The emergence of craniofacial muscles, together with the skull and sensory organs derived from the placodes and neural crest, contributed to the development of the vertebrate head, an evolutionary novelty (Gans and Northcutt, 1983; Glenn Northcutt, 2005). The muscularized pharynx and enhanced sensory abilities facilitated a more active predatory lifestyle during the transition from invertebrates to the early vertebrates (Vyas et al., 2020). In humans, the craniofacial muscles are small-to-medium-sized skeletal muscles, of which there are approximately 70, and they are involved in physiological activities such as facial expressions, food intake, respiration, and speech (Schubert et al., 2019). These muscle functions could be easily impaired by congenital deformities, tumor invasion, and traumatic injury (desJardins-Park et al., 2019). Once affected, craniofacial muscles can only be incompletely repaired with limited treatment modalities, resulting in inadequate muscle restoration and excessive fibrotic tissue production (Schreurs et al., 2020). A lack of knowledge about craniofacial muscle regeneration might contribute to the limited treatment options (Von den Hoff et al., 2019; Schreurs et al., 2020). Because skeletal muscle possesses high regenerative capacity, in that a single transplanted myofiber can generate more than one hundred new myofibers (Collins et al., 2005), therapeutic strategies to boost craniofacial regeneration could be useful.

Unlike the extensively studied limb and trunk muscles, the characteristics of craniofacial muscles have only been investigated in recent years. Researchers have gradually realized that craniofacial muscles have disparate evolutionary origins and undergo distinct developmental trajectories from limb and trunk muscles (Diogo et al., 2015; Schubert et al., 2019; Vyas et al., 2020). Moreover, their specificity in embryonic origin is accompanied by distinct muscle regeneration processes in which the muscle stem cell behavior also differs. Upon injury, the masseter muscle is likely to undergo inadequate muscle regeneration and excessive fibrosis (Ono et al., 2010; Randolph et al., 2015). Nevertheless, the underlying mechanism remains largely unknown.

Skeletal muscle regeneration is orchestrated by the interactions among multiple resident muscle cells (Wosczyna and Rando, 2018). Of these cells, myogenic muscle satellite cells (MuSCs) and mesenchymal fibroadipogenic progenitors (FAPs) have been found to be important. MuSCs play a principal role in the initiation and completion of myogenesis in response to acute or chronic injury (Dumont et al., 2015a; Wosczyna and Rando, 2018). They proliferate and differentiate to produce myogenic progenitor cells, which ultimately fuse with each other to reconstitute myofiber integrity and function (Dumont et al., 2015b). Meanwhile, FAPs have a vital function in extracellular matrix deposition, which is equally indispensable to skeletal muscle regeneration (Bentzinger et al., 2012; Yin et al., 2013; Biferali et al., 2019). Accumulating evidence indicates the distinct cellular behavior of MuSCs and FAPs in the craniofacial region. Understanding the differential regulatory mechanisms involved in muscle regeneration in the craniofacial region and other anatomic regions would support the development of novel muscle repair strategies.

In this review, we aim to provide an up-to-date discussion of the unique properties of the major cell groups in craniofacial muscles, ranging from their embryonic origin to injury response.

Muscle satellite cells are peripherally located myofiber-associated stem cells wedged between the plasma membrane and basal lamina (Dumont et al., 2015a). MuSCs have a large nuclear-to-cytoplasmic ratio, which is a morphological feature typical of stem cells (Dumont et al., 2015a). The MuSC population is characterized by the expression of the transcription factor Pax7 (Dumont et al., 2015b) and can be isolated by flow cytometry via the surface markers Intergrin-α7 and Vcam1 (Dumont et al., 2015b). In healthy, unstressed muscle, MuSCs are mitotically quiescent (Baghdadi et al., 2018). When activated, MuSCs undergo either symmetric or asymmetric division (Forbes and Rosenthal, 2014). Symmetric division enables MuSCs to replenish the stem cell pool, while asymmetric division facilitates myogenic differentiation and MuSC self-renewal (Feige et al., 2018).

Fibro-adipogenic progenitors are muscle stromal cells adjacent to the abluminal side of capillaries between myofibers (Lemos and Duffield, 2018). They express the mesenchymal stem cell markers Sca-1 and Pdgfrα (Wosczyna and Rando, 2018). Acting as structural support and signal guides, FAPs are important participants in the regeneration process. They serve as the progenitors of the muscle connective tissue (Vyas et al., 2020) and are the closest anatomical and functional partners of the myofiber (Sefton and Kardon, 2019). When an injury occurs, FAPs infiltrate the damaged area and synthesize extracellular matrix, which provides a framework for myofiber reconstruction (Biferali et al., 2019). Then FAPs undergo programed cell death to avoid excessive fibrogenesis (Lemos and Duffield, 2018; Wosczyna and Rando, 2018; Biferali et al., 2019). Meanwhile, normal FAPs function is indispensable to MuSCs differentiation because it provides promyogenic factors such as Wisp1 and Il-10 in a paracrine manner (Biferali et al., 2019). The temporal dynamics of FAPs and MuSCs during muscle regeneration overlap substantially, indicating close intercellular communication (Wosczyna and Rando, 2018).

In addition to the above mentioned two types of stem cells, there are other myogenic progenitor cells residing in the interstitial space of muscle. The most studied categories include Pw1+ interstitial cells and Twist2+ cells.

In the attempt to identify specific stem cells responsible for muscle regeneration, Pw1+ cells have garnered a great deal of attention. In fact, the cell stress mediator Pw1 marks two stem cell populations residing in skeletal muscle: Pw1+ /Pax7+ MuSCs and Pw1+ /Pax7- interstitial cells (PICs). PICs are bipotent progenitor cells that give rise to both smooth muscle and skeletal muscle in vitro. Although PICs are not of the Pax7 lineage, they require Pax7 for myogenic specification. In addition, a subgroup of PICs share overlapping surface markers with FAPs (Sca-1 and Pdgfrα) and possess fibrogenic and adipogenic capacity, while the Pdgfrα- PICs are myogenic (Mitchell et al., 2010; Pannerec et al., 2013).

Another type of interstitial stem cell with myogenic potential is the Twist2+ cell. Unlike MuSCs, which contribute to the formation of all types of myofibers, Twist2+ cells are type IIb/x fiber specific myogenic progenitors (Liu et al., 2017). Remarkably, the Twist2+ cells do not express the MuSCs marker Pax7 but can initiate myogenesis both in vitro and in vivo (Liu et al., 2017). Once committed to myogenic lineage, these cells downregulated Twist2 expression and began to express Pax7 (Liu et al., 2017). A notable difference between MuSC and Twsit2+ cells is their opposite roles in muscle regeneration and myofiber size maintenance. MuSCs exert minimal effects on muscle fiber size, as evidenced by the unaffected sarcopenia in Pax7-ablated mice (McCarthy et al., 2011; Fry et al., 2015). In contrast, Twist2+ cells are required for type IIb/x fiber size maintenance, but their ablation does not affect muscle regeneration (Liu et al., 2017). The hypothesis is that Twist2+ cells undergo a pre-myogenic state but are insufficient to guide full regeneration (Liu et al., 2017). It is interesting that Twist2+ cells are specifically excluded from the tongue musculature (Liu et al., 2017). Since type IIb/x fibers are the most abundant in all skeletal muscles and are the most susceptible to aging and disease in mice (Liu et al., 2017), further investigation is needed to characterize the myogenesis mechanism in type IIb/x fibers from different muscle lineages to facilitate the treatment of muscle diseases.

In contrast to trunk/limb muscles, which are ancestral muscles essential to the support and locomotion of the entire body, craniofacial muscles have long been regarded as a variation on the general body muscle scheme. This variation is manifested in many aspects, including developmental origin, myogenic regulatory trajectory, susceptibility to muscular diseases and regenerative capacity. This discrepancy between craniofacial and trunk/limb muscles may to related to their different embryonic origins and to the Hox gene, which conveys positional memory (Leucht et al., 2008). Studies comparing the bone regeneration process of neural crest-derived mandible and mesoderm-derived tibia revealed that heterogeneity in both the embryonic origin and Hox code could account for the ectopic chondrogenesis instead of osteogenesis in tibia-to-mandible periosteum stem cell transplantation assays (Leucht et al., 2008). A similar genetic marker bearing their ancestor’s phenotypic identity, perhaps HoxA and/or HoxB (Rinon et al., 2007; Vieux-Rochas et al., 2013), might also explain the trunk/limb vs. craniofacial muscle differences. Nevertheless, the mechanism underlying the different regenerative response between craniofacial and limb/trunk muscles may vary, because histological differences already exist, in contrast to the histological equivalency with cellular and molecular disparity in the comparison between mandible and tibia (Mitchell et al., 2010). The concept of positional memory should at least be taken into account. Currently, most observations of the features of craniofacial muscles remain at the level of tissue phenotype or cellular behavior. Investigation into the differences in molecular regulatory mechanisms between craniofacial muscles and trunk/limb muscles or among different subgroups of craniofacial muscles, is needed to provide targets for effective muscle regeneration modalities.

XC and JL conceived the idea of writing this manuscript. XC drafted the manuscript. JL drew the figure and critically revised the manuscript. BS critically revised the manuscript. XC and BS provided the funding support. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.