Nintedanib decreases muscle fibrosis and improves muscle function in a murine model of dystrophinopathy

Piñol-Jurado et al. Cell Death and Disease (2018) 9:776 DOI 10.1038/s41419-018-0792-6 Cell Death & Disease ARTICLE Open Access Nintedanib decreases muscle fibrosis and improves muscle function in a murine model of dystrophinopathy Patricia Piñol-Jurado 1 , Xavier Suárez-Calvet 1,2 , Esther Fernández-Simón 1 , Eduard Gallardo 1,2 , Natalia de la Oliva 3,4 , Anna Martínez-Muriana 3,4 , Pedro Gómez-Gálvez 5 ,LuisM.Escudero 5 , María Pérez-Peiró 1 , Lutz Wollin 6 ,NoemideLuna 1,2 , Xavier Navarro 3,4 , Isabel Illa 1,2 and Jordi Díaz-Manera 1,2 Abstract Duchenne muscle dystrophy (DMD) is a genetic disorder characterized by progressive skeletal muscle weakness. Dystrophin deficiency induces instability of the sarcolemma during muscle contraction that leads to muscle necrosis and replacement of muscle by fibro-adipose tissue. Several therapies have been developed to counteract the fibrotic process. We report the effects of nintedanib, a tyrosine kinase inhibitor, in the mdx murine model of DMD. Nintedanib reduced proliferation and migration of human fibroblasts in vitro and decreased the expression of fibrotic genes such as COL1A1,COL3A1,FN1,TGFB1, and PDGFA. We treated seven mdx mice with 60 mg/kg/day nintedanib for 1 month. Electrophysiological studies showed an increase in the amplitude of the motor action potentials and an improvement of the morphology of motor unit potentials in the animals treated. Histological studies demonstrated a significant reduction of the fibrotic areas present in the skeletal muscles. Analysis of mRNA expression from muscles of treated mice showed a reduction in Col1a1,Col3a1,Tgfb1, and Pdgfa. Western blot showed a reduction in the expression of collagen I in skeletal muscles. In conclusion, nintedanib reduced the fibrotic process in a murine model of dystrophinopathy after 1 month of treatment, suggesting its potential use as a therapeutic drug in DMD patients. Introduction Duchenne muscle dystrophy (DMD) is a genetic disorder produced by mutations in the dystrophin gene. DMD patients develop muscle weakness that usually starts at an age of 5–7 years and progresses quickly. At 14 years of age, most of the patients have lost ambulation. At 20 years, patients are completely dependent on care givers or relatives owing to severe muscle weakness. Respiratory muscle failure and/or myocardiopathy are the main causes of mortality in these patients 1,2 . Several therapeutic strategies have been tested in murine models of DMD, and some of them have also been studied in clinical trials in patients 3 . Cell therapy using different stem cells, such as myoblasts or mesoangioblasts, have been tested both in animals and in patients 4 . Although results of animal experiments have shown promising results, tests in humans have not lead to any functional change in the patients treated 5,6 . Gene-based strategies, such as exon-skipping or readthrough of null mutations have demonstrated to restore the expression of dystrophin in muscles fibers both in mice and in patients 7,8 . Ataluren is at present commercialized in Europe to treat patients with non-sense mutations, and eteplirsen is being commercialized in USA to treat DMD patients with skipping of the exon 51 9,10 . CRISPR-Cas9 is a promising genetic strategy, that could be useful in © The Author(s) 2018 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a linktotheCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Correspondence: Jordi Díaz-Manera ([email protected]) 1 Neuromuscular disorders Unit, Neurology department, Universitat Autònoma de Barcelona, Hospital de la Santa Creu I Sant Pau, Barcelona, Spain 2 Centro de Investigación Básica en Red en Enfermedades Raras (CIBERER), Madrid, Spain Full list of author information is available at the end of the article. These authors contributed equally: Patricia Piñol-Jurado, Xavier Suárez-Calvet. Edited by M. Agostini Official journal of the Cell Death Differentiation Association 1234567890():,; 1234567890():,; 1234567890():,; 1234567890():,; patients although it has only been tested in animal models and cell cultures so far 11,12 . Drugs interfering the process of muscle degeneration are another potential strategy to treat muscle dystrophies. The process of muscle degeneration in DMD patients has been thoroughly studied. The absence of dystrophin weakens muscle membrane, leading to contraction-induced muscle fiber damage and death 13 . Muscle fiber loss is associated to expansion of fibro-adipose tissue producing muscle weakness 14 . Several cytokines and growth factors have been related to skeletal muscle fibrosis, although it has been reported that transforming growth factor β(TGF-β) is the most important factor in this process 15–17 . Several therapeutic strategies trying to decrease TGF-βactivity have been developed, which have been shown to reduce fibrous tissue but also to increase inflammatory infiltration 18,19 . These results have promoted the investigation of other growth factors involved in muscle fibrosis. Platelet-derived growth factors (PDGFs) are associated with multiple cellular processes such as proliferation, migration, and cell differentiation 20 . PDGF have been implicated in a broad range of diseases such as cancer, atherosclerosis and fibrosis. Several evidences supporting a role on muscle fibrosis of PDGF-AA have been published justifying therapeutic interventions targeting the PDGF signaling cascade 21,22 . Nintedanib is a tyrosine kinase inhibitor (TKI) also targeting fibroblast growth factor receptor (FGFR) 1 and 2, PDGF receptors α and βand vascular endothelial growth factor receptor (VEGFR) 23 . Nintedanib is approved for the treatment of patients with idiopathic pulmonary fibrosis (IPF), a condition in which expansion of the fibrotic tissue is crucial 24,25 . The anti-fibrotic activity of nintedanib has been confirmed in primary lung fibroblasts from patients with IPF and in dermal fibroblasts from patients with systemic sclerosis 25,26 . In this study, we explored the effect of nintedanib on human fibroblasts obtained from muscle biopsies and on muscle fibrosis and function in the mdx mouse model of DMD. Results Nintedanib reduced proliferation, migration, and mRNA expression of fibrotic markers in human fibroblasts In muscle dystrophies, activated fibroblasts proliferate and express high levels of extracellular proteins leading to the expansion of fibrotic tissue 27,28 . We analyzed whether nintedanib was able to decrease fibroblast proliferation in vitro. Nintedanib significantly reduced fibroblast proliferation in a dose-dependent manner reaching the highest effect using a concentration of 0.4 μM (Fig. 1a) without any cytotoxic effect measured by cell viability and apoptosis assays (Fig. S1A, B). Accordingly, the expression of the gene PT53 (tumor protein p53) and the number of Ki-67+cells, which are well known cell cycle markers, were significantly lower in nintedanibtreated cells than in control non-treated cells (Fig. S1C, D). To study which growth factor signaling cascade targeted by nintedanib affects more prominently fibroblasts proliferation, we cultured human fibroblasts in the presence of recombinant PDGF-AA, recombinant human basic fibroblast growth factor (bFGF), or recombinant human vascular endothelial growth factor (VEGF)-A, with or without nintedanib (Fig. 1b–d). In order to avoid possible interferences with other growth factors present in the fetal bovine serum (FBS), we cultured the cells with 1% FBS only. PDGF-AA significantly increased fibroblast proliferation at days 2, 4, and 6 (Fig. 1b) but not bFGF or VEGF (Fig. 1c, d). This increase was completely blocked by the addition of nintedanib at 0.4 μM (Fig. 1b). After muscle damage, activated fibroblasts gain migratory capabilities that allow them to move to the site of injury 29,30 . Using a scratch assay, we observed that nintedanib blocked cell migration promoted by PDGF-AA, bFGF, VEGF, and collagen tissue growth factor (CTGF) (Fig. 1e). The highest effect of nintedanib on cell migration was observed when PDGF-AA was added to the culture (Fig. 1e, f) and it was dose dependent (Fig. S2A, B). Transwell experiments confirmed these results showing that nintedanib reduced the effect of PDGF-AA in migration (Fig. S2C). Furthermore, nintedanib reverted significantly the effect of PDGF-AA on expression of the genes related with cell migration such as ADAM-17 (ADAM metallopeptidase domain 17), TIMP-1 (tissue inhibitor of metalloproteinase 1), and TIMP-2 (tissue inhibitor of metalloproteinase 2) (Fig. 1g). No significant differences were found in ADAM-12 (ADAM metallopeptidase domain 12), MMP1 (metalloproteinase 1), MMP2 (metalloproteinase 2), and MMP9 (metalloproteinase 9) (Fig. 1g). To know whether nintedanib treatment modified mRNA expression of skeletal muscle fibroblasts in vitro we analyzed the expression of genes related with muscle fibrosis using Real-Time PCR after 4 days of culture. We observed that nintedanib at 0.4 μM reduced significantly the expression of COL1A1 (Collagen type I alpha 1 chain), COL3A1 (collagen type III alpha 1 chain), FN1 (Fibronectin 1), CTGF (connective tissue growth factor), PDGFA (platelet-derived growth factor A), and TGFB1 (transforming growth factor beta 1) (Fig. 1h). In contrast, we did not observe significant differences in the expression of ACTA-2 (actin, alpha 2, smooth muscle, aorta) or PDGF-B (platelet-derived growth factor B). Nintedanib had no effect on myoblast differentiation to myotubes in vitro To analyze whether nintedanib could interfere with myoblast differentiation we cultured confluent human myoblasts in differentiation medium with increasing Piñol-Jurado et al. Cell Death and Disease (2018) 9:776 Page 2 of 15 Official journal of the Cell Death Differentiation Association doses of nintedanib for 7 days. No differences in differentiation index between control myotubes and myotubes treated with nintedanib were observed (Fig. S3A). In addition, we did not find differences in the differentiation index of myoblasts cultured in a medium containing PDGF-AA, b-FGF, or VEGF-A with or without nintedanib (Fig. S3B, C). We also studied whether myotubes treated with nintedanib were more resistant to chemical damage using 0.25 mM sodium dodecyl sulfate (SDS), and we did not see an effect of nintedanib on sarcolemma repair (Figure S3D). Nintedanib treatment improves electrophysiological tests in mdx mice We treated 10 months old mdx mice (n=7) with nintedanib (from now on nintedanib-treated mdx mice) at a Fig. 1 Effect of nintedanib on fibroblasts from healthy human muscle in vitro. Fibroblasts proliferation was analyzed after 2, 4, and 6 days of culture (a–d). We cultured fibroblast with increasing concentrations of nintedanib in a culture containing 10% fetal bovine serum (FBS) and we observed decreased fibroblast proliferation after 4 and 6 days in a dose-dependent manner (a). Fibroblasts were cultured in the presence of platelet-derived growth factor AA (PDGF-AA, 10 ng/mL) (b), basic fibroblast growth factor (FGFb, 10 ng/mL),(c) or vascular endothelial growth factor A (VEGF-A, 50 ng/mL) (d) with 1% FBS with or without nintedanib 0.4 μM. Only PDGF-AA increased cell proliferation, an effect that was reversed with the addition of nintedanib to the culture. Fibroblast migration was analyzed using a scratch assay. Nintedanid treatment at 0.4 μM reverted the promigratory effect of PDGF-AA, FGFb, VEGF, and CTGF (e). Representative images of this assay are shown in (f). Nintedanib treatment at 0.4 μM (blue bars) reverted the effect of PDGF-AA in the expression of ADAM metallopeptidase domain 17 (ADAM-17), tissue inhibitor of metalloproteinase 1 (TIMP-1) and tissue inhibitor of metalloproteinase 2 (TIMP-2) (g). Nintedanib treatment at 0.4 μM (blue bars) produced a statistically significant reduction of collagen type I alpha 1 chain (COL1A1), collagen type III alpha 1 chain (COL3A1),fibronectin 1 (FN1), platelet-derived growth factor A (PDGFA), connective tissue growth factor (CTGF), transforming growth factor beta 1 (TGFβ1) expression compared with control samples (black bars) analyzed by qPCR (h). Data are expressed as means ± SD. N=3 per group. *P< 0.05, **P< 0.01, ***P< 0.005 Piñol-Jurado et al. Cell Death and Disease (2018) 9:776 Page 3 of 15 Official journal of the Cell Death Differentiation Association dose of 60 mg/kg/day orally for 1 month and compared muscle function tests with 10 months old non-treated mdx mice (from now on mdx mice; n=5) and 10 months old C57 wild-type mice (from now on WT mice; n=5). We first performed a DigiGait analysis, which allows identifying differences in the walking pattern of healthy and dystrophic mice. We did not find differences between WT mice, mdx mice, and nintedanib-treated mdx mice, suggesting that the severity of weakness involving muscles of the paws in this murine model at this age is not enough to find differences using this technique (Fig S4). We then used electrophysiological tests to analyze compound muscle action potential (CMAP) amplitude and morphology of the motor unit action potentials (MUAPs). We detected a statistically significant increase of the CMAP of the peroneus nerve registered at the tibialis anterior muscle in nintedanib-treated mdx mice compared with mdx mice (Fig. 2a). Moreover, we also found a non-significant tendency to an increase of CMAPs recorded in gastrocnemius (GM) and plantar interossei muscles (Fig. 2b, c). We did not identify changes in latency or conduction velocity. Electromyographic studies showed a significant change in the morphology of the MUAPs in the tibialis anterior muscle. WT muscles contained a similar proportion of small and medium MUAPs, whereas mdx mice had predominantly small MUAPs. The analysis of nintedanib-treated mdx mice revealed a significant decrease in the proportion of small MUAPs and an increase in the proportion of medium and large MUAPs, which was statistically significant compared with mdx mice (Fig. 2d–g). Nintedanib treatment reduces muscle fibrosis in mdx mice The analysis of muscle histology after 1 month of treatment revealed significant differences between nintedanib-treated and non-treated mdx mice without any effect on apoptosis in the muscle of the animals (Fig. S5). We analyzed the diaphragm, which has been reported to be the most affected muscle in mdx mice 31 , the quadriceps and the tibialis anterior of WT, mdx mice, Fig. 2 Motor nerve conduction tests and electromyographic analysis of WT, mdx, and nintedanib-treated mdx mice after 4 weeks of treatment. a–cMotor nerve conduction test showed that nintedanib administration preserved the compound muscle action potential (CMAP) i`n the tibialis anterior (a), gastrocnemius (b), and plantar (c) muscles. d,eElectromyography (EMG) of tibialis anterior. AUC quantification of electromyography (EMG) responses of the TA muscle after application of Von Frey filaments 5.47 to the ipsilateral hind paw (d). Percentage of MUAPs from each type that were activated (e) and EMG showed a lower proportion of small pathologic motor unit action potentials (MUAPs) in nintedanibtreated mdx mice. f–hRepresentative recordings of EMG bursts after nociceptive stimuli in WT (f), mdx (g), and nintedanib-treated mdx mice (h). Note that mdx animals showed lower amplitude and shorter duration MUAPs compared with WT and nintedanib-treated mdx mice. Data are expressed as means ± SD. Genetic background mouse strain C57BL (WT); n=5, mdx mice (mdx), n=5; nintedanib-treated mdx mice (mdx +Ninte.), n=7. Data are expressed as means ± SD. *P< 0.05, **P< 0.01, **** p< 0.0001 Piñol-Jurado et al. Cell Death and Disease (2018) 9:776 Page 4 of 15 Official journal of the Cell Death Differentiation Association and nintedanib-treated mdx mice. Hematoxilin–eosin staining showed no differences either on the size of the muscle fibers or in the proportion of central nucleated fibers between mdx mice and nintedanib-treated mdx mice (Fig. 3). As nintedanib blocks the VEGF receptor, we also analyzed muscle vascularization. Although we did not find differences in the ratio of CD31-positive vessels per fiber among the three groups of mice, nintedanib treatment tended to reverse the increase observed in the ratio vessels/muscle fibers observed in the mdx mice (Fig.S6). To investigate in vivo the effect of nintedanib in the muscle, we quantified the number of eMHC- positive muscle fibers, which is an early marker of muscle regeneration. We observed a decrease in the number of muscle fibers expressing eMHC, in nintedanib-treated mdx mice compared with mdx mice (Fig. 4). We then analyzed the area of fibrotic tissue present in the skeletal muscles by quantifying collagen VI expression (Fig. 5). We observed a statistically significant reduction in the collagen VI area in the diaphragm (−5.6%), the quadriceps (−8.4%) and in the tibialis anterior muscles (−4.1%) of nintedanib-treated mdx mice compared with mdx mice (Fig. 5). The mRNA expression of the extracellular matrix proteins Col1a1, Col3a1, and Fn1, of the growth factors Pdgfa,Pdgfb,Ctgf, and Tgfb1, and of Adgre1 (Adhesion G protein-coupled receptor E1) encoding F4/80, a protein expressed by murine macrophage populations, were significantly increased or showed a tendency to increase in mdx mice compared with WT mice (Fig. 6). The mRNA expression of Col1a1, Col3a, Tgfb1, Pdgfa, and Pdgfb in quadriceps (Fig. 6d, e), diaphragm (Fig. 6l, m), and tibialis anterior muscles (Fig. 6t, u) of nintedanib-treated mdx mice was reduced compared with mdx mice. Adgre1 mRNA expression was decreased by nintedanib in tibialis anterior and quadriceps muscles (Fig. 6o, r, w, y) from nintedanib-treated mdx mice compared with mdx mice. Surprisingly, nintedanib significantly increased Ctgf in the tibialis anterior muscles of mdx mice (Fig. 6v). A similar trend was also detected in the other muscles. We confirmed the reduced expression of Col1a1 and Col3a1 by WB, with a significant decrease of collagen-1 in the tibialis anterior (p< 0.05) and a trend in the diaphragm (p=0.08) of nintedanib-treated mdx mice compared with mdx mice (Fig. 7). Effect of nintedanib in heart fibrosis of the mdx mouse We quantified the collagen VI and collagen III content in heart sections of WT, mdx mice, and nintedanibtreated mdx mice. We did not observe a significant difference in the area positive for collagen VI between WT and non-treated mdx mice (Fig. S7A–D). In contrast, collagen III was increased in mdx mice but it was not Fig. 3 Histological analysis of muscle samples of WT, mdx and nintedanib-treated mdx mice. Hematoxylin and eosin (H&E) staining of muscle samples obtained (a–i). Representative examples of cross-sectional muscle fibers within quadriceps (a–c), diaphragm (d–f), and tibialis anterior (g–i) of WT, mdx and nintedanib-treated mdx mice. Schematic representation of the fiber mean area in quadriceps (j), diaphragm (k), and tibialis anterior (l) showed no differences on the size of the muscle fibers. Analysis of the proportion of central nucleated fibers in quadriceps (m), diaphragm (n), and tibialis anterior (o) did not present significant differences after 1 month of treatment with nintedanib (60 mg/kg). Data are expressed as means ± SD. Genetic background mouse strain C57BL (WT); n=5, mdx mice (mdx), n=5; nintedanib-treated mdx mice (mdx +Ninte), n=7. Data are expressed as means ± SD. *P< 0.05, **P< 0.01 and ***P< 0.001. Scale bar =100 µm Piñol-Jurado et al. Cell Death and Disease (2018) 9:776 Page 5 of 15 Official journal of the Cell Death Differentiation Association modified by nintedanib (Fig. S7E–H). However, gene expression showed a significant decrease in the Col3a1 mRNA expression in nintedanib-treated mdx mice compared with non-treated mdx (Fig. S7J). mRNA expression of the Col1a1, Fn1,Pdgfa,Pdgfb,Ctgf,Tgfb1, and of Adgre1 was also studied. We only observed a decrease in the expression of the Adgre1 gene after treatment with nintedanib (Fig. S7I–P). Discussion In the present study, we show that nintedanib reduces skeletal muscle fibrosis of 10 months old mdx mice after 4 weeks of treatment. The decrease in fibrotic tissue was associated with an improvement of muscle function evidenced by electrophysiological tests. In vitro experiments showed that nintedanib decreased fibroblasts proliferation and migration and reduced expression of fibrotic genes, supporting an anti-fibrotic effect of the drug. Pharmacological approaches to treat muscle dystrophies aim to target the complex mechanism of fiber necrosis and fibrotic and fatty tissue deposition by restoring the proper expression of the mutated gene or by interfering with the pathological fibrotic cascade triggered by the lack of specific proteins 32 . Advances in the understanding of the fibrotic process have been recently achieved 33,34 . Continuous muscle fiber damage is associated with chronic infiltration of skeletal muscles by inflammatory cells, mainly macrophages. These cells release cytokines that contribute to the fibrotic process by activating fibro-adipogenic progenitor cells and fibroblasts resident in the skeletal muscle 22,35 . Among these cytokines, the growth factors of the PDGF family may have an important role. A series of evidences from the literature suggest that PDGF-AA has a profibrotic role in muscle dystrophies 35–39 . Treatment of mdx mice with imatinib, a TKI inhibiting v-Abl, c-kit, and PDGFRα, decreases muscle fibrosis 40 . The role of PDGF-BB is not so clearly established. Whereas it seems that PDGF-BB could have a role in the fibrosis observed in a number of tissues, several authors suggest that it can also play an important role in muscle regeneration in vivo 41,42 . Nintedanib was shown to inhibit fibroblast proliferation and migration in vitro and to exert anti-fibrotic activity in different animal models of lung fibrosis 25,26 .We observed that muscle fibrosis was reduced in mdx mice after 4 weeks of treatment with nintedanib. This effect is probably explained by a decreased rate of fibroblasts proliferation, migration and by a reduction of the expression of different components of the extracellular matrix. Skeletal muscle fibrosis in DMD is caused by accumulation of different types of collagens, such as Fig. 4 Nintedanib reduced the number of regenerative muscle fibers in mdx mice. Immunostaining of eMHC and quantification of positive fibers in muscle samples of WT, mdxand nintedanib-treated mdx mice (a-l). The number of eMHC-positive fibers (green) is reduced in quadriceps, diaphragm and tibialis anterior in mdx mice treated with nintedanib. Background in red is shown to localize the positivity within the muscle. Data are expressed as means ± SD. *P< 0.05 and **P< 0.01. Scale bar =100 µm Piñol-Jurado et al. Cell Death and Disease (2018) 9:776 Page 6 of 15 Official journal of the Cell Death Differentiation Association type I, III, and VI and also by fibronectin. We focused on collagen VI and III because they are the major components of the extracellular matrix and their levels correlates with muscle destruction 31,36,43,44 . We have observed a reduction in the expression of these extracellular matrix components both in vivo and in vitro. Moreover, we also observed a decrease in the expression of profibrotic growth factors such as PDGF-AA or TGF-β, which can certainly contribute to an attenuation of the fibrotic process in the treated mice. However, we observed an increase in CTGF levels in muscles of nintedanibtreated mdx mice, which could be explained by an enhanced fibrotic pathway not inhibited by our drug. This effect suggests that treatment with a combination of drugs blocking different pathways could have a more powerful anti-fibrotic effect. Using cell culture conditions with only 1% FBS, fibroblast proliferation was stimulated most prominently by PDGF-AA compared with VEGF or bFGF, which had no effect in those culture conditions. Hence, drugs targeting PDGFRα, the main receptor activated by PDGF-AA may be useful in the treatment of muscle dystrophies. The preclinical investigation of new drugs to treat DMD is hampered by the lack of an animal model, which mimics the severity of muscle degeneration and natural history of the disease in patients. In the mdx mouse model we used, it has been shown that significant changes in muscles cannot be observed until 8–10 months of age 45 . Moreover, despite the presence of clear myopathic changes, mice do not present functional impairment until later in their life. Similar findings would apply to heart involvement. This might explain the absence of differences that we have observed in the Digigait test when comparing healthy C57 mice and non-treated mdx mice. Electrophysiology is an objective test that does not need the collaboration of the mice, providing valuable data about muscle integrity 46 . We observed significant changes in the amplitude of the CMAPs and in the morphology of the MUAPs using electrophysiological tests that were restored with treatment. These results are probably the consequence of early changes in the muscle structure produced by nintedanib that occur before clinical symptoms are observed. The effect of specific TKIs, such as nintedanib, on the amount of fibrotic tissue present in the skeletal muscles of the patients may also improve muscle homeostasis. It is well known that muscle microenvironment is important for proper tissue regeneration 47 . In the process Fig. 5 Nintedanib reduced skeletal muscle fibrosis in mdx mice. Representative pictures of immunofluorescence staining for collagen VI in quadriceps (a–c), diaphragm (d–f), and tibialis anterior (g–i) of WT, mdx, and nintedanib-treated mdx mice. Endomysial collagen deposition increased in all mdx muscles compared with WT group and decreased in nintedanib-treated mdx muscles, significantly in quadriceps (j) and diaphragm (k) but not in tibialis anterior (l) compared with non-treated mdx mice. Data are expressed as means ± SD. Genetic background mouse strain C57BL (WT); n=5, mdx mice (mdx), n=5; nintedanib-treated mdx mice (mdx +Ninte), n=7. *P< 0.05. Scale bar of quadriceps and tibialis anterior =200 µm. Scale bar of diaphragm =100 µm. Piñol-Jurado et al. Cell Death and Disease (2018) 9:776 Page 7 of 15 Official journal of the Cell Death Differentiation Association Fig. 6 (See legend on next page.) Piñol-Jurado et al. Cell Death and Disease (2018) 9:776 Page 8 of 15 Official journal of the Cell Death Differentiation Association of muscle degeneration, there is an imbalance between proregenerative and prodegenerative processes. The increase in fibro-adipous tissue and the presence of chronic inflammation seem to impinge satellite cell proliferation, differentiation, and eventually muscle regeneration 48,49 . We hypothesize that nintedanib treatment could have a double benefit: (1) it reduces the amount of fibrotic tissue and (2) it could indirectly, restore the balance between proregenerative and prodegenerative processes. Our results support this hypothesis. We observed that nintedanib did not interfere with the differentiation ability of myoblasts in vitro or the resistance of myotubes to chemical damage. However, we did not see an increase in the number of regenerative muscle fibers in the animals treated suggesting that a reduction of the fibrotic process preserves skeletal muscle structure. Indeed, electromyograhy (EMG) studies showing restoration of MUAPs morphology indicate that muscle structure is more preserved in the treated animals supporting this hypothesis. It is important to take into account that we have observed a reduction of fibrotic tissue in an old animal, in which the fibrotic process has already developed. Treatment of younger animals for longer periods of time could result in a more sustained effect, although this needs to be confirmed in further experiments. (see figure on previous page) Fig. 6 RT-qPCR studying mRNA expression of fibrosis-related genes. Collagen type I alpha 1 chain (Col1a1), Collagen type III alpha 1 chain (Col3a1), Fibronectin 1 (Fn1), Platelet-derived growth factor A (Pdgfa), Platelet-derived growth factor B (Pdgfb), Connective tissue growth factor (Ctgf), Transforming growth factor beta 1 (Tgfβ1), and Adhesion G protein-coupled receptor E1 (Adgre1) showed changes in relative abundance following nintedanib in quadriceps (a–h), diaphragm (i–p), and tibialis anterior (q–y). Col1a1, Col3a1, Pdgfa, Tgfb1, and Adgre1 gene expression was increased in mdx mice compared with WT. Col1a1, Col3a1, Pdgfa, Pdgfb, Tgfb1, and Adgre1 expression was reduced in all muscles analyzed from nintedanibtreated mdx mice compared with mdx mice. In contrast, Ctgf expression was increased in muscles from nintedanib-treated mdx mice compared with mdx mice. Data are expressed as means ± SD. Genetic background mouse strain C57BL (WT); n=5, mdx mice (mdx), n=5; nintedanib-treated mdx mice (mdx +Ninte), n=7. *P< 0.05, **P< 0.01, ***P< 0.005 Fig. 7 Nintedanib decreased collagen I expression in mdx muscles. Western blotting of Collagen I in quadriceps (a), diaphragm (b), and tibialis anterior (c) showed a reduction in the protein expression in diaphragm (e) and tibialis anterior (f) but not in quadriceps (d) of nintedanib-treated mdx mice compared with mdx mice by quantification of Collagen I normalized to α-tubulin levels. Data are expressed as means ± SD. Genetic background mouse strain C57BL (WT); n=4, mdx mice (mdx), n=4; nintedanib-treated mdx mice (mdx +Ninte), n=4. *P≤0.05 and **P≤0.01 Piñol-Jurado et al. Cell Death and Disease (2018) 9:776 Page 9 of 15 Official journal of the Cell Death Differentiation Association