Identification of underexplored mesenchymal and vascular-related cell populations in human skeletal muscle.

Skeletal muscle repair and maintenance are directly and indirectly supported by interstitial cell populations such as vascular cells and fibro-adipogenic progenitors (FAPs), a subset of which express Twist2 and possess direct myogenic potential. Furthermore, work in rodents has highlighted the potential of pericytes to act as progenitor cells, giving rise to muscle cells and transdifferentiating into endothelial cells. However, less is understood about these populations in human skeletal muscle. Here, we performed single-cell RNA sequencing (scRNAseq) on ∼2,000 cells isolated from the human semitendinosus muscle of young individuals. This demonstrated the presence of a vascular-related cell type that expressed pericyte and pan-endothelial genes that we localized to large blood vessels within skeletal muscle cross sections and termed endothelial-like pericytes (ELPCs). RNA velocity analysis indicated that ELPCs may represent a "transition state" between endothelial cells and pericytes. Analysis of published scRNAseq data sets revealed evidence for ELPCs in trunk and heart musculature, which showed transcriptional similarity. In addition, we identified a subset of FAPs expressing TWIST2 mRNA and protein. Human TWIST2-expressing cells were anatomically and transcriptionally comparable to mouse Twist2 cells as they were restricted to the myofiber interstitium, expressed fibrogenic genes but lacked satellite cell markers, and colocalized with the FAPs marker PDGFRα in human muscle cross sections. Taken together, these results highlight the complexity of stromal cells residing in human skeletal muscle and support the utility of scRNAseq for discovery and characterization of poorly described cell populations.

Twist2-expressing cells reside in human skeletal muscle and are responsive to aging and resistance exercise training.

Skeletal muscle is maintained and repaired by sub-laminar, Pax7-expressing satellite cells. However, recent mouse investigations have described a second myogenic progenitor population that resides within the myofiber interstitium and expresses the transcription factor Twist2. Twist2-expressing cells exclusively repair and maintain type IIx/b muscle fibers. Currently, it is unknown if Twist2-expressing cells are present in human skeletal muscle and if they function as myogenic progenitors. Here, we perform a combination of single-cell RNA sequencing analysis and immunofluorescence staining to demonstrate the identity and localization of Twist2-expressing cells in human skeletal muscle. Twist2-expressing cells were identified to be anatomically and transcriptionally comparable to fibro-adipogenic progenitors (FAPs) and lack expression of typical satellite cell markers such as Pax7. Comparative analysis revealed that human and mouse Twist2-expressing cells were highly transcriptionally analogous and resided within the same anatomical structures in vivo. Examination of young and aged skeletal muscle biopsy samples revealed that Twist2-positive cells are more prevalent in aged muscle and increase following 12-weeks of resistance exercise training (RET) in humans. However, the quantity of Twist2-positive cells was not correlated with indices of muscle mass or muscle fiber cross-sectional area (CSA) in young or older muscle, and their abundance was surprisingly, negatively correlated with CSA and myonuclear domain size following RET. Taken together, we have identified cells expressing Twist2 in human skeletal muscle which are responsive to aging and exercise. Further examination of their myogenic potential is warranted.

Muscle injury induces a transient senescence-like state that is required for myofiber growth during muscle regeneration.

Cellular senescence is the irreversible arrest of normally dividing cells and is driven by the cell cycle inhibitors Cdkn2a, Cdkn1a, and Trp53. Senescent cells are implicated in chronic diseases and tissue repair through their increased secretion of pro-inflammatory factors known as the senescence-associated secretory phenotype (SASP). Here, we use spatial transcriptomics and single-cell RNA sequencing (scRNAseq) to demonstrate that cells displaying senescent characteristics are "transiently" present within regenerating skeletal muscle and within the muscles of D2-mdx mice, a model of Muscular Dystrophy. Following injury, multiple cell types including macrophages and fibrog-adipogenic progenitors (FAPs) upregulate senescent features such as senescence pathway genes, SASP factors, and senescence-associated beta-gal (SA-ß-gal) activity. Importantly, when these cells were removed with ABT-263, a senolytic compound, satellite cells are reduced, and muscle fibers were impaired in growth and myonuclear accretion. These results highlight that an "acute" senescent phenotype facilitates regeneration similar to skin and neonatal myocardium.

Loss of dystrophin expression in skeletal muscle is associated with senescence of macrophages and endothelial cells.

Cellular senescence is the irreversible arrest of normally dividing cells and is driven by cell cycle inhibitory proteins such as p16, p21, and p53. When cells enter senescence, they secrete a host of proinflammatory factors known as the senescence-associated secretory phenotype, which has deleterious effects on surrounding cells and tissues. Little is known of the role of senescence in Duchenne muscular dystrophy (DMD), the fatal X-linked neuromuscular disorder typified by chronic inflammation, extracellular matrix remodeling, and a progressive loss in muscle mass and function. Here, we demonstrate using C57-mdx (8-wk-old) and D2-mdx (4-wk-old and 8-wk-old) mice, two mouse models of DMD, that cells displaying canonical markers of senescence are found within the skeletal muscle. Eight-week-old D2-mdx mice, which display severe muscle pathology, had greater numbers of senescent cells associated with areas of inflammation, which were mostly Cdkn1a-positive macrophages, whereas in C57-mdx muscle, senescent populations were endothelial cells and macrophages localized to newly regenerated myofibers. Interestingly, this pattern was similar to cardiotoxin (CTX)-injured wild-type (WT) muscle, which experienced a transient senescent response. Dystrophic muscle demonstrated significant upregulations in senescence pathway genes [Cdkn1a (p21), Cdkn2a (p16INK4A), and Trp53 (p53)], which correlated with the quantity of senescence-associated ß-galactosidase (SA-ß-Gal)-positive cells. These results highlight an underexplored role for cellular senescence in murine dystrophic muscle.