal thick and thin filament assembly. Only after embryonic muscle contraction begins is breakdown of sarcomeric myosin patterning observed, consistent with the previously noted role of titin in maintaining the contractile integrity of mature sarcomeres. This conflicts with the "molecular ruler" model of early sarcomere patterning and supports a titin-independent model of thick filament organization during sarcomerogenesis. These findings are also consistent with the symptoms of human titin myopathies that exhibit a late onset, such as tibial muscular dystrophy.

The development of striated muscle in vertebrates requires the assembly of contractile myofibrils, consisting of highly ordered bundles of protein filaments. Myofibril formation occurs by the stepwise addition of complex proteins, a process that is mediated by a variety of molecular chaperones and quality control factors. Most notably, myosin of the thick filament requires specialized chaperone activity during late myofibrillogenesis, including that of Hsp90 and its cofactor, Unc45b. Unc45b has been proposed to act exclusively as an adaptor molecule, stabilizing interactions between Hsp90 and myosin; however, recent discoveries in zebrafish andC. eleganssuggest the possibility of an earlier role for Unc45b during myofibrillogenesis. This role may involve functional control of nonmuscle myosins during the earliest stages of myogenesis, when premyofibril scaffolds are first formed from dynamic cytoskeletal actin. This paper will outline several lines of evidence that converge to build a model for Unc45b activity during early myofibrillogenesis.

Despite the prevalence of developmental myopathies resulting from muscle fiber defects, the earliest stages of myogenesis remain poorly understood. Unc45b is a molecular chaperone that mediates the folding of thick-filament myosin during sarcomere formation; however, Unc45b may also mediate specific functions of non-muscle myosins (NMMs). unc45b Mutants have specific defects in striated muscle development, which include myocyte detachment indicative of dysfunctional adhesion complex formation. Given the necessity for non-muscle myosin function in the formation of adhesion complexes and premyofibril templates, we tested the hypothesis that the unc45b mutant phenotype is not mediated solely by interaction with muscle myosin heavy chain (mMHC). We used the advantages of a transparent zebrafish embryo to determine the temporal and spatial patterns of expression for unc45b, non-muscle myosins and mMHC in developing somites. We also examined the formation of myocyte attachment complexes (costameres) in wild-type and unc45b mutant embryos. Our results demonstrate co-expression and co-regulation of Unc45b and NMM in myogenic tissue several hours before any muscle myosin heavy chain is expressed. We also note deficiencies in the localization of costamere components and NMM in unc45b mutants that is consistent with an NMM-mediated role for Unc45b during early myogenesis. This represents a novel role for Unc45b in the earliest stages of muscle development that is independent of muscle mMHC folding.

The sarcomeres of striated muscle are among the most elaborate and dynamic eukaryotic cellular protein machinery, and the mechanisms by which these semicrystalline filament networks are initially patterned and assembled remain contentious. In addition to the acto-myosin filaments that provide motor function, the sarcomere contains titin filaments, comprised of individual molecules of the giant Ig-and fibronectin domain-rich protein titin. Titin is the largest known protein, containing many structurally distinct domains with a variety of proposed functions, including sarcomere stabilization, the prevention of over-stretching, and returning to resting length after contraction. One molecule of titin, which binds to both the Z-disk and the M-line, spans a half-sarcomere, and is proposed to serve as a "molecular ruler" that dictates the spacing of sarcomeres. The semirigid rod-like A-band region of titin has also been proposed to act as a scaffold for thick filament formation during muscle development, but despite decades of research, this hypothesis has not been rigorously tested. Recent studies in zebrafish have brought into question the necessity for the A-band region of titin during the early stages of sarcomere patterning. In this review, we give an overview of the many different roles of titin in the development and function of striated muscle, and address the validity of the "molecular ruler" model of myofibrillogenesis in light of the current literature.