The panniculus carnosus muscle: A novel model of striated muscle regeneration that exhibits sex differences in the mdx mouse

The dermal striated muscle panniculus carnosus (PC), prevalent in lower mammals with remnants in humans, is highly regenerative, and whose function is purported to be linked to defence and shivering thermogenesis. Given the heterogeneity of responses of different muscles to disease, we set out to characterize the PC in wild-type and muscular dystrophic mdx mice. The mouse PC contained mainly fast-twitch type IIB myofibers showing body wide distribution. The PC exemplified heterogeneity in myofiber sizes and a prevalence of central nucleated fibres (CNFs), hallmarks of regeneration, in wild-type and mdx muscles, which increased with age. PC myofibers were hypertrophic in mdx compared to wild-type mice. Sexual dimorphism was apparent with a two-fold increase in CNFs in PC from male versus female mdx mice. To evaluate myogenic potential, PC muscle progenitors were isolated from 8-week old wild-type and mdx mice, grown and differentiated for 7-days. Myogenic profiling of PC-derived myocytes suggested that male mdx satellite cells (SCs) were more myogenic than female counterparts, independent of SC density in PC muscles. Muscle regenerative differences in the PC were associated with alterations in expression of calcium handling regulatory proteins. These studies highlight unique aspects of the PC muscle and its potential as a model to study mechanisms of striated muscle regeneration in health and disease.


Results
Muscle fibre types in PC of wild-type and dystrophic mdx mice. Murine skin is composed of multiple layers from the outer epidermis, dermis, panniculus adiposus (PA) and a thin 3-5 myofiber wide (~100 μm) striated muscle layer, located between the PA and interstitial connective tissue (ICT) layers on the dorsum of mice called the panniculus carnosus (PC) (Fig. 1A). These same tissue layers visualised with haematoxylin and eosin (H&E) staining of transverse dorsal skin sections from both wild-type and mdx mice (Fig. 1C). The PC is reported to be composed of striated muscle fibres expressing fast type II MyHC in normal mice 25 . To delineate which subtypes of myosin expressing fibres are present in the PC in wild-type and mdx mice, we subjected muscle sections to a panel of MyHC isoform specific antibodies. The PC fast fibre phenotype was confirmed by cross-reaction with a pan anti-fast myosin antibody in both wild-type and mdx transverse dermal sections (data not shown). Furthermore, the PC had strongest cross-reaction with an anti-MyHC-IIB specific antibody indicating a prevalence of fast-glycolytic type IIB muscle fibres ( Fig. 1C in red). A much lower frequency of muscle fibres immunopositive for an antibody against MyHC-IIA ( Fig. 1C in green) was found, indicating the low presence of fast-oxidative-glycolytic type IIA fibres. There was no cross-reaction of fibres against an anti-slow myosin MyHC-I antibody (data not shown). The muscle genotype of mouse tissues was confirmed by incubating sections with an anti-dystrophin antibody, where wild-type PC tissues were immunopositive for dystrophin in the surrounding sarcolemma, and mdx PC muscle sections were negative for dystrophin (Fig. 1C).

Whole-body wide distribution of PC muscle.
To understand the whole-body distribution of PC in the healthy mouse and the effect of mdx dystrophy on the diverse anatomical locations of PC, we harvested dermal fragments from eleven different locations around the whole body of 1-year old wild-type and mdx mice ( Fig. 2A). Paraffin embedded fragments were sectioned and stained with Trichrome blue (Fig. 2B,C). A thin 3-5 myofiber (~100 μm) wide sub-dermal striated muscle layer consistent with the PC location was robustly stained rich red in all skin cross-sections examined (Fig. 2B,C). The gastrocnemius muscle was included as a control. The PC was widely distributed across almost all the different body skin fragments from both wild-type and mdx mice. Noticeably, the PC muscle layer was thicker (~200 μm) in dermal fragments from mdx compared to wild-type mice (Fig. 2B,C and Suppl. Fig. S2). Furthermore, the PC appeared to be thicker in dorsal compared to ventral sections and with a tendency to be wider in rostral compared to caudal sections in both wild-type and mdx mice.
Regeneration in PC muscle from wild-type and mdx mice. To examine the regeneration levels and the dystrophic histopathology of the PC muscle, we scored myofibers with central nuclei in H & E stained sections at 6 and 12 weeks in both groups (Fig. 3A,B). The mdx fibres showed a greater percentage of CNFs compared to wild-type in both 6-wk and 12-wk old mice. A 4-fold higher CNF level was seen in young mdx mice (29.6 ± 7.1%) when compared to their aged matched counterparts (7.9 ± 1.2%). A further elevation was observed in www.nature.com/scientificreports www.nature.com/scientificreports/ older mdx mice with nearly a 5-fold (53 ± 7.7%) higher frequency of CNFs compared to matching wild-type mice (11.9 ± 1.8%). No significant difference was seen between the percent of centrally nucleated myofibers between the two ages in the control mice (Fig. 3B). As a comparison, CNFs in the gastrocnemius muscles of healthy mice are <1% (see below and Fig. 4D).
Heterogeneity of PC muscle fibre size in wild-type and mdx. To further evaluate the effects of age and dystrophy on the PC muscle, we examined the homogeneity of the PC myofiber size. Dermal tissue sections were fluorescently stained with WGA-Alexa 488 to demarcate the membrane of each myofiber, to facilitate morphometric analysis (Suppl. Fig. S1). As a morphometric index, minimal Feret's diameter was measured on stained PC transverse sections in 6-week and 12-week old wild-type and mdx mice respectively (Fig. 3C,D). The frequency distribution of myofiber diameters was shifted to larger myofibers in mdx compared to age-matched wild-type mice at both ages (Fig. 3C). The mean diameter at 6 weeks (19.9 ± 1.6 µm) was increased by 1.4-fold in mdx compared to its wild-type control (14.2 ± 1 µm) but this was not significant (Fig. 3D). In contrast, the mean diameter at 12 weeks was significantly increased by 1.9-fold in mdx PC (36.8 ± 2.7 µm) above its wild-type control (19.4 ± 1.2 µm) ( Fig. 3D; p < 0.001). These morphometric indices increased with age from 6 to 12 weeks of age in the mdx mouse (Fig. 3C,D; p < 0.001). Haematoxylin & Eosin (H&E) stained transverse section of paraffin embedded skin tissue from the superior-mid dorsum of a 6-week old male mdx mouse. The PC is a 3-5 myofiber wide striated muscle layer, located beneath the dermal white adipose tissue (dWAT). Arrows indicate hair follicles. ICT: Interstitial connective tissue; *PVDF membrane; Scale bar = 200 µm. (C) Sample of paraffin-preserved mid dorsal skin cross-section histochemically stained with H&E demonstrating the anatomical location of the PC muscle in 8-week old male wild-type and mdx skin tissues. Scale bar is 200 µm. PC unfixed frozen cross-sections from 8-week old male wild-type and mdx mice, immunofluorescently stained with anti-myosin IIA (green), anti-myosin IIB antibodies (red) and nuclei stained with DAPI (blue). Scale bar = 50 µm. Immunofluorescent staining for dystrophin confirms the mdx genotype. Positively stained fibres for dystrophin in 8-week old male wild-type mid-dorsal PC striated muscle in contrast to mdx fibres, which are null for dystrophin. Nuclei stained with DAPI (blue). Scale bar = 50 µm.
Noticeably, there was a variation in the thickness of the three dermal layers: dermis, dermal white adipose tissue (dWAT) and PC muscle, between the groups based on age and genotype (Suppl. Fig. S2). Wild-type and mdx 6-week-old mice had similar relative thickness for the three dermal skin layers. In 12-week-old mdx mice there were significant alterations in all dermal layers. The dermis was reduced by half in the mdx compared to wild-type mice (p < 0.01). The dWAT layer was increased 2-fold (p < 0.05) and the PC muscle was increased 2.3-fold (p < 0.001) in the 12-week mdx mice compared their wild-type counterparts. Dermal layers were also significantly different between 6 and 12 week mdx mice with increases in dWAT (p < 0.01) and PC (p < 0.0001) layers, and a decrease in the dermis (p < 0.001).
Sex-related differences in PC and limb muscle degeneration/regeneration in wild-type and mdx mice. To evaluate the effect of sex on the degenerative/regenerative behaviour, PC tissues from 8-week-old male and female wild-type and mdx mice were analysed for CNFs. In the male mdx 35% of PC muscle fibres had CNFs, which is 7-fold higher than the mean percentage recorded in normal male PC (p < 0.001). Furthermore, a  64 . (B,C) Representative histochemical Masson's Trichrome blue stain of paraffin-preserved skin fragment cross-sections from 11 anatomical locations from 1-year old male wild-type and mdx mice. The outer epidermis layer appears red, dermis layer is blue, and the striated muscle PC layer is rich red. A gastrocnemius muscle cross-section is included as a control for skeletal muscle staining. Scale bar = 200 μm.
sex effect was shown in mdx in which the percentage of CNFs in female mdx (15.5 ± 0.02%) was 2.3 times lower than the percentage shown in male mdx (Fig. 4A,B; p < 0.001). In contrast, no differences were found in gastrocnemius limb muscles of male and female, wild-type and mdx mice (  week-old wild-type and mdx mice. CNFs were quantified from 4-5 non-overlapping fields per muscle section at 10x magnification. Data were analysed by 2-way ANOVA and Tukey's multiple comparison test (*p < 0.05, ***p < 0.001; n = 4 mice). (C) Muscle fibre size distribution based on Feret's diameter measurements on PC tissues from 6 and 12-week-old male wild-type and mdx mice. Fibre sizes were determined from 4-5 non-overlapping fields per muscle section at 20x magnification Data bin sizes were analysed by t-tests and are represented as means ± SEM; *p < 0.05,**p < 0.01,***p < 0.001 and ****p < 0.0001; n = 4 mice for each genotype. (D) Mean myofiber diameters in 6-week and 12-week old wild-type and mdx male mice. Data were analysed by 2-way ANOVA and Tukey's multiple comparison posthoc test. Data are represented as means + SEM; ***p < 0.01; n = 4 mice for each sex and genotype. Data are means + SEM with n = 4 mice. (2019) 9:15964 | https://doi.org/10.1038/s41598-019-52071-2 www.nature.com/scientificreports www.nature.com/scientificreports/ Dermospheres started to cluster from day 2 (Suppl. Fig. S3). However, no obvious difference was noted between the dermosphere formation or expansion in wild-type and mdx cultures. Equal densities of progenitor cells were allowed to differentiate for 7 days in differentiation medium. Multi-nucleated myotubes were visualized by the third day and myofiber twitching was observed by the 5 th -6 th day (Suppl. Movies S1). Myogenesis markers Pax7, Myog, MyHC2 (MyHC-2A) and MyHC3 (embryonic myosin) were examined at mRNA level on days 1, 3 and 7. Pax7 was detected on day 1 (Fig. 5A) and was still present on day 3 (data not shown). Myogenin was absent on day 1, detected on day 3 ( Fig. 5A,B) and continued to be detectable on day 7 (data not shown). Myotubes readily expressed the adult MyHC by day 7 and showed the differentiated sarcomere organization (Fig. 5A). Immunofluorescence analysis revealed that while genotype had no influence on myogenin expression, sex had an effect both in wild-type and mdx mice. The mean percentage of myogenin positive nuclei in both female wild-type (8.27 ± 0.56%) and mdx (8.84 ± 0.76%) was significantly lower by 1.5-fold compared to their male counterparts, 13.13 ± 1.6% and 13.39 ± 0.9% respectively ( Fig. 5B; p < 0.05).
To investigate whether genotype/sex has an effect on the myogenesis of PC satellite cells, mRNA levels were examined for Pax7, myogenin, embryonic myosin (MyHC3) and adult myosin (MyHC2) in PC muscle progenitors from both male and female wild-type and mdx mice at day 1, 3 and 7 of differentiation ( Fig. 6A-D). In general Pax 7 levels decreased between 1 and 7 days of differentiation (p < 0.001) except in female mdx mice where they remained unaltered (Fig. 6A). Relative expression of Pax7 showed no genotype effect between wild-type and mdx except it was higher in male mdx on day-3. Although, there was no difference between male and female wild-type cells, Pax7 expression was 4-fold higher in male mdx compared to female mdx on day-3 (p < 0.01). Moreover, a 3-way ANOVA showed a significant interaction between mouse genotype and sex (p < 0.01).
Myogenin peaked on day-3 of differentiation for both sexes and genotypes (p < 0.01). Furthermore, myogenin in male mdx was 4.4 times the evident levels in female mdx on day 3 (Fig. 6B, p < 0.05). Embryonic myosin transcripts (MyHC3) were elevated on day-3 and day-7 compared to day-1 for both sexes and genotypes ( Fig. 6C ANOVA, p < 0.01;). However, MyHC3 levels were generally lower in mdx females for all times points and a 3-way ANOVA indicated a significant interaction for sex and genotype (p < 0.05) despite post-hoc tests not revealing significance for a specific time point.
As expected, transcriptional level of adult myosin heavy chain (MyHC2) was most prevalent on day-7 in all groups. Noticeably, male mdx PC-derived myotubes exhibited 2.4 times the expression of MyHC2 than their wild-type counterparts (p < 0.0001; Fig. 6D). Moreover, male mdx cells demonstrated a 3.5-fold higher MyHC2 mRNA level compared to female mdx on day-7 (p < 0.0001; Fig. 6D). www.nature.com/scientificreports www.nature.com/scientificreports/ Immunodetection and quantification of satellite cells (SCs) in PC muscles from male and female wild-type and mdx mice. Satellite cells (SCs) were identified by expression of Pax7 associated with nuclear staining and juxtaposed to the sarcolemma of muscle fibres as delineated by laminin immunostaining (Fig. 7A). Pax7 immunofluorescently stained PC muscle sections were quantified from both male and female wild-type and mdx mice (Fig. 7B,C). There were no significant differences between groups.
Sex-related differences in the expression of genes encoding Ca 2+ -handling proteins in the PC from male and female mdx mice. The mRNA of five (Serca1a, Sln, Casq, Pvalb and Cam) of the seven transcripts quantified were expressed at significantly higher levels in PC from female mdx compared to the male mdx mice (Fig. 8). Although Serca2a mRNA was expressed at low absolute levels in all PC tissues, male mdx PC exhibited higher www.nature.com/scientificreports www.nature.com/scientificreports/ expression of Serca2a than wild-type male. No significant differences were seen between male and female levels for Capn3 mRNA in wild-type (p = 0.09) or mdx mice (p = 0.07). The comparison shows that the mRNA for five out of seven calcium regulatory proteins were expressed at higher levels in the PC of female compared to male mdx mice.  www.nature.com/scientificreports www.nature.com/scientificreports/

Discussion
The regenerative capacity of different skeletal muscles is heterogeneous. This heterogeneity has been attributed to extrinsic and intrinsic differences including differences in the muscle specific stem cell populations 6,33 . There is an interest in understanding the variation in muscle regenerative capacity of particular muscles when injured or diseased as it may reveal mechanisms for therapy 34,35 . The dermal associated PC is a relatively unfamiliar striated muscle and although more commonly found in lower mammals; in humans, remnants of this tissue do exist as the platysma muscle of the neck, throat, palmaris brevis in the hand 36 and the cremaster muscle in the scrotum 37 . The PC exhibits unique features including high levels of both remodelling and exogenous cell engraftment for reasons not clearly understood 25,28 . The PC also has a major advantage as being very accessible compared to other body musculature. The superficial location and thinness of the PC make it ideal for in vivo live imaging, which has been overlooked for studying muscle degenerative disorders 38 . In view of these features it was of interest to study the regeneration of PC in the mdx mouse model of DMD.
In present study we confirmed that the PC in healthy mice shows heterogeneity in fibre size and elevated central nucleated fibres (CNFs) in the absence of any damage, as previously reported 25 . These are two morphological features of myofibers undergoing remodelling or regeneration in postnatal skeletal muscle 39,40 . The relative high turnover in healthy PC was exacerbated in the mdx mouse, with both an increase in heterogeneity of fibre sizes and frequency of CNFs, typical of affected dystrophic muscles. There was a progressive increase in PC remodelling and regeneration with age, in both wild-type and dystrophic muscles respectively. The high myofiber turnover in the PC is not unique to this striated muscle as the craniofacial muscles including the masticatory muscles (MM), extraocular muscles (EOM) 41,42 and laryngeal muscles (LM) 43 have similar features that collectively set them apart from the majority of muscles (Review 44 ). The craniofacial muscles in common with PC continually undergo myofiber remodelling 45 . The EOM has high regenerative capacity and remodeling activity, due to satellite cells with high proliferation potential and continual cell fusion respectively 34,46,47 . Additionally, the craniofacial muscles are spared from the severe dystrophic phenotype in DMD compared to other muscles. In contrast, our data revealed the PC was not spared in the mdx mouse.
The present study shows that the subdermal PC muscle is widely distributed in the mouse, with variable thickness which is comparable to earlier anatomical reports on PC in rodents and other animal species 48,49 . Noticeably, www.nature.com/scientificreports www.nature.com/scientificreports/ PC myofibers in mdx mice appeared hypertrophic and continued to exhibit a fast fibre phenotype compared to controls which is a phenotype reported in DMD mouse models 50 and patients 51,52 .
Given the interest in sex differences in muscle regeneration and relevance to therapy we examined sexual dimorphism in the PC 15,53-55 . There was a two-fold greater level of CNFs in PC muscle from male compared to female mdx mice. However, these CNF differences may reflect increased muscle regeneration in males and/or decreased levels of degeneration in females. In support of muscle degeneration, it has been reported that female skeletal muscles of dystrophic mice and canines undergo less degeneration than their male counterparts 54,56 .
To gain an enhanced insight into the regenerative activity of PC, we chose to isolate the myogenic precursor cells in PC (satellite cells) from dystrophic mice. PC muscle precursors were examined in vitro when subjected to differentiation as previously undertaken on healthy mice 27,28 . Interestingly, in general we did not detect major differences in the expression of myogenic factors between wild-type and mdx muscle cells, nor between sex for wild-type mice, which is consistent with other reports in limb muscles 14 .
In the present study, we observed no difference in myogenic differentiation between sexes for normal healthy PC muscle progenitor cells in vitro, as reflected in myogenic factor gene expression profile over time from day 1 to day 7 of muscle differentiation. This is consistent with some reports for satellite cells in limb muscles 14 . However, PC cells from male mdx mice expressed differentiation genes earlier than female mdx mice, with elevated expression of myogenic genes Pax7 and myogenin in male compared to female mdx mice. Pax7 reflects a greater number of quiescent myogenic cells with self-renewal ability and elevated myogenin reflects increased myoblast fusion to differentiation. Adult myosin a marker of muscle maturation was more highly expressed in male mdx PC cells compared to female PC cells. These differences in myogenicity detected in cell culture may reflect intrinsic dimorphic differences between mdx male and female PC satellite cells respectively. We inferred myogenic activity from qPCR transcript levels of specific myogenic marker genes and not from protein as this had been previously validated by others for this assay 27,37 . Myogenin scoring was equally greater in both male wild-type and mdx mice compared to female counterparts. We are not sure the reasons for the appearance of a sex difference for myogenin protein staining in wild-type PC muscle and not from the myogenin qPCR assay. It is possible that the battery of qPCR genes is a more accurate assay for myogenicity compared to the relatively low level of detectable myogenin positive myo-nuclei in the differentiated PC cells in culture. We showed that SC densities in PC muscles were not different between muscle sexes nor genotypes. This further suggests that PC derived SCs from male mdx mice are intrinsically different to their female counterparts. We did not detect sex differences in myogenicity for limb muscle progenitor cells from mdx mice (not shown). We conclude that that sexual dimorphism in muscular dystrophic mice maybe another distinct response of the PC muscle compared with other muscles.
Some specialized skeletal muscles including extraocular, laryngeal and masticatory muscles are resistant to degeneration and this has been correlated with higher expression levels of calcium buffering and regulatory proteins 22,23,57 . Conversely, others have provided evidence that calcium handling is not related to muscle sparing of the extraocular muscles in mdx mice 35,58 . However, there is a consensus that calcium dysregulation contributes to the degeneration of muscle fibres in several muscular dystrophies [59][60][61][62] . In the present study, several genes regulating calcium homeostasis were significantly augmented in female versus male PC muscles from mdx mice. However, further examination is required to understand if these changes contribute to less regeneration/degeneration cycling in dystrophic female PC.
We concur that the PC is a fast-twitch muscle demonstrating high turnover in healthy muscle and augmented in a model of muscular dystrophy that further increases with age in mice. PC showed sexual dimorphism in muscular dystrophy with males having a higher muscle regenerative/degenerative activity than females. Sex determined intrinsic differences in the PC satellite cells derived from dystrophic mice included a differential propensity to differentiate and alterations in gene expression of Ca 2+ -handling proteins. PC strongly demonstrated the different phenotypical characteristics of muscular dystrophy and thereby validates this very accessible tissue for studying muscle degenerative disorders. This and other recent studies on PC set a premise to begin to utilize this muscle to expand our knowledge into the heterogeneity of muscle function and regeneration. In particular the use of in vivo imaging (e.g. intra-vital fluorescent microscopy) with fluorescent labelling and/or gene reporters should allow a clearer understanding of muscle regeneration in vivo using the PC as a novel model.

Materials and Methods
Animals. Control (C57BL/10ScSn) and mdx (C57BL/10ScSn-Dmd mdx /J) mice were obtained as breeding pairs from the Jackson Laboratory (Bar Harbor, ME, USA). Animals were housed in the Bio Resource Unit, Bioscience Research Building, NUIG, according to the institutional guidelines. Mice were maintained under controlled conditions (12:12 h light/dark cycle, 20-24 °C) and had access to water and food ad libitum. All experiments were performed and approved in accordance with the guidelines and regulations as defined by the NUIG Galway Animal Care Research Ethics Committee (ACREC) and under the authorization (AE19125/I075) issued by the Health Products Regulatory Authority (HPRA), Ireland.
Tissue preparation. Dorsal skin tissues (1 cm 2 ) and gastrocnemius muscles were harvested from mice after euthanization by CO 2 asphyxiation. The dorsal skin was depilated by shaving and applying a depilatory cream. Tissues were placed on matching size of polyvinylidene difluoride (PVDF) membrane to maintain flat and immersed in 4% paraformaldehyde (PFA) at room temperature (RT) overnight followed by processing (Leica ASP300) and embedding in paraffin. Paraffin embedded tissues were transversely sectioned (5 μm) using a microtome (Leica Biosystems) and placed on glass slides for staining.
Immunofluorescent staining was performed on frozen sections following the preferred protocol for the antibodies used. Dermal tissues (~1 cm 2 ) from the mid-dorsal region were harvested from 8-week-old male and female, wild type and mdx mice. Skin tissues were frozen in liquid-nitrogen-cooled isopentane www.nature.com/scientificreports www.nature.com/scientificreports/ (Merk), embedded in Tissue-Tek O.C.T. compound (Sakura) and stored at −80 °C. Frozen tissues were cryo-sectioned (10 μm) at −26 °C with a Leica CM1850 cryostat (Leica, Germany), placed on glass slides, and stored at −20 °C.

Whole-body pc analysis.
To analyse the distribution of PC muscle throughout the whole-body skin of the mouse, 1 cm 2 sections of skin tissues from different anatomical locations were obtained from 1-year-old male wild-type and mdx mice. Transverse sections representative of each location was histologically stained with Trichrome blue kit (Abcam) following manufacturer's instructions and digital images (10x magnified) captured under a bright-field microscope.
Morphometric analysis. Paraffin sections were stained with Haematoxylin-Eosin (H&E) after processing them through deparaffinization and rehydration. Sections were then dehydrated and mounted with glass cover-slides using DPX mounting media (Sigma-Aldrich). Sections were stained with H&E following a standard approved protocol SOP ID number MDC1A_M.1.2.004, version 1: "Histopathology in Hematoxylin & Eosin stained muscle sections".
Muscle regenerative capacity was assessed by quantifying the percentage of central nucleated muscle fibres (CNFs) on H&E stained wild-type and mdx tissue sections. Sections were imaged under bright-field microscopy using a 10x magnification lens. The proportion of CNFs was measured by analysing up to 1000 fibres per section taken from several non-overlapping images.
Muscle immunofluorescent staining. Frozen cross-sections were equilibrated to RT, fixed in 4% PFA, permeabilized in 0.1% Triton-X100 in phosphate buffered saline (PBS), blocked in 5% goat serum followed by over-night incubation in the appropriate primary antibodies diluted in 1.5% goat serum at 4 °C in a humid environment.
Tissue genotype was confirmed by immunostaining with anti-dystrophin antibody. PC muscle fibre typing was performed by immunostaining serial sections with anti-fast and anti-slow myosin antibodies. To determine the fast isotype, unfixed frozen sections were double-stained with anti-MyHC-IIA and anti-MyHC-IIB to stain type IIA and type IIB myofibers respectively. The primary antibodies were revealed using an appropriate secondary antibody diluted in PBS.
Pax7 and Laminin double immunofluorescent staining was performed as follows. Frozen cross-sections were fixed in 2% paraformaldehyde (PFA), permeabilized in 0.3% Triton/0.1% Glycine, blocked in 3% bovine serum albumin with mouse-on-mouse (MOM) IgG blocking reagent (Vector Labs) followed by over-night incubation in primary antibodies mouse anti-Pax7 and rabbit anti-laminin diluted in 3% BSA/PBS at 4 °C. The primary antibodies were detected using the appropriate secondary antibody and then counterstained with DAPI. Slides were mounted and carefully covered with coverslips and allowed to dry overnight before visualisation under the EVOS M5000 Imaging System (ThermoFisher Scientific). Sections were visualised under GFP, RFP and DAPI channels, with individual and merged images taken for each channel. Non-overlapping microscopic images were taken from two muscle sections for each of four mice in each group. Satellite cells (SC) were determined as those that were co-stained for Pax7 and DAPI, and located along the basal lamina of muscle fibres as delineated with the laminin immunostaining. Analysis was carried out using a manual count method (multi-point tool) on ImageJ software (https://imagej.nih. gov/ij/), with %SCs calculated by number of SCs per total number of muscle fibres x 100.
Muscle fibre sizing. Muscle fibre sizes were analysed on paraffin embedded transverse sections. Paraffin sections were pre-treated as described for immunofluorescence staining method for deparaffinization, dehydration and antigen retrieval. Sections were then treated by 0.1% Triton-X100/PBS for 5 min, washed and incubated with wheat germ agglutinin (WGA) conjugated to Alexa Fluor-488 (W11261; 1-10 µg/mL; ThermoFisher Scientific) for overnight at 4 °C. Images were taken at 20x magnification of 5 non-overlapping areas and analysed using ImageJ software (ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih. gov/ij/). Cross sectional area (CSA) and minimal Feret's diameter (mFD) were measured as indices of muscle fibre size. Frequency distribution graphs were generated for CSA and mFD using bin sizes 100 µm 2 and 5 µm respectively.

Isolation, proliferation and striated muscle differentiation of dermal and gastrocnemius muscle precursor cells.
Precursor cells of the PC muscle were isolated from the entire dorsal skin (superior to inferior) of 8-week-old wild-type and mdx, male and female mice as previously described 26,27 . A day prior to isolation, sterile 12 mm glass coverslips were coated with extracellular matrix (ECM). The ECM was prepared by incubating Cultrex basement membrane extract (2.77 mg/mL; Trevigen) and low molecular weight hyaluronan (2.5 mg/ml; R&D Systems) in pre-cooled phosphate-buffered saline (PBS; pH 7.4). PBS. ECM was applied to coverslips and allowed to polymerize at 37 °C for 24 hours in sterile conditions. Finally, excess PBS was eliminated and coverslips left to dry up for 1 h under laminar flow and UV light, prior to plating cells.
The entire dorsal skin of euthanized mice was shaven, washed in chlorhexidine and wiped with iodine. The skin was incubated for 30 min on ice in Hank's balanced salt solution (HBSS; Gibco ™ ), with 2% Fungizone ™ (Gibco ™ ) and 2% penicillin/streptomycin (P/S; Gibco).
For differentiation, dermospheres after 7 days of proliferation were gently disaggregated with 0.25% trypsin-EDTA solution (Sigma-Aldrich) until a mononucleated cell suspension was formed. Cells were resuspended in the differentiation medium (proliferation medium without growth factors plus 10% fetal bovine serum) before plating onto ECM-coated coverslips at a density of 75,000 cells/cm 2 of each culture per coverslip. Cells were allowed to differentiate for different time points; 1, 3, and 7 days, and the medium was renewed every 2 days. At day 5-7 of differentiation, myoblast fusion and contractile activity were observed. Brightfield microscopic images of cultures as well as video recordings of twitching myotubes were taken by using a Nikon D90 digital camera coupled to a Nikon Eclipse TS100 microscope.
Cell immunofluorescence and myogenesis assay. At the desired time point, cells were washed with PBS, fixed with 4% paraformaldehyde for 10 min at room temperature (RT), and further washed with PBS to wash off the fixative solution. After the cells were permeabilized and blocked in 5% goat serum diluted in 0.3% Triton X-100/PBS (PBST) for 1 hour at RT, the cells were incubated in primary antibodies/PBST for 2 hours at RT. Immunodetection of Pax7 represented the early stage of myogenesis (quiescent/activated satellite cell), myogenin represented the differentiation stage (differentiated myoblast fusion into myotubes) and adult Myosin heavy chain represented the generation of mature myotubes.
Primary antibodies were revealed using appropriate secondary antibodies. Nuclei were highlighted using either Hoechst (10 µg/mL) staining or ProLong ® Gold Antifade Mountant with DAPI (P36935; Molecular Probes ™ ). Fluorescent images were captured using an upright fluorescent microscope (Olympus BX51, Olympus Corporation, Shinjuku, Tokyo, Japan) for tissue sections and Nikon Eclipse 80i microscope (Nikon Instruments Europe B.V., Amsterdam) for cell experiments and analysed by ImageJ software.
To assess myogenin expression, nine different fields from two different coverslips were analysed for myogenin (Myog) immunopositive staining. All myogenin positive nuclei were counted including mono or multi-nucleated myocytes and compared to total cell count derived from nuclei count. Data were analysed by comparing the mean percentages of +MyoG nuclei/total nuclei generated from four different biological replicates for each sample. Gene expression. Total RNA was extracted from cells pooled from three coverslips for each condition at day 1,3 and 7, using the miRNeasy Mini kit (QIAGEN, 217004) and transcribed into cDNA with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems) according to the manufacturer's instructions. Quantitative real-time PCR (qRT-PCR) was conducted as previously described 27 using 25-50 ng cDNA and TaqMan gene expression assays on the 7900HT Fast Real-Time PCR System (ThermoFisher Scientific) for cells or StepOnePlus Real-Time PCR Systems (Applied Biosystems) for tissues. TaqMan assays included Pax7, Myogenin (Myog), embryonic myosin heavy chain (MyHC3) and adult myosin heavy chain (MyHC2). C T values were normalized to TATA-Box Binding Protein housekeeping gene (TBP). Analysis of the relative expression of the studied genes was carried out following the 2 −ΔΔCT method 63 .
Dorsal PC tissues were harvested from the inferior region of the mouse (2 × 1 cm 2 ) and used for the RNA analysis. The subcutaneous layer was mechanically scraped off into an Eppendorf tube on ice. Samples were then lysed in 1 mL Trizol (Sigma) using a tissue homogenizer (TissueRuptor II, QIAGEN). Total RNA was extracted and qRT-PCR was performed as before. TaqMan gene expression assays were run on a StepOnePlus Real-Time PCR System (Applied Biosystems) to detect sarco/endoplasmic reticulum Ca 2+ -ATPase1a (SERCA1a), sarco/ endoplasmic reticulum Ca 2+ -ATPase 2a (SERCA2a), sarcolipin (SLN), calsequestrin1 (CASQ1), calmodulin1 (Calm1), Parvalbumin (Parv) and calpain 3 (Suppl. Table S1). C T values were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) housekeeping gene and 2 −ΔΔCT method was followed for indicating the relative expression of genes of interests.

Statistical analyses.
Statistical analyses were carried out using Excel and GraphPad Prism software. A two-way analysis of variance (ANOVA) was applied where two factors (animal genotype vs. age/sex) were present in an experiment. A three-way ANOVA was performed when 3 factors (animal genotype vs. sex vs. time-point) were present in one experiment (Fig. 6A-D). Subsequently, pairwise multiple comparisons (Tukey's post-hoc test) was used to assess the statistical significance for all data. Data are expressed as mean ± S.E.M., and differences were considered statistically significant at *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001.