Myopathies are a group of heterogenous pathologies affecting the normal functionality of SkM. According to NIH (NIH, 2022), myopathies are neuromuscular disorders in which muscle fiber dysfunctions lead to muscle weakness. They are generally divided into inherited or acquired diseases, depending on whether they are caused by mutations in genes coding for muscle or neural tissues, or by inflammation, toxic exposition or traumas (Stenzel and Schoser, 2017). Among the inherited myopathies (Supplementary Table S1), Duchenne muscular dystrophy (DMD) is one of the most known and severe disorders, characterized by progressive muscle wasting leading to motor, respiratory and cardiac dysfunctions. DMD is an X-linked disorder affecting 1:5,000 male births and is caused by mutations in the Dystrophin gene that lead to the absence or non-functional production of the protein, essential for SkM contraction and maintenance (Mercuri et al., 2019; Duan et al., 2021). Other mutations in genes that encode specific muscle proteins are responsible for severe inherited muscular dystrophies. Sarcoglycanopathies are a family of rare autosomal recessive muscular diseases which affect striated musculature. They are part of the limb girdle muscle dystrophy (LGMD) family, as they mainly affect proximal muscles of the scapular and pelvic girdles. The pathogenesis of these disorders is very heterogeneous, as is the age of onset, progression and severity (Sandona and Betto, 2009). In general, it is the result of mutations in the Sarcoglycan coding genes, part of the essential Dystrophin-associated protein complex (DAPC). These mutations lead to the weakening of the SkM, which causes difficulty in ambulation and breathing that may evolve into premature death.

Different types of metabolic disorders can have severe repercussions in SkM function, as in the case of Pompe disease, a rare (about 1:40,000 births), inherited and often fatal disorder. It is characterized by a defect in the metabolism of glycogen which accumulates abnormally in the tissues causing a slow and progressive weakening of the muscles (Marion and Paljevic, 2020). This occurs when there is a deficiency in the lysosomal enzyme alpha-acid glucosidase (GAA), which is responsible for the glycogen degradation. The disease affects mainly abdominals (with particular reference to the diaphragm), paraspinals, flexors, adductors and abductors of the hip muscles.

The diaphragmatic muscle is generally one of the most affected muscles in myopathies, and due to its main function in respiration, its involvement frequently leads to dramatic consequences. Congenital diaphragmatic hernia (CDH) is a malformation in which the diaphragm fails to properly form, resulting in the abdominal viscera invasion into the thoracic cavity during embryonic development (Sefton et al., 2018). Despite the moderate incidence (1:3,000 new-born), the etiology of the disease is not completely understood due to the heterogeneous and multifactorial players (Bogenschutz et al., 2020). Indeed, high proportion of CDH cases may be due to new dominant mutations, but epigenetic events could also be a causal factor (Pober et al., 2005). As other abdominal wall defects, CDH is a muscle developmental disease, but diaphragmatic hernia can also appear after traumas in young and adult subjects. In this context, another myopathy known as volumetric muscle loss (VML), is a remarkable disease due to the elevated incidence and the strong social implications, as recently discussed in Testa et al. (Testa et al., 2021). VML is an acquired myopathy (Supplementary Table S2) that occurs after trauma or surgery and, as with other muscle defects in which a significant portion of the SkM is missing, often requires regenerative medicine solutions. Recently, other classes of acquired myopathies, for which 3D in vitro models can help understanding the development and progression, have been deeply studied and analyzed. Among them, the idiopathic inflammation myopathies, a group of heterogeneous autoimmune disorders, characterized by chronic inflammation of the muscles often accompanied by extra muscular outcomes (Lundberg et al., 2021). Moreover, many drugs and toxic substances such as the use of statins to reduce cholesterol and triglycerides, or the abuse of alcohol, cause myopathy with diverse and sometimes poorly understood pathogenic mechanisms (Pasnoor and Dimachkie, 2019).

In this scenario, considering the actual lack of definitive cures for several types of myopathies, tissue engineering can provide two important solutions: the generation of transplantable muscles, and the production of in vitro models to perform tests of therapeutic compounds and better investigate both tissue regeneration and pathology progression.

Regardless of the genetic or acquired origin, or whether the disease is a malformation rather than a metabolic dysfunction, the components responsible for the myopathy development and progression are the cells that compose the tissue or the environment in which they live, the ECM.

Cell source for tissue engineering approaches has always raised ethical and technical concerns due to the collection of human muscle biopsies, especially from severely ill patients. An ideal cell population for SkM tissue engineering should respect defined parameters such as good proliferation rate, massive supply, and ability to terminally differentiate in vitro. For the abovementioned reasons, many efforts have been made in the last decades for the maintenance and expansion of human SkM progenitor cell populations (Tamaki et al., 2015; Franzin et al., 2016; Spinazzola and Gussoni, 2017) and for the generation of human induced pluripotent stem (hiPS) cells, the two main categories used in SkM tissue engineering approaches. In vitro SC maintenance represents, by logic, one of the biggest pursuits not only because they are the native SkM stem cells, but also because they possess optimal differentiation ability. Specific protocols were set up for the isolation of human SCs. They include mincing steps of the whole muscle, followed by enzymatic digestion and repetitive trituration of the muscle to break both the connective tissue and the myofibers (Romagnoli et al., 2021). Unfortunately, the invasiveness of the collection methods, the variability in the isolation process efficiency (also due to fibroblast contamination), and the reduced myogenic potential after in vitro cell expansion have prompted the use of commercially available or biopsy-derived committed myoblasts, impacting on the final generation of in vitro muscles depleted in the SCs pool.

In the past, several researchers have focused on the use of adult or fetal stem cells collected from other tissues that possess trans-differentiation properties and assume myogenic capabilities. Through in vitro and in vivo experiments in pre-clinical models, some works have shown at least a partial ability of these non-resident stem cells to recover muscle damages (Pesce et al., 2003; Negroni et al., 2009; Kawamichi et al., 2010; Huri et al., 2018), but overall none have demonstrated the same myogenic differentiation efficiency that distinguishes SCs. These considerations paved the way for the large use of hiPS-derived cells, with which it is possible to obtain the different cell types composing the SkM, including SCs (Langridge et al., 2021).

iPS cells represent a milestone of bioscience progress (Takahashi and Yamanaka, 2016). These cells are reprogrammed to stay in undifferentiated state by forcing the activation of four mainly transcription factors (OCT-3/4, SOX2, c-MYC and KLF4). They can be generated through different strategies following the in vivo or in vitro necessary requirements. The basic principle is the administration of the Yamanaka factors, allowing the generation of undifferentiated cells ranging from somatic cells obtained by non-invasive procedures such as urine and blood collection to skin biopsies (Takahashi et al., 2007). The differentiation of iPS cells into myogenic cells follows a multistep process recapitulating the embryonic development. Embryologically, SkM tissue originates from the mesoderm together with cardiac muscle, connective tissue, bones, blood cells and kidneys. Therefore, all the myogenic progenitors in the body originate from the somites of the mesoderm which, upon specific signals, start to express muscle-specific myogenic factors known as myogenic regulatory factors (MYOD, MYF5, MYOGENIN and MYF6). The obtained myoblasts can fuse together and form muscle fibers (Chal and Pourquie, 2017). This cascade process is recreated in vitro to induce SkM cell generation from iPS cells. Usually, this induction is performed by overexpressing the myogenic transcription factors MYOD or PAX7 (Darabi et al., 2012; Goudenege et al., 2012), or alternatively using recombinant proteins or small molecules (Borchin et al., 2013; Chal et al., 2016; Hicks et al., 2018). Importantly, MYOD has been reported as the master regulator of myogenic differentiation since its overexpression is sufficient to force non-myogenic cells into myogenic fate. This notion is further supported by evidence in mouse models in which MYOD depletion impacts on muscle development and regeneration (Yamamoto et al., 2018). To date, the first evidence of how MYOD overexpression leads to SkM cell generation dates back to the 90s (Dekel et al., 1992). The researchers applied this method not only to obtain healthy, but also myopathic iPS cells, such as those derived from DMD and Pompe diseased patients (Abujarour et al., 2014; Shoji et al., 2015; Sato et al., 2017; Yoshida et al., 2017).

The most used technique for MYOD overexpression has been published by Tanaka and co-workers, using a Piggybac transposon system, allowing cDNA to stably integrate and ensuring efficient gene expression (Tanaka et al., 2013). Notably, other groups used the same strategy, generating cell populations able to recreate specific disease characteristics such as the absence of dystrophin, increased creatine kinase release or altered calcium influx for DMD (Albini et al., 2013; Shoji et al., 2015; Monfoulet et al., 2021).

Another example is the mesangioblast-like stem cells obtained from iPS cells that are capable of differentiating to SkM cells upon forced MYOD expression (Maffioletti et al., 2015). Even though MYOD overexpression is the most used strategy to drive SkM differentiation, it forces cell-cycle to arrest, not allowing the generation of SCs, thus precluding an accurate recapitulation of SkM niches in vitro. In this context, inducing the generation of SkM cells by forcing the expression of PAX3 or PAX7 during development prior to MYOD activation permits to get closer to the real in vivo SkM environment with the in vitro generation of both myofibers and stem cells. Moreover, the use of small molecules and other factors might further overcome this issue, paving the way for safer therapeutic applications. A demonstration of this concept was published in 2016 by Chal et al., that set up a protocol for the generation of PAX7 positive cells from hiPS cells (Chal et al., 2016) due to the dual modulation of Wnt and bone morphogenetic protein (BMP) pathways. These two pathways allow the bypassing of genetic manipulation and cell sorting, two processes that generally characterize other protocols. In this regard, Rao et al. have succeeded with the production of 3D bundles via transient expression of PAX7 in cells differentiated from hiPS cells. Over a month of culture, the 3D muscles increased in structural and molecular maturation. In addition, upon implantation, the 3D constructs displayed progressive vascularization (Rao et al., 2018). Increasing the complexity, Maffioletti and co-workers demonstrated, in an elegant model, the possibility to generate artificial SkM from hiPS cells combining, together with muscle cells, other isogenic cell populations, such as endothelial cells, pericytes and neural progenitors, generating a platform that could bring together regenerative medicine and drug development under the same translational technology (Maffioletti et al., 2018).

The innovative and multidisciplinary strategies adopted in the recent years to rebuild in vitro SkM tissue architecture and complexity can be a game changer. Nevertheless, multiple combinations of different scaffolds, techniques and approaches are very broad and still not well established. In this section, we describe the most commonly used scaffolds and the relative fabrication strategies adopted to model in vitro SkM tissue.

The scaffold is defined as the accommodating material that enhance and promote mechanical support to enable cell engraftment, spatially guide tissue organization, and sustain cell proliferation, differentiation, and final maturation. Nowadays, the main scaffold types for SkM tissue engineering are based on synthetic-derived, natural-derived materials, and a variety of different combinations of hydrogels and ECM peptides that aim to mimic the environmental cues of the natural ECM, or to act like instructive materials able to induce specific cell responses (i.e. Liquid Crystalline Networks promoting myotubes fusion and differentiation) (Martella et al., 2021). Synthetic-derived scaffolds have the advantage that their mechanical and structural properties can be customized with extreme precision according to the final application. Specifically, synthetic or chemical scaffolding techniques are mainly composed of poly-L-lactic-acid (PLLA) and polylactic-co-glycolic-acid (PLGA), polyurethane (PU), polyethylene glycol (PEG), and polyvinyl alcohol (PVA) (Boso et al., 2020). These materials can be tuned in a precise and defined manner to induce or promote the formation of organized and mature tissues. Moreover, they can be loaded with specific bioactive molecules and growth factors to induce cellular and tissue stimulation (Maleiner et al., 2018). Although these materials present many advantages in terms of manufacturing, synthetic scaffolds lack the biochemical components that are indispensable for cellular behavior guidance which make the scaffold a peculiar environment for the function that the tissue should perform. From this perspective, natural-derived materials have an intrinsic affinity with the physiological environment. In recent years, they have been widely used as both plate-coating for 2D cultures (Serena et al., 2010) and scaffolds for the in vitro-grown tissues (Urbani et al., 2018). Organ- and tissue-donor decellularization is one of the most used techniques to obtain tissue-specific and well preserved natural-derived scaffolds for the generation of 3D constructs (Mayorca-Guiliani et al., 2019). Indeed, this approach, allows the preservation of a variety of resident ECM components, maintaining at the same time 3D tissue architecture and ultrastructure (Gerli et al., 2018), features that exert great influence on cellular behavior. All these considerations assume a relevant importance when dealing with muscle tissue, where the myofiber bundle organization and the biomechanical characteristics make this organ a hardly replicable structure with artificial systems. Decellularized scaffolds were employed for the in vitro generation of both linear (Urciuolo et al., 2018) and non-linear muscles (Trevisan et al., 2019), demonstrating to sustain and support in vitro and in vivo cell engraftment, proliferation and maturation of engineered constructs (Jank et al., 2015).

Synthetic, natural derived and combined scaffolds are frequently used as hydrogel formulation. Hydrogels are a family of hydrophilic polymers with high water content and, due to their liquid state, are commonly used to entrap the cells together with biomolecules such as growth factors (Langridge et al., 2021). As described before, the main ECM components are collagens and laminin. Overall, the most common hydrogel-based strategies aim at recapitulating one or more aspects of the SkM ECM such as stiffness, patterning, and biochemical compositions. The combination of collagen-derived hydrogel with Matrigel or fibrin has been demonstrated to produce reliable models that closely recapitulate important aspects of the human SkM physiology and metabolism (Khodabukus and Baar, 2016). For instance, Gholobova and colleagues presented the production of human bio-artificial muscles, named BAM, based on fibrin-derived hydrogel casting procedure. The authors showed a tissue engineered model useful for intramuscular injection studies, for drug testing and screening (Gholobova et al., 2018). More recently, Khodabukus et al. described the use of 3D engineered muscle tissues (myobundles) formed by mixing cells, Matrigel and fibrinogen. The casting technique they employed was aimed at forming the supporting hydrogel scaffold to provide an in vitro platform to: establish patient-specific drug-induced myotoxicity test; assess novel injury biomarkers; and guide preclinical and clinical studies (Khodabukus et al., 2020). Furrer et al. showed how Matrigel scaffolding can be combined with the 3D bioprinting technologies in order to reproduce reliable and standardize protocols to obtain human SkM models, recapitulating exercise and pharmacological responses (Alave Reyes-Furrer et al., 2021). In this research field, the new trend is the use of tissue specific ECM-derived hydrogels based on SkM decellularized ECM as reported by several authors (Choi et al., 2016; Lee et al., 2020). Decellularized ECM has emerged as novel natural-derived hydrogel for muscle tissue engineering because, compared with other hydrogels, even after the decellularization procedures, most of the obtained formulations still contain growth factors, cytokines, proteoglycans, and structural adhesive proteins, which represent tissue-specific biochemical cues. Our group recently described the production of a decellularized ECM-derived hydrogel obtained from porcine diaphragms, representing a useful biological tool for SkM tissue engineering applications, as it is capable of accommodating muscle cells and also of modifying its own biomechanical characteristics following concentration increase or crosslinking.

Several approaches have been developed for SkM regeneration according to their application. They can generally be classified in three main categories: in vitro, in vivo and in situ methods. Each approach has advantages and limitations that determine and circumscribe their use. Importantly, in this context, it is not guarantee that in vitro models, characterized by environmental complexity and functional capabilities, can then be easily adapted for in vivo application. The translation of 3D models can be affected by scalabilities, especially by the huge cell amount required to obtain bundled myofibers for in vivo applications (i.e. for human VML treatment). Moreover, the biomaterials used to mimic ECM and build tissue structure should be bio-compatible with the final recipient (Qazi et al., 2015; Smoak and Mikos, 2020). In the following paragraphs, we focus especially on in vitro models, citing only the most prominent examples of in vivo or in situ application.

For exclusive in vitro purposes, in recent years the work belonging to the group of Professor Gilbert has been of great relevance. They reported a system designed to produce 3D SkM microtissues, using immortalized myoblast lines from healthy and DMD donors, able to self-organize following electrical stimulation (Ebrahimi et al., 2021). This work was preceded by a game-changer paper that filled the lack of an easy-to-use platform for the fabrication of micro-SkM models with high reproducible physiological responses and exhibiting macro-tissue characteristics. The 3D models were generated in reusable 96 multi-well plates using primary human myoblasts directly obtained from muscle biopsies (Afshar et al., 2020).

Generally, the strategies to obtain 3D SkM models rely on self-organized 3D constructs or organoid-like cultures (Li et al., 2011; Takahashi et al., 2013; Takahashi and Okano, 2015; Shin et al., 2022), and scaffold-based 3D models (Jalal et al., 2021) even though the majority of protocols require the presence of scaffolds which can be of different composition and formulation. Currently, cells are embedded in Matrigel, collagen or fibrin hydrogels and the 3D constructs take place due to the presence of two attaching points able to provide proper physical cues as the tensile strength (Ostrovidov et al., 2014; Gholobova et al., 2018; Laternser et al., 2018; Maffioletti et al., 2018; Afshar Bakooshli et al., 2019; Capel et al., 2019; Afshar et al., 2020; Fleming et al., 2020; Rajabian et al., 2021). In these cited papers, hydrogels are manually loaded. However, together with the presence of attaching points, other techniques have been used such as scaffold topography, including micropatterned substrates, to better mimic ECM architecture and mechanical properties, and to guide muscle cell differentiation (reviewed by (Boso et al., 2020)).

To overcome the limitations associated with the manual production of the above-mentioned 3D models, 3D bioprinting technology became more relevant due to the feasibility of finely mimic SkM tissue complexity. The layer-by-layer stratification leads to the deposition of cells and ECM components with high precision and accuracy; recreating complex micro-geometries and mechanical cues. The fidelity of the process depends on several parameters such as the biocompatibility, viscosity, and gelation properties of the synthetic and biologic inks which should be modulated to preserve cell viability. Thus, the inks should provide cell homogeneous distribution, avoiding cell clusters and aggregates formation (Ferris et al., 2013). The most used 3D bioprinting approaches are based on extrusion method; this means that sheer forces can act on cells and inks possibly modifying cell viability and ECM component properties (Jessop et al., 2017). In 2018 Kim and colleagues developed a 3D bioprinting strategy for the generation of implantable muscle using human primary muscle progenitor cells. The bioprinted constructs were able to integrate in vivo and form networks with host vascular and neural tissues in a rodent model. Technically, the constructs were composed of three layers: a hydrogel bioink with embedded cells; a sacrificial acellular gelatin bioink; and a thermoplastic polymeric (PLC) pillar supporting structure. For their production, multiple strips of the bioinks were deposited in parallel in the PLC structure to control cell viability, differentiation, orientation, and oxygen and nutrient diffusion. The authors compared the results of bioprinted and non-printed constructs containing the same cell density. Interestingly, the data showed significant differences, over time, in cell viability, differentiation and proper alignment in favor of the bioprinting approach (Kim et al., 2018). Recently, an incredibly innovative research performed by Urciuolo and colleagues demonstrated the possibility of intravital 3D bioprinting of muscle-derived stem cells and 7-Hydroxycoumarin-3-carboxylate (HCC)-gelatin hydrogel under the epimysium of hindlimb muscle in mice that leads de novo formation of myofibers (Urciuolo et al., 2020). Other groups also confirmed the great potential of 3D bioprinting approach: Choi et al. developed a strategy based on the use of a tissue-derived bioink able to maintain a high cell viability, proper oxygen tensions and de novo muscle formation inside the printed structures upon in vivo implantation (Choi et al., 2019).

The use of cell-embedded hydrogels highlights the self-organization properties of the different cells composing SkM tissue for the generation of in vitro differentiated muscles. However, SkM is a highly mechanically active tissue characterized by rigorous physiological stiffnesses, able to vary according to the developmental stage, physiological status, traumas or injury recovery (Gosselin et al., 1998; Wood et al., 2014; Lacraz et al., 2015). Thus, the possibility to modulate these mechanical aspects represents a crucial point. 3D structure biomechanical properties can be modified not only by controlling hydrogel stiffness (i.e. with different kind of crosslinking) but also by substituting the use of hydrogel with decellularized scaffolds which intrinsically recapitulate the anatomy and the biological composition of SkM ECM. The main difference in the use of decellularized ECM instead of hydrogels concerns the recellularization process: cells are no longer embedded in a liquid solution but need to be injected inside the solid scaffold. This can be achieved, for example, through the perfusion method exploiting the native decellularized vascular tree (Jank et al., 2015), or directly inserted in several points through the use of small stings. Our group demonstrated the possibility to obtain mature and metabolically active 3D constructs by injecting human SkM primary cells inside murine decellularized diaphragmatic muscle (Trevisan et al., 2019). These results were recently confirmed and implemented by introducing the use of custom-made bioreactor able to mimic in vitro the physiological muscle movements leading to the generation of more mature and physiologically aligned structures (Maghin et al., 2022).

Besides the choice of the proper cells and materials in the generation of 3D SkM models a key role is reserved to the external stimuli. Given the high reactivity of the cells, by modifying the microenvironment and thus acting on the ECM compartment, it is possible to trigger or increase proliferation, maturation, and alignment of SkM cells. In the recent decades, several papers have been published describing the efforts to build valid supports and devices for SkM stimulation varying from deformable membranes to proper bioreactors. Generally, these papers present custom-built systems affecting the comparability of different studies. Nevertheless, the main principle is to recapitulate in vitro the passive muscle development and growth (ramp stimulation) and the muscle training and exercise (cyclic stimulation). Using different biomaterials with human SkM cells and applying diverse magnitude strains and frequencies, numerous groups demonstrated the positive effects of mechanical stimulation for the generation of reliable 3D in vitro models. This is confirmed by the spectrum of analyzed outcomes as promotion of ECM remodeling (Auluck et al., 2005; Maghin et al., 2022), increased SkM fibers dimension (Powell et al., 2002; Juhas et al., 2016) and alignment (Matsumoto et al., 2007) or enhanced SCs activation (Fleming et al., 2020). Together with mechanical strain, other kinds of exogenous stimuli such as electrical stimulation, can be added to the 3D muscle constructs (Rangarajan et al., 2014; Khodabukus et al., 2019).