15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Received: 7 April 2021 | Revised: 8 July 2021 | Accepted: 12 July 2021 DOI: 10.1096/fj.202100588R METHODS Isolation of human fibroadipogenic progenitors and satellite cells from frozen muscle biopsies Xavier Suárez-­Calvet1,2 | Esther Fernández-­Simón1,3 | Patricia Piñol-­Jurado1,3 | Jorge Alonso-­Pérez1 | Ana Carrasco-­Rozas1 | Cinta Lleixà1 | Susana López-­Fernández4 | Gemma Pons4 | Laura Soria5 | Anne Bigot6 | Vincent Mouly6 | Isabel Illa1,2 | Eduard Gallardo1,2 | Jyoti K. Jaiswal7 | Jordi Díaz-­Manera1,2,3 1 Neuromuscular Diseases Unit, Neurology Department, Hospital de la Santa Creu i Sant Pau and Biomedical Research Institute Sant Pau (IIB Sant Pau), Barcelona, Spain 2 Centro de Investigaciones Biomédicas en Red en Enfermedades Raras (CIBERER), Madrid, Spain 3 John Walton Muscular Dystrophy Research Center, University of Newcastle, Newcastle upon Tyne, UK 4 Plastic Surgery Department, Hospital de la Santa Creu i Sant Pau, Autonomous University of Barcelona, Barcelona, Spain 5 Orthopaedics and Traumatology Department, Hospital de la Santa Creu i Sant Pau, Autonomous University of Barcelona, Barcelona, Spain 6 Center for Research in Myology, Sorbonne Université, Inserm, Institut de Myologie, U974, Paris, France 7 Department of Genomics and Precision Medicine, George Washington University School of Medicine and Health Sciences, and Center for Genetic Medicine Research, Children’s National Hospital, Washington, DC, USA Correspondence Xavier Suárez-­Calvet, Neuromuscular Abstract Diseases Unit, Neurology Department, Skeletal muscle contains multiple cell types that work together to maintain tissue Hospital de la Santa Creu i Sant Pau and homeostasis. Among these, satellite cells (SC) and fibroadipogenic progenitors cells Biomedical Research Institute Sant Pau (IIB Sant Pau), C/Sant Quintí 77, 08041 (FAPs) are the two main stem cell pools. Studies of these cells using animal models Barcelona, Spain. have shown the importance of interactions between these cells in repair of healthy Email: xsuarez@santpau.cat muscle, and degeneration of dystrophic muscle. Due to the unavailability of fresh pa- Jordi Díaz-­Manera, John Walton Muscular tient muscle biopsies, similar analysis of interactions between human FAPs and SCs Dystrophy Research Center, Newcastle is limited especially among the muscular dystrophy patients. To address this issue University Translational and Clinical Research Institute, Center for Life, Central here we describe a method that allows the use of frozen human skeletal muscle bi- Parkway, Newcastle upon Tyne NE1 3BZ, opsies to simultaneously isolate and grow SCs and FAPs from healthy or dystrophic UK. patients. We show that while the purified SCs differentiate into mature myotubes, Email: jordi.diaz-manera@newcastle.ac.uk purified FAPs can differentiate into adipocytes or fibroblasts demonstrating their Funding information multipotency. We find that these FAPs can be immortalized and the immortalized AFM-­Téléthon (French Muscular Abbreviations: bFGF, basic fibroblast growth factor; CDK4, cyclin dependent kinase 4; DMD, Duchenne muscular dystrophy; DMEM, Dulbecco’s modified eagle medium; DMSO, dimethyl sulfoxide; FABP4, fatty acid-­binding protein 4; FACS, fluorescence-­activated cell sorting; FAP, fibroadipogenic progenitors; FBS, fetal bovine serum; FMO, fluorescence minus one; hTERT, human telomerase reverse transcriptase; ICW, in-­cell western; iDMD, immortalized DMD FAPs; iFAP, immortalized FAPs; IL-­10, interleukin-­10; IL-­13, interleukin-­13; IL-­4, interleukin-­4; IL-­6, interleukin-­6; M199, medium 199; MD, muscular dystrophy; MOI, multiplicity of infection; MyHC, myosin heavy chain; Nm, nanometer; OR, Oil-­Red; PBS, phosphate buffered saline; PDGFRα, platelet derived growth factor receptor alpha; SC, satellite cell; SD, standard deviation; SEM, standard error of the mean; TGFβ, transforming growth factor β. This is an open access article under the terms of the Creat​ive Commo​ns Attri​butio​n-­NonCo​mmerc​ial-­NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-­commercial and no modifications or adaptations are made. © 2021 The Authors. The FASEB Journal published by Wiley Periodicals LLC on behalf of Federation of American Societies for Experimental Biology The FASEB Journal. 2021;35:e21819.  wileyonlinelibrary.com/journal/fsb2 | 1 of 11 https://doi.org/10.1096/fj.202100588R |    15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2 of 11 SUÁREZ-­CALVET et al. Dystrophy Association), Grant/Award Number: 22525; Instituto de Salud Carlos FAPs (iFAPs) retain their multipotency. These approaches open the door for carrying III, Grant/Award Number: CD18/00195, out personalized analysis of patient FAPs and interactions with the SCs that lead to CM19/00178 and PI18/01525; CIBERER, muscle loss. Grant/Award Number: 2019/29 KEYWORDS Duchenne muscular dystrophy, fibroadipogenic progenitors cells, muscle explant muscle biopsies, satellite cells 1 | IN TRO D U C T ION of other cell types apart of SCs such as FAPs from the same patient and therefore allowing the study of the cell interac- Muscle regeneration and degeneration are complex processes tions leading to a better understanding of the degenerative that have been only partially characterized at the cellular process on each muscle disease and on each individual condi- level in human. A number of different cells in the muscle, in- tion. Thus, it can enhance the development of new treatments cluding muscle-­resident stem cells such as satellite cells (SC) for patients with muscle diseases and facilitate testing per- and fibroadipogenic progenitors cells (FAPs), interact among sonalized based-­mechanism therapies targeting single or both them, with the extracellular matrix and with the lymphocytes cells and cellular interactome. Although FAPs have been iso- and macrophages that infiltrate the muscle during injury.1 lated from fresh human muscles,6,10 studies using FAPs from The microscopic environment where all these interactions patients with a genetically confirmed muscular diseases are take place is commonly known as the cell niche, which is lacking because of the unavailability of fresh muscle tissue. dynamic and contributes to both the regenerative and the de- Our study aims to address this issue by describing a new ap- generative process of injured muscles.2,3 For example, after proach that allows the simultaneous isolation of functional an acute muscle damage, muscle fibers release cytokines that SCs and FAPs from frozen human muscle biopsies and the promote muscle infiltration by inflammatory cells activating generation of immortalized FAPS (iFAPs) which retain their SC and FAPs, while inflammatory cells secrete interleukin-­4 multipotency and efficiently differentiate into fibrogenic and and interleukin-­13 that promote FAP proliferation and main- adipogenic lineages. tain a pro-­ regenerative microenvironment that fosters SC differentiation through interleukin-­6, follistatin, and interleu- kin-­10.1,4 This interaction between SCs, FAPs, and inflam- 2 | M ETHODS matory cells is crucial for normal muscle regeneration after acute injury. During chronic muscle damage, as is the case of 2.1 | Muscle samples muscular dystrophies (MD), muscle fibers are progressively lost and there is a persistent infiltration of muscles by M2 Frozen muscle biopsies from 15 healthy controls and from 1 macrophages which secrete transforming growth factor β Duchenne muscular dystrophy (DMD) patient were included (TGFβ) leading to a persistent accumulation of FAPs which in the study. Control samples were obtained during conven- differentiate into fibrocytes and adipocytes contributing to tional surgeries in collaboration with the Orthopaedic Surgery replacement of muscle fibers by fibrotic and fat tissue.5 The and Plastic Surgery Departments at Hospital de la Santa Creu description of the underlying cellular interactions between I Sant Pau (Barcelona) and patient samples were obtained as FAPs, SCs and inflammatory cells has been described using part of the diagnostic process. We included muscles of 5 men cells isolated from different mouse models.6-­9 and 10 women with a mean age of 44 ± 19 years. Muscle While isolation of human cells from muscle is feasi- fragments included in this study were frozen during 2.2 ± 2.7 ble,6,9,10 studying pathological cell interactions in dystrophic years (Table 1). DMD muscle biopsy was obtained from 8-­ muscle is challenging mainly due to the limited access to year patient with exon 18-­44 deletion in the dystrophin gene. fresh muscle biopsies from patients with genetically con- The DMD muscle biopsy was obtained from the right biceps firmed diagnosis. Muscle biopsies are typically performed which was not weak at clinical examination. as part of the diagnostic process of patients. Once obtained, muscle biopsies are flash-­ frozen in liquid nitrogen or at −80℃ for structural analysis and protein quantification. This 2.2 | Human skeletal muscle approach excludes the possibility of isolating live cells for explants culture cellular studies. However, under specific conditions it is possible to store muscle biopsies that allow isolation of live Fresh muscle biopsies were washed and minced into SCs.11 This approach would potentially enable the isolation 2-­ 3 mm fragments in Hanks’ balanced salt solution | 15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License SUÁREZ-­CALVET et al.     3 of 11 T A B L E 1 Frozen muscle samples from Years healthy controls included in this study Sample Gender Age Muscle frozen %FAP %SC 1 F 35 Triceps 2 86 10 2 F 32 Triceps 2 99 0.1 3 M 12 Ticeps 1 56 34 4 F 57 Pectoralis 1 89 4 5 F 47 Pectoralis 1 87 5 6 F 45 Pectoralis 0.5 90 3 7 M 34 Gastrocnemius 2 75 15 8 F 65 Pectoralis 1 month 88 2 9 F 15 Vastus lateralis 2 months 56 27 10 F 54 Biceps 1 week 93 4 11 M 67 Quadriceps 10 90 10 12 M 54 Triceps 4 77 15 13 F 67 Vastus lateralis 3 99 0 14 M 20 Quadriceps 6 98 0 15 F 59 Pectoral 1 88 9 Note: Muscle explants were cultured in medium I and the proportion of FAP and Satellite cells (SC) cells were determined by platelet derived growth factor receptor alpha and CD56 using flow cytometry. Abbreviations: FAP, fibroadipogenic progenitors; SC, satellite cells. with 1% penicillin–­ streptomycin (PS) (Lonza, Basilea, were removed and re-­ seeded if needed. The attached Switzerland). Muscle fragments were placed in a dish with cells were trypsinized and expanded or resuspended in conditioned media containing M199 (Lonza), 35% fetal bo- fluorescence-­ activated cell sorting (FACS) buffer (PBS vine serum (FBS) (ThermoFisher [Gibco], Waltham, MA), with 2% FBS) for flow cytometry analysis. and 1% PS (Lonza) and incubated for 24 hours at 37℃ and 5% CO2. Then, five to ten muscle fragments were placed in a cryotube containing 1 mL of freezing medium consisting 2.3 | Flow cytometry and cell sorting of Dulbecco’s modified eagle medium (DMEM), 10% FBS and 8% dimethyl Sulfoxide (DMSO).11 These specimens Cells from muscle explants were trypsinized, washed, and were then put into cryofreezers at −80℃ and transferred stained with the following antibodies for 30 minutes at 4℃: to liquid nitrogen after 24 hours. Cryotubes were thawed anti-­ PDGFRα-­ biotin at 2.5 µg/mL (Cat. BAF322, R&D at 37℃ and rapidly transferred to coated dishes with 1.5% Systems, Minneapolis, MN), CD56-­ FITC (Cat. 130-­ 114-­ gelatine plus an equal volume of human plasma. We added 549), CD90-­APC (Cat. 130-­114-­903), CD105-­APC-­Vio770 several drops of proliferation medium to each fragment and (Cat. 130-­112-­326), and CD166-­PE-­Vio770 (Cat. 130-­106-­ incubated it for 30 minutes at 37℃ and 5% CO2 to increase 620) (at the recommended dilutions, all from Miltenyi Biotec, the attachment of muscle fragments to the matrix. Then, Bergisch Gladbach, Germany). After washing steps, we incu- proliferation medium was added until cover the muscle ex- bated streptavidin PE-­Cy5 (Cat. 405202; 1:200; Biolegend, plants in the dish. We cultured muscle explants with two San Diego, CA) for 30 minutes at 4℃. Stained cells were ana- different proliferation media. The medium I consisted of lyzed using a cytometer MACSQuant 10 (Miltenyi Biotec) three parts of DMEM-­GlutaMAX (Gibco) and one part of or sorted on a FACSAria using FACSDiva software (Becton M-­199 (Lonza), with 10% FBS (Gibco), 25 ng/mL of basic Dickinson, Ashland, OR). Doublet cells were excluded using fibroblast growth factor (bFGF), 10 ng/mL of epidermal forward scatter area and height and fluorescence minus growth factor (Peprotech, Rocky Hill, NJ), and 1% PS one (FMO) controls were used to determine positivity. (Lonza). Medium II consisted of DMEM-­GlutaMAX sup- Compensations were adjusted according to the single stained plemented with 20% FBS, 1% PS and 2.5 ng/mL bFGF. controls. PDGFRα+/CD56− fraction was defined as FAPs After 5-­7 days, cells started to sprout from muscle explants and the PDGFRα−/CD56+ fraction as SC. FlowJo v10 soft- and when the number of sprouting cells was enough, ap- ware was used for data analysis and data elaboration (Becton proximately after 1 week but it is very variable, explants Dickinson). |    15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4 of 11 SUÁREZ-­CALVET et al. 2.4 | In vitro differentiation assays M11-­41017; 1:100; Invitrogen), mouse anti-­myogenin (Cat. 556358; 1:100; BD) or with mouse anti-­myosin heavy chain Purified FAPs were subcultured with the medium I and (MF-­20; 1:100; R&D systems) followed by the incubation purified SC were subcultured with Skeletal Muscle Cell of the secondary antibodies donkey anti-­mouse Alexa488 Growth Medium (PromoCell, Heidelberg, Germany). FAPs and donkey anti-­goat Alexa594 as described above. Pictures and SC were induced to fibrogenic differentiation culturing of random fields and of replicates were obtained by using them for 3 days with DMEM-­GlutaMAX with 10% FBS an Olympus BX51 microscope coupled to a DP72 camera. with or without 5 ng/mL TGFβ (R&D Systems). For adi- Quantitative analysis was performed using ImageJ software pogenic differentiation, FAPs and SC were cultured for 14 (National Institute of Health).12 days with StemPro adipogenesis differentiation kit (Gibco). Myogenic differentiation was induced by culturing conflu- ent cells with DMEM-­GlutaMAX with 2% FBS for 9 days. 2.6 | Real-­time PCR At the indicated time points, cells were harvested for gene expression analysis and/or cell staining. SC from digestion We performed gene expression assays using the Taqman were isolated using an enzymatic disgregation with 0.2% Fast Advanced Cells-­to-­CT kit (ThermoFisher) following collagenase and Dispase II 2.4U/mL, expanded in vitro and the manufacturer’s instructions. Briefly, cells were washed sorted as described by CD56. with PBS and incubated with lysis solution for 5 minutes. Sample lysate was mixed with the RT Master mix and run the reaction to obtain cDNA. qPCR was performed by trip- 2.5 | Cell staining licate using the Fast Advanced Master Mix and the 7900 HT Fast Real-­ Time PCR System (ThermoFisher). The We analyzed the expression of collagen I in the cell cultures mRNA-­specific probes were: COL1A1 (Hs00164004_m1), exposed to fibrogenic differentiation by quantitative in-­cell ADIPOQ (Hs00605917_m1), PAX7 (Hs00242962_m1), western (ICW) assays in microplates. Muscle differentiation MYOD1 (Hs00159528_m1), MYOG (Hs01072232_m1), of expanded SCs clones was also assessed by ICW by myosin MYH1 (Hs00428600_m1), and GAPDH (Hs99999905_m1) heavy chain staining (MF-­20). Wells were rinsed with PBS (ThermoFisher). Relative quantification was performed and fixed with cold 4% PFA for 10 minutes. After wash- using the comparative Ct method. ing steps and blocking with casein solution for 30 minutes (ThermoFisher), we incubated goat anti-­collagen I (1:200; Cat. 1310-­01; Southern Biotech, Birmingham, AL) for 1 2.7 | FAP immortalization hour. After washing steps, we incubated for 1-­hour donkey anti-­goat IRDye800 or donkey anti-­mouse IRDye800 sec- Human FAPs were transduced with a retroviral vector con- ondary antibody (1:1000; Li-­COR, Lincoln, NE) and CellTag taining a sequence encoding the catalytic subunit of human IRDye 700 (Li-­COR) following the manufacturer’s instruc- telomerase reverse transcriptase (hTERT) alone or with both tions. After washing with PBS, the fluorescent signal of col- hTERT and cyclin-­dependent kinase 4 (Cdk4) as described lagen I was measured in the 800 nm channel by an Odyssey previously.13 cDNAs were cloned into different lentivi- Imaging system (Li-­COR), and normalized by cell number ral vectors containing puromycin and neomycin selection measured in the 700 nm channel. markers, respectively. FAPs were transduced with lentivi- In adipogenic culture conditions, we analyzed the per- ral vectors with a MOI of 5 in the presence of 4 μg/mL of centage of cells that differentiated to adipocyte by FABP4 polybrene (Sigma-­Aldrich). Transduced cell cultures were immunofluorescence. After fixation and incubation with selected with puromycin (0.2 μg/mL) and neomycin (0.3 Ultracruz blocking solution (Santa Cruz Biotechnology, mg/mL) for 10 days. To assess the most efficient method of Dallas, TX), we incubated the primary antibodies goat anti-­ immortalization, we followed the number of cell divisions FABP4 (Cat. AF3150; 1:100; R&D systems) and rabbit anti-­ during 2 months. perilipin A/B (Cat. P1873-­200UL; 1:100; Sigma-­Aldrich, St. Louis, MO) for 1 hour. After washing steps, we incubated for 1 hour donkey anti-­goat Alexa488 and donkey anti-­rabbit 2.8 | Statistics Alexa594 (1:200; ThermoFisher). Oil Red staining was per- formed using the Lipid Staining kit (Promocell) following the Statistical comparisons between non-­induced and induced manufacturer’s protocols. When indicated, cells were stained conditions were performed using the ratio paired t test. with mouse anti-­desmin (1:100; Leica, Wetzlar, Germany), Statistical analyses and graphic representations were per- goat anti-­PDGFRα (Cat. AF1062; 1:100; R&D Systems), formed with GraphPad Prism Software 8 (La Jolla, CA, mouse anti-­Pax7 (DSHB; 1:100), mouse anti-­MyoD (Cat. USA). | 15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License SUÁREZ-­CALVET et al.     5 of 11 2.9 | Study approval if they retained the expression of PDGFRα considering low passage between 1 and 4 and high passages between 5 and 9 The human studies described here were approved by the in- passages. We observed that the expression levels of PDGFRα stitutional review board of the Hospital de la Santa Creu i were stable during expansion of cells (Figure 1D). Similarly, Sant Pau. Informed consent for the donation of muscles to the percentage of positivity for the mesenchymal markers was our sample collection was received prior to the inclusion to not modified during cell passages: CD105: 97% ± 2% versus the study. 94% ± 4%, CD90: 72% ± 18% versus 86% ± 11%, CD166: 94% ± 3% versus 86% ± 13% (mean ± SD; low passages vs. high passages; n = 3). The use of frozen muscle fragments for 3 | R E S U LTS obtaining human FAP cells opens the door to multiple new research projects such as high throughput screening of drugs 3.1 | Human SC and FAPs can be isolated able to counteract or modulate the differentiation capacity from frozen muscle biopsies of FAPs. In fact, drugs able to modulate FAP cells function are being tested both in animal models and in patients with To assess the utility of frozen muscle biopsies for SCs and MD.15-­21 However, one of the limitations of primary cell lines FAP isolation, we thawed skeletal muscle fragments that is their limited ability to proliferate as these cells enter into were frozen in DMEM plus 10% FBS and 8% DMSO after senescence quickly. To address this limitation we attempted culturing them in the conditioned medium for 24 hours. We to immortalize human FAPs as is commonly done in SCs.22 included samples from five different healthy individuals that We tested two different strategies to generate iFAPs (iFAPs): were frozen for 1 week and up to 10 years ago from (Table 1). (1) transduction of hTERT alone, as has been used for the im- We cultured these muscle explants in two different growth mortalization of skin fibroblast cells23 and (2) transduction of media (Figure 1A). Flow cytometry analysis of the mixture both hTERT and CDK4, as has been used for the immortal- of cells that migrated from these explants identified cells ex- ization of SCs from patients with MD.13,24 Cells transduced pressing PDGFRα+, which is a well-­known marker of FAPS, with hTERT had low proliferation rate, undergoing seven in medium I and II (91% ± 5% and 77% ± 17%; n = 5; mean divisions in almost 2 months, while in contrast, cells trans- ± SD) and also cells expressing CD56+, a marker of SC duced with hTERT and CDK4 underwent seven divisions cells, in both cell media (5% ± 4% and 13% ± 40%; Figure in just 1 month (Figure 1E). We further characterized iFAPs 1B). To further characterize the phenotype of the PDGFRα+ generated by hTERT and CDK4 transduction. Flow cytome- fraction (from now on FAP cells) in the different cell culture try showed that after transduction, equivalent percentage of conditions, we analyzed the expression of the mesenchymal the iFAPs kept the expression of PDGFRα which was not markers CD105, CD90 and CD16614 previously described in modified after cell passages (Figure 1E). CD105 and CD166 human FAPs isolated from fresh tissue.10 In both cell culture expression levels remained stable in iFAPs, but in the case of conditions, most of the FAPs expressed CD105 (92% ± 14% CD90 we observed a prominent reduction of its expression and 99% ± 1%) and CD166 (96% ± 5% and 73% ± 11%). A that was partially recovered after passaging (60% low vs. 40% vast majority of FAPs were also positive for CD90 (83% ± high expression; Figure 1E). 19% and 79% ± 23%) although a short fraction were nega- tive in both culture conditions (Figure 1B). We decided to continue characterizing FAPs isolated just from the medium 3.2 | Purified human FAPs and SCs I because cell markers where similar between cells obtained maintain their differentiation capacity in both culture conditions and medium I is the media most commonly used for human FAP culture. To further examine Once we generated stable primary and immortalized human the utility of our approach to isolate both FAPs and SCs from FAP and SC cells from the same muscle biopsy, the next step the same frozen biopsy, we repeated the process in ten ad- was to assess their differentiation potential. FAPs and iFAPs, ditional muscle biopsies. Flow cytometry analysis confirmed but not SC, cultured in adipogenic medium significantly dif- our previous results demonstrating that this cell culture ap- ferentiated into adipocytes (Figure 2A-­C), as indicated by proach enables the growth of both SC and FAPs from frag- the increased expression of adiponectin gene (ADIPOQ) ments of skeletal muscle (FAPs: 84.7% ± 13%; SC: 9% ± (Figure 2A,B), which is specifically expressed in mature 10%; n = 15; mean ± SD; Table 1). adipocytes.25 ADIPOQ was not detected neither in FAPs nor We purified SCs and FAPs by FACS from three healthy iFAPs at baseline, but was upregulated in both primary FAPs controls and expanded them in vitro for further analysis. The and in iFAPs after culture in adipogenic medium (Figure purity of the cultured SCs and FAPs was confirmed by im- 2A,B). Differentiated FAPs and iFAPs also expressed the munostaning of desmin and PDGFRα (Figure 1C). We cul- late adipogenic differentiation markers FABP4 and per- tured explant-­derived FAP cells for multiple passages to test ilipin and produced fat as was demonstrated by oild red |    15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 6 of 11 SUÁREZ-­CALVET et al. staining (Figure 2A,B). Quantification studies of FABP4+ bigger in adipocytes obtained from FAPs than in adipocytes cells showed a variable degree of adipogenic differentiation obtained from iFAPs (Figure 2A,B). To assess fibrogenic dif- (38% ± 29%) of FAPs isolated from different healthy individ- ferentiation of these cells, we treated FAPs, iFAPs and SCs uals. The proportion of iFAPs differentiated into adipocytes with TGFβ and we quantified collagen-­I expression by in-­ was 42% ± 15% (Figure 2B). Lipid droplets were in general cell western-­blot (ICW) and by qPCR. Both FAPs and iFAPs | 15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License SUÁREZ-­CALVET et al.     7 of 11 F I G U R E 1 Simultaneous isolation of human satellite cells (SC) and FAPs from frozen muscle fragments. Fresh muscle biopsies were cultured for 24 hours in conditioned medium and after 24 hours frozen and stored in liquid nitrogen. After thawing, muscle fragments were cultured as muscle explants and after 7-­10 days, the mixture of cells sprouting out from muscle explant were analyzed (A). Muscle explants cultured in two different media were analyzed by flow cytometry for the presence of SC CD56+ and FAPs PDGFRα+ cells. Both cell culture conditions showed that the mixture of cells contained SC and FAPs which expressed the mesenchymal markers CD105, CD90 and CD166. Grey histograms correspond to fluorescence minus one (FMO) controls and cell counts were normalized to mode. Representative data from five different healthy controls is shown (B). SC and FAPs were sorted and verified by flow cytometry and immunofluorescence (C). Purified FAPs were cultured for multiple passages and platelet derived growth factor receptor alpha (PDGFRα) and mesenchymal markers expression levels were compared. Representative data from additional 10 different healthy controls is shown (D). FAPs from a healthy donor were transduced with hTERT alone or with CDK4 for their immortalization (iFAP). The number of cell divisions after the different transduction strategies was compared. PDGFRα and mesenchymal markers expression levels were compared in iFAPs after multiple passages. Grey histograms correspond to FMO control (E) showed collagen-­I expression at baseline which increased both FAPs and SC. Cell sorting and posterior immunola- significantly in response to TGFβ-­treatment (Figure 2D,E). beling with PDGFRα and desmin staining confirmed their In contrast to FAPs, SCs showed no production of collagen-­I identity (Figure 3A). We analyzed the stability of PDGFRα at baseline or after being cultured with TGFβ (Figure 2D). and the mesenchymal markers during the subculture of FAPs The myogenic potential of SCs isolated from muscle explants from DMD. PDGFRα, CD105, and CD166 expression was was confirmed by the staining of the myogenic transcription consistently detected during cell passages (Figure 3B, upper factors Pax7, MyoD and Myogenin (Figure 2F). Moreover, panel). At low passages, CD90 expression was similar to clonogenic assays supported the self-­renewal property of SCs that of healthy controls but its expression was increased pro- obtained as 52% of the clones were able to proliferate and dif- gressively through the passaging (Figure 3B, upper panel). ferentiate (Supplemental Figure S1). We compared the myo- Having confirmed successful isolation of FAPs and SCs from genic potential of SCs obtained from muscle explants and the patient biopsy, next we immortalized the FAPs (iDMD) SCs obtained from enzymatic muscle digestion. We assessed using the hTERT and CDK4 transduction strategy. Flow cy- the expression of the aforementioned myogenic transcription tometry demonstrated a similar profile of PDGFRα, CD105, factors and of the marker of myotube differentiation myosin and CD166 expression in iDMD compared to non-­iFAPs. heavy chain during 9 days while cells were cultured in a dif- However, in iDMD we observed changes in the CD90 ex- ferentiation medium (Figure 2G). At day 3, we observed an pression profile similar to what we observed in FAPs from upregulation of the expression of the early myogenic tran- healthy controls: 60% of FAP iDMD cells had high CD90 scription factors Pax7 and MyoD that decreased at day 6 and expression while the 40% of the cells expressed CD90 at low 9. In the case of Myogenin, which is expressed later during levels and became negative after passages (Figure 3B, down myogenic differentiation, expression levels sharply increase panel). High CD90 expression levels have been related to an at day 3 and then plateaued during day 6 and 9. Myosin heavy undifferentiated status of mesenchymal cells. In fact, a de- chain expression levels progressively increased during the crease in CD90 level can be correlated with the temporal lin- follow-­up peaking up at day 9 (Figure 2G). A similar pro- eage commitment in vitro26 and therefore we could speculate file of expression was observed between SCs obtained from that the decrease of CD90 expression in iFAPs may reflect a explants and SCs obtained from muscle digestion. Although reduction of multilineage capacity although our in vitro ex- the magnitude of the upregulation of MyHC in SCs from di- periments do not reveal such a possible explanation. We then gestion was less pronounced, mature myotubes expressing performed differentiation assays with primary DMD FAPs MyHC in both SC types were observed while no myotubes and the immortalized iDMD FAPs. Adipogenic induction were observed when FAPs or iFAPs were cultured in myo- showed that both FAP cells differentiated into adipocytes genic conditions (Figure 2H). being positive for perilipin and FABP4 and accumulating lipid droplets as was shown using oil red staining (Figure 3C). The differentiation index was similar between primary 3.3 | FAPs and SCs can be purified from and immortalized cells in DMD: 57% ± 8% and 53% ± 14% frozen biopsies of muscular dystrophy patients of cells differentiated into adipocytes, respectively (Figure 3C). Gene expression analysis revealed undetectable adi- In view of the potential of our approach to isolate matched ponectin levels at baseline, which was significantly upregu- FAPs and SCs from frozen muscle biopsies from healthy do- lated in adipogenic culture conditions in both primary and nors, we explored if these cells could also be obtained from immortalized DMD FAPs (Figure 3B). Similarly, treatment frozen muscle biopsies of patients with MD. We used a fro- with TGFβ-­induced DMD FAPs and iDMD FAPs to adopt a zen muscle biopsy of a DMD patient. Flow cytometry analy- fibrogenic fate with a significant increase in their production sis of the cells growing from the cultured explants identified of collagen-­I (Figure 3D). As observed in healthy controls, |    15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 8 of 11 SUÁREZ-­CALVET et al. F I G U R E 2 Differentiation of FAPs, satellite cells (SCs), and immortalized FAPs (iFAPs) obtained from frozen muscle explants. FAPs, iFAPs and SCs from healthy donors were cultured with adipogenic medium. Oil-­red staining (OR) and fatty acid-­binding protein 4 (FABP4) and perilipin co-­staining showed that FAPs (A) and iFAPs (B) differentiated to adipocytes while SC did not (C). The percentage of adipocytes was determined by the number of FABP4-­positive cells. A minimum of four random fields in each of the three healthy controls were analyzed. Adiponectin (ADIPOQ) gene expression was significantly increased in FAPs (P = .037) (A) and in iFAPS (P = .035) (B). Analysis of collagen I by ICW on FAPs, iFAPs and SC cultured in fibrogenic medium showed a significant increase in FAPs (P = .041) and iFAPs (P = .026) of collagen I in fibrogenic conditions, while SC did not (P = .193) (D). qPCR gene expression analysis showed a significant increase of collagen I in FAPs (P = .019), iFAPs (P = .011) but not in SC (P = .37) (E). Immunolabeling of the myogenic transcription factors Pax7, MyoD and myogenin in SC isolated from muscle explants (F). Gene expression analysis of Pax7, MyoD1, Myogenin and myhc1 at 3, 6 and 9 days of differentiating SC isolated from muscle explants or from muscle digestion (G). Confluent culture with low-­serum conditions and MyHC staining was performed to identify mature myotubes. Both SC from explants or from digestion showed the presence of MyHC-­expressing myotubes while FAPs and iFAPs did not (H). Data from FAPs from 3 healthy donors is shown. In experiments with iFAPs, technical replicates are shown. In ICW, data from five replicates is shown and was normalized to the number of cells. The last well of each condition corresponds to the negative control for collagen I staining. qPCR data was normalized to GAPDH endogenous control. Data from SC-­explants from 3 healthy donors are shown while technical replicates from one individual is shown in SC from digestion. Standard error of the mean is shown. ICW, in-­cell western-­blot; MyHC, myosin heavy chain; NT, non-­treated. Statistical comparisons between non-­induced and induced conditions were performed using ratio paired t test. *P ≤ .05 the SC cell fraction obtained from the DMD patient did not expressing MHC were only obtained after culturing DMD differentiate into adipocytes and did not express collagen- SCs, and not DMD FAPs or iDMD FAPs for 9 days in low-­ I in response to TGFβ (Figure 3E). However, myotubes ­ serum culture medium (Figure 3D). In summary, although | 15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License SUÁREZ-­CALVET et al.     9 of 11 F I G U R E 3 Isolation, immortalization and differentiation of FAPs from frozen Duchenne muscular dystrophy (DMD) muscle explant. Frozen muscle explants from a DMD patient were thawed and cultured to allow cells to sprout out from the explant. These cells were analyzed by flow cytometry. CD56+ cells were identified as satellite cells (SCs) and PDGFRα+ as FAPs. Immunofluorescence confirmed the identity of both cell types by the staining of desmin and PDGFRα (A). The expression levels of PDGFRα and the mesenchymal markers CD105, CD90, and CD166 were assessed by flow cytometry in DMD and in iDMD (immortalized) FAP cells during multiple passages. Grey histograms correspond to fluorescence minus one (FMO) controls and cell counts were normalized to mode (B). Adipogenic induction in DMD and iDMD FAP cells promote the presence of lipid droplets observed in the OR staining. Perilipin and FABP4 stainings demonstrated their adipogenic fate, and the percentage of adipocytes was calculated based on FABP4-­positive cells. Gene expression analysis demonstrated a significant upregulation of adiponectin (ADIPOQ) in DMD (P = .007) and in iDMD (P = .007) FAPs (C). ICW showing a significant overexpression of collagen I after culturing DMD (P = .05) and iDMD (P = .04) FAPs with transforming growth factor β (TGFβ) (D). SC fraction from a DMD patient was exposed to adipogenic, myogenic, and fibrogenic medium. OR staining showed the absence of adipocytes, MyHC staining showed the presence of myotubes and ICW showed no effect of TGFβ on collagen I expression (E).; ICW, in-­cell western-­blot; MyHC, myosin-­heavy chain; NT, non-­treated; OR, oil-­Red. Statistical comparisons between non-­induced and induced conditions were performed using ratio paired t test *P ≤ .05; **P ≤ .01 we only tested one sample from a DMD patient, these results FAPs obtained from patients with DMD and other MD and support that FAPs can be obtained from frozen DMD muscle studying their interactions with SCs obtained from the same biopsies, opening the door to further studies characterizing patient. |    15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 10 of 11 SUÁREZ-­CALVET et al. 4 | D IS C U SS ION properties and how the cells interact among them. Such pos- sibilities enhance the impact of the approach described here One of the main limitations for the study of the mechanism for studying human skeletal muscle biology under healthy leading to muscle degeneration in humans is the lack of avail- and pathologic conditions, leading to translational impact on able tissue for the isolation of cells. The approach that we the discovery of new mechanism and subconsequently to the describe here allows freezing muscle in optimum conditions development of new therapies for these disorders. for cell culture enabling the posterior isolation of FAPs and SCs concomitantly from the same patient. We have exten- ACKNOWLEDGEMENTS sively applied our protocol to healthy muscle biopsies and XSC was funded by Instituto de Salud Carlos III (Sara verified our approach in a muscle sample obtained from a Borrell fellowship, CD18/00195) and co-­funded by European DMD patient. We cannot obtain any conclusions from a bio- Regional Development Fund (ERDF)/European Social Fund logical point of view in the sample from a DMD patient we (ESF), “investing in your future”. JAP was funded by Instituto tested in terms of the ratio of FAPs/SCs obtained or the dif- de Salud Carlos III (Rio-­Hortega fellowship, CM19/00178) ferentiation potential of FAPs but this experiment supports and co-­funded by ERDF/ESF, “investing in your future”. that our protocol can be applied to samples from patients This study has been funded by Instituto de Salud Carlos III with muscle diseases. Muscle biopsies are commonly per- (Project number PI18/01525), co-­funded by European Union formed for diagnostic purposes and therefore, the diagnosis (ERDF) “A Way to make Europe”, by AFM (France) through is not known when the muscle biopsy is obtained limiting the the grant number 22525, by Fundación Isabel Gemio and by inclusion of these fresh muscle samples in any predesigned CIBERER ACCI 2019/29. The Neuromuscular Diseases research study. Moreover, the popularization of next genera- Unit is member of the European Reference Network for tion sequencing as standard for diagnostic in many countries Neuromuscular Diseases (ERN-­ EURO-­ NMD). CERCA has reduced the number of biopsies performed. Obtaining a Programme/Generalitat de Catalunya. We thank Myoline, new muscle biopsy from a patient with a confirmed genetic the platform for immortalization of human cells from the disease is ethically questionable, as the procedure is invasive Institute of Myology (Paris, France) and Valentine Hanique and painful for patients. Additionally, MD are rare diseases for technical assistance. and the inclusion of several muscle biopsies in a short pe- riod of time for research studies is commonly not feasible. CONFLICTS OF INTEREST We have demonstrated that FAP cells can be isolated from The authors have declared that no conflict of interest exists. frozen muscle and can be expanded in vitro maintaining the expression of cell specific markers without limiting their AUTHOR CONTRIBUTIONS differentiation potential. FAPs can undergo adipogenic or X. Suárez-­ Calvet, JK. Jaiswal, and J Díaz-­ Manera de- fibrogenic fate using specific culture conditions and SCs dif- signed research. X. Suárez-­ Calvet, E. Fernández-­ Simón, ferentiate producing mature myotubes. The standard use of P. Piñol-­Jurado, J. Alonso-­Pérez, A. Carrasco-­Rozas, and the approach used here by worldwide biobanks and special- C. Lleixà performed research. A. Bigot and V. Mouly per- ized research laboratories would enable long-­term storage of formed immortalization experiments. X. Suárez-­ Calvet, muscles and the design of experiments based in the obtention E. Fernández-­ Simón, and J. Díaz-­ Manera analyzed data. of SCs, FAPs and other muscle cells from the same patient, S. López-­Fernández, G. Pons, L. Soria, I. Illa, and J. Díaz-­ to better understand the pathomechanims of these diseases. In Manera provided human samples. X. Suárez-­ Calvet, this sense, the knowledge on the physiopathology of muscu- E. Gallardo, and J Díaz-­Manera obtained funding. X. Suárez-­ lar diseases has experienced an authentic revolution thanks to Calvet and J. Díaz-­Manera wrote the paper with contribu- new techniques such as single cell RNA sequencing or prot- tions from JK. Jaiswal and E. Gallardo. eomics characterization of muscle cells obtained from animal models. In recent years several subpopulations of muscle res- ORCID ident cells have been identified as the most prevalent at dif- Xavier Suárez-­Calvet https://orcid. ferent stages of disease, remarking the heterogeneity of cells org/0000-0002-5314-6607 participating in the process of muscle degeneration and sug- gesting that this is a highly dynamic process.27-­29 However, R E F E R E NC E S there is a clear gap of knowledge in humans, as there is only 1. Farup J, Madaro L, Puri PL, et al. Interactions between mus- few data coming from a minority of studies done with cells cle stem cells, mesenchymal-­derived cells and immune cells in from healthy donors.30,31 Our approach could enhance basic muscle homeostasis, regeneration and disease. Cell Death Dis. 2015;6:e1830. research in human samples revealing what are the subpopu- 2. Chazaud B. Inflammation and skeletal muscle regeneration: leave lations of muscle resident cells more prevalent in patients at it to the macrophages! Trends Immunol. 2020;41:481-­492. different stages of the disease, what are their main biological | 15306860, 2021, 9, Downloaded from https://faseb.onlinelibrary.wiley.com/doi/10.1096/fj.202100588R by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [22/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License SUÁREZ-­CALVET et al.     11 of 11 3. Munoz-­ Canoves P, Serrano AL. Macrophages decide between 20. Mozzetta C, Consalvi S, Saccone V, et al. Fibroadipogenic progen- regeneration and fibrosis in muscle. Trends Endocrinol Metab. itors mediate the ability of HDAC inhibitors to promote regenera- 2015;26:449-­450. tion in dystrophic muscles of young, but not old Mdx mice. EMBO 4. Heredia J, Mukundan L, Chen F, et al. Type 2 innate signals stimu- Mol Med. 2013;5:626-­639. late fibro/adipogenic progenitors to facilitate muscle regeneration. 21. Sandonà M, Consalvi S, Tucciarone L, et al. HDAC inhibitors tune Cell. 2013;153:376-­388. miRNAs in extracellular vesicles of dystrophic muscle-­resident 5. Uezumi A, Fukada S-­I, Yamamoto N, et al. Mesenchymal progeni- mesenchymal cells. EMBO Rep. 2020;21:e50863. tors distinct from satellite cells contribute to ectopic fat cell forma- 22. Massenet J, Gitiaux C, Magnan M, et al. Derivation and characteriza- tion in skeletal muscle. Nat Cell Biol. 2010;12:143-­152. tion of immortalized human muscle satellite cell clones from muscular 6. Latroche C, Weiss-­Gayet M, Gitiaux C, et al. Cell sorting of var- dystrophy patients and healthy individuals. Cells. 2020;9:1780. ious cell types from mouse and human skeletal muscle. Methods. 23. Auré K, Mamchaoui K, Frachon P, et al. Impact on oxidative 2018;134-­135:50-­55. phosphorylation of immortalization with the telomerase gene. 7. Perdiguero E, Moiseeva V, Munoz-­Canoves P. Simultaneous iso- Neuromuscul Disord. 2007;17:368-­375. lation of stem and niche cells of skeletal muscle: applicability for 24. Thorley M, Duguez S, Mazza EMC, et al. Skeletal muscle charac- aging studies. Methods Mol Biol. 2019;2045:13-­23. teristics are preserved in hTERT/cdk4 human myogenic cell lines. 8. Tucciarone L, Etxaniz U, Sandona M, et al. Advanced methods to Skelet Muscle. 2016;6:43. study the cross talk between fibro-­adipogenic progenitors and mus- 25. Körner A, Wabitsch M, Seidel B, et al. Adiponectin expression in cle stem cells. Methods Mol Biol. 2018;1687:231-­256. humans is dependent on differentiation of adipocytes and down-­ 9. Uezumi A, Kasai T, Tsuchida K. Identification, isolation, and regulated by humoral serum components of high molecular weight. characterization of mesenchymal progenitors in mouse and human Biochem Biophys Res Commun. 2005;337:540-­550. skeletal muscle. Methods Mol Biol. 2016;1460:241-­253. 26. Sibov TT, Severino P, Marti LC, et al. Mesenchymal stem cells from 10. Uezumi A, Fukada S, Yamamoto N, et al. Identification and char- umbilical cord blood: parameters for isolation, characterization and acterization of PDGFRalpha+ mesenchymal progenitors in human adipogenic differentiation. Cytotechnology. 2012;64:511-­521. skeletal muscle. Cell Death Dis. 2014;5:e1186. 27. Giordani L, He GJ, Negroni E, et al. High-­dimensional single-­cell 11. Carter LS, Askanas V. Vital-­freezing of human muscle cultures for cartography reveals novel skeletal muscle-­resident cell popula- storage and reculture. Muscle Nerve. 1981;4:367-­369. tions. Mol Cell. 2019;74(609-­621):e6. 12. Schindelin J, Arganda-­Carreras I, Frise E, et al. Fiji: an open-­ 28. Oprescu SN, Yue F, Qiu J, et al. Temporal dynamics and hetero- source platform for biological-­ image analysis. Nat Methods. geneity of cell populations during skeletal muscle regeneration. 2012;9:676-­682. iScience, 2020;23:100993. 13. Mamchaoui K, Trollet C, Bigot A, et al. Immortalized pathological 29. Petrilli LL, Spada F, Palma A, et al. High-­dimensional single-­cell human myoblasts: towards a universal tool for the study of neuro- quantitative profiling of skeletal muscle cell population dynamics muscular disorders. Skelet Muscle. 2011;1:34. during regeneration. Cells. 2020;9:1723. 14. Halfon S, Abramov N, Grinblat B, et al. Markers distinguishing 30. De Micheli AJ, Spector JA, Elemento O, et al. A reference single-­ mesenchymal stem cells from fibroblasts are downregulated with cell transcriptomic atlas of human skeletal muscle tissue reveals bi- passaging. Stem Cells Dev. 2011;20:53-­66. furcated muscle stem cell populations. Skelet Muscle. 2020;10:19. 15. Consalvi S, Saccone V, Mozzetta C. Histone deacetylase inhibitors: 31. Rubenstein AB, Smith GR, Raue U, et al. Single-­cell transcrip- a potential epigenetic treatment for Duchenne muscular dystrophy. tional profiles in human skeletal muscle. Sci Rep. 2020;10:229. Epigenomics. 2014;6:547-­560. 16. Hogarth MW, Defour A, Lazarski C, et al. Fibroadipogenic pro- genitors are responsible for muscle loss in limb girdle muscular SUPPORTING INFORMATION dystrophy 2B. Nat Commun. 2019;10:2430. Additional supporting information may be found online in 17. Ito T, Ogawa R, Uezumi A, et al. Imatinib attenuates severe mouse the Supporting Information section. dystrophy and inhibits proliferation and fibrosis-­marker expres- sion in muscle mesenchymal progenitors. Neuromuscul Disord. 2013;23:349-­356. How to cite this article: Suárez-­Calvet X, Fernández-­ 18. Kasai T, Nakatani M, Ishiguro N, et al. Promethazine hydrochlo- Simón E, Piñol-­Jurado P, et al. Isolation of human ride inhibits ectopic fat cell formation in skeletal muscle. Am J fibroadipogenic progenitors and satellite cells from Pathol. 2017;187:2627-­2634. frozen muscle biopsies. FASEB J. 2021;35:e21819. 19. Mazala DA, Novak JS, Hogarth MW, et al. TGF-­beta-­driven mus- https://doi.org/10.1096/fj.20210​0588R cle degeneration and failed regeneration underlie disease onset in a DMD mouse model. JCI Insight. 2020;5:e135703.