Muscle spindles (MS) are mechanosensory organs that detect and mediate static and dynamic information about skeletal muscle fibre length, as well as rate and extent of strain (Thornell et al., 2015). There are three sub-types of intrafusal fibre; nuclear bag₁, bag₂ and chain, each have a unique morphology, innervation pattern and protein expression profiles, which contribute to a distinctive functional response to static and dynamic muscle stretch. Nuclear bag fibres have a distinctive fusiform shape with a clustering of nuclei at the equatorial region, nuclear chain fibres have a linear morphology and nuclear alignment (Macefield and Knellwolf, 2018). MS provide a significant contribution to proprioceptive function, which can be described as the sense of position and movement of parts of the body relative to one another (Macefield and Knellwolf, 2018). Impairment of proprioception and dysfunction of the MS is linked with many neuromuscular diseases including; multiple sclerosis (Feys et al., 2011; Prather et al., 2011; Fling et al., 2014) Parkinson’s disease (Conte et al., 2013; Teasdale et al., 2017), muscular dystrophy (Ovalle and Dow, 1986), and peripheral nerve injuries (Cope et al., 1994; Haftel, 2005; Ruijs et al., 2005; Maas et al., 2007; Bullinger et al., 2011; Prather et al., 2011). Proprioceptive function also deteriorates in diabetic patients (van Deursen et al., 1998; Muller et al., 2008; Muramatsu et al., 2017) and during aging (Miwa et al., 1995; Shaffer and Harrison, 2007). Proprioceptive dysfunction can cause a considerable alteration in the regulation of the speed and precision of limb movement. These effects cause significant physical limitations, including; disruption to balance, locomotion and postural stability (van Deursen et al., 1998; van Deursen and Simoneau, 1999; D’Silva et al., 2016; Ettinger et al., 2018; Ferlinc et al., 2019), which can significantly negatively impact upon the individual’s quality of life (Teasdale et al., 2017).

The muscle spindle proprioceptive system have been extensively studied in vivo, whereby studies conducted in mice, rats and cats have formed the basis of much of our current understanding (Banks, 1994, 2015; Macefield and Knellwolf, 2018). Intrafusal fibres develop from primary myotubes, whereby the onset of intrafusal specification coincides with Ia afferent innervation and initiates spindle morphogenesis (Chal and Pourquié, 2017). Approaching Ia afferents release neuregulin-1 (Nrg-1), which binds primary muscle tyrosine kinase receptors ErbB (2-4) (erythroblastic leukemia viral oncogene homologue), resulting in downstream expression of the early growth response protein-3 (Egr3) (Tourtellotte and Milbrandt, 1998; Tourtellotte et al., 2001; Albert et al., 2005; Oliveira Fernandes and Tourtellotte, 2015). Disrupting any of these key mechanisms in vivo results in aberrant MS development and function (Tourtellotte and Milbrandt, 1998; Andrechek et al., 2002; Hippenmeyer et al., 2002; Leu et al., 2003; Jacobson et al., 2004; Albert et al., 2005; Herndon et al., 2014; Oliveira Fernandes and Tourtellotte, 2015). To date most skeletal muscle models often neglect the integrated nature of the neuromuscular system by not accounting for afferent functions mediated via the muscle spindle. Such experiments are expensive, time consuming and extremely difficult in vivo considering the an adult human contains only about 50,000 muscle spindles (Kröger and Watkins, 2021).

Therefore, in vitro models of the MS offer the possibility to study the cellular and molecular mechanisms regulating development, function and the effect of injury and disease, in a highly controlled environment. To develop a representative model in vitro, it is necessary to identify and recapitulate native tissue characteristics. In vivo, postnatal intrafusal fibres contain a higher number of paired-domain transcription factor 7 (PAX7) positive satellite cells and retain the expression of embryonic satellite cell marker, paired-domain transcription factor 3 (PAX3) (Horst et al., 2006; Kirkpatrick et al., 2008, 2010). Their myonuclei maintain the expression of an early myogenic regulatory factor (MRF), Myf5 (Zammit et al., 2004), they remain comparatively small and have reduced myonuclear domains (volume of sarcoplasm per myonucleus) compared to extrafusal fibres (Kozeka and Ontell, 1981). They also have preferential expression of embryonic (MyHC3), neonatal (MyHC8) and specialised (MyHC6, MyHC7b) myosin heavy chains (MyHC) (Walro and Kucera, 1999; Liu et al., 2002), concomitant with the retained expression of developmental protein, Egr3 (Tourtellotte and Milbrandt, 1998; Oliveira Fernandes and Tourtellotte, 2015). To this end, when developing an in vitro model, it would be prudent to measure the characteristics outlined above.

Despite this, characterisation in vitro has largely relied on morphological identification of what have been termed “bag myotubes” (intrafusal nuclear bag fibres), alongside expression of developmental proteins or putative phenotypical proteins of intrafusal fibres as characterised in vivo (Jacobson et al., 2004; Rumsey et al., 2008; Colón et al., 2017, 2020; Guo et al., 2017; Qiao et al., 2018). Previous in vitro studies for the purpose of developing de novo intrafusal muscle fibres have indicated that Nrg-1 treatment (100–160 ng/ml, 12.5–20 nM) causes an increase in bag-like myotubes with an expanded equatorial region and centrally clustered nuclei in primary human and rat cells (Rumsey et al., 2008; Colón et al., 2017; Guo et al., 2017; Qiao et al., 2018), paired with high magnification images of bag myotube specific MyHC6 and Egr3 staining. (Rumsey et al., 2008; Colón et al., 2017). To date, the only quantitative protein (western blot) analysis of myogenic cells following Nrg-1 (8 ng/ml, 1 nM) treatment indicated a sharp increase of approximately 6-fold in both MyHC8 and a slow developmental isoform (Jacobson et al., 2004) (in other papers this refers to the s46 antibody (Tourtellotte et al., 2001; Albert et al., 2005), which recognises MyHC6) (Jacobson et al., 2004). However, unlike the Hickman papers, MyHC6 immunoreactivity is not exclusive to the Nrg-1 treated myotubes and there is no mention of gross morphological changes following Nrg-1 treatment and therefore no mention of bag myotube specificity associated to MyHC6.

Current models have now progressed rapidly to incorporate afferent and efferent innervation with induced pluripotent stem cell (iPSC)-derived intrafusal myotubes, while overlooking some basic structural and molecular characterisation (Colón et al., 2017; Guo et al., 2017). The extent of de novo, in vitro intrafusal fibres structural and physiological similarity to native intrafusal fibres is not yet fully elucidated. Therefore, there is a need to further characterise cell engineered, Nrg-1 induced intrafusal myotubes before they can become platforms to integrate muscle spindle associated proprioceptive physiology, disease state modelling or clinical applications. Furthermore, despite the advances in iPSC and primary cell technologies, there are significant challenges with using these cells. Myogenic cell lines (e.g., the C2C12 cell line) provide established, fusion-competent cells, are affordable, sustainable and are easy to maintain. However, the capacity for intrafusal myotube differentiation following Nrg-1 supplementation in myogenic cell lines is currently unknown.

Current literature relies a subjective determination of intrafusal morphology by the investigator (Rumsey et al., 2008, 2010; Colón et al., 2017, 2020; Guo et al., 2017). To improve objectivity, statistical morphological parameters for assigning intrafusal bag myotubes from a heterogeneous population need to be defined. Secondly, MyHC isoform protein sequences are very similar, which has suggested limitations caused by cross-reactivity of antibodies (Thornell et al., 2015). Since MyHC expression is controlled at the transcriptional level (Schiaffino et al., 2015), MyHC gene expression could be a suitable alternative to detect intrafusal specific patterns (Brown et al., 2012).

In addition, there are muscle spindle specific targets, which have not yet been applied for in vitro characterisation. These are; Ets variant 4 (Etv4) (Arber et al., 2000; Hippenmeyer et al., 2002), Glial cell derived neurotrophic factor (Gdnf) (Shneider et al., 2009; Schiaffino and Reggiani, 2011), neurotrophin-3 (NT3) (Chen et al., 2002; Kröger and Watkins, 2021), low- affinity neurotrophin receptor p75 (Ngfr), somatostatin receptor type 2 (Sstr2) and the type III inter- mediate filament peripherin 1 (Prph1) (Albert et al., 2005).

In vitro skeletal muscle research has gained momentum over the past 10 years, however there are still relatively few publications tackling cell and tissue engineering approaches to generating de novo intrafusal muscle fibres in vitro (Barrett et al., 2020) and they are not without some inconsistencies, described previously. This study presents an objective, novel method for assignment of intrafusal bag morphology and the most in-depth characterisation of myotubes following Nrg-1 treatment to-date. The aim was to determine the degree to which Nrg-1 induces an intrafusal fibre-like morphology and the extent the myotubes recapitulate features of native intrafusal fibres. Developing a well characterised in vitro model of intrafusal fibres using C2C12 myoblasts, will provide an investigative tool for developmental physiology, the framework for multi-cell, innervated, integrated models of the muscle spindle and relevant disease models in a defined, controlled and highly reproducible manner.

There is an important clinical need to develop robust in vitro models of skeletal muscle that account for the sensory function of the muscle spindle through integration of intrafusal skeletal muscle fibres. This will help to investigate the basic molecular and cellular mechanisms regulating their phenotype in health and disease, in a controlled, defined, and ethical environment. In this study, a reductionist, monocellular C2C12 model, for de novo intrafusal skeletal muscle generation is presented. The addition of a recombinant, developmentally associated protein, Nrg-1, is used to replicate an innervating Ia afferent neuron. C2C12 myoblasts offer an accessible, fusion competent, reproducible cell line that are widely used in a number of skeletal muscle research applications. C2C12s facilitate rapid model progression, assay development and also an opportunity to investigate intrafusal fibre development in a reductionist pure myoblast population. Progress in such a cell line will facilitate translation to a primary human, multi-cell, integrated, biomimetic model of the muscle spindle.

To firstly investigate the utility of C2C12s as a model of intrafusal fibre development, Nrg-1 was added to developing C2C12 myotubes, which resulted in an over 4-fold increase of bag myotube formation (Figure 3C). This supports previous in vivo and in vitro experiments (Tourtellotte and Milbrandt, 1998; Tourtellotte et al., 2001; Jacobson et al., 2004; Albert et al., 2005; Rumsey et al., 2008; Akay et al., 2014; Herndon et al., 2014; Oliveira Fernandes and Tourtellotte, 2015; Colón et al., 2017, 2020; Guo et al., 2017) that indicate Nrg-1 is essential for the development of intrafusal skeletal muscle fibres. The fold increase in bag myotubes is comparable to Hickman and others, who reported 4-5-fold increases in both primary rat and human cells (Rumsey et al., 2008; Guo et al., 2017). However, the percentage of bag myotubes to the total myotube population is much higher in our C2C12 model compared to human myoblasts as previously published (41.61 ± 6.77% vs 15.80 ± 6.62%). Nevertheless, we must consider in this study, a novel statistical DDR method was used to more objectively define intrafusal bag myotubes, which may be a contributing factor, rather than suggesting C2C12s are more responsive to Nrg-1.

Myoblast fusion is a required event for skeletal muscle development and regeneration, therefore measuring fusion efficiency in vitro is a good indicator for overall myogenesis (Sampath et al., 2018). Fusion efficiency was not significantly changed upon Nrg-1 addition in this investigation (Figure 3C), which would suggest that the morphological changes observed are independent of other myogenic fusion parameters. There was however, a 1.4-fold increase in total nuclei number per FOV, which did not affect fusion efficiency (Figure 3B). This change can be accounted to the increase of bag myotubes, which contain on average over two more myonuclei than linear fibres (Figure 3C).

Nrg-1 downstream activation of Egr3 is a key signalling event in muscle spindle and intrafusal skeletal muscle fibre development (Tourtellotte and Milbrandt, 1998; Jacobson et al., 2004; Albert et al., 2005; Herndon et al., 2014; Oliveira Fernandes and Tourtellotte, 2015). Ablation of Erg-3 is accompanied by abnormal muscle spindles, reduced running performance and degradation of locomotor pattern in mice (Tourtellotte et al., 2001; Akay et al., 2014; Oliveira Fernandes and Tourtellotte, 2015). To further validate the role of Nrg-1 on C2C12 intrafusal bag myotube formation, Egr3 expression was quantified. Treated cells displayed a significant increase in gene (Figure 3D) and protein expression (Figure 6D) of Egr3, similar to that previously evidenced in primary human (Jacobson et al., 2004) and C2C12s (Williams and Jacobson, 2010; Herndon et al., 2014). Fluorescent micrographs demonstrated increased Egr3 protein staining intensity in Nrg-1 treated and intrafusal bag C2C12 myotubes, suggesting an important role for Egr3 toward C2C12 intrafusal bag myotube development. In contrast to the Hickman papers (Rumsey et al., 2008, 2010; Colón et al., 2017, 2020; Guo et al., 2017), control myotubes had visible expression of Egr3 (Figure 6A), meaning it was not a binary determinant of the intrafusal phenotype as defined by DDR. In previous studies, Egr3 displayed strong immunoreactivity in untreated myogenic cells (Herndon et al., 2014), is involved in in vitro myoblast proliferation (Kurosaka et al., 2017) and upregulated during differentiation (Muñoz et al., 2018; Choi et al., 2020). Therefore, non-binary staining results were expected. Egr3 in Figure 6, paired with the increase in intrafusal bag morphology (Figure 3B) suggests that Nrg-1 elicits endogenous production of Egr3, suggesting activation of the Nrg1/ErbB2/Egr3 signalling pathway, resulting in intrafusal specific differentiation.

To validate an intrafusal phenotype, the most in depth MyHC characterisation of Nrg-1 treated myotubes was completed. MyHCs are key contractile proteins and are major determinants of force velocity properties of muscle, which are differentially distributed across fibres and transiently expressed during development, regeneration and various stimulus such as injury or exercise (Schiaffino et al., 2015). Therefore, MyHCs are often used to provide an indication of the developing, regenerative or mature phenotype of skeletal muscle (Liu et al., 2005). In vivo studies have highlighted MyHC3, MyHC6, MyHC8 and MyHC7b/14 as having preferential or retained expression in mature intrafusal muscle fibres (Kucera et al., 1992; Liu et al., 2002; Österlund et al., 2011; Schiaffino and Reggiani, 2011; Thornell et al., 2015) and therefore are suitable candidates for characterisation in vitro. MyHC6 and MyHC8 have been formerly shown to display increased protein expression in human cells in vitro following Nrg-1 treatment (Jacobson et al., 2004), even in the absence of bag myotube morphology. In vitro bag myotubes have displayed MyHC6 specific staining compared to their morphologically linear neighbours in primary rat (Rumsey et al., 2008), primary human (Guo et al., 2017) and iPSCs (Colón et al., 2020). In contrast, this current paper presents that MyHC3,6 and 8 are all downregulated transcriptionally when analysed by RT-qPCR (Figures 4A,B), translationally when analysed by western blots (Figures 5A–C), and exhibit no clear preferential immunofluorescent staining in C2C12 myotubes with a bag morphology (Figures 5A–C). MyHC6 presented sparse staining in Nrg-1 vs control, as compared to the immature isoforms. Although MyHC6 is a predominately cardiac isoform, it is expressed in slow-twitch skeletal muscle tissue and its reduced expression in Nrg-1 treated cells could be related to a faster MyHC profile, as indicated by the increased Myh4 gene expression (Stuart et al., 2016) Furthermore, s46 (DSHB, United States), an antibody for avian slow developmental isoform, is repeatedly stated as being the best intrafusal specific marker in vivo and in vitro (Schiaffino and Reggiani, 2011; Schiaffino et al., 2015) gave no immunoreactivity in the present model (data not shown). The only Myh increase was a 2-fold increase in Myh4 following Nrg-1 treatment (Figure 4A), representing a rise from 33.24 ± 0.85% to 60.13 ± 0.82% of total Myh expression (Figure 4B). Myh4 codes for MyHC-IIb, the most prominent isoform (Pellegrino et al., 2003) expressed in adult mice, associated with high forces of contraction combined with rapid contractile characteristics (Harrison et al., 2011). MyHC4/IIb Immunofluorescence and western blotting were attempted using both BF-F3 (DSHB, United States) and 20140-1-AP (Proteintech, United Kingdom) antibodies. Unfortunately, they proved unsuccessful in achieving a clean and strong signal which would be sufficient for use. Intrafusal bag fibres, especially of the type 1 subset, are associated with a slow phenotype, however intrafusal chain and bag₂ fibres express a fast phenotype in both human and rats (Kucera et al., 1992; Soukup et al., 1995; Liu et al., 2005; Österlund et al., 2013). Therefore, an increase in Myh4 is not indicative of poor intrafusal myotube generation in this model, but rather an indication of a more mature, glycolytic and faster contractile phenotype following Nrg-1 supplementation (Kucera et al., 1992; Soukup et al., 1995; Liu et al., 2005; Österlund et al., 2013). Furthermore, previous studies have demonstrated the increased expression of Myh4 mRNA in C2C12 myotubes in similar time course of experimentation (Brown et al., 2012). In addition, rat intrafusal fibres are believed to originate from three sequential generations of myotubes, the MyHC expression displays regional variability across the fibre length, differs throughout development and is dependent on interactions with external factors such as innervation and intracapsular niche (Kucera and Walro, 1990, 1995). This highlights the complexity of using MyHC expression to define an intrafusal myotube in vitro in the absence of sensory innervation and functional outputs (Kucera et al., 1992; Liu et al., 2002).

To characterise this further, supplementary intrafusal markers were employed. Myf5, an early myogenic regulatory factor, retains expression into adult intrafusal muscle fibres in-vivo (Zammit et al., 2004). Etv4 transcription factor, is another early stage intrafusal muscle fibre development marker, which similar to Egr3 is induced by afferent innervation (Arber et al., 2000; Hippenmeyer et al., 2002). However, Etv4 is largely involved in patterning motor innervation rather than in spindle morphogenesis per se (Albert et al., 2005). Gdnf is another established intrafusal fibre marker that has an essential role in fusimotor survival (Shneider et al., 2009; Schiaffino and Reggiani, 2011). Additionally, a murine knockdown model identified Sstr2 and Prph1 as novel intrafusal specific markers (Albert et al., 2005), which are yet to be utilised for in vitro characterisation. Despite this, gene expression changes for the aforementioned targets, aside from a significant 2-fold increase in Etv4 (Figure 6E), were insignificant following Nrg-1 treatment (Figures 3F, 6E). Activation of Etv4 expression further clarifies Nrg-1 is replicating afferent signalling pathways in this model. Although Myf5 gene expression was not modified, Myod1 was significantly upregulated. As an intermediate MRF, Myod1 is required for myogenic cells to exit the cell cycle and to enter the differentiation process (Zanou and Gailly, 2013). With the evidence for increased cell proliferation during the differentiation phase in Nrg-1 treated culture (Figure 7A), the increase in Myod1 compared to control could be accounted to a greater availability of myoblasts for fusion at the given time point. Myod1 also has a causative relationship with increased Myh4 expression in mice, which may also be a contributing factor to the upregulation of Myh4 discussed above (Wheeler et al., 1999; Seward et al., 2001; Ekmark et al., 2007; Zammit, 2017).

Intrafusal fibres have an increased expression of PAX 7 satellite cells and decreased myonuclear domain compared to their extrafusal neighbours (Kirkpatrick et al., 2008). This highlights the importance of satellite cell and myonuclei distribution to muscle spindle and intrafusal fibre function (Kirkpatrick et al., 2008). In corroboration to this hypothesis, adult mice ablated of satellite cells exhibit no detectable changes in extrafusal muscle fibre morphology, fibre-type composition, aerobic capacity or force generation. However, they display gross motor coordination defects, increased muscle spindle extracellular matrix deposition and decreased intrafusal fibre cross-sectional area (Jackson et al., 2015). A similar phenotype is seen with skeletal muscle specific Egr3 knockout mice, whom are deficient of a functional muscle spindle (Tourtellotte and Milbrandt, 1998; Akay et al., 2014; Oliveira Fernandes and Tourtellotte, 2015). Increased satellite cell number and smaller myonuclear domains are indicative of greater capacities for growth, regeneration, and repair (Rosser et al., 2002; Allouh et al., 2008; Shenkman et al., 2010). In general, higher satellite cells numbers corresponds with greater contractile activity (Gibson and Schultz, 1982, 1983; Yin et al., 2013) and myonuclear domains are smaller in slower, frequently activated fibres (Roy et al., 2005; Aravamudan et al., 2006). Therefore, intrafusal fibres, which are continuously sending and receiving information to the central nervous system, even at rest (Macefield and Knellwolf, 2018), fit these criteria. In this model, C2C12s express a significant upregulation in cell number (Figure 7A) and cell viability (Figure 7B) alongside reduced myotube area per myonuclei, following Nrg-1 treatment (Figure 7C). Therefore, recapitulating several in vivo phenotypes discussed above. Nrg-1/ErbB2 signalling activates many genes associated with specific cellular processes including proliferation, differentiation, survival, apoptosis, and migration (Geissler et al., 2020). To this extent, it has previously been associated with and not limited to, upregulating proliferation in myoblasts (Ford et al., 2003), fibroblasts (Kirabo et al., 2017), cardiomyocytes (Geissler et al., 2020), Schwann cells (Fallon et al., 2004) and epithelial cells (Liu and Kern, 2002). Additionally, Egr3 expression has been linked C2C12 myoblast proliferation (Kurosaka et al., 2017), therefore it is hard to speculate on the downstream signalling responsible for increased proliferation following Nrg-1 treatment, without further cell type specific investigations.

We must also consider the maturity of the model, during skeletal muscle development and regeneration, MyHCs and MRFs are transiently expressed, and results in vitro will vary depending on the culture environment and time spent differentiating before terminal analysis (Brown et al., 2012; Stuart et al., 2016; Rao et al., 2018). C2C12 myotubes after 8 days differentiation in both control and Nrg-1 treated conditions are still exhibiting an immature phenotype, as determined by high proportions of Myh3 gene expression (Figure 4B) and strong staining for MyHC3 and MyHC8 (Figures 5A,B). In vivo work used to identify intrafusal specific markers are mostly based from adult, or new-born immunohistochemistry samples (Kucera et al., 1992, 1993; Kucera and Walro, 1992, 1995; Tourtellotte and Milbrandt, 1998; Walro and Kucera, 1999; Liu et al., 2002; Zammit et al., 2004; Horst et al., 2006; Kirkpatrick et al., 2008, 2010; Oliveira Fernandes and Tourtellotte, 2015). To overcome this, future model iterations should look to increase the differentiation time span. 2D monolayer cultures will not facilitate this, as spontaneous contractions in C2C12 myotubes detach them from the surface when cultured much past 8 days. Therefore researchers should look to adopt a 3D tissue engineering approach, not only will this facilitate longer culture times (Vandenburgh et al., 1988; Juhas et al., 2014; Rao et al., 2018), but it will better replicate the in vivo environment, allowing cell to ECM and cell to cell interactions in three directions and give the user control over orientation, porosity and stiffness (Torii et al., 2018). In turn, 3D skeletal muscle cultures permit more mature molecular, structural and functional differentiation, better representing native adult muscle (Rao et al., 2018; Capel et al., 2019; Wang et al., 2019; Zhuang et al., 2020) and may facilitate identification of retained immature or specialised intrafusal specific phenotypes using C2C12 cells. Other novel future approaches are discussed in detail in a recent review paper by our group (Barrett et al., 2020).

Without a definitive intrafusal phenotype in vitro, we cannot speculate on the origins of intrafusal fibres. If a single bipotential population of myocytes can develop into both fibre type (i.e., intrafusal and extrafusal), then theoretically we should be able to achieve a pure, intrafusal-only C2C12 cell culture population. To achieve this, a reliable intrafusal linear and bag fibre marker is required. If a distinct lineage of myoblast intrinsically committed to differentiation into intrafusal fibres exists, then C2C12s may be an incompatible source for an intrafusal fibre model. The data attained in this study highlights several hallmarks of intrafusal fibres within Nrg-1 treated C2C12 myotubes. These warrant future endeavours using more sophisticated, biomimetic cell cultures to tissue engineer functional intrafusal muscle for the applications of both basic science and clinical studies, towards improving complete neuromuscular function in disease and injury.

In conclusion, this study presents a novel minimalistic, monocellular C2C12 model for progression towards de novo intrafusal skeletal muscle generation. Recombinant addition of intrafusal muscle developmentally associated protein Nrg-1 was employed to replicate an innervating Ia afferent neuron. The exogenous addition of Nrg-1 elicits elevated endogenous production of Egr3, resulting in intrafusal-like specific differentiation, as determined by the novel morphological characterisation of intrafusal bag myotubes. Concurrently, Nrg-1 increased cell proliferation during the differentiation phase of the protocol, resulting in increased nuclei per FOV and paired with less myotube area per nuclei. Extensive mRNA and Protein analysis of MyHCs is strongly suggestive of a unique phenotype following Nrg-1 addition, however both control and treated myotubes remain in an immature state. All putative intrafusal specific targets apart from developmental proteins Egr3 and Etv4 are not preferentially expressed in Nrg-1 treated myotubes. Egr3 staining, although significantly increased, was not a binary determinant for intrafusal bag myotubes. The suitability for C2C12s to generate intrafusal muscle fibres is still unclear. There is enough promise from the results presented here to encourage future research toward biomimetic tissue engineering approaches, in the hope of producing mature, Nrg-1 treated myotubes, characteristic of native, in vivo intrafusal skeletal muscle. These future models could provide platforms for developmental or disease state studies, pre-clinical screening, or clinical applications, including regeneration or replacement of diseased or dysfunctional tissue (Barrett et al., 2020; Kröger and Watkins, 2021).