Keeping your strength up: induced pluripotent stem cell-based approaches for the treatment and investigation of skeletal muscle disorders

Abstract

Skeletal muscle comprises the major organ system in our bodies and is the effector tissue for fundamental processes, such as movement, breathing and glucose homeostasis. However, muscle regeneration can be insufficient to repair major injuries or become exhausted by the ongoing fiber damage that occurs in muscular dystrophy. Cell therapy to provide a source of myogenic stem cells that repopulate the lost muscle fibers is an attractive option to treat skeletal muscle injury or dystrophy. Induced pluripotent stem cells (iPSCs) have the potential to be an ideal source of donor muscle cells. This review will discuss major research applications of iPSCs to facilitate skeletal muscle repair, which involve myogenic cell transplantation, in vitro modeling of normal and dystrophic muscle tissue, and drug screening protocols. Recent advances in these fields will be discussed in this review, such as the production of neuromuscular junctions between muscle fibers and motor neurons, along with an overview of the problems that currently prevent iPSCs from being a clinically valid source of transplantable skeletal muscle stem cells. Potential future applications of iPSCs to enhance skeletal muscle regeneration will also be discussed.

KEYWORDS:

• Induced pluripotent stem cells
• iPSCs
• skeletal muscle
• volumetric muscle loss
• muscular dystrophy
• biomaterials

Introduction: skeletal muscle structure, function and regenerative capacity

Skeletal muscle is a highly specialized, post-mitotic tissue and the largest organ system in the body, comprising around 40% of body weight (Kim et al. 2016). The structure of skeletal muscle is shown in Figure 1. The muscle consists of bundles of aligned, multinucleate contractile cells termed myofibers (Mukund and Subramaniam 2020). Each myofiber is composed of thousands of myofibrils, which are the contractile units of skeletal muscle, being made up of thick myosin and thin actin myofilaments. The myofilaments are organized into structural units termed sarcomeres. The sarcolemma (also known as the myolemma) is a specialized cell membrane system that surrounds each myofiber. Muscle contraction is initiated by synaptic input from the motor neuron and muscle fiber at the neuromuscular junction. Contraction takes place via the process of excitation–contraction coupling (Sandow 1952), in which the signal is carried along muscle fibers by a network of transverse tubules (t-tubules). Depolarization of the inner portion of the myofiber activates dihydropyridine receptors in enlarged areas of the sarcoplasmic reticulum surrounding the t-tubules, known as the terminal cisternae. The activated dihydropyridine receptors are in close proximity to ryanodine receptors in the adjacent sarcoplasmic reticulum, and physically interact with the ryanodine receptors to activate them. The ryanodine receptors open, and calcium (Ca²⁺) is released from the sarcoplasmic reticulum into the local junctional space. Ca²⁺ diffuses into the cytoplasm to produce a Ca²⁺ spark (Niggli 1999). The released Ca²⁺ in the cytosol binds troponin C by the actin filaments that results in cross-bridge cycling and the production of force (Kuo and Ehrlich 2015).

Figure 1. Highly organized structure of skeletal muscle. Individual muscles are surrounded by a layer of connective tissue called the perimysium. Each muscle is composed of bundles of contractile cells termed muscle fibers. These fibers contain sarcomeres, which are repeating units of myofibrils that give skeletal muscle its banding pattern. Two contractile proteins, myosin and actin, form the thick and thin filaments of the sarcomere, respectively. The interaction between myosin and actin produces muscle contraction.

Skeletal muscle is also highly vascularized due to the metabolic demands required for contraction (Latroche et al. 2015). Numerous primary arteries run parallel to the myofibers. These branch into arterioles and capillaries that carry blood to the myofibers. Connective tissue plays an important role in maintaining the structure of skeletal muscle (Purslow 2010). Individual myofibers are surrounded by a thin layer of connective tissue termed the endomysium, containing nerves and capillaries. Bundles of myofibers are grouped together into fascicles surrounded by the perimysium connective tissue. Finally, the entire muscle is surrounded by a fibrous connective tissue termed the epimysium.

Myofibers can be subdivided into two major subtypes, termed slow (type 1) or fast (type 2), which is dependent on their functional properties and the expression of ATPase or myosin heavy chain isoforms (Cretoiu et al. 2018). Type 1 myofibers are slow twitch, utilize oxidative metabolism and possess relatively greater endurance than type 2 myofibers, whereas type 2 myofibers are fast twitch and designed for strength. In humans, type 2 fibers can be further subdivided into type 2A (fast oxidative/glycolytic) and type 2X (fast glycolytic). Additional fiber types exist in other mammals, for example type 2b, based on myosin heavy chain expression. It should be noted that myofibers are capable of fiber type switching in response to workload or disease (Pette and Staron 1997; Lee et al. 2018). For example, heart failure induces type 1 myofibers to switch to type 2 (Philippou et al. 2020). Individual muscles vary in their proportion of type 1 and type 2 myofibers that also reflects their function. For example, the soleus muscle, which is involved in standing, contains around 70% type 1 myofibers, whereas the vastus lateralis, which is the most powerful part of the quadriceps femoris, contains around 70% type 2 myofibers (Edgerton et al. 1975). A separate group are the craniofacial muscles, which are evolutionarily, morphologically and molecularly distinct from those of the trunk muscle and account for over 10% of the skeletal muscle system (Sambasivan et al. 2011a; Kim et al. 2020). For example, the extraocular muscles are spared from pathology in Duchenne muscular dystrophy (Khurana et al. 1995).

Skeletal muscle possesses an impressive capacity for regeneration (Carlson and Faulkner 1983). This is due to the presence of a population of quiescent muscle stem cells (termed satellite cells) that reside beneath the basal lamina and above the sarcolemma of myofibers (Dhawan and Rando 2005). Satellite cells (also termed myosatellite cells) are mitotically quiescent in mature muscle, but retain the potential to provide myoblasts for muscle hypertrophy and repair (Scharner and Zammit 2011). In response to muscle injury, satellite cells become activated, re-enter the cell cycle and proliferate to supply a population of myoblasts (the precursors to skeletal muscle cells), which can repair or replace damaged myofibers by differentiation and fusion with existing myofibers or other myoblasts (Kadi et al. 2004; Siegel et al. 2011). Alternatively, a portion of the activated satellite cells may return to a quiescent state (Kuang et al. 2007).

The regenerative capacity of skeletal muscle is not unlimited. Myopathies that produce cycles of myofiber degeneration, such as Duchenne muscular dystrophy, eventually exhaust these regeneration mechanisms and lead to fibrosis and adipose tissue infiltration (Mah 2018). In addition, aging is associated with a decline in skeletal muscle function that becomes noticeable from the fifth decade of life and can progress to severe muscle loss in old age, which is termed sarcopenia (Vicenti et al. 2020). Injuries that produce major skeletal muscle tissue injuries, for example those caused by firearms and automobile collisions, can also overwhelm the regenerative response if the connective tissue scaffold is destroyed.

Satellite cell activity during the process of muscle regeneration is tightly controlled by intrinsic factors within the satellite cells and extrinsic factors produced by the local microenvironment (also known as the muscle stem cell niche) (Yin et al. 2013). For example, although satellite cells are a source muscle stem cells for skeletal muscle regeneration, this process is also regulated by diffusible molecules released from other cell types in the muscle tissue, such as motor neurons (e.g. nerve growth factor (Yin et al. 2013), endothelial cells (e.g. vascular endothelial growth factor (Prior et al. 2003), fibro-adipogenic progenitors (e.g. interleukin-6 (Joe et al. 2010), immune cells (e.g. tumor necrosis factor-α secreted by macrophages (Koike et al. 2022), and connective tissue cells (e.g. angiopoietin 1 secreted by pericytes Kostallari et al. 2015. The extracellular matrix can also be a source of factors that influence the efficiency of regeneration.

It should be noted that laboratory studies have indicated that some non-satellite cell populations in skeletal muscle possess the potential for myogenic differentiation. For example, pericytes and ‘side population’ cells (harvested by cell sorting) can commit to myogenesis after muscle injury or co-culture with myoblasts (Tanaka et al. 2009; Dellavalle et al. 2011; Wang and Rudnicki 2011). However, satellite cell ablation experiments have indicated that these cells do not effectively contribute to skeletal muscle regeneration or replenish the satellite cell population (Lepper et al. 2011; Wang and Rudnicki 2011; Sambasivan et al. 2011b).

Cell therapy is an attractive option for replenishing or complimenting the endogenous satellite cell population when their capacity for regeneration is insufficient to maintain muscle mass and function. Autologous muscle stem cell therapy would be an ideal approach, in which the patient’s satellite cells are obtained from a muscle biopsy, propagated by culture in vitro, and then transplanted back into the muscle tissue to enhance regeneration capacity. Studies in mice have shown that the transplantation of as few as 900 purified satellite cells into the tibialis anterior muscle can improve function in models of Duchenne muscular dystrophy and sarcopenia (Cosgrove et al. 2014; Judson and Rossi 2020). However, this success was not translated to human trials and it became apparent that the need to expand the satellite cell population by culture causes their conversion into a muscle precursor cell termed myoblasts, which have much reduced myogenic potential and survival post-transplantation. Research efforts have adopted a number of approaches to attempt to circumvent this problem, including; 1) small molecule compounds and/or growth factor cocktails that promote myoblast engraftment and differentiation, 2) biophysical factors that model the muscle ECM during satellite cell culture, and 3) inclusion of biomaterials that mimic the structure of muscle to enhance transplantation (reviewed in (Judson and Rossi 2020).

Due to the inability of satellite cells to be sufficiently expanded in vitro, alternative sources of myogenic cells have been considered. One appealing approach is to utilize stem cells that are pluripotent (the ability to differentiate into all of the structures in the adult body) and induce myogenic commitment prior to transplantation. Theoretically, embryonic stem cells (ESCs) derived from the inner cell mass of the blastocyst before implantation into the uterus can provide a reliable source of high quality pluripotent cells. A major stumbling block for this approach is the ethical issues associated with ESC research, such as the reliance on discarded embryos to produce ESC lines (Jung et al. 2014). Embryos can also be created for therapeutic cloning using somatic cell nuclear transfer, although this still involves the destruction of the donor egg cell nucleus (de Wert and Mummery 2003). These issues necessitated the need to develop alternative sources of pluripotent cells for skeletal muscle cell therapy and disease modeling.

iPSCs: history and applicability to skeletal muscle regeneration

A major breakthrough in the generation of patient-derived cells with a similar differentiation potential to ECSs was the development of mouse induced pluripotent stem cells (iPSCs) in 2006. Readily available somatic cells can be reprogramed into iPSCs by the forced expression of transcription factors linked to dedifferentiation or the maintenance of stem cell behavior, such as the classic ‘Yamanaka’ factors: Oct3/4, Sox2, Klf4, c-Myc (Takahashi and Yamanaka 2006). In 2007, the iPSC protocol had been modified to use Nanog as the reporter gene for pluripotency (instead of Fbx15), generating iPSCs that achieved the ‘gold standard’ for pluripotency: production of chimeric mice and contribution to the germ line (Maherali et al. 2007; Okita et al. 2007; Wernig et al. 2007). By late 2007, it was reported that iPSC technology had been successfully applied to human cells, using the same four transcription factors or Oct4, Sox2, Nanog, and Lin28 (Takahashi et al. 2007; Yu et al. 2007). iPSC reprograming methods have been optimized to reduce reliance on potentially oncogenic transcription factors and improve the efficiency of iPSC generation, such as the substitution of one or more transcription factors with small molecules or utilizing microRNAs to inhibit repressors of the reprograming process (Bao et al. 2013; Hou et al. 2013).

The potential for iPSC-based technology to produce large numbers of myogenic cells was of great interest to researchers specializing in skeletal muscle degenerative disorders, because it would circumvent problems associated with utilizing ESCs or non-muscle cells with myogenic potential. These cells could be utilized for cell therapy studies or in vitro modeling of the disease phenotype, for applications such as drug screening or the generation of 3D cultures of human skeletal muscle tissue. In the next section of this review, we will discuss recent progress in two major therapeutic applications of iPSC-derived myogenic cells: Duchenne muscular dystrophy and volumetric muscle loss.

iPSC-based therapeutic approaches for Duchenne muscular dystrophy

Disease course and current treatment options

Duchenne muscular dystrophy (DMD) is a severe form of X-chromosome linked muscular dystrophy with an incidence of approximately 1 in 5,000 male births (Emery 2002). Much more rarely, females with compound heterozygous mutations may also be affected, with an incidence of approximately 1:50,000,000 births (Nozoe et al. 2016). Muscle weakness presents at around 4 years of age with symptoms such as Gowers's sign (difficulty standing up) and increased frequency of falls (Takeshita et al. 2017; Hong et al. 2019). Walking ability usually disappears by 13 years of age and patients become essentially paralyzed from the neck down by age 21 (Rowland 1985). With high quality health care, patients may survive until the fourth or even fifth decade, although the mean life expectancy is estimated to be around 28 years (Landfeldt et al. 2020; Broomfield et al. 2021). The most common cause of death is cardio-respiratory failure (Van Ruiten et al. 2016). DMD is caused by mutations in the dystrophin gene, resulting in much reduced or absent expression of the dystrophin protein. Dystrophin maintains the structural integrity of muscle fibers by connecting actin filaments to the ECM via the dystrophin-dystroglycan protein complex at the fiber sarcolemma (Matsumura et al. 1999). The absence of dystrophin allows unregulated calcium entry through the sarcolemma, producing mitochondrial dysfunction, reactive oxygen species (ROS) production and fiber death in a process that is not yet completely understood (Zablocka et al. 2021).

The standard drug regime for DMD is treatment with the glucocorticoids prednisone, prednisolone, or deflazacort. Even though glucocorticoids are known to cause skeletal muscle atrophy, they are thought to prolong patient survival by dampening the inflammatory response observed in degenerating muscles (Werneck et al. 2019). Recent research progress in the development of exon-skipping oligonucleotides has provided new, U.S. Food and Drug Administration approved therapies for DMD, although they are still in their infancy and their long term effects on DMD progression are not fully known (Dzierlega and Yokota 2020). Examples include eteplirsen (brand name Exondys 51), which targets mutations found in around 14% of DMD patients, and viltolarsen (Viltepso), which targets mutations found in around 8% of patients (Iftikhar et al. 2021). The European Medicines Agency approved the oxadiazole, ataluren (Translarna), which is applicable for the premature stop codon mutation found in around 10% of patients (Hoffman 2020). Current research for new treatments include in vivo gene editing techniques, such as CRISPR/Cas9-mediated genome editing, which have the potential to repair the genetic defect in the dystrophin gene in situ (reviewed in (Lim et al. 2018). However, CRISPR/Cas9 technology is not yet applicable to humans. Consequently, there is an urgent need to develop clinically effective therapies for DMD.

Development of cell therapy approaches for DMD

Cell therapy has been investigated as a treatment for DMD since the demonstration that genetically normal mouse myoblasts transplanted into dystrophin-deficient muscles of the mdx model persisted and gave rise to dystrophin positive muscle fibers (Partridge et al. 1989). As discussed above, a major caveat to this approach is the generation of sufficient numbers of myogenic cells for intramuscular transplantation. Although iPSCs have the potential to form all of the tissues in the body, some tissue types are simpler than others to produce in the laboratory. This applies to muscle, with the differentiation of iPSCs into cardiac cells proving to be relatively straightforward compared to skeletal muscle cells (Pourquie et al. 2018). In the next section, we will describe approaches to enhance myogenesis in iPSCs, firstly focusing on mouse iPSCs and then describing progress using human iPSCs (hiPSCs).

Mouse iPSC-based cell therapy and validation in animal models

Due to the difficulty in producing myogenic cells from iPSC cultures in vitro, an alternative approach is the utilization of iPSCs at an earlier stage of potency, followed by in vivo intramuscular delivery and reliance on the host myogenic environment for inducing differentiation into muscle tissues. This was recently demonstrated by Chan, et al., who generated teratomas (tumors containing immature and/or fully differentiated tissues, including muscle) in mice by intramuscular transplantation of undifferentiated mouse iPSCs (Chan et al. 2018). Muscle-derived teratoma cells were shown to contain myogenic cells, as defined α7-integrin and vascular cell adhesion molecule-1 (VCAM-1) expression. These cells could be purified by FACS and showed relatively high regeneration potential, compared to myogenic cells derived from iPSCs using standard protocols, and regenerated approximately 80% of the tibialis anterior muscle volume (Chan et al. 2018). Remarkably, the authors showed that 1.5 × 10³ teratoma-derived myogenic cells were functionally equivalent to 400 freshly isolated Pax7 expressing satellite cells in their ability to produce muscle fibers with resident satellite cells (Chan et al. 2018).

Another area of DMD research where iPSCs are facilitating progress is the development of in vivo gene editing strategies based on clustered regularly interspaced short palindromic repeats/CRISPR-associated 9 (CRISPR/Cas9) technology. For example, Jin, et al., used CRISPR/Cas9 in vitro to delete exon 23 of the dystrophin gene in mouse iPSCs derived from the mdx model of DMD (Jin et al. 2020). This deletion is sufficient to restore dystrophin expression, but is traditionally very inefficient to achieve. This was overcome using a homology-directed repair donor vector, which carries floxed fluorescent protein and antibiotic selection genes. These mouse iPSCs were differentiated into myogenic cells via upregulated expression of MyoD, and shown to form dystrophin expressing fibers in vivo after transplantation into the tibialis anterior muscle of mdx mice. More recently, Domenig, et al., applied CRISPR/Cas9 editing of the dystrophin gene in vitro to mouse iPSCs derived from mdx mouse fibroblasts, which were converted into myogenic precursors via MyoD overexpression in concert with small molecule treatment (Domenig et al. 2022). In vivo transplantation into the TA muscle restored dystrophin expression and the level of engraftment was not affected by the dystrophin gene editing protocol.

Development of iPSC-based therapies using human cells

Some early efforts to potentiate the myogenic differentiation of hiPSCs focused on viral-mediated delivery of key transcription factors, such (a regulator of muscle precursor cell proliferation that binds target genes as a heterodimer with PAX3) (Darabi et al. 2012), and MYOD1 (one of the first markers of commitment to myogenic differentiation) (Buckingham and Relaix 2007; Young et al. 2016; Judson and Rossi 2020). This increased the numbers of myogenic cells that produce muscle tissue in vitro and in vivo, although doubts remained over the degree to which these cells recapitulate myogenesis during embryonic development (Hicks and Pyle 2015; Judson and Rossi 2020). Nonetheless, it should also be noted that adult myogenesis in satellite cells does not fully recapitulate myogenesis during embryonic development. For example, Lepper et al demonstrated a differential requirement for Pax7 expression during embryonic, juvenile and adult myogenesis (Lepper et al. 2009; Wang and Conboy 2010).

Additionally, the use of viral vectors that produce genomic integrations raises concerns about their use in the clinical setting and suitability for disease modeling (Kaiser et al. 2015; Judson and Rossi 2020). To circumvent these issues, more recent methodologies have relied upon manipulating the iPSC culture environment in vitro with growth factors and bioactive small molecules that induce the myogenic program (Judson and Rossi 2020). A large number of protocols have been developed based on the developmental signals that guide skeletal muscle formation from precursor cells in the somites of the embryo paraxial mesoderm (for a review, see (Chal et al. 2018). Based on these methods, cultures of PAX7-expressing myogenic cells, myosin heavy chain positive myofibers and other cell types can be produced from hiPSCs in vitro within 30–50 days (Chal et al. 2016). Subsequent protocols have aimed to reduce the time required to generate myogenic cells from hiPSCs in vitro and enrich their presence in the cell population, for example by cell sorting using the markers ERBB3⁺/NGFR⁺ or CD10⁺/CD24^- (Shelton et al. 2016; Wu et al. 2018). These optimizations have produced myogenic precursor cells with the ability to generate large numbers of dystrophin expressing myofibers in vivo in the mdx/nsg immunocompromised mouse model of DMD. Further studies have shown that hiPSC-derived myogenic cells can form 3D contractile skeletal muscle in vivo in immunocompromised mice that fuses with adjacent, damaged tissue (Rao et al. 2018). These cells also showed enhanced in vitro differentiation into myotubes, indicating their usefulness for skeletal muscle disease modeling (Wu et al. 2018). Other modifications to induce myogenic cell formation from hiPSCs in vitro include the generation of EZ spheres (cell aggregates to derive pre-rosette multipotent stem cells) by treatment with high concentrations of basic fibroblast growth factor (FGF2) and epidermal growth factor (EGF), or activation of the Wnt signaling pathway (Hosoyama et al. 2014; Sakai-Takemura et al. 2018) (see (Sato 2020) for a review of these methodologies). A summary of the various protocols to enhance myogenesis in iPSC cultures is shown in Table 1.

Table 1. Culture strategies to enhance iPSC differentiation into myogenic precursor cells.

Myogenic progenitor induction	Terminal differentiation	Culture	Myogenic cell selection	Ref.
MYOD1 (tetracycline-inducible system) StemFit AK02N, 0.5% pen/strep (1 d); primate ES Cell media, 0.5% pen/strep (1 d); same medium as before, 1 μg/mL dox (1-2 days); α-MEM, 0.5% pen/strep, 5% KSR, 200 μM 2-ME, 1 μg/mL dox (6–7 days) Sorting markers GE: MYOD1, MYOG, MYH PE: MYH2, MYOG	DMEM, 0.5% pen/strep, 2 mM L-glutamine, 200 μM 2-ME, 5% HS, 10 ng/mL IGF-1, 5 μM SB431542 (2-3 days); DMEM, 0.5% pen/strep, 2 mM L-glutamine, 200 μM 2-ME, 2% HS, 10 ng/mL IGF-1, 5 μM SB431542 (7 days) Sorting markers GE: n/a PE: MYH2	Adherent	n/a	(Uchimura et al. 2017)
PAX7 (dox-inducible lentiviral system) Matrigel® coating, with E6 media, 10 µM CHIR99021 (2 days); E6 media, 1 µg/mL dox (18 days) Sorting markers GE: MYHs PE: MYOD1, PAX3, MYF5, MYOG, sarcomeric α-actinin, PAX7	logDMEM, 10% FBS, Fetuin (500X), hEGF (1000X), DE (1000X), pen (100 unit/mL), strep (50 µg/mL) until 80% confluence (for 2D culture) or 4 days (for 3D culture); logDMEM, N-2 Supplement (100X), pen G (100 unit/mL) (∼ 2 weeks) Sorting markers GE: n/a PE: sarcomeric α-actinin, ACh receptor, MYH, Ca^2+ handling genes	Adherent vs 3D	Myogenic progenitors: FACS for iPAX7+ (GFP+)	(Rao et al. 2018)
Collagen I coating with SKM-01, 5% HS, 3 µM CHIR99021, 2 µM Alk5 Inhibitor, 10 ng/mL EGF, 10 µg/mL insulin, 0.4 µg/mL DE, 200 µM AsAc (10 days); SKM-02, 5% HS, 10 µg/mL insulin, 10 µg/mL EGF, 20 ng/mL HGF, 10 ng/mL PDGF, 20 ng/mL FGF2, 20 µg/mL oncostatin, 10 ng/mL IGF-1, 2 mM SB431542, 200 µM AsAc (8 days) Sorting markers GE: n/a PE: PAX7, DES	SKM-03, 10 µg/mL insulin, 20 µg/mL oncostatin, 50 nM necrosulfonamide, 200 µM AsAc (7 days) Sorting markers GE: n/a PE: MYH	3D	n/a	(Uchimura et al. 2020)
iMatrix coating, DMEM/F12, 1% ITS, 3 μM CHIR99021, 0.5 μM LDN193189 (3 days); same medium as before with 20 ng/mL FGF2 (3 days); DMEM/F12, 10 ng/mL HGF, 2 ng/mL IGF-1, 20 ng/mL FGF2, 0.5 μM LDN-193189 (2 days); DMEM/F12, 15% KOSR, 2 ng/mL IGF-1 (4 days). Induction of EZ sphere culture in Stemline, 100 ng/mL FGF2, 100 ng/mL EGF, 5 µg/mL heparin sodium salt for at different time points (6–10 weeks) (Lim et al. 2018) Sorting markers GE: T, TBX6, PAX3, PAX7 PE: MYOD, PAX7, MYOG	Collagen coating, DMEM, 10% FBS (4 weeks); DMEM, 10% FBS, 10 μM SB431542, and/or 10 μM DAPT Sorting markers GE: MRF4, MYH2, MSTN, MYH8, PAX7 PE: MYH2, MYOG, PAX7	Combination of adherent 2D culture and EZ spheres	Myogenic progenitors: FACS for CD57−/CD108− /CD271+/ERBB3+	(Sakai-Takemura et al. 2018)
DMEM/F12, 1% ITS, 1% pen/strep/L-glutamine, 3.5 µM CHIR99021 (5 days); same medium as before with 20 ng/mL FGF2 (14 days) Sorting markers GE: n/a PE: PAX7	DMEM/F12, 1% ITS, 1% pen/strep/L-glutamine (16 days) Sorting markers GE: RNA seq PE: MYH, TTN, α-ACTN	Adherent	Myogenic progenitors: FACS for c-MET+ /HNK-1-	(van der Wal et al. 2018)
PAX7 (dox-inducible lentiviral system)	KO DMEM, 20% KOSR, 1% NEAA, 1% GlutaMAX^TM, 1% pen/strep, 10 µM SB431542, 10 µM DAPT, 10 µM DE, 10 µM PD0325901, 10 µM Forskolin (5 days)	EBs then adherent	Myogenic progenitors: FACS for iPAX7+ (GFP+)	(Selvaraj et al. 2019)
IMDM basal medium, 15% FBS, 10% HS, 1% pen/strep, 1% GlutaMAX^TM, 1% KOSR, 50 µg/mL AsAc, 4.5 mM MTG (altogether defined ‘myogenic medium’), 10 µM CHIR99021 (2 days); myogenic medium, 200 nM LDN193189, 10 µM SB431542 (1 d); myogenic medium, 1 µg/mL dox (3 days); gelatin coating, myogenic medium, 1 µg/mL dox, 5 ng/mL FGF2 (4 days) Sorting markers GE: MEOX1, TCF15, PAX3, FOXC2 PE: n/a	Sorting markers GE: MYOG, MYOD1, MYH2, MYH3, MYH8, MYH7 PE: MYH8, TTN, DES
Matrigel® coating, DMEM/F12, 20 IU/mL pen, 0.02 mg/mL strep (2%), 1% NEAA, 1% ITS, 3 μM CHIR99021, 0.5 μM LDN193189 (3 days); same medium as before with 20 ng/mL FGF2 (3 days); DMEM/F12, 15% KOSR, 1% NEAA, 2% pen/strep, 0.1 mM 2-ME, 10 ng/mL HGF, 2 ng/mL IGF-1, 20 ng/mL FGF2, 0.5 μM LDN193189 (2 days); DMEM/F12, 15% KOSR, 1% NEAA, 2% pen/strep, 0.1 mM 2-ME, 2 ng/mL IGF-1 (4 days); same medium as before with 10 ng/mL HGF (9 days); DMEM/F12, 1% ITS, 2% pen/strep, 1% N-2 Supplement, 1% L-glutamine Sorting markers GE: n/a PE: TBX6, PAX3, MYOD1, MYOG, PAX7, MYH2, TTN, DMD	Skeletal muscle growth medium (SkGM-2, Lonza), 10 µM ROCK inhibitor (1 d); SkGM-2 (1-2 days); DMEM/F12, 2% KSR, 1 µM Chiron, 0.2% pen/strep, 1× ITS (1-2 weeks) Sorting markers GE: n/a PE: PAX7, MYH2	Adherent	Myogenic progenitors (dissociated after 21 days of differentiation): FACS for PAX7^Venus	(Al Tanoury et al. 2020)
Geltrex matrix, E6 medium, 1% ITS, 10 µM CHIR99021 (2 days); E6 medium, 1% ITS, 5 mM LiCl, 10 ng FGF2, 10 ng IGF-1, 50 µg/mL EVs from MT (4 days); cell splitting; collagen I coating, E6 medium, 1% ITS, 10 ng FGF2, 10 ng IGF-1, 50 µg/mL EVs from MT (10 days) Sorting markers GE: MSGN1, PAX3, PAX7, MYOD1, MYOG PE: n/a	E6 medium, 1% ITS (10 days) Sorting markers GE: MYOG, MYH8, CKM, MYH PE: MYH2, NCAM1, MYOD1, MYOG, MYH2	Adherent	n/a	(Baci et al. 2020)
STEM Diff Apel medium (STEMCELL Technologies), 10 ng/mL FGF2, 0.5 mM BIO, 20 mM forskolin (7 days) Sorting markers GE: n/a PE: PAX7	Matrigel® coating, DMEM, 2% HS (26 days) Sorting markers GE: n/a PE: PAX7, MYF5, MYH4, DES, DMD	EBs and then adherent	n/a	(Uchimura et al. 2020)

Due to a large proportion of DMD patients succumbing to respiratory failure, cell therapy strategies focusing on the diaphragm have the potential to increase average lifespan. To assess the feasibility of this approach, Miura, et al., developed a protocol to study in vivo cell transplantation in the diaphragm of mdx mice, by injection at the edge of the muscle via the abdomen using stereomicroscopy (Miura et al. 2022). Muscle stem cells derived from hiPSCs were shown to survive the injection process and produce dystrophin-expressing myofibers. Moreover, suspending these iPSC-derived cells in a polymer solution enhanced their transplantation into the rapidly contracting diaphragm muscle (Miura et al. 2022).

A novel strategy to enhance myogenic cell differentiation from iPSCs in vitro was recently reported by Mashinchian, et al. (Mashinchian et al. 2022). Utilizing heterotypic three dimensional embryoid cultures, the cell types that were most effective for directing myogenesis in hiPSCs were identified by screening. Embryonic endothelial cells and fibroblasts were found to be highly permissive for myogenesis, and triple in vitro culture with hiPSCs plus Wnt and FGF pathway activation enhanced differentiation into Pax7-positive myogenic cells. In a DMD model, these cells outperformed human myoblasts in terms of re-population of the satellite cell niche in vivo and muscle repair over repeated cycles of degeneration. This study demonstrates the importance of fully characterizing the stem cell niche to realize the full therapeutic potential of iPSCs.

In addition to being a robust source of myogenic cells for in vivo transplantation, iPSCs are being utilized in vitro to improve our understanding of DMD disease progression, with the aim of providing novel therapeutic targets or treatment modalities. For example, in vitro myogenesis in iPSCs derived from normal or skeletal muscle disease conditions can be compared to elucidate differences in the molecular framework regulating muscle development and maintenance (Lim et al. 2021). This approach has identified signaling pathways that are differentially activated during the stages of in vitro myogenesis in hiPSCs derived from normal and DMD patient tissues, such as aberrant activation of the bone morphogenetic protein (BMP) and transforming growth factor (TGF)-β signaling pathways (Lim et al. 2021). Myogenesis was inhibited in the DMD iPSCs in vitro, as shown by reduced myotube formation. As a validation of this analysis, pharmacological inhibition of the SMAD activity, the downstream effector of TGF-β signaling, restored myotube formation in the DMD iPSCs (Lim et al. 2021). Recently, a multi-omics strategy based on published transcriptomic and proteomic datasets was used to investigate early onset abnormalities in the in vitro myogenesis of hiPSCs obtained from DMD patients (Mournetas et al. 2021). It was observed that DMD iPSCs follow the same developmental steps of myogenesis in vitro as normal iPSCs (mesoderm, somite, and skeletal muscle), but abnormalities in the transcriptome occur as early as the somite stage. Approximately 10% of the expressed genes were affected, including mitochondrial genes which showed increased transcriptome abnormalities as myogenesis progressed (Mournetas et al. 2021). Significantly, fibrosis-related pathways were identified as an intrinsic feature of early stage myogenesis in DMD iPSCs in vitro, even though fibrosis is generally thought of as a complication of mature muscle fiber degeneration in DMD patients. These fibrosis pathways were also independent of TGF-β signaling, which is thought to be a major driver of fibrosis in DMD muscle (Ismaeel et al. 2019). This use of cutting-edge technologies has provided new insights into the molecular pathogenesis of DMD, as a disease that also affects the early steps of myogenesis.

The role of TGF-β signaling in DMD progression was also investigated in skeletal muscle cells differentiated in vitro from DMD patient-derived iPSCs. Caputo, et al., utilized a rapid myogenesis protocol for DMD hiPSCs based on the forced expression of MyoD and BAF60C, a protein the functions in chromatin remodeling (Caputo et al. 2020). Myotubes derived by this in vitro method showed high levels of TGF-β- SMAD2/3 signaling, which was further exacerbated by electrical pacing-induced contraction. Using this approach, the authors demonstrated that in vitro-based models of DMD skeletal muscle could recapitulate defects in ECM-derived signals or mechanical cues and may be suitable for the identification of new therapeutic targets (Caputo et al. 2020).

Development of hiPSC-based therapies for cardiomyopathy in DMD

The use of iPSC-derived cardiomyocytes to study the pathogenesis of DMD is also gaining research prominence, because heart failure is becoming an increasingly common cause of mortality (D’Amario et al. 2017). It has previously been shown that heart failure and skeletal muscle wasting are linked (Ding et al. 2018). Yasutake, et al., investigated the role of yes-associated protein (YAP) signaling in hiPSC-derived cardiomyocytes from DMD patients in vitro (Yasutake et al. 2021). YAP activity was compared to hiPSC-derived cardiomyocytes with the DMD mutation corrected using CRISPR/Cas9. It was observed that YAP signaling and proliferation was reduced in cardiomyocytes with the DMD mutation, which was due to disorganization of the actin filament network, suggesting that the activation of YAP signaling could be a target for preventing cardiac disease in DMD. A recent report by Willi, et al., demonstrated in vitro bioenergetic and metabolic impairments in hiPSC-derived cardiomyocytes from DMD patients, such as a 75% decrease in the mitochondrial ATP production rate (Willi et al. 2022).

hiPSCs for the study of drug mechanisms

iPSC-based models of DMD are also providing novel insights into the mechanisms of prescribed therapeutics. Tanoury, et al., generated late stage myogenic cells from hiPSCs with a DMD mutation (Al Tanoury et al. 2021). Compared to normal iPSC derived muscle cells, the DMD mutation produced numerous abnormalities, such as increased fusion and myofiber branching, and defective contraction force. Treatment with the prescribed glucocorticoid drug, prednisolone, reduced these abnormalities in the DMD mutated myofibers. These findings shed new light on the action mechanism of prednisolone, because it was previously thought that this drug functioned by immunosuppressant effects, rather than directly acting on the myofibers (Al Tanoury et al. 2021).

Increasing access to hiPSC lines from DMD patients

New protocols are also being developed to increase the availability of hiPSCs from DMD patients. Recently, Ghori and Wahid demonstrated that hiPSCs could be generated from urine-derived cells harvested from DMD patients using an integration-free episomal vector system (Ghori and Wahid 2021). This can provide a low-cost, non-invasive, painless, and repeatable source of hiPSCs. To reduce the need to obtain hiPSCs from donors with DMD, Valetdinova, et al., have generated three hiPSC lines from a DMD patient using non-integrating episomal plasmid vectors (Valetdinova et al. 2020). The unique stem cell lines identifiers are ICGi002-A, ICGi002-B, and ICGi002-C and have been deposited in a cell line repository. These hiPSC lines were validated by their ESC-like morphology, expression of pluripotency markers, normal karyotype and tri-lineage differentiation potential. Similarly, Guan, et al., have generated an hiPSC line (SDQLCHi007-A) from a patient with deletions in exons 49–50 of the dystrophin gene, via reprograming blood mononuclear cells using non-integrating vectors (Guan et al. 2020). More recently, Ahamd, et al., reported the generation of two hiPSC lines (IGIBi006-A and IGIBi008-A) from the lateral collateral ligament of two patients of Indian origin carrying exonic deletions of the dystrophin gene (Ahmad et al. 2022).

CRISPR/Cas editing in hiPSCs and targeting long noncoding RNAs

In addition to the CRISPR/Cas9 based editing of mouse mdx iPSCs, described above, Morisaka, et al., tested the multi-component Class I Cas 3 CRISPR system as a tool to induce exon skipping of the DMD mutation in patient-derived hiPSCs. This approach provided a unique editing system and alternative to the more widely used single-component Class 2 CRISPR systems (Morisaka et al. 2019).

Long noncoding RNAs (lncRNA) regulate gene expression at multiple levels (epigenetic, transcriptional, and post-transcriptional) (Wei et al. 2018). iPSC-based production of skeletal muscle cells from patients has allowed the investigation of lncRNAs in the pathogenesis of DMD. Zhang, et al., demonstrated that lncRNA H19 interacts with dystrophin and inhibits E3-ligase-dependent degradation using hiPSC-derived muscle cells from patients with Becker muscular dystrophy, which also results from mutations in the dystrophin gene (Zhang et al. 2020). Exon skipping approaches that circumvented a silent mutation that was responsible for preventing the H19-dystrophin interaction were shown to be effective at preventing muscle atrophy in the mdx model of DMD.

hiPSC-based drug screening protocols

An important application of iPSC-based technologies is the development of disease- and patient-specific drug screening protocols. In the case of DMD, Sun, et al. produced myoblasts from DMD patient hiPSCs for a screening platform based on high-content imaging parameters, such as morphological changes during differentiation and myogenic marker protein expression (Sun et al. 2020). This approach led to the identification of two compounds (ginsenoside Rd and fenofibrate), that improved disease phenotype in the mdx mouse model of DMD. Similarly, Xuan, et al., reported a drug discovery advance using hiPSCs (Xuan et al. 2021). In this study, a candidate compound that promoted iPSC differentiation into skeletal muscle progenitor cells was tested. The histone deacetylase inhibitor givinostat was shown to enhance differentiation compared to the standard protocol, and produce progenitor cells that effectively repaired damaged muscle and restored dystrophin expression in the mdx model of DMD.

In summary, iPSC-based methods have significantly contributed to DMD research by enhancing our understanding of the disease process, providing a robust source of myogenic cells for transplantation, and indicating candidate compounds for drug development. Due to constraints related to the pathogenesis of DMD, as a severe myopathy that affects all muscles of the body, cell therapy approaches may be limited by the poor migration of transplanted myogenic cells from the site of delivery, and the lack of widespread engraftment after vascular delivery. In the next section of this review, we describe a cell therapy-based application of iPSC technology that may be suitable for a skeletal muscle disorder: volumetric muscle loss.

iPSC-based therapeutic approaches for volumetric muscle loss (VML)

Socioeconomic impact of VML

As described above, skeletal muscle possesses an impressive capacity for regeneration. However, injuries or insults (such as trauma, surgery or degeneration) may occur that exceed this capacity, resulting in connective tissue infiltration, abnormal morphology, and loss of muscle function (Shayan and Huang 2020). This occurs when the bio-scaffold of the muscle is damaged to such an extent that it cannot support satellite cell proliferation and myogenesis (Wu et al. 2021). Unfortunately, the incidence of VML is not well documented, because it is not tracked by civilian trauma data (Langridge et al. 2021). Some estimation can be gained from records of open bone fractures: approximately 150, 000 occur annually in the United States with the majority having soft tissue loss, and approximately 58% of severe open tibial fractures have significant muscle damage (Court-Brown and McBirnie 1995; Bosse et al. 2002; Langridge et al. 2021). Military data shows that VML accounts for 65% of disability following severe open tibial fractures and produces a lifetime disability cost of around $400,000 per patient, in addition to medical fees (Corona et al. 2015; Langridge et al. 2021). VML lacks a standard treatment protocol to replace lost muscle tissue and restore strength (Shayan and Huang 2020). Surgical approaches aim to restore tissue defects and involve autologous free flap grafting, scar tissue debridement or minced skeletal tissue transfer, although they are limited by factors such as tissue availability, high level of surgical skill, morbidity at the donor tissue site, and a lack of demonstrable clinical efficacy (Lin et al. 2007; Corona et al. 2016; Shayan and Huang 2020).

Experimental therapies for VML

Multiple potential cell types could be utilized to enhance muscle regeneration for VML repair, such as patient-derived muscle stem cells (Shayan and Huang 2020). Patient-derived hiPSCs can be advantageous for this approach, because they are obtained from any tissue source, expanded to provide sufficient donor cells in vitro, and differentiated into the various cell types that comprise skeletal muscle tissue (Shayan and Huang 2020). However, it should be noted that a number of alternative strategies for treating VML are also being investigated in clinical and pre-clinical trials. For example, in patients acellular biological scaffolds composed of ECM proteins known to promote muscle regeneration have undergone clinical investigation and shown promising results (Dziki et al. 2016). Protocols for physical therapy in patients are also undergoing continuous modification in response to new clinical data (Washington and Wolchok 2023). In animal models, the local delivery of regeneration-promoting cytokines, such as interleukin-10, has also been shown to promote muscle growth after VML in rats (Huynh et al. 2023). FDA-approved drug repositioning has also been applied to a mouse model of VML, with the demonstration that the β2 adrenergic receptor agonist, formoterol, is a translational option (Raymond-Pope et al. 2023). These approaches are interesting, depending on their efficacy, because they do not rely on the production of stem cell populations. Due to iPSCs not being currently approved for clinical use, other sources of progenitor cells with the potential for muscle repair are under investigated in animal models, such as mesenchymal stem cells and mesoangioblasts (Shayan and Huang 2020). An interesting alternative to treat VML is skeletal muscle transplantation. As a source of donor tissue (because post-mortem human muscle is not viable), humanized skeletal muscle is being developed in pigs using a combination of gene editing, somatic cell nuclear transfer and blastocyst complementation (Greising et al. 2022). If successful, this strategy could provide large quantities of donor muscle tissue for VML patients.

iPSC-based studies for the treatment of VML

Studies of iPSCs as a source of myogenic cells for VML have utilized numerous approaches, such in vitro myogenesis enhanced using scaffolds constructed from biomaterials and 3D cell culture systems, followed by validation in mouse models of VML. It has been shown that myotubes differentiated from hiPSCs can be matured in 3D cell culture, integrate into sites of muscle damage and induce vascularization (Rao et al. 2018; Langridge et al. 2021). Using a mouse model of VML, which involves tibialis anterior muscle excision to produce 30-40% muscle loss, Wu, et al., recently demonstrated that hiPSCs differentiated into myogenic cells by activation of Wnt signaling, followed by inhibition of BMP and TGF-β signaling, were transplanted into the area of muscle resection within a fibrin hydrogel (Wu et al. 2021). The authors reported abundant formation of donor-derived myofibers and satellite cells, and restoration of the lost bio-scaffold (as shown by laminin staining). Moreover, donor myofibers expressed components of the neuromuscular junction, indicating functional coupling with motor neurons. This was supported by the demonstration of increased in situ contraction of muscles that were transplanted with myogenic cells from hiPSCs. A recent preprint by Jodat, et al., describes the use of 3D bioprinting to create hiPSC-derived skeletal muscle implants for VML (Jodat et al. 2021). Using tailored hydrogels with chemical, structural and mechanical properties based on skeletal muscle ECM, hiPSC-derived myoblasts differentiated into linear, contractile immature myofibers in the millimeter length range. Implantation of these pre-vascularized constructs into a mouse model of VML produced successful graft-host interaction and the de novo production of muscle tissue.

iPSC-based methodologies for the effective formation of neuromuscular junctions

As mentioned above, skeletal muscle is a complex tissue composed of highly structured contractile elements, neuromuscular junctions with motor neurons, and a rich capillary network. Some tissue engineering approaches have specifically focused on effectively re-building these structures after injury or disease to ensure effective functional recovery. The integrity of the regenerated structures has been assessed using in vitro and in vivo approaches. Much research attention has focused on the neuromuscular junction (NMJ), because it is critical for initiating muscle contraction and iPSCs have the potential to differentiate into both muscle fibers and motor neurons.

In vitro studies of NMJ formation from mouse and hiPSCs

NMJ formation using iPSC cultures has been investigated in vitro, which can also increase understanding of the molecular mechanisms underlying this process (see (Luttrell et al. 2021) for a dedicated review). Recently, 3D in vitro culture of mouse skeletal muscle and motor neurons (derived from mouse iPSCs) were constructed using magnetic force-based tissue engineering (Mag-TE) (Yoshioka et al. 2020). This produced functional NMJs with excitation-contraction coupling and potential applicability for ALS drug screening.

Co-cultures of hiPSCs and mouse skeletal muscle primary cells also results in NMJ formation with neurite processes associated with acetylcholine receptor clustering and muscle twitching (Liu et al. 2016; Barbeau et al. 2020). Lin, et al., demonstrated that single hiPSC cultures can be differentiated into motor neurons and myotubes with mature functioning NMJs containing pre- and post- synaptic components, and Schwann cells (Lin et al. 2020). This was achieved by switching between neurogenic and myogenic induction media during a one month differentiation period. The development of robust, reproducible, step-wise protocols for the development of NMJs from hiPSCs in vitro can be useful for characterizing the physiology and pathology of human NMJs (Lin et al. 2019). In addition, such simplified protocols may facilitate drug development for neuromuscular disease, via their adoption by laboratories which specialize in medicinal chemistry, rapid high throughput screening, and/or target druggability.

3D- and co-culture techniques to enhance NMJ formation

The optimization of culture protocols for differentiating hiPSCs into muscle precursor cells (discussed above) has made possible the co-culture of healthy and disease iPSC-derived motor neurons and muscle from the same human donor (Mazaleyrat et al. 2020). NMJ formation showed acceptable functionality, as determined by AChR clustering, muscle contractions and calcium imaging. To identify new drug targets and therapies for patients with nervous system disorders affecting NMJ function, it is important to develop a system that allows objective quantitative methods to assess NMJ formation and degeneration. Recently, Vila, et al., combined the direct differentiation of myotubes and motor neurons from hiPSCs in a 3D model with optogenetics, microfabrication, optoelectronics and video processing to allow the automated stimulation and recording of photosensitive NMJ function (Vila et al. 2021). The adjustable analysis parameters related to light-induced electrical depolarization were baseline time, peak threshold, and minimum minima height and width, with sufficient sensitivity for drug development applications.

To further recreate the skeletal muscle tissue milieu, culture strategies have moved towards three dimensional systems rather than traditional two dimensional environments. In addition, increasing culture time and including trophic factors for different cell fates have been used to improve iPSC differentiation into the various cell types present in skeletal muscle. Complex isogenic multi-lineage 3D cultures have been developed that allow hiPSCs differentiation into skeletal muscle cells, vascular endothelial cells, pericytes, and motor neurons (Maffioletti et al. 2018). 3D culture systems have also been combined with microfluidics, which allows human muscle cells and human motor neurons to be compartmentalised while promoting spatial guidance and integration using bioactive hydrogels (Osaki et al. 2018; Vila et al. 2019). These advances could facilitate the development of mature ‘all human’ NMJs between muscle and motor neurons in vitro, with applications such as drug development for neuromuscular diseases involving NMJ degeneration (e.g. amyotrophic lateral sclerosis (ALS)). A detailed review of research progress for NMJ formation in vitro is provided by Barbeau, et al. (Barbeau et al. 2020).

Alongside the approaches to improve NMJ formation described above, research has also focused on driving myogenesis in iPSCs towards fully differentiated myofibers, rather than immature myotubes. These approaches are discussed in the next section of this review.

Approaches to improve iPSC differentiation into mature skeletal muscle tissue

In vitro myogenesis of primary skeletal muscle cells produces cell fusion into multinucleate syncytia termed myotubes, which can be striated and undergo spontaneous contractions in the culture dish. However, skeletal muscle tissue is composed of myofibers, which are much larger and more highly structured than myotubes. In addition, myofibers are further sub-divided into fast- and slow-twitch fibers, and individual muscles may be predominantly composed of fast, slow or a mixture of fiber types depending on their function in the body, e.g. contraction with high force versus the maintenance of posture (Mukund and Subramaniam 2020). Protocols than can enhance myogenesis in iPSCs to generate more mature myofibers in vitro would facilitate neuromuscular disease research and drug screening.

hiPSC differentiation into craniofacial muscles in vitro

The MyoD-induced microRNA-494-3p has been shown to inhibit the formation of fast oxidative fibers during hiPSC myogenesis (Iwasaki et al. 2015). Iwasaki, et al., demonstrated that this microRNA targets the histone acetyltransferase, E1A-binding protein p300, which down-regulates expression of myosin heavy chain 2 (Iwasaki et al. 2021). Thus, strategies to suppress the activity of microRNA-494-3p during hiPSC myogenesis may promote the development of fast oxidative fibers in the muscle cultures. The craniofacial muscles, which control food consumption, facial expressions, and eye movement, have a distinct embryonal lineage from the limb and trunk muscles (Buckingham et al. 2003). The formation of craniofacial muscles is also regulated by specific combinations of transcription factors and these muscles are preferentially affected by some myopathies, such as myotonic, facioscapulohumeral and oculopharyngeal muscular dystrophy (Kim et al. 2020). However, myogenesis protocols for hiPSCs are generally based on observations of limb and trunk muscle formation. To address this issue, Kim, et al., developed an optimized protocol mimicking the key signaling pathways during craniofacial embryonic myogenesis with pharyngeal mesoderm induction media, which controlled BMP signaling via notch inhibition (Kim et al. 2020). Transcriptomic profiling indicated that the muscle tissue generated by this protocol had a gene expression pattern that was highly similar to craniofacial myoblasts and distinct from primary myoblasts obtained from the limb muscle.

hiPSC differentiation into muscle cells modeling adult myofibers

As mentioned above, it is difficult to fully recapitulate the myogenesis program in vitro to produce mature myofibers. A recent study by Chen, et al., utilized a 3D tissue-engineered model of human skeletal muscle, termed ‘myobundle’ that can be composed of primary myoblasts or iPSC-derived myogenic progenitors (Rao et al. 2018; Chen et al. 2021). The myobundles show enhanced myotube differentiation compared to 2D cultures, electrically or chemically induced twitch and tetanic contractions, positive force-frequency relationship, strong calcium transients, and physiological responses to different drugs. Moreover, this culture system was validated as a model of adult skeletal muscle by the discovery that exercise-mimetic electrical stimulation or Janus kinase pathway inhibition can prevent myobundle atrophy mediated by interferon-γ, which has been linked to aging, infection-related skeletal muscle atrophy (Chen et al. 2021).

Future challenges

This review has described the multiple applications of iPSC technology for facilitating research into skeletal muscle biology and neuromuscular disorders. iPSCs are also being studied as a source of therapeutic cells for treating myopathies or muscle injury. Although impressive progress has been made, there remain a number of technical challenges that should be addressed before the full potential of iPSCs can be realized.

From the standpoint of iPSC-based cell therapy for myopathies, a major problem is the effective delivery of therapeutic cells to the sites of muscle damage throughout the body. This issue also applies to other types of cell therapy, such as myoblast transplantation. One option under investigation is to develop a delivery system that utilizes the vasculature. For example, Darabi, et al., showed that myogenic cells derived from ESCs can be delivered to degenerating skeletal muscle via intra-arterial delivery and contribute to muscle repair (Darabi et al. 2011). The potential of intra-arterial delivery was also demonstrated by Tedesco, et al., using genetically corrected iPSC-derived mesoangioblasts (myogenic progenitor cells derived from pericytes located in skeletal muscle adjacent to blood vessels) in a model of limb-girdle muscular dystrophy 2D (Tedesco et al. 2012). Alternatively, iPSC-derived muscle precursor cells have been used to identify factors that can enhance their migration from the transplantation site. Using a combination of bioinformatics and tissue engineering studies, Choi, et al., recently demonstrated the treatment with Delta-like 4 and platelet-derived growth factor-BB can increase the migration and therapeutic effect of transplanted cells (Choi et al. 2022). However, it appears that the majority of studies concerning iPSC-based cell therapy for myopathies still test the therapeutic potential of transplanted cells using intramuscular delivery to a single muscle site of interest. Due to these constraints on achieving the widespread delivery of myogenic cells to the major muscle groups, a more achievable option for iPSC-based cell therapy is VML that involves one or a few muscles. It may also be noted that, while most studies of iPSC-based approaches for myopathies have focused on generating myogenic cells, it may be preferable to utilize other stem/precursor cell types, such as mesoangioblasts. This is because of their potential to into different mesodermal lineages (including muscle) and their capacity for efficient extravasation across blood vessels into muscle tissue, making them more suitable for systemic delivery (Biressi et al. 2020).

Although this review has discussed research approaches to develop an iPSC-based cell therapy for DMD, a number of major obstacles remain before this treatment can be successfully applied in the clinical setting. One issue is the safety of the transplanted cells, because of the genetic recombination events used to produce iPSCs, along with the overexpression of myogenic factors employed in some protocols (Darabi et al. 2012; Young et al. 2016; Danisovic et al. 2018; Jin et al. 2020). In addition, although these protocols report a high level of terminal differentiation, the possible occurrence of residual undifferentiated cells could potentially lead to teratoma formation (Suman et al. 2019). Thus, there is a need to develop methods that do not rely on stable genetic integration, coupled with methodologies that can remove undifferentiated cells from the myogenic cultures. Due to these limitations, iPSC-based cell therapies are not currently approved for patients (Dahlke et al. 2021; Kim et al. 2022b). An additional obstacle to the development of iPSC-based cell therapies for DMD is the problem of reproducing the later stages of the disease process in animal models. As discussed above, the mdx mouse is the most commonly used model of DMD for cell therapy research. However, this model fails to present the disease phenotype observed in the muscle of patients with advanced DMD (extensive fibrosis, fatty infiltration, severe cardiomyopathy and physical disability) (Zaynitdinova et al. 2021). An exception is the mdx diaphragm, which develops extensive fibrosis but is technically difficult to study (Henry et al. 2017). An alternative model may be the dystrophin-utrophin (which has high homology to dystrophin and is upregulated during fetal development) double knockout (dKO) mouse that presents a more severe disease progression more representative of a late-stage phenotype (Deconinck et al. 1997; Grady et al. 1997). However, the dKO model is not commonly used in cell therapy research compared to the mdx model. Another major stumbling block for iPSC-based cell therapy for DMD is the uniform lack of success of myoblast transplantation clinical trials in humans (Piga et al. 2019). Transplanted myoblasts showed limited engraftment ability and efficacy of muscle regeneration. Differences in disease progression and underlying molecular and cellular mechanisms between DMD patients and the mdx model may have contributed to the failure of these clinical trials, similar to numerous drug clinical trials that showed efficacy in the mdx model but lacked efficacy and produced previously unreported collateral effects in patients (De Luca 2012).

The ability of transplanted hiPSC-derived myogenic cells to contribute to skeletal muscle regeneration and contribute to the satellite cell pool has been repeatedly demonstrated in animal models. However, issues remain concerning how effectively the transplanted cells could migrate into the skeletal muscle niche areas, their capacity for long-term engraftment and transition to dormant/quiescent status, and cellular maturation from progenitor cells into the adult cell phenotype (Lim et al. 2021). Thus, a more rewarding approach for future studies with hiPSCs for muscular dystrophies may be transcriptional profiling, such as the subsets skeletal muscle precursor/stem cells produced during the step-wise myogenesis. This may reveal abnormalities in the signaling or transcriptional networks of patient-derived iPSCs and indicate novel targets for drug development (Lim et al. 2021).

As mentioned above, a major stumbling block in the use of iPSCs for cell therapy is the risk of teratoma formation after transplantation, which currently precludes the approval of iPSC-based therapies by the FDA, and mice derived from iPSCs have an increased rate of cancer-related mortality (Aoi et al. 2008; Jarrige et al. 2021). There is also debate about the potential immunogenicity of iPSCs (Zhao et al. 2011; Araki et al. 2013). Consequently, there are a number of technical, regulatory and ethical issues currently preventing the clinical translation of this technology (Moradi et al. 2019). Nevertheless, a number of clinical trials of iPSC transplantation are ongoing (Kim et al. 2022b). The majority are nontherapeutic and aim to evaluate tolerance to the iPSCs. Therapeutic trials are mainly focused on circulatory and visual disorders. The outcomes of these trials will shed further light on the applicability of iPSCs as a cell therapy for skeletal muscular disorders.

The majority of iPSC-based cell therapy studies for myopathies have focused on DMD due to its prevalence in society. However, these studies can also be applied to less common myopathies, such as nemaline rod myopathy, for which iPSC lines have recently become available (Ma et al. 2020; Houweling et al. 2021; Suleski et al. 2022). As discussed above, iPSC lines from DMD patients are becoming more widely available in cell repositories. A larger amount of lines with a greater range of mutations in the dystrophin gene will be required, because assessing therapeutic approaches such as exon skipping will vary in their methodology depending on the mutation type (Eser and Topaloglu 2022). In addition to genetic muscle disorders, iPSCs can be source of myogenic cells for other prominent muscle disorders. For example, aging-related loss of muscle mass and function (sarcopenia) is a major health problem in developed countries with demographic aging (Ding et al. 2018; Kim et al. 2022a). Although the potential application of iPSCs for treating/preventing sarcopenia has been discussed (Pareja-Galeano et al. 2016a; Pareja-Galeano et al. 2016b), there have been few published studies. A recent study has demonstrated the effectiveness of extracellular vesicles purified from myogenic cells derived from iPSCs for inhibiting muscle aging (Xuan et al. 2022), which may indicated renewed interest in this disease application.

Although myogenic cell therapy is a major application of iPSC technology for muscle disorders, products derived from iPSCs are also being investigated as potential treatment options. These so-called ‘acellular therapies’ are based on the two types of iPSC-derived extracellular vesicles: exosomes and microvesicles (Jarrige et al. 2021). The rationale for this methodology is based on the findings that beneficial effects of stem cell transplantation may arise from paracrine effects from the donor cells, rather than a direct functional contribution to the diseased tissue. Extracellular vesicles can cross biological barriers and can carry therapeutic molecules to target cells or tissues via fusion with the plasma membrane (Jarrige et al. 2021). Recently, cardiomyocytes derived from normal iPSCs were utilized to show that exosomes can produce beneficial effects in DMD cardiomyocytes, whereas exosomes purified from cardiomyocytes derived from DMD iPSCs were detrimental (Gartz et al. 2020). However, from a clinical perspective, there are challenges due to the lack of development of robust procedures to isolate and purify high yields of extracellular vesicles with sufficient purity while maintaining their structure and activity. Currently, there is no consensus on a ‘gold standard’ protocol for the isolation of extracellular vesicles for therapeutic applications or basic research (Jarrige et al. 2021).

Due to the large distances between the cell body of the motor neuron and muscle fiber in humans, it is currently not possible to entirely replicate the NMJ in vitro. Some studies of NMJ formation in vitro have reported the observation of myotube contractions as supporting evidence of the construction of fully functional NMJs. However, it should be noted that spontaneous contractions have also been observed in cultures of hiPSCs undergoing myogenesis in the absence of motor neurons (for example (Fujiwara et al. 2022) or primary myoblast cultures (Vaughan and Lamia 2019). This may be due to factors such as the presence of calcium ions in the culture media. Congruent with this issue, it has been shown that contractions in cultured cardiomyocytes can be prevented by the calcium ion chelator EGTA (Neco et al. 2010; Koh et al. 2017).

Although much research has focused on producing in vitro models of the NMJ from stem cells, the role of the muscle spindle (which detects changes in muscle length) is also important and undervalued in neuromuscular disease (Badiola-Mateos et al. 2022). To address this issue, protocols have been developed that allow the efficient differentiation of proprioceptors, which are mechanosensory neurons contained in muscle spindles and develop synaptic bouton-like structures with co-cultured skeletal muscle cells (Badiola-Mateos et al. 2022).

Concerning the use of biomaterials-based constructs for de novo skeletal muscle formation, there are issues regarding the adequate diffusion of nutrients and oxygen into the constructs (Shayan and Huang 2020). Additionally, poor levels of innervation remains a main problem for achieving full functional recovery after VML and re-establishing the myotendinous junction is still an unmet challenge (Shayan and Huang 2020). Another challenge will be culture media formulation for the multiple cell types derived from iPSCs that are required to re-build contractile muscle, such as endothelial cells, tenocytes, motor neurons, and connective tissue cells (Shayan and Huang 2020; Jarrige et al. 2021). Recent technological advances, such as Slide-seq, which allows mapping of the multiple cell types patterned in the biomaterials scaffold, may facilitate the correct arrangement of these diverse cell types (Rodriques et al. 2019; Shayan and Huang 2020). The use of patient-derived iPSC cells for the clinical application of biomaterials may also be hampered by the highly inefficient process of cellular reprograming (Langridge et al. 2021).

Figure 2. Schematic of iPSC-based cell therapy for DMD patients. iPSCs are generated from a tissue biopsy, and the genetic mutation in dystrophin corrected. After expansion in culture to generate sufficient cell numbers for transplantation, the cells are then induced to enter the myogenic pathway to yield cultures of muscle precursor cells. These cells are then transplanted into the patient’s skeletal muscle and produce dystrophin positive myofibers and satellite cells.

Figure 3. Schematic of iPSC-based cell therapy for VML patients. Similar to DMD patients, a tissue biopsy is used to generate iPSCs, which are then expanded in culture and induced to enter the myogenic pathway to produce muscle precursor cells. Alternatively, culture conditions may be modified to yield cultures containing multiple cells types found in skeletal muscle, such as motor neurons and immature myotubes. Due to the loss of the connective tissue scaffold in VML, a biomaterials-based matrix is transplanted with the iPSC-derived cells to support skeletal muscle regeneration. The transplanted cells utilize the artificial matrix enter myogenesis and effectively replace the loss muscle tissue.

Concluding remarks

Since their discovery in 2006, iPSCs have made a tremendous contribution to the field of regenerative medicine. Skeletal muscle possesses impressive inherent regenerative capacity, but this can be hampered by widespread injury (VML) or become exhausted in muscular dystrophies, such as DMD. iPSCs can be expanded and differentiated in vitro to provide a large pool of myogenic cells for transplantation studies. Dissecting the individual steps and culture conditions required for myogenesis in iPSC cultures has facilitated our understanding of this process during embryogenic development. Mapping the trajectory of myogenesis in patient-derived iPSCs has shed new light on disease mechanisms and possible new drug targets, such as TGF-β signaling in DMD muscle stem cells (Caputo et al. 2020). These cultures protocols have been advanced to the stage where specific subtypes of skeletal muscle cells can be produced, such as the craniofacial muscles. 3D culture techniques, combined with advances in the production of biomaterials scaffolds, allows the patterned culture of the multiple cell types found in skeletal muscle to produce tissues with greater resemblance to mature muscle, such as contractile myobundles or myotubes with functional innervation by motor neurons (Maffioletti et al. 2018; Chen et al. 2021). The advances in gene-editing techniques, such as CRISPR/Cas9, allow the study of gene editing in muscle cells differentiated from patient-derived iPSCs. Moreover, these muscle cultures can be utilized for disease-specific drug screening. From the perspective of cell therapy, iPSC-derived myogenic cells may be most suited for VML, which involves delivery to one/few affected muscles. It should be noted that the majority of research publications have focused on the DMD application, even though VML could be a more achievable aim. The development of acellular therapies relying on extracellular vesicles isolated from iPSC-derived myogenic cells may broaden cell therapy applications to the muscular dystrophies, because they can be applicable for systemic delivery to reach all muscles in the body. Overall, iPSC technology has greatly increased our understanding of skeletal muscle biology in health and disease, and the further advancement of this technology may yield effective treatments for VML, genetic disorders, or defective regeneration.

Author contributions

Kim, H. carried out the literature search, designed the figures, produced the table, and drafted the paper. Jung, D-W. and Williams, D.R. carried out the literature search, drafted the paper, revised it critically for intellectual content, and gave the final approval of the version to be published. All authors agree to be accountable for all aspects of the work.

Acknowledgements

Figures 1–3 were created with BioRender.com.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Additional information

Funding

This work was supported by the National Research Foundation of Korea (NRF) funded by the Korean government (MSIT) [grant no NRF-2022R1A2C1008322] and the Bio & Medical Technology Development Program of the National Research Foundation (NRF) funded by the Korean government (MSIT) [grant no. NRF-2020M3A9G3080282]. This work was partly funded by the Korean government (MSIP) through the Institute for Information and Communications Technology Promotion (IITP) grant (No. 2019-0-00567, Development of Intelligent SW systems for uncovering genetic variation and developing personalized medicine for cancer patients with unknown molecular genetic mechanisms), and a ‘GIST Research Institute (GRI) IIBR’ grant funded by the GIST in 2023.