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

Learn More →

Novel and optimized strategies for inducing fibrosis in vivo: focus on Duchenne Muscular Dystrophy

Novel and optimized strategies for inducing fibrosis in vivo: focus on Duchenne Muscular Dystrophy Background: Fibrosis, an excessive collagen accumulation, results in scar formation, impairing function of vital organs and tissues. Fibrosis is a hallmark of muscular dystrophies, including the lethal Duchenne muscular dystrophy (DMD), which remains incurable. Substitution of muscle by fibrotic tissue also complicates gene/cell therapies for DMD. Yet, no optimal models to study muscle fibrosis are available. In the widely used mdx mouse model for DMD, extensive fibrosis develops in the diaphragm only at advanced adulthood, and at about two years of age in the ‘easy-to-access’ limb muscles, thus precluding fibrosis research and the testing of novel therapies. Methods: We developed distinct experimental strategies, ranging from chronic exercise to increasing muscle damage on limb muscles of young mdx mice, by myotoxin injection, surgically induced trauma (laceration or denervation) or intramuscular delivery of profibrotic growth factors (such as TGFβ). We also extended these approaches to muscle of normal non-dystrophic mice. Results: These strategies resulted in advanced and enhanced muscle fibrosis in young mdx mice, which persisted over time, and correlated with reduced muscle force, thus mimicking the severe DMD phenotype. Furthermore, increased fibrosis was also obtained by combining these procedures in muscles of normal mice, mirroring aberrant repair after severe trauma. Conclusions: We have developed new and improved experimental strategies to accelerate and enhance muscle fibrosis in vivo. These strategies will allow rapidly assessing fibrosis in the easily accessible limb muscles of young mdx mice, without necessarily having to use old animals. The extension of these fibrogenic regimes to the muscle of non-dystrophic wild-type mice will allow fibrosis assessment in a wide array of pre-existing transgenic mouse lines, which in turn will facilitate understanding the mechanisms of fibrogenesis. These strategies should improve our ability to combat fibrosis-driven dystrophy progression and aberrant regeneration. Background dystrophies and is caused by loss of the dystrophin protein In skeletal muscle, accumulation of collagens (fibrosis) in due to genetic mutations. As a result, the sarcolemma be- the extracellular matrix (ECM) is most often associated comes fragile and susceptible to contraction-induced with the muscular dystrophies, characterized by muscle damage [1]. Skeletal muscle stem cells (satellite cells) me- wasting, leading to loss of patient mobility. Duchenne diate the repair process, but in the absence of dystrophin, muscular dystrophy (DMD) is one of the severest of the the muscle undergoes continuous cycles of degeneration and regeneration, eventually leading to satellite cell deple- * Correspondence: ebrandan@bio.puc.cl; pura.munoz@upf.edu tion and myofiber loss [2-4]. The severity of this childhood- Department of Cell and Molecular Biology, Catholic University of Chile, associated pathology may also be exacerbated by the Avenida Libertador Bernardo O’Higgins, 340, Santiago, Chile Cell Biology Group, Department of Experimental and Health Sciences, CIBER growth of myofibers that occurs in boys with DMD over on Neurodegenerative Diseases (CIBERNED), Pompeu Fabra University (UPF), many years [5]. Affected children eventually succumb to Dr. Aiguader, 88, 08003 Barcelona, Spain 3 muscle wasting, with muscle progressively being replaced Institució Catalana de Recerca i Estudis Avançats (ICREA), Dr. Aiguader, 88, 08003 Barcelona, Spain © 2014 Pessina et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 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. Pessina et al. Skeletal Muscle 2014, 4:7 Page 2 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 by fat and fibrotic tissue, leading to premature death in the dystrophic C57Bl/10scsn-mdx (mdx) male mice were used late teens or early twenties from respiratory and cardiac in experiments: the background strain for mdx mice is failure [6]. There are currently only palliative treatments for similar to, but not identical with, the C57Bl/6 J strain. All DMD patients. Importantly, no effective clinical treatments operations were performed after injection intraperitoneal are available yet to combat or attenuate fibrosis in patients (i.p.) of ketamine/metedomidine anesthesia (50 mg/kg with DMD. Halting or diminishing the development of fi- and 1 mg/kg body weight). Atipamezol (1.0 mg/kg body brosis could not only ameliorate DMD progression, but weight) by subcutaneous injection was used to reverse could also increase the success of new cell- and gene-based the effects of anesthesia. Mice were sacrificed at the in- therapies [7,8]. The mdx mouse strain, the most widely dicated ages and the tissues were immediately processed used animal model for studying human DMD, has a non- to avoid artifacts, either by direct freezing in liquid nitro- sensemutationin dystrophinexon23leading to dystrophin gen for protein and RNA extraction or in 2-methylbutane protein absence [9]. Although mdx mice and DMD patients cooled with liquid nitrogen for histological analysis, as de- share many genetic, biochemical and histological similar- scribed below. ities, the clinical manifestations are generally less severe in mdx mice[10]. WhileDMD individualshaveahighdegree Skeletal muscle fibrogenic treatments of muscle fibrosis, mdx mice present extensive fibrosis ex- clusively in the diaphragm muscle. In the limb muscles of – Chronic exercise: Mdx mice were exercised three mdx mice, however, fibrosis only becomes apparent around times per week on a treadmill for 30 minutes at a 20 months of age [11]. Therefore, despite our awareness of speed of 12 meters per minute, with a rest of the importance of fibrosis in DMD, there is a lack of ap- 5 minutes every 10 minutes of exercise. Mdx mice propriate mouse models for studying dystrophic skeletal of three, four and five months of age were exercised muscle fibrosis in accessible muscles, such as limb mus- for three, two and one month, respectively and were cles, without requiring nearly two years for fibrosis to ap- sacrificed at the age of six months, together with pear. Thus, there is a genuine need to develop mouse age-matched unexercised control mdx mice. At the models that present fibrosis at early stages in life and that end of the training period muscles were collected more closely mimic human DMD. and processed for further analyses. In this manuscript, we present several experimental – Single treatments in WT and mdx mice: strategies to simply and effectively advance and enhance  Myotoxin-induced injuries: muscle fibrosis in young mdx mice. We use both physio- Cardiotoxin injury: Tibialis anterior (TA) muscles logical exercise, as well as more direct tissue-damaging of three-month-old WT or mdx mice were –5 procedures or delivery of profibrotic growth factors to injected with 50 μlof10 M cardiotoxin (CTX; limb muscle of young dystrophic mice, and demonstrate Latoxan, Rosans, France). Muscles were collected sustained collagen deposition reminiscent of aged mdx at the indicated times on each set of experiments, diaphragm muscle, and muscle of human DMD patients. which was usually two weeks after myotoxin Notably, we could extend these strategies to induce fi- injection. Muscle samples were also obtained at brosis in muscles of normal, non-dystrophic mice, which one month post-injection (in WT mice) and two will facilitate studying fibrosis in a wide array of genetic- months post-injection (in mdx mice). ally modified mouse lines, and this will in turn increase Barium chloride injury: TA muscle of three- our understanding of the cells and molecules involved in month-old WT mice was injected with 50 μlof fibrosis development. Thus, we offer for the first time, to 0.2% barium chloride (BaCl )and was isolatedafter the best of our knowledge, a comparative and quantita- two weeks. For repeated BaCl injuries, BaCl injec- 2 2 tive set of new and improved strategies for inducing tions were made in the same muscle, one per week muscle tissue fibrosis, which will greatly foster our abil- for six weeks, and muscles were isolated and col- ity to combat fibrosis-dependent dystrophy progression. lected for analysis two weeks after the last injection. Traumatic injuries: Methods Laceration: TA muscles of three-month-old WT Mice handling and sample collection or mdx mice were subjected to laceration (LAC) All experiments were approved by the Ethics Committee as previously described [11,12]. Briefly, the skin of the Pompeu Fabra University (UPF) and performed was carefully cut and separated from the according to Spanish and European legislation. Mice were underlying tissue, then the TA muscle of one leg housed in standard cages under 12-hour light–dark cycles was cut horizontally at its middle of the length by and fed ad libitum with a standard chow diet. Three- making a lesion through 75% of their width and month-old normal C57Bl/6 J mice (the classic standard la- 50% of the muscle thickness with a scalpel. boratory mouse strain, hereafter referred to as WT) and Contralateral control muscles were sham-operated. Pessina et al. Skeletal Muscle 2014, 4:7 Page 3 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 In WT mice, muscles were collected at two weeks Dystrophic patients study and one month post-surgery. In mdx mice, Human samples were provided from Dr. J. Colomer muscles were obtained at two weeks and two (Hospital Sant Joan de Deu, Barcelona, Spain). DMD diag- months post-surgery. nosis was established on a total absence of dystrophin by Denervation: Muscle denervation (DEN) was immunohistochemistry and Western blotting. Muscle sam- performed as previously described [13,14]. In brief, a ples were obtained by a standard quadriceps muscle biopsy 5 mm segment of the sciatic nerve was surgically from six DMD patients (ranging from five to eleven years removed down to the gluteus maximum from the of age) and five healthy male human controls of similar age right legs. In three-month-old WT mice, TA (seven to fourteen years). Quantification of fibrosis was muscles were isolated for analysis at two weeks and carried out by color image segmentation and automatic one month post-surgery. In three-month-old mdx measurement using Fiji image analysis software [16]. The mice, TA muscles were collected at two weeks and ratio of the total area of fibrosis to the total biopsy area two months post-surgery. Contralateral muscles of was used to estimate the extent of fibrosis (fibrosis index). sham-operated mice were used as controls on every Histological analysis was performed similarly to mouse single treatment (CTX, BaCl ,LAC andDEN). samples, as explained in the next section. Profibrotic growth factor treatments: Transforming growth factor beta treatment:50 ng Histological analysis and immunohistochemistry of transforming growth factor beta 1 (TGFβ1) Cryosections (10 μm thickness) were stained with (recombinant human TGFβ1; R&D Systems, hematoxylin/eosin (H&E) or Sirius red (Sigma-Aldrich, Minneapolis, MN, USA) were injected in the TA St Louis, MO, USA). Quantification of collagen content muscle in a volume of 50 μl of phosphate-buffered in muscle was performed according to Ardite et al. [11]. saline (PBS, vehicle). Two injections (one per week) Briefly, 10 cryosections were collected in a tube and were were made in TA muscle of three-month-old mdx sequentially incubated with a solution containing 0.1% mice, and muscles were collected for analysis two Fast green in saturated picric acid, washed with distillated weeks or two months after the first injection. water, incubated with 0.1% Fast green and 0.1% Sirius red Connective tissue growth factor delivery:TA in saturated picric acid, washed with distillated water, and muscles of three-month-old mdx mice were gently resuspended in a solution of 0.1 M NaOH in abso- injected with 1 × 10 viral particles of Ad-m lute methanol (1 vol:1 vol). Absorbance was measured in a connective tissue growth factor (CTGF) or spectrophotometer at 540 and 605 nm wavelengths and Ad-GFP or with PBS (vehicle) [15] as a control in used to calculate total protein and collagen. a total volume of 50 μl. Muscles were collected Immunohistochemistry on frozen sections was per- two weeks after injury. formed using the following primary antibodies: rabbit poly- – Combined treatments in WT mice: clonal collagen I (Coll 1) (Millipore, Billerica, MA, USA), Cardiotoxin injury combined with denervation: TA rabbit polyclonal fibronectin (FN) (Abcam, Cambridge, muscles of three-month-old WT mice were injected MA, USA) and rabbit polyclonal phosphorylated-Smad2/3 –5 with 50 μlof10 M CTX immediately after DEN, (P-Smad2/3) (Abcam). For immunoperoxidase staining, as indicated above. Muscles were collected at two labeling of sections was performed using the peroxidase weeks and one month after the treatments. staining kit (Vector Laboratories, Burlingame, CA, USA) Contralateral muscles of non-denervated left legs according to the manufacturer’s instructions. For immuno- were used as controls. fluorescence, secondary antibodies were coupled to Alexa Cardiotoxin injury combined with TGFβ/CTGF Fluor 488 or 568 fluorochromes (Invitrogen, Carlsbad, treatment: TA muscles of three-month-old WT mice CA, USA). Stained sections were photographed on a –5 were injected with 50 μlof10 M CTX. TGFβ1 Leica DM6000B microscope (Leica Microsystems, Wetzler, was injected intramuscularly twice at day 7 and 10 Germany). after cardiotoxin injection. Muscles were collected at two weeks and one month after the cardiotoxin RNA isolation, reverse transcription (RT) and real-time injection. Contralateral muscles of sham-operated quantitative PCR legs were used as controls. When indicated, CTGF Total RNA was isolated from muscle tissue using Trizol adenoviral delivery was performed immediately after (Invitrogen). cDNA was synthesized from 1 μg of total cardiotoxin injection. RNA using the First Strand cDNA Synthesis kit and ran- dom priming according to the manufacturer’s instructions Specific information about starting and sampling ages (Promega, Madison, WI, USA). RT-PCR was performed of mice after the different experimental protocols is in- on a LightCycler 480 System using LightCycler 480 SYBR cluded in Table S1 in Additional file 1. Green I Master Mix (Roche, Basel, Switzerland) with Pessina et al. Skeletal Muscle 2014, 4:7 Page 4 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 10 μM each primer and normalized to L7 ribosomal RNA Results as a housekeeping gene: mL7 5′-GAAGCTCATCTATG Mdx mice reproduce the human DMD fibrotic phenotype AGAAGGC–3′ and 5′–AAGACGAAGGAGCTGCAGA in aging diaphragm muscle AC-3′; mCollagen I, 5′-GGTATGCTTGATCTGTATCT To recreate as closely as possible the fibrosis status of GC-3′ and 5′-AGTCCAGTTCTTCATTGCATT-3′;mC human DMD in animal models, we first sought to TGF, 5′-CAGGCTGGAGAAGCAGAGTCGT-3′ and 5′- characterize in detail distinct fibrosis-associated parame- CTGGTGCAGCCAGAAAGCTCAA–3′;mTIMP-1 5′- ters in muscle biopsies of DMD patients. Compared to TTCCAGTAAGGCCTGTAGC-3′ and 5′-TTATGACCA muscles of healthy individuals, we found an increased col- GGTCCGAGTT-3′;mTGFβ 5′-TATGACCAGGTCCGA lagen content in DMD patients, based on Sirius red stain- GTT-3′ and 5′-CTGGTGCAGCCAGAAAGCTCAA-3′; ing and collagen quantification, where fibrotic tissue had hFibronectin: 5′- GGATGACAAGGAAAATAGCCCTG- replaced the myofiber area (Figure 1A and B). Transform- 3′ and 5′-GAACATCGGTCACTTGCATCT-3′; hTIMP- ing growth factor-β (TGFβ) has been shown to be a profi- 15′-CTTCTGCAATTCCGACCTCGT-3′ and 5′-CCCT brotic cytokine in many types of fibrotic tissues and is a AAGGCTTGGAACCCTTT-3′; hTGFβ 5′-CCTAA GGC potent stimulator of matrix production, including colla- CAGATCCTGTCCAAGC-3′ and 5′- GTGGGTTTCCA gen, by fibroblasts [13,18-22]. We found higher levels of CCATTAGCAC-3′;hCTGF 5′- CAAGGGCCTCTTCTG activated TGFβ protein in muscle biopsies from dys- TGACT-3′ and 5′-ACGTGCACTGGTACTTGCAG-3′. trophic children compared to healthy controls (Figure 1C). Consistent with this, we found enhanced levels of active Smad2/3 (as indicated by phosphorylated Smad2/3) Quantification of TGFβ protein (Figure 1D) and TGFβ target genes such as Coll I, FN, The protein concentration of active and total (active tissue inhibitor of metalloproteinases 1 (TIMP-1) and plus latent) TGFβ1 levels in dystrophic muscle was quan- CTGF, indicative of functional TGFβ signaling in fi- tified by ELISA (Promega), following the manufacturer’s brotic DMD muscle (Figure 1E). instructions. The most common experimental model of DMD is the mdx mouse [23]. We examined the TA limb muscle and the diaphragm muscle by hematoxylin and eosin (H&E) Muscle force measurement and Sirius red staining from young (three months of Muscle strength was determined as described previously age), adult (nine months) and old mdx mice (eighteen to [17]. Briefly, after the indicated days of treatment, mice twenty-four months) in comparison to age-matched WT were sacrificed and the TA was rapidly excised into a muscles. Significant fibrosis, similar to that observed in dish containing oxygenated Krebs-Ringer solution. The human patients, was found in TA muscles of mdx mice optimum muscle length (Lo) was determined from mi- only at old age (>18 months) (Figure 2A, upper panels cromanipulations of muscle length to produce the max- and Figure 2D), while adult mdx TA muscles presented imum isometric twitch force. Maximum isometric-specific milder fibrosis. In diaphragm muscle, fibrosis increased tetanic force was determined from the plateau of the age-dependently, reaching near maximum levels in adult curve of the relationship between specific isometric force mice of nine months of age and plateauing thereafter with a stimulation frequency (Hz) ranging from 1 to (Figure 2A, lower panels). Furthermore, in TA muscles of 200 Hz for 450 ms, with 2 minutes of rest between stimuli. mdx mice, collagen content, activated TGFβ and expres- The force was normalized per total muscle fiber cross- sion of ECM-associated molecules started to increase at sectional area (CSA), to calculate the specific net force adult age but were much higher at old age (Figure 2B, D, F, (mN/mm ). G); in the diaphragm muscle, these parameters were mod- erately increased already at young age (Figure 2C and E, Statistical analysis and Figure S1A and B in Additional file 2). The limited de- Comparison between groups was done using the non- velopment of fibrosis (compared to the diaphragm muscle) parametric Mann–Whitney U test for independent sam- in the easily accessible limb muscles of mdx mice until old ples, with a confidence level of 95% being considered age, reinforces the need for developing new protocols that statistically significant. One-way or two-way analysis of will advance muscle fibrosis in young mdx mice. variance (ANOVA) was used for comparisons between multiple groups as appropriate, and post hoc analysis Exercise training triggers fibrosis in muscles of young was performed using Tukey’s test. All statistical analyses dystrophic mice were performed using GraphPad Prism 5.0 (GraphPad In a first attempt to induce and advance muscle fibrosis, Software, San Diego, CA, USA). The number of samples young mdx mice were subjected to a chronic exercise analyzed per group is detailed on each figure. Differences training routine, known to exacerbate the muscle degen- were considered to be statistically significant at P <0.05. eration/regeneration process [24]. Three-month-old mdx Pessina et al. Skeletal Muscle 2014, 4:7 Page 5 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 Human Normal Patient Normal ** Patient Sirius red Human 1800 Normal Patient Normal ** Patient P-Smad2/3 Co ll I FN TIMP-1 E CTGF Normal * Patient * 10 100 3 8 50 5 4 0 0 0 0 0 Figure 1 Quantification of fibrosis in human dystrophic muscle. (A) Representative Sirius red staining of healthy and dystrophic human muscle sections reveals the extent of collagen deposition in patients with Duchenne muscular dystrophy (DMD). (B) Percentage of fibrosis (collagen content) in healthy and DMD muscles as measured by Sirius red staining in muscle sections. Data correspond to the mean ± SEM; n = 6 for DMD group and n = 5 for control group. Non-parametric Mann–Whitney U test was used for comparison. **P <0.01 versus healthy controls. (C) Active transforming growth factor beta 1 (TGFβ1) protein levels measured by ELISA in muscle biopsy material from healthy and DMD muscle. Data correspond to the mean ± SEM; n = 5 on each group. Non-parametric Mann–Whitney U test; **P <0.01 versus healthy controls. (D) Immuno- histochemistry for phosphorylated-Smad2/3 (P-Smad2/3) in healthy and dystrophic human muscle sections. (E) Quantitative RT-PCR for collagen I (Coll 1), fibronectin (FN), tissue inhibitor of metalloproteinases 1(TIMP-1), TGFβ1 and connective tissue growth factor (CTGF) in DMD muscles compared to healthy muscles (which were given the arbitrary value of 1). Data correspond to the mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test *P <0.05. Scale bars = 50 μm. mice were exercised on a treadmill three times per week respect to normally active non-exercised mdx control for up to three months, for 30 minutes each time, at a mice (Figure 3A, and Figure S2A in Additional file 3). The speed of 12 meters per minute, with a rest of 5 minutes increased fibrosis observed by Sirius red staining was con- every 10 minutes [25]. Muscles of mice exercised for one, firmed by Coll I immunofluorescence (Figure 3B, upper two and three months were compared with age- and sex- panels, and Figure S2B in Additional file 3), and the greater matched unexercised mice. After one month, exercised deposition of FN (Figure 3B, and Figure S2B in Additional mdx mice already showed a worsening of the dystrophic file 3, lower panels) that is normally only observed in old phenotype (compared to age-matched controls), and this mdx limb muscles (Figure 2G). Consistent with this, colla- condition was further aggravated by continued adherence gen content and the expression of TGFβ1and CTGF to the exercise regime. Hindlimb muscles (gastrocnemius mRNA, and the levels of P-Smad2/3 proteins, were in- and TA) of one-month exercised mice displayed a higher creased in exercised dystrophic mdx muscles, compared to degree of fibrosis, identified by Sirius red staining, with non-exercised controls (Figure 3C, D, E, and Figure S2C in TGF 1 TGFβ1 active protein Relative expression pg/mg protein % fibrosis Pessina et al. Skeletal Muscle 2014, 4:7 Page 6 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 WT Young Adult Old WT mdx *** ** WT 600 mdx *** *** 200 ** TA Diaphragm D E 100 *** WT WT *** 40 *** mdx mdx ** *** Young Adult Old Young Adult Old Collagen I CTGF TGFβ1 TIMP-1 40 30 6 *** Young mdx *** * *** 8 Adult mdx 20 Old mdx 10 2 ** 0 0 0 0 mdx WT Young Adult Old Figure 2 Quantification of fibrosis in muscle of mdx mice. (A) Sirius red and H&E staining of mdx tibialis anterior (TA) (upper panels) and diaphragm (lower panels) muscles at different ages compared to adult wild-type (WT) muscle. ‘Young’ corresponds to muscles of three-month-old mice, ‘Adult’ to nine months and ‘Old’ to eighteen to twenty-four months of age. (B) and (C) Active transforming growth factor beta 1 (TGFβ1) protein quantifi- cation by ELISA in TA and diaphragm muscles of WT and mdx mice at the indicated ages, respectively. Data correspond to the mean ± SEM values; n = 4 for each group. Two-way analysis of variance with Tukey’s post hoc multiple comparison test. **P <0.01, ***P <0.001 versus age-matched WT. (D) and (E) Quantification of collagen content in TA and diaphragm muscles of WT and mdx mice at different ages. Values are mean ± SEM; n = 4 for each group. Two-way analysis of variance with Tukey’s post hoc multiple comparison test. **P <0.01, ***P <0.001 versus control WT values. (F) Relative expression of collagen I, connective tissue growth factor (CTGF), tissue inhibitor of metalloproteinases 1(TIMP-1) and TGFβ1mRNAbyquantitativeRT-PCRinmdx TA muscles at the indicated ages with respect to WT muscles (baseline set arbitrarily to 1). Values are mean ± SEM; n = 3 for each group. Two-way analysis of variance with Tukey’s post hoc multiple comparison test. *P <0.05, **P <0.01, ***P <0.001, versus age-matched WT. (G) Representative pictures of immuno- fluorescence staining for collagen I (green) and fibronectin (red) in young, adult and old mdx TA, compared to WT muscle. Scale bars = 50 μm. Young Adult Old Young Adult Old Diaphragm TA Relative expression Fibronectin Collagen I μg collagen/mg protein (versus WT) μg collagen/mg protein TGFβ1 active protein TGFβ1 active protein pg/mg protein pg/mg protein Pessina et al. Skeletal Muscle 2014, 4:7 Page 7 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 Months of exercise 0 1 2 3 CD CTGF TGF 1 4 40 ** *** *** ** ** *** ** 1 10 0 0 0 01 2 3 01 2 3 01 2 3 Months of exercise F Tetanic force Unexercised Exercised ** 03 Months of exercise P-Smad2/3 Figure 3 (See legend on next page.) Relative expression Fibronectin Collagen I H&E Sirius red mN/mm μg collagen/mg protein Pessina et al. Skeletal Muscle 2014, 4:7 Page 8 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 (See figure on previous page.) Figure 3 Effect of exercise on muscle fibrosis in mdx mice. (A) Sirius Red and H&E staining of gastrocnemius muscle of mdx mice that were exercised three times weekly, for 30 minutes at a speed of 12 meters per minute with a rest of 5 minutes each 10 minutes of exercise, for one, two and three months, compared to sections of muscle from unexercised mdx mice. All the samples were collected when the animals were six months old (see the Methods section). (B) Representative immunofluorescence for collagen I and fibronectin in muscle sections of control or exercised mice as shown in (A). (C) Quantitative RT-PCR of connective tissue growth factor (CTGF) and transforming growth factor beta 1 (TGFβ1) mRNA levels after exercising for the indicated period as compared to unexercised age-matched mdx mice. Data correspond to the mean ± SEM; n = 4 sedentary and 4 exercised mdx mice for each exercise time point. One-way analysis of variance with Tukey’s post hoc multiple comparison test; *P <0.05, **P <0.01, ***P <0.001 versus control. (D) Biochemical quantification of collagen protein content in mdx gastrocnemius muscle after exercising for the indicated periods, as compared to unexercised age-matched mdx mice. Data correspond to the mean ± SEM; n = 4 sedentary and 4 exercised mdx mice for each exercise time point. One-way analysis of variance with Tukey’s post hoc multiple comparison test; *P <0.05, **P <0.01, ***P <0.001 versus control. (E) Immunofluorescence for phosphorylated-Smad2/3 proteins on sections from gastrocnemius muscle of six-month-old mdx mice after three months of exercise, as evidence for TGFβ activation, compared to unexercised age-matched control mdx mice. (F) Ex vivo maximum isometric force (tetanic force) of gastrocnemius muscle of age-matched unexercised and three-month-trained mdx mice. Values as mean ± SEM; n = 7 on each group. Non-parametric Mann–Whitney U test; **P <0.01 versus non-exercised. Scale bars = 50 μm. Additional file 3). Furthermore maximum force of the biochemical parameters (see below, Figure 4F and G). This muscles of mdx mice subjected to the exercise regime was extended fibrotic status reinforces the utility of these two decreased with respect to non-exercised mice (Figure 3F, methods as drivers of limb muscle fibrosis in young mdx and Figure S2D in Additional file 3). These data confirm mice, after which the tissue more closely resembles the that exercise in young mdx mice can activate fibrogenesis, more severe phenotype of old mdx mice, as well as human and in particular the profibrotic TGFβ pathway, and DMD patients. Furthermore, these procedures have the thereby enhance and anticipate muscle tissue fibrosis. advantage of not requiring exercise devices, nor the time and labor of the three-month exercise protocol. Surgical muscle injuries advance and enhance fibrosis in young dystrophic mice We next sought alternative and faster ways than long- Raising TGFβ levels in dystrophic muscle of young mdx term exercise training to induce fibrosis in limb muscles mice accelerates fibrosis and accentuates disease severity of young mdx mice, based on inflicting increased surgical Despite the profibrotic effect of the surgical methods on or chemical damage. Since CTX-induced muscle injury is a mdx muscle, each one has particularities and limitations. widely used and well-characterized experimental model for In the LAC model, the injury is confined to a small area inducing skeletal muscle degeneration/regeneration [26-28], of the muscle and this reduces the amount of tissue we hypothesized that superimposing CTX-induced damage available for further studies, whereas, for reasons of ani- on young dystrophic mdx muscle would promote fibrosis. mal welfare, DEN can only be performed in one leg of Despite an early increase in collagen content, two weeks the mouse, affecting only the muscles under the knee. –5 after intramuscular CTX injection (50 μlof10 M), TA Therefore, based on our observation of the elevated levels mdx muscle showed a similar quantity of deposited colla- of TGFβ in human and mouse dystrophic muscle (Figures 1 gen compared to non-injured (NI) mdx TA indicating that and 2), and its correlation with the extent of dystrophy- thefibrogeniceffect of CTX-induced damage was transient associated fibrosis, we reasoned that exogenous delivery of (Figure 4A). To increase and prolong collagen deposition, TGFβ1 to muscle of young dystrophic mice might increase we superimposed on young mdx limb muscle two more and accelerate the development of fibrosis. We therefore extreme, but distinct, experimental paradigms: lacer- performed two intramuscular TA injections of TGFβ1 ation (LAC) and denervation (DEN). The DEN model (50 ng of TGFβ1in50 μl of PBS per injection), spaced involves severing the sciatic nerve thus causing atrophy sevendaysapart, inanattempt to sustainthe profibrogenic of the denervated myofibers [14,29], while the LAC action of this growth factor. Contralateral control muscles model consists in a deep cut across the muscle, which received the same number of injections of PBS. Analysis of causes a delay in the healing process [11,12]. Muscle of the muscles histologically by H&E and Sirius red staining dystrophic mdx mice at two weeks after DEN showed showed that TGFβ1 delivery lead to substantial increase in an increased deposition of collagen relative to CTX- collagen deposition already at two weeks after the first in- injured mdx muscle (Figure 4A, C). Lacerated dystrophic jection, which persisted for up to two months and this was muscle also showed increased fibrosis after two weeks, also confirmed by biochemical quantification of muscle which was even higher than in denervated muscle after extracts (Figure 4B and C). Of note, local muscle overex- the same time period (Figure 4A, C). Importantly, the pression of the TGFβ1 target gene product CTGF also in- mdx muscle fibrosis induced by both methods persisted creased fibrogenesis in limb muscle of young mdx mice for up to two months, as indicated by histological and (Figure S3A, B in Additional file 4). Pessina et al. Skeletal Muscle 2014, 4:7 Page 9 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 N.I. CTX DEN LAC TGFβ C D Collagen I TGF 1 CTGF TIMP-1 N.I. 50 40 5 3 10 CTX * * ** 40 4 30 DEN * * LAC 30 3 6 * TGF 10 * 20 2 4 10 1 0 0 0 0 N.I CTX DEN LAC TGFβ N.I. CTX DEN LAC TGFβ Tetanic force N.I. H N.I. CTX CTX * * DEN DEN LAC LAC TGF TGF ** Figure 4 (See legend on next page.) μg collagen/mg protein μg collagen/mg protein H&E Sirius red Fibronectin Collagen I H&E Sirius red Relative expression mN/mm Pessina et al. Skeletal Muscle 2014, 4:7 Page 10 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 (See figure on previous page.) Figure 4 Induction of fibrosis after chemical and surgical muscle damage in young mdx mice. (A) Sirius red and hematoxylin and eosin –5 (H&E) staining of tibialis anterior (TA) muscles of young (three-month-old) mdx mice two weeks after cardiotoxin (CTX)-injury (50 μlof10 M), denervation (DEN) and laceration (LAC), compared to non-injured (NI) muscle of sham-operated mdx mice. (B) Sirius red and H&E staining of young mdx TA muscle analyzed after two sequential treatments with recombinant transforming growth factor beta 1 (TGFβ1) (50 ng in 50 μl phosphate-buffered saline (PBS)) spaced seven days apart. (C) Biochemical quantification of collagen protein content in mdx TA, two weeks after different treatments relative to uninjured mdx control. Values represent mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (D) Quantitative RT-PCR of collagen I, connective tissue growth factor (CTGF), tissue inhibitor of metalloproteinases 1(TIMP-1) and TGFβ1 mRNA expression in mdx muscle two weeks after different injuries versus control mdx mice. Values represent mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (E) Representative immunostaining for collagen I (green) and fibronectin (red) on sections of young mdx TA muscles two weeks after injury relative to control. (F) Sirius Red and H&E staining of mdx TA muscle two months after CTX injury, DEN, LAC or injection of TGFβ in three-month-old mdx muscle compared to NI mdx control muscle. (G) Quantification of collagen content in TA muscle of young mdx mice two months after different treatments, as described above. Values represent mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (H) Ex vivo maximum isometric force (tetanic force) of TA muscle of young mdx mice two months after treatments. Values as mean ± SEM; n = 4 to 5 on each group. Non-parametric Mann–Whitney U test; *P <0.05; **P <0.01 versus NI. Scale bars = 50 μm. Overall, comparing the distinct biochemical and func- repair. Thus, we designed easy-to-perform profibrotic tional parameters in all the procedures tested revealed procedures in non-dystrophic WT muscle, which could that LAC and TGFβ treatments gave statistically higher ideally be extended to a wide variety of transgenic mouse quantitative measures of collagen than NI age-matched lines for research or therapeutic purposes. control mdx muscles. The collagen values for LAC and We applied the surgical/chemical methods previously TGFβ1 treatments were comparable to the values re- used on muscle of dystrophic mdx mice (see above), ei- corded in limb muscles of old mdx mice (see Figure 2D), ther alone or in combination, to induce muscle fibrosis indicating that either one of these methods advances fibro- in WT mice. First, we performed CTX injury in TA sis by the equivalent of about fourteen months (that is in- muscle of WT mice and assessed fibrosis development. ducing fibrosis at four months of age instead of eighteen We observed a mild and transient deposition of ECM –5 months). DEN also significantly increases muscle collagen between days 5 and 7 after CTX (50 μlof10 M) content over mdx controls, but to a lesser extent than LAC muscle injury (Figure S4A and B in Additional file 5); or TGFβ1 treatment (Figure 4A and C). Interestingly, the however, it did not persist beyond this stage. Indeed, two levels of endogenous TGFβ1 mRNA were increased in weeks after CTX injury, collagen content returned to young dystrophic muscle in response to LAC, DEN and ex- near basal levels, in agreement with efficient muscle re- ogenous TGFβ1 delivery, but not CTX. Consistent with covery (Figure 5A-C and Figure 6H). this, the expression of TGFβ-dependent signaling fibrotic We next subjected WT muscle to the more severe target genes, such as, Coll I, CTGF, TIMP-1, were increased LAC and DEN procedures and compared the fibrosis in mdx limb muscle after all three treatments, but not in index of the affected muscles to that of CTX-injured CTX-damaged muscle (Figure 4D). Finally, immunostain- muscle at similar time points. LAC in TA muscle of WT ing for FN and Coll I on sections from the different dam- mice disrupted the tissue quite extensively and for a pro- aged mdx muscles showed greater ECM production than longed period of time (over one month) resulting in sus- uninjured (or CTX-injured) dystrophic muscle (Figure 4E). tained fibrosis, which correlated with the slow kinetics Remarkably, at two months after injury, collagen depos- for regeneration (Figure 5A, B). DEN, in turn, did not ition still persisted in TGFβ-treated young dystrophic mus- alter ECM production significantly, as revealed by H&E cles as it did in lacerated and denervated muscles, as and Sirius red staining, or immunostaining for Coll I revealed by histological and biochemical analysis (Figure 4F and FN, despite inducing the expected myofiber atrophy. and G). In agreement with this, and demonstrating the Consistent with these findings, we only observed statisti- deleterious physiological consequences of the increased fi- cally significant increases in the expression of TGFβ1 brosis in young mdx muscles, the maximum force of the and the fibrotic markers Coll I, FN, CTGF and TIMP-1 in muscles subjected to the distinct treatments decreased with lacerated muscle, but not in denervated or CTX-injured respect to NI mdx muscles (Figure 4H), therefore better muscles, compared to uninjured muscle (Figure 5C). mimicking the severe phenotype of the human condition. These results suggest that LAC is the most fibrotic of the traumatic models tested in non-dystrophic mice. Induction of fibrosis in non-dystrophic, wild-type muscle As stated above, one of the limitations of the LAC pro- by combining surgical injury and growth factor delivery cedure is the restricted availability of biopsy material. Try- Fibrosis persistence has negative consequences on tissue ing to induce fibrosis by methods that would render more wound healing. Severe muscle injuries caused by trauma fibrotic tissue available for analysis, we decided to combine often result in scar formation at the expense of tissue CTX injury, which individually was a poor fibrosis-inducing Pessina et al. Skeletal Muscle 2014, 4:7 Page 11 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 A N.I. CTX LAC DEN N.I. CTX LAC DEN C Collagen I CTGF TIMP-1 TGF 1 4 4 12 * 4 N.I. * CTX LAC DEN 2 2 6 0 0 Figure 5 Quantification of muscle fibrosis after chemical and surgical damage in wild-type mice. (A) Sirius red, hematoxylin and eosin (H&E), collagen I (green) and fibronectin (red) staining on wild-type (WT) tibialis anterior (TA) muscles two weeks after cardiotoxin (CTX) injury –5 (50 μlof10 M), laceration (LAC) and denervation (DEN) compared to non-injured (NI) muscle of sham-operated WT mice. (B) Quantification of collagen content in WT muscle after different injuries. Data correspond to the mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (C) Quantitative RT-PCR for collagen I, connective tissue growth factor (CTGF), tissue inhibitor of metalloproteinases 1 (TIMP-1) and transforming growth factor beta 1 (TGFβ1) mRNA in muscles after the different injuries (values are means ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI). Scale bar = 50 μm. method, with either DEN or co-injection of TGFβ1in TA muscles of WT mice were first subjected to CTX injec- –5 muscle of WT mice, methods which we previously showed tion (50 μlof10 M) and subsequently denervated or were able to increase fibrosis in young mdx muscle injected twice with TGFβ1(50 ng of TGFβ1in 50 μlPBS (Figure 4). Both DEN and injection of TGFβ1failedto per injection) (at day 7 and 10 after CTX injection) and induce fibrosis in WT muscles when used alone (Figure 5, muscles were collected two and four weeks later. We found and Figure S4C and D in Additional file 5). Accordingly, that the combination of CTX injury with DEN or TGFβ1 Fibronectin Collagen I H&E Sirius red Relative expression μg collagen/mg protein Pessina et al. Skeletal Muscle 2014, 4:7 Page 12 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 A CTX CTX+DEN CTX+TGFβ CTX CTX+DEN * CTX+TGF Collagen I C CTGF TIMP-1 TGF 1 CTX 5 3 20 5 CTX+DEN * CTX+TGF * 15 3 3 10 * 0 0 0 0 CTX CTX+ DEN CTX+TGFβ E N.I. CTX LAC CTX + DEN CTX + TGFβ F N.I. N.I. Tetanic force * CTX CTX LAC * LAC CTX+DEN * 200 * CTX+DEN CTX+TGF CTX+TGF Figure 6 Synergistic effect on fibrosis induction in muscle of wild-type mice by combined treatments. (A) Sirius red and hematoxylin and –5 eosin (H&E) staining of wild-type (WT) tibialis anterior (TA) muscles subjected to a combination of cardiotoxin (CTX) injury (50 μlof 10 M) and denerv- ation or transforming growth factor beta 1 (TGFβ1) (50 ng in 50 μl phosphate-buffered saline (PBS)) injection (as described in the Methods section), respectively, compared to CTX injury alone. (B) Quantification of collagen content in muscle after each treatment. Data correspond to the mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus CTX injury. (C) Quantitative RT-PCR for collagen I, connective tissue growth factor (CTGF), tissue inhibitor of metalloproteinases 1 (TIMP-1) and TGFβ1 after the different treatments (n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05, compared to CTX injury). (D) Representative immunostaining for collagen I (green) and fibronectin (red) on sections of WT muscle subjected to the different fibrosis-inducing methods. (E-G) Analysis of long-term fibrosis in WT muscle at one month after injury. Data are compared to non-injured (NI) muscle of sham-operated WT mice. (E) Sirius red and H&E staining of CTX-injured, lacerated, CTX/denervation and CTX/TGFβ-injured muscles at one month after injury. (F) Quantification of collagen content one month after injury of WT muscle. Values represent mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (G) Ex vivo maximum isometric force (tetanic force) of TA muscle. Values as mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. Scale bars = 50 μm. H&E Sirius red H&E Sirius red Fibronectin Collagen I Relative expression μg collagen/mg protein mN/mm μg collagen/mg protein Pessina et al. Skeletal Muscle 2014, 4:7 Page 13 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 delivery induced fibrosis significantly compared to CTX in- muscular degeneration, of which DMD is one of the se- jury alone, as shown by Sirius red staining (Figure 6A), col- verest. Progressive replacement of skeletal muscle by fat lagen quantification as well as expression of fibrotic and fibrotic tissue not only exacerbates disease progres- markers by quantitative RT-PCR and immunohistochemis- sion, but also impairs the efficiency of gene- and stem cell- try analyses, after 14 days (Figure 6B-D), correlating with based therapies [30,31]. Yet, there is no effective clinical delayed regeneration kinetics (Figure 6A). Of note, a com- treatment to reverse or attenuate fibrosis in DMD patients, bination of CTX injury and CTGF local overexpression except for promising new agents such as halofuginone produced similar profibrotic effects as combining CTX in- [32]. To a great extent, this deficit may derive from the jury and TGFβ1 delivery (Figure S3C in Additional file 4), poor understanding of the mechanisms underlying fibro- suggesting that part of the TGFβ profibrotic actions are genesis in muscular dystrophy. Indeed, chronic inflamma- likely to be mediated by CTGF. tion and production of collagens by myoblasts are among We next compared the persistence of fibrosis over the few reported causal factors promoting progression to time and the consequences on muscle function of each fibrosis in dystrophic muscle [33-38]. The largely unknown of the distinct fibrogenic regimes on WT muscle. Sirius etiology of fibrogenesis in DMD in turn may be principally red staining and collagen quantification showed that due to the lack of adequate animal models of muscle fibro- muscle fibrosis still persisted after four weeks of either sis. Here we report the application of simple models of laceration or CTX combined with TGFβ1 or DEN, com- tissue damage that are able to significantly enhance the pared to muscle injured with CTX alone or NI muscle fibrotic response in skeletal muscle and which may be (Figure 6E and 6F). The relevance of these results was useful for investigating therapeutic strategies for DMD. supported by functional studies of WT muscle after the Studies using mdx mice, the most common mouse combined profibrotic treatments (CTX combined with model of DMD, may not be translated directly to dys- TGFβ1 or DEN). Indeed, dual treatments on muscles trophic patients due to the mild phenotype they display. exerted a synergistic effect, resulting in increased fibrosis In particular, limb muscles of mdx mice show a relatively and reduced net force compared to uninjured muscle or efficient regeneration and no significantly aberrant de- muscle injured with CTX alone (Figure 6G). These re- position of ECM proteins until very old age. Progressive sults suggest that in WT mice, LAC, as well as a com- endomysial fibrosis only develops in diaphragm muscle, bination of CTX injury with either DEN or TGFβ1, but is still not significantly advanced until well into proved to be effective fibrosis-inducing models that trig- adulthood [39]. To try to accelerate or exacerbate this ger a rapid accumulation of fibrotic tissue that is sus- phenotype, other mouse models have been generated such tained for an extended period of time, with negative as mdx mice lacking arginase-2, PAI-1 (plasminogen acti- consequences on muscle function. vator inhibitor-1) or Cmah (cytidine monophosphate- Finally, and in order to further expand the variety of sialic acid hydroxylase) [11,40,41] and previously the mdx/ +/− fibrogenic-inducing procedures to the maximum number utrn mouse line (mdx mice with haploinsufficiency of +/− of laboratories working on skeletal muscle, we tested the utrophin) [42]. However, mdx/utrn mice show early fibrosis-inducing effect of a widely used muscle-damaging mortality and the manipulation of the line requires time method involving BaCl injection in WT muscle. We and resources in genotyping and breeding. Moreover, the found that, as for CTX injection, one intramuscular injec- genetics of these mouse models do not adequately reflect tion of BaCl (50 μlof0.2%BaCl ) only induced a very mild human DMD patients. Therefore, the need for fibrotic 2 2 and transient accumulation of ECM. Of note, repeated in- models that do not require waiting for the natural physio- jections (spaced one week) for up to six weeks resulted in logical onset of fibrosis in the hindlimb of old mice, and significant ECM accumulation after eight weeks from the that recapitulate the human DMD phenotype becomes in- first injection (that is two weeks after the last injection), creasingly more important. One recent attempt to address although no major change in muscle force was observed this problem came from Desguerre and colleagues (2012) (Figure 7A-C). Thus, repeated damaging with myotoxins who described a model of mechanical muscle injury by may be a fibrosis-inducing alternative in non-dystrophic daily repeated micro-punctures in mdx hindlimb muscle muscle, although development of fibrosis requires up to [43]. Induction of endomysial fibrosis in dystrophic muscle eight weeks, and involves weekly mouse manipulation for through this method is ascribed to a small fibrotic area six weeks, compared to the less labor-consuming and and requires daily animal manipulation during two weeks. more rapid fibrogenic effect (with additional impact on In addition, this procedure does not seem to induce fibro- muscle force) of the combined treatments. sis in WT mice [43]. The strategies we propose here are valid alternatives to Discussion both hasten the appearance and prolong the duration of Muscular dystrophies constitute a heterogeneous group fibrosis in hindlimb muscles of young mdx mice, with of inherited myopathies, characterized by progressive very limited (non-daily) animal manipulation, which Pessina et al. Skeletal Muscle 2014, 4:7 Page 14 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 A N.I. BaCl 1 round BaCl 6 rounds 2 2 Collagen I Fibronectin N.I. 3 2 BaCl 1 round BaCl 6 rounds 0 0 Tetanic force N.I. BaCl 1 round 300 BaCl 6 rounds Figure 7 Fibrosis induction in muscle of wild-type mice after repeated BaCl injuries. (A) Sirius red, H&E and fibronectin staining on wild-type (WT) tibialis anterior (TA) muscles subjected to one or six consecutive weekly rounds of BaCl injections (50 μlof 0.2% BaCl ), compared to non-injured 2 2 (NI) muscle of sham-operated WT mice, sampled two weeks after the final injection. (B) Quantitative expression of fibronectin and collagen I in the distinct muscle samples (values are means ± SEM; n = 4 to 5 on each group. Non-parametric Mann–Whitney U test; **P <0.05 versus NI). (C) Ex vivo maximum isometric force (tetanic force) of TA muscle. Values as mean ± SEM; n = 4 for each group; non-parametric Mann–Whitney U test; no significant differences P >0.05. Scale bar = 50 μm. Fibronectin H&E Sirius red Relative expression mN/mm Relative expression Pessina et al. Skeletal Muscle 2014, 4:7 Page 15 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 notably are also able to induce relatively sustained fibro- is relatively stable over long periods of time, despite af- sis in WT muscle. Therefore, these methods would be fecting only a localized small tissue area. However, the applicable to other genetically modified mice, and this combination of regimes showed an improved capacity to will help further delineating the cellular and genetic generate fibrosis in WT muscle for a sustained period of basis of muscle fibrosis. Exercise training of young mdx time, correlating with reduction in muscle force, indicat- mice induced endomysial fibrosis, resembling the pheno- ing that they mimic in WT animals pathophysiological type of old hindlimb dystrophic muscles; however, situations of severe muscle trauma that result in aber- although this method can be considered more physio- rant regeneration, scar deposition and functional impair- logical, it still requires a lengthy time period to obtain a ment. We propose that this variety of fibrosis-inducing fibrotic muscle tissue, in addition to significant amount of methodologies will enable fibrosis to be studied in a vast effort and time, since exercise protocols need to be applied array of transgenic mouse lines (with no apparent under- several times a week for ideally three months. At variance, lying muscle pathology) or after crossing them with dys- the methods based on muscle growth factor delivery and trophic strains such as mdx mice. surgical injuries that we present here offer a faster and less labor-intensive alternative. The rationale for the proposed Conclusions profibrotic growth factor-based methods relies on the ob- Collectively, through this study, we propose novel and/or servation that, in fibrotic muscles of human DMD patients optimized experimental strategies to accelerate, anticipate and old mdx mice, TGFβ1 (and its downstream target and boost muscle fibrosis in young dystrophic mice or to CTGF) is present at high levels [22,44], correlating with drive de novo fibrosis onset in WT mice. We think that the increased activation of Smad2/3 transcriptional me- our findings provide very useful methodologies that will fa- diators (see Figure S5 in Additional file 6). Of the surgi- cilitate research in the emerging field of skeletal muscle fi- cal methods tested, muscle laceration proved to be the brosis. In particular, these rapid and feasible procedures for most effective for inducing sustained fibrosis; however, most laboratories will help getting deeper insight into the this method has the disadvantage that the affected area mechanisms underlying muscle fibrosis, as well as develop- is relatively small (and only one muscle per mouse can ing therapeutic strategies aimed to reduce its magnitude in be lesioned due to the severity of the procedure) thereby dystrophic diseases and to ameliorate dystrophy progres- limiting the amount of material available for downstream sion. Since fibrosis is also a main obstacle for stem cell en- processing. Subsequent cellular analysis of fibrotic muscle graftment, availability of appropriate fibrosis models will be by techniques such as fluorescence-activated cell sorting a determinant factor in the research toward successful (FACS) may not be possible in this type of model without gene/cell therapy-based strategies in muscular dystrophy. vast improvements of sensitivity or without increasing the number of animals used, which has extra cost and ethical Additional files implications. Sciatic nerve denervation of mdx mice gen- erates increased collagen deposition, as a possible mech- Additional file 1: Table S1. Methods and sampling times of the anism to replace the tissue volume lost due to myofiber different fibrosis-inducing procedures in mdx and wild-type (WT) mice. atrophy. All of these fibrogenesis-inducing methods per- Additional file 2: Figure S1. Quantification of fibrosis in mdx diaphragm muscle. (A) Relative expression of collagen I, connective tissue sist with time, since at two months after injury muscles growth factor (CTGF), tissue inhibitor of metalloproteinases 1(TIMP-1) and still displays a fibrotic phenotype. Moreover, consistent transforming growth factor beta 1 (TGFβ1) mRNA by quantitative RT-PCR with the idea that fibrosis aggravates muscle dysfunction in mdx diaphragm muscles at the indicated ages respect to wild-type (WT) muscles. Values are mean ± SEM; n = 4 for each group; non- in DMD, we showed that maximal muscle force was also parametric Mann–Whitney U test; *P <0.05 versus age-matched WT. (B) reduced in young mdx mice after fibrosis induction Representative pictures of immunofluorescence staining for collagen I through the different protocols. (green) and fibronectin (red) in young, adult and old mdx diaphragm, compared to age-matched WT muscle. Scale bars = 50 μm. Finally, to be able to investigate fibrosis development Additional file 3: Figure S2. Effect of exercise on tibialis anterior (TA) and therapeutic options in muscle of non-dystrophic mdx muscle. (A) Sirius red and hematoxylin and eosin (H&E) staining of models, we sought to apply these methods to WT mice. TA muscle of mdx mice that were exercised three times weekly, for To date, studies on muscle damage in non-dystrophic 30 minutes at a speed of 12 meters per minute with a rest of 5 minutes each 10 minutes of exercise, for one, two and three months, compared models have been performed classically with a single in- to sections of muscle from unexercised mdx mice. All the samples were jection of myotoxins (for example, CTX or BaCl ). Des- collected when the animals were six months old (see the Methods pite the general use, we have shown in this study that section). (B) Representative immunofluorescence for collagen I and fibronectin in muscle sections of control and exercised mice as shown in these standard single-injury methods are not appropriate (A). (C) Biochemical quantification of collagen protein content in mdx TA fibrosis-inducing models, as the resolution of the dam- muscle after exercising for the indicated period as compared to unexercised age occurs rapidly and collagen deposition is very mild age-matched mdx mice. Data correspond to the mean ± SEM; n = 4 sedentary and 4 exercised mdx mice for each exercise time point. One-way analysis of and only transient. On the contrary, muscle laceration of variance with Tukey’s post hoc multiple comparison test; **P <0.01, ***P <0.001 WT muscle induces a massive collagen deposition that Pessina et al. Skeletal Muscle 2014, 4:7 Page 16 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 3. Serrano AL, Munoz-Canoves P: Regulation and dysregulation of fibrosis in versus control. (D) Ex vivo maximum isometric force (tetanic force) of TA skeletal muscle. Exp Cell Res 2010, 316:3050–3058. muscle of age-matched unexercised and three-month-trained mdx mice. 4. Yablonka-Reuveni Z, Anderson JE: Satellite cells from dystrophic (mdx) Values as mean ± SEM; n = 7 on each group. Non-parametric Mann–Whitney mice display accelerated differentiation in primary cultures and in U test; **P <0.01 versus non-exercised. Scale bars = 50 μm. isolated myofibers. Dev Dynamics 2006, 235:203–212. Additional file 4: Figure S3. Fibrosis induction in muscle by viral 5. Grounds MD, Shavlakadze T: Growing muscle has different sarcolemmal delivery of connective tissue growth factor (CTGF). (A) Mdx mice: Sirius properties from adult muscle: a proposal with scientific and clinical red and hematoxylin and eosin (H&E) staining of mdx tibialis anterior (TA) implications: reasons to reassess skeletal muscle molecular dynamics, muscles overexpressing mouse CTGF after intramuscular injection of cellular responses and suitability of experimental models of muscle 50 μl of 2x10 particles of adenovirus (AdV) in three-month-old mice. disorders. Bio Essays 2011, 33:458–468. (B) Collagen content quantification. Data correspond to the mean ± SEM; 6. Muntoni F: Cardiac complications of childhood myopathies. J Child Neurol n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 2003, 18:191–202. versus NI. (C) Wild-type (WT) mice: H&E of WT muscles after adenoviral 7. Benedetti S, Hoshiya H, Tedesco FS: Repair or replace? Exploiting novel CTGF delivery coupled with cardiotoxin (CTX) injury; representative gene and cell therapy strategies for muscular dystrophies. FEBS J 2013, immunostaining for collagen I (green) and fibronectin (red) on sections 280:4263–4280. of AdV-transduced muscle overexpressing CTGF. Scale bars = 50 μm. 8. Tedesco FS, Hoshiya H, D’Antona G, Gerli MF, Messina G, Antonini S, Additional file 5: Figure S4. Collagen deposition after cardiotoxin Tonlorenzi R, Benedetti S, Berghella L, Torrente Y, Kazuki Y, Bottinelli R, (CTX)-induced muscle injury and transforming growth factor beta 1 Oshimura M, Cossu G: Stem cell-mediated transfer of a human artificial (TGFβ1) delivery alone is quickly resolved in wild-type (WT) mice. (A) chromosome ameliorates muscular dystrophy. Sci Trans Med 2011, Sirius red, hematoxylin and eosin (H&E), collagen I (green) and fibronectin 3:96ra78. (red) staining on WT tibialis anterior (TA) muscles after five and eight days 9. Sicinski P, Geng Y, Ryder-Cook AS, Barnard EA, Darlison MG, Barnard PJ: The from CTX injury, compared to non-injured (NI) muscle of sham-operated molecular basis of muscular dystrophy in the mdx mouse: a point WT mice. (B) Quantification of collagen content in muscle after treatment. mutation. Science 1989, 244:1578–1580. Data correspond to the mean ± SEM, n = 4 on each group. Non-parametric 10. Carnwath JW, Shotton DM: Muscular dystrophy in the mdx mouse: Mann–Whitney U test; *P <0.05 versus NI. (C) Sirius red, H&E, collagen I (green) histopathology of the soleus and extensor digitorum longus muscles. and fibronectin (red) staining on WT TA muscle two weeks after two sequen- J Neuro Sci 1987, 80:39–54. tial treatments with recombinant TGFβ1 (50 ng in 50 μl phosphate-buffered 11. Ardite E, Perdiguero E, Vidal B, Gutarra S, Serrano AL, Munoz-Canoves P: PAI- saline (PBS)), spaced seven days apart. (D) Quantification of collagen content 1-regulated miR-21 defines a novel age-associated fibrogenic pathway in muscle after injection of TGFβ1 or PBS (vehicle). Data are mean ± SEM, n = 4 in muscular dystrophy. J Cell Biol 2012, 196:163–175. for each group. Non-parametric Mann–Whitney U test; no significant differ- 12. Menetrey J, Kasemkijwattana C, Fu FH, Moreland MS, Huard J: Suturing ences P >0.05. Scale bars = 50 μm. versus immobilization of a muscle laceration. A morphological and functional study in a mouse model. Am J Sports Med 1999, Additional file 6: Figure S5. Smad2/3 protein phosphorylation in 27:222–229. injured muscles. Immunofluorescence for phosphorylated-Smad2/3 proteins 13. Serrano AL, Murgia M, Pallafacchina G, Calabria E, Coniglio P, Lomo T, on sections from tibialis anterior (TA) muscle of mdx (A) and wild-type (WT) Schiaffino S: Calcineurin controls nerve activity-dependent specification (B) mice after the indicated treatments. Scale bars = 50 μm. of slow skeletal muscle fibers but not muscle growth. Proc Natl Acad Sci USA 2001, 98:13108–13113. Abbreviations 14. Glass DJ: Skeletal muscle hypertrophy and atrophy signaling pathways. BaCl : barium chloride; Coll I: collagen I; CTGF: connective tissue growth Int J Biochem Cell Biol 2005, 37:1974–1984. factor; CTX: cardiotoxin; DEN: denervation; DMD: Duchenne muscular 15. Morales MG, Cabello-Verrugio C, Santander C, Cabrera D, Goldschmeding R, dystrophy; ECM: extracellular matrix; FN: fibronectin; H&E: hematoxylin and Brandan E: CTGF/CCN-2 over-expression can directly induce features of eosin; LAC: laceration; NI: non-injured; PBS: phosphate-buffered saline; P-Smad2/ skeletal muscle dystrophy. J Pathol 2011, 225:490–501. 3: phosphorylated Smad2/3; TA: tibialis anterior; TGFβ1: transforming growth 16. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, factor beta1; TIMP-1: tissue inhibitor of metalloproteinases 1; WT: wild-type. Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A: Fiji: an open-source Competing interests platform for biological-image analysis. Nat Methods 2012, 9:676–682. The authors declare that they have no competing interests. 17. Cabello-Verrugio C, Morales MG, Cabrera D, Vio CP, Brandan E: Angiotensin II receptor type 1 blockade decreases CTGF/CCN2-mediated damage Authors’ contributions and fibrosis in normal and dystrophic skeletal muscles. J Cellular Mol Med PMC, EB and ALS conceived and designed the project. PP performed and 2012, 16:752–764. analyzed most of the experiments and was assisted by DC, MGM, CAR and 18. Biernacka A, Frangogiannis NG: Aging and cardiac fibrosis. Aging Dis 2011, JG for Figures 3 and 7. PMC and PP wrote the manuscript and ALS and EB 2:158–173. revised and edited it. All authors read and approved the final manuscript. 19. Brandan E, Gutierrez J: Role of proteoglycans in the regulation of the skeletal muscle fibrotic response. FEBS J 2013, 280:4109–4117. Acknowledgements 20. MacDonald EM, Cohn RD: TGFbeta signaling: its role in fibrosis formation We are indebted to E. Perdiguero, V. Lukesova, L. Correa, A. Vasquez, C. Mann and myopathies. Cur Opinion Rheumatol 2012, 24:628–634. and M. Raya for their continuous help and advice. We also thank previous 21. Mann CJ, Perdiguero E, Kharraz Y, Aguilar S, Pessina P, Serrano AL, members of our laboratories, especially E. Ardite and B. Vidal, for setting up Munoz-Canoves P: Aberrant repair and fibrosis development in skeletal the basis of this study, and J. Martín-Caballero for assistance in the PRBB muscle. Skelet Muscle 2011, 1:21. animal facility. The authors acknowledge funding from MINECO-Spain 22. Morales MG, Cabrera D, Cespedes C, Vio CP, Vazquez Y, Brandan E, (SAF2012-38547, FIS-PS09/01267, FIS-PI13/02512, PLE2009-0124), AFM, E-Rare, Cabello-Verrugio C: Inhibition of the angiotensin-converting enzyme Fundació-MaratóTV3, Duchenne PP-NL, EU-FP7 (Myoage, Optistem and decreases skeletal muscle fibrosis in dystrophic mice by a diminution in Endostem), MDA, CARE PFB12/2007 and FONDECYT 1110426. the expression and activity of connective tissue growth factor (CTGF/ CCN-2). Cell Tissue Res 2013, 353:173–187. Received: 29 October 2013 Accepted: 20 January 2014 23. Dangain J, Vrbova G: Muscle development in mdx mutant mice. Muscle Published: 25 August 2014 Nerve 1984, 7:700–704. 24. De Luca A, Pierno S, Liantonio A, Cetrone M, Camerino C, Fraysse B, References Mirabella M, Servidei S, Ruegg UT, Conte Camerino D: Enhanced 1. Emery AE: The muscular dystrophies. Lancet 2002, 359:687–695. dystrophic progression in mdx mice by exercise and beneficial effects 2. Briggs D, Morgan JE: Recent progress in satellite cell/myoblast of taurine and insulin-like growth factor-1. J Pharm Exp Thera 2003, engraftment - relevance for therapy. FEBS J 2013, 280:4281–4293. 304:453–463. Pessina et al. Skeletal Muscle 2014, 4:7 Page 17 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 25. Morales MG, Gutierrez J, Cabello-Verrugio C, Cabrera D, Lipson KE, 43. Desguerre I, Arnold L, Vignaud A, Cuvellier S, Yacoub-Youssef H, Gherardi Goldschmeding R, Brandan E: Reducing CTGF/CCN2 slows down mdx RK, Chelly J, Chretien F, Mounier R, Ferry A, Chazaud B: A new model of muscle dystrophy and improves cell therapy. Hum Mol Gen 2013, experimental fibrosis in hindlimb skeletal muscle of adult mdx mouse 22:4938–4951. mimicking muscular dystrophy. Muscle Nerve 2012, 45:803–814. 26. Perdiguero E, Sousa-Victor P, Ruiz-Bonilla V, Jardi M, Caelles C, Serrano AL, 44. Bernasconi P, Di Blasi C, Mora M, Morandi L, Galbiati S, Confalonieri P, Munoz-Canoves P: p38/MKP-1-regulated AKT coordinates macrophage Cornelio F, Mantegazza R: Transforming growth factor-beta1 and fibrosis transitions and resolution of inflammation during tissue repair. J Cell Biol in congenital muscular dystrophies. Neuromusc Dis 1999, 9:28–33. 2011, 195:307–322. 27. Suelves M, Lopez-Alemany R, Lluis F, Aniorte G, Serrano E, Parra M, Carmeliet P, doi:10.1186/2044-5040-4-7 Munoz-Canoves P: Plasmin activity is required for myogenesis in vitro and Cite this article as: Pessina et al.: Novel and optimized strategies for skeletal muscle regeneration in vivo. Blood 2002, 99:2835–2844. inducing fibrosis in vivo: focus on Duchenne Muscular Dystrophy. Skeletal Muscle 2014 4:7. 28. Suelves M, Vidal B, Serrano AL, Tjwa M, Roma J, Lopez-Alemany R, Luttun A, de Lagran MM, Diaz-Ramos A, Jardi M, Roig M, Dierssen M, Dewerchin M, Carmeliet P, Muñoz-Cánoves P: uPA deficiency exacerbates muscular dystrophy in MDX mice. J Cell Biol 2007, 178:1039–1051. 29. Carlson BM, Billington L, Faulkner J: Studies on the regenerative recovery of long-term denervated muscle in rats. Rest Neurol Neurosci 1996, 10:77–84. 30. Gargioli C, Coletta M, De Grandis F, Cannata SM, Cossu G: PlGF-MMP-9- expressing cells restore microcirculation and efficacy of cell therapy in aged dystrophic muscle. Nat Med 2008, 14:973–978. 31. Muir LA, Chamberlain JS: Emerging strategies for cell and gene therapy of the muscular dystrophies. Expert Rev Mol Med 2009, 11:e18. 32. Turgeman T, Hagai Y, Huebner K, Jassal DS, Anderson JE, Genin O, Nagler A, Halevy O, Pines M: Prevention of muscle fibrosis and improvement in muscle performance in the mdx mouse by halofuginone. Neuro Dis 2008, 18:857–868. 33. Alexakis C, Partridge T, Bou-Gharios G: Implication of the satellite cell in dystrophic muscle fibrosis: a self-perpetuating mechanism of collagen overproduction. Am J Physiol Cell Physiol 2007, 293:C661–C669. 34. Morrison J, Palmer DB, Cobbold S, Partridge T, Bou-Gharios G: Effects of T-lymphocyte depletion on muscle fibrosis in the mdx mouse. Am J Pathol 2005, 166:1701–1710. 35. Vidal B, Serrano AL, Tjwa M, Suelves M, Ardite E, De Mori R, Baeza-Raja B, Martinez de Lagran M, Lafuste P, Ruiz-Bonilla V, Jardí M, Gherardi R, Christov C, DierssenM,Carmeliet P, DegenJL, Dewerchin M, Muñoz-Cánoves P: Fibrinogen drives dystrophic muscle fibrosis via a TGFbeta/alternative macrophage activation pathway. Genes Dev 2008, 22:1747–1752. 36. Vidal B, Ardite E, Suelves M, Ruiz-Bonilla V, Janue A, Flick MJ, Degen JL, Serrano AL, Munoz-Canoves P: Amelioration of Duchenne muscular dystrophy in mdx mice by elimination of matrix-associated fibrin-driven inflammation coupled to the alphaMbeta2 leukocyte integrin receptor. Hum Mol Gen 2012, 21:1989–2004. 37. Villalta SA, Nguyen HX, Deng B, Gotoh T, Tidball JG: Shifts in macrophage phenotypes and macrophage competition for arginine metabolism affect the severity of muscle pathology in muscular dystrophy. Hum Mol Gen 2009, 18:482–496. 38. Kharraz Y, Guerra J, Mann CJ, Serrano AL, Munoz-Canoves P: Macrophage plasticity and the role of inflammation in skeletal muscle repair. Mediators Inflam 2013, 2013:491497. 39. Stedman HH, Sweeney HL, Shrager JB, Maguire HC, Panettieri RA, Petrof B, Narusawa M, Leferovich JM, Sladky JT, Kelly AM: The mdx mouse diaphragm reproduces the degenerative changes of Duchenne muscular dystrophy. Nature 1991, 352:536–539. 40. Chandrasekharan K, Yoon JH, Xu Y, deVries S, Camboni M, Janssen PM, Varki A, Martin PT: A human-specific deletion in mouse Cmah increases disease severity in the mdx model of Duchenne muscular dystrophy. Sci Trans Med 2010, 2:42ra54. 41. Wehling-Henricks M, Jordan MC, Gotoh T, Grody WW, Roos KP, Tidball JG: Submit your next manuscript to BioMed Central Arginine metabolism by macrophages promotes cardiac and muscle and take full advantage of: fibrosis in mdx muscular dystrophy. PloS one 2010, 5:e10763. 42. Zhou L, Rafael-Fortney JA, Huang P, Zhao XS, Cheng G, Zhou X, Kaminski • Convenient online submission HJ, Liu L, Ransohoff RM: Haploinsufficiency of utrophin gene worsens • Thorough peer review skeletal muscle inflammation and fibrosis in mdx mice. JNeurol Sci 2008, 264:106–111. • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Skeletal Muscle Springer Journals

Novel and optimized strategies for inducing fibrosis in vivo: focus on Duchenne Muscular Dystrophy

Loading next page...
 
/lp/springer-journals/novel-and-optimized-strategies-for-inducing-fibrosis-in-vivo-focus-on-OF5nAIDRYk
Publisher
Springer Journals
Copyright
Copyright © 2014 by Pessina et al.; licensee BioMed Central Ltd.
Subject
Life Sciences; Cell Biology; Developmental Biology; Biochemistry, general; Systems Biology; Biotechnology
eISSN
2044-5040
DOI
10.1186/2044-5040-4-7
pmid
25157321
Publisher site
See Article on Publisher Site

Abstract

Background: Fibrosis, an excessive collagen accumulation, results in scar formation, impairing function of vital organs and tissues. Fibrosis is a hallmark of muscular dystrophies, including the lethal Duchenne muscular dystrophy (DMD), which remains incurable. Substitution of muscle by fibrotic tissue also complicates gene/cell therapies for DMD. Yet, no optimal models to study muscle fibrosis are available. In the widely used mdx mouse model for DMD, extensive fibrosis develops in the diaphragm only at advanced adulthood, and at about two years of age in the ‘easy-to-access’ limb muscles, thus precluding fibrosis research and the testing of novel therapies. Methods: We developed distinct experimental strategies, ranging from chronic exercise to increasing muscle damage on limb muscles of young mdx mice, by myotoxin injection, surgically induced trauma (laceration or denervation) or intramuscular delivery of profibrotic growth factors (such as TGFβ). We also extended these approaches to muscle of normal non-dystrophic mice. Results: These strategies resulted in advanced and enhanced muscle fibrosis in young mdx mice, which persisted over time, and correlated with reduced muscle force, thus mimicking the severe DMD phenotype. Furthermore, increased fibrosis was also obtained by combining these procedures in muscles of normal mice, mirroring aberrant repair after severe trauma. Conclusions: We have developed new and improved experimental strategies to accelerate and enhance muscle fibrosis in vivo. These strategies will allow rapidly assessing fibrosis in the easily accessible limb muscles of young mdx mice, without necessarily having to use old animals. The extension of these fibrogenic regimes to the muscle of non-dystrophic wild-type mice will allow fibrosis assessment in a wide array of pre-existing transgenic mouse lines, which in turn will facilitate understanding the mechanisms of fibrogenesis. These strategies should improve our ability to combat fibrosis-driven dystrophy progression and aberrant regeneration. Background dystrophies and is caused by loss of the dystrophin protein In skeletal muscle, accumulation of collagens (fibrosis) in due to genetic mutations. As a result, the sarcolemma be- the extracellular matrix (ECM) is most often associated comes fragile and susceptible to contraction-induced with the muscular dystrophies, characterized by muscle damage [1]. Skeletal muscle stem cells (satellite cells) me- wasting, leading to loss of patient mobility. Duchenne diate the repair process, but in the absence of dystrophin, muscular dystrophy (DMD) is one of the severest of the the muscle undergoes continuous cycles of degeneration and regeneration, eventually leading to satellite cell deple- * Correspondence: ebrandan@bio.puc.cl; pura.munoz@upf.edu tion and myofiber loss [2-4]. The severity of this childhood- Department of Cell and Molecular Biology, Catholic University of Chile, associated pathology may also be exacerbated by the Avenida Libertador Bernardo O’Higgins, 340, Santiago, Chile Cell Biology Group, Department of Experimental and Health Sciences, CIBER growth of myofibers that occurs in boys with DMD over on Neurodegenerative Diseases (CIBERNED), Pompeu Fabra University (UPF), many years [5]. Affected children eventually succumb to Dr. Aiguader, 88, 08003 Barcelona, Spain 3 muscle wasting, with muscle progressively being replaced Institució Catalana de Recerca i Estudis Avançats (ICREA), Dr. Aiguader, 88, 08003 Barcelona, Spain © 2014 Pessina et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 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. Pessina et al. Skeletal Muscle 2014, 4:7 Page 2 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 by fat and fibrotic tissue, leading to premature death in the dystrophic C57Bl/10scsn-mdx (mdx) male mice were used late teens or early twenties from respiratory and cardiac in experiments: the background strain for mdx mice is failure [6]. There are currently only palliative treatments for similar to, but not identical with, the C57Bl/6 J strain. All DMD patients. Importantly, no effective clinical treatments operations were performed after injection intraperitoneal are available yet to combat or attenuate fibrosis in patients (i.p.) of ketamine/metedomidine anesthesia (50 mg/kg with DMD. Halting or diminishing the development of fi- and 1 mg/kg body weight). Atipamezol (1.0 mg/kg body brosis could not only ameliorate DMD progression, but weight) by subcutaneous injection was used to reverse could also increase the success of new cell- and gene-based the effects of anesthesia. Mice were sacrificed at the in- therapies [7,8]. The mdx mouse strain, the most widely dicated ages and the tissues were immediately processed used animal model for studying human DMD, has a non- to avoid artifacts, either by direct freezing in liquid nitro- sensemutationin dystrophinexon23leading to dystrophin gen for protein and RNA extraction or in 2-methylbutane protein absence [9]. Although mdx mice and DMD patients cooled with liquid nitrogen for histological analysis, as de- share many genetic, biochemical and histological similar- scribed below. ities, the clinical manifestations are generally less severe in mdx mice[10]. WhileDMD individualshaveahighdegree Skeletal muscle fibrogenic treatments of muscle fibrosis, mdx mice present extensive fibrosis ex- clusively in the diaphragm muscle. In the limb muscles of – Chronic exercise: Mdx mice were exercised three mdx mice, however, fibrosis only becomes apparent around times per week on a treadmill for 30 minutes at a 20 months of age [11]. Therefore, despite our awareness of speed of 12 meters per minute, with a rest of the importance of fibrosis in DMD, there is a lack of ap- 5 minutes every 10 minutes of exercise. Mdx mice propriate mouse models for studying dystrophic skeletal of three, four and five months of age were exercised muscle fibrosis in accessible muscles, such as limb mus- for three, two and one month, respectively and were cles, without requiring nearly two years for fibrosis to ap- sacrificed at the age of six months, together with pear. Thus, there is a genuine need to develop mouse age-matched unexercised control mdx mice. At the models that present fibrosis at early stages in life and that end of the training period muscles were collected more closely mimic human DMD. and processed for further analyses. In this manuscript, we present several experimental – Single treatments in WT and mdx mice: strategies to simply and effectively advance and enhance  Myotoxin-induced injuries: muscle fibrosis in young mdx mice. We use both physio- Cardiotoxin injury: Tibialis anterior (TA) muscles logical exercise, as well as more direct tissue-damaging of three-month-old WT or mdx mice were –5 procedures or delivery of profibrotic growth factors to injected with 50 μlof10 M cardiotoxin (CTX; limb muscle of young dystrophic mice, and demonstrate Latoxan, Rosans, France). Muscles were collected sustained collagen deposition reminiscent of aged mdx at the indicated times on each set of experiments, diaphragm muscle, and muscle of human DMD patients. which was usually two weeks after myotoxin Notably, we could extend these strategies to induce fi- injection. Muscle samples were also obtained at brosis in muscles of normal, non-dystrophic mice, which one month post-injection (in WT mice) and two will facilitate studying fibrosis in a wide array of genetic- months post-injection (in mdx mice). ally modified mouse lines, and this will in turn increase Barium chloride injury: TA muscle of three- our understanding of the cells and molecules involved in month-old WT mice was injected with 50 μlof fibrosis development. Thus, we offer for the first time, to 0.2% barium chloride (BaCl )and was isolatedafter the best of our knowledge, a comparative and quantita- two weeks. For repeated BaCl injuries, BaCl injec- 2 2 tive set of new and improved strategies for inducing tions were made in the same muscle, one per week muscle tissue fibrosis, which will greatly foster our abil- for six weeks, and muscles were isolated and col- ity to combat fibrosis-dependent dystrophy progression. lected for analysis two weeks after the last injection. Traumatic injuries: Methods Laceration: TA muscles of three-month-old WT Mice handling and sample collection or mdx mice were subjected to laceration (LAC) All experiments were approved by the Ethics Committee as previously described [11,12]. Briefly, the skin of the Pompeu Fabra University (UPF) and performed was carefully cut and separated from the according to Spanish and European legislation. Mice were underlying tissue, then the TA muscle of one leg housed in standard cages under 12-hour light–dark cycles was cut horizontally at its middle of the length by and fed ad libitum with a standard chow diet. Three- making a lesion through 75% of their width and month-old normal C57Bl/6 J mice (the classic standard la- 50% of the muscle thickness with a scalpel. boratory mouse strain, hereafter referred to as WT) and Contralateral control muscles were sham-operated. Pessina et al. Skeletal Muscle 2014, 4:7 Page 3 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 In WT mice, muscles were collected at two weeks Dystrophic patients study and one month post-surgery. In mdx mice, Human samples were provided from Dr. J. Colomer muscles were obtained at two weeks and two (Hospital Sant Joan de Deu, Barcelona, Spain). DMD diag- months post-surgery. nosis was established on a total absence of dystrophin by Denervation: Muscle denervation (DEN) was immunohistochemistry and Western blotting. Muscle sam- performed as previously described [13,14]. In brief, a ples were obtained by a standard quadriceps muscle biopsy 5 mm segment of the sciatic nerve was surgically from six DMD patients (ranging from five to eleven years removed down to the gluteus maximum from the of age) and five healthy male human controls of similar age right legs. In three-month-old WT mice, TA (seven to fourteen years). Quantification of fibrosis was muscles were isolated for analysis at two weeks and carried out by color image segmentation and automatic one month post-surgery. In three-month-old mdx measurement using Fiji image analysis software [16]. The mice, TA muscles were collected at two weeks and ratio of the total area of fibrosis to the total biopsy area two months post-surgery. Contralateral muscles of was used to estimate the extent of fibrosis (fibrosis index). sham-operated mice were used as controls on every Histological analysis was performed similarly to mouse single treatment (CTX, BaCl ,LAC andDEN). samples, as explained in the next section. Profibrotic growth factor treatments: Transforming growth factor beta treatment:50 ng Histological analysis and immunohistochemistry of transforming growth factor beta 1 (TGFβ1) Cryosections (10 μm thickness) were stained with (recombinant human TGFβ1; R&D Systems, hematoxylin/eosin (H&E) or Sirius red (Sigma-Aldrich, Minneapolis, MN, USA) were injected in the TA St Louis, MO, USA). Quantification of collagen content muscle in a volume of 50 μl of phosphate-buffered in muscle was performed according to Ardite et al. [11]. saline (PBS, vehicle). Two injections (one per week) Briefly, 10 cryosections were collected in a tube and were were made in TA muscle of three-month-old mdx sequentially incubated with a solution containing 0.1% mice, and muscles were collected for analysis two Fast green in saturated picric acid, washed with distillated weeks or two months after the first injection. water, incubated with 0.1% Fast green and 0.1% Sirius red Connective tissue growth factor delivery:TA in saturated picric acid, washed with distillated water, and muscles of three-month-old mdx mice were gently resuspended in a solution of 0.1 M NaOH in abso- injected with 1 × 10 viral particles of Ad-m lute methanol (1 vol:1 vol). Absorbance was measured in a connective tissue growth factor (CTGF) or spectrophotometer at 540 and 605 nm wavelengths and Ad-GFP or with PBS (vehicle) [15] as a control in used to calculate total protein and collagen. a total volume of 50 μl. Muscles were collected Immunohistochemistry on frozen sections was per- two weeks after injury. formed using the following primary antibodies: rabbit poly- – Combined treatments in WT mice: clonal collagen I (Coll 1) (Millipore, Billerica, MA, USA), Cardiotoxin injury combined with denervation: TA rabbit polyclonal fibronectin (FN) (Abcam, Cambridge, muscles of three-month-old WT mice were injected MA, USA) and rabbit polyclonal phosphorylated-Smad2/3 –5 with 50 μlof10 M CTX immediately after DEN, (P-Smad2/3) (Abcam). For immunoperoxidase staining, as indicated above. Muscles were collected at two labeling of sections was performed using the peroxidase weeks and one month after the treatments. staining kit (Vector Laboratories, Burlingame, CA, USA) Contralateral muscles of non-denervated left legs according to the manufacturer’s instructions. For immuno- were used as controls. fluorescence, secondary antibodies were coupled to Alexa Cardiotoxin injury combined with TGFβ/CTGF Fluor 488 or 568 fluorochromes (Invitrogen, Carlsbad, treatment: TA muscles of three-month-old WT mice CA, USA). Stained sections were photographed on a –5 were injected with 50 μlof10 M CTX. TGFβ1 Leica DM6000B microscope (Leica Microsystems, Wetzler, was injected intramuscularly twice at day 7 and 10 Germany). after cardiotoxin injection. Muscles were collected at two weeks and one month after the cardiotoxin RNA isolation, reverse transcription (RT) and real-time injection. Contralateral muscles of sham-operated quantitative PCR legs were used as controls. When indicated, CTGF Total RNA was isolated from muscle tissue using Trizol adenoviral delivery was performed immediately after (Invitrogen). cDNA was synthesized from 1 μg of total cardiotoxin injection. RNA using the First Strand cDNA Synthesis kit and ran- dom priming according to the manufacturer’s instructions Specific information about starting and sampling ages (Promega, Madison, WI, USA). RT-PCR was performed of mice after the different experimental protocols is in- on a LightCycler 480 System using LightCycler 480 SYBR cluded in Table S1 in Additional file 1. Green I Master Mix (Roche, Basel, Switzerland) with Pessina et al. Skeletal Muscle 2014, 4:7 Page 4 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 10 μM each primer and normalized to L7 ribosomal RNA Results as a housekeeping gene: mL7 5′-GAAGCTCATCTATG Mdx mice reproduce the human DMD fibrotic phenotype AGAAGGC–3′ and 5′–AAGACGAAGGAGCTGCAGA in aging diaphragm muscle AC-3′; mCollagen I, 5′-GGTATGCTTGATCTGTATCT To recreate as closely as possible the fibrosis status of GC-3′ and 5′-AGTCCAGTTCTTCATTGCATT-3′;mC human DMD in animal models, we first sought to TGF, 5′-CAGGCTGGAGAAGCAGAGTCGT-3′ and 5′- characterize in detail distinct fibrosis-associated parame- CTGGTGCAGCCAGAAAGCTCAA–3′;mTIMP-1 5′- ters in muscle biopsies of DMD patients. Compared to TTCCAGTAAGGCCTGTAGC-3′ and 5′-TTATGACCA muscles of healthy individuals, we found an increased col- GGTCCGAGTT-3′;mTGFβ 5′-TATGACCAGGTCCGA lagen content in DMD patients, based on Sirius red stain- GTT-3′ and 5′-CTGGTGCAGCCAGAAAGCTCAA-3′; ing and collagen quantification, where fibrotic tissue had hFibronectin: 5′- GGATGACAAGGAAAATAGCCCTG- replaced the myofiber area (Figure 1A and B). Transform- 3′ and 5′-GAACATCGGTCACTTGCATCT-3′; hTIMP- ing growth factor-β (TGFβ) has been shown to be a profi- 15′-CTTCTGCAATTCCGACCTCGT-3′ and 5′-CCCT brotic cytokine in many types of fibrotic tissues and is a AAGGCTTGGAACCCTTT-3′; hTGFβ 5′-CCTAA GGC potent stimulator of matrix production, including colla- CAGATCCTGTCCAAGC-3′ and 5′- GTGGGTTTCCA gen, by fibroblasts [13,18-22]. We found higher levels of CCATTAGCAC-3′;hCTGF 5′- CAAGGGCCTCTTCTG activated TGFβ protein in muscle biopsies from dys- TGACT-3′ and 5′-ACGTGCACTGGTACTTGCAG-3′. trophic children compared to healthy controls (Figure 1C). Consistent with this, we found enhanced levels of active Smad2/3 (as indicated by phosphorylated Smad2/3) Quantification of TGFβ protein (Figure 1D) and TGFβ target genes such as Coll I, FN, The protein concentration of active and total (active tissue inhibitor of metalloproteinases 1 (TIMP-1) and plus latent) TGFβ1 levels in dystrophic muscle was quan- CTGF, indicative of functional TGFβ signaling in fi- tified by ELISA (Promega), following the manufacturer’s brotic DMD muscle (Figure 1E). instructions. The most common experimental model of DMD is the mdx mouse [23]. We examined the TA limb muscle and the diaphragm muscle by hematoxylin and eosin (H&E) Muscle force measurement and Sirius red staining from young (three months of Muscle strength was determined as described previously age), adult (nine months) and old mdx mice (eighteen to [17]. Briefly, after the indicated days of treatment, mice twenty-four months) in comparison to age-matched WT were sacrificed and the TA was rapidly excised into a muscles. Significant fibrosis, similar to that observed in dish containing oxygenated Krebs-Ringer solution. The human patients, was found in TA muscles of mdx mice optimum muscle length (Lo) was determined from mi- only at old age (>18 months) (Figure 2A, upper panels cromanipulations of muscle length to produce the max- and Figure 2D), while adult mdx TA muscles presented imum isometric twitch force. Maximum isometric-specific milder fibrosis. In diaphragm muscle, fibrosis increased tetanic force was determined from the plateau of the age-dependently, reaching near maximum levels in adult curve of the relationship between specific isometric force mice of nine months of age and plateauing thereafter with a stimulation frequency (Hz) ranging from 1 to (Figure 2A, lower panels). Furthermore, in TA muscles of 200 Hz for 450 ms, with 2 minutes of rest between stimuli. mdx mice, collagen content, activated TGFβ and expres- The force was normalized per total muscle fiber cross- sion of ECM-associated molecules started to increase at sectional area (CSA), to calculate the specific net force adult age but were much higher at old age (Figure 2B, D, F, (mN/mm ). G); in the diaphragm muscle, these parameters were mod- erately increased already at young age (Figure 2C and E, Statistical analysis and Figure S1A and B in Additional file 2). The limited de- Comparison between groups was done using the non- velopment of fibrosis (compared to the diaphragm muscle) parametric Mann–Whitney U test for independent sam- in the easily accessible limb muscles of mdx mice until old ples, with a confidence level of 95% being considered age, reinforces the need for developing new protocols that statistically significant. One-way or two-way analysis of will advance muscle fibrosis in young mdx mice. variance (ANOVA) was used for comparisons between multiple groups as appropriate, and post hoc analysis Exercise training triggers fibrosis in muscles of young was performed using Tukey’s test. All statistical analyses dystrophic mice were performed using GraphPad Prism 5.0 (GraphPad In a first attempt to induce and advance muscle fibrosis, Software, San Diego, CA, USA). The number of samples young mdx mice were subjected to a chronic exercise analyzed per group is detailed on each figure. Differences training routine, known to exacerbate the muscle degen- were considered to be statistically significant at P <0.05. eration/regeneration process [24]. Three-month-old mdx Pessina et al. Skeletal Muscle 2014, 4:7 Page 5 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 Human Normal Patient Normal ** Patient Sirius red Human 1800 Normal Patient Normal ** Patient P-Smad2/3 Co ll I FN TIMP-1 E CTGF Normal * Patient * 10 100 3 8 50 5 4 0 0 0 0 0 Figure 1 Quantification of fibrosis in human dystrophic muscle. (A) Representative Sirius red staining of healthy and dystrophic human muscle sections reveals the extent of collagen deposition in patients with Duchenne muscular dystrophy (DMD). (B) Percentage of fibrosis (collagen content) in healthy and DMD muscles as measured by Sirius red staining in muscle sections. Data correspond to the mean ± SEM; n = 6 for DMD group and n = 5 for control group. Non-parametric Mann–Whitney U test was used for comparison. **P <0.01 versus healthy controls. (C) Active transforming growth factor beta 1 (TGFβ1) protein levels measured by ELISA in muscle biopsy material from healthy and DMD muscle. Data correspond to the mean ± SEM; n = 5 on each group. Non-parametric Mann–Whitney U test; **P <0.01 versus healthy controls. (D) Immuno- histochemistry for phosphorylated-Smad2/3 (P-Smad2/3) in healthy and dystrophic human muscle sections. (E) Quantitative RT-PCR for collagen I (Coll 1), fibronectin (FN), tissue inhibitor of metalloproteinases 1(TIMP-1), TGFβ1 and connective tissue growth factor (CTGF) in DMD muscles compared to healthy muscles (which were given the arbitrary value of 1). Data correspond to the mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test *P <0.05. Scale bars = 50 μm. mice were exercised on a treadmill three times per week respect to normally active non-exercised mdx control for up to three months, for 30 minutes each time, at a mice (Figure 3A, and Figure S2A in Additional file 3). The speed of 12 meters per minute, with a rest of 5 minutes increased fibrosis observed by Sirius red staining was con- every 10 minutes [25]. Muscles of mice exercised for one, firmed by Coll I immunofluorescence (Figure 3B, upper two and three months were compared with age- and sex- panels, and Figure S2B in Additional file 3), and the greater matched unexercised mice. After one month, exercised deposition of FN (Figure 3B, and Figure S2B in Additional mdx mice already showed a worsening of the dystrophic file 3, lower panels) that is normally only observed in old phenotype (compared to age-matched controls), and this mdx limb muscles (Figure 2G). Consistent with this, colla- condition was further aggravated by continued adherence gen content and the expression of TGFβ1and CTGF to the exercise regime. Hindlimb muscles (gastrocnemius mRNA, and the levels of P-Smad2/3 proteins, were in- and TA) of one-month exercised mice displayed a higher creased in exercised dystrophic mdx muscles, compared to degree of fibrosis, identified by Sirius red staining, with non-exercised controls (Figure 3C, D, E, and Figure S2C in TGF 1 TGFβ1 active protein Relative expression pg/mg protein % fibrosis Pessina et al. Skeletal Muscle 2014, 4:7 Page 6 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 WT Young Adult Old WT mdx *** ** WT 600 mdx *** *** 200 ** TA Diaphragm D E 100 *** WT WT *** 40 *** mdx mdx ** *** Young Adult Old Young Adult Old Collagen I CTGF TGFβ1 TIMP-1 40 30 6 *** Young mdx *** * *** 8 Adult mdx 20 Old mdx 10 2 ** 0 0 0 0 mdx WT Young Adult Old Figure 2 Quantification of fibrosis in muscle of mdx mice. (A) Sirius red and H&E staining of mdx tibialis anterior (TA) (upper panels) and diaphragm (lower panels) muscles at different ages compared to adult wild-type (WT) muscle. ‘Young’ corresponds to muscles of three-month-old mice, ‘Adult’ to nine months and ‘Old’ to eighteen to twenty-four months of age. (B) and (C) Active transforming growth factor beta 1 (TGFβ1) protein quantifi- cation by ELISA in TA and diaphragm muscles of WT and mdx mice at the indicated ages, respectively. Data correspond to the mean ± SEM values; n = 4 for each group. Two-way analysis of variance with Tukey’s post hoc multiple comparison test. **P <0.01, ***P <0.001 versus age-matched WT. (D) and (E) Quantification of collagen content in TA and diaphragm muscles of WT and mdx mice at different ages. Values are mean ± SEM; n = 4 for each group. Two-way analysis of variance with Tukey’s post hoc multiple comparison test. **P <0.01, ***P <0.001 versus control WT values. (F) Relative expression of collagen I, connective tissue growth factor (CTGF), tissue inhibitor of metalloproteinases 1(TIMP-1) and TGFβ1mRNAbyquantitativeRT-PCRinmdx TA muscles at the indicated ages with respect to WT muscles (baseline set arbitrarily to 1). Values are mean ± SEM; n = 3 for each group. Two-way analysis of variance with Tukey’s post hoc multiple comparison test. *P <0.05, **P <0.01, ***P <0.001, versus age-matched WT. (G) Representative pictures of immuno- fluorescence staining for collagen I (green) and fibronectin (red) in young, adult and old mdx TA, compared to WT muscle. Scale bars = 50 μm. Young Adult Old Young Adult Old Diaphragm TA Relative expression Fibronectin Collagen I μg collagen/mg protein (versus WT) μg collagen/mg protein TGFβ1 active protein TGFβ1 active protein pg/mg protein pg/mg protein Pessina et al. Skeletal Muscle 2014, 4:7 Page 7 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 Months of exercise 0 1 2 3 CD CTGF TGF 1 4 40 ** *** *** ** ** *** ** 1 10 0 0 0 01 2 3 01 2 3 01 2 3 Months of exercise F Tetanic force Unexercised Exercised ** 03 Months of exercise P-Smad2/3 Figure 3 (See legend on next page.) Relative expression Fibronectin Collagen I H&E Sirius red mN/mm μg collagen/mg protein Pessina et al. Skeletal Muscle 2014, 4:7 Page 8 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 (See figure on previous page.) Figure 3 Effect of exercise on muscle fibrosis in mdx mice. (A) Sirius Red and H&E staining of gastrocnemius muscle of mdx mice that were exercised three times weekly, for 30 minutes at a speed of 12 meters per minute with a rest of 5 minutes each 10 minutes of exercise, for one, two and three months, compared to sections of muscle from unexercised mdx mice. All the samples were collected when the animals were six months old (see the Methods section). (B) Representative immunofluorescence for collagen I and fibronectin in muscle sections of control or exercised mice as shown in (A). (C) Quantitative RT-PCR of connective tissue growth factor (CTGF) and transforming growth factor beta 1 (TGFβ1) mRNA levels after exercising for the indicated period as compared to unexercised age-matched mdx mice. Data correspond to the mean ± SEM; n = 4 sedentary and 4 exercised mdx mice for each exercise time point. One-way analysis of variance with Tukey’s post hoc multiple comparison test; *P <0.05, **P <0.01, ***P <0.001 versus control. (D) Biochemical quantification of collagen protein content in mdx gastrocnemius muscle after exercising for the indicated periods, as compared to unexercised age-matched mdx mice. Data correspond to the mean ± SEM; n = 4 sedentary and 4 exercised mdx mice for each exercise time point. One-way analysis of variance with Tukey’s post hoc multiple comparison test; *P <0.05, **P <0.01, ***P <0.001 versus control. (E) Immunofluorescence for phosphorylated-Smad2/3 proteins on sections from gastrocnemius muscle of six-month-old mdx mice after three months of exercise, as evidence for TGFβ activation, compared to unexercised age-matched control mdx mice. (F) Ex vivo maximum isometric force (tetanic force) of gastrocnemius muscle of age-matched unexercised and three-month-trained mdx mice. Values as mean ± SEM; n = 7 on each group. Non-parametric Mann–Whitney U test; **P <0.01 versus non-exercised. Scale bars = 50 μm. Additional file 3). Furthermore maximum force of the biochemical parameters (see below, Figure 4F and G). This muscles of mdx mice subjected to the exercise regime was extended fibrotic status reinforces the utility of these two decreased with respect to non-exercised mice (Figure 3F, methods as drivers of limb muscle fibrosis in young mdx and Figure S2D in Additional file 3). These data confirm mice, after which the tissue more closely resembles the that exercise in young mdx mice can activate fibrogenesis, more severe phenotype of old mdx mice, as well as human and in particular the profibrotic TGFβ pathway, and DMD patients. Furthermore, these procedures have the thereby enhance and anticipate muscle tissue fibrosis. advantage of not requiring exercise devices, nor the time and labor of the three-month exercise protocol. Surgical muscle injuries advance and enhance fibrosis in young dystrophic mice We next sought alternative and faster ways than long- Raising TGFβ levels in dystrophic muscle of young mdx term exercise training to induce fibrosis in limb muscles mice accelerates fibrosis and accentuates disease severity of young mdx mice, based on inflicting increased surgical Despite the profibrotic effect of the surgical methods on or chemical damage. Since CTX-induced muscle injury is a mdx muscle, each one has particularities and limitations. widely used and well-characterized experimental model for In the LAC model, the injury is confined to a small area inducing skeletal muscle degeneration/regeneration [26-28], of the muscle and this reduces the amount of tissue we hypothesized that superimposing CTX-induced damage available for further studies, whereas, for reasons of ani- on young dystrophic mdx muscle would promote fibrosis. mal welfare, DEN can only be performed in one leg of Despite an early increase in collagen content, two weeks the mouse, affecting only the muscles under the knee. –5 after intramuscular CTX injection (50 μlof10 M), TA Therefore, based on our observation of the elevated levels mdx muscle showed a similar quantity of deposited colla- of TGFβ in human and mouse dystrophic muscle (Figures 1 gen compared to non-injured (NI) mdx TA indicating that and 2), and its correlation with the extent of dystrophy- thefibrogeniceffect of CTX-induced damage was transient associated fibrosis, we reasoned that exogenous delivery of (Figure 4A). To increase and prolong collagen deposition, TGFβ1 to muscle of young dystrophic mice might increase we superimposed on young mdx limb muscle two more and accelerate the development of fibrosis. We therefore extreme, but distinct, experimental paradigms: lacer- performed two intramuscular TA injections of TGFβ1 ation (LAC) and denervation (DEN). The DEN model (50 ng of TGFβ1in50 μl of PBS per injection), spaced involves severing the sciatic nerve thus causing atrophy sevendaysapart, inanattempt to sustainthe profibrogenic of the denervated myofibers [14,29], while the LAC action of this growth factor. Contralateral control muscles model consists in a deep cut across the muscle, which received the same number of injections of PBS. Analysis of causes a delay in the healing process [11,12]. Muscle of the muscles histologically by H&E and Sirius red staining dystrophic mdx mice at two weeks after DEN showed showed that TGFβ1 delivery lead to substantial increase in an increased deposition of collagen relative to CTX- collagen deposition already at two weeks after the first in- injured mdx muscle (Figure 4A, C). Lacerated dystrophic jection, which persisted for up to two months and this was muscle also showed increased fibrosis after two weeks, also confirmed by biochemical quantification of muscle which was even higher than in denervated muscle after extracts (Figure 4B and C). Of note, local muscle overex- the same time period (Figure 4A, C). Importantly, the pression of the TGFβ1 target gene product CTGF also in- mdx muscle fibrosis induced by both methods persisted creased fibrogenesis in limb muscle of young mdx mice for up to two months, as indicated by histological and (Figure S3A, B in Additional file 4). Pessina et al. Skeletal Muscle 2014, 4:7 Page 9 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 N.I. CTX DEN LAC TGFβ C D Collagen I TGF 1 CTGF TIMP-1 N.I. 50 40 5 3 10 CTX * * ** 40 4 30 DEN * * LAC 30 3 6 * TGF 10 * 20 2 4 10 1 0 0 0 0 N.I CTX DEN LAC TGFβ N.I. CTX DEN LAC TGFβ Tetanic force N.I. H N.I. CTX CTX * * DEN DEN LAC LAC TGF TGF ** Figure 4 (See legend on next page.) μg collagen/mg protein μg collagen/mg protein H&E Sirius red Fibronectin Collagen I H&E Sirius red Relative expression mN/mm Pessina et al. Skeletal Muscle 2014, 4:7 Page 10 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 (See figure on previous page.) Figure 4 Induction of fibrosis after chemical and surgical muscle damage in young mdx mice. (A) Sirius red and hematoxylin and eosin –5 (H&E) staining of tibialis anterior (TA) muscles of young (three-month-old) mdx mice two weeks after cardiotoxin (CTX)-injury (50 μlof10 M), denervation (DEN) and laceration (LAC), compared to non-injured (NI) muscle of sham-operated mdx mice. (B) Sirius red and H&E staining of young mdx TA muscle analyzed after two sequential treatments with recombinant transforming growth factor beta 1 (TGFβ1) (50 ng in 50 μl phosphate-buffered saline (PBS)) spaced seven days apart. (C) Biochemical quantification of collagen protein content in mdx TA, two weeks after different treatments relative to uninjured mdx control. Values represent mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (D) Quantitative RT-PCR of collagen I, connective tissue growth factor (CTGF), tissue inhibitor of metalloproteinases 1(TIMP-1) and TGFβ1 mRNA expression in mdx muscle two weeks after different injuries versus control mdx mice. Values represent mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (E) Representative immunostaining for collagen I (green) and fibronectin (red) on sections of young mdx TA muscles two weeks after injury relative to control. (F) Sirius Red and H&E staining of mdx TA muscle two months after CTX injury, DEN, LAC or injection of TGFβ in three-month-old mdx muscle compared to NI mdx control muscle. (G) Quantification of collagen content in TA muscle of young mdx mice two months after different treatments, as described above. Values represent mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (H) Ex vivo maximum isometric force (tetanic force) of TA muscle of young mdx mice two months after treatments. Values as mean ± SEM; n = 4 to 5 on each group. Non-parametric Mann–Whitney U test; *P <0.05; **P <0.01 versus NI. Scale bars = 50 μm. Overall, comparing the distinct biochemical and func- repair. Thus, we designed easy-to-perform profibrotic tional parameters in all the procedures tested revealed procedures in non-dystrophic WT muscle, which could that LAC and TGFβ treatments gave statistically higher ideally be extended to a wide variety of transgenic mouse quantitative measures of collagen than NI age-matched lines for research or therapeutic purposes. control mdx muscles. The collagen values for LAC and We applied the surgical/chemical methods previously TGFβ1 treatments were comparable to the values re- used on muscle of dystrophic mdx mice (see above), ei- corded in limb muscles of old mdx mice (see Figure 2D), ther alone or in combination, to induce muscle fibrosis indicating that either one of these methods advances fibro- in WT mice. First, we performed CTX injury in TA sis by the equivalent of about fourteen months (that is in- muscle of WT mice and assessed fibrosis development. ducing fibrosis at four months of age instead of eighteen We observed a mild and transient deposition of ECM –5 months). DEN also significantly increases muscle collagen between days 5 and 7 after CTX (50 μlof10 M) content over mdx controls, but to a lesser extent than LAC muscle injury (Figure S4A and B in Additional file 5); or TGFβ1 treatment (Figure 4A and C). Interestingly, the however, it did not persist beyond this stage. Indeed, two levels of endogenous TGFβ1 mRNA were increased in weeks after CTX injury, collagen content returned to young dystrophic muscle in response to LAC, DEN and ex- near basal levels, in agreement with efficient muscle re- ogenous TGFβ1 delivery, but not CTX. Consistent with covery (Figure 5A-C and Figure 6H). this, the expression of TGFβ-dependent signaling fibrotic We next subjected WT muscle to the more severe target genes, such as, Coll I, CTGF, TIMP-1, were increased LAC and DEN procedures and compared the fibrosis in mdx limb muscle after all three treatments, but not in index of the affected muscles to that of CTX-injured CTX-damaged muscle (Figure 4D). Finally, immunostain- muscle at similar time points. LAC in TA muscle of WT ing for FN and Coll I on sections from the different dam- mice disrupted the tissue quite extensively and for a pro- aged mdx muscles showed greater ECM production than longed period of time (over one month) resulting in sus- uninjured (or CTX-injured) dystrophic muscle (Figure 4E). tained fibrosis, which correlated with the slow kinetics Remarkably, at two months after injury, collagen depos- for regeneration (Figure 5A, B). DEN, in turn, did not ition still persisted in TGFβ-treated young dystrophic mus- alter ECM production significantly, as revealed by H&E cles as it did in lacerated and denervated muscles, as and Sirius red staining, or immunostaining for Coll I revealed by histological and biochemical analysis (Figure 4F and FN, despite inducing the expected myofiber atrophy. and G). In agreement with this, and demonstrating the Consistent with these findings, we only observed statisti- deleterious physiological consequences of the increased fi- cally significant increases in the expression of TGFβ1 brosis in young mdx muscles, the maximum force of the and the fibrotic markers Coll I, FN, CTGF and TIMP-1 in muscles subjected to the distinct treatments decreased with lacerated muscle, but not in denervated or CTX-injured respect to NI mdx muscles (Figure 4H), therefore better muscles, compared to uninjured muscle (Figure 5C). mimicking the severe phenotype of the human condition. These results suggest that LAC is the most fibrotic of the traumatic models tested in non-dystrophic mice. Induction of fibrosis in non-dystrophic, wild-type muscle As stated above, one of the limitations of the LAC pro- by combining surgical injury and growth factor delivery cedure is the restricted availability of biopsy material. Try- Fibrosis persistence has negative consequences on tissue ing to induce fibrosis by methods that would render more wound healing. Severe muscle injuries caused by trauma fibrotic tissue available for analysis, we decided to combine often result in scar formation at the expense of tissue CTX injury, which individually was a poor fibrosis-inducing Pessina et al. Skeletal Muscle 2014, 4:7 Page 11 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 A N.I. CTX LAC DEN N.I. CTX LAC DEN C Collagen I CTGF TIMP-1 TGF 1 4 4 12 * 4 N.I. * CTX LAC DEN 2 2 6 0 0 Figure 5 Quantification of muscle fibrosis after chemical and surgical damage in wild-type mice. (A) Sirius red, hematoxylin and eosin (H&E), collagen I (green) and fibronectin (red) staining on wild-type (WT) tibialis anterior (TA) muscles two weeks after cardiotoxin (CTX) injury –5 (50 μlof10 M), laceration (LAC) and denervation (DEN) compared to non-injured (NI) muscle of sham-operated WT mice. (B) Quantification of collagen content in WT muscle after different injuries. Data correspond to the mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (C) Quantitative RT-PCR for collagen I, connective tissue growth factor (CTGF), tissue inhibitor of metalloproteinases 1 (TIMP-1) and transforming growth factor beta 1 (TGFβ1) mRNA in muscles after the different injuries (values are means ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI). Scale bar = 50 μm. method, with either DEN or co-injection of TGFβ1in TA muscles of WT mice were first subjected to CTX injec- –5 muscle of WT mice, methods which we previously showed tion (50 μlof10 M) and subsequently denervated or were able to increase fibrosis in young mdx muscle injected twice with TGFβ1(50 ng of TGFβ1in 50 μlPBS (Figure 4). Both DEN and injection of TGFβ1failedto per injection) (at day 7 and 10 after CTX injection) and induce fibrosis in WT muscles when used alone (Figure 5, muscles were collected two and four weeks later. We found and Figure S4C and D in Additional file 5). Accordingly, that the combination of CTX injury with DEN or TGFβ1 Fibronectin Collagen I H&E Sirius red Relative expression μg collagen/mg protein Pessina et al. Skeletal Muscle 2014, 4:7 Page 12 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 A CTX CTX+DEN CTX+TGFβ CTX CTX+DEN * CTX+TGF Collagen I C CTGF TIMP-1 TGF 1 CTX 5 3 20 5 CTX+DEN * CTX+TGF * 15 3 3 10 * 0 0 0 0 CTX CTX+ DEN CTX+TGFβ E N.I. CTX LAC CTX + DEN CTX + TGFβ F N.I. N.I. Tetanic force * CTX CTX LAC * LAC CTX+DEN * 200 * CTX+DEN CTX+TGF CTX+TGF Figure 6 Synergistic effect on fibrosis induction in muscle of wild-type mice by combined treatments. (A) Sirius red and hematoxylin and –5 eosin (H&E) staining of wild-type (WT) tibialis anterior (TA) muscles subjected to a combination of cardiotoxin (CTX) injury (50 μlof 10 M) and denerv- ation or transforming growth factor beta 1 (TGFβ1) (50 ng in 50 μl phosphate-buffered saline (PBS)) injection (as described in the Methods section), respectively, compared to CTX injury alone. (B) Quantification of collagen content in muscle after each treatment. Data correspond to the mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus CTX injury. (C) Quantitative RT-PCR for collagen I, connective tissue growth factor (CTGF), tissue inhibitor of metalloproteinases 1 (TIMP-1) and TGFβ1 after the different treatments (n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05, compared to CTX injury). (D) Representative immunostaining for collagen I (green) and fibronectin (red) on sections of WT muscle subjected to the different fibrosis-inducing methods. (E-G) Analysis of long-term fibrosis in WT muscle at one month after injury. Data are compared to non-injured (NI) muscle of sham-operated WT mice. (E) Sirius red and H&E staining of CTX-injured, lacerated, CTX/denervation and CTX/TGFβ-injured muscles at one month after injury. (F) Quantification of collagen content one month after injury of WT muscle. Values represent mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. (G) Ex vivo maximum isometric force (tetanic force) of TA muscle. Values as mean ± SEM; n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 versus NI. Scale bars = 50 μm. H&E Sirius red H&E Sirius red Fibronectin Collagen I Relative expression μg collagen/mg protein mN/mm μg collagen/mg protein Pessina et al. Skeletal Muscle 2014, 4:7 Page 13 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 delivery induced fibrosis significantly compared to CTX in- muscular degeneration, of which DMD is one of the se- jury alone, as shown by Sirius red staining (Figure 6A), col- verest. Progressive replacement of skeletal muscle by fat lagen quantification as well as expression of fibrotic and fibrotic tissue not only exacerbates disease progres- markers by quantitative RT-PCR and immunohistochemis- sion, but also impairs the efficiency of gene- and stem cell- try analyses, after 14 days (Figure 6B-D), correlating with based therapies [30,31]. Yet, there is no effective clinical delayed regeneration kinetics (Figure 6A). Of note, a com- treatment to reverse or attenuate fibrosis in DMD patients, bination of CTX injury and CTGF local overexpression except for promising new agents such as halofuginone produced similar profibrotic effects as combining CTX in- [32]. To a great extent, this deficit may derive from the jury and TGFβ1 delivery (Figure S3C in Additional file 4), poor understanding of the mechanisms underlying fibro- suggesting that part of the TGFβ profibrotic actions are genesis in muscular dystrophy. Indeed, chronic inflamma- likely to be mediated by CTGF. tion and production of collagens by myoblasts are among We next compared the persistence of fibrosis over the few reported causal factors promoting progression to time and the consequences on muscle function of each fibrosis in dystrophic muscle [33-38]. The largely unknown of the distinct fibrogenic regimes on WT muscle. Sirius etiology of fibrogenesis in DMD in turn may be principally red staining and collagen quantification showed that due to the lack of adequate animal models of muscle fibro- muscle fibrosis still persisted after four weeks of either sis. Here we report the application of simple models of laceration or CTX combined with TGFβ1 or DEN, com- tissue damage that are able to significantly enhance the pared to muscle injured with CTX alone or NI muscle fibrotic response in skeletal muscle and which may be (Figure 6E and 6F). The relevance of these results was useful for investigating therapeutic strategies for DMD. supported by functional studies of WT muscle after the Studies using mdx mice, the most common mouse combined profibrotic treatments (CTX combined with model of DMD, may not be translated directly to dys- TGFβ1 or DEN). Indeed, dual treatments on muscles trophic patients due to the mild phenotype they display. exerted a synergistic effect, resulting in increased fibrosis In particular, limb muscles of mdx mice show a relatively and reduced net force compared to uninjured muscle or efficient regeneration and no significantly aberrant de- muscle injured with CTX alone (Figure 6G). These re- position of ECM proteins until very old age. Progressive sults suggest that in WT mice, LAC, as well as a com- endomysial fibrosis only develops in diaphragm muscle, bination of CTX injury with either DEN or TGFβ1, but is still not significantly advanced until well into proved to be effective fibrosis-inducing models that trig- adulthood [39]. To try to accelerate or exacerbate this ger a rapid accumulation of fibrotic tissue that is sus- phenotype, other mouse models have been generated such tained for an extended period of time, with negative as mdx mice lacking arginase-2, PAI-1 (plasminogen acti- consequences on muscle function. vator inhibitor-1) or Cmah (cytidine monophosphate- Finally, and in order to further expand the variety of sialic acid hydroxylase) [11,40,41] and previously the mdx/ +/− fibrogenic-inducing procedures to the maximum number utrn mouse line (mdx mice with haploinsufficiency of +/− of laboratories working on skeletal muscle, we tested the utrophin) [42]. However, mdx/utrn mice show early fibrosis-inducing effect of a widely used muscle-damaging mortality and the manipulation of the line requires time method involving BaCl injection in WT muscle. We and resources in genotyping and breeding. Moreover, the found that, as for CTX injection, one intramuscular injec- genetics of these mouse models do not adequately reflect tion of BaCl (50 μlof0.2%BaCl ) only induced a very mild human DMD patients. Therefore, the need for fibrotic 2 2 and transient accumulation of ECM. Of note, repeated in- models that do not require waiting for the natural physio- jections (spaced one week) for up to six weeks resulted in logical onset of fibrosis in the hindlimb of old mice, and significant ECM accumulation after eight weeks from the that recapitulate the human DMD phenotype becomes in- first injection (that is two weeks after the last injection), creasingly more important. One recent attempt to address although no major change in muscle force was observed this problem came from Desguerre and colleagues (2012) (Figure 7A-C). Thus, repeated damaging with myotoxins who described a model of mechanical muscle injury by may be a fibrosis-inducing alternative in non-dystrophic daily repeated micro-punctures in mdx hindlimb muscle muscle, although development of fibrosis requires up to [43]. Induction of endomysial fibrosis in dystrophic muscle eight weeks, and involves weekly mouse manipulation for through this method is ascribed to a small fibrotic area six weeks, compared to the less labor-consuming and and requires daily animal manipulation during two weeks. more rapid fibrogenic effect (with additional impact on In addition, this procedure does not seem to induce fibro- muscle force) of the combined treatments. sis in WT mice [43]. The strategies we propose here are valid alternatives to Discussion both hasten the appearance and prolong the duration of Muscular dystrophies constitute a heterogeneous group fibrosis in hindlimb muscles of young mdx mice, with of inherited myopathies, characterized by progressive very limited (non-daily) animal manipulation, which Pessina et al. Skeletal Muscle 2014, 4:7 Page 14 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 A N.I. BaCl 1 round BaCl 6 rounds 2 2 Collagen I Fibronectin N.I. 3 2 BaCl 1 round BaCl 6 rounds 0 0 Tetanic force N.I. BaCl 1 round 300 BaCl 6 rounds Figure 7 Fibrosis induction in muscle of wild-type mice after repeated BaCl injuries. (A) Sirius red, H&E and fibronectin staining on wild-type (WT) tibialis anterior (TA) muscles subjected to one or six consecutive weekly rounds of BaCl injections (50 μlof 0.2% BaCl ), compared to non-injured 2 2 (NI) muscle of sham-operated WT mice, sampled two weeks after the final injection. (B) Quantitative expression of fibronectin and collagen I in the distinct muscle samples (values are means ± SEM; n = 4 to 5 on each group. Non-parametric Mann–Whitney U test; **P <0.05 versus NI). (C) Ex vivo maximum isometric force (tetanic force) of TA muscle. Values as mean ± SEM; n = 4 for each group; non-parametric Mann–Whitney U test; no significant differences P >0.05. Scale bar = 50 μm. Fibronectin H&E Sirius red Relative expression mN/mm Relative expression Pessina et al. Skeletal Muscle 2014, 4:7 Page 15 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 notably are also able to induce relatively sustained fibro- is relatively stable over long periods of time, despite af- sis in WT muscle. Therefore, these methods would be fecting only a localized small tissue area. However, the applicable to other genetically modified mice, and this combination of regimes showed an improved capacity to will help further delineating the cellular and genetic generate fibrosis in WT muscle for a sustained period of basis of muscle fibrosis. Exercise training of young mdx time, correlating with reduction in muscle force, indicat- mice induced endomysial fibrosis, resembling the pheno- ing that they mimic in WT animals pathophysiological type of old hindlimb dystrophic muscles; however, situations of severe muscle trauma that result in aber- although this method can be considered more physio- rant regeneration, scar deposition and functional impair- logical, it still requires a lengthy time period to obtain a ment. We propose that this variety of fibrosis-inducing fibrotic muscle tissue, in addition to significant amount of methodologies will enable fibrosis to be studied in a vast effort and time, since exercise protocols need to be applied array of transgenic mouse lines (with no apparent under- several times a week for ideally three months. At variance, lying muscle pathology) or after crossing them with dys- the methods based on muscle growth factor delivery and trophic strains such as mdx mice. surgical injuries that we present here offer a faster and less labor-intensive alternative. The rationale for the proposed Conclusions profibrotic growth factor-based methods relies on the ob- Collectively, through this study, we propose novel and/or servation that, in fibrotic muscles of human DMD patients optimized experimental strategies to accelerate, anticipate and old mdx mice, TGFβ1 (and its downstream target and boost muscle fibrosis in young dystrophic mice or to CTGF) is present at high levels [22,44], correlating with drive de novo fibrosis onset in WT mice. We think that the increased activation of Smad2/3 transcriptional me- our findings provide very useful methodologies that will fa- diators (see Figure S5 in Additional file 6). Of the surgi- cilitate research in the emerging field of skeletal muscle fi- cal methods tested, muscle laceration proved to be the brosis. In particular, these rapid and feasible procedures for most effective for inducing sustained fibrosis; however, most laboratories will help getting deeper insight into the this method has the disadvantage that the affected area mechanisms underlying muscle fibrosis, as well as develop- is relatively small (and only one muscle per mouse can ing therapeutic strategies aimed to reduce its magnitude in be lesioned due to the severity of the procedure) thereby dystrophic diseases and to ameliorate dystrophy progres- limiting the amount of material available for downstream sion. Since fibrosis is also a main obstacle for stem cell en- processing. Subsequent cellular analysis of fibrotic muscle graftment, availability of appropriate fibrosis models will be by techniques such as fluorescence-activated cell sorting a determinant factor in the research toward successful (FACS) may not be possible in this type of model without gene/cell therapy-based strategies in muscular dystrophy. vast improvements of sensitivity or without increasing the number of animals used, which has extra cost and ethical Additional files implications. Sciatic nerve denervation of mdx mice gen- erates increased collagen deposition, as a possible mech- Additional file 1: Table S1. Methods and sampling times of the anism to replace the tissue volume lost due to myofiber different fibrosis-inducing procedures in mdx and wild-type (WT) mice. atrophy. All of these fibrogenesis-inducing methods per- Additional file 2: Figure S1. Quantification of fibrosis in mdx diaphragm muscle. (A) Relative expression of collagen I, connective tissue sist with time, since at two months after injury muscles growth factor (CTGF), tissue inhibitor of metalloproteinases 1(TIMP-1) and still displays a fibrotic phenotype. Moreover, consistent transforming growth factor beta 1 (TGFβ1) mRNA by quantitative RT-PCR with the idea that fibrosis aggravates muscle dysfunction in mdx diaphragm muscles at the indicated ages respect to wild-type (WT) muscles. Values are mean ± SEM; n = 4 for each group; non- in DMD, we showed that maximal muscle force was also parametric Mann–Whitney U test; *P <0.05 versus age-matched WT. (B) reduced in young mdx mice after fibrosis induction Representative pictures of immunofluorescence staining for collagen I through the different protocols. (green) and fibronectin (red) in young, adult and old mdx diaphragm, compared to age-matched WT muscle. Scale bars = 50 μm. Finally, to be able to investigate fibrosis development Additional file 3: Figure S2. Effect of exercise on tibialis anterior (TA) and therapeutic options in muscle of non-dystrophic mdx muscle. (A) Sirius red and hematoxylin and eosin (H&E) staining of models, we sought to apply these methods to WT mice. TA muscle of mdx mice that were exercised three times weekly, for To date, studies on muscle damage in non-dystrophic 30 minutes at a speed of 12 meters per minute with a rest of 5 minutes each 10 minutes of exercise, for one, two and three months, compared models have been performed classically with a single in- to sections of muscle from unexercised mdx mice. All the samples were jection of myotoxins (for example, CTX or BaCl ). Des- collected when the animals were six months old (see the Methods pite the general use, we have shown in this study that section). (B) Representative immunofluorescence for collagen I and fibronectin in muscle sections of control and exercised mice as shown in these standard single-injury methods are not appropriate (A). (C) Biochemical quantification of collagen protein content in mdx TA fibrosis-inducing models, as the resolution of the dam- muscle after exercising for the indicated period as compared to unexercised age occurs rapidly and collagen deposition is very mild age-matched mdx mice. Data correspond to the mean ± SEM; n = 4 sedentary and 4 exercised mdx mice for each exercise time point. One-way analysis of and only transient. On the contrary, muscle laceration of variance with Tukey’s post hoc multiple comparison test; **P <0.01, ***P <0.001 WT muscle induces a massive collagen deposition that Pessina et al. Skeletal Muscle 2014, 4:7 Page 16 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 3. Serrano AL, Munoz-Canoves P: Regulation and dysregulation of fibrosis in versus control. (D) Ex vivo maximum isometric force (tetanic force) of TA skeletal muscle. Exp Cell Res 2010, 316:3050–3058. muscle of age-matched unexercised and three-month-trained mdx mice. 4. Yablonka-Reuveni Z, Anderson JE: Satellite cells from dystrophic (mdx) Values as mean ± SEM; n = 7 on each group. Non-parametric Mann–Whitney mice display accelerated differentiation in primary cultures and in U test; **P <0.01 versus non-exercised. Scale bars = 50 μm. isolated myofibers. Dev Dynamics 2006, 235:203–212. Additional file 4: Figure S3. Fibrosis induction in muscle by viral 5. Grounds MD, Shavlakadze T: Growing muscle has different sarcolemmal delivery of connective tissue growth factor (CTGF). (A) Mdx mice: Sirius properties from adult muscle: a proposal with scientific and clinical red and hematoxylin and eosin (H&E) staining of mdx tibialis anterior (TA) implications: reasons to reassess skeletal muscle molecular dynamics, muscles overexpressing mouse CTGF after intramuscular injection of cellular responses and suitability of experimental models of muscle 50 μl of 2x10 particles of adenovirus (AdV) in three-month-old mice. disorders. Bio Essays 2011, 33:458–468. (B) Collagen content quantification. Data correspond to the mean ± SEM; 6. Muntoni F: Cardiac complications of childhood myopathies. J Child Neurol n = 4 on each group. Non-parametric Mann–Whitney U test; *P <0.05 2003, 18:191–202. versus NI. (C) Wild-type (WT) mice: H&E of WT muscles after adenoviral 7. Benedetti S, Hoshiya H, Tedesco FS: Repair or replace? Exploiting novel CTGF delivery coupled with cardiotoxin (CTX) injury; representative gene and cell therapy strategies for muscular dystrophies. FEBS J 2013, immunostaining for collagen I (green) and fibronectin (red) on sections 280:4263–4280. of AdV-transduced muscle overexpressing CTGF. Scale bars = 50 μm. 8. Tedesco FS, Hoshiya H, D’Antona G, Gerli MF, Messina G, Antonini S, Additional file 5: Figure S4. Collagen deposition after cardiotoxin Tonlorenzi R, Benedetti S, Berghella L, Torrente Y, Kazuki Y, Bottinelli R, (CTX)-induced muscle injury and transforming growth factor beta 1 Oshimura M, Cossu G: Stem cell-mediated transfer of a human artificial (TGFβ1) delivery alone is quickly resolved in wild-type (WT) mice. (A) chromosome ameliorates muscular dystrophy. Sci Trans Med 2011, Sirius red, hematoxylin and eosin (H&E), collagen I (green) and fibronectin 3:96ra78. (red) staining on WT tibialis anterior (TA) muscles after five and eight days 9. Sicinski P, Geng Y, Ryder-Cook AS, Barnard EA, Darlison MG, Barnard PJ: The from CTX injury, compared to non-injured (NI) muscle of sham-operated molecular basis of muscular dystrophy in the mdx mouse: a point WT mice. (B) Quantification of collagen content in muscle after treatment. mutation. Science 1989, 244:1578–1580. Data correspond to the mean ± SEM, n = 4 on each group. Non-parametric 10. Carnwath JW, Shotton DM: Muscular dystrophy in the mdx mouse: Mann–Whitney U test; *P <0.05 versus NI. (C) Sirius red, H&E, collagen I (green) histopathology of the soleus and extensor digitorum longus muscles. and fibronectin (red) staining on WT TA muscle two weeks after two sequen- J Neuro Sci 1987, 80:39–54. tial treatments with recombinant TGFβ1 (50 ng in 50 μl phosphate-buffered 11. Ardite E, Perdiguero E, Vidal B, Gutarra S, Serrano AL, Munoz-Canoves P: PAI- saline (PBS)), spaced seven days apart. (D) Quantification of collagen content 1-regulated miR-21 defines a novel age-associated fibrogenic pathway in muscle after injection of TGFβ1 or PBS (vehicle). Data are mean ± SEM, n = 4 in muscular dystrophy. J Cell Biol 2012, 196:163–175. for each group. Non-parametric Mann–Whitney U test; no significant differ- 12. Menetrey J, Kasemkijwattana C, Fu FH, Moreland MS, Huard J: Suturing ences P >0.05. Scale bars = 50 μm. versus immobilization of a muscle laceration. A morphological and functional study in a mouse model. Am J Sports Med 1999, Additional file 6: Figure S5. Smad2/3 protein phosphorylation in 27:222–229. injured muscles. Immunofluorescence for phosphorylated-Smad2/3 proteins 13. Serrano AL, Murgia M, Pallafacchina G, Calabria E, Coniglio P, Lomo T, on sections from tibialis anterior (TA) muscle of mdx (A) and wild-type (WT) Schiaffino S: Calcineurin controls nerve activity-dependent specification (B) mice after the indicated treatments. Scale bars = 50 μm. of slow skeletal muscle fibers but not muscle growth. Proc Natl Acad Sci USA 2001, 98:13108–13113. Abbreviations 14. Glass DJ: Skeletal muscle hypertrophy and atrophy signaling pathways. BaCl : barium chloride; Coll I: collagen I; CTGF: connective tissue growth Int J Biochem Cell Biol 2005, 37:1974–1984. factor; CTX: cardiotoxin; DEN: denervation; DMD: Duchenne muscular 15. Morales MG, Cabello-Verrugio C, Santander C, Cabrera D, Goldschmeding R, dystrophy; ECM: extracellular matrix; FN: fibronectin; H&E: hematoxylin and Brandan E: CTGF/CCN-2 over-expression can directly induce features of eosin; LAC: laceration; NI: non-injured; PBS: phosphate-buffered saline; P-Smad2/ skeletal muscle dystrophy. J Pathol 2011, 225:490–501. 3: phosphorylated Smad2/3; TA: tibialis anterior; TGFβ1: transforming growth 16. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, factor beta1; TIMP-1: tissue inhibitor of metalloproteinases 1; WT: wild-type. Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A: Fiji: an open-source Competing interests platform for biological-image analysis. Nat Methods 2012, 9:676–682. The authors declare that they have no competing interests. 17. Cabello-Verrugio C, Morales MG, Cabrera D, Vio CP, Brandan E: Angiotensin II receptor type 1 blockade decreases CTGF/CCN2-mediated damage Authors’ contributions and fibrosis in normal and dystrophic skeletal muscles. J Cellular Mol Med PMC, EB and ALS conceived and designed the project. PP performed and 2012, 16:752–764. analyzed most of the experiments and was assisted by DC, MGM, CAR and 18. Biernacka A, Frangogiannis NG: Aging and cardiac fibrosis. Aging Dis 2011, JG for Figures 3 and 7. PMC and PP wrote the manuscript and ALS and EB 2:158–173. revised and edited it. All authors read and approved the final manuscript. 19. Brandan E, Gutierrez J: Role of proteoglycans in the regulation of the skeletal muscle fibrotic response. FEBS J 2013, 280:4109–4117. Acknowledgements 20. MacDonald EM, Cohn RD: TGFbeta signaling: its role in fibrosis formation We are indebted to E. Perdiguero, V. Lukesova, L. Correa, A. Vasquez, C. Mann and myopathies. Cur Opinion Rheumatol 2012, 24:628–634. and M. Raya for their continuous help and advice. We also thank previous 21. Mann CJ, Perdiguero E, Kharraz Y, Aguilar S, Pessina P, Serrano AL, members of our laboratories, especially E. Ardite and B. Vidal, for setting up Munoz-Canoves P: Aberrant repair and fibrosis development in skeletal the basis of this study, and J. Martín-Caballero for assistance in the PRBB muscle. Skelet Muscle 2011, 1:21. animal facility. The authors acknowledge funding from MINECO-Spain 22. Morales MG, Cabrera D, Cespedes C, Vio CP, Vazquez Y, Brandan E, (SAF2012-38547, FIS-PS09/01267, FIS-PI13/02512, PLE2009-0124), AFM, E-Rare, Cabello-Verrugio C: Inhibition of the angiotensin-converting enzyme Fundació-MaratóTV3, Duchenne PP-NL, EU-FP7 (Myoage, Optistem and decreases skeletal muscle fibrosis in dystrophic mice by a diminution in Endostem), MDA, CARE PFB12/2007 and FONDECYT 1110426. the expression and activity of connective tissue growth factor (CTGF/ CCN-2). Cell Tissue Res 2013, 353:173–187. Received: 29 October 2013 Accepted: 20 January 2014 23. Dangain J, Vrbova G: Muscle development in mdx mutant mice. Muscle Published: 25 August 2014 Nerve 1984, 7:700–704. 24. De Luca A, Pierno S, Liantonio A, Cetrone M, Camerino C, Fraysse B, References Mirabella M, Servidei S, Ruegg UT, Conte Camerino D: Enhanced 1. Emery AE: The muscular dystrophies. Lancet 2002, 359:687–695. dystrophic progression in mdx mice by exercise and beneficial effects 2. Briggs D, Morgan JE: Recent progress in satellite cell/myoblast of taurine and insulin-like growth factor-1. J Pharm Exp Thera 2003, engraftment - relevance for therapy. FEBS J 2013, 280:4281–4293. 304:453–463. Pessina et al. Skeletal Muscle 2014, 4:7 Page 17 of 17 http://www.skeletalmusclejournal.com/content/4/1/7 25. Morales MG, Gutierrez J, Cabello-Verrugio C, Cabrera D, Lipson KE, 43. Desguerre I, Arnold L, Vignaud A, Cuvellier S, Yacoub-Youssef H, Gherardi Goldschmeding R, Brandan E: Reducing CTGF/CCN2 slows down mdx RK, Chelly J, Chretien F, Mounier R, Ferry A, Chazaud B: A new model of muscle dystrophy and improves cell therapy. Hum Mol Gen 2013, experimental fibrosis in hindlimb skeletal muscle of adult mdx mouse 22:4938–4951. mimicking muscular dystrophy. Muscle Nerve 2012, 45:803–814. 26. Perdiguero E, Sousa-Victor P, Ruiz-Bonilla V, Jardi M, Caelles C, Serrano AL, 44. Bernasconi P, Di Blasi C, Mora M, Morandi L, Galbiati S, Confalonieri P, Munoz-Canoves P: p38/MKP-1-regulated AKT coordinates macrophage Cornelio F, Mantegazza R: Transforming growth factor-beta1 and fibrosis transitions and resolution of inflammation during tissue repair. J Cell Biol in congenital muscular dystrophies. Neuromusc Dis 1999, 9:28–33. 2011, 195:307–322. 27. Suelves M, Lopez-Alemany R, Lluis F, Aniorte G, Serrano E, Parra M, Carmeliet P, doi:10.1186/2044-5040-4-7 Munoz-Canoves P: Plasmin activity is required for myogenesis in vitro and Cite this article as: Pessina et al.: Novel and optimized strategies for skeletal muscle regeneration in vivo. Blood 2002, 99:2835–2844. inducing fibrosis in vivo: focus on Duchenne Muscular Dystrophy. Skeletal Muscle 2014 4:7. 28. Suelves M, Vidal B, Serrano AL, Tjwa M, Roma J, Lopez-Alemany R, Luttun A, de Lagran MM, Diaz-Ramos A, Jardi M, Roig M, Dierssen M, Dewerchin M, Carmeliet P, Muñoz-Cánoves P: uPA deficiency exacerbates muscular dystrophy in MDX mice. J Cell Biol 2007, 178:1039–1051. 29. Carlson BM, Billington L, Faulkner J: Studies on the regenerative recovery of long-term denervated muscle in rats. Rest Neurol Neurosci 1996, 10:77–84. 30. Gargioli C, Coletta M, De Grandis F, Cannata SM, Cossu G: PlGF-MMP-9- expressing cells restore microcirculation and efficacy of cell therapy in aged dystrophic muscle. Nat Med 2008, 14:973–978. 31. Muir LA, Chamberlain JS: Emerging strategies for cell and gene therapy of the muscular dystrophies. Expert Rev Mol Med 2009, 11:e18. 32. Turgeman T, Hagai Y, Huebner K, Jassal DS, Anderson JE, Genin O, Nagler A, Halevy O, Pines M: Prevention of muscle fibrosis and improvement in muscle performance in the mdx mouse by halofuginone. Neuro Dis 2008, 18:857–868. 33. Alexakis C, Partridge T, Bou-Gharios G: Implication of the satellite cell in dystrophic muscle fibrosis: a self-perpetuating mechanism of collagen overproduction. Am J Physiol Cell Physiol 2007, 293:C661–C669. 34. Morrison J, Palmer DB, Cobbold S, Partridge T, Bou-Gharios G: Effects of T-lymphocyte depletion on muscle fibrosis in the mdx mouse. Am J Pathol 2005, 166:1701–1710. 35. Vidal B, Serrano AL, Tjwa M, Suelves M, Ardite E, De Mori R, Baeza-Raja B, Martinez de Lagran M, Lafuste P, Ruiz-Bonilla V, Jardí M, Gherardi R, Christov C, DierssenM,Carmeliet P, DegenJL, Dewerchin M, Muñoz-Cánoves P: Fibrinogen drives dystrophic muscle fibrosis via a TGFbeta/alternative macrophage activation pathway. Genes Dev 2008, 22:1747–1752. 36. Vidal B, Ardite E, Suelves M, Ruiz-Bonilla V, Janue A, Flick MJ, Degen JL, Serrano AL, Munoz-Canoves P: Amelioration of Duchenne muscular dystrophy in mdx mice by elimination of matrix-associated fibrin-driven inflammation coupled to the alphaMbeta2 leukocyte integrin receptor. Hum Mol Gen 2012, 21:1989–2004. 37. Villalta SA, Nguyen HX, Deng B, Gotoh T, Tidball JG: Shifts in macrophage phenotypes and macrophage competition for arginine metabolism affect the severity of muscle pathology in muscular dystrophy. Hum Mol Gen 2009, 18:482–496. 38. Kharraz Y, Guerra J, Mann CJ, Serrano AL, Munoz-Canoves P: Macrophage plasticity and the role of inflammation in skeletal muscle repair. Mediators Inflam 2013, 2013:491497. 39. Stedman HH, Sweeney HL, Shrager JB, Maguire HC, Panettieri RA, Petrof B, Narusawa M, Leferovich JM, Sladky JT, Kelly AM: The mdx mouse diaphragm reproduces the degenerative changes of Duchenne muscular dystrophy. Nature 1991, 352:536–539. 40. Chandrasekharan K, Yoon JH, Xu Y, deVries S, Camboni M, Janssen PM, Varki A, Martin PT: A human-specific deletion in mouse Cmah increases disease severity in the mdx model of Duchenne muscular dystrophy. Sci Trans Med 2010, 2:42ra54. 41. Wehling-Henricks M, Jordan MC, Gotoh T, Grody WW, Roos KP, Tidball JG: Submit your next manuscript to BioMed Central Arginine metabolism by macrophages promotes cardiac and muscle and take full advantage of: fibrosis in mdx muscular dystrophy. PloS one 2010, 5:e10763. 42. Zhou L, Rafael-Fortney JA, Huang P, Zhao XS, Cheng G, Zhou X, Kaminski • Convenient online submission HJ, Liu L, Ransohoff RM: Haploinsufficiency of utrophin gene worsens • Thorough peer review skeletal muscle inflammation and fibrosis in mdx mice. JNeurol Sci 2008, 264:106–111. • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit

Journal

Skeletal MuscleSpringer Journals

Published: Aug 25, 2014

References