The Myonuclear Domain in Adult Skeletal Muscle Fibres: Past, Present, and Future

Abstract

Most cells in the body are mononuclear whereas skeletal muscle fibres are uniquely multinuclear. The nuclei of muscle fibres (myonuclei) are usually situated peripherally which complicates the equitable distribution of gene products. Myonuclear abundance can also change under conditions such as hypertrophy and atrophy. Specialized zones in muscle fibres have different functions and thus distinct synthetic demands from myonuclei. The complex structure and regulatory requirements of multinuclear muscle cells understandably led to the hypothesis that myonuclei govern defined “domains” to maintain homeostasis and facilitate adaptation. The purpose of this review is to provide historical context for the myonuclear domain and evaluate its veracity with respect to mRNA and protein distribution resulting from myonuclear transcription. We synthesize insights from past and current in vitro and in vivo genetically modified models for studying the myonuclear domain under dynamic conditions. We also cover the most contemporary knowledge on mRNA and protein transport in muscle cells. Insights from emerging technologies such as single myonuclear RNA-sequencing further inform our discussion of the myonuclear domain. We broadly conclude: 1) the myonuclear domain can be flexible during muscle fibre growth and atrophy, 2) the mechanisms and role of myonuclear loss and motility deserve further consideration, 3) mRNA in muscle is actively transported via microtubules and locally restricted, but proteins may travel far from a myonucleus of origin, and 4) myonuclear transcriptional specialization extends beyond the classic neuromuscular and myotendinous populations. A deeper understanding of the myonuclear domain in muscle may promote effective therapies for ageing and disease.

Graphical Abstract

The “myonuclear domain” is the theoretically finite area that an individual muscle fibre nucleus provides RNA for within the multinuclear cell. The myonuclear domain may only expand to a certain extent during muscle fibre growth (hypertrophy) before the addition of a new myonucleus from a muscle stem cell (satellite cell) is required. Conversely, the myonuclear domain may shrink during muscle atrophy and myonuclei may eventually be lost (top left panel). mRNAs, as well as the proteins that are made from them from a given myonucleus, are actively transported along microtubule “tracks” in adult muscle. Myonuclei can also move along tracks using molecular motors to facilitate specialized gene expression and protein synthesis where needed (top right panel). mRNAs and proteins from an individual myonucleus can travel different distances in the cell based on specific molecular characteristics as well as enviromental conditions (bottom left panel). Furthermore, myonuclear gene expression is specialized based on location in the cell (bottom right panel). Collectively, various mechanisms may influence the size of the myonuclear domain. Undertanding how the myonuclear domain is regulated and maintained has implications for muscle adaptation and the etiology of disease.

Introduction

As long cylindrical and voluminous cells containing hundreds to thousands of nuclei (myonuclei), mammalian skeletal muscle fibres require unique regulation to maintain homeostasis. The engines of locomotion, muscle fibres are densely packed with motor proteins (i.e. myosin heavy chains) in their core. The numerous control centers of the muscle fibre, the myonuclei, are pushed to the periphery as a consequence of this architecture. Peripheral myonuclear placement in muscle fibres leaves relatively large areas of the cell without a nucleus in the vicinity; this causes myonuclei to transcriptionally oversee relatively large “territories” or “domains”. The highly plastic nature of muscle fibres to grow (hypertrophy) or shrink (atrophy) in response to various perturbations places additional demands on coordinating the regulation of transcription among myonuclei.

Muscle fibres can differ widely in contractile myosin composition or “type” (e.g. slow-twitch versus fast-twitch) as well as metabolic profile (e.g. oxidative versus glycolytic) (Pette & Staron, 1997, 2001; Zierath & Hawley, 2004). The biochemical and functional characteristics of a given fibre-type can even vary from muscle to muscle (Luden et al., 2008; Gejl et al., 2021) as well as between sexes (Miller et al., 2013; Callahan et al., 2014; Jeon et al., 2019). Myonuclei appear to be organized in a regular pattern throughout the muscle fibre which may confer a functional advantage (Bruusgaard et al., 2003; Metzger et al., 2012; Hansson et al., 2020b). Myonuclear organization may change with age and differ according to fibre type (Cristea et al., 2010). Some evidence suggests myonuclear patterning along a muscle fibre, specifically in oxidative fibres, is in part dependent on vascular organization (Ralston et al., 2006). Other myonuclei are clustered and have specialized functions for maintaining certain regions of the myofibre such as the myotendinous junction (MTJ) or neuromuscular junction (NMJ) (Merlie & Sanes, 1985; Englander & Rubin, 1987; Fontaine et al., 1988; Perillo & Folker, 2018; de Lima et al., 2021; Yaseen et al., 2021). The clustering of specialized nuclei makes their local domains smaller than for individual nuclei in other areas of the fibre and hints at differential myonuclear domain regulation. Recent single myonuclear RNA-sequencing (smnRNA-seq) studies provide deep insights on myonuclear specialization and how muscle fibre characteristics such as myosin heavy chain type (MyHC) and metabolic profile are coordinated and maintained within the fibre syncytium (Chemello et al., 2020; Dos Santos et al., 2020; Englund et al., 2020a; Garry et al., 2020; Kim et al., 2020; Petrany et al., 2020; de Lima et al., 2021; Orchard et al., 2021; Wen et al., 2021; Eraslan et al., 2022; Lin et al., 2022). Collectively, studying how myonuclei contribute to their intracellular environment and maintain muscle fibre homeostasis is exceedingly complex, an area of historic and ongoing interest, and has consequences for understanding muscle remodeling in health and disease.

Enesco and Puddy were among the first to observe a positive relationship between DNA content and muscle mass during post-natal growth (Enesco & Puddy, 1964). Shortly thereafter, Moss reported a linear relationship between muscle fibre growth and nuclear abundance during juvenile development (Moss, 1968). These early findings provided the basis for the concept of a “DNA unit” in muscle. The DNA unit was later expanded upon and formally defined by Cheek and colleagues (Cheek, 1985). These authors stated, “While certain types of exercise cause increments in muscle growth, it is of interest that the protein/DNA ratio does not change…which supports the concept that the nucleus dominates a finite volume of cytoplasm, the DNA unit…” (Cheek et al., 1971). Similar observations have been made in cardiomyocytes that can also contain several nuclei per cell (Landim-Vieira et al., 2020). With myonuclei as the main sites of transcription, emphasis has been placed on defining the theoretically finite “transcriptional jurisdiction” of each myonucleus within the muscle fibre syncytium. Maintenance of the proposed transcriptional “myonuclear domain” becomes of particular interest during muscle fibre growth, when the upper threshold of each domain within the syncytium might be exceeded and become rate-limiting (Petrella et al., 2008; Conceicao et al., 2018; Murach et al., 2018a; Snijders et al., 2020; Snijders et al., 2021). Thus, the myonuclear domain has long been considered a key determinant of muscle mass regulation and function in adult skeletal muscle (Allen et al., 1999; Van der Meer et al., 2011; Murach et al., 2018a).

In recent years, innovative in vivo and in vitro approaches have shed new light on the myonuclear domain with respect to transcript and protein distribution throughout muscle cells. The purpose of this review is to: 1) place the myonuclear domain in historical context and provide recent perspectives as it relates to the control of muscle cell transcription and translation, 2) discuss the potential roles of the myonuclear domain during muscle adaptation, and 3) propose new directions of research to further define its regulation and limits.

The Past

Foundations of the Myonuclear Domain Concept

As early as the 1950’s, it was recognized that each myonucleus in a muscle fibre syncytium may control a local “territory” (Ruska & Edwards, 1957). In 1974, Landing, Dixon, and Wells published “Studies on Isolated Human Skeletal Muscle Fibres: Including a Proposed Pattern of Nuclear Distribution and a Concept of Nuclear Territories” (Landing et al., 1974). This study was perhaps the first detailed investigation into what would eventually become formally recognized as the myonuclear domain. These authors insightfully speculated that “the ratio of the diffusion distance along tile surface (radius of a nuclear territory) to the vertical diffusion distance (the radius of the fibre) is critical to some biochemical function(s) of the skeletal muscle fibre” (Landing et al., 1974). Building on the earlier work of Mintz and Baker using “mosaic” mice to study muscle enzyme production (Mintz & Baker, 1967), Frair and Peterson found that different GPL-1 isozyme proteins were not localized to their myonucleus of origin, but rather uniformly distributed along the length of the muscle fibre (Frair & Peterson, 1983). Similar results were later reported for dystrophin protein using heterozygous mdx mice (Watkins et al., 1989). These studies were limited in scope and did not track the fate of mRNAs or proteins from specific myonuclei at high resolution. Notwhistanding, these in vivo observations laid the groundwork for more detailed investigation into the myonuclear domain in skeletal muscle cells.

Early In Vitro Studies of the Myonuclear Domain

In the 1980s, a deeper appreciation was emerging for the idea that myonuclei could be specialized in zones depending on their location within the muscle fibre. Several studies identified specialized myonuclei associated with the neuromuscular junction (NMJ). These myonuclei produced specific transcripts and proteins that were restricted to the area surrounding the motor endplate (Merlie & Sanes, 1985; Fontaine et al., 1988), providing preliminary clues for the existence of a myonuclear domain.

The earliest investigations aimed at directly studying the myonuclear domain at single-nucleus resolution used in vitro experiments involving heterokaryons. Building on their earlier work (Miller et al., 1988), the Blau laboratory reported on the localization of gene products in syncytial muscle cells in 1989 (Pavlath et al., 1989). In brief, the authors generated chimeric multinuclear myotubes using myogenic progenitor cells from mice and humans and determined the localization of species-specific proteins around myonuclei. They concluded that a muscle membrane-, golgi apparatus-, and sarcomeric-associated protein were primarily localized to the vicinity around the myonucleus of origin. Ralston and Hall performed similar experiments at the same time as Pavlath and coworkers using stable cell lines with a reporter gene and arrived at similar conclusions (Ralston & Hall, 1989b). Related in vitro and in vivo studies on NMJ myonuclei and the localization and/or transcriptional rate of their gene products around the NMJ provided further evidence to support the concept of a myonuclear domain (Bursztajn et al., 1989; Rotundo, 1990; Sanes et al., 1991; Jasmin et al., 1993; Grubič et al., 1995; Rossi et al., 2000; Jevsek et al., 2006; Ravel-Chapuis et al., 2007; Kim et al., 2020). Interestingly, Ralston and Hall also presented evidence that a membrane protein derived from a single myonucleus can be widely distributed at the surface of a myotube (Ralston & Hall, 1989a). This observation raised the possibility that defining myonuclear domains could be complicated and context-specific. These same authors went on to show that while mRNAs may be localized around their nucleus of origin, some proteins can be distributed more broadly throughout the cell depending on their function (Ralston & Hall, 1992). With respect to transcriptional territories around myonuclei, they hypothesized that “…mRNA, shortly after its appearance in the cytoplasm of muscle cells, binds to a cytoskeletal or membrane component that restricts its further movement” (Ralston & Hall, 1992). In a follow-up study, Ralston et al. provided evidence that mRNA distribution in myotubes could be determined by ribosome availability at the endoplasmic reticulum, and not necessarily transcript stability (i.e. rate of RNA degradation) (Ralston et al., 1997). These data are consistent with ribosomes influencing mRNA localization in neurons (Lu et al., 1998).

The seminal efforts at defining the myonuclear domain in muscle were generally limited to in vitro experiments in which a few select mRNAs or proteins were studied. Collectively, though, these studies yielded fundamental insights on how myonuclei control muscle homeostasis and emphasized two main ideas: 1) mRNAs from specific myonuclei remain localized, which could be related to ribosomes, and 2) proteins from specific myonuclei can be restricted but can also be distributed widely, likely dependent upon the function of the protein.

The Present

Modern Technology to Study the Myonuclear Domain

Skeletal muscle fibres are highly differentiated, post-mitotic, and syncytial cells; each of these characteristics presents unique experimental challenges. Recent work has made progress implementing modern technologies - including genetic engineering, sensitive fluorescent reporters, single molecule imaging, and single nucleus transcriptomics - in muscle model systems in vitro and in vivo to provide information on the myonuclear domain. A key limitation to studying the myonuclear domain is that adult/mature myofibres are difficult to model in vitro. Cutting edge methods in molecular and cell biology are also difficult to apply in vivo. In recent years, researchers studying the myonuclear domain have taken parallel approaches: 1) devising contemporary in vitro systems to more faithfully model mature myofibres, and 2) adapting cutting edge technology to in vivo settings where possible. These advancements have confirmed many of the hypotheses discussed above while also revealing novel insights.

Methodological Considerations - Advancing In Vitro Models for Studying the Myonuclear Domain

Modeling a fully differentiated, mature myofibre in vitro is notoriously difficult. The C2C12 mouse myoblast cell line has long been used as a simple, manipulable system to study myogenesis in vitro (Yaffe & Saxel, 1977; Blau et al., 1983). C2C12s were used for some of the foundational work described on the myonuclear domain above. These cells are easily obtained from cell line repositories, can be passaged for long periods in vitro, and induced to differentiate using a simple serum starvation protocol. Unfortunately, differentiated C2C12 myotubes do not recapitulate all the features of mature myofibres (Abdelmoez et al., 2020), and their utility is generally limited to studying early events in myogenesis or immature myotubes. Nevertheless, C2C12 is an attractive model due to its tractability. Some work has attempted to push C2C12 myotubes further into maturity using a variety of engineered culture systems. Engineered substrates as opposed to standard tissue culture plastic has been particularly successful. Its widespread adoption is unfortunately limited by complex implementations that are inaccessible to most laboratories. A simple recent method that uses patterned hydrogel substrates to force myotube alignment before fusion was successful in generating more mature myotubes without complicated or expensive equipment (Denes et al., 2019). Additional work has shown that tuning substrate stiffness can promote alignment without the need for patterning (Jensen et al., 2020), which is promising as an even more minimal system. Engineered substrates can facilitate more mature cultures for molecular studies and fixed cell imaging. Due to the requirement to image through the substrate, however, high-resolution live-cell imaging using standard confocal or wide-field optics becomes complicated. Combining engineered substrates with light-sheet microscopy, which uses water-dipping objectives above the sample, is a promising solution for studying myonuclear behavior in the future.

A common alternative to C2C12 cells are primary mouse myoblasts (Rando and Blau, 1994). These cells can be isolated from mouse tissue relatively easily using fluorescent activated cell sorting and differentiated into myotubes in vitro using straightforward protocols (Liu et al., 2015). Primary myoblasts typically generate more mature myotubes than those derived from C2C12, and can be used to study the effects of sex (C2C12 were derived from a female donor). Even more translatable, primary myoblasts can be obtained from human skeletal muscle for generating mature myotubes. Combining primary cells with engineered substrates appears to magnify the maturity of myotubes. Methods based on embedding cells in commercially available Matrigel show impressive results, generating long myotubes that recapitulate elements of myonuclear spacing (Pimentel et al., 2017; Alave Reyes-Furrer et al., 2021; Brunetti et al., 2021). Unfortunately, primary cells can not be passaged indefinitely due to replicative senescence, which limits genome engineering applications that require single-cell cloning. While C2C12 can repopulate from a single clone and retain differentiation potential, they do not have a stable diploid karyotype and are subsequently not ideal for genome engineering applications. This same problem has led scientists in other fields to move away from similarly unstable cell lines like HeLa towards immortalized primary cells or pluripotent stem cells that have relatively stable karyotypes. The development and widespread adoption of immortalized human myoblast lines capable of differentiation into mature myotubes and amenable to cutting-edge gene editing methods would be a key resource for the field moving forward (Lathuiliere et al., 2022).

Overall, leveraging the aforementioned advances in muscle cell culture will allow for more granular insights into ageing, exercise adaptation, and disease with respect to studying the myonuclear domain. These models are already being combined with in vitro exercise or injury paradigms that use optogenetics or laser ablation to simulate in vivo muscle physiology (Roman et al., 2021; Hennig et al., 2022). Combined myoblast/myotube models may also provide more detailed investigation on the role of myonuclear accretion in response to contraction (Kneppers et al., 2018).

Fluorescent Reporters to Assess Protein Distribution in the Myonuclear Domain

The use of genetically encoded fluorescent reporters in vitro and in vivo have been particularly successful for understanding functional-dependence of protein distribution within and between myonuclear domains. Work in the 1990’s documented the transport of nuclear-localized β-galactosidase protein between myonuclei in vitro and in vivo with differing results; some studies reported restricted localization, while others reported translocation between myonuclei (Ono et al., 1994; Yang et al., 1997; Blaveri et al., 1999). Utilizing more sensitive fluorescent reporters featuring a nuclear localization signal (NLS) and clever in vitro experimental designs, Cutler et al. found that nuclear proteins can be transported to neighboring myonuclei, but the import of proteins harboring a NLS can vary between myonuclei (Cutler et al., 2018). Heterogeneous myonuclear import of proteins might reflect functional myonuclear specialization. Myonuclear specialization is further supported by evidence from Drosophila muscle development, where the myonucleus from the “founder cell” differs transcriptionally from other fused-in myonuclei upon syncytial muscle cell growth (Bataillé et al., 2017). Another recent study using NLS tracking of proteins in vitro reported similar findings regarding translocation to neighboring myonuclei (Taylor-Weiner et al., 2020). These authors also observed that medium and large proteins travel furthest from their nucleus of origin during chemically-induced hypertrophy (Taylor-Weiner et al., 2020).

In vivo, the transport of proteins with a NLS from satellite cell-derived myonuclei (resulting from exercise-induced fusion) is widespread throughout the muscle fibre (Masschelein et al., 2020). Furthermore, fusion of a limited number of GFP-expressing cells from the circulation to dystrophic muscle fibres results in ubiquitous cytoplasmic expression of GFP throughout muscle fibres, but only regional expression of dystrophin (Chretien et al., 2005). A similar result using a β-galactosidase reporter with myoblast transfer in muscle was noted (Kinoshita et al., 1998). Recent evidence using a fluorescent dystrophin reporter indicates that dystrophin is highly compartmentalized to sarcolemmal domains around myonuclei in muscle fibres, and that dystrophin is enriched at the myotendinous junction (Morin et al., 2023). This compartmentalization has implications regarding the efficiency of genetic therapeutic approaches targeting myonuclei for muscular dystrophy. The current evidence suggests that proteins generated from one myonucleus can travel outside the transcriptional domain of that nucleus, but the distance may depend on the function of the protein. Using the most up-to-date technology, studies should expand on the classic but limited work mentioned above to explore the distribution and localization of specific cytoplasmic and membrane proteins.

Methods to Study RNA Production and Distribution in the Myonuclear Domain

Recent work has applied modern techniques to investigate the myonuclear domain at the level of RNA. This work can be stratified into two broad categories: production and distribution. While this work has largely agreed with the original in vitro studies, the ability of recent studies to apply new methods to mature, in vivo muscle has revealed novel mechanistic insights.

Regarding RNA production, a number of modern studies have characterized transcriptional profiles of myonuclei and other cell-types from muscle tissue. Single-cell transcriptomics methods have been widely applied in other tissues and led to the description of a previously under-appreciated diversity of molecularly-defined cell types. The syncytial nature of muscle cells initially complicated application of single-cell RNA sequencing methods to muscle tissue; however, technical advances allowing single-myonucleus sequencing were eventually developed (Dos Santos et al., 2020; Petrany et al., 2020). These protocols were dependent on more efficient molecular biology, allowing single nucleus sequencing libraries to be generated from limited starting material, and optimized nuclear isolation protocols. smnRNA-seq experiments identified known populations of transcriptionally-specialized myonuclei at junctional regions and also revealed heterogeneity in the transcriptional profiles of myonuclei throughout the fibre. The implications of this work are discussed in greater detail below.

With respect to RNA distribution, recent studies emphasize the importance of understanding patterns and mechanisms of RNA localization in the sarcoplasm in greater detail. RNA localization is well-studied in neurons, but an appreciation for its importance has lagged behind in muscle despite the well-defined involvement of RNA binding proteins in a variety of muscle diseases, such as myotonic dystrophy (Timchenko et al., 1996). This gap is in part due to technical challenges in applying the tools used to study RNA localization in other systems (such as neurons) to multinuclear muscle cells. Nevertheless, recent progress been made in adapting both live and fixed cell RNA imaging techniques to study RNA localization in muscle systems (see Tutucci et al., 2018 for a detailed review of RNA imaging strategies).

The MS2 system, used for live-cell RNA imaging, is based on a bacteriophage RNA-binding protein (MS2 coat protein) that associates with high affinity to the MS2 stem-loop RNA sequence. By inserting this sequence into a reporter RNA and expressing the reporter along with a MCP-GFP fusion protein, transcripts can be visualized in live cells. The MS2 system has been used extensively in cultured cells and neurons to reveal the mechanisms that transport RNAs in the cytoplasm. The general paradigm in the RNA localization field is that cis sequence elements in RNA molecules are bound by RNA-binding proteins that direct their localization to specific regions through interactions with either anchoring adapters or molecular motors. These cis elements, termed “zipcodes”, and the RNA-binding proteins that bind them work together to define the final steady-state localization pattern for any given RNA. The MS2 system has now been applied to both Drosophila muscle and C2C12 to assess live cell RNA localization dynamics, revealing some unique features of muscle cells.

The first study to use a MS2 reporter in myotubes was in the context of a Drosophila development model. This study found that mRNAs were restricted in individual myonuclear domains (Van Gemert et al., 2009), which supports the earlier observations of Ralston and Hall. The MS2 technology allowed bulk measurements of RNA dynamics via photobleaching methods. Unfortunately, the authors were unable to resolve single RNA molecules and thus could not examine the mechanisms governing RNA mobility in detail. More recent work using the MS2 system in C2C12 myotubes achieved single-molecule resolution imaging using a HaloTag-MCP fusion and far-red fluorescent dyes with more favorable photochemical properties. This study characterized reporter RNA mobility in detail, finding that microtubule-dependent transport was a critical determinant allowing RNAs to translocate in the sarcoplasm of myotubes (Denes et al., 2021). The result was somewhat surprising, as directed transport is typically thought to move zipcode-directed RNAs to specific locations, while in this study a generic reporter underwent efficient directed transport and was otherwise immobile in the sarcoplasm. These studies set the stage for expanded future work that should investigate the impacts of zipcodes on RNA mobility to determine whether and how RNA localization specificity is attained in myotubes.

The MS2 system allows live cell RNA dynamics to be observed, but it requires genetic manipulation. The cost and technical demands of such a approach can make the implementation of high sensitivity imaging in vivo prohibitive. Single molecule RNA fluorescence in situ hybridization (smFISH), on the other hand, does not require any genetic manipulation, can resolve the localization of endogenous targets with single-molecule precision, and in principle can be applied to tissues. Despite these advantages, applying smFISH to muscle tissue is hampered by high autofluorescence and difficulty in quantifying localization patterns. A number of recent studies have overcome these barriers using signal amplification approaches and single myofibre isolation along with computational pipelines to quantify RNA abundance and spatial organization. Because single myofibres can be cultured ex vivo, these approaches can be combined with pulse/chase treatment strategies to infer mechanisms and dynamics of RNA localization from steady-state smFISH snapshots. Recent complementary studies in mouse myofibres used this strategy to evaluate mechanisms of RNA transport, finding that microtubule-based transport and RNA size played important roles in the distance RNAs travel within myonuclear domains (Denes et al., 2021; Pinheiro et al., 2021)

The early conclusions from genetic labeling and tracking studies in muscle are broadly in agreement with earlier observations: many transcripts localize around their nucleus of origin, but once the corresponding protein is synthesized, can be transported more widely throughout the muscle cell depending on its function. Single nucleus RNA-seq has been powerful for understanding heterogeneity in myonuclear domain gene expression states, but it does not preserve spatial information. RNA imaging techniques, on the other hand, reveal high-resolution spatial information but have generally only been applied to a limited number of transcripts. Future work should aim to bring cutting-edge spatial transcriptomics approaches in muscle systems (McKellar et al., 2021; McKellar et al., 2022) to the context of exercise and the myonuclear domain. Adapting modern approaches to resolving the spatial localization patterns of the full transcriptome in muscle promises to clarify the relationship between myonuclear domains and tissue organization and provide the statistical power to evaluate mechanisms underlying localization specificity.

Mechanisms of Spatial Transcriptional and Translational Control in Muscle Cells

Understanding how the myonuclear domain is regulated requires not just characterization of steady-state localization patterns of biomolecules, but also characterization of the mechanisms that arrange these molecules in the muscle syncytium. Perhaps the most striking recent observation from multiple groups is that the movement of mRNAs from the perinuclear region is highly dependent on microtubule- and kinesin-based transport, becoming operative upon myogenic differentiation (Denes et al., 2021; Pinheiro et al., 2021). Microtubule dependence for mRNA transport is consistent with what has previously been observed for protein (Pizon et al., 2005; Wang et al., 2013) and organelle transport in muscle (Elhanany-Tamir et al., 2012; Metzger et al., 2012; Azevedo & Baylies, 2020; Collins et al., 2021). Interestingly, a similar requirement for microtubules and kinesins was demonstrated in a distinct but related system – the cardiomyocyte. Here, inhibiting the microtubule transport of mRNA away from the perinuclear space reduced protein synthesis and limited the extent of cardiac hypertrophy in response to adrenergic stress (Scarborough et al., 2021). A large body of literature describes microtubule-dependent RNA transport in neurons (Goldstein & Yang, 2000; Hirokawa, 2006; Fernandopulle et al., 2021), and the general principles of RNA transport (Suter, 2018) in muscle appear consistent with this work. Interestingly, in both neuron and muscle cells, passive RNA diffusion seems limited relative to proliferating cells in culture. Understanding the nature of constrained RNA mobility in muscle will be important going forward. It is conceivable that the physical constraints imposed by the increasing density of myofibrils during differentiation limits the passive diffusion of mRNAs, though further work is required to determine if this scenario is true. Binding of RNAs to organelles or anchored ribosomes could impose constraints as well.

In myofibres, highly abundant mRNAs are disseminated efficiently such that they achieve a random distribution, while low abundance transcripts show a slight concentration gradient, or “source” effect around the myonucleus of origin. This transcript localization is perhaps due to the burst-like nature of transcription (Fukaya et al., 2016; Larsson et al., 2019) and the time-scale of RNA movement away from the nucleus. Future work to optimize transcriptome-scale RNA visualization methods, such as seqFISH (Shah et al., 2017; Shah et al., 2018) or MERFISH (Wang et al., 2018; Xia et al., 2019a; Xia et al., 2019b) for use in muscle tissue promises to reveal the relationships between RNA localization, physical properties of RNAs, and other features such as ontological categories.

The microtubule-dependent trafficking of mRNAs in muscle fibres raises important questions about how generalized mRNA transport is achieved and whether there is specificity in the transport machinery that carries mRNAs along microtubule networks. Of the 46 known kinesin genes, three constitute ~75% of the kinesins expressed in skeletal muscle; orthologues Kif1c and Kif1bα make up 65% of all kinesin-encoding transcripts with an additional ~10% from Kif5b mRNA (Wang et al., 2013). Determining how muscle cells link every RNA molecule to a kinesin motor for transport, and how specificity can be achieved in light of this general requirement will be the focus of future investigation. RNA-binding proteins (RBPs) may act as adaptors to link RNA molecules to kinesins. It will be important to elucidate the putative “code” underlying the combinatorial interactions between mRNA sequences, RBPs, and motors. Another promising substrate for generalized transport is RNA “hitchhiking” on trafficked vesicles, recently observed in neurons, which may reveal its own code of vesicle markers and RNA adapters (Liao et al., 2019).

What are the consequences of pertrurbing mRNA transport in muscle? In the cardiomyocyte, preventing the microtubule-dependent transport of mRNA to peripheral regions of the cell limited protein synthesis and cell growth. In skeletal muscle, protein synthesis outside of the perinuclear region was also eliminated by disrupting RNA distribution (Denes et al., 2021); further work is required to determine the functional impact on homeostasis and growth. Importantly, it remains unclear why RNA distribution is required to maintain protein synthesis, especially in light of relatively high protein mobility. One possibility is that different environments (i.e. perinuclear versus nucleus-distal regions) are differentially permissible to translation of specific mRNAs. Another possibility is that ribosomes become limiting when RNAs are concentrated around the nucleus. Since ribosomes and polymerases are assumed to be limiting (Lin & Amir, 2018), local ribosomes could perhaps become “saturated” if RNAs do not move to other areas of the cell. The microtubule network is disrupted in Duchenne muscular dystrophy (Khairallah et al., 2012) and in cardiomyopathies (Caporizzo et al., 2019). How the presumed disruption of RNA transport and local protein synthesis contributes to these pathologies represents an exciting area of future investigation with important considerations for general myofibre homeostasis. Finally, gene therapies for Duchenne and other myopathies that rely on delivery and expression of genetic material from relatively few myonuclei in the muscle syncytium will need to reckon with the necessity of efficient RNA distribution to facilitate maximal expression of therapeutic cargoes (Morin et al., 2023).

New Insights on the Myonuclear Domain from Single Myonuclear RNA-Sequencing (smnRNA-seq)

The application of single nuclear RNA-sequencing to skeletal muscle has broadened our understanding of regional regulation of transcription within muscle fibres (Williams et al., 2022). The initial smnRNA-seq studies in mice emphasized specialization of myonuclei as well as transitional states during development. For instance, smnRNA-seq provided more granular insight into specialized myonuclei of the neuromuscular junction (NMJ) and myotendinous junction (MTJ). NMJ-specific myonuclei cluster closely and have a unique arrangement, aligned adjacent to acetylcholine receptors (Hastings et al., 2020). In addition to known factors such as acetylcholinesterases, myonuclei of the NMJ are enriched for several additional genes such as Musk, Lrp4, Colq, and Etv5 (Dos Santos et al., 2020; Petrany et al., 2020). Myonuclei of the MTJ are also enriched for specific transcripts such as Col22a1 and Ankrd1 (Marp1) (Dos Santos et al., 2020; Petrany et al., 2020), and can be classified into two distinct populations (Kim et al., 2020). Ankrd1 is induced by an acute bout of loading in muscle (Barash et al., 2004; Murach et al., 2022) and locks titin to the thin filament, regulates passive force, and protects the sarcomere from mechanical damage (van der Pijl et al., 2021). Since myonuclei near the MTJ are likely experiencing high deformation with contraction, we speculate that Anrkrd1-expressing myonuclei near the MTJ are primed for responsiveness to muscle loading. The expression of myosin also differs across myonuclei according to the muscle of origin. Nuclei expressing several myosins are more abundant in the oxidative soleus than the glycolytic extensor digitorum longus (Dos Santos et al., 2020). Varying myosin expression patterns across myonuclei potentially sheds light on the regional regulation of myosin heavy chain (MyHC) fibre type within a myofibre (Staron & Pette, 1987; Zhang et al., 2010; Sawano et al., 2016; Murach et al., 2019b). Myosin co-expressing myonuclei could also have consequences for maintenance of muscle fibre type in conditions such as disease, ageing, and inactivity where “hybrid fibres” become prevalent (Murach et al., 2019b). Heterogeneous myonuclei with unique transcriptional profiles may also be dispersed throughout muscle fibres, and this could depend on where fibres are anatomically located within the muscle (i.e. superficial versus deep) (Kim et al., 2020).

The heterogeneity of myonuclei is further revealed under dynamic conditions. Subsets of neighboring myonuclei adopt distinct transcriptional profiles associated with repair in muscular dystrophy, a condition characterized by chronic membrane damage and degeneration/regeneration (Kim et al., 2020). Some myonuclei in dystrophic muscle are also enriched for apoptotic markers that are localized specifically in necrotic fibres (Chemello et al., 2020). During developmental muscle growth, a specific subset of myonuclei emerge throughout muscle fibres that are in “sarcomere assembly states” and could be specialized for growth (Petrany et al., 2020). Following denervation, myonuclear subgroups undergo transcriptional reprogramming and express a distinct signature that includes increased Runx1 and Gadd4 (Lin et al., 2022). Runx1 is enriched in myonuclei during loading-induced hypertrophy (Murach et al., 2022), suggesting Runx1 expression during atrophy is a compensatory response. In response to a bout of endurance/resistance exercise in the absence of satellite cells, a population of “cryptic” myonuclei arise that lack a clear identity (Wen et al., 2021). Transcriptionally unspecified myonuclei may in part explain blunted hypertrophy when myonuclear addition is prevented during exercise training (Englund et al., 2019; Englund et al., 2020a; Murach et al., 2021b). The location of these myonuclei in the muscle fibre remains to be determined. Satellite cell-derived myonuclei may also specifically provide ribosomal proteins during loading-induced adult muscle hypertrophy (Murach et al., 2021a). We speculate this addition occurs throughout the myofibre and could have implications for the myonuclear domain (Ralston et al., 1997). Some preliminary evidence also suggests that myonuclei from newly-fused satellite cells may contribute certain transcription factors to growing muscle fibres (Murach et al., 2020b; Murach et al., 2021a). The specific triggers for myonuclear accretion by satellite cells during exercise is discussed at length elsewhere (Murach et al., 2021b), but may be driven by contraction-induced signals from the muscle fibre (Serrano et al., 2008; Ross et al., 2017; Ross et al., 2018; Battey et al., 2022; Noviello et al., 2022).

Collectively, transcriptional compartmentalization by myonuclei in muscle fibres at rest and during remodeling could have consequences for myonuclear domain regulation. Combining smnRNA-seq with emerging techniques such as RNA-scope (Kann & Krauss, 2019) and spatial transcriptomics in muscle (McKellar et al., 2021; McKellar et al., 2022; Young et al., 2022) will further define myonuclear subpopulations and their influence on myonuclear specialization and compartmentalization. Future work should focus on understanding the mechanisms that can establish and maintain specialized domains within a shared syncytia. At the neuromuscular junction, for example, microtubule organization is dramatically different (Osseni et al., 2020; Parato & Bartolini, 2021) and may serve to preserve transcripts produced in subsynaptic myonuclei – including important NMJ transcription factors – nearby, creating a positive feedback loop. Understanding the transcriptional drivers and regulatory networks responsible for various myonuclear states will shed light on development and disease.

Evidence for Flexibility within the Myonuclear Domain

The flexibility of the myonuclear domain pertaining to rodents and humans has been discussed at length elsewhere (Murach et al., 2018a; Murach et al., 2021b; Prasad & Millay, 2021), but several points are worth mentioning here. Skeletal muscle myonuclear domain size differs between species and broadly scales according to body size, but is quite similar between mice and humans (Liu et al., 2009). Myonuclear domains are also reportedly smaller in women compared to men (Horwath et al., 2021). Myonuclear domain size appears highly dependent on myosin heavy chain (MyHC) fibre type, with a general continuum of smaller myonuclear domains in slow-twitch fibres (MyHC I) and larger domains in fast-twitch fibre (MyHC IIa, IIa/IIx, and IIx) in humans and larger mammals (Liu et al., 2009). Regardless of species, sex, or fibre type, myonuclear domain size can change rapidly during times of muscle fibre hypertrophy or atrophy.

The extent to which the myonuclear domain can adjust its size without the addition of new myonuclei during growth, and if there is an upper limit, is of importance for understanding skeletal muscle homeostasis in health and disease. Given that myonuclei are thought to be post-mitotic and unable to replenish themselves, the fusion of Pax7+ satellite cells is the primary source of new myonuclei to muscle fibres (Lepper et al., 2011; McCarthy et al., 2011; Murphy et al., 2011; Sambasivan et al., 2011). When satellite cell fusion is prevented during post-natal developmental growth in mice (Cramer et al., 2020; Hansson et al., 2020a) or when satellite cells are deleted (Bachman et al., 2018; Bachman et al., 2020; Bachman & Chakkalakal, 2021), muscle fibres are smaller but the myonuclear domains are large. Similarly, satellite cell depletion early in the pathology of muscular dystrophy results in muscle fibres with large myonuclear domains (Boyer et al., 2022). Plasticity during development is indicative of transcriptional reserve capacity in resident myonuclei which likely underlies the observed flexibility of the myonuclear domain in adult muscle fibres.

When the muscles of adult mice (>4 months old) are mechanically loaded in the absence of satellite cells, an induction of transcription and expansion of the myonuclear domain made possible by transcriptional reserve in resident myonuclei is apparent during short-term hypertrophy (McCarthy et al., 2011; Kirby et al., 2016; Murach et al., 2017; Murach et al., 2020b). Without myonuclear accretion from satellite cells, both oxidative and glycolytic muscles can hypertrophy to a significant degree with loading in the long-term in mice (≥8 weeks) (Fry et al., 2014a; Fry et al., 2017; Englund et al., 2020a; Englund et al., 2020b; Kaneshige et al., 2021; Murach et al., 2021c). Eventually, however, adult muscle fibres do not hypertrophy to the same extent with exercise in the absence of satellite cells (Englund et al., 2020a; Englund et al., 2020b; Murach et al., 2021b). Exercise-induced hypertrophy is also blunted in muscle of adult mice when satellite cell fusion is prevented via genetic manipulation (Goh et al., 2019). As is often speculated, there is likely an upper limit to the “myonuclear domain” during adult skeletal muscle hypertrophy (Petrella et al., 2008; Conceicao et al., 2018; Murach et al., 2021b; Murach et al., 2021c). Some evidence also suggests the myonuclear domain during hypertrophy could be influenced by the oxidative capacity of a muscle fibre (Fry et al., 2014b; Omairi et al., 2016). The non-fusion mediated effects of satellite cells on muscle hypertrophy should not be overlooked, however, since satellite cells communicate to muscle fibres at the onset of hypertophy prior to fusion (Murach et al., 2020b). Upon exceeding a potential ceiling, additional myonuclei may need to be added in order to stabilize the myonuclear domain and allow further growth. Future work should more closely examine the interplay between hypertrophy, the myonuclear domain, and fibre type transitions in mice during exercise since fibre size varies significantly between types at baseline (Zhu et al., 2021). How the myonuclear domain is measured and/or normalized could influence interpretation, so other factors that may be worth considering with regard to studying the myonuclear domain during hypertrophy are muscle fibre perimeter versus cross-sectional area (Moesgaard et al., 2022), muscle fibre surface area-to-volume ratio (Prasad & Millay, 2021), as well as the initial size and myonuclear density of a fibre prior to hypertrophy (Snijders et al., 2016; Snijders et al., 2021).

The myonuclear domain is flexible during muscle atrophy. It can shrink concomitant with the maintenance of resident myonuclei (Bruusgaard et al., 2012) independent from satellite cells (Jackson et al., 2012) in the short term. After a prolonged period of atrophy, however, emerging evidence suggests that myonuclei are lost (Sandonà et al., 2012; Murach et al., 2018b; Dungan et al., 2019; Snijders et al., 2019; Murach et al., 2020a; Viggars et al., 2022b). Myonuclear loss perhaps occurs as a strategy to stabilize the myonuclear domain. The prevalence and potential mechanisms of myonuclear removal was recently debated at length (Kirby & Dupont-Versteegden, 2022a; Kirby & Dupont-Versteegden, 2022b; Schwartz & Gundersen, 2022a; Schwartz & Gundersen, 2022b) and is an open area of inquiry (Murach et al., 2019a; Eftestøl et al., 2020; Snijders et al., 2020; Rahmati et al., 2022).

The role of myonuclear motility is a burgeoning area of interest with respect to the myonuclear domain during adult muscle adaptation. It is well-established in invertebrates that myonuclei move within the muscle fibre in response to various stimuli (Manhart et al., 2018; Roman & Gomes, 2018; Windner et al., 2019; Azevedo & Baylies, 2020), perhaps for the purpose of influencing myonuclear domains. This movement is mediated by motor proteins such as kinesin and dynein (Folker et al., 2014; Gache et al., 2017; Azevedo & Baylies, 2020; Roman et al., 2021). In Drosophila, myonuclei move closer to the myotendinous junction after stretch, possibly to promote repair (Perillo & Folker, 2018). Myonuclear movement may be facilitated by contraction in mammalian muscle fibres (Roman et al., 2017). Mammalian myonuclei are capable of moving to sites of focal injury to assist with sarcomere reconstruction via targeted protein synthesis, independent from satellite cell participation (Roman et al., 2021). The movement of myonuclei in adult muscle could be related to the activity of the microtubule-associated protein MACF1 (Ghasemizadeh et al., 2021). Resident myonuclei may relocate centrally during adaptation to exercise in vivo in adult muscle (Murach et al., 2020a); this could be tied to muscle repair. Central myonuclei are reportedly the most transcriptionally active during muscle repair in response to severe injury (Buckley et al., 2022). These data provide some evidence for a spatial component to myonuclear transcriptional activity. Myonuclei also change shape in response to exercise training in rodents and humans (Battey et al., 2023; Murach et al., 2020a; Rader & Baker, 2022), which could be linked to myonuclear transcriptional activity and/or movement (Folker et al., 2014; Kirby & Lammerding, 2016; Kirby & Lammerding, 2018). More work is needed on the causes and consequences of changes in myonuclear morphology under different conditions, and the implications for the myonuclear domain.

Collectively, there appears to be flexibility in the myonuclear domain during developmental muscle growth as well as in adult muscle. The flexibility of the myonuclear domain is likely made possible via transcriptional reserve of resident myonuclei, and perhaps the ability of myonuclei to relocate and facilitate specialized gene expression and protein synthesis during times of stress (Roman & Muñoz-Cánoves, 2022).

The Future

Conclusions and Open Areas of Inquiry

Our understanding of the myonuclear domain has advanced significantly, especially in the last 20 years with the development of new technologies in cell biology. We now appreciate that the transcriptional output of myonuclei within a muscle fibre is heterogeneous, that mRNAs initially localize around myonuclei and then are transported along microtubules, and certain transcripts and proteins are transported long distances within the fibre. We also have evidence showing that myonuclei are motile in adult muscle and can be gained and lost under various conditions. All together, these discoveries reveal the challenges in coordinating transcriptional and translational output in voluminous syncytial muscle cells. It should be considered that the heterogeneity across distinct myonuclear domains is primarily based on static examinations of gene and protein expression at specific time points. This ‘snapshot’ is unable to observe pulsatile gene expression patterns (Newlands et al., 1998) and myonuclear movements in muscle fibres (Roman et al., 2021), which may make the myonuclear domain more homogenous over time.

Moving forward, it will be important to identify the limits of the myonuclear domain during adult muscle growth and the potential mechanisms of myonuclear removal during atrophy, as well as the triggers and mechanisms for myonuclear movement during muscle remodeling. Information on how mRNA transport from myonuclei may be perturbed in pathological conditions could inform specific myonuclear-targeted gene therapies. A more detailed understanding of myonuclear specialization during health and disease could result in an updated framework for the basic mechanisms of muscle mass regulation and metabolism.

Using emerging technologies, effort should be directed toward studying the mechanisms by which myonuclei communicate and coordinate with one another; this should be done so with consideration of sex, age, patient populations, MyHC fibre type, and muscle of origin. Most information on the myonuclear domain has related to the production/localization of mRNA and the protein from that message; essentially nothing is known about non-coding RNA in this context. Recent advances in spatial transcriptomic profiling may illuminate how non-coding and ribosomal RNA behave with respect to myonuclear domains (McKellar et al., 2022). Advances in spatial proteomics will further augment these efforts (Lundberg & Borner, 2019; Mund et al., 2022). Attention should also be directed toward understanding what satellite cell-derived myonuclei specifically contribute to adult muscle fibres at the molecular level (Murach et al., 2020b; Murach et al., 2021a). The focus regarding satellite cell contributions to the myonuclear domain has understandably been on myonuclei and RNA, but perhaps the delivery of mitochondria (which has its own DNA) is also worth considering (Wang et al., 2022). Modern technology applied to the study of myonuclei could shed new light on dogmatic beliefs in muscle, such as whether mammalian myonuclei are truly post-mitotic (Stockdale & Holtzer, 1961; Moss & Leblond, 1970; Pullman & Yeoh, 1978; Clegg & Hauschka, 1987), which has recently been challenged (Borowik et al., 2022). The identification of myonucleus-specific protein markers could help address these questions and others across species; at the current time, a highly-specific marker is seemingly lacking (Winje et al., 2018; Viggars et al., 2022a). Finally, continued energy toward linking alterations in the myonuclear domain to contractile function (Qaisar et al., 2012), and what may underlie this relationship, is of practical significance for understanding what maintains muscle health throughout the lifespan as well as the etiology of disease (Levy et al., 2018; Ross et al., 2018; Battey et al., 2020; Liu et al., 2020). Addressing the fundamental gaps in our understanding of the myonuclear domain could lead to interventions that improve human muscle health throughout the lifespan.

Key Points.

• Various mechanisms and conditions contribute to regulation of the “myonuclear domain”, or theoretically finite area that an individual myoncleus provides gene products to within a multinuclear muscle fiibre

• mRNA from an individual myonucleus and the protein it produces are actively transported by molecular motors along microtubules in muscle fibres, as are myonuclei

• The distance that an mRNA or protein travels away from its myonucleus of origin is variable based on a variety of known and unknown factors

• Current evidence indicates that myonuclei are added during muscle hypertrophy fibre and lost during atrophy, perhaps to stabilize the myonuclear domain

• Myonuclei within a muscle fibre are transcriptionally and functionally diverse, which may have consequences for myonuclear domain maintenance and muscle fibre homeostasis

Biographies

James R. “Jimmy” Bagley, PhD, is an Associate Professor of Kinesiology and Director of the Muscle Physiology Lab at San Francisco State University. His research focuses on exercise physiology, with the overarching goal to better understand human performance. To study skeletal muscle cells and connective tissue, his Lab utilizes advanced imaging techniques including super-resolution confocal and atomic force microscopy. Dr. Bagley earned his PhD in Human Bioenergetics from the Ball State University Human Performance Lab, MS from Cal State University-Fullerton, and BS from Cal Poly-San Luis Obispo.

Lance Denes, PhD, is a post-doctoral fellow in the Institute for Systems Genetics at New York University – Langone Health. He did his undergraduate work at the University of Florida as well as obtained his PhD there working with Eric Wang, PhD. Lance uses fluorescent microscopy and molecular biology techniques to understand how molecules in cells are transported.