Pax3 is a key transcription factor that regulates skeletal muscle development and regeneration and plays a critical role in the specification, proliferation, migration and differentiation of muscle progenitor cells during embryogenesis (Buckingham and Relaix 2015). Progenitor cells expressing Pax3 and its paralog Pax7, give rise to adult muscle satellite cells (Muscle Stem Cells, MuSCs). While Pax7 is known to be essential for the specification and maintenance of MuSCs (Seale et al. 2000), the role of Pax3, which is expressed in a subpopulation of muscle stem cells (Buckingham and Relaix 2007), is less understood. This review explores the complex regulation of Pax3 in adult MuSCs, its influence on satellite cell homeostasis, regeneration, and stress response, and how Pax3-driven satellite cell heterogeneity is a key factor shaping muscle stem cell function.

Skeletal muscle satellite cells

Skeletal muscle stem cells (MuSCs), also known as satellite cells, play a crucial role in postnatal muscle growth, homeostasis, and repair. In adult muscles, MuSCs reside in a specialized niche between the basal lamina and the plasma membrane of muscle fibers, where they remain quiescent under normal conditions (Relaix et al. 2021). During postnatal growth, MuSCs proliferate and fuse to sustain muscle expansion until approximately three weeks of age in mice (White et al. 2010). From the fetal stage to postnatal week three, proliferating muscle precursors progressively exit the cell cycle, becoming embedded beneath the basal lamina and adopting the anatomical position that defines them as satellite cells (Lepper and Fan 2010; Relaix and Zammit 2012). After this period, MuSCs remain largely quiescent, undergoing sporadic activation to maintain muscle homeostasis.

Muscle injury and damaged fibers trigger a cascade of signals that activate the pool of quiescent satellite cells. These cues include diffusible factors, mechanical signals, and changes in the local environment (Evano and Tajbakhsh 2018; Relaix et al. 2021). Activated satellite cells initiate an early stress program (Machado et al. 2017, 2021), become metabolically active, re-enter the cell cycle, proliferate, and differentiate into myogenic cells. These cells subsequently fuse either with each other or with existing myofibers to promote muscle repair (Collins et al. 2024).

This process is tightly regulated by members of the Myogenic Regulatory Factor (MRF) family of bHLH transcription factors, including MyoD, Myf5, MRF4, and Myogenin. During embryonic development and postnatal myogenesis, Myf5, MyoD, and MRF4 specify the muscle lineage, while Myogenin is essential for myogenic differentiation (Lima and Relaix 2021). Furthermore, during the late stage of the regeneration process, MuSCs undergo self-renewal to maintain a stem cell reservoir for future regenerative needs (Okafor et al. 2023; Pawlikowski et al. 2019; Zammit et al. 2004). The activity of MuSCs during regeneration is governed by intrinsic signaling pathways and interactions with the regenerative environment, including neighboring cells such as macrophages and fibro-adipogenic progenitors, through direct contact or paracrine signaling. This dynamic interplay ensures that skeletal muscle adapts to tissue damage while preserving its repair capacity throughout life (Relaix et al. 2021).

Muscle satellite cell heterogeneity

Satellite cell heterogeneity has emerged as a crucial aspect of muscle stem cell biology, with implications for muscle regeneration and homeostasis. This heterogeneity is evident in several key features, including label-retaining capacity, Pax7 expression levels, Myf5 lineage tracing and asymmetrical cell division patterns (Chakkalakal et al. 2014; Kuang et al. 2007; Rocheteau et al. 2012). Label-retaining cells (LRCs) represent a subpopulation of satellite cells with distinct properties and functions within the heterogeneous satellite cell pool. Chakkalakal et al. identified LRCs using a TetO-H2B-GFP reporter system, which allows for tracking of cell proliferation history (Chakkalakal et al. 2014). These cells comprise approximately 30% of the adult satellite cell population and demonstrate slower division rates, suggesting a more stem-like state. LRCs exhibit enhanced self-renewal capacity and stem cell-like functions in comparison to non-label-retaining cells (nLRCs). They sustain a less differentiated state during replication and demonstrated greater transplantation potential. Significantly, LRCs have the capacity to give rise to both LRCs and nLRCs, with the former exhibiting self-renewal capability while the latter being constrained to differentiation. This hierarchical organization within the satellite cell population contributes to the balance between the maintenance of the stem cell pool and the generation of committed progenitors (Chakkalakal et al. 2014).

The Myf5-Cre/ROSA26-YFP lineage tracing system has been instrumental in demonstrating satellite cell heterogeneity and asymmetric division. Kuang et al. demonstrated that approximately 10% of satellite cells are upstream stem cells, capable of both self-renewal and generating committed progenitors through apical-basal asymmetric division (Kuang et al. 2007). In these asymmetric divisions, the satellite stem cell (Myf5-negative) remains attached to the basal lamina, while the committed progenitor (Myf5-positive) is localized near the myofiber. The use of this system showed that approximately 30–40% of the initial divisions of YFP-negative cells exhibit a symmetrical configuration, while the residual divisions adopt an asymmetric fate (Kuang et al. 2007).

The equilibrium between symmetric and asymmetric divisions allows satellite cell fate choice required for regeneration, thereby promoting either the expansion of the stem cell population or the generation of myogenic cells. The balance between these two types of division, influenced by a multitude of intrinsic and extrinsic factors, is crucial for ensuring effective muscle regeneration and homeostasis throughout life (Guilhot et al. 2024). Studies, including live imaging (Webster et al. 2016), functional studies of progenitor expansion and renewal (Robinson et al. 2021) and three-dimensional reconstruction of injured muscles (Collins et al. 2024), in combination with EdU experiments (Pawlikowski et al. 2019), have demonstrated that asymmetric divisions and satellite cell renewal are minimal before five days post-injury.

Pax7 expression levels also contribute to satellite cell heterogeneity. Rocheteau et al. demonstrated that satellite cells with high Pax7 expression asymmetrically segregate their DNA during cell division, while Pax7-low cells distribute their DNA randomly. The Pax7-high population exhibits greater stemness and self-renewal potential after serial transplantations, whereas the Pax7-low cells are myogenically committed (Rocheteau et al. 2012). Moreover, Pax7-high subpopulation is enriched in Notch signaling gene expression known to maintain quiescence in satellite cells (Mourikis et al. 2012).

It is important to note that the methods used to study satellite cell heterogeneity, such as label-retaining assays, Myf5-Cre lineage tracing, and Pax7 expression level analysis, constitute indirect approaches that do not rely on direct gene expression profiling. As Tierney & Sacco have noted, these methods may not fully capture the dynamic nature of satellite cell states, or the potential artefacts introduced by the experimental techniques themselves (Tierney and Sacco 2016). Furthermore, the isolation of satellite cells alters their state and gene expression patterns (Machado et al. 2017, 2021). Additionally, it was hypothesized that the heterogeneity observed through these methods might represent different functional states of a single satellite cell population, as opposed to distinct subpopulations (Chakkalakal et al. 2014).

This perspective is supported by single-cell RNA sequencing studies, which reveal a continuum of gene expression states rather than discrete subpopulations (Cho and Doles 2017; Micheli et al. 2020; McKellar et al. 2021; Walter et al. 2024). Moreover, multicolor lineage tracing studies, such as those performed by Tierney et al., demonstrated that satellite cell heterogeneity can be dynamic and influenced by environmental factors (Tierney et al. 2018). This suggests that the observed differences may not be fixed characteristics but rather reflect the cells’ adaptability to different conditions (Tierney et al. 2018). These considerations underscore the need to exercise caution when interpreting results from indirect heterogeneity assays and highlight the importance of identifying genetic markers for muscle satellite cell subpopulations.

Pax3 and Pax7 are evolutionary conserved transcription factors

Pax3 and Pax7 are members of the Pax family of transcription factors, which play essential roles in organogenesis, particularly in myogenic development, neural crest cell formation, and the specification of the dorsal neural tube (Buckingham and Relaix 2015). These proteins exhibit a high degree of structural similarity and overlapping functional roles during embryonic development (Relaix et al. 2004; 2005; 2006). The structural features of Pax3 and Pax7 reflect their shared evolutionary history while underpinning their divergent biological functions. Cyclostomes (e.g., lamprey) possess only a single Pax3/7-like gene (Modrell et al. 2014). The duplication of a common ancestral Pax3/7 gene at the base of the vertebrate lineage led to the emergence of Pax3 and Pax7, enabling functional divergence and specialization (Paixão-Côrtes, Salzano, and Bortolini 2015).

The evolutionary origins of Pax3 and Pax7 can be traced back to a single ancestral Pax3/7 gene present in early chordates, such as amphioxus (Holland et al. 1999). Comparative phylogenetic analyses confirm that this divergence occurred early in the evolution of vertebrates. This enabled functional specialization while preserving core developmental roles shared by both paralogs. Subsequent gene duplication events contributed to Pax3/7 diversification, as evidenced by the analysis of the cephalochordate amphioxus (Barton-Owen, Ferrier, and Somorjai 2018). In teleost fish, whole-genome duplication events led to expansion of the Pax3 and Pax7 gene repertoire, resulting in the presence of multiple paralogs (e.g., SsPax3a, SsPax3b, SsPax7a, SsPax7b)(Barton-Owen, Ferrier, and Somorjai 2018).

Despite their high degree of sequence similarity, Pax3 and Pax7 have evolved distinct, non-redundant roles in vertebrates. Pax3 is essential for the long-range migration of muscle progenitors during limb development (Relaix et al. 2003), whereas Pax7 plays a predominant role in satellite cell maintenance during postnatal growth (Seale et al. 2000).

Gene replacement experiments in mice, involving the replacement of Pax3 with Pax7 (Relaix et al. 2004) and the generation of Pax3/Pax7 double mutant embryos demonstrated that this functional divergence during development primarily arises from their distinct expression profiles as Pax3 and Pax7 can largely compensate for each other’s function when expressed in the appropriate context (Relaix et al. 2004, 2005).

However, despite the high functional conservation observed in gene replacement experiments, the important variability observed in their coding sequences outside the DNA-binding domains suggests that Pax3 and Pax7 may have acquired additional, unique functions that extend beyond their shared core functions. These distinct molecular properties remain to be elucidated and may contribute to their specialized roles in tissue-specific gene regulation, protein interactions, or post-translational modifications, ultimately influencing cell fate decisions and regenerative potential. This specialization has enabled these transcription factors to evolve unique roles while preserving their fundamental contributions to development, growth, and tissue repair.

Structure of Pax3/Pax7 proteins

Both Pax3 and Pax7 possess a modular structure characteristic of Pax family proteins. This structure consists of three conserved domains, including a highly conserved DNA-binding motif formed by two subdomains connected by a linker. The paired-box and homeodomains are responsible for recognizing specific DNA sequences to regulate gene expression (Shaw, Barr, and Üren 2024). While these two domains are highly conserved between Pax3 and Pax7, subtle differences in their sequences may contribute to variations in DNA-binding affinity and target specificity. Notably, both proteins can bind separately to a set of target genes in adult muscles, with Pax7 uniquely promoting proliferation and inhibiting differentiation, partly due to its higher DNA-binding affinity in the homeodomain. This underscores the distinct functional roles of Pax3 and Pax7 in myogenesis (Soleimani et al. 2012).

During development, however, Pax3 and Pax7 bind together, with their two DNA-binding domains working in concert to provide additional specificity to target genes (Buckingham and Relaix 2015). A conserved octapeptide motif is thought to facilitate protein-protein interactions, allowing Pax proteins to recruit co-factors essential for transcriptional regulation, including repression via Groucho factors (Abraham et al. 2015; Cai 2003; Jennings and Ish-Horowicz 2008). The C-terminal region is less conserved between Pax3 and Pax7 and contains sequences that likely contribute to their functional divergence. This region influences protein stability and interaction with specific co-activators (Fig. 1).

Pax3 interacts with a variety of proteins that modulate its transcriptional activity and functional roles in myogenesis, development, and disease (Buckingham and Relaix 2015). Notably, a significant interaction occurs with Pax7, leading to the formation of Pax3/Pax7 heterodimers, which influence DNA binding to specific palindromic sequences, thereby regulating downstream gene expression that is critical for muscle development (Schäfer et al. 1994). Therefore, we hypothesize that the Pax3/Pax7 modular architecture, comprising paired-box domains, homeodomains, octapeptide motifs, and variable C-terminal regions, may enable them to regulate distinct yet overlapping sets of genes in adult muscle stem cells. More recently, Magli et al. identified Pax3 interaction with members of the chromatin looping complex by using mass spectrometry technology (Magli et al. 2019). Notably, this study identified Pax3 interaction with the LIM-domain binding protein 1 (LDB1), showing how the Lbd1 recruitment at a subset of Pax3 sites play an important role in deposition of epigenetic marks, and ultimately, chromatin looping of the enhancers associated with the activation of the myogenic program (Magli et al. 2019).

Pax7 is a critical regulator of adult muscle satellite cells

Pax7 is indispensable for the specification, maintenance, and function of satellite cells. While Pax7-deficient neonates contain a normal population of satellite cells, these are rapidly lost during postnatal growth and adult Pax7 mutant mice lack satellite cells, resulting in severe muscle atrophy and impaired regenerative capacity, underscoring its critical role in satellite cell lineage maintenance (Lepper, Conway, and Fan 2009; Oustanina, Hause, and Braun 2004; Seale et al. 2000; Relaix et al. 2006). Deletion of Pax7 specifically in adult satellite cells, achieved by combining a Pax7-CreERT2 line with a Pax7 flox conditional allele, results in cell-cycle arrest, dysregulation of myogenic regulatory factors, and regeneration failure, which confirms its essential function in myogenesis (Günther et al. 2013).

Pax7 preserves satellite cell stemness by repressing differentiation-promoting genes while maintaining the expression of self-renewal markers (Maltzahn et al. 2013). It transcriptionally regulates genes such as Id3 to prevent premature differentiation of quiescent satellite cells, ensuring proper myogenic progression (Kumar et al. 2009). Additionally, Pax7 activates gene transcription by recruiting the Wdr5/Ash2L/MLL2 histone methyltransferase complex, which catalyzes H3K4 trimethylation at target loci such as Myf5, promoting myogenic progression (McKinnell et al. 2008). This chromatin modification facilitates the transcriptional activation of Pax7 target genes, highlighting its role in regulating satellite cell proliferation and differentiation.

Post-translational modifications, such as methylation and acetylation, fine-tune Pax7 function. Pax7 interacts with the arginine methyltransferase CARM1, which methylates specific arginine in its N-terminus. This methylation enhances Pax7 ability to recruit the MLL1/2 complex to regulatory regions of Myf5, linking epigenetic modifications to asymmetric satellite cell divisions and myogenic commitment (Addicks et al. 2019; Kawabe et al. 2012). The acetyltransferase MYST1 and deacetylase SIRT2 also modulate Pax7 acetylation levels in response to metabolic cues, including the availability of Acetyl-CoA and NAD+. Loss of Pax7 acetylation disrupts the balance between symmetric and asymmetric divisions, leading to an expanded satellite cell pool and reduced differentiation. This shift alters metabolic adaptation and ultimately compromises muscle regeneration after cardiotoxin injury (Sincennes et al. 2021). These findings emphasize Pax7 dual role in transcriptional regulation and epigenetic remodeling, both essential for maintaining satellite cell function and muscle regeneration.

Pax3 is expressed in a subset of skeletal muscle stem cells

Although Pax3-expressing myogenic cell lines lacking Pax7 expression have been identified (Richardson et al. 2022), Pax7 is universally expressed in adult muscle satellite cells. In contrast, Pax3 shows a more complex and heterogeneous expression pattern among satellite cells, varying between different muscles. This difference is due to the number of Pax3-expressing MuSCs which vary significantly among muscle groups, with higher numbers observed in trunk and forelimb muscles compared to hindlimb muscles (Collins et al. 2009; Vartanian et al. 2019; Relaix et al. 2006; Young and Wagers 2010). Pax3 is not expressed in head muscles, which originate from a distinct embryonic lineage and do not express Pax3 during development (Lima and Relaix 2021; Schubert et al. 2019). Pax3 levels are very low in hindlimb muscles, including the plantaris, quadriceps