Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You and Your Team.

Learn More →

Origins, potency, and heterogeneity of skeletal muscle fibro-adipogenic progenitors—time for new definitions

Origins, potency, and heterogeneity of skeletal muscle fibro-adipogenic progenitors—time for new... Striated muscle is a highly plastic and regenerative organ that regulates body movement, temperature, and metabolism—all the functions needed for an individual’s health and well-being. The muscle connective tissue’s main components are the extracellular matrix and its resident stromal cells, which continuously reshape it in embryonic development, homeostasis, and regeneration. Fibro-adipogenic progenitors are enigmatic and transformative muscle-resident interstitial cells with mesenchymal stem/stromal cell properties. They act as cellular sentinels and physiological hubs for adult muscle homeostasis and regeneration by shaping the microenvironment by secreting a complex cocktail of extracellular matrix components, diffusible cytokines, ligands, and immune- modulatory factors. Fibro-adipogenic progenitors are the lineage precursors of specialized cells, including activated fibroblasts, adipocytes, and osteogenic cells after injury. Here, we discuss current research gaps, potential druggable developments, and outstanding questions about fibro-adipogenic progenitor origins, potency, and heterogeneity. Finally, we took advantage of recent advances in single-cell technologies combined with lineage tracing to unify the diversity of stromal fibro-adipogenic progenitors. Thus, this compelling review provides new cellular and molecular insights in comprehending the origins, definitions, markers, fate, and plasticity of murine and human fibro-adipogenic progenitors in muscle development, homeostasis, regeneration, and repair. Keywords: Mesenchymal stromal/stem cell, Fibro/adipogenic progenitor, Fibroblast, Adipocyte, Regeneration, Single-cell RNAseq Background Mammalian adult skeletal muscle has extraordinary re- In mammals, skeletal muscle represents ~ 30–40% of the generation capabilities upon injury, making the organ a total body mass, regulating body temperature, metabolism, perfect model to study regeneration and repair, and inves- and physical activity. Comprising the musculoskeletal sys- tigate the contribution of adult stem and interstitial cells tem, striated muscles are responsible for voluntary and in settings of acute or chronic injury. The muscle connect- non-voluntary movements. Skeletal muscles are recog- ive tissue (MCT) components are the extracellular matrix nized as highly plastic tissue, illustrated by atrophic or (ECM) and its stromal cells, which actively produce, main- hypertrophic changes when disused or trained. tain, and remodel this dynamic scaffold during develop- ment, homeostasis, and after trauma. Among the several cell types that participate in muscle re- * Correspondence: o.contreras@victorchang.edu.au; mtheret@brc.ubc.ca Developmental and Stem Cell Biology Division, Victor Chang Cardiac generation, tissue-resident mesenchymal progenitors play a Research Institute, Darlinghurst, NSW 2010, Australia crucial role by providing signaling cues that modulate other Biomedical Research Centre, Department of Medical Genetics and School of muscle-resident cells’ function, and actively remodel the Biomedical Engineering, University of British Columbia, Vancouver, BC V6T 1Z3, Canada ECM during this process. Fibro-adipogenic progenitors Full list of author information is available at the end of the article © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Contreras et al. Skeletal Muscle (2021) 11:16 Page 2 of 25 (FAPs) have been identified as platelet-derived growth factor known as muscle fibroblasts [17–19]. However, com- receptor alpha (PDGFRα,alsoknown as PDGFRA)express- pared with the ever-growing knowledge about adult ing cells [1, 2]. Agrowing bodyofevidenceshows that MCT, the composition and the dynamic remodeling of PDGFRα+ FAPs provide regenerative cues to control muscle embryonic MCT are poorly understood. While evidence stem cell (MuSC) expansion, fate, and myogenesis after acute about the ontogeny of interstitial muscle cells exists damage and aging [1–7]. Furthermore, the ablation of stro- [20–22], only a paucity of studies have reported their mal cells by using mice model expressing the diphtheria embryonic determination, and hence, the developmental toxin receptor (DTR) under the control of the fibroblast acti- origin and role of these ECM-embedded cells are not yet vation protein alpha promoter (FAPα-DTR) impairs the fully appreciated and understood. long-term maintenance of hematopoiesis, muscle mass, and Kardon and colleagues published early evidence of the cachexia [8]. To note, FAPα+ cells are found in most tissues function of these cells in the formation of limb muscles such as bone, salivary gland, visceral adipose tissue, skeletal in the 2000s [23]. The authors described that a mesoder- muscle, and pancreas; express CD90, CD140a, and SCA-1; mal population of TCF7L2+ cells (formerly known as T- and so are most likely to be mesenchymal progenitors, hence cell factor 4 or TCF-4, a TCF/LEF transcription factor FAPs in skeletal muscle [8]. These findings have been con- downstream the canonical Wnt/β-catenin signaling) reg- C- firmed by the Rando laboratory using a knock-in PDGFRα ulates the spatiotemporal determination and differenti- reER DTA :Rosa26 mice model [7], and more recently, by ation of myogenic progenitors and, therefore, regulates Tsuchida’s group using a similar cell ablation strategy [9]. In- limb muscle development in chicks [23]. Limb TCF7L2+ deed, genetic ablation of PDGFRα+ lineage cells leads to im- precursors derive from the lateral plate mesoderm in a paired MuSC expansion and leucocyte infiltration, leading to muscle-specific pattern, but are different from myogenic deficient skeletal muscle regeneration after acute chemical precursors since they do not form muscle nor express injury and neuromuscular defects and muscle atrophy [7, 9]. classical myogenic markers (e.g., Pax7)[6, 23, 24]. Thus, In addition, following limb ischemia, proper muscle revascu- myogenic precursors are patterned by extrinsic cues, larization and repair are lost after ablating FAPs [10]. Hence, mostly coming from the MCT, after the cells have mi- PDGFRα+ FAPs are required for successful muscle regener- grated through the limb rather than being embryonically ation, repair, and maintenance during tissue homeostasis and predetermined to form particular muscle anatomical in pathological states. structures [23, 25]. These MCT progenitors also influ- Muscle-resident PDGFRα+ cells readily initiate fi- ence the myofiber type of limb and diaphragm muscles broblastic colonies (also called fibroblast colony- in a paracrine fashion [24]. Interestingly, not all limb formingunits,CFU-F,(Fig. 1a)) and can clonally dif- muscles contain TCF7L2+ cells during mouse embryo ferentiate not only into activated fibroblasts/myofibro- development, which suggest a distinct patterning and blasts and adipocytes but also into chondrogenic and three-dimensional distribution of these cells in different osteogenic lineages depending on the context [1, 2, subtypes, or the existence of MCT progenitors that do 10–15]. The plasticity and clonal expansion of muscle not express this marker [26]. Nevertheless, TCF7L2 la- FAPs are also seen in humans [16]. However, the ef- bels a significant proportion of mammalian stromal non- fects of damage-induced signals and cues on their myogenic precursors at birth and during adulthood [24, plasticity, fate, and functions have only recently begun 27, 28]. Additionally, MuSCs and endothelial cells also to be explored. The development of new in vivo express Tcf7l2 mRNA and protein, albeit at low levels lineage tracing tools used to identify and track cells compared with FAPs [7, 24, 28]. expressing specific markers in various animal and Researchers have argued that vertebrate muscles de- damage models in parallel with the recent emergence rive from several developmental sources, adding com- of single-cell omics have allowed the identification of plexity to our understanding of the different origins of a broad spectrum of specific stromal populations and MCT in muscle development. For comprehensive re- their relative contribution to muscle homeostasis, re- views, see: [20–22]. Like myogenic precursors, MCT generation, and repair. progenitors originate from different and distinct struc- tural origins during embryonic development. In mam- The developmental ontology of muscle-resident mals, these include the somites for axial-trunk muscles mesenchymal progenitors [29], the lateral plate mesoderm for limb muscles [23, From the embryo to the adult: role of MCT mesenchymal 30], the neural crest cells (NCCs) for head and neck progenitors on muscle development muscles [31–34], and the transient developmental Adult MCT is mainly composed of ECM, largely fibrillar structure originating from the somites called pleuroper- collagens type 1 and 3, elastin, fibronectin and proteo- itoneal folds (PPFs) for the diaphragm [35]. Remark- glycans, and the supportive matrix-resident stromal cells, ably, Merrell and colleagues demonstrated that PPF- + + also called mesenchymal progenitors or traditionally resident TCF7L2 /GATA4 CT precursors regulate the Contreras et al. Skeletal Muscle (2021) 11:16 Page 3 of 25 Fig. 1 a Illustration of FAP cellular properties, including the high expression of PDGFRα, quiescency, CFU-F, and mesenchymal/stromal cell multipotency. Skeletal muscle fibro-adipogenic progenitors form clonal CFU-F following in vitro cell culture. b Z-stack confocal images showing the localization of PDGFRα-EGFP cells in tibialis anterior muscle sections of adult PDGFRαH2BEGFP/+ knock-in mice. Pictures show different skeletal muscle anatomical locations of muscle FAPs. Laminin (magenta) and nuclei (Hoechst, blue) were also stained. Scale bars: 50μm development of the diaphragm and participate in the and Tbx5, in determining the formation of muscles and etiology of congenital diaphragm hernias (CDH), a type tendons of the musculoskeletal system [36]. Interest- of fibroproliferative developmental disorder [35]. The ingly, they found that the myoblast-specific loss of Tbx5 authors also demonstrated that Gata4 null mutations does not affect the correct positioning of myogenic pre- in CT progenitors expressing Paired related homeobox cursors. However, genetic deletion of Tbx5 and Tbx4 in 1(Prx1) could cause CDH during diaphragm develop- the mesenchyme (paired related homeobox (Prx) ex- ment. These studies indicate that the aberrant behavior pressing lineage), resulted in the perturbation of MCT of PPFs CT progenitors can cause congenital muscle organization, and therefore, caused mispatterned muscle diseases like CDH [35]. limbs. Although the authors observed no changes in the The studies of Logan’s group have also helped to ad- expression of Tcf7l2 in the absence of Tbx4/5, the lack vance our understanding of the developmental role of of these transcription factors impaired the spatiotempo- MCT precursors in muscle morphogenesis. Initially, ral distribution of TCF7L2+ cells [36]. Remarkably, the through a combination of conditional deletion and ad- Holt-Oram syndrome, known for leading to skeletal ab- vanced imaging techniques, they demonstrated the cru- normalities and congenital heart disease, is caused by cial participation of T-box transcription factors, Tbx4 mutations in the Tbx5 gene [37]. The study of Hasson Contreras et al. Skeletal Muscle (2021) 11:16 Page 4 of 25 et al. reinforces the model in which MCT gives rise to regionalization of muscles and tendons, independently muscle pre-patterned structures to guide myogenic pre- of bone defects [46]. Altogether, these results not only cursors during development and further demonstrated demonstrate a previously unappreciated function of that extrinsic MCT-derived cues are critical for muscle Hox genes for proper patterning and integration of morphogenesis. Without surprise, the Transforming muscles, tendons, and bones but also illustrated that Growth Factor beta (TGF-β) signaling pathway is in- CT spatiotemporal dynamics participate in the integra- volved in this process. Indeed, Kutchuk et al. demon- tion of the musculoskeletal system as a whole. Further strated that embryonic myofibers and C2C12 myoblasts studies should detail how, when and what factor(s) express Lysyl Oxidase (Lox, an enzyme required for modulate the spatiotemporal dynamics and positional cross-linkage formation in elastin and collagen) and that fate of muscle connective tissue cells. −/− its deletion upregulates the TGF-β signaling. Lox mu- tants display MCT disorganization and delayed myogen- Searching for cell-type-specific markers of muscle stromal esis [38]. Thus, this study illustrates the homeostatic fibro-adipogenic progenitors cross-talk between MCT and muscle cells during limb In adult tissue, two studies characterized a population of musculoskeletal system development. interstitial muscle-resident progenitors with spontaneous The above-proposed model was recently corroborated mesenchymal stem/stromal cell (MSC) potential towards in detail by Besse and colleagues [39]. These authors fibrous myofibroblast and fatty differentiation [1, 2]. took full advantage of an array of labeling and imaging- Using fluorescence-activated cell sorting (FACS) of based studies, mouse genetics, and transcriptomic ana- digested mouse skeletal muscle, our laboratory identified lyses to establish how individual muscle bundles are gen- and named these cells as fibro-adipogenic progenitors erated and established, shedding profound lights on the (FAPs) based on their spontaneous differentiation along role of MCT precursors on muscle morphogenesis at these lineages [1]. We characterized these progenitors as unprecedented resolution. They provided a compelling lineage-negative (Lin−, not expressing hematopoietic demonstration that muscle morphogenesis is primarily (CD45), endothelial (CD31, also known as PECAM-1) or orchestrated by CT mesenchymal progenitors via the se- myogenic markers (α7-INTEGRIN) and positive for cretion of matrix-modifying proteoglycans [39]. Thereby, Stem cell antigen-1 (SCA-1) and CD34 cell-surface anti- through the expression and the secretion of a myriad of gen expression. Interestingly, while quiescent MuSCs, chemoattractants, ECM components, and growth fac- endothelial cells, and a subset of hematopoietic cells ex- tors, these stromal cells promote a variety of responses press CD34, its genetic deletion impairs MuSC but not in myogenic precursors, including repulsion, attraction, FAP proliferation [47]. We also demonstrated that most migration, and patterning [20]. Hence, the MCT creates Lin-/α7 INTEGRIN-/SCA-1+ cells express high levels of a developmental pre-pattern that orientates and controls the receptor tyrosine kinase PDGFRα [1]. Similarly, the positioning of myogenic precursors that differentiate Uezumi et al. characterized the same population using a into myofibers forming muscle bundles and, conse- different gating strategy. They used CD45, CD31, and quently, will serve to define the size and shape of mus- Sm/C2.6 (MuSC marker) as a negative selection and cles, the orientation of its myofibers, and points of origin CD140a (PDGFRα) as positive. They showed that Lin- and insertion on bones [23, 25, 36, 39, 40]. PDGFRα+ cells express a low level of PDGFRβ and can The notion that MCT cells participate in muscle differentiate in adipocytes, myofibroblasts, and chondro- morphogenesis leads to wonder what determines the cytes in vitro [2]. They also observed that muscle PDGF spatiotemporal dynamics and positional information of Rα+ cells were perivascular but did not co-localize with MCT precursors. Hox genesare aset of genescoding NG2, suggesting that PDGFRα+ cells are not pericytes for transcription factors that specify segment identity [2]. and provide positional information during animal de- PDGFRα+ cells reside in the muscle interstitium and velopment [41]. Among them, the caudal Hox11 genes are more abundant in the epimysium and perimysium participate in determining the proximal-distal axis of than in the endomysium. Although most muscle- the musculoskeletal system of limbs [42–45]. Hoxa11 is resident PDGFRα+ progenitors are in close association broadly expressed through the distal primordium of with blood vessels [1, 2, 48], they are distinct from peri- limb buds at E10.5, but later on, at E14.5, it is exclu- cytes. Indeed, pericytes are embedded within the endo- sively expressed in the CT of tendons, perichondrium, thelium basement membrane, but PDGFRα+ cells reside and TCF7L2+ cells, but not in endothelial cells, chon- outside of vessels. The localization of FAPs is evident drocytes, osteocytes, nor myogenic precursors [46]. around large blood vessels, in which they adopt an ad- Genetic deletion of Hoxa11/Hoxd11 paralogs, which ventitial position. With rare exceptions in organs other have a prominent role in patterning bones during de- than muscle, PDGFRα cells do not express defining peri- velopment, leads to severe defects in the pattern and cyte markers like Cspg4 (NG2), Rgs5, Pdgfrβ,or Mcam Contreras et al. Skeletal Muscle (2021) 11:16 Page 5 of 25 (CD146) [2, 48, 49]. Notably, while FAPs were initially Recently, Gli1 (also known as glioma-associated oncogene described in murine muscles, growing evidence indi- 1) expression has been shown to label a subpopulation of cates that human FAPs have a similar phenotype and muscle FAPs with higher clonogenicity and reduced adipo- functions to mouse FAPs [16, 50–54]. In summary, FAPs genic differentiation than Gli negative FAPs [65]. Perivascu- (historically called fibroblasts) and the ECM they actively lar cells expressing the zinc finger protein Gli1 undergo secrete and modify are both significant constituents of the proliferative expansion and generate myofibroblasts after interstitium and perivascular CT. kidney, lung, liver, and heart injury [68]and heartinjury Distinct subpopulations of CT progenitors exist and [6], suggesting that Gli1+ cells are likely a FAP subpopula- express an array of proteins and transcription factors, al- tion as recently shown in skeletal muscles [69]. beit at variable levels. In the mouse embryo, CT progeni- In humans, cell-surface markers like PDGFRα, CD201, tor markers include PDGFRα, TCF7L2, TBX3/4/5, CD166, CD105, CD90, CD73, and CD15 identify skeletal HOX11, and the Odd-skipped transcription factors muscle FAPs (Table 2)[16, 51, 53, 54, 64, 71, 72]. Remark- OSR1 and OSR2 [21, 23, 26, 40, 55, 56] (Table 1). In ably, the expression of SCA-1 defines a particular cluster of murine adult muscles, the large majority of CT fibro- stromal cells within the murine FAP population with differ- adipogenic progenitors express PDGFRα, SCA-1 (also ent potency and properties in vivo and in vitro, both in the known as Ly6A/E), CD90 (THY1), CD34, TCF7L2, skeletal muscle and heart [66, 73]. However, as SCA-1 does HIC1, VIMENTIN, DECORIN, and ADAM12 but few of not have a human homolog, its use to identify FAPs is lim- these markers are specific and unique for this heteroge- ited by the absence of this antigen in humans. neous population of cells (discussed below) [1, 11, 12, Recently, we showed that the majority of cells express- 28, 57, 59, 61, 63, 66] (Table 1). Of note, murine adult ing the protein-coding gene Hypermethylated in Cancer muscle PDGFRα+ FAPs express low levels of Osr1, 1(Hic1) correspond to quiescent muscle-resident FAPs which increases upon acute injury in a small subset of in mice [48]. In adult muscles, HIC1+ progenitors reside FAPs, suggesting the participation of regulatory mecha- in the interstitial space and the myotendinous junctions. nisms that tightly turn on the expression of Osr1 resem- In addition to FAPs, small subsets of pericytes (SCA-1−, bling developmental-like programs [67]. Remarkably, RSG5+ cells) and tenogenic cells (SCA-1−, SCX+ and damage-activated OSR1+ FAPs proliferate faster com- FMOD+ cells) express HIC1. Therefore, the expression pared with OSR1- FAPs [67], suggesting that either of Hic1 comprises a larger proportion of mesenchymal OSR1 modulates the expansion and functions of FAPs, stromal progenitors compared with the expression of PDGF or it represents an activation marker whose expression Rα, which is limited to FAPs [48]. Along with others, we increases in proliferating cells. have confirmed that cardiac PDGFRα+ cells also exhibit Table 1 Summary of endogenous murine skeletal muscle fibro-adipogenic progenitors Murine Canonical Alternative Negative Localization Differentiation potential Additional comments References cell Markers markers markers Embryonic- PDGFRα Osr2 CD45 Muscle- Robust in vitro adipogenic Little is known about their [24]; [26]; [57]; fetal FAPs TCF7L2/ Hox11 CD31 associated and fibrogenic differentiation origin, fate, gene TCF4 Tbx3 Ter119 connective but low chondrogenic and regulation, function, Osr1 Tbx4 α7- tissue and no detectable osteogenic or stemness, and self- Tbx5 Integrin muscle myogenic potential. Osr1+ renewal Sca-1 interstitium progenitors also give rise to CD34 embryonic fibroblast-like cells Adam-12 in the dermis and FABP4+ Tie-2 adipocytes in white fat pads Adult FAPs PDGFRα Hic1 CD45 Fascia, Adipocytes, myofibroblasts, Required for adult skeletal [1]; [11, 12, 27, 28]; SCA-1 CD90 CD31 epimysium, osteocytes, and muscle regeneration and [13]; [14]; [58]; [58]; Decorin Ter119 perimysium, and chondrocytes after muscle homeostasis; cellular and [59]; [60]; [61]; [62], (Dcn) α7- endomysium; injury and in vitro, with no molecular dysfunction in [2, 15, 53, 63, 64]; PDGFRβ Integrin abundant as myogenic potential pathology and disease Col1a1 NG2/ perivascular cells TCF7L2/ Cspg4 TCF4 Rsg5 CD34 Adam-12 Tie-2 Gli1 These markers have not been studied in the embryo with detail These markers are also expressed by different cell types, including satellite cells, pericytes, and endothelial cells Adam-12 and Tie-2 expression appears to be restricted for a subpopulation of FAPs Gli1 defines a subpopulation of murine muscle FAPs with pro-myogenic and anti-adipogenic functions [65] Contreras et al. Skeletal Muscle (2021) 11:16 Page 6 of 25 Table 2 Summary of endogenous human skeletal muscle fibro-adipogenic progenitors Human Canonical Alternative Negative Localization Differentiation potential Additional References cell Markers markers markers comments Embryonic- PDGFRA DCN PAX3 Similar to what is Not evaluated but probably No information [70] fetal FAPs FN1 PAX7 found in mouse similar to what is found in about their origin, LUM development, mouse development gene regulation, OSR1 although not function, and POSTN evaluated in detail potency FAP THY1/CD90 VIM NT5E/CD73 COL1A1 COL1A2 COL3A1 PTN OGN FBLN5 Adult FAPs PDGFRA, CD34 CD201 CD31 Fascia, epimysium, Adipocytes, myofibroblasts, Increased numbers [16]; [51]; (when negative CD166 CD45 perimysium, and osteocytes, and chondrocytes in diverse [71]; [72]; for CD56, CD31 CD105 CD56 endomysium; in diseased states and in vitro. pathologies [52]; [53, and CD45) CD90 α7- abundant as Lack of myogenic potential 54, 64] CD73 Integrin perivascular cells CD34 NG2/ CD15 CSPG4 COL1A1 RSG5 TCF7L2/ TCF4 These other alternative markers suggested by Pyle and colleagues are based on scRNAseq data ([70]) multilineage properties in vivo and in vitro [11, 12, 73–75]. highlighting the contribution of endogenous PDGFRα+ In themurineheart, HIC1+progenitors represent a signifi- cells to mammalian skeletal muscle homeostasis, regen- cant proportion of cardiac FAPs [73]. HIC1 (also known as eration, and repair, it is worth revisiting the terminology ZBTB29) is a transcription factor involved in quiescence and in this area. Muscle regeneration is defined as the spe- cell cycle control [76, 77]. Consistent with these roles, the cific substitution or replacement of lost tissue, eventually conditional deletion of Hic1 induces aberrant cell activation leading to full restoration of muscle strength. This re- and proliferation of FAPs, impairing muscle regeneration fol- generative capacity relies on resident adult unipotent lowing acute damage and leading to spontaneous develop- stem cells (also known as satellite cells), which are quies- ment of arrhythmogenic cardiomyopathy-like pathology and cent but activate to rebuild this tissue upon injury [6, signs in the mouse heart [48, 73]. Thus, the unrestrained ac- 78–80]. tivation of these progenitor cells and the consequent gener- In comparison, skeletal muscle repair aims to safe- ation of differentiated progeny are potential pathological guard the remaining function of muscle following solu- drivers of disease. tions of continuity after partial loss of tissue due to We claim that the heterogeneity of FAP markers massive traumas or chronic insults such as repetitive in- makes sense in a context where the upregulation and juries, disease, and aging. Thus, muscle repair often en- downregulation of cell-specific makers participate in tails replacing lost myofibers with scar tissue, which acts modulating the commitment of FAPs into a transitional as a bridge between areas still capable of contraction (for cell state or differentiation process during lineage pro- review of tissue regeneration and repair see [81–83]. gression in response to injury or in disease states. Hence, Therefore, while repair restores muscle integrity, regen- FAP heterogeneity might contribute to restricting or eration accounts for restoring tissue function. As ob- priming the multipotency of PDGFRα+ FAPs. These are served in other mammalian regenerative tissues such as important issues to explore in future research. the liver [81, 84], the form and periodicity of damage can impair the ability of the skeletal muscle to return to Adult muscle connective tissue and PDGFRα+ homeostasis [59, 85, 86]. Therefore, the current estab- progenitors lished dual role that PDGFRα+ cells play in acute (re- Adult skeletal muscle contains several cell types that generative) and chronic damage (reparative/ work in unison under tightly regulated conditions to degenerative) suggests that the organization of their maintain homeostasis. Adult mammalian muscle is a re- intracellular signaling network may integrate opposite markable exception to the low regenerative potential of complementary signals whose relative strength mainly several organs and tissues like the heart. Before we start depends on the type, extension, and frequency of the Contreras et al. Skeletal Muscle (2021) 11:16 Page 7 of 25 injury. For this review, we redefine the fate and hetero- radically increased the attention paid to MCT [93–100]. geneity of muscle-resident PDGFRα+ progenitors and In order to understand MCT development, establish- explain their multilineage potentials. ment, and remodeling, it is crucial to consider the stro- mal cells that participate in these processes. A critical The adult muscle connective tissue and challenging step towards a complete understanding The muscle environment is complex in structure and of MCT biology has been the identification of a hetero- several heterogeneous cell types co-exist within it to geneous population as the primary effector of ECM de- regulate its function and structure. Although skeletal position and remodeling [11, 12, 17, 18, 48, 62]. muscles have an intricate network of blood vessels and Increasing evidence suggests that there are distinct sub- nervous tissue, most of their mass is comprised of myo- sets of stromal cells located in discrete yet similar ana- fibers. In adults, MCT, which accounts for 1–20% of the tomical positions during muscle development and into total dry mass of muscles, surrounds, protects, and inter- adulthood [20–22]. This stromal cellular diversity and connects these primary components [19, 87, 88]. The heterogeneity have been an obstacle to attributing the amount of CT varies significantly from one muscle to primary role for matrix deposition to a specific subset of another, depending on the anatomical location and stromal cells. physiological function of particular muscles [89]. The Jackson and colleagues reported the existence of adult MCT follows the nomenclature of fascia, epimy- tissue-resident mesenchymal progenitors with multiline- sium, perimysium, and endomysium accordingly to its age differentiation capabilities in damaged human location and arrangement within the tissue [90, 91] (Fig. muscle over a decade ago [101]. Today, thanks to the 1b). The topological organization of the covering con- great effort of many researchers, we know that adult nective tissue from the outside is described as follows: MCT is mainly produced by muscle-resident PDGFRα+ the fascia, the CT outside the epimysium that surrounds cells with multilineage progenitor properties and a and separates the muscles; the epimysium, which sur- fibroblast-like phenotype, called FAPs. Increasing evi- rounds each muscle group, linking them to the tendons dence suggests that these muscle-resident cells are the at the myotendinous junctions; the perimysium, which primary cellular source of regenerative matrix deposition consists of collagen-rich structures that surround the as well as scarring following muscle injury, disease, fascicles and interconnect with the epimysium; and the neuromuscular disorders, or aging [1, 2, 5, 9, 11, 12, 14, endomysium, which represents a modified basement 15, 27, 53, 54, 59, 61, 102–105]. Vallecillo-García and membrane unsheathing individual myofibers and inter- colleagues showed that the source of developmental connects to the perimysium [19, 91, 92] (Fig. 1b). These ECM in limb muscles is a heterogeneous population of four levels of stromal organization describe the intercon- PDGFRα-expressing progenitors called embryonic FAPs, nected ECM compartments within muscles. Although closely resembling the population of adult stromal cells each compartment is distinguished by its anatomical we have described, along with other groups [1, 2, 6, 23, position, it is difficult to discriminate each of these ECM 24, 26]. These findings led to some confusion in the no- compartments in terms of their protein and cellular menclature, with some publications distinguishing be- composition. Remarkably, MCT not only determines the tween FAPs and fibroblasts, some using the term FAPs macro and microstructures of embryonic and adult mus- as better representing their predominant fibrogenic and cles but also connects the myofibers to produce and adipogenic developmental potential, and some remaining transmit force. As a result, it increases not only the effi- faithful to the historical term fibroblast, which are also ciency of force generation but also protects myofibers known for being heterogeneous and plastic cells. Here, from excessive stretching, supporting muscle regener- we propose that these muscle-resident multipotent pro- ation and cellular mechanosensation [17–19]. genitors, whether called FAPs or fibroblasts, are the same cells. Muscle-resident fibro-adipogenic progenitors: definitions From this point on, the term PDGFRα+ FAPs will and identity refer to muscle-resident CT mesenchymal progenitor Historically, the observation of ECM proteins, such as with multilineage developmental properties. As dis- collagens, being produced and deposited in skeletal cussed below, recent advances in single-cell RNA se- muscle suggested the existence of a resident collagen- quencing demonstrated that FAPs comprise multiple producing cell within the tissue [17, 18, 90]. Later, nu- sub-populations, some of which could be bona fide dif- merous observations of CT hyperplasia and interstitial ferentiated cells with little developmental potential left proliferation associated with healing scars in skeletal [16, 48, 62, 106–109]. This may create a problem with muscle diseases, including congenital muscular dystro- nomenclature diversity, speculation, and high cellular phies, immobilized muscles, and neuromuscular disor- heterogeneity within the adult stromal lineage [110]. ders (e.g., amyotrophic lateral sclerosis and denervation) FAP heterogeneity is also known to increase following Contreras et al. Skeletal Muscle (2021) 11:16 Page 8 of 25 injury and disease, which also complicates their classifi- These results further confirm our idea that muscle FAPs cation and nomenclature [48, 62, 66, 108, 111]. cannot be solely identified using collagen I reporter The muscle community has historically described mice, but as previously suggested, we strongly recom- interstitial cells with MSC capability (i.e., fibrogenic, adi- mend employing PDGFRα expression. Since most of the pogenic, chondrogenic, and osteogenic potency). In work related to FAPs biology refers to models of single, addition to PDGFRα+ cells, muscle-resident pericytes or repeated rounds of injury, we believe that further have also been proposed to be MSCs that have adapted studies will likely uncover the role of PDGFRα+ cells in to the specialized functions required by their adjacent atrophy-related pathologies such as aging-related sarco- vascular niche. However, although PDGFRα+ FAPs be- penia, cachexia, myasthenia gravis, polytrauma, and have as and present defined canonical MSC properties, neuromuscular disorders. Further research is needed to FAPs are different from tissue-resident pericytic “MSCs.” clarify the existence of subtle differences within stromal Indeed, pericytes' cell-surface profile is CD34-/CD45- cells that might have functional impacts and conse- and CD146+ [112, 113]. Remarkably, Bianco and col- quences in muscle physiology, not only during mainten- leagues revisited the MSC origins and differentiation po- ance but also in pathological and disease states. tential using a broad set of human MSC-like cells (HLA class I, CD73, CD90, CD105, and CD146 positive cells). Multipotency of muscle-resident PDGFRα+ fibro- The authors showed that the cell surface phenotype of adipogenic progenitors “MSCs” isolated from bone marrow, skeletal muscle, In healthy adult muscles, we and others have demon- periosteum, and cord blood, although quite identical, did strated that PDGFRα+ cells represent between ~ 5–15% not reflect these cells’ cell transcriptomic identity, func- of the total nuclei and ~ 20–30% of the interstitial mono- tion, and therefore, their differentiation properties. Thus, nuclear cells at homeostasis [11, 12, 106, 107, 116]. Stro- “MSCs” are separated from each other, as the authors mal PDGFRα+ FAPs display MSC properties and can defined it, by a developmental origin factor [113]. Not- spontaneously differentiate into adipocytes (rounded, ably, the authors also showed that CD146+ pericytes are single-vacuole lipid-rich cells, perilipin+ and peroxisome not true MSCs in most of the analyzed tissues, with the proliferator-activated receptor gamma+ (PPARγ)), acti- possible exception of the bone marrow, where they in- vated fibroblasts (long-shaped contractile cells with herently form bone and bone marrow stroma but lack fibroblast-like morphology, αSMA (Acta2), and highly chondrogenic potential in vivo or myogenic in vitro. On producing ECM cells), as well as chondrocytes/osteoblasts the contrary, in skeletal muscle, CD146+ perivascular when bulk cultured, and in clonal assays in vitro and pericytes are rather inherently myogenic than skeleto- in vivo [1, 2, 11–13 15, 50, 51, 59, 104, 117]. Notably, genic [113]. Remarkably, skeletal muscle pericytes are a HGFA, an injury-induced systemic cue, activates muscle distinct cell type from MuSCs (CD56+/CD146-) and FAPs, priming these cells to transition from quiescence CD34+/CD146+ endothelial cells that possess a latent into a cellular state with enhanced regenerative potential myogenic gene signature and potential, and hence, also known as G alert state [118]. In the following chap- muscle pericytes are committed myogenic progenitors ters, we discuss FAP multipotency (Fig. 2). [113, 114]. These pivotal studies have challenged the loose and non-specific MSC nomenclature. However, Fibrogenic potential of PDGFRα+ FAPs further studies with lineage tracing and clonal assays are When FAPs are cultured in vitro using standard growth needed to deeply understand stromal cell dynamics in media and 20% oxygen, a large proportion of them will development, homeostasis, and injury and, therefore, to spontaneously differentiate into activated fibroblasts finally faithfully unify their markers, nomenclatures, and with αSMA stress fibers [1, 11, 12, 15] (Fig. 2). This definitions. demonstrates that FAPs have intrinsic capabilities to dif- The abundance of collagen, especially the most abun- ferentiate, which is unleashed following their activation dant protein in animals, type I collagen, determines the and makes in vitro studies easily feasible. However, the stiffness of mammalian tissues [115]. Notably, increased mechanisms regulating the fibrogenic potential of FAPs production and deposition of type I collagen fibrils are remain underexplored. found after muscle damage. Several cell sources have been suggested as producers of collagen proteins. Using Transforming growth factor-beta signaling a murine model of increased of increased muscle fibro- One of the most studied signaling pathways in regulating sis, Chapman et al. corroborated that at least three dif- the behavior and fate of muscle FAPs is the transforming ferent muscle-resident cell populations express collagen growth factor-beta (TGF-β) signaling pathway. The I, among them PDGFRα+ FAPs. However, muscle pro- TGF-β sub-family of cytokines (TGF-β1, TGF-β2, and genitors (α7-INTEGRIN)+ and SCA-1+ cells also ex- TGF-β3) are secreted proteins that participate in cell- press the mRNA for this fibrillar matrix protein [17, 18]. and tissue-specific biological processes such as wound Contreras et al. Skeletal Muscle (2021) 11:16 Page 9 of 25 Fig. 2 Skeletal muscle FAPs are quiescent cells with multipotency to differentiate towards all the mesenchymal lineages, depending on the degree of activation and tissue damage. Tissue injury and its associated biochemical cues and cell-secreted factors activate muscle FAPs. Activated FAPs act as immunomodulatory stromal cells and signaling hubs before their commitment to more specialized cells. Usually, muscle injury induces the differentiation of them into activated fibroblasts and adipocytes. Severe damage and chronic pathologies tip their differentiation also into chondrogenic and osteogenic lineages. The figure also shows different molecules and factors as well as ligands that regulate their differentiation potential and fate. Notably, many of these molecules hold several steps of FAP life. As quiescent FAPs find their way into activation and cell differentiation, they lose the expression of quiescence markers and their FAP identity but gain cell differentiation markers healing, angiogenesis, immune regulation, apoptosis, release from the LAP-LTBP complex, which can occur tumorigenesis, and proliferation. In pathological condi- via proteolytic rupture or through ECM-cell forces gen- tions, they strongly associate with tissue damage, dys- erated by cell traction via the integrin complexes [126– function, and fibrosis and are notably mis-expressed 129]. After release, TGF-β binds to its heteromeric (Burks & Cohn, [119, 120]). The complexity of the TGF- serine/threonine kinase type 1 and 2 receptors β pathway is exemplified by its pleiotropic effects, indu- [TGFBR1/ALK5 and TGFBR2, respectively], and cing growth arrest in some cell types but promoting the TGFBR3 (also known as betaglycan) co-receptor on the proliferation of others [121, 122]. TGF-β enhances the cell surface of the target cell. Of interest, while TGF- proliferation and differentiation of several cell types, in- β family ligands can bind TGFBR3, this receptor does cluding stromal cells (for review, see [123, 124]. not have signaling activity on its own, but it modifies the When secreted, TGF-β associates non-covalently to a affinity of TGFBR1 and 2 to TGF-β ligands [130]. In- large complex consisting of the latency-associated pep- deed, TGFBR3 acts as a co-receptor, amplifying TGF-β tide (LAP) and latent TGF-β-binding protein (LTBP) signaling activation [131]. TGFBR3 also binds other proteins [125]. Extracellular TGF-β is activated after its TGF-β-family ligands such as ACTIVINS, INHIBINS, Contreras et al. Skeletal Muscle (2021) 11:16 Page 10 of 25 and bone morphogenetic proteins (BMPs), which we in vitro and in vivo [28]. Thus, as these cells activate, know are primordial proteins for ECM remodeling in proliferate, and differentiate they lose or reduce the ex- skeletal muscle [132, 133]. This co-receptor can also be pression of their progenitor state markers (Fig. 2). soluble (by a mechanism called shedding [134]) and could, in some cases, act as an inhibitor of the TGF-β Wnt/β-catenin signaling signaling by sequestration of its various ligands [131, The Wnt/β-catenin pathway relies on the binding of 135]. Nevertheless, the function of TGFBR3 in FAP or Wnt ligands to Frizzled receptors and the co-receptors mesenchymal progenitor behavior has not been studied LRP5 and LRP6 at the cell surface to initiate a cascade yet. Its modulation could be a powerful tool as TGFBR3 that regulates the intracellular proteostasis of β-catenin misexpression is associated with cancer and metastasis (for recent reviews about the Wnt/β-catenin signaling [136], in which ECM remodeling is known to be highly see [156, 157]. At steady state, the β-catenin pool that is active. not participating in cell adhesion is bound to a destruc- Then TGF-β canonical downstream effectors SMAD2 tion complex, where it becomes phosphorylated and tar- and SMAD3 (R-SMADs) are phosphorylated throughout geted for degradation in a process mediated by the TGFBR1/ALK5 kinase activity and form a cytoplasmic ubiquitin-proteasome system (UPS) [158]. The Wnt heteromeric complex with SMAD4 (co-SMAD) [121, ligand-mediated destabilization of the β-catenin destruc- 122]. This ternary protein complex translocates to the nu- tion complex leads to the accumulation of activated β- cleus where it recognizes SMAD-binding elements (SBE) catenin (unphosphorylated). Accumulated cytoplasmatic in the DNA to regulate the expression of diverse target β-catenin subsequently translocates to the nucleus and genes [123, 137]. In parallel, SMAD6 and SMAD7 act as associates with DNA-binding T-cell factor (TCF) or inhibitors (also called I-SMADs). Their model of action lymphoid enhancer factor (LEF)–TCF/LEF− transcrip- can be various (via TGFBR1 or SMAD4), but their activa- tion factors (TFs) [159]. The binding of β-catenin and tion is often a result of a negative feedback loop aiming to TCF/LEF recruits transcriptional partners and chromatin downregulate the TGF-β or the BMP signaling pathway remodeling complexes to regulate the expression of [138–140]. TGF-β also activates non-canonical down- TCF/LEF target genes [160, 161]. stream signaling pathways such as ABL, PI3K-AKT, RHO, Despite the increasing knowledge about the Wnt sig- TAK1, ERK1/2, JNK, and p38-MAPK [124, 141]. In vitro naling pathway, the participation of Wnt proteins and and in vivo experiments suggest that both canonical and signaling in modulating FAP fate has not been investi- non-canonical TGF-β pathways are involved in fibroblast gated until recently. Skeletal muscle SCA-1+ cells (FAPs) proliferation and myofibroblast differentiation, and are abundant in the muscles of the mdx mice (model of thereby modulate TGF-β-induced fibrosis and ECM re- the Duchenne Muscular Dystrophy (DMD)), and modeling [11, 12, 120, 126, 128, 141–143]. However, the WNT3a treatment promotes their proliferation and col- specific role of TGF-β canonical and non-canonical path- lagen expression both in vitro and in vivo [162]. Interest- ways in regulating muscle-resident PDGFRα+ FAP plasti- ingly, the treatment of dystrophic mice with DKK1 city and fate remains underexplored. (Dickkopf 1, a WNT inhibitor) reduced β-catenin pro- In response to muscle injury, TGF-β is produced and tein levels and muscle fibrosis [162]. On the other hand, secreted by macrophages, FAPs, and regenerating myofi- increased canonical Wnt/β-catenin signaling regulates bers [11, 12, 59, 144, 145]. Muscle FAPs express the satellite cell fate and fibrogenic commitment via cross- three TGF-β isoforms (TGF-β1, TGF-β2, and TGF-β3) talk with TGF-β2 in dystrophic mdx muscles [163]. Ac- and TGF-β receptors (TGFBR1, TGFBR2, and TGFBR3) cordingly, we also observed increased β-catenin protein [11, 12]. TGF-β ligands through TGFBRs induce FAP- levels upon acute glycerol muscle injury [28]. Xiang and myofibroblast differentiation and ECM production [11, colleagues showed that the conditional genetic loss of β- 12, 144, 146]. In addition, TGF-β inhibits the adipogenic catenin in heart fibroblast (Transcription factor 21 priming of muscle FAPs [11], and is pro-mitogenic, and (TCF21) + cells) and activated fibroblasts and myofibro- hence, stimulates the proliferation of PDGFRα+ FAPs blasts (Periostin+) lineages reduces fibrosis and amelio- [11, 12, 59, 144] (Fig. 2). TGF-β signaling pathway acti- rates cardiac hypertrophy induced by pressure overload vation also seems to be required for FAP survival since [164]. In agreement, the sole transgenic overexpression the in vivo treatment of mice with SB431542—a selective of canonical WNT10B is sufficient to induce fibrosis and potent ALK4, ALK5, and ALK7 receptor inhibitor— in vivo [165]. Overall, the Wnt/β-catenin pathway regu- reduced the number of expanded FAPs following rotator lates the expression of several ECM genes in fibroblasts cuff tear injury [147] (Table 3). Remarkably, we also from different tissues and organs following injury and showed that TGFBR1 and the p38-MAPK protein are re- disease [28, 164–166]. sponsible for TGF-β-mediated downregulation of PDGF The outcomes of Wnt/β-catenin signaling depend on Rα [11], associated with a decrease in TCFL2 expression the TCF/LEF TFs. However, the potential roles of them Contreras et al. Skeletal Muscle (2021) 11:16 Page 11 of 25 Table 3 Summary of drug strategies to target muscle fibro-adipogenic progenitor differentiation and fate Therapy Target Cell Proliferation Cell death/ Fibrogenesis Adipogenesis References survival apoptosis AG1296 PDGFR kinase activity Not Reduced? Not evaluated Reduced Not evaluated [11] inhibitor evaluated AICAR AMPK activator Reduced Not Induced Not Reduced [148] evaluated evaluated Azathioprine Immunosuppressant Not Reduced Not affected Not affected Reduced [149] affected Batimastat MMPs inhibitor Not Not affected Not affected Not affected Reduced [14][104] (including MMP14) affected BMS493 Pan-retinoic acid Not Reduced Not evaluated Reduced Induced [69] receptor (RAR) inverse evaluated spontaneous agonist differentiation Dexamethasone Glucocorticoid Induced Induced Not affected Not Induced [150] receptor evaluated HDAC inhibitors (TSA and HDACs Not Not Not evaluated Reduced Reduced [28][151] Pracinostat) evaluated evaluated [152]; LY2090314 & other GSK GSK3 inhibitors Slightly Not affected Not affected Mixed results Reduced [153] inhibitors decreased Metformin AMPK activator Not Reduced Not evaluated Not Reduced [16] evaluated evaluated Molsidomine NO donating molecule Reduced? Reduced? Not evaluated Reduced Reduced [154] Promethazine hydrochloride H1 histamine receptor Not Not affected Not affected Not Reduced [72] affected evaluated SB525334/SB431542 TGFBR kinase activity Reduced Reduced Induced after Reduced Not evaluated [11, 12][147] inhibitor long treatment TKIs (imatinib, nilotinib, Abl, PDGFRs, Kit, DDRs, Reduced Reduced Induced Reduced Reduced and/or [11, 12]; [16]; crenolanib, sorafenib, and p38 Induced [59]; [155]; masitinib) [146]; HDACs-mediated effects on FAP fate are seen only in young mdx but not aged mdx mice [16] reported that imatinib enhances the amount of perilipin+ FAP-derived adipocytes in vitro in muscle FAPs are underexplored. These TFs recognize idiopathic pulmonary fibrosis and heart fibroblasts [28]. TCF/LEF-binding elements and regulatory regions of Hence, our work confirms the cross-talk between the target genes to regulate gene expression. In this context, Wnt and TGF-β pathways that controls the fate of we showed the expression of the four Wnt TCF/LEF PDGFRα+ cells and potentially fibrosis (Table 3). In members in MSC and fibroblast cell lines, as well as summary, the Wnt cascade modulates TGF-β-mediated tissue-resident FAPs from skeletal muscle and cardiac effects in fibroblasts, and vice versa [28, 167–169]. tissues [28]. We observed that Tcf7l2 and Tcf7l1 were the two most highly expressed members, whereas the Platelet-derived growth factor signaling fibroblast lineage, including FAPs, express Tcf7 and Lef1 The platelet-derived growth factor (PDGF) signaling at lower levels. Moreover, treatment with TGF-β de- pathway regulates not only vascular development and creases both the mRNA and protein levels of TCF7L2 in angiogenesis [170] but also plays crucial roles during de- PDGFRα+ cells. We described that this regulatory mech- velopment, stem cell fate, migration, and proliferation. anism requires the transcriptional regulation activity of PDGF receptors (PDGFRs) are the cell membrane- histone deacetylases (HDACs) and the participation of bound tyrosine kinase receptors for PDGF ligands [171– the UPS [28]. 174]. PDGFs were initially described as serum-derived Interestingly, TGF-β activates the canonical Wnt/β-ca- mitogens essential for fibroblast and smooth muscle cell tenin cascade and induces nuclear accumulation of β- growth [175, 176]. PDGFs ligands are four gene products catenin, which in turn reduced the expression of the consisting of five dimeric isoforms: the homodimers WNT inhibitor DKK1 [165]. In agreement with our PDGF-AA, PDGF-BB, PDGF-CC, PDGF-DD, and the most recent results showing that TGF-β reduces the ex- PDGF-AB heterodimer [177]. PDGFs are known for be- pression of several TCF7L2 target genes, whereas it pro- ing released from α-granules of platelets and are potent motes the expression of ECM remodeling genes in chemoattractants and mitogens for cells of mesenchymal Contreras et al. Skeletal Muscle (2021) 11:16 Page 12 of 25 origin [178]. However, several other cell types express proliferative and differentiation-related downstream sig- and secrete these ligands, such as inflammatory cells naling pathways such as PI3K-AKT, ERK1/2, p38- (e.g., macrophages) and fibroblasts [179]. Post- MAPK, and STAT3 in PDGFRα expressing cells [11, translational proteolytic processing of PDGFs is neces- 12]. Recently, Farup et al. showed that PDGF-AA treat- sary for their activation. It occurs extracellularly for ment increases the expression of collagen type I in FAPs, PDGF-C and PDGF-D but intracellularly for PDGF-A, whereas it reduces their adipogenic differentiation (Fig. PDGF-B, and PDGF-AB [178, 179]. A biologically active 2). Notably, the PDGF-AA-mediated fibrogenic fate of PDGF ligand is a dimer of two single PDGF chains, FAPs associates with a metabolic switch that promotes which binds one PDGFR. enhanced glucose consumption [16]. Hence, PDGF sig- PDGFRs genes (PDGFRA and PDGFRB) encode naling could regulate the potency and fate of skeletal single-pass transmembrane receptors with an extracellu- muscle FAPs (Fig. 2). lar portion of five immunoglobulin-like domains, a In the heart, Asli et al. showed that PDGF-AB treat- transmembrane segment, a juxtamembrane segment, a ment promotes colony formation and self-renewal of tyrosine kinase domain, and a carboxy-terminal tail cardiac fibroblast, whereas the PDGFR inhibitor, [180]. PDGFRs are monomeric before exposure to PDGF AG1296, suppressed these activities [186]. Interestingly, [181]. Its ligand binding-induced dimerization causes activated PDGFRαH2BEGFP-mid fibroblasts formed at their activation, and therefore, later PDGFR de- the expense of resting PDGFRαH2BEGFP-high fibro- repression and activation of the receptor's tyrosine kin- blasts [73, 186]. These results are in agreement with our ase activity [180, 182, 183]. Three known functional recent findings where the expression of PDGFRα dimer forms of the receptors exist. They consist of the changes dynamically during muscle regeneration and re- PDGFRα/α and PDGFRβ/β homodimers and the PDGF pair [11]. Moreover, in vivo PDGF-AB treatment of un- Rα/β heterodimer [179, 180]. PDGF-AA, PDGF-AB, injured hearts did not cause fibroblast activation; PDGF-BB, and PDGF-CC promote PDGFRα/α homodi- however, it increased the number of PDGFRαH2BEGFP- mer formation, PDGF-BB, PDGF-CC, PDGF-DD, and mid fibroblasts after myocardial infarction [186]. There- PDGF-AB promote PDGFRα/β heterodimer assembly. fore, PDGF-AB isoform targets tissue-resident fibroblasts PDGFRβ/β homodimer can only be induced by PDGF- by increasing the activated fibroblast pool after injury. BB and PDGF-DD isoforms [177, 178, 180]. Although Interestingly, genetic loss of Pdgfra in the resident car- the precise role of PDGF and its receptors in vivo in diac fibroblast lineage (TCF21+ cells) results in an over- muscle-resident FAPs is unknown, PDGF signaling all reduction in the fibroblast population in adult hearts, seems to regulate FAP survival, activation, proliferation, demonstrating that PDGFRα regulates fibroblast main- migration, and fate. In this review, we focused on PDGF tenance and homeostasis [187]. Consistently, lineage- ligands and PDGFRα in skeletal muscle health and specific deletion of Pdgfra in tubulin polymerization- pathophysiology. promoting protein family member 3 expressing cell Treatment of ex vivo FAPs with PDGF-AA and PDGF- population (Tppp3+ tendon stem cells) caused impaired BB ligands activates the PDGF cascade inducing FAP ac- tendon regeneration, and therefore, corroborates the cell tivation and proliferation (Fig. 2)[11, 53]. In addition, requirements of PDGFRα signaling for proper tendon upregulated expression of ECM genes and activated healing [188]. Remarkably, the passaging of plastic ad- downstream ERK1/2, PI3K-AKT, and SMAD2/3 signal- herent FAPs obtained from muscles reduces the protein ing pathways is observed in ex vivo FAPs in response to levels of PDGFRα, which associates with their differenti- PDGF-AA treatment [53, 184]. By utilizing a pharmaco- ation [12]. Thus, cellular PDGFRα bioavailability may be logical inhibitor of PDGFR signaling, Mendias and col- a modulating factor in PDGF-mediated responses of leagues showed that PDGFR signaling modulates muscle FAP lineage during survival, fate decisions, and damage- ECM remodeling and angiogenesis upon synergist abla- associated behaviors. tion surgery to induce postnatal muscle growth or hypertrophy [185]. In addition, the treatment with Adipogenic potential of PDGFRα+ FAP cells PDGF-AA induces the phosphorylation of PDGFRα and Infiltration and deposition of fatty adipose tissue are the proliferation of PDGFRα+ cells (Fig. 2)[53]. The au- hallmarks of several skeletal muscle pathologies. How- thors also suggested, using pharmacological inhibitors, ever, the cellular and molecular mechanisms underlying that both PI3K-Akt and MEK2-MAPK signaling path- fatty infiltration of muscles have not been extensively in- ways are necessary for PDGFRα-induced proliferation vestigated compared with the ever-growing research in [53, 54]. However, persistent PDGF ligand exposure and muscle fibrosis. A better understanding of such a enhanced PDGFRα signaling levels can cause patho- discrete fat compartment between myofibers and fascia, logical muscle fibrosis [53, 54, 155, 184]. We have re- also called intra/intermuscular adipose tissue (IMAT), cently shown that PDGF-BB treatment activates Contreras et al. Skeletal Muscle (2021) 11:16 Page 13 of 25 may allow for the targeting of these adipogenic progeni- to multiply and hypertrophy [195, 196]. On the other tors to increase muscle regeneration and repair. hand, brown fat cells are equipped with smaller droplets The lack of reliable cell-specific markers for fat pre- and large mitochondria concentration, giving the tissue cursor cells has been the main limitation of studying its chestnut hue. Hence, in brown adipose tissue, mito- IMAT. As described above, the studies of Joe et al. and chondria produce heat using these fatty droplets, a Uezumi et al. helped to clarify many aspects of the process also known as thermogenesis [197]. The role of muscle adipogenic precursor cells. One major focus of FAP-derived fat cells, whether brown/beige or white, in these approaches was determining whether IMAT- skeletal muscle health, regeneration, and disease is associated adipocytes were in vivo derived from pre- unknown. existent muscle-resident PDGFRα+ cells, other muscle- Perhaps the most serious disadvantage of these studies resident cells, or circulating cells. The work led by Liu is that they do not directly address the in vivo adipo- et al. in murine skeletal muscle is a classic example of genic differentiation potential of adult PDGFRα+ FAPs. these efforts. The authors suggested that IMAT derives The definitive proof that muscle PDGFRα+ cells are the from a lineage of cells not expressing Pax3 (i.e., non- main, if not the only, source of injury-induced adipo- myogenic). They also showed that the genetic ablation cytes came from lineage tracing experiments using CreERT EYFP of intramuscular adipogenic progenitors based on Ap2 Pdgfra :Rosa26 transgenic mice [14]. The au- (also known as fatty acid-binding protein 4 (FABP4)) ex- thors demonstrated that seven days after acute intramus- pression leads to impaired skeletal muscle regeneration, cular injury, a large proportion of perilipin+ adipocytes suggesting for the first time that damage-induced fatty derived from PDGFRα+ FAPs, indicating that PDGFRα tissue may support efficient regeneration upon acute in- expressing progenitors are the major source of damage- jury [189]. However, AP2/FABP4 expression is com- induced fat cells in normal muscle regeneration and in monly thought to be restricted to committed or muscular dystrophy. Indeed, using similar lineage tracing differentiated adipocytes than progenitor cells [190], strategies we have demonstrated that cardiac PDGFRα+ questioning the interpretation of the results. Marinkovic FAPs can cause fibrofatty infiltration within the myocar- et al. [111] showed that Notch signaling is a pivotal dium in an arrhythmogenic cardiomyopathy mouse pathway regulating FAP adipogenesis in wild-type cells model driven by the conditional deletion of the and that dystrophic FAPs are insensitive to Notch- quiescence-associated factor Hic1 in heart FAPs [73]. mediated adipogenic inhibition compared with acute Intriguingly, PDGFRα+ FAPs are ciliated cells and thus injury-derived FAPs [111]. Hence, these results demon- possess primary cilium. Conditional deletion of a gene strate that wild-type and dystrophic muscle PDGFRα+ required for ciliogenesis, Ift88, in FAPs impaired the FAPs are in different functional states, which influences injury-induced formation of adipocytes [14]. Mechanis- their fate and responsiveness to extracellular cues, as tically, the cilia-dependent modulation of FAP adipogen- previously suggested [62]. esis involves the participation of Sonic Hedgehog (SHH) Human PDGFRα+ FAPs exist in healthy and DMD signaling, which is repressed in the absence of cilia. In- pathological muscles, being bona fide counterparts of deed, constitutive activation of the Shh-pathway via gen- the PDGFRα+ cells found in mouse muscles [16, 50, 51, etic deletion of the repressor Ptch1 was sufficient to 53, 54, 109]. Remarkably, FACS-isolated human FAPs block adipocytes’ generation following injury [14]. Re- (CD15+/PDGFRα+/CD56−) differentiate towards fully markably, elimination of the primary cilium in PDGF mature adipocytes, phenocopying the in vitro differenti- Rα+ FAPs led to enhanced regeneration of myofibers by ation kinetic and potential of adipose stromal cells ob- reducing fatty degeneration of dystrophic muscles, tained from subcutaneous adipose tissue depots [51]. which was also associated with increased myofiber Moreover, when transplanted into a glycerol-damaged size. The authors also showed that tissue inhibitor of muscle (an injury model that promotes adipogenesis) metalloproteinase 3 (TIMP3), an ECM modifier, in- [191, 192], murine FAPs readily differentiate into adipo- hibits adipocyte formation by muscle PDGFRα+pro- cytes. In concordance to the in vitro report of Liu and genitors. Interestingly, aiming to mimic TIMP3 colleagues using mouse muscle samples, Arrighi et al. activity, the authors utilized batimastat and showed also showed that FAP-derived adipocytes from human that the treatment with this pharmacological inhibitor muscle biopsies are white rather than beige/brown fat of metalloproteinases prevented injury-induced adipo- cells. In contrast, Gorski et al. showed increased expres- genesis in vivo [14](Table 3). sion of UCP1, a brown/beige fat cell marker [193], in In a different study, Jaiswal and colleagues showed that muscle as well as in FAP cultures following induction of the treatment with Batimastat prevented FAP spontan- IMAT by glycerol injection [194]. Throughout the body, eous adipogenesis and reduced fat in dysferlinopathic white fat cells store energy in large, often single, oily muscle of dysferlin-deficient (B6A/J) mice [104]. Hence, droplets. Obesity causes these white adipose tissue cells the authors suggested that the accumulation and Contreras et al. Skeletal Muscle (2021) 11:16 Page 14 of 25 adipogenic differentiation of FAP are critical contribu- Feeley and colleagues showed that rotator cuff tears en- tors to limb-girdle muscular dystrophy type 2B. Surpris- hanced HDAC activity in FAPs and trichostatin A inhib- ingly, the authors observed no changes in either FAP ited it. HDAC inhibition prevented FAP-mediated fatty accumulation, proliferation, or fibrosis as a result of bati- infiltration in supraspinatus muscles. Also, trichostatin mastat treatment (Table 3). Nevertheless, the batima- A regulates muscle FAP adipogenesis by promoting FAP stat’s off-targets on other tissue-resident cells such as browning (Table 3)[152]. MuSCs, myofibers, endothelial cells, pericytes, or infil- These studies demonstrated that HDACs-mediated trating CD45+ cells have not been evaluated yet. pharmacological intervention might counter DMD pro- Altogether, these findings suggest novel strategies to gression and chronic muscle injury by increasing regen- combat fatty degeneration of chronically damaged mus- eration by inhibiting fibro-fatty degeneration while cles by targeting the adipogenic conversion of PDGFRα favoring the interplay and communication between FAPs expressing FAPs to inhibit the deposition of injury- and and myogenic progenitors. Recently, we have shown that disease-induced intramuscular fat. two well-characterized pan-HDACi reduce TGF-β- To date, there is not a single clinically approved drug induced ECM gene expression and also block TGF-β- used to prevent IMAT accumulation in muscle disease. mediated downregulation of Tcf7l2 expression [28]. However, significant pre-clinical advances have been Mechanistically, histone deacetylase inhibitors modulate made. In vivo treatment of mdx mice with molsido- TGF-β-mediated changes in the expression of TCF7L2 mine—a nitric oxide (NO) donating molecule—reduced transcription factor target genes of the Wnt pathway muscle pathology, IMAT accumulation, and fibrosis [28]. Further investigations should unravel the mechan- [154]. These improvements were at least in part medi- ism by which HDACs regulate the fate of FAPs and how ated by the inhibition of NO-mediated FAP adipogenesis could this be used to target muscle-associated diseases. (Table 3). Hence, altered synthesis of NO, a typical find- In a recent study, Reggio and colleagues used a large ing in DMD, could contribute to enhanced fat depos- drug library screen with pharmacological approaches to ition. On the search for adipogenic inhibitors, Uezumi demonstrate that the inhibition of the cytoplasmic sig- and colleagues found that promethazine hydrochloride naling protein, glycogen synthase kinase 3 (GSK3), re- inhibits, through binding to the H1 histamine receptor, duces PDGFRα+ FAP adipogenesis in vitro, while also the in vitro and in vivo formation of ectopic adipocytes repressing muscle glycerol-induced fatty degeneration derived from PDGFRα+ lineage cells in the muscle [72] [153] (Table 3). GSK3 is composed of 2 isoforms (α and (Table 3). Promethazine hydrochloride is a first- β) and is part of the destruction complex of β-catenin, generation antagonist of the H1 histamine receptor, and which we showed earlier to play a modulatory role in therefore, this family of drugs emerge as attractive novel FAP fate (see the “Wnt/β signaling” section). Mechanis- therapeutics against ectopic fat formation in muscle tically, the authors suggested that UPS-targeted β- pathologies. catenin degradation causes an imbalance in the adipo- Histone deacetylation leads to the repression of gene genic fate of dystrophic mdx FAPs. The authors also expression, and histone deacetylase inhibitors (HDACi, exploited single-cell data and in silico modeling to show like trichostatin A) provide an exciting means to treat that PDGFRα+ FAPs compose the core of the stromal DMD. HDACi have been used in both pre-clinical and cells in the muscle cell niche by expressing Wnt compo- clinical studies to improve muscle regeneration and re- nents and also for being the primary source of Wnt li- pair in DMD [151, 198–200]. As HDACi treatment in- gands. FAPs seem to actively communicate with hibits fibro-fatty differentiation of PDGFRα+ FAPs, it endothelial cells, tenocytes, and MuSCs through the reduces the dystrophic pathology through increasing production of Wnt ligands. Among the Wnt ligands, muscle regeneration [151]. Remarkably, in dystrophic they observed that dystrophic FAPs downregulate FAPs, an HDAC–myomiR–BAF60 molecular network Wnt5a expression compared with wild-type cells. regulates FAP fate, and old FAPs become resistant to Moreover, WNT5a treatment reduced FAP-induced HDACi-induced chromatin remodeling compared with adipogenesis in vitro by repressing PPARγ expression young FAPs [201]. Also, HDACi restore the dystrophic- throughout the activation of β-catenin, suggesting mediated loss of intercellular communication between that the Wnt signaling modulates the adipogenic PDGFRα+ FAPs and myogenic progenitors required for commitment of FAPs in dystrophic muscles (Fig. 2). proper muscle regeneration [151], and as recently sug- On the other hand, Zhao and colleagues recently de- gested through an extracellular vesicle-mediated transfer scribed that the supplementation of retinoic acid (RA) of miRNAs [200]. Interestingly, aging and DMD disease enhances the proliferation of FAPs at the expense of progression limit HDACi-mediated effects [151], which inhibiting their adipogenic and fibrogenic differentiation suggests that aging affects the fate of FAPs, as recently [69]. Additionally, treatment of isolated FAPs with a detailed by Lukjanenko and colleagues [5]. Recently, pan-retinoic acid receptor antagonist, BMS493, blocked Contreras et al. Skeletal Muscle (2021) 11:16 Page 15 of 25 the RA-mediated effects. Notably, the authors also Adiponectin in skeletal muscle FAPs [11, 12]. We also showed that RA treatment rescued obesity-impaired showed that the adipogenic differentiation of FAPs re- skeletal muscle regeneration. These findings showed a presses the expression of PDGFRα [11]. Taken together, FAP-type specific effect of RA signaling that regulates these studies demonstrate that IMAT-associated adipo- skeletal muscle regeneration and repair by means of pre- cytes can derive from pre-existent muscle-resident fibro- serving their progenitor state. Taken together, these adipogenic progenitors. findings suggest a novel potential retinoic acid-based strategy to combat chronic skeletal muscle fibro-fatty Osteogenic differentiation of PDGFRα+ FAP cells degeneration of obese patients. Muscle PDGFRα+ FAPs have osteogenic potential On the contrary, several factors positively regulate in vitro [2] and when transplanted can successfully en- muscle FAP adipogenesis. For instance, the matricellular graft and form calcification-rich structures using an protein CCN family member 1 (CCN1/CYR61) is ele- in vivo heterotopic ossification (HO) model [208]. HO is vated in the serum and sarcopenic muscles of a murine a musculoskeletal disorder distinguished by the patho- model of chronic kidney disease and induces FAP adipo- logic formation of extraskeletal bone in muscle, tendon, genesis [202]. In vivo treatment of mice with the gluco- ligaments, and fascia [209]. BMP2 promotes intramuscu- corticoid dexamethasone enhanced IMAT deposition lar HO regardless of damage; however, BMP9-induced following acute injury (Dong et al., 2014 [150]). Dexa- HO requires skeletal muscle injury [210] (Fig. 2). The methasone also induces FAP proliferation while increas- authors described that intramuscular HO might involve ing their adipogenesis, possibly involving the reduction a population of Lin-SCA-1+ cells—likely FAPs [210]. of IL-4 expression (Dong et al., 2014 [150]). Remarkably, Moreover, Lin /TIE2+/PDGFRα+ progenitors respond IL-4 administration reduces dexamethasone-induced to BMP2-stimulated osteogenic commitment and con- FAP-derived adipocyte formation, suggesting a novel tribute to HO in mice [211]. Additionally, muscle- therapeutic use of IL-4 to reduce IMAT accumulation derived MSCs contribute to fracture repair in a tumor due to glucocorticoid use in DMD patients (Fig. 2). Per- necrosis factor-alpha (TNFα) dependent manner [212]. petuini et al. showed that the glucocorticoid-related The above findings are consistent with a recent study of molecules, dexamethasone, and budesonide, inhibited Goldhamer’s group, where the authors employed and the insulin-induced adipocyte formation from mdx- characterized a transgenic mouse model that recapitu- derived FAPs. However, both drugs have a pro- lates a rare autosomal-dominant disorder called fibro- adipogenic impact when the adipogenic mix contains dysplasia ossificans progressiva (FOP), which results factors that increase the concentration of cyclic AMP. from a single activating mutation in ACVR1; the type I The authors also showed that, only in anti-adipogenic BMP receptor also known as ACVR1/ALK2. The Tie2- conditions, budesonide suppresses the expression of driven expression of the mutation Acvr1 R206H is suffi- Pparg, a master adipogenic regulator, via the cient to phenocopy the spectrum of HO observed in glucocorticoid-induced-leucine-zipper (GILZ/TSC22D3), FOP patients [60]. Moreover, they also showed that and the glucocorticoid antagonist mifepristone alleviates intramuscular transplantation of mutant Acvr1R206H/+ such inhibitory effect [203] (Table 3). This study may FAPs into immunodeficient mice resulted in the forma- shed light on some of the mechanisms underlying the tion of HO in an Activin A-dependent fashion. Overall, use of glucocorticoids in DMD patients under this kind these data established TIE2+/SCA-1+/PDGFRα+ FAPs of treatment. The use of glucocorticoids to treat DMD as the predominant cell-of-origin and driver of patho- patients is so far the most common treatment available logical HO. However, it has been suggested that TIE2 is to delay muscle necrosis and degeneration up to date a nonspecific marker for a subset of PDGFRα+ cells [204–206]. Finally, the same group, using a similar since its expression overlaps with other cell populations chemical library-based approach, identified an immuno- like endothelial cells, MuSCs, and subsets of suppressant drug, azathioprine, that negatively perturbs hematopoietic cells [2, 213, 214]. Hence, the precise the intrinsic adipogenic fate, also via PPARγ repression, mechanisms and the populations of cells involved in the of wild type and mdx PDGFRα+ FAPs (Table 3). formation and remodeling of HO remained unknown On the other hand, we recently showed that TGF-β until then. We recently took advantage of a novel PDGF treatment negatively affects FAP differentiation to adipo- Rα lineage tracing reporter mouse (Pdgfrα-CreERT2- cytes while inducing FAP-to-myofibroblast commitment TdTomato) to further explore the cellular source of (Fig. 2). TGF-β1 impairs basal PDGFRα+ FAP differenti- muscle ossification [13]. Using a model of BMP2- ation into the adipogenic lineage, by reducing the stimulated intramuscular HO, we showed that a large steady-state percentage of adipocytes but increasing the proportion (~80%) of differentiated osteogenic cells were number of myofibroblasts [11, 207]. Mechanistically, TdTomato+ after 21 days of muscle injury. Thus, the TGF-β treatment reduces the expression of Pparγ and cell-source responsible for forming ectopic bone in Contreras et al. Skeletal Muscle (2021) 11:16 Page 16 of 25 muscle is a subpopulation of muscle-resident PDGFRα+ single-cell omics technologies to refine our understand- progenitors [13]. Overall, these studies demonstrate that ing of cell heterogeneity by using a plethora of genes FAPs are a significant cellular source of chondrogenic and proteins to identify a particular cluster or subpopu- cells and osteogenic cells in severely damaged muscles. lation of cells, they are significantly more accurate com- Remarkably, intramuscular calcium deposits serve as a pared with the use of a single marker to identify cell pathohistological feature of DMD [215]. Notably, the de- types. gree of osteogenic commitment of FAPs appears to The classical view of cellular muscle composition is match the model of muscle damage and degeneration/ that most of the non-myogenic cells play a positive role regeneration used. Using the severe D2-mdx (DBA/2J- and generate a pro-regenerative transitional niche, mdx) dystrophic mice, which better recapitulates the hu- which, among other functions, support MuSC-driven man characteristics of DMD myopathology, Mázala et al. myogenesis following acute damage [217]. These popula- demonstrated that PDGFRα+ FAPs accumulate within tions of non-myogenic cells include endothelial cells calcified deposits in degenerative muscles [117]. Also, (CD31+) [218, 219], FAPs (PDGFRα+) [1, 2], connective the in vitro osteogenic differentiation of these cells posi- tissue fibroblasts TCF7L2+ (significantly overlapping tively correlates with the degree and extension of muscle with FAPs) [6, 24, 27, 28], pericytes (NG2+, RGS5+) degeneration and TGF-β levels, which supports previous [48], mesoangioblasts [220, 221], tenocytes (TNMD+, studies showing that FAPs vastly expand and accumulate SCX+), glial cells (PIP1+, KCNA1+) [48, 58], and a com- accordingly with the extension of damage, TGF-β levels, plex array of immune cells [222–224]. and fibrosis [11, 12, 15, 27, 53, 54, 59, 103, 117]. In sum- With single-cell omics technologies, the transcriptional mary, FAP activity and responses are highly contextual, identity in homeostasis and lineage trajectories of which suggests that signals emanating from the local muscle-resident FAPs during the regenerative response niche determine their phenotypic multi-lineage-fate. and in disease states have started to be discovered. Male- Why are different muscle groups affected to a different cova and colleagues were the first to initially show the extent in muscular dystrophy or neuromuscular disor- existence of cellular heterogeneity in muscle FAPs using ders? Although several hypotheses might explain this, in- single-cell RT-PCR and showed that it increases in re- cluding muscle fiber type, muscle fiber innervation, generating muscles [62] (Table 4). The authors showed muscle of origin, calcium homeostasis, and muscle activ- that a specific subpopulation of vascular cell adhesion ity, we still lack information of the role that FAPs play in molecule VCAM-1+ FAPs vastly expands and drives these processes. muscle fibrosis in acute damaged muscles and adult dys- trophic muscles of the mdx mice [62]. Notably, the Fibro-adipogenic cell diversity: Single-cell omics VCAM-1+ subFAPs are absent in uninjured muscles, unveil stromal populations in muscles suggesting that Vcam1 may be an activation marker. Re- The recent revolution in single-cell omics technologies, cently, a population of quiescent HIC1+ mesenchymal including single-cell RNA sequencing (scRNAseq), progenitor cells has been described. This population single-cell epigenomics (e.i. scATACseq), and single-cell contains precursor cells for several mesenchymal line- mass cytometry (e.i., CyTOF) has helped to uncover the ages including muscle FAPs (Pdgfra+, Ly6a+), tenocytes mysteries of muscle cellular composition and heterogen- (Tnmd+), pericytes (Rgs5+), and a new subset of cells eity as well as to faithfully recreate a more precise cellu- called myotenocytes (Col22a1+) [48]. However, further lar atlas of murine and human adult skeletal muscle in functional analyses will be required to confirm if the homeostasis, regeneration, and repair [48, 58, 62, 106– myotenocyte cells present at the regenerated myotendi- 109, 111, 216] (Fig. 3). Muscle single-cell analyses faith- nous junction represent an independent subpopulation fully recapitulate key cellular events involved in skeletal of HIC1+ progenitors with a specific function or a differ- muscle regeneration and repair, derived from studies entiated cell state of tenocyte progenitors in the myoten- over many years. Such tools and information led us to dinous niche. Even though HIC1 is a broad stromal realize that a complex array of non-myogenic cells (tis- progenitor marker, FAP gene signatures segregate from sue-resident and non-tissue-resident) engage in active other interstitial populations of mesenchymal cells like cross-talk between each other and with MuSCs to re- pericytes and tenocytes [48]. Remarkably, the gene sig- store tissue function following damage. Single-cell stud- nature of muscle PDGFRα+ FAPs is heterogeneous and ies evaluate molecular signatures and expression levels progresses over time after acute muscle injury, which re- of genes or cell surface protein abundance in large num- veals their dynamic role in regeneration [48, 108]. bers of individual cells. They aim to describe at an un- Therefore, FAPs acquire a unique plastic transcriptome precedented resolution the total interstitial populations that changes as the inflammation progresses and damage of cells in a resting state and to understand their flux in resolves through regeneration. Concerning mouse mus- response to injury and disease. Owing to the ability of cles, two major FAP populations have been described Contreras et al. Skeletal Muscle (2021) 11:16 Page 17 of 25 Fig. 3 a Single-cell RNA sequencing analyses to map muscle-resident FAP mononuclear landscape in murine (left graph) and human (right graph) skeletal muscle tissue. Three different studies, utilizing mice, agree with the existence of at least two principal muscle FAP subpopulations (here shown as FAPs 1 and FAPs 2; see text for details). On the other hand, FAP clustering and FAP subpopulations greatly vary in human muscles. Two different bioinformatic techniques for the presentation of large scRNA-seq datasets and their dimensionality reduction are shown: uniform manifold approximation and projection (UMAP) algorithm and t-Distributed Stochastic Neighbor Embedding (t-SNE). Colored dots represent individual FAP cells. Dotted lines illustrate the different studies discussed in this review. b FAP cell trajectories are based on the gene signatures of single cells following damage [48, 108]. The transcriptomes of FAPs indicate high cellular heterogeneity within the FAP populations in response to injury. In mouse muscles, two major FAP subpopulations (Dpp4 FAPs and Cxcl14 FAPs) are present in homeostatic conditions (for detailed markers, see Table 4). Analysis of the pseudotime trajectory of different FAP subpopulations suggests that FAP cells follow a continuum and diverge into two major subclusters upon damage Contreras et al. Skeletal Muscle (2021) 11:16 Page 18 of 25 Table 4 scRNA-seq gene signatures used for FAP identification and clustering in muscle homeostasis Genes/markers FAP subpopulations Species Reference low low Sca1, Cd34, Pdgfra SubFAPs: Tie2 (Tek), Vcam1 Mouse [62] high high SubFAPs: Tie2 (Tek), Vcam1 Ly6a (Sca1), Ly6e, Pdgfra, Dcn Not determined Mouse [58] Pdgfra, Ly6a (Sca1), Hic1 FAP1: Cxcl14, Col4, Col6, Col15, Lum, Sparcl1, Podn, Smoc2, Mgp, and Bgn Mouse [48] FAP2: Dpp4, Sfrp4, Igfbp5, Sema3c, Tgfrb2, and Wnt2 Pdgfra, Ly6a (Sca1), Dcn, Cd34 Not determined Mouse [216] Pdgfra, Col3a1, Dcn, and Gsn Not determined Mouse [107] Pdgfra, Ly6a (Sca1), Cd34 FAP1: Cxcl14, Enpp2 (Autotaxin), Crispld2, Hsd11b1, Smoc2, Ccl11, Gsn, and Dcn Mouse [108] FAP2: Dpp4, Pi16, Wnt2, Igfbp5, Igfbp6, Fbn1, and Ugdh PDGFRA, CD34, COLLAGEN 1, LUMICAN (LUM) FAP: LUM, DCN, CXCL14, COLLAGEN 4, and COLLAGEN 15, SMOC2, and GSN Mouse [109] COLLAGEN 3, and COLLAGEN 6 and FIBRILLIN 1 (FBN1) FAP: FNB1, MFAP5, LOXL1, PRG4, ELN, IGFBP5, and FSTL1 human PDGFRA FAP1 (fibroblasts 1): COL1A1, SFRP4, SERPINE1, and CCL2 Human [106] FAP2 (Fibroblast 2): FBN1, MFAP5, and CD55 FAP3 (Fibroblast 3): SMOC2, ADH1B, and ABC18 PDGFRA, CD34, COL1A1, COL6A3, FAP1: PCOLCE2, MFAP5, IGFBP6, ENNP1, CD55, and AXL Human [16] TCF7L2 FAP2: LUM, MYOC, CCL2, ADH1B, SFRP2, CXCL14, and MGP FAP3: TNXB, C3, COL15A1, SMOC2, ABCA8, COL6A1, and HMCN2 FAP4: IGF1, CRLF1, SCN7A, ITIH5, PTGDS and NOV FAP5: SEMA3C, PRG4, DEFB1, CCDC80, LINC01133, and IGFBP5 By re-clustering the FAP population, the authors described the existence of 7 different FAP subpopulations in human muscles [16] (FAP1 and FAP2) in undamaged muscles (Fig. 3a). FAP1 expressing chemokine genes like Cxcl5, Cxcl3, Ccl7, and associates with the ECM gene signatures, such as colla- Ccl2) at 0.5 and 2-day post-injury (DPI), then progres- gens Col4, Col6 and Col15, Lum, Sparcl1, Podn, Smoc2, sing into WISP1+ FAPs at 3.5 and 5 DPI (highly ex- Mgp, Cxcl14, and Bgn. On the other hand, the FAP2 pressing Postn, Csrp2, Sfrp2, Ptn, Cilp, and Cthrc1), subpopulation expresses genes involved in cell signaling followed by DLK1+ FAPs at 10 DPI (expressing Itm2a, and migration, including Sfrp4, Igfbp5, Sema3c, Dpp4, B830012L14Rik, Meg3, Airn, Peg3, Zim1, H19, and Igf2), Tgfrb2, and Wnt2 [48] (Table 4). In contrast, another and finally two FAP subpopulations at day 21 DPI, study described no segregation of the FAP population, OSR1+ (expressing Gsn, Ccl1, Bmp4, Bmp5, and Wnt5a) primarily based on the expression of a few FAP markers and fibroblast FAPs (expressing Col3a1, Col1a1, Col1a2, [216]. These subtle differences might be explained by a Col6a3, and Meg3)[108] (Table 4 and Fig. 3b). Notably, number of technical factors such as muscle groups used the authors showed that a proportion of the OSR1+ for scRNAseq, tissue digestion and cell-type enrichment FAPs at 21 DPI diverge into the two populations ob- methods, single-cell RNA sequencing platform—al- served in undamaged muscle: DPP4+ FAPs and though most of them used Chromium 10× Genomics, CXCL14+ FAPs (Fig. 3b). Therefore, the gene expression the number of cells captured and recovered after se- of single-cell FAPs is highly diverse, representing a con- quencing, and downstream data processing (e.g., version tinuum state during skeletal muscle regeneration (Fig. of Seurat R package used) and interpretation. 3b). Owing to the high degree of FAPs diversity, we Oprescu et al. [108] reported that murine CXCL14+ speculate that FAP subpopulations have adapted to play FAPs (also expressing Smoc2, Ccl11, Gsn, and Dcn) and supportive and distinct roles during regeneration. These DPP4+ (expressing Pi16, Igfbp5, Igfbp6, Fbn1, and Ugdh) data also suggest that the transcriptional diversity of FAPs represent two different FAP subtypes present in PDGFRα+ FAPs at the single-cell level might reflect non-injured murine tibialis anterior muscle, suggesting their differential developmental potential. that FAPs could represent two distinct subpopulations In addition to these studies, Rubenstein et al. [109] de- of interstitial cells in resting conditions, as it was previ- scribed two human FAP subpopulations as LUMICAN ously shown by us [48] (Table 4). However, in response (LUM)+ FAP and FIBRILLIN 1 (FBN1)+ FAP subtypes to injury, the two populations follow a linear trajectory (Table 4 and Fig. 3a). Interestingly, both FAP subpopula- into a single population of activated FAPs (highly tions showed specific differences in the expression of Contreras et al. Skeletal Muscle (2021) 11:16 Page 19 of 25 Collagen types. The authors also validated the existence their fate and plasticity within muscles. Although of these two distinct FAP subtypes by using scRNAseq there has been encouraging progress in understanding of mouse quadriceps and diaphragm muscles [109]. FAP phenotypic variability and activities, future re- ECM gene expression was also consistent among mouse search should look to translating this knowledge into and human muscles, with COLLAGEN 1, COLLAGEN 3, efficient medical applications. and COLLAGEN 6 broadly expressed across FAP subpop- ulations. In contrast, LUM+ FAPs express COLLAGEN 4, Future perspectives and COLLAGEN 15 predominantly compared with Skeletal muscle requires a complex orchestra of special- FBN1+ FAPs (Table 4). Interestingly, the authors reported ized populations of cells to perform its crucial functions. differences in the gene expression of the precursor gene of The origin, behavioral activities, lineage potency, and ex- TIE2 protein among the two species, which was expressed pression of markers associated with stem or progenitor only in the FBN1+ FAP subtype found in mouse muscle cell states define these specialized cell types. Here, we but none of the FAP subtypes found in human muscles. have focused on the unappreciated role of PDGFRα+ The meaning of these subtle differences in gene expres- FAPs in muscle biology, health, structure, and regener- sion across mouse and human muscle FAP subtypes ation. Apart from the accepted structural part that the should be address in future research. connective tissue provides for proper muscle develop- De Micheli and collegues collected and integrated ~ ment, the complex cues and matrix that stromal cells 22,000 single-cell transcriptomes generating for the first produce are essential to sustain myogenesis and support time a consensus cell atlas of human skeletal muscles proper muscle morphogenesis. FAPs are implicated in [106] (Table 4 and Fig. 3a). The authors described three muscle scarring, disease, and pathology. Although sub- subpopulations of fibroblasts (likely FAPs) in which stantial progress has been made in understanding FAP COLLAGEN 1, SFRP4, SERPINE1 and CCL2 are highly behavior, they remain poorly characterized, and the rela- expressed by fibroblast 1; FBN1+, MFAP5, and CD55 are tionships with other stromal cells are not well under- expressed by fibroblast 2, whereas fibroblast 3 highly ex- stood. PDGFRα+ FAPs and their descendant lineages, presses SMOC2 [106, 107] (Table 4). including activated-fibroblasts/myofibroblasts, adipo- Recently, Farup and colleagues described 5 subpop- cytes, chondrogenic and osteogenic cells, modulate ulations of human muscle FAPs (Table 4 and Fig. 3a). muscle regeneration and repair. These plastic cells play However, by sub-setting the FAP population and re- broad roles as sentinels, stress sensors, immune regula- clustering, the number of clusters increased to 7. The tors, cellular hubs, and paracrine factories, which are still authors reported that the expression of THY1/CD90 under active research in multiple pathological settings. is enriched in cluster 4, whereas PDGFRA gene ex- As discussed above, lineage tracing technologies com- pression is broadly distributed among the FAP sub- bined with single-cell sequencing strategies should by- populations (Table 4). Remarkably, the CD90+ pass the significant limitations that historically subpopulation of FAPs is associated with increased prevented us from deconvolving the complexity of stro- fibro-fatty infiltration and seems to drive the muscle mal cell populations. The diverse fibroblast nomencla- degeneration found in obese and type-2 diabetes pa- ture has periodically led to confusing claims in muscle tients [16]. Although the composition of each cellular biology and ensuing turmoil in the literature. Thus, re- interstitial compartment changes dramatically after in- solving the regenerative vs. reparative dichotomy of jury and in disease settings in mice, there is no infor- muscle-resident mesenchymal progenitors, and distin- mation about how the FAP population behaves guishing true lineage heterogeneity from the diverse following injury or how degenerative diseases alter its functional states that these cells can dynamically and re- activities in humans. Nonetheless, we and others have versibly acquire remains a high-priority issue for the detected a diverse range of mesenchymal stromal cells field. Despite these various uncertainties, in this review, including quiescent subsets, which rapidly expand fol- we establish a baseline for the contribution of fibro- lowing injury and secrete cytokines modulating in- adipogenic progenitors to muscle development, homeo- flammation, trophic factors, and regenerative cues to stasis, regeneration, and repair. promote skeletal muscle maintenance, MuSC renewal, and regeneration. Conclusions In conclusion, further studies should focus on un- In this review, we document new insights about the vari- derstanding the mechanisms by which FAP cell het- ous properties of muscle-resident PDGFRα+ FAPs and erogeneity arises. We aim to understand the lineage discuss the current state of knowledge on their origins restriction of FAPs by gene regulatory networks and and lineage capabilities. Here we propose to define a cell epigenetic factors that, in combination with the ex- as FAP if they present the following characteristics: 1. trinsic effects of the spatial context could regulate Express PDGFRα at the gene and protein level. 2. It is Contreras et al. Skeletal Muscle (2021) 11:16 Page 20 of 25 located in the tissue's interstitium and behaves as a peri- Availability of data and materials All data generated or analyzed during this study are included in this vascular cell but not residing in the blood vessel cavity. published article. 3. It can form colonies in vitro. 4. Can differentiate into activated fibroblasts, adipocytes, chondrocytes, and oste- Declarations ocytes in vitro and in vivo. Ethics approval and consent to participate We illustrated their importance in maintaining proper Not applicable. muscle function and critical role during the onset and establishment of scarring in pathology and disease. Consent for publication Not applicable. Growing evidence shows that PDGFRα+ cells are hetero- geneous and act as signaling hubs by providing regenera- Competing interests tive cues and integrating these signals in the muscle The authors declare that they have no competing interests. niche. Thus, FAPs influence other populations of cells Author details within skeletal muscle and vice versa. Ultimately, by un- Developmental and Stem Cell Biology Division, Victor Chang Cardiac derstanding and manipulating the complexity and vari- Research Institute, Darlinghurst, NSW 2010, Australia. St. Vincent’s Clinical School, Faculty of Medicine, UNSW Sydney, Kensington 2052, Australia. ability of the stromal compartment, specifically the FAP Departamento de Biología Celular y Molecular and Center for Aging and lineage, we aim to develop novel therapeutics to treat Regeneration (CARE-ChileUC), Facultad de Ciencias Biológicas, Pontificia several scar-forming pathologies. It remains plausible to Universidad Católica de Chile, 8331150 Santiago, Chile. Biomedical Research Centre, Department of Medical Genetics and School of Biomedical foresee a future where clinical leaps could be made Engineering, University of British Columbia, Vancouver, BC V6T 1Z3, Canada. based on these cells and where severe muscle injury could be treated without prolonged myodegeneration Received: 22 January 2021 Accepted: 22 March 2021 and muscle malfunctioning. References Abbreviations 1. Joe AWB, et al. Muscle injury activates resident fibro/adipogenic progenitors aSMA: Alpha smooth muscle actin; BMP: Bone morphogenetic protein; that facilitate myogenesis. Nat Cell Biol. 2010;12(2):153–63 https://doi.org/1 CFU: Colony-forming unit; CDHs: Congenital diaphragmatic hernias; 0.1038/ncb2015. CT: Connective tissue; DMD: Duchenne muscular dystrophy; DTA: Diphteria 2. Uezumi A, et al. Mesenchymal progenitors distinct from satellite cells toxin A; DTR: Diphteria toxin receptor; ECM: Extracellular matrix; contribute to ectopic fat cell formation in skeletal muscle. Nat Cell Biol. FAPα: Fibroblast activation protein alpha; FAPs: Fibro-adipogenic progenitors; 2010;12(2):143–52 https://doi.org/10.1038/ncb2014. FOP: Fibrodysplasia ossificans progressiva; GSK: Glycogen Synthase Kinase-3; 3. Fiore D, et al. Pharmacological blockage of fibro/adipogenic progenitor HO: Heterotopic ossification; IL: Interleukin; IMAT: Intermuscular adipose expansion and suppression of regenerative fibrogenesis is associated with tissue; LGMD: Limb-girdle muscular dystrophy; Lox: Lysyl oxidase; impaired skeletal muscle regeneration. Stem Cell Res. 2016;17(1):161–9 MSCs: Mesenchymal stem cells; MCT: Muscle connective tissue; https://doi.org/10.1016/j.scr.2016.06.007. MuSCs: Muscle stem cells; NCCs: Neural crest cells; PDGFRα: Platelet-derived 4. Heredia JE, et al. Type 2 innate signals stimulate fibro/adipogenic growth factor receptor alpha; PDGFRβ: Platelet-derived growth factor progenitors to facilitate muscle regeneration. Cell. 2013;153(2):376–88 receptor beta; PDGF: Platelet-derived growth factor; PPFs: Pleuroperitoneal https://doi.org/10.1016/j.cell.2013.02.053. folds; OSR: Odd-skipped-related; Shh: Sonic Hedgehog signaling; Sca-1: Stem 5. Lukjanenko L, et al. Aging disrupts muscle stem cell function by impairing cell antigen-1; Tbx: T-box transcription factor; TIMP3: Tissue inhibitor of matricellular WISP1 Secretion from fibro-adipogenic progenitors. Cell Stem metalloproteinases 3; TCF21: Transcription factor 21; TGF-β: Transforming Cell United States. 2019;24(3):433–446.e7 https://doi.org/10.1016/j.stem.201 growth factor beta; UPS: Ubiquitin-proteasome system 8.12.014. 6. Murphy MM, et al. Satellite cells, connective tissue fibroblasts and their interactions are crucial for muscle regeneration. Development. 2011;138(17): Acknowledgements 3625–37 https://doi.org/10.1242/dev.064162. The authors acknowledge Yen Tran and Ralph Patrick for helpful suggestions 7. Wosczyna MN, et al. Mesenchymal Stromal Cells Are Required for to the single-cell omics chapter, and Lucas Rempel for his inputs improving Regeneration and Homeostatic Maintenance of Skeletal Muscle. Cell Rep. the review. Figures were created using Illustrator (Adobe Inc.) and Key- 2019;27(7):2029–2035.e5 https://doi.org/10.1016/j.celrep.2019.04.074. note (Apple Inc.) for macOS. 8. Roberts EW, et al. Depletion of stromal cells expressing fibroblast activation protein-α from skeletal muscle and bone marrow results in cachexia and anemia. J Exp Med. 2013;210(6):1137–51 https://doi.org/10.1084/jem.2 Authors’ contributions O.C. and M.T. drafted the review and figures. F.M.V.R. revised and reviewed 9. Uezumi A, et al. Mesenchymal Bmp3b expression maintains skeletal muscle the manuscript. All authors read and approved the final manuscript. integrity and decreases in age-related sarcopenia. J Clin Investig. 2021; 131(1) https://doi.org/10.1172/JCI139617. Funding 10. Santini MP, et al. Tissue-Resident PDGFRα+ Progenitor Cells Contribute to This work was supported by Comisión Nacional de Ciencia y Tecnología Fibrosis versus Healing in a Context- and Spatiotemporally Dependent CONICYT Beca Doctorado Nacional 2014 folio 21140378 “National Doctorate Manner. Cell Rep. 2020;30(2):555–570.e7 https://doi.org/10.1016/j.celrep.201 Fellowship”, by Centro Basal de Excelencia en Envejecimiento y 9.12.045. Regeneración (CONICYT-AFB 170005), and by the Victor Chang Cardiac 11. Contreras O, Cruz-Soca M, Theret M, Soliman H, Tung LW, Groppa E, et al. Research Institute to O.C.; by Fondation pour la Recherche Médicale (FRM, Cross-talk between TGF-β and PDGFRα signaling pathways regulates the 40248), by the European Molecular Biology Organization (EMBO, ALTF 115- fate of stromal fibro–adipogenic progenitors. J Cell Sci. 2019;132:jcs232157 2016), by the Association contre les myopathies (AFM, 22576), and by Mi- https://doi.org/10.1242/jcs.232157. chael Smith Foundation for Health Research (MSFHR, 18351) to M.T.; and by 12. Contreras O, Rossi FM, Brandan E. Adherent muscle connective tissue the Canadian Institutes of Health Research (CIHR-FDN-159908) to F.M.V.R. The fibroblasts are phenotypically and biochemically equivalent to stromal fibro/ funding agencies had no role in the design of the study, data collection and adipogenic progenitors. Matrix Biol Plus. 2019;2:100006 https://doi.org/10.1 analysis, the decision to publish, or preparation of the manuscript. 016/j.mbplus.2019.04.003. Contreras et al. Skeletal Muscle (2021) 11:16 Page 21 of 25 13. Eisner C, et al. Murine Tissue-Resident PDGFRα+ Fibro-Adipogenic 35. Merrell AJ, et al. Muscle connective tissue controls development of the Progenitors Spontaneously Acquire Osteogenic Phenotype in an Altered diaphragm and is a source of congenital diaphragmatic hernias. Nat Genet. Inflammatory Environment. J Bone Miner Res. 2020;35(8):1525–34 https:// 2015;47(5):496–504 https://doi.org/10.1038/ng.3250. doi.org/10.1002/jbmr.4020. 36. Hasson P, et al. Tbx4 and Tbx5 Acting in Connective Tissue Are Required for 14. Kopinke D, Roberson EC, Reiter JF. Ciliary Hedgehog Signaling Restricts Limb Muscle and Tendon Patterning. Dev Cell. 2010;18(1):148–56 https:// Injury-Induced Adipogenesis. Cell. 2017;170(2):340–351.e12 https://doi.org/1 doi.org/10.1016/j.devcel.2009.11.013. 0.1016/j.cell.2017.06.035. 37. Li QY, et al. Holt-Oram syndrome is caused by mutations in TBX5, a member of the Brachyury (T) gene family. Nat Genet. 1997:21–9 https://doi. 15. Uezumi A, et al. Fibrosis and adipogenesis originate from a common org/10.1038/ng0197-21. mesenchymal progenitor in skeletal muscle. J Cell Sci. 2011;124(21):3654–64 https://doi.org/10.1242/jcs.086629. 38. Kutchuk L, et al. Muscle composition is regulated by a lox-TGFβ feedback 16. Farup J, et al. Human skeletal muscle CD90+ fibro-adipogenic progenitors loop. Development (Cambridge). 2015; https://doi.org/10.1242/dev.113449. are associated with muscle degeneration in type 2 diabetic patients. 39. Besse L, et al. Individual Limb Muscle Bundles Are Formed through bioRxiv. 2020:2020.08.25.243907 https://doi.org/10.1101/2020.08.25.243907. Progressive Steps Orchestrated by Adjacent Connective Tissue Cells during 17. Chapman MA, et al. Three distinct cell populations express extracellular Primary Myogenesis. Cell Rep. 2020;30(10):3552–3565.e6 https://doi.org/10.1 matrix proteins and increase in number during skeletal muscle fibrosis. Am 016/j.celrep.2020.02.037. J Physiol Cell Physiol. 2016a;312(2):C131–43 https://doi.org/10.1152/ajpcell. 40. Colasanto MP, et al. Development of a subset of forelimb muscles and their 00226.2016. attachment sites requires the ulnar-mammary syndrome gene Tbx3. DMM 18. Chapman MA, Meza R, Lieber RL. Skeletal muscle fibroblasts in health and Dis Models Mech. 2016;9(11):1257–69 https://doi.org/10.1242/dmm.025874. disease. Differentiation. 2016b:108–15 https://doi.org/10.1016/j.diff.2016.05. 41. Duboule D, Dolle P. The structural and functional organization of the murine HOX gene family resembles that of Drosophila homeotic genes. 19. Gillies AR, Lieber RL. Structure and function of the skeletal muscle EMBO J. 1989;8(5):1497–505 https://doi.org/10.1002/j.1460-2075.1989.tb03 extracellular matrix. Muscle Nerve. 2011:318–31 https://doi.org/10.1002/ 534.x. mus.22094. 42. Boulet AM, Capecchi MR. Duplication of the Hoxd11 gene causes alterations 20. Helmbacher F, Stricker S. Tissue cross talks governing limb muscle in the axial and appendicular skeleton of the mouse. Dev Biol. 2002;249(1): development and regeneration. Semin Cell Dev Biol. 2020:14–30 https://doi. 96–107 https://doi.org/10.1006/dbio.2002.0755. org/10.1016/j.semcdb.2020.05.005. 43. Boulet AM, Capecchi MR. Multiple roles of Hoxa11 and Hoxd11 in the 21. Nassari S, Duprez D, Fournier-Thibault C. Non-myogenic contribution to formation of the mammalian forelimb zeugopod. Development. 2004; muscle development and homeostasis: The role of connective tissues. Front 131(2):299–309 https://doi.org/10.1242/dev.00936. Cell Dev Biol. 2017; https://doi.org/10.3389/fcell.2017.00022. 44. Davis AP, et al. Absence of radius and ulna in mice lacking hoxa-11 andhoxd-11. Nature. 1995;375(6534):791–5 https://doi.org/10.1038/375791a0. 22. Sefton EM, Kardon G. Connecting muscle development, birth defects, and evolution: An essential role for muscle connective tissue. Curr Top Dev Biol. 45. Iimura T, Pourquié O. Hox genes in time and space during vertebrate body 2019:137–76 https://doi.org/10.1016/bs.ctdb.2018.12.004. formation. Develop Growth Differ. 2007:265–75 https://doi.org/10.1111/j.144 23. Kardon G, Harfe BD, Tabin CJ. A Tcf4-positive mesodermal population 0-169X.2007.00928.x. provides a prepattern for vertebrate limb muscle patterning. Dev Cell. 2003; 46. Swinehart IT, et al. Hox11 genes are required for regional patterning and 5(6):937–44 https://doi.org/10.1016/S1534-5807(03)00360-5. integration of muscle, tendon and bone. Development (Cambridge). 2013; 24. Mathew SJ, et al. Connective tissue fibroblasts and Tcf4 regulate 140(22):4574–82 https://doi.org/10.1242/dev.096693. myogenesis. Development. 2011;138(2):371–84 https://doi.org/10.1242/dev. 47. Alfaro LAS, et al. CD34 promotes satellite cell motility and entry into 057463. proliferation to facilitate efficient skeletal muscle regeneration. Stem Cells. 25. Kardon G, Campbell JK, Tabin CJ. Local extrinsic signals determine muscle 2011;29(12):2030–41 https://doi.org/10.1002/stem.759. and endothelial cell fate and patterning in the vertebrate limb. Dev Cell. 48. Scott RW, et al. Hic1 Defines Quiescent Mesenchymal Progenitor 2002;3(4):533–45 https://doi.org/10.1016/S1534-5807(02)00291-5. Subpopulations with Distinct Functions and Fates in Skeletal Muscle 26. Vallecillo-García P, et al. Odd skipped-related 1 identifies a population of Regeneration. Cell Stem Cell. 2019;25(6):797–813.e9 https://doi.org/10.1016/j. embryonic fibro-adipogenic progenitors regulating myogenesis during limb stem.2019.11.004. development. Nat Commun. 2017;8(1) https://doi.org/10.1038/s41467-017- 49. Dellavalle A, et al. Pericytes resident in postnatal skeletal muscle 01120-3. differentiate into muscle fibres and generate satellite cells. Nat Commun. 27. Contreras O, Rebolledo DL, Oyarzún JE, Olguín HC, Brandan E. Connective 2011;2(1) https://doi.org/10.1038/ncomms1508. tissue cells expressing fibro/adipogenic progenitor markers increase under 50. Agley CC, et al. Human skeletal muscle fibroblasts, but not myogenic cells, chronic damage: relevance in fibroblast-myofibroblast differentiation and readily undergo adipogenic differentiation. J Cell Sci. 2013;126(24):5610–25 skeletal muscle fibrosis. Cell Tissue Res. 2016;364:647–60 https://doi.org/10.1 https://doi.org/10.1242/jcs.132563. 007/s00441-015-2343-0. 51. Arrighi N, et al. Characterization of adipocytes derived from fibro/ 28. Contreras O, Soliman H, Theret M, Rossi FMV, Brandan E. TGF-β-driven adipogenic progenitors resident in human skeletal muscle. Cell Death Dis. downregulation of the transcription factor TCF7L2 affects Wnt/β-catenin 2015;6(4) https://doi.org/10.1038/cddis.2015.79. signaling in PDGFRα+ fibroblasts. J Cell Sci. 2020;133 https://doi.org/1 52. Mackey AL, et al. Human skeletal muscle fibroblasts stimulate in vitro 0.1242/jcs.242297. myogenesis and in vivo muscle regeneration. J Physiol. 2017;595(15):5115– 29. Nowicki JL, Takimoto R, Burke AC. The lateral somitic frontier: Dorso-ventral 27 https://doi.org/10.1113/JP273997. aspects of anterio-posterior regionalization in avian embryos. Mech Dev. 53. Uezumi A, et al. Identification and characterization of PDGFR + 2003;120(2):227–40 https://doi.org/10.1016/S0925-4773(02)00415-X. mesenchymal progenitors in human skeletal muscle. Cell Death Dis. 2014a; 30. Pearse RV, et al. A cellular lineage analysis of the chick limb bud. Dev Biol. 5(4) https://doi.org/10.1038/cddis.2014.161. 2007;310(2):388–400 https://doi.org/10.1016/j.ydbio.2007.08.002. 54. Uezumi A, Ikemoto-Uezumi M, Tsuchida K. Roles of nonmyogenic 31. Noden DM. The embryonic origins of avian cephalic and cervical muscles mesenchymal progenitors in pathogenesis and regeneration of skeletal and associated connective tissues. Am J Anat. 1983a;168(3):257–76 https:// muscle. Front Physiol. 2014b; https://doi.org/10.3389/fphys.2014.00068. doi.org/10.1002/aja.1001680302. 55. Stricker S, et al. Comparative expression pattern of Odd-skipped related 32. Noden DM. The role of the neural crest in patterning of avian cranial genes Osr1 and Osr2 in chick embryonic development. Gene Expr Patterns. skeletal, connective, and muscle tissues. Dev Biol. 1983b;96(1):144–65 2006;6(8):826–34 https://doi.org/10.1016/j.modgep.2006.02.003. https://doi.org/10.1016/0012-1606(83)90318-4. 56. Stricker S, et al. Odd-skipped related genes regulate differentiation of 33. Olsson L, et al. Cranial neural crest cells contribute to connective tissue in embryonic limb mesenchyme and bone marrow mesenchymal stromal cranial muscles in the anuran amphibian, Bombina orientalis. Dev Biol. 2001; cells. Stem Cells Dev. 2012;21(4):623–33 https://doi.org/10.1089/scd.2011. 237(2):354–67 https://doi.org/10.1006/dbio.2001.0377. 0154. 34. Theis S, et al. The occipital lateral plate mesoderm is a novel source for 57. Dulauroy S, et al. Lineage tracing and genetic ablation of ADAM12 + vertebrate neck musculature. Development. 2010;137(17):2961–71 https:// perivascular cells identify a major source of profibrotic cells during acute doi.org/10.1242/dev.049726. tissue injury. Nat Med. 2012;18(8):1262–70 https://doi.org/10.1038/nm.2848. Contreras et al. Skeletal Muscle (2021) 11:16 Page 22 of 25 58. Giordani L, et al. High-Dimensional Single-Cell Cartography Reveals Novel 80. Sambasivan R, et al. Pax7-expressing satellite cells are indispensable for Skeletal Muscle-Resident Cell Populations. Mol Cell. 2019;74(3):609–621.e6 adult skeletal muscle regeneration. Development. 2011;138(17):3647–56 https://doi.org/10.1016/j.molcel.2019.02.026. https://doi.org/10.1242/dev.067587. 59. Lemos DR, et al. Nilotinib reduces muscle fibrosis in chronic muscle injury 81. Iismaa, S. E. et al. (2018) Comparative regenerative mechanisms across by promoting TNF-mediated apoptosis of fibro/adipogenic progenitors. Nat different mammalian tissues, npj Regenerative Medicine. doi: https://doi. Med. 2015;21(7):786–94 https://doi.org/10.1038/nm.3869. org/10.1038/s41536-018-0044-5. 60. Lees-Shepard JB, et al. Palovarotene reduces heterotopic ossification in 82. Sandoval-Guzmán T, Currie JD. The journey of cells through regeneration. juvenile fop mice but exhibits pronounced skeletal toxicity. eLife. 2018;7 Curr Opin Cell Biol. 2018:36–41 https://doi.org/10.1016/j.ceb.2018.05.008. https://doi.org/10.7554/eLife.40814. 83. Wells JM, Watt FM. Diverse mechanisms for endogenous regeneration and 61. Madaro L, et al. Denervation-activated STAT3–IL-6 signalling in fibro- repair in mammalian organs. Nature. 2018:322–8 https://doi.org/10.1038/s41 adipogenic progenitors promotes myofibres atrophy and fibrosis. Nat Cell 586-018-0073-7. Biol. 2018;20(8):917–27 https://doi.org/10.1038/s41556-018-0151-y. 84. Cordero-Espinoza L, Huch M. The balancing act of the liver: tissue 62. Malecova B, et al. Dynamics of cellular states of fibro-adipogenic regeneration versus fibrosis. J Clin Investig. 2018:85–96 https://doi.org/10.11 progenitors during myogenesis and muscular dystrophy. Nat Commun. 72/JCI93562. 2018;9(1) https://doi.org/10.1038/s41467-018-06068-6. 85. Dadgar S, et al. Asynchronous remodeling is a driver of failed regeneration 63. Petrilli LL, et al. High-Dimensional Single-Cell Quantitative Profiling of in Duchenne muscular dystrophy. J Cell Biol. 2014;207(1):139–58 https://doi. Skeletal Muscle Cell Population Dynamics during Regeneration. Cells. 2020; org/10.1083/jcb.201402079. 9(7) https://doi.org/10.3390/cells9071723. 86. Pessina P, et al. Novel and optimized strategies for inducing fibrosis in vivo: 64. Uezumi A, et al. Cell-Surface Protein Profiling Identifies Distinctive Markers Focus on Duchenne Muscular Dystrophy. Skelet Muscle. 2014;4(1) https:// of Progenitor Cells in Human Skeletal Muscle. Stem Cell Rep. 2016;7(2):263– doi.org/10.1186/2044-5040-4-7. 78 https://doi.org/10.1016/j.stemcr.2016.07.004. 87. Bendall JR. The elastin content of various muscles of beef animals. J Sci 65. Yao L, et al. Gli1 defines a subset of fibroadipogenic progenitors that Food Agric. 1967;18(12):553–8 https://doi.org/10.1002/jsfa.2740181201. promote skeletal muscle regeneration with less fat accumulation. J Bone 88. Dransfield E. Intramuscular composition and texture of beef muscles. J Sci Miner Res. 2021; https://doi.org/10.1002/jbmr.4265. Food Agric. 1977;28(9):833–42 https://doi.org/10.1002/jsfa.2740280910. 66. Giuliani G, Vumbaca S, Fuoco C, Gargioli C, Giorda E, Massacci G, et al. 89. Vognarová I, Dvorák Z, Böhm R. Collagen and Elastin in Different Cuts of SCA-1 micro-heterogeneity in the fate decision of dystrophic fibro/ Veal and Beef. J Food Sci. 1968;33(4):339–43 https://doi.org/10.1111/j.1365-2 adipogenic progenitors. Cell Death Dis. 2021;12 https://doi.org/10.1038/ 621.1968.tb03626.x. s41419-021-03408-1. 90. Light N, Champion AE. Characterization of muscle epimysium, perimysium 67. Stumm J, et al. Odd skipped-related 1 (Osr1) identifies muscle-interstitial and endomysium collagens. Biochem J. 1984;219(3):1017–26 https://doi. fibro-adipogenic progenitors (FAPs) activated by acute injury. Stem Cell Res. org/10.1042/bj2191017. 2018;32:8–16 https://doi.org/10.1016/j.scr.2018.08.010. 91. Purslow PP. The Structure and Role of Intramuscular Connective Tissue 68. Kramann R, et al. Perivascular Gli1+ progenitors are key contributors to in Muscle Function. Front Physiol. 2020; https://doi.org/10.3389/fphys.2 injury-induced organ fibrosis. Cell Stem Cell. 2015;16(1):51–66 https://doi. 020.00495. org/10.1016/j.stem.2014.11.004. 92. Borg TK, Caulfield JB. Morphology of connective tissue in skeletal 69. Zhao L, et al. Retinoic acid signalling in fibro/adipogenic progenitors muscle. Tissue Cell. 1980;12(1):197–207 https://doi.org/10.1016/0040-81 robustly enhances muscle regeneration. EBioMedicine. 2020;60 https://doi. 66(80)90061-0. org/10.1016/j.ebiom.2020.103020. 93. Carnwath JW, Shotton DM. Muscular dystrophy in the mdx mouse: 70. Xi H, Langerman J, Sabri S, Chien P, Young CS, Younesi S, et al. A human Histopathology of the soleus and extensor digitorum longus muscles. skeletal muscle atlas identifies the trajectories of stem and progenitor cells J Neurol Sci. 1987;80(1):39–54 https://doi.org/10.1016/0022-510X(87)90219-X. across development and from human pluripotent stem cells. Cell Stem Cell. 94. Duance VC, et al. A role for collagen in the pathogenesis of muscular 2020;27(1):158–76.e10 https://doi.org/10.1016/j.stem.2020.04.017. dystrophy? Nature. 1980;284(5755):470–2 https://doi.org/10.1038/284470a0. 71. Goloviznina NA, et al. Prospective isolation of human fibroadipogenic 95. Gatchalian CL, Schachner M, Sanes JR. Fibroblasts that proliferate near progenitors with CD73. Heliyon. 2020;6(7) https://doi.org/10.1016/j.heliyon.2 denervated synaptic sites in skeletal muscle synthesize the adhesive 020.e04503. molecules tenascin(J1), N-CAM, fibronectin, and a heparan sulfate 72. Kasai T, et al. Promethazine Hydrochloride Inhibits Ectopic Fat Cell proteoglycan. J Cell Biol. 1989;108(5):1873–90 https://doi.org/10.1083/jcb.1 Formation in Skeletal Muscle. Am J Pathol. 2017;187(12):2627–34 https://doi. 08.5.1873. org/10.1016/j.ajpath.2017.08.008. 96. Klingler W, et al. The role of fibrosis in Duchenne muscular dystrophy. Acta 73. Soliman H, et al. Pathogenic Potential of Hic1-Expressing Cardiac Stromal Myologica. 2012;31(3):184–95. Progenitors. Cell Stem Cell. 2020;26(2):205–220.e8 https://doi.org/10.1016/j. 97. Lieber RL, Ward SR. Cellular mechanisms of tissue fibrosis. 4. structural and stem.2019.12.008. functional consequences of skeletal muscle fibrosis. Am J Physiol Cell 74. Chong JJH, et al. Adult cardiac-resident MSC-like stem cells with a Physiol. 2013;305(3) https://doi.org/10.1152/ajpcell.00173.2013. proepicardial origin. Cell Stem Cell. 2011;9(6):527–40 https://doi.org/10.1016/ 98. Morrison J, et al. T-cell-dependent fibrosis in the mdx dystrophic mouse. j.stem.2011.10.002. Lab Investig. 2000;80(6):881–91 https://doi.org/10.1038/labinvest.3780092. 75. Noseda M, et al. PDGFRα demarcates the cardiogenic clonogenic Sca1+ 99. Serrano AL, et al. Cellular and molecular mechanisms regulating fibrosis in stem/progenitor cell in adult murine myocardium. Nat Commun. 2015;6 skeletal muscle repair and disease. Curr Top Dev Biol. 2011; https://doi.org/1 https://doi.org/10.1038/ncomms7930. 0.1016/B978-0-12-385940-2.00007-3. 76. Chen WY, et al. Heterozygous disruption of Hic1 predisposes mice to a 100. Williams PE, Goldspink G. Connective tissue changes in immobilised muscle. gender-dependent spectrum of malignant tumors. Nat Genet. 2003;33(2): J Anat. 1984;138(Pt 2):343–50 Available at: http://www.ncbi.nlm.nih.gov/ 197–202 https://doi.org/10.1038/ng1077. pubmed/6715254%0A, http://www.pubmedcentral.nih.gov/articlerender. 77. Van Rechem C, et al. Differential Regulation of HIC1 Target Genes by CtBP fcgi?artid=PMC1164074. and NuRD, via an Acetylation/SUMOylation Switch, in Quiescent versus 101. Jackson WM, et al. Mesenchymal progenitor cells derived from traumatized Proliferating Cells. Mol Cell Biol. 2010;30(16):4045–59 https://doi.org/10.1128/ human muscle. J Tissue Eng Regen Med. 2009;3(2):129–38 https://doi.org/1 mcb.00582-09. 0.1002/term.149. 78. Lepper C, Partridge TA, Fan CM. An absolute requirement for pax7-positive 102. Dammone G, et al. PPARγ controls ectopic adipogenesis and cross-talks satellite cells in acute injury-induced skeletal muscle regeneration. with myogenesis during skeletal muscle regeneration. Int J Mol Sci. 2018; Development. 2011;138(17):3639–46 https://doi.org/10.1242/dev.067595. 19(7) https://doi.org/10.3390/ijms19072044. 79. Relaix F, Zammit PS. Satellite cells are essential for skeletal muscle 103. Gonzalez D, et al. ALS skeletal muscle shows enhanced TGF-β signaling, regeneration: The cell on the edge returns centre stage. Development fibrosis and induction of fibro/adipogenic progenitor markers. PLoS One. (Cambridge). 2012;139(16):2845–56 https://doi.org/10.1242/dev.069088. 2017;12(5) https://doi.org/10.1371/journal.pone.0177649. Contreras et al. Skeletal Muscle (2021) 11:16 Page 23 of 25 104. Hogarth MW, et al. Fibroadipogenic progenitors are responsible for muscle 126. Györfi AH, Matei AE, Distler JHW. Targeting TGF-β signaling for the loss in limb girdle muscular dystrophy 2B. Nat Commun. 2019;10(1) https:// treatment of fibrosis. Matrix Biol. 2018:8–27 https://doi.org/10.1016/j.ma doi.org/10.1038/s41467-019-10438-z. tbio.2017.12.016. 105. Lukjanenko L, et al. Loss of fibronectin from the aged stem cell niche affects 127. Hinz B. Myofibroblasts. Exp Eye Res. 2015:56–70 https://doi.org/10.1016/j. the regenerative capacity of skeletal muscle in mice. Nat Med. 2016;22(8): exer.2015.07.009. 897–905 https://doi.org/10.1038/nm.4126. 128. Klingberg F, et al. Prestress in the extracellular matrix sensitizes latent TGF- 106. De Micheli AJ, Spector JA, et al. A reference single-cell transcriptomic atlas β1 for activation. J Cell Biol. 2014;207(2):283–97 https://doi.org/10.1083/jcb.2 of human skeletal muscle tissue reveals bifurcated muscle stem cell populations. Skelet Muscle. 2020a;10(1) https://doi.org/10.1186/s13395-020- 129. Reed NI, et al. The αvβ1integrinplays acriticalinvivoroleintissue 00236-3. fibrosis. Sci Transl Med. 2015;7(288) https://doi.org/10.1126/scitra 107. De Micheli AJ, Laurilliard EJ, et al. Single-cell analysis of the muscle stem cell nslmed.aaa5094. hierarchy identifies heterotypic communication signals involved in skeletal 130. Wang XF, et al. Expression cloning and characterization of the TGF-β type III muscle regeneration. Cell Rep. 2020b;30(10):3583–3595.e5 https://doi.org/1 receptor. Cell. 1991; https://doi.org/10.1016/0092-8674(91)90074-9. 0.1016/j.celrep.2020.02.067. 131. López-Casillas F, et al. Betaglycan can act as a dual modulator of TGF-β 108. Oprescu SN, et al. Temporal Dynamics and Heterogeneity of Cell access to signaling receptors: Mapping of ligand binding and GAG Populations during Skeletal Muscle Regeneration. iScience. 2020;23(4) attachment sites. J Cell Biol. 1994; https://doi.org/10.1083/jcb.124.4.557. https://doi.org/10.1016/j.isci.2020.100993. 132. Lewis KA, et al. Betaglycan binds inhibin and can mediate functional 109. Rubenstein AB, et al. Single-cell transcriptional profiles in human antagonism of activin signalling. Nature. 2000; https://doi.org/10.1038/3 skeletal muscle. Sci Rep. 2020;10(1) https://doi.org/10.1038/s41598-019- 57110-6. 133. Wiater E, Vale W. Inhibin is an antagonist of bone morphogenetic protein 110. Riquelme-Guzmán C, Contreras O. Single-cell revolution unveils the signaling. J Biol Chem. 2003; https://doi.org/10.1074/jbc.M209710200. mysteries of the regenerative mammalian digit tip. Dev Biol. 2020;461:107–9 134. Velasco-Loyden G, Arribas J, López-Casillas F. The Shedding of Betaglycan Is https://doi.org/10.1016/j.ydbio.2020.02.002. Regulated by Pervanadate and Mediated by Membrane Type Matrix 111. Marinkovic M, et al. Fibro-adipogenic progenitors of dystrophic mice are Metalloprotease-1. J Biol Chem. 2004; https://doi.org/10.1074/jbc.M3064992 insensitive to NOTCH regulation of adipogenesis. Life Sci Alliance. 2019;2(3) 00. https://doi.org/10.26508/lsa.201900437. 135. Vilchis-Landeros MM, et al. Recombinant soluble betaglycan is a potent and 112. Ciuffreda MC, et al. Protocols for in vitro differentiation of human isoform-selective transforming growth factor-β neutralizing agent. Biochem mesenchymal stem cells into osteogenic, chondrogenic and adipogenic J. 2001; https://doi.org/10.1042/0264-6021:3550215. lineages. Methods Mol Biol. 2016:149–58 https://doi.org/10.1007/978-1-493 136. Bilandzic M, et al. Betaglycan blocks metastatic behaviors in human 9-3584-0_8. granulosa cell tumors by suppressing NFκB-mediated induction of MMP2. 113. Sacchetti B, et al. No identical “mesenchymal stem cells” at different Cancer Lett. 2014; https://doi.org/10.1016/j.canlet.2014.07.039. times and sites: Human committed progenitors of distinct origin and 137. Shi Y, Massagué J. Mechanisms of TGF-β signaling from cell membrane to differentiation potential are incorporated as adventitial cells in the nucleus. Cell. 2003:685–700 https://doi.org/10.1016/S0092-8674(03 microvessels. Stem Cell Rep. 2016;6(6):897–913 https://doi.org/10.1016/j. )00432-X. stemcr.2016.05.011. 138. Imamura T, et al. Smad6 inhibits signalling by the TGF-β superfamily. 114. Mierzejewski B, et al. Mouse CD146+ muscle interstitial progenitor cells Nature. 1997; https://doi.org/10.1038/39355. differ from satellite cells and present myogenic potential. Stem Cell Res 139. Jung SM, et al. Smad6 inhibits non-canonical TGF-β1 signalling by recruiting Ther. 2020;11(1) https://doi.org/10.1186/s13287-020-01827-z. the deubiquitinase A20 to TRAF6. Nat Commun. 2013; https://doi.org/10.103 115. Swift J, et al. Nuclear lamin-A scales with tissue stiffness and enhances 8/ncomms3562. matrix-directed differentiation. Science. 2013;341(6149) https://doi.org/1 140. Nakao A, et al. Identification of Smad7, a TGFβ-inducible antagonist of TGF- 0.1126/science.1240104. β signalling. Nature. 1997; https://doi.org/10.1038/39369. 116. Lee C, et al. Rotator Cuff Fibro-Adipogenic Progenitors Demonstrate Highest 141. Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways Concentration, Proliferative Capacity, and Adipogenic Potential Across in TGF-β family signalling. Nature. 2003:577–84 https://doi.org/10.1038/na Muscle Groups. J Orthop Res. 2020;38(5):1113–21 https://doi.org/10.1002/ ture02006. jor.24550. 142. Khalil H, et al. Fibroblast-specific TGF-β-Smad2/3 signaling underlies cardiac 117. Mázala DAG, et al. TGF-β-driven muscle degeneration and failed fibrosis, in. J Clin Investig. 2017:3770–83 https://doi.org/10.1172/JCI94753. regeneration underlie disease onset in a DMD mouse model. JCI Insight. 143. Kim KK, Sheppard D, Chapman HA. TGF-β1 signaling and tissue fibrosis. 2020;5(6) https://doi.org/10.1172/jci.insight.135703. Cold Spring Harb Perspect Biol. 2018;10(4) https://doi.org/10.1101/ 118. Rodgers JT, et al. HGFA Is an Injury-Regulated Systemic Factor that Induces cshperspect.a022293. the Transition of Stem Cells into GAlert. Cell Rep. 2017;19(3):479–86 https:// 144. Juban G, et al. AMPK Activation Regulates LTBP4-Dependent TGF-β1 doi.org/10.1016/j.celrep.2017.03.066. Secretion by Pro-inflammatory Macrophages and Controls Fibrosis in 119. Burks TN, Cohn RD. Role of TGF-β signaling in inherited and acquired Duchenne Muscular Dystrophy. Cell Rep. 2018;25(8):2163–2176.e6 https:// myopathies. Skelet Muscle. 2011;1(1):19 https://doi.org/10.1186/2044-504 doi.org/10.1016/j.celrep.2018.10.077. 0-1-19. 145. McLennan IS, Koishi K. Cellular localisation of transforming growth factor- 120. Lodyga M, Hinz B. TGF-β1 - A truly transforming growth factor in fibrosis beta 2 and -beta 3 (TGF-β2, TGF-β3) in damaged and regenerating skeletal and immunity. Semin Cell Dev Biol. 2020:123–39 https://doi.org/10.1016/j. muscles. Dev Dyn. 1997;208(2):278–89 https://doi.org/10.1002/(sici)1097-01 semcdb.2019.12.010. 77(199702)208:2<278::aid-aja14>3.0.co;2-%23. 121. Budi EH, Duan D, Derynck R. Transforming Growth Factor-β Receptors and 146. Theret, Marine, Marcela Low, Lucas Rempel, Fang Fang Li, Lin Wei Tung, Smads: Regulatory Complexity and Functional Versatility. Trends Cell Biol. Osvaldo Contreras, Chih-Kai Chang, Andrew Wu, Hesham Soliman, and 2017:658–72 https://doi.org/10.1016/j.tcb.2017.04.005. Fabio M V Rossi. “In Vitro Assessment of Anti-Fibrotic Drug Activity Does 122. Wu MY, Hill CS. TGF-β Superfamily Signaling in Embryonic Development Not Predict in Vivo Efficacy in Murine Models of Duchenne Muscular and Homeostasis. Dev Cell. 2009:329–43 https://doi.org/10.1016/j.devcel.2 Dystrophy.” Life Sciences 279 (2021):119482. https://doi.org/10.1016/j.lfs.2 009.02.012. 021.119482. 123. David CJ, Massagué J. Contextual determinants of TGFβ action in 147. Davies MR, et al. TGF-β small molecule inhibitor sb431542 reduces rotator development, immunity and cancer. Nat Rev Mol Cell Biol. 2018:419–35 cuff muscle fibrosis and fatty infiltration by promoting fibro/ adipogenic https://doi.org/10.1038/s41580-018-0007-0. progenitor apoptosis. PLoS One. 2016;11(5) https://doi.org/10.1371/journal. 124. Derynck R, Budi EH. Specificity, versatility, and control of TGF-b family pone.0155486. signaling. Sci Signal. 2019; https://doi.org/10.1126/scisignal.aav5183. 148. Saito Y, Chikenji TS, Matsumura T, Nakano M, Fujimiya M. Exercise enhances 125. Hinz B, et al. The myofibroblast: One function, multiple origins. Am J Pathol. skeletal muscle regeneration by promoting senescence in fibro-adipogenic 2007;170(6):1807–16 https://doi.org/10.2353/ajpath.2007.070112. progenitors. Nat Commun. 2020;11:889 https://doi.org/10.1038/s41467-02 0-14734-x. Contreras et al. Skeletal Muscle (2021) 11:16 Page 24 of 25 149. Reggio A, Spada F, Rosina M, Massacci G, Zuccotti A, Fuoco C, et al. The 172. Gronwald RGK, et al. Cloning and expression of a cDNA coding for the immunosuppressant drug azathioprine restrains adipogenesis of muscle human platelet-derived growth factor receptor: Evidence for more than one Fibro/Adipogenic Progenitors from dystrophic mice by affecting AKT receptor class. Proc Natl Acad Sci U S A. 1988;85(10):3435–9 https://doi. signaling. Sci Rep. 2019;9 https://doi.org/10.1038/s41598-019-39538-y. org/10.1073/pnas.85.10.3435. 150. Dong Y, Augusto K, Silva S, Dong Y, Zhang L. Glucocorticoids increase 173. Matsui T, et al. Isolation of a novel receptor cDNA establishes the existence adipocytes in muscle by affecting IL-4 regulated FAP activity. FASEB J. 2014; of two PDGF receptor genes. Science. 1989;243(4892):800–4 https://doi. 28(9):4123–32 https://doi.org/10.1096/fj.14-254011. org/10.1126/science.2536956. 151. Mozzetta C, et al. Fibroadipogenic progenitors mediate the ability of HDAC 174. Yarden Y, et al. Structure of the receptor for platelet-derived growth factor inhibitors to promote regeneration in dystrophic muscles of young, but not helps define a family of closely related growth factor receptors. Nature. old Mdx mice. EMBO Mol Med. 2013;5(4):626–39 https://doi.org/10.1002/ 1986;323(6085):226–32 https://doi.org/10.1038/323226a0. emmm.201202096. 175. Heldin C-H, Westermark B. Role of Platelet-Derived Growth Factor in Vivo. 152. Liu X, et al. Trichostatin A regulates fibro/adipogenic progenitor Mol Cell Biol Wound Repair. 1988:249–73 https://doi.org/10.1007/978-1-4 adipogenesis epigenetically and reduces rotator cuff muscle fatty 899-0185-9_7. infiltration. J Orthop Res. 2020; https://doi.org/10.1002/jor.24865. 176. Kohler N, Lipton A. Platelets as a source of fibroblast growth-promoting 153. Reggio A, et al. Adipogenesis of skeletal muscle fibro/adipogenic activity. Exp Cell Res. 1974;87(2):297–301 https://doi.org/10.1016/0014-482 progenitors is affected by the WNT5a/GSK3/β-catenin axis. Cell Death Differ. 7(74)90484-4. 2020;27(10):2921–41 https://doi.org/10.1038/s41418-020-0551-y. 177. Hoch RV, Soriano P. Roles of PDGF in animal development. Development. 154. Cordani N, et al. Nitric oxide controls fat deposition in dystrophic skeletal 2003;130(20):4769–84 https://doi.org/10.1242/dev.00721. muscle by regulating fibro-adipogenic precursor differentiation. Stem Cells. 178. Heldin CH, Westermark B. Mechanism of action and in vivo role of platelet- 2014;32(4):874–85 https://doi.org/10.1002/stem.1587. derived growth factor, physiological reviews. Am Physiol Soc. 1999;79(4): 155. Ieronimakis N, et al. PDGFRα signalling promotes fibrogenic responses in 1283–316 https://doi.org/10.1152/physrev.1999.79.4.1283. collagen-producing cells in Duchenne muscular dystrophy. J Pathol. 2016; 179. Roskoski R. The role of small molecule platelet-derived growth factor 240(4):410–24 https://doi.org/10.1002/path.4801. receptor (PDGFR) inhibitors in the treatment of neoplastic disorders. 156. Astudillo P. Extracellular matrix stiffness and Wnt/β-catenin signaling in Pharmacol Res. 2018:65–83 https://doi.org/10.1016/j.phrs.2018.01.021. physiology and disease. Biochem Soc Trans. 2020:1187–98 https://doi.org/1 180. Kazlauskas A. PDGFs and their receptors. Gene. 2017:1–7 https://doi.org/10.1 0.1042/BST20200026. 016/j.gene.2017.03.003. 157. Nusse R, Clevers H. Wnt/β-Catenin Signaling. Dis Emerg Ther Modalities Cell. 181. Kanakaraj P, et al. Ligand-Induced Interaction between α- and β-Type 2017:985–99 https://doi.org/10.1016/j.cell.2017.05.016. Platelet-Derived Growth Factor (PDGF) Receptors: Role of Receptor 158. Aberle H, et al. Β-Catenin Is a Target for the Ubiquitin-Proteasome Pathway. Heterodimers in Kinase Activation. Biochemistry. 1991;30(7):1761–7 https:// EMBO J. 1997;16(13):3797–804 https://doi.org/10.1093/emboj/16.13.3797. doi.org/10.1021/bi00221a005. 159. Valenta T, Hausmann G, Basler K. The many faces and functions of β- 182. Kelly JD, et al. Platelet-derived growth factor (PDGF) stimulates PDGF catenin. EMBO J. 2012:2714–36 https://doi.org/10.1038/emboj.2012.150. receptor subunit dimerization and intersubunit trans-phosphorylation. 160. Cadigan KM, Waterman ML. TCF/LEFs and Wnt signaling in the nucleus. J Biol Chem. 1991;266(14):8987–92 https://doi.org/10.1016/s0021-9258(1 Cold Spring Harb Perspect Biol. 2012;4(11) https://doi.org/10.1101/ 8)31541-2. cshperspect.a007906. 183. Wang Y, et al. Platelet-derived Growth Factor Receptor-mediated Signal Transduction from Endosomes. J Biol Chem. 2004;279(9):8038–46 https://doi. 161. Clevers H. Wnt/β-Catenin Signaling in Development and Disease. Cell. 2006: org/10.1074/jbc.M311494200. 469–80 https://doi.org/10.1016/j.cell.2006.10.018. 162. Trensz F, et al. A muscle resident cell population promotes fibrosis in 184. Mueller AA, et al. Intronic polyadenylation of PDGFRα in resident stem cells hindlimb skeletal muscles of mdx mice through the Wnt canonical attenuates muscle fibrosis. Nature England. 2016;540(7632):276–9 https://doi. pathway. Am J Physiol Cell Physiol. 2010;299(5) https://doi.org/10.1152/a org/10.1038/nature20160. jpcell.00253.2010. 185. Sugg KB, et al. Inhibition of platelet-derived growth factor signaling 163. Biressi S, et al. A Wnt-TGF2 axis induces a fibrogenic program in muscle prevents muscle fiber growth during skeletal muscle hypertrophy. FEBS stem cells from dystrophic mice. Sci Transl Med. 2014;6(267) https://doi. Letters. 2017;591(5):801–9 https://doi.org/10.1002/1873-3468.12571. John org/10.1126/scitranslmed.3008411. Wiley & Sons, Ltd. 164. Xiang FL, Fang M, Yutzey KE. Loss of β-catenin in resident cardiac fibroblasts 186. Asli NS, et al. PDGFRα signaling in cardiac fibroblasts modulates quiescence, attenuates fibrosis induced by pressure overload in mice. Nat Commun. metabolism and self-renewal, and promotes anatomical and functional 2017;8(1):712 https://doi.org/10.1038/s41467-017-00840-w. repair. bioRxiv. 2019a:225979 https://doi.org/10.1101/225979. 187. Ivey MJ, et al. Platelet-derived growth factor receptor-α is essential for 165. Akhmetshina, A. et al. (2012) Activation of canonical Wnt signalling is cardiac fibroblast survival. Am J Physiol Heart Circ Physiol. 2019;317(2):H330– required for TGF-β-mediated fibrosis, Nature Communications, 3. doi: 44 https://doi.org/10.1152/ajpheart.00054.2019. https://doi.org/10.1038/ncomms1734. 166. Hamburg-Shields E, et al. Sustained β-catenin activity in dermal fibroblasts 188. Harvey T, Flamenco S, Fan CM. A Tppp3 + Pdgfra + tendon stem cell promotes fibrosis by up-regulating expression of extracellular matrix population contributes to regeneration and reveals a shared role for PDGF protein-coding genes. J Pathol. 2015;235(5):686–97 https://doi.org/10.1002/ signalling in regeneration and fibrosis. Nat Cell Biol. 2019;21(12):1490–503 path.4481. https://doi.org/10.1038/s41556-019-0417-z. 167. Działo E, Tkacz K, Błyszczuk P. Crosstalk between the TGF-β and WNT 189. Liu W, et al. Intramuscular adipose is derived from a non-Pax3 lineage and signalling pathways during cardiac fibrogenesis. Acta Biochim Pol. 2018; required for efficient regeneration of skeletal muscles. Dev Biol. 2012;361(1): 65(3):341–9. https://doi.org/10.18388/abp.2018_2635. 27–38 https://doi.org/10.1016/j.ydbio.2011.10.011. 190. Tchoukalova YD, Sarr MG, Jensen MD. Measuring committed preadipocytes 168. Girardi F, Le Grand F. Wnt Signaling in Skeletal Muscle Development and in human adipose tissue from severely obese patients by using adipocyte Regeneration. Prog Mol Biol Transl Sci. 2018:157–79 https://doi.org/10.1016/ fatty acid binding protein. Am J Physiol Regul Integr Comp Physiol. 2004; bs.pmbts.2017.11.026. 287(5 56-5) https://doi.org/10.1152/ajpregu.00337.2004. 169. Piersma B, Bank RA, Boersema M. Signaling in fibrosis: TGF-β, WNT, and YAP/TAZ converge. Front Med. 2015; https://doi.org/10.3389/fmed.2015. 191. Biltz NK, Meyer GA. A novel method for the quantification of fatty 00059. infiltration in skeletal muscle. Skelet Muscle. 2017;7(1) https://doi.org/10.11 170. Lindahl P, et al. Pericyte loss and microaneurysm formation in PDGF-B- 86/s13395-016-0118-2. deficient mice. Science. 1997;277(5323):242–5 https://doi.org/10.1126/ 192. Pisani DF, et al. Mouse model of skeletal muscle adiposity: A glycerol science.277.5323.242. treatment approach. Biochem Biophys Res Commun. 2010;396(3):767–73 https://doi.org/10.1016/j.bbrc.2010.05.021. 171. Claesson-Welsh L, et al. cDNA cloning and expression of the human A-type 193. Shabalina IG, et al. UCP1 in Brite/Beige adipose tissue mitochondria is platelet-derived growth factor (PDGF) receptor establishes structural functionally thermogenic. Cell Rep. 2013;5(5):1196–203 https://doi.org/10.1 similarity to the B-type PDGF receptor. Proc Natl Acad Sci U S A. 1989; 016/j.celrep.2013.10.044. 86(13):4917–21 https://doi.org/10.1073/pnas.86.13.4917. Contreras et al. Skeletal Muscle (2021) 11:16 Page 25 of 25 194. Gorski T, Mathes S, Krützfeldt J. Uncoupling protein 1 expression in 216. Pawlikowski, B. et al. (2019) A cellular atlas of skeletal muscle regeneration adipocytes derived from skeletal muscle fibro/adipogenic progenitors is and aging, bioRxiv, p. 635805. doi: https://doi.org/10.1101/635805. under genetic and hormonal control. J Cachexia Sarcopenia Muscle. 2018; 217. Bentzinger CF, et al. Cellular dynamics in the muscle satellite cell niche. 9(2):384–99 https://doi.org/10.1002/jcsm.12277. EMBO Rep. 2013:1062–72 https://doi.org/10.1038/embor.2013.182. 195. Saely CH, Geiger K, Drexel H. Brown versus white adipose tissue: A mini- 218. Chiristov C, Chrétien F, Abou-Khalil R, Bassez G, Vallet G, Authier FJ, et al. review. Gerontology. 2011:15–23 https://doi.org/10.1159/000321319. Muscle satellite cells and endothelial cells: Close neighbors and privileged partners. Mol Biol Cell. 2007;18:1397–409 https://doi.org/10.1091/mbc.E06- 196. Vishvanath L, Gupta RK. Contribution of adipogenesis to healthy adipose 08-0693. tissue expansion in obesity. J Clin Invest. 2019:4022–31 https://doi.org/10.11 219. Latroche C, et al. Skeletal muscle microvasculature: A highly dynamic 72/JCI129191. lifeline. Physiology. 2015:417–27 https://doi.org/10.1152/physiol.00026.2015. 197. Schulz TJ, Tseng YH. Brown adipose tissue: Development, metabolism and 220. Bonfanti C, et al. PW1/Peg3 expression regulates key properties that beyond. Biochem J. 2013:167–78 https://doi.org/10.1042/BJ20130457. determine mesoangioblast stem cell competence. Nat Commun. 2015;6 198. Bettica P, et al. Histological effects of givinostat in boys with Duchenne https://doi.org/10.1038/ncomms7364. muscular dystrophy. Neuromuscul Disord. 2016;26(10):643–9 https://doi. 221. Tedesco FS, Moyle LA, Perdiguero E. Muscle interstitial cells: A brief field org/10.1016/j.nmd.2016.07.002. guide to non-satellite cell populations in skeletal muscle. Methods Mol Biol. 199. Minetti GC, et al. Functional and morphological recovery of dystrophic 2017:129–47 https://doi.org/10.1007/978-1-4939-6771-1_7. muscles in mice treated with deacetylase inhibitors. Nat Med. 2006;12(10): 222. Chazaud B. Inflammation and Skeletal Muscle Regeneration: Leave It to the 1147–50 https://doi.org/10.1038/nm1479. Macrophages! Trends Immunol. 2020:481–92 https://doi.org/10.1016/j.it.202 200. Sandonà M, et al. HDAC inhibitors tune miRNAs in extracellular vesicles of 0.04.006. dystrophic muscle-resident mesenchymal cells. EMBO Rep. 2020a;21(9) 223. Juban G, Chazaud B. Metabolic regulation of macrophages during tissue https://doi.org/10.15252/embr.202050863. repair: insights from skeletal muscle regeneration. FEBS Letter. 2017:3007–21 201. Saccone V, et al. HDAC-regulated myomiRs control BAF60 variant exchange https://doi.org/10.1002/1873-3468.12703. and direct the functional phenotype of fibro-adipogenic progenitors in 224. Theret M, Mounier R, Rossi F. The origins and non-canonical functions of dystrophic muscles. Genes Dev. 2014;28(8):841–57 https://doi.org/10.1101/ macrophages in development and regeneration. Development (Cambridge). gad.234468.113. 2019;146(9) https://doi.org/10.1242/dev.156000. 202. Hu F, et al. CCN1 induces adipogenic differentiation of fibro/adipogenic progenitors in a chronic kidney disease model. Biochem Biophys Res Commun. 2019;520(2):385–91 https://doi.org/10.1016/j.bbrc.2019.10.047. Publisher’sNote 203. Cerquone Perpetuini A, et al. Janus effect of glucocorticoids on Springer Nature remains neutral with regard to jurisdictional claims in differentiation of muscle fibro/adipogenic progenitors. Sci Rep. 2020;10(1) published maps and institutional affiliations. https://doi.org/10.1038/s41598-020-62194-6. 204. Fardet L, Petersen I, Nazareth I. Prevalence of long-term oral glucocorticoid prescriptions in the UK over the past 20 years. Rheumatology (Oxford, England). 2011;50(11):1982–90 https://doi.org/10.1093/rheumatology/ker017. 205. McDonald CM, et al. Long-term effects of glucocorticoids on function, quality of life, and survival in patients with Duchenne muscular dystrophy: a prospective cohort study. Lancet. 2018;391(10119):451–61 https://doi.org/1 0.1016/S0140-6736(17)32160-8. 206. Ricotti V, et al. Long-term benefits and adverse effects of intermittent versus daily glucocorticoids in boys with Duchenne muscular dystrophy. J Neurol Neurosurg Psychiatry. 2013;84(6):698–705 https://doi.org/10.1136/jnnp-2 012-303902. 207. Palma, A. et al. (2019) Myo-REG: A portal for signaling interactions in muscle regeneration, Frontiers in Physiology, 10(SEP). doi: https://doi.org/10.3389/ fphys.2019.01216. 208. Oishi T, et al. Osteogenic Differentiation Capacity of Human Skeletal Muscle- Derived Progenitor Cells. PLoS One. 2013;8(2) https://doi.org/10.1371/journa l.pone.0056641. 209. Meyers C, et al. Heterotopic Ossification: A Comprehensive Review. JBMR Plus. 2019;3(4):e10172 https://doi.org/10.1002/jbm4.10172. 210. Leblanc E, et al. BMP-9-induced muscle heterotopic ossification requires changes to the skeletal muscle microenvironment. J Bone Miner Res. 2011; 26(6):1166–77 https://doi.org/10.1002/jbmr.311. 211. Wosczyna MN, et al. Multipotent progenitors resident in the skeletal muscle interstitium exhibit robust BMP-dependent osteogenic activity and mediate heterotopic ossification. J Bone Miner Res. 2012;27(5):1004–17 https://doi. org/10.1002/jbmr.1562. 212. Glass GE, et al. TNF-α promotes fracture repair by augmenting the recruitment and differentiation of muscle-derived stromal cells. Proc Natl Acad Sci U S A. 2011;108(4):1585–90 https://doi.org/10.1073/pnas.10185011 213. Abou-Khalil R, et al. Autocrine and Paracrine Angiopoietin 1/Tie-2 Signaling Promotes Muscle Satellite Cell Self-Renewal. Cell Stem Cell. 2009;5(3):298– 309 https://doi.org/10.1016/j.stem.2009.06.001. 214. De Palma M, et al. Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell. 2005;8(3):211–26 https://doi. org/10.1016/j.ccr.2005.08.002. 215. Bozycki L, et al. Whole-body clearing, staining and screening of calcium deposits in the mdx mouse model of Duchenne muscular dystrophy. Skelet Muscle. 2018;8(1) https://doi.org/10.1186/s13395-018-0168-8. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Skeletal Muscle Springer Journals

Origins, potency, and heterogeneity of skeletal muscle fibro-adipogenic progenitors—time for new definitions

Loading next page...
 
/lp/springer-journals/origins-potency-and-heterogeneity-of-skeletal-muscle-fibro-adipogenic-ASgw5SDYJj
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2044-5040
DOI
10.1186/s13395-021-00265-6
Publisher site
See Article on Publisher Site

Abstract

Striated muscle is a highly plastic and regenerative organ that regulates body movement, temperature, and metabolism—all the functions needed for an individual’s health and well-being. The muscle connective tissue’s main components are the extracellular matrix and its resident stromal cells, which continuously reshape it in embryonic development, homeostasis, and regeneration. Fibro-adipogenic progenitors are enigmatic and transformative muscle-resident interstitial cells with mesenchymal stem/stromal cell properties. They act as cellular sentinels and physiological hubs for adult muscle homeostasis and regeneration by shaping the microenvironment by secreting a complex cocktail of extracellular matrix components, diffusible cytokines, ligands, and immune- modulatory factors. Fibro-adipogenic progenitors are the lineage precursors of specialized cells, including activated fibroblasts, adipocytes, and osteogenic cells after injury. Here, we discuss current research gaps, potential druggable developments, and outstanding questions about fibro-adipogenic progenitor origins, potency, and heterogeneity. Finally, we took advantage of recent advances in single-cell technologies combined with lineage tracing to unify the diversity of stromal fibro-adipogenic progenitors. Thus, this compelling review provides new cellular and molecular insights in comprehending the origins, definitions, markers, fate, and plasticity of murine and human fibro-adipogenic progenitors in muscle development, homeostasis, regeneration, and repair. Keywords: Mesenchymal stromal/stem cell, Fibro/adipogenic progenitor, Fibroblast, Adipocyte, Regeneration, Single-cell RNAseq Background Mammalian adult skeletal muscle has extraordinary re- In mammals, skeletal muscle represents ~ 30–40% of the generation capabilities upon injury, making the organ a total body mass, regulating body temperature, metabolism, perfect model to study regeneration and repair, and inves- and physical activity. Comprising the musculoskeletal sys- tigate the contribution of adult stem and interstitial cells tem, striated muscles are responsible for voluntary and in settings of acute or chronic injury. The muscle connect- non-voluntary movements. Skeletal muscles are recog- ive tissue (MCT) components are the extracellular matrix nized as highly plastic tissue, illustrated by atrophic or (ECM) and its stromal cells, which actively produce, main- hypertrophic changes when disused or trained. tain, and remodel this dynamic scaffold during develop- ment, homeostasis, and after trauma. Among the several cell types that participate in muscle re- * Correspondence: o.contreras@victorchang.edu.au; mtheret@brc.ubc.ca Developmental and Stem Cell Biology Division, Victor Chang Cardiac generation, tissue-resident mesenchymal progenitors play a Research Institute, Darlinghurst, NSW 2010, Australia crucial role by providing signaling cues that modulate other Biomedical Research Centre, Department of Medical Genetics and School of muscle-resident cells’ function, and actively remodel the Biomedical Engineering, University of British Columbia, Vancouver, BC V6T 1Z3, Canada ECM during this process. Fibro-adipogenic progenitors Full list of author information is available at the end of the article © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Contreras et al. Skeletal Muscle (2021) 11:16 Page 2 of 25 (FAPs) have been identified as platelet-derived growth factor known as muscle fibroblasts [17–19]. However, com- receptor alpha (PDGFRα,alsoknown as PDGFRA)express- pared with the ever-growing knowledge about adult ing cells [1, 2]. Agrowing bodyofevidenceshows that MCT, the composition and the dynamic remodeling of PDGFRα+ FAPs provide regenerative cues to control muscle embryonic MCT are poorly understood. While evidence stem cell (MuSC) expansion, fate, and myogenesis after acute about the ontogeny of interstitial muscle cells exists damage and aging [1–7]. Furthermore, the ablation of stro- [20–22], only a paucity of studies have reported their mal cells by using mice model expressing the diphtheria embryonic determination, and hence, the developmental toxin receptor (DTR) under the control of the fibroblast acti- origin and role of these ECM-embedded cells are not yet vation protein alpha promoter (FAPα-DTR) impairs the fully appreciated and understood. long-term maintenance of hematopoiesis, muscle mass, and Kardon and colleagues published early evidence of the cachexia [8]. To note, FAPα+ cells are found in most tissues function of these cells in the formation of limb muscles such as bone, salivary gland, visceral adipose tissue, skeletal in the 2000s [23]. The authors described that a mesoder- muscle, and pancreas; express CD90, CD140a, and SCA-1; mal population of TCF7L2+ cells (formerly known as T- and so are most likely to be mesenchymal progenitors, hence cell factor 4 or TCF-4, a TCF/LEF transcription factor FAPs in skeletal muscle [8]. These findings have been con- downstream the canonical Wnt/β-catenin signaling) reg- C- firmed by the Rando laboratory using a knock-in PDGFRα ulates the spatiotemporal determination and differenti- reER DTA :Rosa26 mice model [7], and more recently, by ation of myogenic progenitors and, therefore, regulates Tsuchida’s group using a similar cell ablation strategy [9]. In- limb muscle development in chicks [23]. Limb TCF7L2+ deed, genetic ablation of PDGFRα+ lineage cells leads to im- precursors derive from the lateral plate mesoderm in a paired MuSC expansion and leucocyte infiltration, leading to muscle-specific pattern, but are different from myogenic deficient skeletal muscle regeneration after acute chemical precursors since they do not form muscle nor express injury and neuromuscular defects and muscle atrophy [7, 9]. classical myogenic markers (e.g., Pax7)[6, 23, 24]. Thus, In addition, following limb ischemia, proper muscle revascu- myogenic precursors are patterned by extrinsic cues, larization and repair are lost after ablating FAPs [10]. Hence, mostly coming from the MCT, after the cells have mi- PDGFRα+ FAPs are required for successful muscle regener- grated through the limb rather than being embryonically ation, repair, and maintenance during tissue homeostasis and predetermined to form particular muscle anatomical in pathological states. structures [23, 25]. These MCT progenitors also influ- Muscle-resident PDGFRα+ cells readily initiate fi- ence the myofiber type of limb and diaphragm muscles broblastic colonies (also called fibroblast colony- in a paracrine fashion [24]. Interestingly, not all limb formingunits,CFU-F,(Fig. 1a)) and can clonally dif- muscles contain TCF7L2+ cells during mouse embryo ferentiate not only into activated fibroblasts/myofibro- development, which suggest a distinct patterning and blasts and adipocytes but also into chondrogenic and three-dimensional distribution of these cells in different osteogenic lineages depending on the context [1, 2, subtypes, or the existence of MCT progenitors that do 10–15]. The plasticity and clonal expansion of muscle not express this marker [26]. Nevertheless, TCF7L2 la- FAPs are also seen in humans [16]. However, the ef- bels a significant proportion of mammalian stromal non- fects of damage-induced signals and cues on their myogenic precursors at birth and during adulthood [24, plasticity, fate, and functions have only recently begun 27, 28]. Additionally, MuSCs and endothelial cells also to be explored. The development of new in vivo express Tcf7l2 mRNA and protein, albeit at low levels lineage tracing tools used to identify and track cells compared with FAPs [7, 24, 28]. expressing specific markers in various animal and Researchers have argued that vertebrate muscles de- damage models in parallel with the recent emergence rive from several developmental sources, adding com- of single-cell omics have allowed the identification of plexity to our understanding of the different origins of a broad spectrum of specific stromal populations and MCT in muscle development. For comprehensive re- their relative contribution to muscle homeostasis, re- views, see: [20–22]. Like myogenic precursors, MCT generation, and repair. progenitors originate from different and distinct struc- tural origins during embryonic development. In mam- The developmental ontology of muscle-resident mals, these include the somites for axial-trunk muscles mesenchymal progenitors [29], the lateral plate mesoderm for limb muscles [23, From the embryo to the adult: role of MCT mesenchymal 30], the neural crest cells (NCCs) for head and neck progenitors on muscle development muscles [31–34], and the transient developmental Adult MCT is mainly composed of ECM, largely fibrillar structure originating from the somites called pleuroper- collagens type 1 and 3, elastin, fibronectin and proteo- itoneal folds (PPFs) for the diaphragm [35]. Remark- glycans, and the supportive matrix-resident stromal cells, ably, Merrell and colleagues demonstrated that PPF- + + also called mesenchymal progenitors or traditionally resident TCF7L2 /GATA4 CT precursors regulate the Contreras et al. Skeletal Muscle (2021) 11:16 Page 3 of 25 Fig. 1 a Illustration of FAP cellular properties, including the high expression of PDGFRα, quiescency, CFU-F, and mesenchymal/stromal cell multipotency. Skeletal muscle fibro-adipogenic progenitors form clonal CFU-F following in vitro cell culture. b Z-stack confocal images showing the localization of PDGFRα-EGFP cells in tibialis anterior muscle sections of adult PDGFRαH2BEGFP/+ knock-in mice. Pictures show different skeletal muscle anatomical locations of muscle FAPs. Laminin (magenta) and nuclei (Hoechst, blue) were also stained. Scale bars: 50μm development of the diaphragm and participate in the and Tbx5, in determining the formation of muscles and etiology of congenital diaphragm hernias (CDH), a type tendons of the musculoskeletal system [36]. Interest- of fibroproliferative developmental disorder [35]. The ingly, they found that the myoblast-specific loss of Tbx5 authors also demonstrated that Gata4 null mutations does not affect the correct positioning of myogenic pre- in CT progenitors expressing Paired related homeobox cursors. However, genetic deletion of Tbx5 and Tbx4 in 1(Prx1) could cause CDH during diaphragm develop- the mesenchyme (paired related homeobox (Prx) ex- ment. These studies indicate that the aberrant behavior pressing lineage), resulted in the perturbation of MCT of PPFs CT progenitors can cause congenital muscle organization, and therefore, caused mispatterned muscle diseases like CDH [35]. limbs. Although the authors observed no changes in the The studies of Logan’s group have also helped to ad- expression of Tcf7l2 in the absence of Tbx4/5, the lack vance our understanding of the developmental role of of these transcription factors impaired the spatiotempo- MCT precursors in muscle morphogenesis. Initially, ral distribution of TCF7L2+ cells [36]. Remarkably, the through a combination of conditional deletion and ad- Holt-Oram syndrome, known for leading to skeletal ab- vanced imaging techniques, they demonstrated the cru- normalities and congenital heart disease, is caused by cial participation of T-box transcription factors, Tbx4 mutations in the Tbx5 gene [37]. The study of Hasson Contreras et al. Skeletal Muscle (2021) 11:16 Page 4 of 25 et al. reinforces the model in which MCT gives rise to regionalization of muscles and tendons, independently muscle pre-patterned structures to guide myogenic pre- of bone defects [46]. Altogether, these results not only cursors during development and further demonstrated demonstrate a previously unappreciated function of that extrinsic MCT-derived cues are critical for muscle Hox genes for proper patterning and integration of morphogenesis. Without surprise, the Transforming muscles, tendons, and bones but also illustrated that Growth Factor beta (TGF-β) signaling pathway is in- CT spatiotemporal dynamics participate in the integra- volved in this process. Indeed, Kutchuk et al. demon- tion of the musculoskeletal system as a whole. Further strated that embryonic myofibers and C2C12 myoblasts studies should detail how, when and what factor(s) express Lysyl Oxidase (Lox, an enzyme required for modulate the spatiotemporal dynamics and positional cross-linkage formation in elastin and collagen) and that fate of muscle connective tissue cells. −/− its deletion upregulates the TGF-β signaling. Lox mu- tants display MCT disorganization and delayed myogen- Searching for cell-type-specific markers of muscle stromal esis [38]. Thus, this study illustrates the homeostatic fibro-adipogenic progenitors cross-talk between MCT and muscle cells during limb In adult tissue, two studies characterized a population of musculoskeletal system development. interstitial muscle-resident progenitors with spontaneous The above-proposed model was recently corroborated mesenchymal stem/stromal cell (MSC) potential towards in detail by Besse and colleagues [39]. These authors fibrous myofibroblast and fatty differentiation [1, 2]. took full advantage of an array of labeling and imaging- Using fluorescence-activated cell sorting (FACS) of based studies, mouse genetics, and transcriptomic ana- digested mouse skeletal muscle, our laboratory identified lyses to establish how individual muscle bundles are gen- and named these cells as fibro-adipogenic progenitors erated and established, shedding profound lights on the (FAPs) based on their spontaneous differentiation along role of MCT precursors on muscle morphogenesis at these lineages [1]. We characterized these progenitors as unprecedented resolution. They provided a compelling lineage-negative (Lin−, not expressing hematopoietic demonstration that muscle morphogenesis is primarily (CD45), endothelial (CD31, also known as PECAM-1) or orchestrated by CT mesenchymal progenitors via the se- myogenic markers (α7-INTEGRIN) and positive for cretion of matrix-modifying proteoglycans [39]. Thereby, Stem cell antigen-1 (SCA-1) and CD34 cell-surface anti- through the expression and the secretion of a myriad of gen expression. Interestingly, while quiescent MuSCs, chemoattractants, ECM components, and growth fac- endothelial cells, and a subset of hematopoietic cells ex- tors, these stromal cells promote a variety of responses press CD34, its genetic deletion impairs MuSC but not in myogenic precursors, including repulsion, attraction, FAP proliferation [47]. We also demonstrated that most migration, and patterning [20]. Hence, the MCT creates Lin-/α7 INTEGRIN-/SCA-1+ cells express high levels of a developmental pre-pattern that orientates and controls the receptor tyrosine kinase PDGFRα [1]. Similarly, the positioning of myogenic precursors that differentiate Uezumi et al. characterized the same population using a into myofibers forming muscle bundles and, conse- different gating strategy. They used CD45, CD31, and quently, will serve to define the size and shape of mus- Sm/C2.6 (MuSC marker) as a negative selection and cles, the orientation of its myofibers, and points of origin CD140a (PDGFRα) as positive. They showed that Lin- and insertion on bones [23, 25, 36, 39, 40]. PDGFRα+ cells express a low level of PDGFRβ and can The notion that MCT cells participate in muscle differentiate in adipocytes, myofibroblasts, and chondro- morphogenesis leads to wonder what determines the cytes in vitro [2]. They also observed that muscle PDGF spatiotemporal dynamics and positional information of Rα+ cells were perivascular but did not co-localize with MCT precursors. Hox genesare aset of genescoding NG2, suggesting that PDGFRα+ cells are not pericytes for transcription factors that specify segment identity [2]. and provide positional information during animal de- PDGFRα+ cells reside in the muscle interstitium and velopment [41]. Among them, the caudal Hox11 genes are more abundant in the epimysium and perimysium participate in determining the proximal-distal axis of than in the endomysium. Although most muscle- the musculoskeletal system of limbs [42–45]. Hoxa11 is resident PDGFRα+ progenitors are in close association broadly expressed through the distal primordium of with blood vessels [1, 2, 48], they are distinct from peri- limb buds at E10.5, but later on, at E14.5, it is exclu- cytes. Indeed, pericytes are embedded within the endo- sively expressed in the CT of tendons, perichondrium, thelium basement membrane, but PDGFRα+ cells reside and TCF7L2+ cells, but not in endothelial cells, chon- outside of vessels. The localization of FAPs is evident drocytes, osteocytes, nor myogenic precursors [46]. around large blood vessels, in which they adopt an ad- Genetic deletion of Hoxa11/Hoxd11 paralogs, which ventitial position. With rare exceptions in organs other have a prominent role in patterning bones during de- than muscle, PDGFRα cells do not express defining peri- velopment, leads to severe defects in the pattern and cyte markers like Cspg4 (NG2), Rgs5, Pdgfrβ,or Mcam Contreras et al. Skeletal Muscle (2021) 11:16 Page 5 of 25 (CD146) [2, 48, 49]. Notably, while FAPs were initially Recently, Gli1 (also known as glioma-associated oncogene described in murine muscles, growing evidence indi- 1) expression has been shown to label a subpopulation of cates that human FAPs have a similar phenotype and muscle FAPs with higher clonogenicity and reduced adipo- functions to mouse FAPs [16, 50–54]. In summary, FAPs genic differentiation than Gli negative FAPs [65]. Perivascu- (historically called fibroblasts) and the ECM they actively lar cells expressing the zinc finger protein Gli1 undergo secrete and modify are both significant constituents of the proliferative expansion and generate myofibroblasts after interstitium and perivascular CT. kidney, lung, liver, and heart injury [68]and heartinjury Distinct subpopulations of CT progenitors exist and [6], suggesting that Gli1+ cells are likely a FAP subpopula- express an array of proteins and transcription factors, al- tion as recently shown in skeletal muscles [69]. beit at variable levels. In the mouse embryo, CT progeni- In humans, cell-surface markers like PDGFRα, CD201, tor markers include PDGFRα, TCF7L2, TBX3/4/5, CD166, CD105, CD90, CD73, and CD15 identify skeletal HOX11, and the Odd-skipped transcription factors muscle FAPs (Table 2)[16, 51, 53, 54, 64, 71, 72]. Remark- OSR1 and OSR2 [21, 23, 26, 40, 55, 56] (Table 1). In ably, the expression of SCA-1 defines a particular cluster of murine adult muscles, the large majority of CT fibro- stromal cells within the murine FAP population with differ- adipogenic progenitors express PDGFRα, SCA-1 (also ent potency and properties in vivo and in vitro, both in the known as Ly6A/E), CD90 (THY1), CD34, TCF7L2, skeletal muscle and heart [66, 73]. However, as SCA-1 does HIC1, VIMENTIN, DECORIN, and ADAM12 but few of not have a human homolog, its use to identify FAPs is lim- these markers are specific and unique for this heteroge- ited by the absence of this antigen in humans. neous population of cells (discussed below) [1, 11, 12, Recently, we showed that the majority of cells express- 28, 57, 59, 61, 63, 66] (Table 1). Of note, murine adult ing the protein-coding gene Hypermethylated in Cancer muscle PDGFRα+ FAPs express low levels of Osr1, 1(Hic1) correspond to quiescent muscle-resident FAPs which increases upon acute injury in a small subset of in mice [48]. In adult muscles, HIC1+ progenitors reside FAPs, suggesting the participation of regulatory mecha- in the interstitial space and the myotendinous junctions. nisms that tightly turn on the expression of Osr1 resem- In addition to FAPs, small subsets of pericytes (SCA-1−, bling developmental-like programs [67]. Remarkably, RSG5+ cells) and tenogenic cells (SCA-1−, SCX+ and damage-activated OSR1+ FAPs proliferate faster com- FMOD+ cells) express HIC1. Therefore, the expression pared with OSR1- FAPs [67], suggesting that either of Hic1 comprises a larger proportion of mesenchymal OSR1 modulates the expansion and functions of FAPs, stromal progenitors compared with the expression of PDGF or it represents an activation marker whose expression Rα, which is limited to FAPs [48]. Along with others, we increases in proliferating cells. have confirmed that cardiac PDGFRα+ cells also exhibit Table 1 Summary of endogenous murine skeletal muscle fibro-adipogenic progenitors Murine Canonical Alternative Negative Localization Differentiation potential Additional comments References cell Markers markers markers Embryonic- PDGFRα Osr2 CD45 Muscle- Robust in vitro adipogenic Little is known about their [24]; [26]; [57]; fetal FAPs TCF7L2/ Hox11 CD31 associated and fibrogenic differentiation origin, fate, gene TCF4 Tbx3 Ter119 connective but low chondrogenic and regulation, function, Osr1 Tbx4 α7- tissue and no detectable osteogenic or stemness, and self- Tbx5 Integrin muscle myogenic potential. Osr1+ renewal Sca-1 interstitium progenitors also give rise to CD34 embryonic fibroblast-like cells Adam-12 in the dermis and FABP4+ Tie-2 adipocytes in white fat pads Adult FAPs PDGFRα Hic1 CD45 Fascia, Adipocytes, myofibroblasts, Required for adult skeletal [1]; [11, 12, 27, 28]; SCA-1 CD90 CD31 epimysium, osteocytes, and muscle regeneration and [13]; [14]; [58]; [58]; Decorin Ter119 perimysium, and chondrocytes after muscle homeostasis; cellular and [59]; [60]; [61]; [62], (Dcn) α7- endomysium; injury and in vitro, with no molecular dysfunction in [2, 15, 53, 63, 64]; PDGFRβ Integrin abundant as myogenic potential pathology and disease Col1a1 NG2/ perivascular cells TCF7L2/ Cspg4 TCF4 Rsg5 CD34 Adam-12 Tie-2 Gli1 These markers have not been studied in the embryo with detail These markers are also expressed by different cell types, including satellite cells, pericytes, and endothelial cells Adam-12 and Tie-2 expression appears to be restricted for a subpopulation of FAPs Gli1 defines a subpopulation of murine muscle FAPs with pro-myogenic and anti-adipogenic functions [65] Contreras et al. Skeletal Muscle (2021) 11:16 Page 6 of 25 Table 2 Summary of endogenous human skeletal muscle fibro-adipogenic progenitors Human Canonical Alternative Negative Localization Differentiation potential Additional References cell Markers markers markers comments Embryonic- PDGFRA DCN PAX3 Similar to what is Not evaluated but probably No information [70] fetal FAPs FN1 PAX7 found in mouse similar to what is found in about their origin, LUM development, mouse development gene regulation, OSR1 although not function, and POSTN evaluated in detail potency FAP THY1/CD90 VIM NT5E/CD73 COL1A1 COL1A2 COL3A1 PTN OGN FBLN5 Adult FAPs PDGFRA, CD34 CD201 CD31 Fascia, epimysium, Adipocytes, myofibroblasts, Increased numbers [16]; [51]; (when negative CD166 CD45 perimysium, and osteocytes, and chondrocytes in diverse [71]; [72]; for CD56, CD31 CD105 CD56 endomysium; in diseased states and in vitro. pathologies [52]; [53, and CD45) CD90 α7- abundant as Lack of myogenic potential 54, 64] CD73 Integrin perivascular cells CD34 NG2/ CD15 CSPG4 COL1A1 RSG5 TCF7L2/ TCF4 These other alternative markers suggested by Pyle and colleagues are based on scRNAseq data ([70]) multilineage properties in vivo and in vitro [11, 12, 73–75]. highlighting the contribution of endogenous PDGFRα+ In themurineheart, HIC1+progenitors represent a signifi- cells to mammalian skeletal muscle homeostasis, regen- cant proportion of cardiac FAPs [73]. HIC1 (also known as eration, and repair, it is worth revisiting the terminology ZBTB29) is a transcription factor involved in quiescence and in this area. Muscle regeneration is defined as the spe- cell cycle control [76, 77]. Consistent with these roles, the cific substitution or replacement of lost tissue, eventually conditional deletion of Hic1 induces aberrant cell activation leading to full restoration of muscle strength. This re- and proliferation of FAPs, impairing muscle regeneration fol- generative capacity relies on resident adult unipotent lowing acute damage and leading to spontaneous develop- stem cells (also known as satellite cells), which are quies- ment of arrhythmogenic cardiomyopathy-like pathology and cent but activate to rebuild this tissue upon injury [6, signs in the mouse heart [48, 73]. Thus, the unrestrained ac- 78–80]. tivation of these progenitor cells and the consequent gener- In comparison, skeletal muscle repair aims to safe- ation of differentiated progeny are potential pathological guard the remaining function of muscle following solu- drivers of disease. tions of continuity after partial loss of tissue due to We claim that the heterogeneity of FAP markers massive traumas or chronic insults such as repetitive in- makes sense in a context where the upregulation and juries, disease, and aging. Thus, muscle repair often en- downregulation of cell-specific makers participate in tails replacing lost myofibers with scar tissue, which acts modulating the commitment of FAPs into a transitional as a bridge between areas still capable of contraction (for cell state or differentiation process during lineage pro- review of tissue regeneration and repair see [81–83]. gression in response to injury or in disease states. Hence, Therefore, while repair restores muscle integrity, regen- FAP heterogeneity might contribute to restricting or eration accounts for restoring tissue function. As ob- priming the multipotency of PDGFRα+ FAPs. These are served in other mammalian regenerative tissues such as important issues to explore in future research. the liver [81, 84], the form and periodicity of damage can impair the ability of the skeletal muscle to return to Adult muscle connective tissue and PDGFRα+ homeostasis [59, 85, 86]. Therefore, the current estab- progenitors lished dual role that PDGFRα+ cells play in acute (re- Adult skeletal muscle contains several cell types that generative) and chronic damage (reparative/ work in unison under tightly regulated conditions to degenerative) suggests that the organization of their maintain homeostasis. Adult mammalian muscle is a re- intracellular signaling network may integrate opposite markable exception to the low regenerative potential of complementary signals whose relative strength mainly several organs and tissues like the heart. Before we start depends on the type, extension, and frequency of the Contreras et al. Skeletal Muscle (2021) 11:16 Page 7 of 25 injury. For this review, we redefine the fate and hetero- radically increased the attention paid to MCT [93–100]. geneity of muscle-resident PDGFRα+ progenitors and In order to understand MCT development, establish- explain their multilineage potentials. ment, and remodeling, it is crucial to consider the stro- mal cells that participate in these processes. A critical The adult muscle connective tissue and challenging step towards a complete understanding The muscle environment is complex in structure and of MCT biology has been the identification of a hetero- several heterogeneous cell types co-exist within it to geneous population as the primary effector of ECM de- regulate its function and structure. Although skeletal position and remodeling [11, 12, 17, 18, 48, 62]. muscles have an intricate network of blood vessels and Increasing evidence suggests that there are distinct sub- nervous tissue, most of their mass is comprised of myo- sets of stromal cells located in discrete yet similar ana- fibers. In adults, MCT, which accounts for 1–20% of the tomical positions during muscle development and into total dry mass of muscles, surrounds, protects, and inter- adulthood [20–22]. This stromal cellular diversity and connects these primary components [19, 87, 88]. The heterogeneity have been an obstacle to attributing the amount of CT varies significantly from one muscle to primary role for matrix deposition to a specific subset of another, depending on the anatomical location and stromal cells. physiological function of particular muscles [89]. The Jackson and colleagues reported the existence of adult MCT follows the nomenclature of fascia, epimy- tissue-resident mesenchymal progenitors with multiline- sium, perimysium, and endomysium accordingly to its age differentiation capabilities in damaged human location and arrangement within the tissue [90, 91] (Fig. muscle over a decade ago [101]. Today, thanks to the 1b). The topological organization of the covering con- great effort of many researchers, we know that adult nective tissue from the outside is described as follows: MCT is mainly produced by muscle-resident PDGFRα+ the fascia, the CT outside the epimysium that surrounds cells with multilineage progenitor properties and a and separates the muscles; the epimysium, which sur- fibroblast-like phenotype, called FAPs. Increasing evi- rounds each muscle group, linking them to the tendons dence suggests that these muscle-resident cells are the at the myotendinous junctions; the perimysium, which primary cellular source of regenerative matrix deposition consists of collagen-rich structures that surround the as well as scarring following muscle injury, disease, fascicles and interconnect with the epimysium; and the neuromuscular disorders, or aging [1, 2, 5, 9, 11, 12, 14, endomysium, which represents a modified basement 15, 27, 53, 54, 59, 61, 102–105]. Vallecillo-García and membrane unsheathing individual myofibers and inter- colleagues showed that the source of developmental connects to the perimysium [19, 91, 92] (Fig. 1b). These ECM in limb muscles is a heterogeneous population of four levels of stromal organization describe the intercon- PDGFRα-expressing progenitors called embryonic FAPs, nected ECM compartments within muscles. Although closely resembling the population of adult stromal cells each compartment is distinguished by its anatomical we have described, along with other groups [1, 2, 6, 23, position, it is difficult to discriminate each of these ECM 24, 26]. These findings led to some confusion in the no- compartments in terms of their protein and cellular menclature, with some publications distinguishing be- composition. Remarkably, MCT not only determines the tween FAPs and fibroblasts, some using the term FAPs macro and microstructures of embryonic and adult mus- as better representing their predominant fibrogenic and cles but also connects the myofibers to produce and adipogenic developmental potential, and some remaining transmit force. As a result, it increases not only the effi- faithful to the historical term fibroblast, which are also ciency of force generation but also protects myofibers known for being heterogeneous and plastic cells. Here, from excessive stretching, supporting muscle regener- we propose that these muscle-resident multipotent pro- ation and cellular mechanosensation [17–19]. genitors, whether called FAPs or fibroblasts, are the same cells. Muscle-resident fibro-adipogenic progenitors: definitions From this point on, the term PDGFRα+ FAPs will and identity refer to muscle-resident CT mesenchymal progenitor Historically, the observation of ECM proteins, such as with multilineage developmental properties. As dis- collagens, being produced and deposited in skeletal cussed below, recent advances in single-cell RNA se- muscle suggested the existence of a resident collagen- quencing demonstrated that FAPs comprise multiple producing cell within the tissue [17, 18, 90]. Later, nu- sub-populations, some of which could be bona fide dif- merous observations of CT hyperplasia and interstitial ferentiated cells with little developmental potential left proliferation associated with healing scars in skeletal [16, 48, 62, 106–109]. This may create a problem with muscle diseases, including congenital muscular dystro- nomenclature diversity, speculation, and high cellular phies, immobilized muscles, and neuromuscular disor- heterogeneity within the adult stromal lineage [110]. ders (e.g., amyotrophic lateral sclerosis and denervation) FAP heterogeneity is also known to increase following Contreras et al. Skeletal Muscle (2021) 11:16 Page 8 of 25 injury and disease, which also complicates their classifi- These results further confirm our idea that muscle FAPs cation and nomenclature [48, 62, 66, 108, 111]. cannot be solely identified using collagen I reporter The muscle community has historically described mice, but as previously suggested, we strongly recom- interstitial cells with MSC capability (i.e., fibrogenic, adi- mend employing PDGFRα expression. Since most of the pogenic, chondrogenic, and osteogenic potency). In work related to FAPs biology refers to models of single, addition to PDGFRα+ cells, muscle-resident pericytes or repeated rounds of injury, we believe that further have also been proposed to be MSCs that have adapted studies will likely uncover the role of PDGFRα+ cells in to the specialized functions required by their adjacent atrophy-related pathologies such as aging-related sarco- vascular niche. However, although PDGFRα+ FAPs be- penia, cachexia, myasthenia gravis, polytrauma, and have as and present defined canonical MSC properties, neuromuscular disorders. Further research is needed to FAPs are different from tissue-resident pericytic “MSCs.” clarify the existence of subtle differences within stromal Indeed, pericytes' cell-surface profile is CD34-/CD45- cells that might have functional impacts and conse- and CD146+ [112, 113]. Remarkably, Bianco and col- quences in muscle physiology, not only during mainten- leagues revisited the MSC origins and differentiation po- ance but also in pathological and disease states. tential using a broad set of human MSC-like cells (HLA class I, CD73, CD90, CD105, and CD146 positive cells). Multipotency of muscle-resident PDGFRα+ fibro- The authors showed that the cell surface phenotype of adipogenic progenitors “MSCs” isolated from bone marrow, skeletal muscle, In healthy adult muscles, we and others have demon- periosteum, and cord blood, although quite identical, did strated that PDGFRα+ cells represent between ~ 5–15% not reflect these cells’ cell transcriptomic identity, func- of the total nuclei and ~ 20–30% of the interstitial mono- tion, and therefore, their differentiation properties. Thus, nuclear cells at homeostasis [11, 12, 106, 107, 116]. Stro- “MSCs” are separated from each other, as the authors mal PDGFRα+ FAPs display MSC properties and can defined it, by a developmental origin factor [113]. Not- spontaneously differentiate into adipocytes (rounded, ably, the authors also showed that CD146+ pericytes are single-vacuole lipid-rich cells, perilipin+ and peroxisome not true MSCs in most of the analyzed tissues, with the proliferator-activated receptor gamma+ (PPARγ)), acti- possible exception of the bone marrow, where they in- vated fibroblasts (long-shaped contractile cells with herently form bone and bone marrow stroma but lack fibroblast-like morphology, αSMA (Acta2), and highly chondrogenic potential in vivo or myogenic in vitro. On producing ECM cells), as well as chondrocytes/osteoblasts the contrary, in skeletal muscle, CD146+ perivascular when bulk cultured, and in clonal assays in vitro and pericytes are rather inherently myogenic than skeleto- in vivo [1, 2, 11–13 15, 50, 51, 59, 104, 117]. Notably, genic [113]. Remarkably, skeletal muscle pericytes are a HGFA, an injury-induced systemic cue, activates muscle distinct cell type from MuSCs (CD56+/CD146-) and FAPs, priming these cells to transition from quiescence CD34+/CD146+ endothelial cells that possess a latent into a cellular state with enhanced regenerative potential myogenic gene signature and potential, and hence, also known as G alert state [118]. In the following chap- muscle pericytes are committed myogenic progenitors ters, we discuss FAP multipotency (Fig. 2). [113, 114]. These pivotal studies have challenged the loose and non-specific MSC nomenclature. However, Fibrogenic potential of PDGFRα+ FAPs further studies with lineage tracing and clonal assays are When FAPs are cultured in vitro using standard growth needed to deeply understand stromal cell dynamics in media and 20% oxygen, a large proportion of them will development, homeostasis, and injury and, therefore, to spontaneously differentiate into activated fibroblasts finally faithfully unify their markers, nomenclatures, and with αSMA stress fibers [1, 11, 12, 15] (Fig. 2). This definitions. demonstrates that FAPs have intrinsic capabilities to dif- The abundance of collagen, especially the most abun- ferentiate, which is unleashed following their activation dant protein in animals, type I collagen, determines the and makes in vitro studies easily feasible. However, the stiffness of mammalian tissues [115]. Notably, increased mechanisms regulating the fibrogenic potential of FAPs production and deposition of type I collagen fibrils are remain underexplored. found after muscle damage. Several cell sources have been suggested as producers of collagen proteins. Using Transforming growth factor-beta signaling a murine model of increased of increased muscle fibro- One of the most studied signaling pathways in regulating sis, Chapman et al. corroborated that at least three dif- the behavior and fate of muscle FAPs is the transforming ferent muscle-resident cell populations express collagen growth factor-beta (TGF-β) signaling pathway. The I, among them PDGFRα+ FAPs. However, muscle pro- TGF-β sub-family of cytokines (TGF-β1, TGF-β2, and genitors (α7-INTEGRIN)+ and SCA-1+ cells also ex- TGF-β3) are secreted proteins that participate in cell- press the mRNA for this fibrillar matrix protein [17, 18]. and tissue-specific biological processes such as wound Contreras et al. Skeletal Muscle (2021) 11:16 Page 9 of 25 Fig. 2 Skeletal muscle FAPs are quiescent cells with multipotency to differentiate towards all the mesenchymal lineages, depending on the degree of activation and tissue damage. Tissue injury and its associated biochemical cues and cell-secreted factors activate muscle FAPs. Activated FAPs act as immunomodulatory stromal cells and signaling hubs before their commitment to more specialized cells. Usually, muscle injury induces the differentiation of them into activated fibroblasts and adipocytes. Severe damage and chronic pathologies tip their differentiation also into chondrogenic and osteogenic lineages. The figure also shows different molecules and factors as well as ligands that regulate their differentiation potential and fate. Notably, many of these molecules hold several steps of FAP life. As quiescent FAPs find their way into activation and cell differentiation, they lose the expression of quiescence markers and their FAP identity but gain cell differentiation markers healing, angiogenesis, immune regulation, apoptosis, release from the LAP-LTBP complex, which can occur tumorigenesis, and proliferation. In pathological condi- via proteolytic rupture or through ECM-cell forces gen- tions, they strongly associate with tissue damage, dys- erated by cell traction via the integrin complexes [126– function, and fibrosis and are notably mis-expressed 129]. After release, TGF-β binds to its heteromeric (Burks & Cohn, [119, 120]). The complexity of the TGF- serine/threonine kinase type 1 and 2 receptors β pathway is exemplified by its pleiotropic effects, indu- [TGFBR1/ALK5 and TGFBR2, respectively], and cing growth arrest in some cell types but promoting the TGFBR3 (also known as betaglycan) co-receptor on the proliferation of others [121, 122]. TGF-β enhances the cell surface of the target cell. Of interest, while TGF- proliferation and differentiation of several cell types, in- β family ligands can bind TGFBR3, this receptor does cluding stromal cells (for review, see [123, 124]. not have signaling activity on its own, but it modifies the When secreted, TGF-β associates non-covalently to a affinity of TGFBR1 and 2 to TGF-β ligands [130]. In- large complex consisting of the latency-associated pep- deed, TGFBR3 acts as a co-receptor, amplifying TGF-β tide (LAP) and latent TGF-β-binding protein (LTBP) signaling activation [131]. TGFBR3 also binds other proteins [125]. Extracellular TGF-β is activated after its TGF-β-family ligands such as ACTIVINS, INHIBINS, Contreras et al. Skeletal Muscle (2021) 11:16 Page 10 of 25 and bone morphogenetic proteins (BMPs), which we in vitro and in vivo [28]. Thus, as these cells activate, know are primordial proteins for ECM remodeling in proliferate, and differentiate they lose or reduce the ex- skeletal muscle [132, 133]. This co-receptor can also be pression of their progenitor state markers (Fig. 2). soluble (by a mechanism called shedding [134]) and could, in some cases, act as an inhibitor of the TGF-β Wnt/β-catenin signaling signaling by sequestration of its various ligands [131, The Wnt/β-catenin pathway relies on the binding of 135]. Nevertheless, the function of TGFBR3 in FAP or Wnt ligands to Frizzled receptors and the co-receptors mesenchymal progenitor behavior has not been studied LRP5 and LRP6 at the cell surface to initiate a cascade yet. Its modulation could be a powerful tool as TGFBR3 that regulates the intracellular proteostasis of β-catenin misexpression is associated with cancer and metastasis (for recent reviews about the Wnt/β-catenin signaling [136], in which ECM remodeling is known to be highly see [156, 157]. At steady state, the β-catenin pool that is active. not participating in cell adhesion is bound to a destruc- Then TGF-β canonical downstream effectors SMAD2 tion complex, where it becomes phosphorylated and tar- and SMAD3 (R-SMADs) are phosphorylated throughout geted for degradation in a process mediated by the TGFBR1/ALK5 kinase activity and form a cytoplasmic ubiquitin-proteasome system (UPS) [158]. The Wnt heteromeric complex with SMAD4 (co-SMAD) [121, ligand-mediated destabilization of the β-catenin destruc- 122]. This ternary protein complex translocates to the nu- tion complex leads to the accumulation of activated β- cleus where it recognizes SMAD-binding elements (SBE) catenin (unphosphorylated). Accumulated cytoplasmatic in the DNA to regulate the expression of diverse target β-catenin subsequently translocates to the nucleus and genes [123, 137]. In parallel, SMAD6 and SMAD7 act as associates with DNA-binding T-cell factor (TCF) or inhibitors (also called I-SMADs). Their model of action lymphoid enhancer factor (LEF)–TCF/LEF− transcrip- can be various (via TGFBR1 or SMAD4), but their activa- tion factors (TFs) [159]. The binding of β-catenin and tion is often a result of a negative feedback loop aiming to TCF/LEF recruits transcriptional partners and chromatin downregulate the TGF-β or the BMP signaling pathway remodeling complexes to regulate the expression of [138–140]. TGF-β also activates non-canonical down- TCF/LEF target genes [160, 161]. stream signaling pathways such as ABL, PI3K-AKT, RHO, Despite the increasing knowledge about the Wnt sig- TAK1, ERK1/2, JNK, and p38-MAPK [124, 141]. In vitro naling pathway, the participation of Wnt proteins and and in vivo experiments suggest that both canonical and signaling in modulating FAP fate has not been investi- non-canonical TGF-β pathways are involved in fibroblast gated until recently. Skeletal muscle SCA-1+ cells (FAPs) proliferation and myofibroblast differentiation, and are abundant in the muscles of the mdx mice (model of thereby modulate TGF-β-induced fibrosis and ECM re- the Duchenne Muscular Dystrophy (DMD)), and modeling [11, 12, 120, 126, 128, 141–143]. However, the WNT3a treatment promotes their proliferation and col- specific role of TGF-β canonical and non-canonical path- lagen expression both in vitro and in vivo [162]. Interest- ways in regulating muscle-resident PDGFRα+ FAP plasti- ingly, the treatment of dystrophic mice with DKK1 city and fate remains underexplored. (Dickkopf 1, a WNT inhibitor) reduced β-catenin pro- In response to muscle injury, TGF-β is produced and tein levels and muscle fibrosis [162]. On the other hand, secreted by macrophages, FAPs, and regenerating myofi- increased canonical Wnt/β-catenin signaling regulates bers [11, 12, 59, 144, 145]. Muscle FAPs express the satellite cell fate and fibrogenic commitment via cross- three TGF-β isoforms (TGF-β1, TGF-β2, and TGF-β3) talk with TGF-β2 in dystrophic mdx muscles [163]. Ac- and TGF-β receptors (TGFBR1, TGFBR2, and TGFBR3) cordingly, we also observed increased β-catenin protein [11, 12]. TGF-β ligands through TGFBRs induce FAP- levels upon acute glycerol muscle injury [28]. Xiang and myofibroblast differentiation and ECM production [11, colleagues showed that the conditional genetic loss of β- 12, 144, 146]. In addition, TGF-β inhibits the adipogenic catenin in heart fibroblast (Transcription factor 21 priming of muscle FAPs [11], and is pro-mitogenic, and (TCF21) + cells) and activated fibroblasts and myofibro- hence, stimulates the proliferation of PDGFRα+ FAPs blasts (Periostin+) lineages reduces fibrosis and amelio- [11, 12, 59, 144] (Fig. 2). TGF-β signaling pathway acti- rates cardiac hypertrophy induced by pressure overload vation also seems to be required for FAP survival since [164]. In agreement, the sole transgenic overexpression the in vivo treatment of mice with SB431542—a selective of canonical WNT10B is sufficient to induce fibrosis and potent ALK4, ALK5, and ALK7 receptor inhibitor— in vivo [165]. Overall, the Wnt/β-catenin pathway regu- reduced the number of expanded FAPs following rotator lates the expression of several ECM genes in fibroblasts cuff tear injury [147] (Table 3). Remarkably, we also from different tissues and organs following injury and showed that TGFBR1 and the p38-MAPK protein are re- disease [28, 164–166]. sponsible for TGF-β-mediated downregulation of PDGF The outcomes of Wnt/β-catenin signaling depend on Rα [11], associated with a decrease in TCFL2 expression the TCF/LEF TFs. However, the potential roles of them Contreras et al. Skeletal Muscle (2021) 11:16 Page 11 of 25 Table 3 Summary of drug strategies to target muscle fibro-adipogenic progenitor differentiation and fate Therapy Target Cell Proliferation Cell death/ Fibrogenesis Adipogenesis References survival apoptosis AG1296 PDGFR kinase activity Not Reduced? Not evaluated Reduced Not evaluated [11] inhibitor evaluated AICAR AMPK activator Reduced Not Induced Not Reduced [148] evaluated evaluated Azathioprine Immunosuppressant Not Reduced Not affected Not affected Reduced [149] affected Batimastat MMPs inhibitor Not Not affected Not affected Not affected Reduced [14][104] (including MMP14) affected BMS493 Pan-retinoic acid Not Reduced Not evaluated Reduced Induced [69] receptor (RAR) inverse evaluated spontaneous agonist differentiation Dexamethasone Glucocorticoid Induced Induced Not affected Not Induced [150] receptor evaluated HDAC inhibitors (TSA and HDACs Not Not Not evaluated Reduced Reduced [28][151] Pracinostat) evaluated evaluated [152]; LY2090314 & other GSK GSK3 inhibitors Slightly Not affected Not affected Mixed results Reduced [153] inhibitors decreased Metformin AMPK activator Not Reduced Not evaluated Not Reduced [16] evaluated evaluated Molsidomine NO donating molecule Reduced? Reduced? Not evaluated Reduced Reduced [154] Promethazine hydrochloride H1 histamine receptor Not Not affected Not affected Not Reduced [72] affected evaluated SB525334/SB431542 TGFBR kinase activity Reduced Reduced Induced after Reduced Not evaluated [11, 12][147] inhibitor long treatment TKIs (imatinib, nilotinib, Abl, PDGFRs, Kit, DDRs, Reduced Reduced Induced Reduced Reduced and/or [11, 12]; [16]; crenolanib, sorafenib, and p38 Induced [59]; [155]; masitinib) [146]; HDACs-mediated effects on FAP fate are seen only in young mdx but not aged mdx mice [16] reported that imatinib enhances the amount of perilipin+ FAP-derived adipocytes in vitro in muscle FAPs are underexplored. These TFs recognize idiopathic pulmonary fibrosis and heart fibroblasts [28]. TCF/LEF-binding elements and regulatory regions of Hence, our work confirms the cross-talk between the target genes to regulate gene expression. In this context, Wnt and TGF-β pathways that controls the fate of we showed the expression of the four Wnt TCF/LEF PDGFRα+ cells and potentially fibrosis (Table 3). In members in MSC and fibroblast cell lines, as well as summary, the Wnt cascade modulates TGF-β-mediated tissue-resident FAPs from skeletal muscle and cardiac effects in fibroblasts, and vice versa [28, 167–169]. tissues [28]. We observed that Tcf7l2 and Tcf7l1 were the two most highly expressed members, whereas the Platelet-derived growth factor signaling fibroblast lineage, including FAPs, express Tcf7 and Lef1 The platelet-derived growth factor (PDGF) signaling at lower levels. Moreover, treatment with TGF-β de- pathway regulates not only vascular development and creases both the mRNA and protein levels of TCF7L2 in angiogenesis [170] but also plays crucial roles during de- PDGFRα+ cells. We described that this regulatory mech- velopment, stem cell fate, migration, and proliferation. anism requires the transcriptional regulation activity of PDGF receptors (PDGFRs) are the cell membrane- histone deacetylases (HDACs) and the participation of bound tyrosine kinase receptors for PDGF ligands [171– the UPS [28]. 174]. PDGFs were initially described as serum-derived Interestingly, TGF-β activates the canonical Wnt/β-ca- mitogens essential for fibroblast and smooth muscle cell tenin cascade and induces nuclear accumulation of β- growth [175, 176]. PDGFs ligands are four gene products catenin, which in turn reduced the expression of the consisting of five dimeric isoforms: the homodimers WNT inhibitor DKK1 [165]. In agreement with our PDGF-AA, PDGF-BB, PDGF-CC, PDGF-DD, and the most recent results showing that TGF-β reduces the ex- PDGF-AB heterodimer [177]. PDGFs are known for be- pression of several TCF7L2 target genes, whereas it pro- ing released from α-granules of platelets and are potent motes the expression of ECM remodeling genes in chemoattractants and mitogens for cells of mesenchymal Contreras et al. Skeletal Muscle (2021) 11:16 Page 12 of 25 origin [178]. However, several other cell types express proliferative and differentiation-related downstream sig- and secrete these ligands, such as inflammatory cells naling pathways such as PI3K-AKT, ERK1/2, p38- (e.g., macrophages) and fibroblasts [179]. Post- MAPK, and STAT3 in PDGFRα expressing cells [11, translational proteolytic processing of PDGFs is neces- 12]. Recently, Farup et al. showed that PDGF-AA treat- sary for their activation. It occurs extracellularly for ment increases the expression of collagen type I in FAPs, PDGF-C and PDGF-D but intracellularly for PDGF-A, whereas it reduces their adipogenic differentiation (Fig. PDGF-B, and PDGF-AB [178, 179]. A biologically active 2). Notably, the PDGF-AA-mediated fibrogenic fate of PDGF ligand is a dimer of two single PDGF chains, FAPs associates with a metabolic switch that promotes which binds one PDGFR. enhanced glucose consumption [16]. Hence, PDGF sig- PDGFRs genes (PDGFRA and PDGFRB) encode naling could regulate the potency and fate of skeletal single-pass transmembrane receptors with an extracellu- muscle FAPs (Fig. 2). lar portion of five immunoglobulin-like domains, a In the heart, Asli et al. showed that PDGF-AB treat- transmembrane segment, a juxtamembrane segment, a ment promotes colony formation and self-renewal of tyrosine kinase domain, and a carboxy-terminal tail cardiac fibroblast, whereas the PDGFR inhibitor, [180]. PDGFRs are monomeric before exposure to PDGF AG1296, suppressed these activities [186]. Interestingly, [181]. Its ligand binding-induced dimerization causes activated PDGFRαH2BEGFP-mid fibroblasts formed at their activation, and therefore, later PDGFR de- the expense of resting PDGFRαH2BEGFP-high fibro- repression and activation of the receptor's tyrosine kin- blasts [73, 186]. These results are in agreement with our ase activity [180, 182, 183]. Three known functional recent findings where the expression of PDGFRα dimer forms of the receptors exist. They consist of the changes dynamically during muscle regeneration and re- PDGFRα/α and PDGFRβ/β homodimers and the PDGF pair [11]. Moreover, in vivo PDGF-AB treatment of un- Rα/β heterodimer [179, 180]. PDGF-AA, PDGF-AB, injured hearts did not cause fibroblast activation; PDGF-BB, and PDGF-CC promote PDGFRα/α homodi- however, it increased the number of PDGFRαH2BEGFP- mer formation, PDGF-BB, PDGF-CC, PDGF-DD, and mid fibroblasts after myocardial infarction [186]. There- PDGF-AB promote PDGFRα/β heterodimer assembly. fore, PDGF-AB isoform targets tissue-resident fibroblasts PDGFRβ/β homodimer can only be induced by PDGF- by increasing the activated fibroblast pool after injury. BB and PDGF-DD isoforms [177, 178, 180]. Although Interestingly, genetic loss of Pdgfra in the resident car- the precise role of PDGF and its receptors in vivo in diac fibroblast lineage (TCF21+ cells) results in an over- muscle-resident FAPs is unknown, PDGF signaling all reduction in the fibroblast population in adult hearts, seems to regulate FAP survival, activation, proliferation, demonstrating that PDGFRα regulates fibroblast main- migration, and fate. In this review, we focused on PDGF tenance and homeostasis [187]. Consistently, lineage- ligands and PDGFRα in skeletal muscle health and specific deletion of Pdgfra in tubulin polymerization- pathophysiology. promoting protein family member 3 expressing cell Treatment of ex vivo FAPs with PDGF-AA and PDGF- population (Tppp3+ tendon stem cells) caused impaired BB ligands activates the PDGF cascade inducing FAP ac- tendon regeneration, and therefore, corroborates the cell tivation and proliferation (Fig. 2)[11, 53]. In addition, requirements of PDGFRα signaling for proper tendon upregulated expression of ECM genes and activated healing [188]. Remarkably, the passaging of plastic ad- downstream ERK1/2, PI3K-AKT, and SMAD2/3 signal- herent FAPs obtained from muscles reduces the protein ing pathways is observed in ex vivo FAPs in response to levels of PDGFRα, which associates with their differenti- PDGF-AA treatment [53, 184]. By utilizing a pharmaco- ation [12]. Thus, cellular PDGFRα bioavailability may be logical inhibitor of PDGFR signaling, Mendias and col- a modulating factor in PDGF-mediated responses of leagues showed that PDGFR signaling modulates muscle FAP lineage during survival, fate decisions, and damage- ECM remodeling and angiogenesis upon synergist abla- associated behaviors. tion surgery to induce postnatal muscle growth or hypertrophy [185]. In addition, the treatment with Adipogenic potential of PDGFRα+ FAP cells PDGF-AA induces the phosphorylation of PDGFRα and Infiltration and deposition of fatty adipose tissue are the proliferation of PDGFRα+ cells (Fig. 2)[53]. The au- hallmarks of several skeletal muscle pathologies. How- thors also suggested, using pharmacological inhibitors, ever, the cellular and molecular mechanisms underlying that both PI3K-Akt and MEK2-MAPK signaling path- fatty infiltration of muscles have not been extensively in- ways are necessary for PDGFRα-induced proliferation vestigated compared with the ever-growing research in [53, 54]. However, persistent PDGF ligand exposure and muscle fibrosis. A better understanding of such a enhanced PDGFRα signaling levels can cause patho- discrete fat compartment between myofibers and fascia, logical muscle fibrosis [53, 54, 155, 184]. We have re- also called intra/intermuscular adipose tissue (IMAT), cently shown that PDGF-BB treatment activates Contreras et al. Skeletal Muscle (2021) 11:16 Page 13 of 25 may allow for the targeting of these adipogenic progeni- to multiply and hypertrophy [195, 196]. On the other tors to increase muscle regeneration and repair. hand, brown fat cells are equipped with smaller droplets The lack of reliable cell-specific markers for fat pre- and large mitochondria concentration, giving the tissue cursor cells has been the main limitation of studying its chestnut hue. Hence, in brown adipose tissue, mito- IMAT. As described above, the studies of Joe et al. and chondria produce heat using these fatty droplets, a Uezumi et al. helped to clarify many aspects of the process also known as thermogenesis [197]. The role of muscle adipogenic precursor cells. One major focus of FAP-derived fat cells, whether brown/beige or white, in these approaches was determining whether IMAT- skeletal muscle health, regeneration, and disease is associated adipocytes were in vivo derived from pre- unknown. existent muscle-resident PDGFRα+ cells, other muscle- Perhaps the most serious disadvantage of these studies resident cells, or circulating cells. The work led by Liu is that they do not directly address the in vivo adipo- et al. in murine skeletal muscle is a classic example of genic differentiation potential of adult PDGFRα+ FAPs. these efforts. The authors suggested that IMAT derives The definitive proof that muscle PDGFRα+ cells are the from a lineage of cells not expressing Pax3 (i.e., non- main, if not the only, source of injury-induced adipo- myogenic). They also showed that the genetic ablation cytes came from lineage tracing experiments using CreERT EYFP of intramuscular adipogenic progenitors based on Ap2 Pdgfra :Rosa26 transgenic mice [14]. The au- (also known as fatty acid-binding protein 4 (FABP4)) ex- thors demonstrated that seven days after acute intramus- pression leads to impaired skeletal muscle regeneration, cular injury, a large proportion of perilipin+ adipocytes suggesting for the first time that damage-induced fatty derived from PDGFRα+ FAPs, indicating that PDGFRα tissue may support efficient regeneration upon acute in- expressing progenitors are the major source of damage- jury [189]. However, AP2/FABP4 expression is com- induced fat cells in normal muscle regeneration and in monly thought to be restricted to committed or muscular dystrophy. Indeed, using similar lineage tracing differentiated adipocytes than progenitor cells [190], strategies we have demonstrated that cardiac PDGFRα+ questioning the interpretation of the results. Marinkovic FAPs can cause fibrofatty infiltration within the myocar- et al. [111] showed that Notch signaling is a pivotal dium in an arrhythmogenic cardiomyopathy mouse pathway regulating FAP adipogenesis in wild-type cells model driven by the conditional deletion of the and that dystrophic FAPs are insensitive to Notch- quiescence-associated factor Hic1 in heart FAPs [73]. mediated adipogenic inhibition compared with acute Intriguingly, PDGFRα+ FAPs are ciliated cells and thus injury-derived FAPs [111]. Hence, these results demon- possess primary cilium. Conditional deletion of a gene strate that wild-type and dystrophic muscle PDGFRα+ required for ciliogenesis, Ift88, in FAPs impaired the FAPs are in different functional states, which influences injury-induced formation of adipocytes [14]. Mechanis- their fate and responsiveness to extracellular cues, as tically, the cilia-dependent modulation of FAP adipogen- previously suggested [62]. esis involves the participation of Sonic Hedgehog (SHH) Human PDGFRα+ FAPs exist in healthy and DMD signaling, which is repressed in the absence of cilia. In- pathological muscles, being bona fide counterparts of deed, constitutive activation of the Shh-pathway via gen- the PDGFRα+ cells found in mouse muscles [16, 50, 51, etic deletion of the repressor Ptch1 was sufficient to 53, 54, 109]. Remarkably, FACS-isolated human FAPs block adipocytes’ generation following injury [14]. Re- (CD15+/PDGFRα+/CD56−) differentiate towards fully markably, elimination of the primary cilium in PDGF mature adipocytes, phenocopying the in vitro differenti- Rα+ FAPs led to enhanced regeneration of myofibers by ation kinetic and potential of adipose stromal cells ob- reducing fatty degeneration of dystrophic muscles, tained from subcutaneous adipose tissue depots [51]. which was also associated with increased myofiber Moreover, when transplanted into a glycerol-damaged size. The authors also showed that tissue inhibitor of muscle (an injury model that promotes adipogenesis) metalloproteinase 3 (TIMP3), an ECM modifier, in- [191, 192], murine FAPs readily differentiate into adipo- hibits adipocyte formation by muscle PDGFRα+pro- cytes. In concordance to the in vitro report of Liu and genitors. Interestingly, aiming to mimic TIMP3 colleagues using mouse muscle samples, Arrighi et al. activity, the authors utilized batimastat and showed also showed that FAP-derived adipocytes from human that the treatment with this pharmacological inhibitor muscle biopsies are white rather than beige/brown fat of metalloproteinases prevented injury-induced adipo- cells. In contrast, Gorski et al. showed increased expres- genesis in vivo [14](Table 3). sion of UCP1, a brown/beige fat cell marker [193], in In a different study, Jaiswal and colleagues showed that muscle as well as in FAP cultures following induction of the treatment with Batimastat prevented FAP spontan- IMAT by glycerol injection [194]. Throughout the body, eous adipogenesis and reduced fat in dysferlinopathic white fat cells store energy in large, often single, oily muscle of dysferlin-deficient (B6A/J) mice [104]. Hence, droplets. Obesity causes these white adipose tissue cells the authors suggested that the accumulation and Contreras et al. Skeletal Muscle (2021) 11:16 Page 14 of 25 adipogenic differentiation of FAP are critical contribu- Feeley and colleagues showed that rotator cuff tears en- tors to limb-girdle muscular dystrophy type 2B. Surpris- hanced HDAC activity in FAPs and trichostatin A inhib- ingly, the authors observed no changes in either FAP ited it. HDAC inhibition prevented FAP-mediated fatty accumulation, proliferation, or fibrosis as a result of bati- infiltration in supraspinatus muscles. Also, trichostatin mastat treatment (Table 3). Nevertheless, the batima- A regulates muscle FAP adipogenesis by promoting FAP stat’s off-targets on other tissue-resident cells such as browning (Table 3)[152]. MuSCs, myofibers, endothelial cells, pericytes, or infil- These studies demonstrated that HDACs-mediated trating CD45+ cells have not been evaluated yet. pharmacological intervention might counter DMD pro- Altogether, these findings suggest novel strategies to gression and chronic muscle injury by increasing regen- combat fatty degeneration of chronically damaged mus- eration by inhibiting fibro-fatty degeneration while cles by targeting the adipogenic conversion of PDGFRα favoring the interplay and communication between FAPs expressing FAPs to inhibit the deposition of injury- and and myogenic progenitors. Recently, we have shown that disease-induced intramuscular fat. two well-characterized pan-HDACi reduce TGF-β- To date, there is not a single clinically approved drug induced ECM gene expression and also block TGF-β- used to prevent IMAT accumulation in muscle disease. mediated downregulation of Tcf7l2 expression [28]. However, significant pre-clinical advances have been Mechanistically, histone deacetylase inhibitors modulate made. In vivo treatment of mdx mice with molsido- TGF-β-mediated changes in the expression of TCF7L2 mine—a nitric oxide (NO) donating molecule—reduced transcription factor target genes of the Wnt pathway muscle pathology, IMAT accumulation, and fibrosis [28]. Further investigations should unravel the mechan- [154]. These improvements were at least in part medi- ism by which HDACs regulate the fate of FAPs and how ated by the inhibition of NO-mediated FAP adipogenesis could this be used to target muscle-associated diseases. (Table 3). Hence, altered synthesis of NO, a typical find- In a recent study, Reggio and colleagues used a large ing in DMD, could contribute to enhanced fat depos- drug library screen with pharmacological approaches to ition. On the search for adipogenic inhibitors, Uezumi demonstrate that the inhibition of the cytoplasmic sig- and colleagues found that promethazine hydrochloride naling protein, glycogen synthase kinase 3 (GSK3), re- inhibits, through binding to the H1 histamine receptor, duces PDGFRα+ FAP adipogenesis in vitro, while also the in vitro and in vivo formation of ectopic adipocytes repressing muscle glycerol-induced fatty degeneration derived from PDGFRα+ lineage cells in the muscle [72] [153] (Table 3). GSK3 is composed of 2 isoforms (α and (Table 3). Promethazine hydrochloride is a first- β) and is part of the destruction complex of β-catenin, generation antagonist of the H1 histamine receptor, and which we showed earlier to play a modulatory role in therefore, this family of drugs emerge as attractive novel FAP fate (see the “Wnt/β signaling” section). Mechanis- therapeutics against ectopic fat formation in muscle tically, the authors suggested that UPS-targeted β- pathologies. catenin degradation causes an imbalance in the adipo- Histone deacetylation leads to the repression of gene genic fate of dystrophic mdx FAPs. The authors also expression, and histone deacetylase inhibitors (HDACi, exploited single-cell data and in silico modeling to show like trichostatin A) provide an exciting means to treat that PDGFRα+ FAPs compose the core of the stromal DMD. HDACi have been used in both pre-clinical and cells in the muscle cell niche by expressing Wnt compo- clinical studies to improve muscle regeneration and re- nents and also for being the primary source of Wnt li- pair in DMD [151, 198–200]. As HDACi treatment in- gands. FAPs seem to actively communicate with hibits fibro-fatty differentiation of PDGFRα+ FAPs, it endothelial cells, tenocytes, and MuSCs through the reduces the dystrophic pathology through increasing production of Wnt ligands. Among the Wnt ligands, muscle regeneration [151]. Remarkably, in dystrophic they observed that dystrophic FAPs downregulate FAPs, an HDAC–myomiR–BAF60 molecular network Wnt5a expression compared with wild-type cells. regulates FAP fate, and old FAPs become resistant to Moreover, WNT5a treatment reduced FAP-induced HDACi-induced chromatin remodeling compared with adipogenesis in vitro by repressing PPARγ expression young FAPs [201]. Also, HDACi restore the dystrophic- throughout the activation of β-catenin, suggesting mediated loss of intercellular communication between that the Wnt signaling modulates the adipogenic PDGFRα+ FAPs and myogenic progenitors required for commitment of FAPs in dystrophic muscles (Fig. 2). proper muscle regeneration [151], and as recently sug- On the other hand, Zhao and colleagues recently de- gested through an extracellular vesicle-mediated transfer scribed that the supplementation of retinoic acid (RA) of miRNAs [200]. Interestingly, aging and DMD disease enhances the proliferation of FAPs at the expense of progression limit HDACi-mediated effects [151], which inhibiting their adipogenic and fibrogenic differentiation suggests that aging affects the fate of FAPs, as recently [69]. Additionally, treatment of isolated FAPs with a detailed by Lukjanenko and colleagues [5]. Recently, pan-retinoic acid receptor antagonist, BMS493, blocked Contreras et al. Skeletal Muscle (2021) 11:16 Page 15 of 25 the RA-mediated effects. Notably, the authors also Adiponectin in skeletal muscle FAPs [11, 12]. We also showed that RA treatment rescued obesity-impaired showed that the adipogenic differentiation of FAPs re- skeletal muscle regeneration. These findings showed a presses the expression of PDGFRα [11]. Taken together, FAP-type specific effect of RA signaling that regulates these studies demonstrate that IMAT-associated adipo- skeletal muscle regeneration and repair by means of pre- cytes can derive from pre-existent muscle-resident fibro- serving their progenitor state. Taken together, these adipogenic progenitors. findings suggest a novel potential retinoic acid-based strategy to combat chronic skeletal muscle fibro-fatty Osteogenic differentiation of PDGFRα+ FAP cells degeneration of obese patients. Muscle PDGFRα+ FAPs have osteogenic potential On the contrary, several factors positively regulate in vitro [2] and when transplanted can successfully en- muscle FAP adipogenesis. For instance, the matricellular graft and form calcification-rich structures using an protein CCN family member 1 (CCN1/CYR61) is ele- in vivo heterotopic ossification (HO) model [208]. HO is vated in the serum and sarcopenic muscles of a murine a musculoskeletal disorder distinguished by the patho- model of chronic kidney disease and induces FAP adipo- logic formation of extraskeletal bone in muscle, tendon, genesis [202]. In vivo treatment of mice with the gluco- ligaments, and fascia [209]. BMP2 promotes intramuscu- corticoid dexamethasone enhanced IMAT deposition lar HO regardless of damage; however, BMP9-induced following acute injury (Dong et al., 2014 [150]). Dexa- HO requires skeletal muscle injury [210] (Fig. 2). The methasone also induces FAP proliferation while increas- authors described that intramuscular HO might involve ing their adipogenesis, possibly involving the reduction a population of Lin-SCA-1+ cells—likely FAPs [210]. of IL-4 expression (Dong et al., 2014 [150]). Remarkably, Moreover, Lin /TIE2+/PDGFRα+ progenitors respond IL-4 administration reduces dexamethasone-induced to BMP2-stimulated osteogenic commitment and con- FAP-derived adipocyte formation, suggesting a novel tribute to HO in mice [211]. Additionally, muscle- therapeutic use of IL-4 to reduce IMAT accumulation derived MSCs contribute to fracture repair in a tumor due to glucocorticoid use in DMD patients (Fig. 2). Per- necrosis factor-alpha (TNFα) dependent manner [212]. petuini et al. showed that the glucocorticoid-related The above findings are consistent with a recent study of molecules, dexamethasone, and budesonide, inhibited Goldhamer’s group, where the authors employed and the insulin-induced adipocyte formation from mdx- characterized a transgenic mouse model that recapitu- derived FAPs. However, both drugs have a pro- lates a rare autosomal-dominant disorder called fibro- adipogenic impact when the adipogenic mix contains dysplasia ossificans progressiva (FOP), which results factors that increase the concentration of cyclic AMP. from a single activating mutation in ACVR1; the type I The authors also showed that, only in anti-adipogenic BMP receptor also known as ACVR1/ALK2. The Tie2- conditions, budesonide suppresses the expression of driven expression of the mutation Acvr1 R206H is suffi- Pparg, a master adipogenic regulator, via the cient to phenocopy the spectrum of HO observed in glucocorticoid-induced-leucine-zipper (GILZ/TSC22D3), FOP patients [60]. Moreover, they also showed that and the glucocorticoid antagonist mifepristone alleviates intramuscular transplantation of mutant Acvr1R206H/+ such inhibitory effect [203] (Table 3). This study may FAPs into immunodeficient mice resulted in the forma- shed light on some of the mechanisms underlying the tion of HO in an Activin A-dependent fashion. Overall, use of glucocorticoids in DMD patients under this kind these data established TIE2+/SCA-1+/PDGFRα+ FAPs of treatment. The use of glucocorticoids to treat DMD as the predominant cell-of-origin and driver of patho- patients is so far the most common treatment available logical HO. However, it has been suggested that TIE2 is to delay muscle necrosis and degeneration up to date a nonspecific marker for a subset of PDGFRα+ cells [204–206]. Finally, the same group, using a similar since its expression overlaps with other cell populations chemical library-based approach, identified an immuno- like endothelial cells, MuSCs, and subsets of suppressant drug, azathioprine, that negatively perturbs hematopoietic cells [2, 213, 214]. Hence, the precise the intrinsic adipogenic fate, also via PPARγ repression, mechanisms and the populations of cells involved in the of wild type and mdx PDGFRα+ FAPs (Table 3). formation and remodeling of HO remained unknown On the other hand, we recently showed that TGF-β until then. We recently took advantage of a novel PDGF treatment negatively affects FAP differentiation to adipo- Rα lineage tracing reporter mouse (Pdgfrα-CreERT2- cytes while inducing FAP-to-myofibroblast commitment TdTomato) to further explore the cellular source of (Fig. 2). TGF-β1 impairs basal PDGFRα+ FAP differenti- muscle ossification [13]. Using a model of BMP2- ation into the adipogenic lineage, by reducing the stimulated intramuscular HO, we showed that a large steady-state percentage of adipocytes but increasing the proportion (~80%) of differentiated osteogenic cells were number of myofibroblasts [11, 207]. Mechanistically, TdTomato+ after 21 days of muscle injury. Thus, the TGF-β treatment reduces the expression of Pparγ and cell-source responsible for forming ectopic bone in Contreras et al. Skeletal Muscle (2021) 11:16 Page 16 of 25 muscle is a subpopulation of muscle-resident PDGFRα+ single-cell omics technologies to refine our understand- progenitors [13]. Overall, these studies demonstrate that ing of cell heterogeneity by using a plethora of genes FAPs are a significant cellular source of chondrogenic and proteins to identify a particular cluster or subpopu- cells and osteogenic cells in severely damaged muscles. lation of cells, they are significantly more accurate com- Remarkably, intramuscular calcium deposits serve as a pared with the use of a single marker to identify cell pathohistological feature of DMD [215]. Notably, the de- types. gree of osteogenic commitment of FAPs appears to The classical view of cellular muscle composition is match the model of muscle damage and degeneration/ that most of the non-myogenic cells play a positive role regeneration used. Using the severe D2-mdx (DBA/2J- and generate a pro-regenerative transitional niche, mdx) dystrophic mice, which better recapitulates the hu- which, among other functions, support MuSC-driven man characteristics of DMD myopathology, Mázala et al. myogenesis following acute damage [217]. These popula- demonstrated that PDGFRα+ FAPs accumulate within tions of non-myogenic cells include endothelial cells calcified deposits in degenerative muscles [117]. Also, (CD31+) [218, 219], FAPs (PDGFRα+) [1, 2], connective the in vitro osteogenic differentiation of these cells posi- tissue fibroblasts TCF7L2+ (significantly overlapping tively correlates with the degree and extension of muscle with FAPs) [6, 24, 27, 28], pericytes (NG2+, RGS5+) degeneration and TGF-β levels, which supports previous [48], mesoangioblasts [220, 221], tenocytes (TNMD+, studies showing that FAPs vastly expand and accumulate SCX+), glial cells (PIP1+, KCNA1+) [48, 58], and a com- accordingly with the extension of damage, TGF-β levels, plex array of immune cells [222–224]. and fibrosis [11, 12, 15, 27, 53, 54, 59, 103, 117]. In sum- With single-cell omics technologies, the transcriptional mary, FAP activity and responses are highly contextual, identity in homeostasis and lineage trajectories of which suggests that signals emanating from the local muscle-resident FAPs during the regenerative response niche determine their phenotypic multi-lineage-fate. and in disease states have started to be discovered. Male- Why are different muscle groups affected to a different cova and colleagues were the first to initially show the extent in muscular dystrophy or neuromuscular disor- existence of cellular heterogeneity in muscle FAPs using ders? Although several hypotheses might explain this, in- single-cell RT-PCR and showed that it increases in re- cluding muscle fiber type, muscle fiber innervation, generating muscles [62] (Table 4). The authors showed muscle of origin, calcium homeostasis, and muscle activ- that a specific subpopulation of vascular cell adhesion ity, we still lack information of the role that FAPs play in molecule VCAM-1+ FAPs vastly expands and drives these processes. muscle fibrosis in acute damaged muscles and adult dys- trophic muscles of the mdx mice [62]. Notably, the Fibro-adipogenic cell diversity: Single-cell omics VCAM-1+ subFAPs are absent in uninjured muscles, unveil stromal populations in muscles suggesting that Vcam1 may be an activation marker. Re- The recent revolution in single-cell omics technologies, cently, a population of quiescent HIC1+ mesenchymal including single-cell RNA sequencing (scRNAseq), progenitor cells has been described. This population single-cell epigenomics (e.i. scATACseq), and single-cell contains precursor cells for several mesenchymal line- mass cytometry (e.i., CyTOF) has helped to uncover the ages including muscle FAPs (Pdgfra+, Ly6a+), tenocytes mysteries of muscle cellular composition and heterogen- (Tnmd+), pericytes (Rgs5+), and a new subset of cells eity as well as to faithfully recreate a more precise cellu- called myotenocytes (Col22a1+) [48]. However, further lar atlas of murine and human adult skeletal muscle in functional analyses will be required to confirm if the homeostasis, regeneration, and repair [48, 58, 62, 106– myotenocyte cells present at the regenerated myotendi- 109, 111, 216] (Fig. 3). Muscle single-cell analyses faith- nous junction represent an independent subpopulation fully recapitulate key cellular events involved in skeletal of HIC1+ progenitors with a specific function or a differ- muscle regeneration and repair, derived from studies entiated cell state of tenocyte progenitors in the myoten- over many years. Such tools and information led us to dinous niche. Even though HIC1 is a broad stromal realize that a complex array of non-myogenic cells (tis- progenitor marker, FAP gene signatures segregate from sue-resident and non-tissue-resident) engage in active other interstitial populations of mesenchymal cells like cross-talk between each other and with MuSCs to re- pericytes and tenocytes [48]. Remarkably, the gene sig- store tissue function following damage. Single-cell stud- nature of muscle PDGFRα+ FAPs is heterogeneous and ies evaluate molecular signatures and expression levels progresses over time after acute muscle injury, which re- of genes or cell surface protein abundance in large num- veals their dynamic role in regeneration [48, 108]. bers of individual cells. They aim to describe at an un- Therefore, FAPs acquire a unique plastic transcriptome precedented resolution the total interstitial populations that changes as the inflammation progresses and damage of cells in a resting state and to understand their flux in resolves through regeneration. Concerning mouse mus- response to injury and disease. Owing to the ability of cles, two major FAP populations have been described Contreras et al. Skeletal Muscle (2021) 11:16 Page 17 of 25 Fig. 3 a Single-cell RNA sequencing analyses to map muscle-resident FAP mononuclear landscape in murine (left graph) and human (right graph) skeletal muscle tissue. Three different studies, utilizing mice, agree with the existence of at least two principal muscle FAP subpopulations (here shown as FAPs 1 and FAPs 2; see text for details). On the other hand, FAP clustering and FAP subpopulations greatly vary in human muscles. Two different bioinformatic techniques for the presentation of large scRNA-seq datasets and their dimensionality reduction are shown: uniform manifold approximation and projection (UMAP) algorithm and t-Distributed Stochastic Neighbor Embedding (t-SNE). Colored dots represent individual FAP cells. Dotted lines illustrate the different studies discussed in this review. b FAP cell trajectories are based on the gene signatures of single cells following damage [48, 108]. The transcriptomes of FAPs indicate high cellular heterogeneity within the FAP populations in response to injury. In mouse muscles, two major FAP subpopulations (Dpp4 FAPs and Cxcl14 FAPs) are present in homeostatic conditions (for detailed markers, see Table 4). Analysis of the pseudotime trajectory of different FAP subpopulations suggests that FAP cells follow a continuum and diverge into two major subclusters upon damage Contreras et al. Skeletal Muscle (2021) 11:16 Page 18 of 25 Table 4 scRNA-seq gene signatures used for FAP identification and clustering in muscle homeostasis Genes/markers FAP subpopulations Species Reference low low Sca1, Cd34, Pdgfra SubFAPs: Tie2 (Tek), Vcam1 Mouse [62] high high SubFAPs: Tie2 (Tek), Vcam1 Ly6a (Sca1), Ly6e, Pdgfra, Dcn Not determined Mouse [58] Pdgfra, Ly6a (Sca1), Hic1 FAP1: Cxcl14, Col4, Col6, Col15, Lum, Sparcl1, Podn, Smoc2, Mgp, and Bgn Mouse [48] FAP2: Dpp4, Sfrp4, Igfbp5, Sema3c, Tgfrb2, and Wnt2 Pdgfra, Ly6a (Sca1), Dcn, Cd34 Not determined Mouse [216] Pdgfra, Col3a1, Dcn, and Gsn Not determined Mouse [107] Pdgfra, Ly6a (Sca1), Cd34 FAP1: Cxcl14, Enpp2 (Autotaxin), Crispld2, Hsd11b1, Smoc2, Ccl11, Gsn, and Dcn Mouse [108] FAP2: Dpp4, Pi16, Wnt2, Igfbp5, Igfbp6, Fbn1, and Ugdh PDGFRA, CD34, COLLAGEN 1, LUMICAN (LUM) FAP: LUM, DCN, CXCL14, COLLAGEN 4, and COLLAGEN 15, SMOC2, and GSN Mouse [109] COLLAGEN 3, and COLLAGEN 6 and FIBRILLIN 1 (FBN1) FAP: FNB1, MFAP5, LOXL1, PRG4, ELN, IGFBP5, and FSTL1 human PDGFRA FAP1 (fibroblasts 1): COL1A1, SFRP4, SERPINE1, and CCL2 Human [106] FAP2 (Fibroblast 2): FBN1, MFAP5, and CD55 FAP3 (Fibroblast 3): SMOC2, ADH1B, and ABC18 PDGFRA, CD34, COL1A1, COL6A3, FAP1: PCOLCE2, MFAP5, IGFBP6, ENNP1, CD55, and AXL Human [16] TCF7L2 FAP2: LUM, MYOC, CCL2, ADH1B, SFRP2, CXCL14, and MGP FAP3: TNXB, C3, COL15A1, SMOC2, ABCA8, COL6A1, and HMCN2 FAP4: IGF1, CRLF1, SCN7A, ITIH5, PTGDS and NOV FAP5: SEMA3C, PRG4, DEFB1, CCDC80, LINC01133, and IGFBP5 By re-clustering the FAP population, the authors described the existence of 7 different FAP subpopulations in human muscles [16] (FAP1 and FAP2) in undamaged muscles (Fig. 3a). FAP1 expressing chemokine genes like Cxcl5, Cxcl3, Ccl7, and associates with the ECM gene signatures, such as colla- Ccl2) at 0.5 and 2-day post-injury (DPI), then progres- gens Col4, Col6 and Col15, Lum, Sparcl1, Podn, Smoc2, sing into WISP1+ FAPs at 3.5 and 5 DPI (highly ex- Mgp, Cxcl14, and Bgn. On the other hand, the FAP2 pressing Postn, Csrp2, Sfrp2, Ptn, Cilp, and Cthrc1), subpopulation expresses genes involved in cell signaling followed by DLK1+ FAPs at 10 DPI (expressing Itm2a, and migration, including Sfrp4, Igfbp5, Sema3c, Dpp4, B830012L14Rik, Meg3, Airn, Peg3, Zim1, H19, and Igf2), Tgfrb2, and Wnt2 [48] (Table 4). In contrast, another and finally two FAP subpopulations at day 21 DPI, study described no segregation of the FAP population, OSR1+ (expressing Gsn, Ccl1, Bmp4, Bmp5, and Wnt5a) primarily based on the expression of a few FAP markers and fibroblast FAPs (expressing Col3a1, Col1a1, Col1a2, [216]. These subtle differences might be explained by a Col6a3, and Meg3)[108] (Table 4 and Fig. 3b). Notably, number of technical factors such as muscle groups used the authors showed that a proportion of the OSR1+ for scRNAseq, tissue digestion and cell-type enrichment FAPs at 21 DPI diverge into the two populations ob- methods, single-cell RNA sequencing platform—al- served in undamaged muscle: DPP4+ FAPs and though most of them used Chromium 10× Genomics, CXCL14+ FAPs (Fig. 3b). Therefore, the gene expression the number of cells captured and recovered after se- of single-cell FAPs is highly diverse, representing a con- quencing, and downstream data processing (e.g., version tinuum state during skeletal muscle regeneration (Fig. of Seurat R package used) and interpretation. 3b). Owing to the high degree of FAPs diversity, we Oprescu et al. [108] reported that murine CXCL14+ speculate that FAP subpopulations have adapted to play FAPs (also expressing Smoc2, Ccl11, Gsn, and Dcn) and supportive and distinct roles during regeneration. These DPP4+ (expressing Pi16, Igfbp5, Igfbp6, Fbn1, and Ugdh) data also suggest that the transcriptional diversity of FAPs represent two different FAP subtypes present in PDGFRα+ FAPs at the single-cell level might reflect non-injured murine tibialis anterior muscle, suggesting their differential developmental potential. that FAPs could represent two distinct subpopulations In addition to these studies, Rubenstein et al. [109] de- of interstitial cells in resting conditions, as it was previ- scribed two human FAP subpopulations as LUMICAN ously shown by us [48] (Table 4). However, in response (LUM)+ FAP and FIBRILLIN 1 (FBN1)+ FAP subtypes to injury, the two populations follow a linear trajectory (Table 4 and Fig. 3a). Interestingly, both FAP subpopula- into a single population of activated FAPs (highly tions showed specific differences in the expression of Contreras et al. Skeletal Muscle (2021) 11:16 Page 19 of 25 Collagen types. The authors also validated the existence their fate and plasticity within muscles. Although of these two distinct FAP subtypes by using scRNAseq there has been encouraging progress in understanding of mouse quadriceps and diaphragm muscles [109]. FAP phenotypic variability and activities, future re- ECM gene expression was also consistent among mouse search should look to translating this knowledge into and human muscles, with COLLAGEN 1, COLLAGEN 3, efficient medical applications. and COLLAGEN 6 broadly expressed across FAP subpop- ulations. In contrast, LUM+ FAPs express COLLAGEN 4, Future perspectives and COLLAGEN 15 predominantly compared with Skeletal muscle requires a complex orchestra of special- FBN1+ FAPs (Table 4). Interestingly, the authors reported ized populations of cells to perform its crucial functions. differences in the gene expression of the precursor gene of The origin, behavioral activities, lineage potency, and ex- TIE2 protein among the two species, which was expressed pression of markers associated with stem or progenitor only in the FBN1+ FAP subtype found in mouse muscle cell states define these specialized cell types. Here, we but none of the FAP subtypes found in human muscles. have focused on the unappreciated role of PDGFRα+ The meaning of these subtle differences in gene expres- FAPs in muscle biology, health, structure, and regener- sion across mouse and human muscle FAP subtypes ation. Apart from the accepted structural part that the should be address in future research. connective tissue provides for proper muscle develop- De Micheli and collegues collected and integrated ~ ment, the complex cues and matrix that stromal cells 22,000 single-cell transcriptomes generating for the first produce are essential to sustain myogenesis and support time a consensus cell atlas of human skeletal muscles proper muscle morphogenesis. FAPs are implicated in [106] (Table 4 and Fig. 3a). The authors described three muscle scarring, disease, and pathology. Although sub- subpopulations of fibroblasts (likely FAPs) in which stantial progress has been made in understanding FAP COLLAGEN 1, SFRP4, SERPINE1 and CCL2 are highly behavior, they remain poorly characterized, and the rela- expressed by fibroblast 1; FBN1+, MFAP5, and CD55 are tionships with other stromal cells are not well under- expressed by fibroblast 2, whereas fibroblast 3 highly ex- stood. PDGFRα+ FAPs and their descendant lineages, presses SMOC2 [106, 107] (Table 4). including activated-fibroblasts/myofibroblasts, adipo- Recently, Farup and colleagues described 5 subpop- cytes, chondrogenic and osteogenic cells, modulate ulations of human muscle FAPs (Table 4 and Fig. 3a). muscle regeneration and repair. These plastic cells play However, by sub-setting the FAP population and re- broad roles as sentinels, stress sensors, immune regula- clustering, the number of clusters increased to 7. The tors, cellular hubs, and paracrine factories, which are still authors reported that the expression of THY1/CD90 under active research in multiple pathological settings. is enriched in cluster 4, whereas PDGFRA gene ex- As discussed above, lineage tracing technologies com- pression is broadly distributed among the FAP sub- bined with single-cell sequencing strategies should by- populations (Table 4). Remarkably, the CD90+ pass the significant limitations that historically subpopulation of FAPs is associated with increased prevented us from deconvolving the complexity of stro- fibro-fatty infiltration and seems to drive the muscle mal cell populations. The diverse fibroblast nomencla- degeneration found in obese and type-2 diabetes pa- ture has periodically led to confusing claims in muscle tients [16]. Although the composition of each cellular biology and ensuing turmoil in the literature. Thus, re- interstitial compartment changes dramatically after in- solving the regenerative vs. reparative dichotomy of jury and in disease settings in mice, there is no infor- muscle-resident mesenchymal progenitors, and distin- mation about how the FAP population behaves guishing true lineage heterogeneity from the diverse following injury or how degenerative diseases alter its functional states that these cells can dynamically and re- activities in humans. Nonetheless, we and others have versibly acquire remains a high-priority issue for the detected a diverse range of mesenchymal stromal cells field. Despite these various uncertainties, in this review, including quiescent subsets, which rapidly expand fol- we establish a baseline for the contribution of fibro- lowing injury and secrete cytokines modulating in- adipogenic progenitors to muscle development, homeo- flammation, trophic factors, and regenerative cues to stasis, regeneration, and repair. promote skeletal muscle maintenance, MuSC renewal, and regeneration. Conclusions In conclusion, further studies should focus on un- In this review, we document new insights about the vari- derstanding the mechanisms by which FAP cell het- ous properties of muscle-resident PDGFRα+ FAPs and erogeneity arises. We aim to understand the lineage discuss the current state of knowledge on their origins restriction of FAPs by gene regulatory networks and and lineage capabilities. Here we propose to define a cell epigenetic factors that, in combination with the ex- as FAP if they present the following characteristics: 1. trinsic effects of the spatial context could regulate Express PDGFRα at the gene and protein level. 2. It is Contreras et al. Skeletal Muscle (2021) 11:16 Page 20 of 25 located in the tissue's interstitium and behaves as a peri- Availability of data and materials All data generated or analyzed during this study are included in this vascular cell but not residing in the blood vessel cavity. published article. 3. It can form colonies in vitro. 4. Can differentiate into activated fibroblasts, adipocytes, chondrocytes, and oste- Declarations ocytes in vitro and in vivo. Ethics approval and consent to participate We illustrated their importance in maintaining proper Not applicable. muscle function and critical role during the onset and establishment of scarring in pathology and disease. Consent for publication Not applicable. Growing evidence shows that PDGFRα+ cells are hetero- geneous and act as signaling hubs by providing regenera- Competing interests tive cues and integrating these signals in the muscle The authors declare that they have no competing interests. niche. Thus, FAPs influence other populations of cells Author details within skeletal muscle and vice versa. Ultimately, by un- Developmental and Stem Cell Biology Division, Victor Chang Cardiac derstanding and manipulating the complexity and vari- Research Institute, Darlinghurst, NSW 2010, Australia. St. Vincent’s Clinical School, Faculty of Medicine, UNSW Sydney, Kensington 2052, Australia. ability of the stromal compartment, specifically the FAP Departamento de Biología Celular y Molecular and Center for Aging and lineage, we aim to develop novel therapeutics to treat Regeneration (CARE-ChileUC), Facultad de Ciencias Biológicas, Pontificia several scar-forming pathologies. It remains plausible to Universidad Católica de Chile, 8331150 Santiago, Chile. Biomedical Research Centre, Department of Medical Genetics and School of Biomedical foresee a future where clinical leaps could be made Engineering, University of British Columbia, Vancouver, BC V6T 1Z3, Canada. based on these cells and where severe muscle injury could be treated without prolonged myodegeneration Received: 22 January 2021 Accepted: 22 March 2021 and muscle malfunctioning. References Abbreviations 1. Joe AWB, et al. Muscle injury activates resident fibro/adipogenic progenitors aSMA: Alpha smooth muscle actin; BMP: Bone morphogenetic protein; that facilitate myogenesis. Nat Cell Biol. 2010;12(2):153–63 https://doi.org/1 CFU: Colony-forming unit; CDHs: Congenital diaphragmatic hernias; 0.1038/ncb2015. CT: Connective tissue; DMD: Duchenne muscular dystrophy; DTA: Diphteria 2. Uezumi A, et al. Mesenchymal progenitors distinct from satellite cells toxin A; DTR: Diphteria toxin receptor; ECM: Extracellular matrix; contribute to ectopic fat cell formation in skeletal muscle. Nat Cell Biol. FAPα: Fibroblast activation protein alpha; FAPs: Fibro-adipogenic progenitors; 2010;12(2):143–52 https://doi.org/10.1038/ncb2014. FOP: Fibrodysplasia ossificans progressiva; GSK: Glycogen Synthase Kinase-3; 3. Fiore D, et al. Pharmacological blockage of fibro/adipogenic progenitor HO: Heterotopic ossification; IL: Interleukin; IMAT: Intermuscular adipose expansion and suppression of regenerative fibrogenesis is associated with tissue; LGMD: Limb-girdle muscular dystrophy; Lox: Lysyl oxidase; impaired skeletal muscle regeneration. Stem Cell Res. 2016;17(1):161–9 MSCs: Mesenchymal stem cells; MCT: Muscle connective tissue; https://doi.org/10.1016/j.scr.2016.06.007. MuSCs: Muscle stem cells; NCCs: Neural crest cells; PDGFRα: Platelet-derived 4. Heredia JE, et al. Type 2 innate signals stimulate fibro/adipogenic growth factor receptor alpha; PDGFRβ: Platelet-derived growth factor progenitors to facilitate muscle regeneration. Cell. 2013;153(2):376–88 receptor beta; PDGF: Platelet-derived growth factor; PPFs: Pleuroperitoneal https://doi.org/10.1016/j.cell.2013.02.053. folds; OSR: Odd-skipped-related; Shh: Sonic Hedgehog signaling; Sca-1: Stem 5. Lukjanenko L, et al. Aging disrupts muscle stem cell function by impairing cell antigen-1; Tbx: T-box transcription factor; TIMP3: Tissue inhibitor of matricellular WISP1 Secretion from fibro-adipogenic progenitors. Cell Stem metalloproteinases 3; TCF21: Transcription factor 21; TGF-β: Transforming Cell United States. 2019;24(3):433–446.e7 https://doi.org/10.1016/j.stem.201 growth factor beta; UPS: Ubiquitin-proteasome system 8.12.014. 6. Murphy MM, et al. Satellite cells, connective tissue fibroblasts and their interactions are crucial for muscle regeneration. Development. 2011;138(17): Acknowledgements 3625–37 https://doi.org/10.1242/dev.064162. The authors acknowledge Yen Tran and Ralph Patrick for helpful suggestions 7. Wosczyna MN, et al. Mesenchymal Stromal Cells Are Required for to the single-cell omics chapter, and Lucas Rempel for his inputs improving Regeneration and Homeostatic Maintenance of Skeletal Muscle. Cell Rep. the review. Figures were created using Illustrator (Adobe Inc.) and Key- 2019;27(7):2029–2035.e5 https://doi.org/10.1016/j.celrep.2019.04.074. note (Apple Inc.) for macOS. 8. Roberts EW, et al. Depletion of stromal cells expressing fibroblast activation protein-α from skeletal muscle and bone marrow results in cachexia and anemia. J Exp Med. 2013;210(6):1137–51 https://doi.org/10.1084/jem.2 Authors’ contributions O.C. and M.T. drafted the review and figures. F.M.V.R. revised and reviewed 9. Uezumi A, et al. Mesenchymal Bmp3b expression maintains skeletal muscle the manuscript. All authors read and approved the final manuscript. integrity and decreases in age-related sarcopenia. J Clin Investig. 2021; 131(1) https://doi.org/10.1172/JCI139617. Funding 10. Santini MP, et al. Tissue-Resident PDGFRα+ Progenitor Cells Contribute to This work was supported by Comisión Nacional de Ciencia y Tecnología Fibrosis versus Healing in a Context- and Spatiotemporally Dependent CONICYT Beca Doctorado Nacional 2014 folio 21140378 “National Doctorate Manner. Cell Rep. 2020;30(2):555–570.e7 https://doi.org/10.1016/j.celrep.201 Fellowship”, by Centro Basal de Excelencia en Envejecimiento y 9.12.045. Regeneración (CONICYT-AFB 170005), and by the Victor Chang Cardiac 11. Contreras O, Cruz-Soca M, Theret M, Soliman H, Tung LW, Groppa E, et al. Research Institute to O.C.; by Fondation pour la Recherche Médicale (FRM, Cross-talk between TGF-β and PDGFRα signaling pathways regulates the 40248), by the European Molecular Biology Organization (EMBO, ALTF 115- fate of stromal fibro–adipogenic progenitors. J Cell Sci. 2019;132:jcs232157 2016), by the Association contre les myopathies (AFM, 22576), and by Mi- https://doi.org/10.1242/jcs.232157. chael Smith Foundation for Health Research (MSFHR, 18351) to M.T.; and by 12. Contreras O, Rossi FM, Brandan E. Adherent muscle connective tissue the Canadian Institutes of Health Research (CIHR-FDN-159908) to F.M.V.R. The fibroblasts are phenotypically and biochemically equivalent to stromal fibro/ funding agencies had no role in the design of the study, data collection and adipogenic progenitors. Matrix Biol Plus. 2019;2:100006 https://doi.org/10.1 analysis, the decision to publish, or preparation of the manuscript. 016/j.mbplus.2019.04.003. Contreras et al. Skeletal Muscle (2021) 11:16 Page 21 of 25 13. Eisner C, et al. Murine Tissue-Resident PDGFRα+ Fibro-Adipogenic 35. Merrell AJ, et al. Muscle connective tissue controls development of the Progenitors Spontaneously Acquire Osteogenic Phenotype in an Altered diaphragm and is a source of congenital diaphragmatic hernias. Nat Genet. Inflammatory Environment. J Bone Miner Res. 2020;35(8):1525–34 https:// 2015;47(5):496–504 https://doi.org/10.1038/ng.3250. doi.org/10.1002/jbmr.4020. 36. Hasson P, et al. Tbx4 and Tbx5 Acting in Connective Tissue Are Required for 14. Kopinke D, Roberson EC, Reiter JF. Ciliary Hedgehog Signaling Restricts Limb Muscle and Tendon Patterning. Dev Cell. 2010;18(1):148–56 https:// Injury-Induced Adipogenesis. Cell. 2017;170(2):340–351.e12 https://doi.org/1 doi.org/10.1016/j.devcel.2009.11.013. 0.1016/j.cell.2017.06.035. 37. Li QY, et al. Holt-Oram syndrome is caused by mutations in TBX5, a member of the Brachyury (T) gene family. Nat Genet. 1997:21–9 https://doi. 15. Uezumi A, et al. Fibrosis and adipogenesis originate from a common org/10.1038/ng0197-21. mesenchymal progenitor in skeletal muscle. J Cell Sci. 2011;124(21):3654–64 https://doi.org/10.1242/jcs.086629. 38. Kutchuk L, et al. Muscle composition is regulated by a lox-TGFβ feedback 16. Farup J, et al. Human skeletal muscle CD90+ fibro-adipogenic progenitors loop. Development (Cambridge). 2015; https://doi.org/10.1242/dev.113449. are associated with muscle degeneration in type 2 diabetic patients. 39. Besse L, et al. Individual Limb Muscle Bundles Are Formed through bioRxiv. 2020:2020.08.25.243907 https://doi.org/10.1101/2020.08.25.243907. Progressive Steps Orchestrated by Adjacent Connective Tissue Cells during 17. Chapman MA, et al. Three distinct cell populations express extracellular Primary Myogenesis. Cell Rep. 2020;30(10):3552–3565.e6 https://doi.org/10.1 matrix proteins and increase in number during skeletal muscle fibrosis. Am 016/j.celrep.2020.02.037. J Physiol Cell Physiol. 2016a;312(2):C131–43 https://doi.org/10.1152/ajpcell. 40. Colasanto MP, et al. Development of a subset of forelimb muscles and their 00226.2016. attachment sites requires the ulnar-mammary syndrome gene Tbx3. DMM 18. Chapman MA, Meza R, Lieber RL. Skeletal muscle fibroblasts in health and Dis Models Mech. 2016;9(11):1257–69 https://doi.org/10.1242/dmm.025874. disease. Differentiation. 2016b:108–15 https://doi.org/10.1016/j.diff.2016.05. 41. Duboule D, Dolle P. The structural and functional organization of the murine HOX gene family resembles that of Drosophila homeotic genes. 19. Gillies AR, Lieber RL. Structure and function of the skeletal muscle EMBO J. 1989;8(5):1497–505 https://doi.org/10.1002/j.1460-2075.1989.tb03 extracellular matrix. Muscle Nerve. 2011:318–31 https://doi.org/10.1002/ 534.x. mus.22094. 42. Boulet AM, Capecchi MR. Duplication of the Hoxd11 gene causes alterations 20. Helmbacher F, Stricker S. Tissue cross talks governing limb muscle in the axial and appendicular skeleton of the mouse. Dev Biol. 2002;249(1): development and regeneration. Semin Cell Dev Biol. 2020:14–30 https://doi. 96–107 https://doi.org/10.1006/dbio.2002.0755. org/10.1016/j.semcdb.2020.05.005. 43. Boulet AM, Capecchi MR. Multiple roles of Hoxa11 and Hoxd11 in the 21. Nassari S, Duprez D, Fournier-Thibault C. Non-myogenic contribution to formation of the mammalian forelimb zeugopod. Development. 2004; muscle development and homeostasis: The role of connective tissues. Front 131(2):299–309 https://doi.org/10.1242/dev.00936. Cell Dev Biol. 2017; https://doi.org/10.3389/fcell.2017.00022. 44. Davis AP, et al. Absence of radius and ulna in mice lacking hoxa-11 andhoxd-11. Nature. 1995;375(6534):791–5 https://doi.org/10.1038/375791a0. 22. Sefton EM, Kardon G. Connecting muscle development, birth defects, and evolution: An essential role for muscle connective tissue. Curr Top Dev Biol. 45. Iimura T, Pourquié O. Hox genes in time and space during vertebrate body 2019:137–76 https://doi.org/10.1016/bs.ctdb.2018.12.004. formation. Develop Growth Differ. 2007:265–75 https://doi.org/10.1111/j.144 23. Kardon G, Harfe BD, Tabin CJ. A Tcf4-positive mesodermal population 0-169X.2007.00928.x. provides a prepattern for vertebrate limb muscle patterning. Dev Cell. 2003; 46. Swinehart IT, et al. Hox11 genes are required for regional patterning and 5(6):937–44 https://doi.org/10.1016/S1534-5807(03)00360-5. integration of muscle, tendon and bone. Development (Cambridge). 2013; 24. Mathew SJ, et al. Connective tissue fibroblasts and Tcf4 regulate 140(22):4574–82 https://doi.org/10.1242/dev.096693. myogenesis. Development. 2011;138(2):371–84 https://doi.org/10.1242/dev. 47. Alfaro LAS, et al. CD34 promotes satellite cell motility and entry into 057463. proliferation to facilitate efficient skeletal muscle regeneration. Stem Cells. 25. Kardon G, Campbell JK, Tabin CJ. Local extrinsic signals determine muscle 2011;29(12):2030–41 https://doi.org/10.1002/stem.759. and endothelial cell fate and patterning in the vertebrate limb. Dev Cell. 48. Scott RW, et al. Hic1 Defines Quiescent Mesenchymal Progenitor 2002;3(4):533–45 https://doi.org/10.1016/S1534-5807(02)00291-5. Subpopulations with Distinct Functions and Fates in Skeletal Muscle 26. Vallecillo-García P, et al. Odd skipped-related 1 identifies a population of Regeneration. Cell Stem Cell. 2019;25(6):797–813.e9 https://doi.org/10.1016/j. embryonic fibro-adipogenic progenitors regulating myogenesis during limb stem.2019.11.004. development. Nat Commun. 2017;8(1) https://doi.org/10.1038/s41467-017- 49. Dellavalle A, et al. Pericytes resident in postnatal skeletal muscle 01120-3. differentiate into muscle fibres and generate satellite cells. Nat Commun. 27. Contreras O, Rebolledo DL, Oyarzún JE, Olguín HC, Brandan E. Connective 2011;2(1) https://doi.org/10.1038/ncomms1508. tissue cells expressing fibro/adipogenic progenitor markers increase under 50. Agley CC, et al. Human skeletal muscle fibroblasts, but not myogenic cells, chronic damage: relevance in fibroblast-myofibroblast differentiation and readily undergo adipogenic differentiation. J Cell Sci. 2013;126(24):5610–25 skeletal muscle fibrosis. Cell Tissue Res. 2016;364:647–60 https://doi.org/10.1 https://doi.org/10.1242/jcs.132563. 007/s00441-015-2343-0. 51. Arrighi N, et al. Characterization of adipocytes derived from fibro/ 28. Contreras O, Soliman H, Theret M, Rossi FMV, Brandan E. TGF-β-driven adipogenic progenitors resident in human skeletal muscle. Cell Death Dis. downregulation of the transcription factor TCF7L2 affects Wnt/β-catenin 2015;6(4) https://doi.org/10.1038/cddis.2015.79. signaling in PDGFRα+ fibroblasts. J Cell Sci. 2020;133 https://doi.org/1 52. Mackey AL, et al. Human skeletal muscle fibroblasts stimulate in vitro 0.1242/jcs.242297. myogenesis and in vivo muscle regeneration. J Physiol. 2017;595(15):5115– 29. Nowicki JL, Takimoto R, Burke AC. The lateral somitic frontier: Dorso-ventral 27 https://doi.org/10.1113/JP273997. aspects of anterio-posterior regionalization in avian embryos. Mech Dev. 53. Uezumi A, et al. Identification and characterization of PDGFR + 2003;120(2):227–40 https://doi.org/10.1016/S0925-4773(02)00415-X. mesenchymal progenitors in human skeletal muscle. Cell Death Dis. 2014a; 30. Pearse RV, et al. A cellular lineage analysis of the chick limb bud. Dev Biol. 5(4) https://doi.org/10.1038/cddis.2014.161. 2007;310(2):388–400 https://doi.org/10.1016/j.ydbio.2007.08.002. 54. Uezumi A, Ikemoto-Uezumi M, Tsuchida K. Roles of nonmyogenic 31. Noden DM. The embryonic origins of avian cephalic and cervical muscles mesenchymal progenitors in pathogenesis and regeneration of skeletal and associated connective tissues. Am J Anat. 1983a;168(3):257–76 https:// muscle. Front Physiol. 2014b; https://doi.org/10.3389/fphys.2014.00068. doi.org/10.1002/aja.1001680302. 55. Stricker S, et al. Comparative expression pattern of Odd-skipped related 32. Noden DM. The role of the neural crest in patterning of avian cranial genes Osr1 and Osr2 in chick embryonic development. Gene Expr Patterns. skeletal, connective, and muscle tissues. Dev Biol. 1983b;96(1):144–65 2006;6(8):826–34 https://doi.org/10.1016/j.modgep.2006.02.003. https://doi.org/10.1016/0012-1606(83)90318-4. 56. Stricker S, et al. Odd-skipped related genes regulate differentiation of 33. Olsson L, et al. Cranial neural crest cells contribute to connective tissue in embryonic limb mesenchyme and bone marrow mesenchymal stromal cranial muscles in the anuran amphibian, Bombina orientalis. Dev Biol. 2001; cells. Stem Cells Dev. 2012;21(4):623–33 https://doi.org/10.1089/scd.2011. 237(2):354–67 https://doi.org/10.1006/dbio.2001.0377. 0154. 34. Theis S, et al. The occipital lateral plate mesoderm is a novel source for 57. Dulauroy S, et al. Lineage tracing and genetic ablation of ADAM12 + vertebrate neck musculature. Development. 2010;137(17):2961–71 https:// perivascular cells identify a major source of profibrotic cells during acute doi.org/10.1242/dev.049726. tissue injury. Nat Med. 2012;18(8):1262–70 https://doi.org/10.1038/nm.2848. Contreras et al. Skeletal Muscle (2021) 11:16 Page 22 of 25 58. Giordani L, et al. High-Dimensional Single-Cell Cartography Reveals Novel 80. Sambasivan R, et al. Pax7-expressing satellite cells are indispensable for Skeletal Muscle-Resident Cell Populations. Mol Cell. 2019;74(3):609–621.e6 adult skeletal muscle regeneration. Development. 2011;138(17):3647–56 https://doi.org/10.1016/j.molcel.2019.02.026. https://doi.org/10.1242/dev.067587. 59. Lemos DR, et al. Nilotinib reduces muscle fibrosis in chronic muscle injury 81. Iismaa, S. E. et al. (2018) Comparative regenerative mechanisms across by promoting TNF-mediated apoptosis of fibro/adipogenic progenitors. Nat different mammalian tissues, npj Regenerative Medicine. doi: https://doi. Med. 2015;21(7):786–94 https://doi.org/10.1038/nm.3869. org/10.1038/s41536-018-0044-5. 60. Lees-Shepard JB, et al. Palovarotene reduces heterotopic ossification in 82. Sandoval-Guzmán T, Currie JD. The journey of cells through regeneration. juvenile fop mice but exhibits pronounced skeletal toxicity. eLife. 2018;7 Curr Opin Cell Biol. 2018:36–41 https://doi.org/10.1016/j.ceb.2018.05.008. https://doi.org/10.7554/eLife.40814. 83. Wells JM, Watt FM. Diverse mechanisms for endogenous regeneration and 61. Madaro L, et al. Denervation-activated STAT3–IL-6 signalling in fibro- repair in mammalian organs. Nature. 2018:322–8 https://doi.org/10.1038/s41 adipogenic progenitors promotes myofibres atrophy and fibrosis. Nat Cell 586-018-0073-7. Biol. 2018;20(8):917–27 https://doi.org/10.1038/s41556-018-0151-y. 84. Cordero-Espinoza L, Huch M. The balancing act of the liver: tissue 62. Malecova B, et al. Dynamics of cellular states of fibro-adipogenic regeneration versus fibrosis. J Clin Investig. 2018:85–96 https://doi.org/10.11 progenitors during myogenesis and muscular dystrophy. Nat Commun. 72/JCI93562. 2018;9(1) https://doi.org/10.1038/s41467-018-06068-6. 85. Dadgar S, et al. Asynchronous remodeling is a driver of failed regeneration 63. Petrilli LL, et al. High-Dimensional Single-Cell Quantitative Profiling of in Duchenne muscular dystrophy. J Cell Biol. 2014;207(1):139–58 https://doi. Skeletal Muscle Cell Population Dynamics during Regeneration. Cells. 2020; org/10.1083/jcb.201402079. 9(7) https://doi.org/10.3390/cells9071723. 86. Pessina P, et al. Novel and optimized strategies for inducing fibrosis in vivo: 64. Uezumi A, et al. Cell-Surface Protein Profiling Identifies Distinctive Markers Focus on Duchenne Muscular Dystrophy. Skelet Muscle. 2014;4(1) https:// of Progenitor Cells in Human Skeletal Muscle. Stem Cell Rep. 2016;7(2):263– doi.org/10.1186/2044-5040-4-7. 78 https://doi.org/10.1016/j.stemcr.2016.07.004. 87. Bendall JR. The elastin content of various muscles of beef animals. J Sci 65. Yao L, et al. Gli1 defines a subset of fibroadipogenic progenitors that Food Agric. 1967;18(12):553–8 https://doi.org/10.1002/jsfa.2740181201. promote skeletal muscle regeneration with less fat accumulation. J Bone 88. Dransfield E. Intramuscular composition and texture of beef muscles. J Sci Miner Res. 2021; https://doi.org/10.1002/jbmr.4265. Food Agric. 1977;28(9):833–42 https://doi.org/10.1002/jsfa.2740280910. 66. Giuliani G, Vumbaca S, Fuoco C, Gargioli C, Giorda E, Massacci G, et al. 89. Vognarová I, Dvorák Z, Böhm R. Collagen and Elastin in Different Cuts of SCA-1 micro-heterogeneity in the fate decision of dystrophic fibro/ Veal and Beef. J Food Sci. 1968;33(4):339–43 https://doi.org/10.1111/j.1365-2 adipogenic progenitors. Cell Death Dis. 2021;12 https://doi.org/10.1038/ 621.1968.tb03626.x. s41419-021-03408-1. 90. Light N, Champion AE. Characterization of muscle epimysium, perimysium 67. Stumm J, et al. Odd skipped-related 1 (Osr1) identifies muscle-interstitial and endomysium collagens. Biochem J. 1984;219(3):1017–26 https://doi. fibro-adipogenic progenitors (FAPs) activated by acute injury. Stem Cell Res. org/10.1042/bj2191017. 2018;32:8–16 https://doi.org/10.1016/j.scr.2018.08.010. 91. Purslow PP. The Structure and Role of Intramuscular Connective Tissue 68. Kramann R, et al. Perivascular Gli1+ progenitors are key contributors to in Muscle Function. Front Physiol. 2020; https://doi.org/10.3389/fphys.2 injury-induced organ fibrosis. Cell Stem Cell. 2015;16(1):51–66 https://doi. 020.00495. org/10.1016/j.stem.2014.11.004. 92. Borg TK, Caulfield JB. Morphology of connective tissue in skeletal 69. Zhao L, et al. Retinoic acid signalling in fibro/adipogenic progenitors muscle. Tissue Cell. 1980;12(1):197–207 https://doi.org/10.1016/0040-81 robustly enhances muscle regeneration. EBioMedicine. 2020;60 https://doi. 66(80)90061-0. org/10.1016/j.ebiom.2020.103020. 93. Carnwath JW, Shotton DM. Muscular dystrophy in the mdx mouse: 70. Xi H, Langerman J, Sabri S, Chien P, Young CS, Younesi S, et al. A human Histopathology of the soleus and extensor digitorum longus muscles. skeletal muscle atlas identifies the trajectories of stem and progenitor cells J Neurol Sci. 1987;80(1):39–54 https://doi.org/10.1016/0022-510X(87)90219-X. across development and from human pluripotent stem cells. Cell Stem Cell. 94. Duance VC, et al. A role for collagen in the pathogenesis of muscular 2020;27(1):158–76.e10 https://doi.org/10.1016/j.stem.2020.04.017. dystrophy? Nature. 1980;284(5755):470–2 https://doi.org/10.1038/284470a0. 71. Goloviznina NA, et al. Prospective isolation of human fibroadipogenic 95. Gatchalian CL, Schachner M, Sanes JR. Fibroblasts that proliferate near progenitors with CD73. Heliyon. 2020;6(7) https://doi.org/10.1016/j.heliyon.2 denervated synaptic sites in skeletal muscle synthesize the adhesive 020.e04503. molecules tenascin(J1), N-CAM, fibronectin, and a heparan sulfate 72. Kasai T, et al. Promethazine Hydrochloride Inhibits Ectopic Fat Cell proteoglycan. J Cell Biol. 1989;108(5):1873–90 https://doi.org/10.1083/jcb.1 Formation in Skeletal Muscle. Am J Pathol. 2017;187(12):2627–34 https://doi. 08.5.1873. org/10.1016/j.ajpath.2017.08.008. 96. Klingler W, et al. The role of fibrosis in Duchenne muscular dystrophy. Acta 73. Soliman H, et al. Pathogenic Potential of Hic1-Expressing Cardiac Stromal Myologica. 2012;31(3):184–95. Progenitors. Cell Stem Cell. 2020;26(2):205–220.e8 https://doi.org/10.1016/j. 97. Lieber RL, Ward SR. Cellular mechanisms of tissue fibrosis. 4. structural and stem.2019.12.008. functional consequences of skeletal muscle fibrosis. Am J Physiol Cell 74. Chong JJH, et al. Adult cardiac-resident MSC-like stem cells with a Physiol. 2013;305(3) https://doi.org/10.1152/ajpcell.00173.2013. proepicardial origin. Cell Stem Cell. 2011;9(6):527–40 https://doi.org/10.1016/ 98. Morrison J, et al. T-cell-dependent fibrosis in the mdx dystrophic mouse. j.stem.2011.10.002. Lab Investig. 2000;80(6):881–91 https://doi.org/10.1038/labinvest.3780092. 75. Noseda M, et al. PDGFRα demarcates the cardiogenic clonogenic Sca1+ 99. Serrano AL, et al. Cellular and molecular mechanisms regulating fibrosis in stem/progenitor cell in adult murine myocardium. Nat Commun. 2015;6 skeletal muscle repair and disease. Curr Top Dev Biol. 2011; https://doi.org/1 https://doi.org/10.1038/ncomms7930. 0.1016/B978-0-12-385940-2.00007-3. 76. Chen WY, et al. Heterozygous disruption of Hic1 predisposes mice to a 100. Williams PE, Goldspink G. Connective tissue changes in immobilised muscle. gender-dependent spectrum of malignant tumors. Nat Genet. 2003;33(2): J Anat. 1984;138(Pt 2):343–50 Available at: http://www.ncbi.nlm.nih.gov/ 197–202 https://doi.org/10.1038/ng1077. pubmed/6715254%0A, http://www.pubmedcentral.nih.gov/articlerender. 77. Van Rechem C, et al. Differential Regulation of HIC1 Target Genes by CtBP fcgi?artid=PMC1164074. and NuRD, via an Acetylation/SUMOylation Switch, in Quiescent versus 101. Jackson WM, et al. Mesenchymal progenitor cells derived from traumatized Proliferating Cells. Mol Cell Biol. 2010;30(16):4045–59 https://doi.org/10.1128/ human muscle. J Tissue Eng Regen Med. 2009;3(2):129–38 https://doi.org/1 mcb.00582-09. 0.1002/term.149. 78. Lepper C, Partridge TA, Fan CM. An absolute requirement for pax7-positive 102. Dammone G, et al. PPARγ controls ectopic adipogenesis and cross-talks satellite cells in acute injury-induced skeletal muscle regeneration. with myogenesis during skeletal muscle regeneration. Int J Mol Sci. 2018; Development. 2011;138(17):3639–46 https://doi.org/10.1242/dev.067595. 19(7) https://doi.org/10.3390/ijms19072044. 79. Relaix F, Zammit PS. Satellite cells are essential for skeletal muscle 103. Gonzalez D, et al. ALS skeletal muscle shows enhanced TGF-β signaling, regeneration: The cell on the edge returns centre stage. Development fibrosis and induction of fibro/adipogenic progenitor markers. PLoS One. (Cambridge). 2012;139(16):2845–56 https://doi.org/10.1242/dev.069088. 2017;12(5) https://doi.org/10.1371/journal.pone.0177649. Contreras et al. Skeletal Muscle (2021) 11:16 Page 23 of 25 104. Hogarth MW, et al. Fibroadipogenic progenitors are responsible for muscle 126. Györfi AH, Matei AE, Distler JHW. Targeting TGF-β signaling for the loss in limb girdle muscular dystrophy 2B. Nat Commun. 2019;10(1) https:// treatment of fibrosis. Matrix Biol. 2018:8–27 https://doi.org/10.1016/j.ma doi.org/10.1038/s41467-019-10438-z. tbio.2017.12.016. 105. Lukjanenko L, et al. Loss of fibronectin from the aged stem cell niche affects 127. Hinz B. Myofibroblasts. Exp Eye Res. 2015:56–70 https://doi.org/10.1016/j. the regenerative capacity of skeletal muscle in mice. Nat Med. 2016;22(8): exer.2015.07.009. 897–905 https://doi.org/10.1038/nm.4126. 128. Klingberg F, et al. Prestress in the extracellular matrix sensitizes latent TGF- 106. De Micheli AJ, Spector JA, et al. A reference single-cell transcriptomic atlas β1 for activation. J Cell Biol. 2014;207(2):283–97 https://doi.org/10.1083/jcb.2 of human skeletal muscle tissue reveals bifurcated muscle stem cell populations. Skelet Muscle. 2020a;10(1) https://doi.org/10.1186/s13395-020- 129. Reed NI, et al. The αvβ1integrinplays acriticalinvivoroleintissue 00236-3. fibrosis. Sci Transl Med. 2015;7(288) https://doi.org/10.1126/scitra 107. De Micheli AJ, Laurilliard EJ, et al. Single-cell analysis of the muscle stem cell nslmed.aaa5094. hierarchy identifies heterotypic communication signals involved in skeletal 130. Wang XF, et al. Expression cloning and characterization of the TGF-β type III muscle regeneration. Cell Rep. 2020b;30(10):3583–3595.e5 https://doi.org/1 receptor. Cell. 1991; https://doi.org/10.1016/0092-8674(91)90074-9. 0.1016/j.celrep.2020.02.067. 131. López-Casillas F, et al. Betaglycan can act as a dual modulator of TGF-β 108. Oprescu SN, et al. Temporal Dynamics and Heterogeneity of Cell access to signaling receptors: Mapping of ligand binding and GAG Populations during Skeletal Muscle Regeneration. iScience. 2020;23(4) attachment sites. J Cell Biol. 1994; https://doi.org/10.1083/jcb.124.4.557. https://doi.org/10.1016/j.isci.2020.100993. 132. Lewis KA, et al. Betaglycan binds inhibin and can mediate functional 109. Rubenstein AB, et al. Single-cell transcriptional profiles in human antagonism of activin signalling. Nature. 2000; https://doi.org/10.1038/3 skeletal muscle. Sci Rep. 2020;10(1) https://doi.org/10.1038/s41598-019- 57110-6. 133. Wiater E, Vale W. Inhibin is an antagonist of bone morphogenetic protein 110. Riquelme-Guzmán C, Contreras O. Single-cell revolution unveils the signaling. J Biol Chem. 2003; https://doi.org/10.1074/jbc.M209710200. mysteries of the regenerative mammalian digit tip. Dev Biol. 2020;461:107–9 134. Velasco-Loyden G, Arribas J, López-Casillas F. The Shedding of Betaglycan Is https://doi.org/10.1016/j.ydbio.2020.02.002. Regulated by Pervanadate and Mediated by Membrane Type Matrix 111. Marinkovic M, et al. Fibro-adipogenic progenitors of dystrophic mice are Metalloprotease-1. J Biol Chem. 2004; https://doi.org/10.1074/jbc.M3064992 insensitive to NOTCH regulation of adipogenesis. Life Sci Alliance. 2019;2(3) 00. https://doi.org/10.26508/lsa.201900437. 135. Vilchis-Landeros MM, et al. Recombinant soluble betaglycan is a potent and 112. Ciuffreda MC, et al. Protocols for in vitro differentiation of human isoform-selective transforming growth factor-β neutralizing agent. Biochem mesenchymal stem cells into osteogenic, chondrogenic and adipogenic J. 2001; https://doi.org/10.1042/0264-6021:3550215. lineages. Methods Mol Biol. 2016:149–58 https://doi.org/10.1007/978-1-493 136. Bilandzic M, et al. Betaglycan blocks metastatic behaviors in human 9-3584-0_8. granulosa cell tumors by suppressing NFκB-mediated induction of MMP2. 113. Sacchetti B, et al. No identical “mesenchymal stem cells” at different Cancer Lett. 2014; https://doi.org/10.1016/j.canlet.2014.07.039. times and sites: Human committed progenitors of distinct origin and 137. Shi Y, Massagué J. Mechanisms of TGF-β signaling from cell membrane to differentiation potential are incorporated as adventitial cells in the nucleus. Cell. 2003:685–700 https://doi.org/10.1016/S0092-8674(03 microvessels. Stem Cell Rep. 2016;6(6):897–913 https://doi.org/10.1016/j. )00432-X. stemcr.2016.05.011. 138. Imamura T, et al. Smad6 inhibits signalling by the TGF-β superfamily. 114. Mierzejewski B, et al. Mouse CD146+ muscle interstitial progenitor cells Nature. 1997; https://doi.org/10.1038/39355. differ from satellite cells and present myogenic potential. Stem Cell Res 139. Jung SM, et al. Smad6 inhibits non-canonical TGF-β1 signalling by recruiting Ther. 2020;11(1) https://doi.org/10.1186/s13287-020-01827-z. the deubiquitinase A20 to TRAF6. Nat Commun. 2013; https://doi.org/10.103 115. Swift J, et al. Nuclear lamin-A scales with tissue stiffness and enhances 8/ncomms3562. matrix-directed differentiation. Science. 2013;341(6149) https://doi.org/1 140. Nakao A, et al. Identification of Smad7, a TGFβ-inducible antagonist of TGF- 0.1126/science.1240104. β signalling. Nature. 1997; https://doi.org/10.1038/39369. 116. Lee C, et al. Rotator Cuff Fibro-Adipogenic Progenitors Demonstrate Highest 141. Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways Concentration, Proliferative Capacity, and Adipogenic Potential Across in TGF-β family signalling. Nature. 2003:577–84 https://doi.org/10.1038/na Muscle Groups. J Orthop Res. 2020;38(5):1113–21 https://doi.org/10.1002/ ture02006. jor.24550. 142. Khalil H, et al. Fibroblast-specific TGF-β-Smad2/3 signaling underlies cardiac 117. Mázala DAG, et al. TGF-β-driven muscle degeneration and failed fibrosis, in. J Clin Investig. 2017:3770–83 https://doi.org/10.1172/JCI94753. regeneration underlie disease onset in a DMD mouse model. JCI Insight. 143. Kim KK, Sheppard D, Chapman HA. TGF-β1 signaling and tissue fibrosis. 2020;5(6) https://doi.org/10.1172/jci.insight.135703. Cold Spring Harb Perspect Biol. 2018;10(4) https://doi.org/10.1101/ 118. Rodgers JT, et al. HGFA Is an Injury-Regulated Systemic Factor that Induces cshperspect.a022293. the Transition of Stem Cells into GAlert. Cell Rep. 2017;19(3):479–86 https:// 144. Juban G, et al. AMPK Activation Regulates LTBP4-Dependent TGF-β1 doi.org/10.1016/j.celrep.2017.03.066. Secretion by Pro-inflammatory Macrophages and Controls Fibrosis in 119. Burks TN, Cohn RD. Role of TGF-β signaling in inherited and acquired Duchenne Muscular Dystrophy. Cell Rep. 2018;25(8):2163–2176.e6 https:// myopathies. Skelet Muscle. 2011;1(1):19 https://doi.org/10.1186/2044-504 doi.org/10.1016/j.celrep.2018.10.077. 0-1-19. 145. McLennan IS, Koishi K. Cellular localisation of transforming growth factor- 120. Lodyga M, Hinz B. TGF-β1 - A truly transforming growth factor in fibrosis beta 2 and -beta 3 (TGF-β2, TGF-β3) in damaged and regenerating skeletal and immunity. Semin Cell Dev Biol. 2020:123–39 https://doi.org/10.1016/j. muscles. Dev Dyn. 1997;208(2):278–89 https://doi.org/10.1002/(sici)1097-01 semcdb.2019.12.010. 77(199702)208:2<278::aid-aja14>3.0.co;2-%23. 121. Budi EH, Duan D, Derynck R. Transforming Growth Factor-β Receptors and 146. Theret, Marine, Marcela Low, Lucas Rempel, Fang Fang Li, Lin Wei Tung, Smads: Regulatory Complexity and Functional Versatility. Trends Cell Biol. Osvaldo Contreras, Chih-Kai Chang, Andrew Wu, Hesham Soliman, and 2017:658–72 https://doi.org/10.1016/j.tcb.2017.04.005. Fabio M V Rossi. “In Vitro Assessment of Anti-Fibrotic Drug Activity Does 122. Wu MY, Hill CS. TGF-β Superfamily Signaling in Embryonic Development Not Predict in Vivo Efficacy in Murine Models of Duchenne Muscular and Homeostasis. Dev Cell. 2009:329–43 https://doi.org/10.1016/j.devcel.2 Dystrophy.” Life Sciences 279 (2021):119482. https://doi.org/10.1016/j.lfs.2 009.02.012. 021.119482. 123. David CJ, Massagué J. Contextual determinants of TGFβ action in 147. Davies MR, et al. TGF-β small molecule inhibitor sb431542 reduces rotator development, immunity and cancer. Nat Rev Mol Cell Biol. 2018:419–35 cuff muscle fibrosis and fatty infiltration by promoting fibro/ adipogenic https://doi.org/10.1038/s41580-018-0007-0. progenitor apoptosis. PLoS One. 2016;11(5) https://doi.org/10.1371/journal. 124. Derynck R, Budi EH. Specificity, versatility, and control of TGF-b family pone.0155486. signaling. Sci Signal. 2019; https://doi.org/10.1126/scisignal.aav5183. 148. Saito Y, Chikenji TS, Matsumura T, Nakano M, Fujimiya M. Exercise enhances 125. Hinz B, et al. The myofibroblast: One function, multiple origins. Am J Pathol. skeletal muscle regeneration by promoting senescence in fibro-adipogenic 2007;170(6):1807–16 https://doi.org/10.2353/ajpath.2007.070112. progenitors. Nat Commun. 2020;11:889 https://doi.org/10.1038/s41467-02 0-14734-x. Contreras et al. Skeletal Muscle (2021) 11:16 Page 24 of 25 149. Reggio A, Spada F, Rosina M, Massacci G, Zuccotti A, Fuoco C, et al. The 172. Gronwald RGK, et al. Cloning and expression of a cDNA coding for the immunosuppressant drug azathioprine restrains adipogenesis of muscle human platelet-derived growth factor receptor: Evidence for more than one Fibro/Adipogenic Progenitors from dystrophic mice by affecting AKT receptor class. Proc Natl Acad Sci U S A. 1988;85(10):3435–9 https://doi. signaling. Sci Rep. 2019;9 https://doi.org/10.1038/s41598-019-39538-y. org/10.1073/pnas.85.10.3435. 150. Dong Y, Augusto K, Silva S, Dong Y, Zhang L. Glucocorticoids increase 173. Matsui T, et al. Isolation of a novel receptor cDNA establishes the existence adipocytes in muscle by affecting IL-4 regulated FAP activity. FASEB J. 2014; of two PDGF receptor genes. Science. 1989;243(4892):800–4 https://doi. 28(9):4123–32 https://doi.org/10.1096/fj.14-254011. org/10.1126/science.2536956. 151. Mozzetta C, et al. Fibroadipogenic progenitors mediate the ability of HDAC 174. Yarden Y, et al. Structure of the receptor for platelet-derived growth factor inhibitors to promote regeneration in dystrophic muscles of young, but not helps define a family of closely related growth factor receptors. Nature. old Mdx mice. EMBO Mol Med. 2013;5(4):626–39 https://doi.org/10.1002/ 1986;323(6085):226–32 https://doi.org/10.1038/323226a0. emmm.201202096. 175. Heldin C-H, Westermark B. Role of Platelet-Derived Growth Factor in Vivo. 152. Liu X, et al. Trichostatin A regulates fibro/adipogenic progenitor Mol Cell Biol Wound Repair. 1988:249–73 https://doi.org/10.1007/978-1-4 adipogenesis epigenetically and reduces rotator cuff muscle fatty 899-0185-9_7. infiltration. J Orthop Res. 2020; https://doi.org/10.1002/jor.24865. 176. Kohler N, Lipton A. Platelets as a source of fibroblast growth-promoting 153. Reggio A, et al. Adipogenesis of skeletal muscle fibro/adipogenic activity. Exp Cell Res. 1974;87(2):297–301 https://doi.org/10.1016/0014-482 progenitors is affected by the WNT5a/GSK3/β-catenin axis. Cell Death Differ. 7(74)90484-4. 2020;27(10):2921–41 https://doi.org/10.1038/s41418-020-0551-y. 177. Hoch RV, Soriano P. Roles of PDGF in animal development. Development. 154. Cordani N, et al. Nitric oxide controls fat deposition in dystrophic skeletal 2003;130(20):4769–84 https://doi.org/10.1242/dev.00721. muscle by regulating fibro-adipogenic precursor differentiation. Stem Cells. 178. Heldin CH, Westermark B. Mechanism of action and in vivo role of platelet- 2014;32(4):874–85 https://doi.org/10.1002/stem.1587. derived growth factor, physiological reviews. Am Physiol Soc. 1999;79(4): 155. Ieronimakis N, et al. PDGFRα signalling promotes fibrogenic responses in 1283–316 https://doi.org/10.1152/physrev.1999.79.4.1283. collagen-producing cells in Duchenne muscular dystrophy. J Pathol. 2016; 179. Roskoski R. The role of small molecule platelet-derived growth factor 240(4):410–24 https://doi.org/10.1002/path.4801. receptor (PDGFR) inhibitors in the treatment of neoplastic disorders. 156. Astudillo P. Extracellular matrix stiffness and Wnt/β-catenin signaling in Pharmacol Res. 2018:65–83 https://doi.org/10.1016/j.phrs.2018.01.021. physiology and disease. Biochem Soc Trans. 2020:1187–98 https://doi.org/1 180. Kazlauskas A. PDGFs and their receptors. Gene. 2017:1–7 https://doi.org/10.1 0.1042/BST20200026. 016/j.gene.2017.03.003. 157. Nusse R, Clevers H. Wnt/β-Catenin Signaling. Dis Emerg Ther Modalities Cell. 181. Kanakaraj P, et al. Ligand-Induced Interaction between α- and β-Type 2017:985–99 https://doi.org/10.1016/j.cell.2017.05.016. Platelet-Derived Growth Factor (PDGF) Receptors: Role of Receptor 158. Aberle H, et al. Β-Catenin Is a Target for the Ubiquitin-Proteasome Pathway. Heterodimers in Kinase Activation. Biochemistry. 1991;30(7):1761–7 https:// EMBO J. 1997;16(13):3797–804 https://doi.org/10.1093/emboj/16.13.3797. doi.org/10.1021/bi00221a005. 159. Valenta T, Hausmann G, Basler K. The many faces and functions of β- 182. Kelly JD, et al. Platelet-derived growth factor (PDGF) stimulates PDGF catenin. EMBO J. 2012:2714–36 https://doi.org/10.1038/emboj.2012.150. receptor subunit dimerization and intersubunit trans-phosphorylation. 160. Cadigan KM, Waterman ML. TCF/LEFs and Wnt signaling in the nucleus. J Biol Chem. 1991;266(14):8987–92 https://doi.org/10.1016/s0021-9258(1 Cold Spring Harb Perspect Biol. 2012;4(11) https://doi.org/10.1101/ 8)31541-2. cshperspect.a007906. 183. Wang Y, et al. Platelet-derived Growth Factor Receptor-mediated Signal Transduction from Endosomes. J Biol Chem. 2004;279(9):8038–46 https://doi. 161. Clevers H. Wnt/β-Catenin Signaling in Development and Disease. Cell. 2006: org/10.1074/jbc.M311494200. 469–80 https://doi.org/10.1016/j.cell.2006.10.018. 162. Trensz F, et al. A muscle resident cell population promotes fibrosis in 184. Mueller AA, et al. Intronic polyadenylation of PDGFRα in resident stem cells hindlimb skeletal muscles of mdx mice through the Wnt canonical attenuates muscle fibrosis. Nature England. 2016;540(7632):276–9 https://doi. pathway. Am J Physiol Cell Physiol. 2010;299(5) https://doi.org/10.1152/a org/10.1038/nature20160. jpcell.00253.2010. 185. Sugg KB, et al. Inhibition of platelet-derived growth factor signaling 163. Biressi S, et al. A Wnt-TGF2 axis induces a fibrogenic program in muscle prevents muscle fiber growth during skeletal muscle hypertrophy. FEBS stem cells from dystrophic mice. Sci Transl Med. 2014;6(267) https://doi. Letters. 2017;591(5):801–9 https://doi.org/10.1002/1873-3468.12571. John org/10.1126/scitranslmed.3008411. Wiley & Sons, Ltd. 164. Xiang FL, Fang M, Yutzey KE. Loss of β-catenin in resident cardiac fibroblasts 186. Asli NS, et al. PDGFRα signaling in cardiac fibroblasts modulates quiescence, attenuates fibrosis induced by pressure overload in mice. Nat Commun. metabolism and self-renewal, and promotes anatomical and functional 2017;8(1):712 https://doi.org/10.1038/s41467-017-00840-w. repair. bioRxiv. 2019a:225979 https://doi.org/10.1101/225979. 187. Ivey MJ, et al. Platelet-derived growth factor receptor-α is essential for 165. Akhmetshina, A. et al. (2012) Activation of canonical Wnt signalling is cardiac fibroblast survival. Am J Physiol Heart Circ Physiol. 2019;317(2):H330– required for TGF-β-mediated fibrosis, Nature Communications, 3. doi: 44 https://doi.org/10.1152/ajpheart.00054.2019. https://doi.org/10.1038/ncomms1734. 166. Hamburg-Shields E, et al. Sustained β-catenin activity in dermal fibroblasts 188. Harvey T, Flamenco S, Fan CM. A Tppp3 + Pdgfra + tendon stem cell promotes fibrosis by up-regulating expression of extracellular matrix population contributes to regeneration and reveals a shared role for PDGF protein-coding genes. J Pathol. 2015;235(5):686–97 https://doi.org/10.1002/ signalling in regeneration and fibrosis. Nat Cell Biol. 2019;21(12):1490–503 path.4481. https://doi.org/10.1038/s41556-019-0417-z. 167. Działo E, Tkacz K, Błyszczuk P. Crosstalk between the TGF-β and WNT 189. Liu W, et al. Intramuscular adipose is derived from a non-Pax3 lineage and signalling pathways during cardiac fibrogenesis. Acta Biochim Pol. 2018; required for efficient regeneration of skeletal muscles. Dev Biol. 2012;361(1): 65(3):341–9. https://doi.org/10.18388/abp.2018_2635. 27–38 https://doi.org/10.1016/j.ydbio.2011.10.011. 190. Tchoukalova YD, Sarr MG, Jensen MD. Measuring committed preadipocytes 168. Girardi F, Le Grand F. Wnt Signaling in Skeletal Muscle Development and in human adipose tissue from severely obese patients by using adipocyte Regeneration. Prog Mol Biol Transl Sci. 2018:157–79 https://doi.org/10.1016/ fatty acid binding protein. Am J Physiol Regul Integr Comp Physiol. 2004; bs.pmbts.2017.11.026. 287(5 56-5) https://doi.org/10.1152/ajpregu.00337.2004. 169. Piersma B, Bank RA, Boersema M. Signaling in fibrosis: TGF-β, WNT, and YAP/TAZ converge. Front Med. 2015; https://doi.org/10.3389/fmed.2015. 191. Biltz NK, Meyer GA. A novel method for the quantification of fatty 00059. infiltration in skeletal muscle. Skelet Muscle. 2017;7(1) https://doi.org/10.11 170. Lindahl P, et al. Pericyte loss and microaneurysm formation in PDGF-B- 86/s13395-016-0118-2. deficient mice. Science. 1997;277(5323):242–5 https://doi.org/10.1126/ 192. Pisani DF, et al. Mouse model of skeletal muscle adiposity: A glycerol science.277.5323.242. treatment approach. Biochem Biophys Res Commun. 2010;396(3):767–73 https://doi.org/10.1016/j.bbrc.2010.05.021. 171. Claesson-Welsh L, et al. cDNA cloning and expression of the human A-type 193. Shabalina IG, et al. UCP1 in Brite/Beige adipose tissue mitochondria is platelet-derived growth factor (PDGF) receptor establishes structural functionally thermogenic. Cell Rep. 2013;5(5):1196–203 https://doi.org/10.1 similarity to the B-type PDGF receptor. Proc Natl Acad Sci U S A. 1989; 016/j.celrep.2013.10.044. 86(13):4917–21 https://doi.org/10.1073/pnas.86.13.4917. Contreras et al. Skeletal Muscle (2021) 11:16 Page 25 of 25 194. Gorski T, Mathes S, Krützfeldt J. Uncoupling protein 1 expression in 216. Pawlikowski, B. et al. (2019) A cellular atlas of skeletal muscle regeneration adipocytes derived from skeletal muscle fibro/adipogenic progenitors is and aging, bioRxiv, p. 635805. doi: https://doi.org/10.1101/635805. under genetic and hormonal control. J Cachexia Sarcopenia Muscle. 2018; 217. Bentzinger CF, et al. Cellular dynamics in the muscle satellite cell niche. 9(2):384–99 https://doi.org/10.1002/jcsm.12277. EMBO Rep. 2013:1062–72 https://doi.org/10.1038/embor.2013.182. 195. Saely CH, Geiger K, Drexel H. Brown versus white adipose tissue: A mini- 218. Chiristov C, Chrétien F, Abou-Khalil R, Bassez G, Vallet G, Authier FJ, et al. review. Gerontology. 2011:15–23 https://doi.org/10.1159/000321319. Muscle satellite cells and endothelial cells: Close neighbors and privileged partners. Mol Biol Cell. 2007;18:1397–409 https://doi.org/10.1091/mbc.E06- 196. Vishvanath L, Gupta RK. Contribution of adipogenesis to healthy adipose 08-0693. tissue expansion in obesity. J Clin Invest. 2019:4022–31 https://doi.org/10.11 219. Latroche C, et al. Skeletal muscle microvasculature: A highly dynamic 72/JCI129191. lifeline. Physiology. 2015:417–27 https://doi.org/10.1152/physiol.00026.2015. 197. Schulz TJ, Tseng YH. Brown adipose tissue: Development, metabolism and 220. Bonfanti C, et al. PW1/Peg3 expression regulates key properties that beyond. Biochem J. 2013:167–78 https://doi.org/10.1042/BJ20130457. determine mesoangioblast stem cell competence. Nat Commun. 2015;6 198. Bettica P, et al. Histological effects of givinostat in boys with Duchenne https://doi.org/10.1038/ncomms7364. muscular dystrophy. Neuromuscul Disord. 2016;26(10):643–9 https://doi. 221. Tedesco FS, Moyle LA, Perdiguero E. Muscle interstitial cells: A brief field org/10.1016/j.nmd.2016.07.002. guide to non-satellite cell populations in skeletal muscle. Methods Mol Biol. 199. Minetti GC, et al. Functional and morphological recovery of dystrophic 2017:129–47 https://doi.org/10.1007/978-1-4939-6771-1_7. muscles in mice treated with deacetylase inhibitors. Nat Med. 2006;12(10): 222. Chazaud B. Inflammation and Skeletal Muscle Regeneration: Leave It to the 1147–50 https://doi.org/10.1038/nm1479. Macrophages! Trends Immunol. 2020:481–92 https://doi.org/10.1016/j.it.202 200. Sandonà M, et al. HDAC inhibitors tune miRNAs in extracellular vesicles of 0.04.006. dystrophic muscle-resident mesenchymal cells. EMBO Rep. 2020a;21(9) 223. Juban G, Chazaud B. Metabolic regulation of macrophages during tissue https://doi.org/10.15252/embr.202050863. repair: insights from skeletal muscle regeneration. FEBS Letter. 2017:3007–21 201. Saccone V, et al. HDAC-regulated myomiRs control BAF60 variant exchange https://doi.org/10.1002/1873-3468.12703. and direct the functional phenotype of fibro-adipogenic progenitors in 224. Theret M, Mounier R, Rossi F. The origins and non-canonical functions of dystrophic muscles. Genes Dev. 2014;28(8):841–57 https://doi.org/10.1101/ macrophages in development and regeneration. Development (Cambridge). gad.234468.113. 2019;146(9) https://doi.org/10.1242/dev.156000. 202. Hu F, et al. CCN1 induces adipogenic differentiation of fibro/adipogenic progenitors in a chronic kidney disease model. Biochem Biophys Res Commun. 2019;520(2):385–91 https://doi.org/10.1016/j.bbrc.2019.10.047. Publisher’sNote 203. Cerquone Perpetuini A, et al. Janus effect of glucocorticoids on Springer Nature remains neutral with regard to jurisdictional claims in differentiation of muscle fibro/adipogenic progenitors. Sci Rep. 2020;10(1) published maps and institutional affiliations. https://doi.org/10.1038/s41598-020-62194-6. 204. Fardet L, Petersen I, Nazareth I. Prevalence of long-term oral glucocorticoid prescriptions in the UK over the past 20 years. Rheumatology (Oxford, England). 2011;50(11):1982–90 https://doi.org/10.1093/rheumatology/ker017. 205. McDonald CM, et al. Long-term effects of glucocorticoids on function, quality of life, and survival in patients with Duchenne muscular dystrophy: a prospective cohort study. Lancet. 2018;391(10119):451–61 https://doi.org/1 0.1016/S0140-6736(17)32160-8. 206. Ricotti V, et al. Long-term benefits and adverse effects of intermittent versus daily glucocorticoids in boys with Duchenne muscular dystrophy. J Neurol Neurosurg Psychiatry. 2013;84(6):698–705 https://doi.org/10.1136/jnnp-2 012-303902. 207. Palma, A. et al. (2019) Myo-REG: A portal for signaling interactions in muscle regeneration, Frontiers in Physiology, 10(SEP). doi: https://doi.org/10.3389/ fphys.2019.01216. 208. Oishi T, et al. Osteogenic Differentiation Capacity of Human Skeletal Muscle- Derived Progenitor Cells. PLoS One. 2013;8(2) https://doi.org/10.1371/journa l.pone.0056641. 209. Meyers C, et al. Heterotopic Ossification: A Comprehensive Review. JBMR Plus. 2019;3(4):e10172 https://doi.org/10.1002/jbm4.10172. 210. Leblanc E, et al. BMP-9-induced muscle heterotopic ossification requires changes to the skeletal muscle microenvironment. J Bone Miner Res. 2011; 26(6):1166–77 https://doi.org/10.1002/jbmr.311. 211. Wosczyna MN, et al. Multipotent progenitors resident in the skeletal muscle interstitium exhibit robust BMP-dependent osteogenic activity and mediate heterotopic ossification. J Bone Miner Res. 2012;27(5):1004–17 https://doi. org/10.1002/jbmr.1562. 212. Glass GE, et al. TNF-α promotes fracture repair by augmenting the recruitment and differentiation of muscle-derived stromal cells. Proc Natl Acad Sci U S A. 2011;108(4):1585–90 https://doi.org/10.1073/pnas.10185011 213. Abou-Khalil R, et al. Autocrine and Paracrine Angiopoietin 1/Tie-2 Signaling Promotes Muscle Satellite Cell Self-Renewal. Cell Stem Cell. 2009;5(3):298– 309 https://doi.org/10.1016/j.stem.2009.06.001. 214. De Palma M, et al. Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell. 2005;8(3):211–26 https://doi. org/10.1016/j.ccr.2005.08.002. 215. Bozycki L, et al. Whole-body clearing, staining and screening of calcium deposits in the mdx mouse model of Duchenne muscular dystrophy. Skelet Muscle. 2018;8(1) https://doi.org/10.1186/s13395-018-0168-8.

Journal

Skeletal MuscleSpringer Journals

Published: Jul 1, 2021

Keywords: Mesenchymal stromal/stem cell; Fibro/adipogenic progenitor; Fibroblast; Adipocyte; Regeneration; Single-cell RNAseq

References