Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 https://doi.org/10.1186/s12891-020-03756-7 RESEARCH ARTICLE Open Access Proteasome inhibitors reduce thrombospondin-1 release in human dysferlin-deficient myotubes Esther Fernández-Simón1, Cinta Lleixà1, Xavier Suarez-Calvet1,2, Jordi Diaz-Manera1,2,3, Isabel Illa1,2,3, Eduard Gallardo1,2*† and Noemí de Luna1,2*† Abstract Background: Dysferlinopathies are a group of muscle disorders causing muscle weakness and absence or low levels of dysferlin, a type-II transmembrane protein and the causative gene of these dystrophies. Dysferlin is implicated in vesicle fusion, trafficking, and membrane repair. Muscle biopsy of patients with dysferlinopathy is characterized by the presence of inflammatory infiltrates. Studies in the muscle of both human and mouse models of dysferlinopathy suggest dysferlin deficient muscle plays a role in this inflammation by releasing thrombospondin- 1. It has also been reported that vitamin D3 treatment enhances dysferlin expression. The ubiquitin-proteasome system recognizes and removes proteins that fail to fold or assemble properly and previous studies suggest that its inhibition could have a therapeutic effect in muscle dystrophies. Here we assessed whether inhibition of the ubiquitin proteasome system prevented degradation of dysferlin in immortalized myoblasts from a patients with two missense mutations in exon 44. Methods: To assess proteasome inhibition we treated dysferlin deficient myotubes with EB1089, a vitamin D3 analog, oprozomib and ixazomib. Western blot was performed to analyze the effect of these treatments on the recovery of dysferlin and myogenin expression. TSP-1 was quantified using the enzyme-linked immunosorbent assay to analyze the effect of these drugs on its release. A membrane repair assay was designed to assess the ability of treated myotubes to recover after membrane injury and fusion index was also measured with the different treatments. Data were analyzed using a one-way ANOVA test followed by Tukey post hoc test and analysis of variance. A p ≤ 0.05 was considered statistically significant. Results: Treatment with proteasome inhibitors and EB1089 resulted in a trend towards an increase in dysferlin and myogenin expression. Furthermore, EB1089 and proteasome inhibitors reduced the release of TSP-1 in myotubes. However, no effect was observed on the repair of muscle membrane after injury. Conclusions: Our findings indicate that the ubiquitin-proteasome system might not be the main mechanism of mutant dysferlin degradation. However, its inhibition could help to improve muscle inflammation by reducing TSP-1 release. Keywords: Dysferlin, Proteasome, Vitamin D3, Myogenin, Sarcolemma, Thrombospondin-1 * Correspondence: egallardo@santpau.cat; nluna@santpau.cat † Eduard Gallardo and Noemí de Luna contributed equally to this work. 1 Neuromuscular Diseases group. Institut de Recerca Hospital de Sant Pau, Biomedical Research Institute Sant Pau (IIB Sant Pau), Universitat Autónoma de Barcelona, c/Sant Antoni Mª Claret 167, 08025 Barcelona, Spain Full list of author information is available at the end of the article © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 Page 2 of 10 Background Duchenne and Becker muscular dystrophy [20–22]. Mutations in the dysferlin gene (DYSF) that lead to ab- Bortezomib, another proteasome inhibitor, promoted sence or marked reduction of the protein are the cause the expression and membrane localization of dys- of dysferlinopathy [1]. The disease is an autosomal reces- trophin and dystrophin-associated proteins in mdx sive muscle disorder and has several phenotypes, includ- mice, but the drug had only a modest effect in myo- ing limb girdle 2B [2], Miyoshi myopathy [3] and distal blasts from patients with DMD [22]. Since patients anterior compartment myopathy [4]. Progression of the carrying missense mutations in DYSF have residual ex- disease is characterized by muscle weakness and atrophy, pression of dysferlin, prevention of its degradation by and muscle biopsy shows mainly necrosis of muscle fibers the UPS could be used as a therapeutic approach. and inflammatory infiltrates. Although several genetic and Apart from avoiding protein degradation, other studies pharmacologic treatments have been tested [5, 6], no have focused on increasing dysferlin expression by acti- curative treatment is yet available for dysferlinopathy [7]. vating gene expression. In peripheral blood monocytes DYSF contains 55 exons and 14 isoforms. Fifty-three (PBM) and myotubes of dysferlinopathy carriers bearing percent of patients have nonsense mutations, 43% have one mutation in DYSF, dysferlin expression is increased missense mutations, and 4% have small in-frame inser- after vitamin D3 treatment [8]. Vitamin D3 binds to the tions or deletions [1]. While nonsense mutations pro- vitamin D3 response element (VDRE) present in the dys- duce a complete absence of the protein, a residual ferlin promoter and enhances dysferlin expression. expression of dysferlin is detected in primary myotubes EB1089 is an analogue of vitamin D3 that shows higher from patients carrying missense mutations [8, 9]. efficacy and less hypercalcemic activity [23], but its effect Dysferlin protein is highly expressed in skeletal muscle, on dysferlin expression has not been studied. cardiac muscle and blood monocytes [10]. It is a type-II As there are no effective treatments for dysferlinopa- transmembrane protein that contains seven C2 domains thies, we assessed whether inhibition of the ubiquitin (Ca2+-binding) involved in vesicle fusion, trafficking, proteasome system could prevent degradation of dysfer- membrane repair and regulation of calcium homeostasis lin in immortalized myoblasts from a patient carrying [11, 12]. However, it is also associated with other pro- two missense mutations. We used oprozomib and ixazo- cesses, including intracellular signaling and myoblast dif- mib, 2 s generation proteasome inhibitors that have re- ferentiation [13]. Studies using primary cultures of human duced toxicity and off-targets, increasing effectiveness. skeletal muscle show that dysferlin and myogenin may We also studied the role of EB1089 in the expression of share a pathway involved in differentiation of skeletal dysferlin, alone or in combination with proteasome in- muscle “in vitro” as dysferlin-deficient (dysf mut/mut) hibitor. We evaluated the effect of these treatments by myotubes have reduced expression of myogenin and are studying the expression of myogenin, the ability of myo- poorly differentiated [14]. It has been reported that tubes to repair the sarcolemma after injury and the re- thrombospondin-1 (TSP-1) expression is increased in dysf lease of TSP-1. mut/mut myotubes and may play a role in the inflamma- tion observed in muscle biopsies of patients with dysferlin Material and methods myopathy [15]. In fact, it has been shown that serum TSP- Samples 1 levels correlate with macrophage inflammation and In all experiments we used immortalized myoblasts from muscle damage in dysferlin-deficient BlaJ mice [16]. a dysferlinopathy patient with homozygous missense The ubiquitin-proteasome system (UPS) is the major mutations (c.4882G > A/p.G1628R) in exon 44 of DYSF proteolytic pathway as more than 80% of cellular pro- and from a healthy control. This patient corresponds to teins are degraded therein [17]. The UPS is an ATP- number 16 in our previous work and was described dependent system that removes unfolded or misfolded showing some degree of dysferlin expression in sarco- proteins [18]. It interacts only with proteins labelled with plasm [24]. These cells present a residual expression of a polyubiquitin chain [18]. Proteins bind proteasome dysferlin of about 1–2% compared to WT myotubes. through the α-rings of the 20S subunit and then pass These cells were kindly provided by Dr. Mouly [25]. through the β-rings, where they are degraded by the pro- Myoblasts were expanded using skeletal muscle medium teasome particles: chymotrypsin-like (CT-L), caspase- (SMM; Promocell, Heidelberg, Germany) until confluence like (C-L), and trypsin-like (T-L) proteolytic sites [19]. was achieved. Media was then changed to differentiation Previous studies have shown that proteasome inhibitors medium (75% Dulbecco’s Modified Eagle’s Medium and could have a therapeutic implication in muscle dystrophies. 25% M199, supplemented with 2% FBS (Lonza, Basel, It has been reported that the proteasome inhibitor MG-132 Switzerland), 10 μg/ml insulin (Sigma-Aldrich, St Louis, rescued the expression of the dystrophin-glycoprotein com- MO, USA), 2 mmol/l glutamine (Lonza) and penicillin- plex (DGC) in mouse models of Duchenne muscular dys- streptomycin (Lonza)) for 7 to 9 days. Differentiated myo- trophy and in skeletal muscle cultures from patients with tubes were treated with vitamin D3 or EB1089 (Sigma Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 Page 3 of 10 Aldrich) at 100 nM, ixazomib (Selleckchem, Munich, read using Victor 3v Multilabel Plate Reader (Perkin Germany) (25 nM, 50 nM and 100 nM) and oprozomib Elmer, Waltham, MA, USA). (ONX 0912)(Selleckchem) (10 nM, 50 nM and 100 nM). Ixazomib and oprozomib were combined with EB1089 at Thrombospondin-1 Enzyme-Linked Immunosorbent Assay 8 h and 24 h, respectively. (ELISA) Immortalized myoblasts from the dysferlinopathy patient were seeded at 5000 cells/cm2 and expanded until con- Western blot fluence. The media was then changed to differentiation Cell pellets corresponding to each condition were lysed medium to form myotubes. After treatment with prote- in RIPA buffer (Sigma-Aldrich) containing a protease asome inhibitors together with or without EB1089, we and phosphatase inhibitor cocktail (Roche, Basel, removed the media and added 1 ml of basal DMEM Switzerland). Lysates were centrifuged at 4 °C at 13000 x (Lonza) to the culture for 24 h. Cell culture supernatants g for 20 min and supernatants were stored at − 80 °C. were concentrated using Amicon Ultra Centrifugal Filters Protein concentrations were determined using Pierce™ 100 kb (Merck Millipore, Darmstadt, Germany). TSP-1 BCA Protein Assay (Thermo Fisher Scientific, Waltham, present in the culture media was detected using the human MA, USA). Thirty micrograms of protein were resolved TSP-1 Immunoassay (Quanti-Kine ELISA, R&D Systems, in a 10% sodium dodecyl sulfate (SDS) polyacrylamide Minneapolis, MN), following the manufacturer’s instruc- gel and transferred to nitrocellulose membranes. Unspe- tions. The detection limit of the assay was 0.355 ng/mL. cific binding sites were blocked by incubation for 1 h in casein diluted 1:1 in tris-buffered saline (TBS). Blots Membrane repair assay were incubated overnight with the primary mouse Immortalized myoblasts from the dysferlinopathy patient monoclonal antibodies anti-dysferlin (NCL-Hamlet, and the healthy control were grown until confluence in Novocastra, Newcastle, UK) and anti-myogenin (5FD chambered wells and then differentiated to myotubes. clone) (Santa Cruz Biotechnology, Dallas, TX). To The assay was performed as recently described in human normalize the results, mouse anti-desmin (Novocastra) muscle primary cultures [26]. SDS is an anionic surfac- was added simultaneously with the primary antibody. tant that causes the leakage of intracellular components The secondary antibody for anti-dysferlin was a goat due to its ability to affect membranes [27]. Membrane anti-mouse labeled with IR-Dye 800 (Li-Cor, Lincoln, injury was induced by detergent treatment as follows: Nebraska, USA). When the anti-myogenin antibody was after the cell cultures were washed with Hank’s Balanced used as a primary antibody, biotin-labeled horse anti- Salt Solution (HBSS) (Lonza), the injury solution (HBSS mouse (Jackson ImmunoResearch, Ely, UK) was used as with 0.12 mM or 0.25 mM SDS (Sigma-Aldrich)) was ap- a secondary antibody. Secondary antibodies were incu- plied for 2 min. Following exposure to injury solution, bated for 1 h at room temperature. Membranes incu- cells were washed in HBSS and then incubated in recov- bated with biotinylated secondary antibody were washed ery solution consisting of proliferation media for 90 s and incubated for 1 h with IRDye-680-labeled streptavi- and 10 min. The injury and recovery steps were per- din (Li-Cor). After extensive washing, the immunoreac- formed at 37 °C. Cells with injured permeable mem- tive bands were visualized using the Odyssey Infrared branes were identified by exposure to propidium iodide ImagingSystem (Li-Cor). The amount of protein was (PI) (Sigma-Aldrich) (20 μg/mL in HBSS). If cell mem- quantified using Image Studio Lite software (Li-Cor). brane is not repaired, PI enters the cell and penetrates Desmin expression was used as a loading control. As a the nucleus wherein it intercalates with DNA and emits reference value we used dysferlin or myogenin expres- fluorescence. After 2 min of exposure to the dye, cells sion of the healthy control. Protein expression in the were washed with HBSS and fixed with 4% paraformal- remaining conditions was quantified over the expression dehyde in PBS. Nuclei were stained with Hoescht 33,342 in WT myotubes and expressed as fold-change. (Invitrogen, Thermo-Fisher Scientific). In each experi- mental condition, we counted the total number of nuclei Assessment of proteasome-like activity and the PI positive nuclei using Fiji software [28]. Cells were seeded in a 96- well black plate (Sarstedt, Nümbrecht, Germany) at 5000 cells/well in 3 replicates Fusion index until confluence. The media was then changed to differ- Once membrane repair assay was performed, myotubes entiation medium. Cells were treated with the corre- were permeabilized with ethanol for 5 min at room sponding drug for the indicated time. The CT-L, C-T temperature. Samples were then blocked with the Ultra- and T-L activity was assayed by chemiluminiscence Cruz Blocking Reagent (Santa Cruz Biotechnology, using the Proteasome-Glo™ 3-substrate System cell based Dallas, TX) for 1 h at RT. Myotubes were stained using a assay (Promega, Madison, WI, USA) and the plate was mouse anti-myosin heavy chain (MyHC) antibody (MF- Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 Page 4 of 10 20)(Bio-Rad, Hercules, CA, USA) for 1 h at RT and then in combination with proteasome inhibitors to increase the goat anti-mouse secondary antibody Alexa Fluor 594 dysferlin expression. (ThermoFisher, Wal- tham, MA) was used as a secondary antibody for 1 h at RT. Following the incubation with the Proteasome inhibition secondary antibody, nuclei were stained with Hoescht We analyzed the time points at which ixazomib and opro- 3342 (Invitrogen, Thermo-Fisher Scientific). Images were zomib produced the highest increase in dysferlin expres- acquired and analysed using Fiji software. sion in dysf mut/mut myotubes. The effect of ixazomib The fusion index was determined as the percentage of was highest at 8 h and the effect of oprozomib was highest nuclei included in MF-20-expressing myotubes (contain- at 24 h (data not shown). We also analyzed whether the ing at least 3 nuclei) divided by the total number of nuclei. addition of EB1089 to the proteasome inhibitors had an effect on dysferlin expression at the same time points. Statistical analysis Treatment of dysf mut/mut myotubes with 10 nM Multiple comparisons were analyzed using a one-way oprozomib for 24 h showed high selectivity towards the ANOVA test followed by Tukey post hoc test. When CT-L active site of the proteasome. However, at higher statistical comparisons were performed between two doses (50 nM and 100 nM) oprozomib also inhibited C- groups, the nonparametric Mann-Whitney test was used. L and T-L activity significantly (Fig. 2a-c). GraphPad Prism 5.0 software was used (LaJolla, CA, Ixazomib treatment inhibited all 3 proteasome sub- USA) for graphic representation. units (Fig. 2d-f). When EB1089 was added alone to the culture, no pro- Results teasome inhibition in any subunit was found. Moreover, EB1089 and vitamin D3 treatment when it was combined with ixazomib and oprozomib, no Vitamin D3 and EB1089 were added to healthy control effect was observed in the inhibition of each subunit of myotubes for 24 h at 100 nM in order to assess their ef- the proteasome. fect on dysferlin expression. Both Vitamin D3 and EB1089 increased dysferlin expression. However, only Dysferlin and myogenin expression treatment with EB1089 showed statistically significant We analyzed the expression of dysferlin and myogenin differences compared to non-treated myotubes (Fig. 1). in dysf mut/mut myotubes treated with EB1089, oprozo- Therefore, from this point on, we decided to use EB1089 mib and ixazomib. Fig. 1 Effect of Vitamin D3 and EB1089 on healthy control myotubes. a Quantification of WB bands of untreated wild-type myotubes (C- ), Vitamin D3 and EB1089 treated wild-type myotubes for 24 h. Vitamin D3 treatment produced an increase in dysferlin expression compared to non-treated myotubes. However, only EB1089 showed a significant increase compared to non-treated myotubes. b Representative WB of dysferlin with vitamin D3 and EB1089 treatment. Desmin was used as a loading control. Data are represented as mean of 3 replicates ± standard deviation. Fold-change is calculated over the results of dysferlin expression in non-treated WT myotubes. Results were statistically analyzed using one-way Anova followed by Tukey post hoc test. Statistical significance was set at p < 0.05. *p < 0.05 Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 Page 5 of 10 Fig. 2 Proteasome inhibition profile. The three distinct ATPase-independent proteolytic activities (CT-L, C-L and T-L) of the proteasome were assayed using a chemioluminiscence-based method. Dysf mut/mut myotubes were pretreated with oprozomib and EB1089 for 24 h. Oprozomib treatment presented inhibition at: a CT-L at all concentrations tested b) C-L, and c) T-L activity also were inhibited by oprozomib, but only at high doses (50 nM and 100 nM). Pre-treatment with ixazomib and EB1089 for 8 h was also assayed and d) CT-L, e) C-L and f) T-L activity were inhibited at all doses tested with significant differences. Data are represented as mean of 3 replicates ± standard deviation. Results were statistically analyzed using one-way Anova followed by Tukey post hoc test. Statistical significance was set at p < 0.05. *p < 0.05; **p < 0.01; ***p < 0.0001 Treatment with oprozomib at 24 h showed a trend to- 1 release at 8 h. TSP-1 levels also decreased significantly wards an increase in myogenin expression (Fig. 3a-b). in dysf mut/mut cells treated with oprozomib and When EB1089 was added together with oprozomib, myo- EB1089 separately (Fig. 4a). TSP-1 levels decreased even genin presented a trend towards decreased expression more significantly when cells were treated with the com- compared to oprozomib treatment alone. Dysferlin pre- bination of EB1089 and oprozomib. sented a trend towards increased expression with 10 nM Ixazomib treatment significantly decreased TSP-1 only of oprozomib with and without EB1089. No statistical sig- when combined with EB1089 (Fig. 4b). However, al- nificances were observed in any experimental conditions. though EB1089 and ixazomib alone did not produce Treatment with ixazomib (Fig. 3c-d) showed no sig- significant changes in TSP-1 expression, expression was nificant increase in the expression of dysferlin. Myogenin lower than in untreated myotubes. levels showed a trend towards increased expression after Raw western-blot data are available in Supplementary ixazomib treatment, being 100 nM the concentration at Figure 1. which the myogenin expression was the highest. The addition of EB1089 together with ixazomib presented a Sarcolemma repair assay trend towards decreased expression of myogenin com- Considering that dysferlin is involved in membrane re- pared to ixazomib treatment alone, but the results did pair after injury, we analyzed the ability of dysf mut/mut not reach statistical significance. myotubes, treated with proteasome inhibitors and EB1089, to repair the sarcolemma after injury. We found Expression of TSP-1 that wild-type myotubes repaired the membrane at both We evaluated the secretion of TSP-1 by dysf mut/mut low and high concentration of SDS (0.12 mM SDS and myotubes treated with EB1089, oprozomib and ixazo- at 0.25 mM SDS), showed by PI exclusion, 10 min after mib. Treatment with EB1089 significantly reduced TSP- injury (Fig. 5a). However, dysf mut/mut myotubes did Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 Page 6 of 10 Fig. 3 Dysferlin and myogenin expression in dysf mut/mut muscle cells. a Representative WB of myogenin and dysferlin with oprozomib and EB1089 treatment. C+: untreated wild-type myotubes, (C-) untreated dysf mut/mut myotubes. Desmin was used as loading control. b Quantification of WB bands of dysferlin or myogenin expression. The highest expression of dysferlin and myogenin was reached using oprozomib at 10 nM. c Representative WB of myogenin and dysferlin with ixazomib and EB1089. d Quantification of WB showed that 100 nM of ixazomib presented the highest expression of dysferlin and myogenin. Data are represented as the mean of 3 replicates ± standard deviation. The fold- change is calculated over the results of dysferlin expression in WT non-treated myotubes. Results were statistically analyzed using one-way Anova followed by Tukey post hoc test. Statistical significance was set at p < 0.05 not present any sarcolemma repair in any condition Treatment with oprozomib, ixazomib and EB1089 in tested (Fig. 5a-c). myotubes from a dysferlinopathy patient promoted a trend towards increased expression of dysferlin and Fusion index myogenin, but this low increase did not translate into Fusion index results showed that dysf mut/mut myotubes higher sarcolemmal repair neither higher fusion index presented a statistically significant lower fusion index than that in untreated myotubes. However, TSP-1 re- compared to wild-type myotubes (Fig. 5e). Ixazomib, lease from myotubes decreased after treatment. oporozomib and EB1089 did not show any effect on the Western blot analysis showed a trend towards in- fusion index of dysf mut/mut myotubes (Fig. 5f-g). creased expression of dysferlin and myogenin when oprozomib and EB1089 were added to the culture. Ixa- Discussion zomib produced a trend towards increased expression The aim of this study was to test if blockade of prote- of myogenin but not dysferlin. These results are in asome could restore dysferlin expression and conse- agreement with a previous work by Fujita E et al. They quently rescue the muscle functions that are impaired found that mutant-dysferlin was mainly degraded by au- when dysferlin is missing. It has been reported that dys- tophagy while wild type-dysferlin was degraded by the ferlin deficient myoblasts display reduced levels of myo- ubiquitin-proteasome system, although they used a pro- genin and consequently fusion index is reduced [14]. teasome inhibitor different from those used in this study Muscles biopsies of dysferlinopathy patients are charac- [31]. The small increase in dysferlin expression that we terized by the presence of abundant inflammatory infil- observed after treating dysf mut/mut myotubes with our trates [29], TSP-1 seems to have a prominent role in different ubiquitin-proteasome inhibitors confirms that muscle inflammation in this pathology [15]. Finally, dys- this system is not the main pathway to recycle mutated ferlin is involved in sarcolemma repair [26, 30]. dysferlin. Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 Page 7 of 10 Fig. 4 TSP-1 expression by dysf mut/mut myotubes treated with oprozomib, ixazomib and EB1089. a TSP-1 expression with treatment of oprozomib at 10 nM and EB1089 100 nM was significantly decreased compared to C- (non-treated dysf mut/mut myotubes), and the combination of both treatments was also significantly decreased compared to oprozomib treatment alone. b There were no significant differences in TSP-1 levels between ixazomib and EB1089 compared to non-treated myotubes. However, the combination of both treatments showed a significant reduction compared to non-treated myotubes. Data are represented as mean of 5 replicates ± standard deviation. Results were statistically analyzed using one-way Anova followed by Tukey post hoc test. Statistical significance was set at p < 0.05. **p < 0.01; ***p < 0.0001 The other treatment that we studied, the analogue of factors modulate its release. It is expressed by endothe- Vitamin D3, EB1089, produced a trend towards increased lial cells, fibroblasts, neutrophils, macrophages, T cells, expression of dysferlin. In other studies, authors found and myotubes [37]. We observed a significant decrease that Vitamin D3 treatment had no effect on dysferlin ex- in TSP-1 release when dysf mut/mut myotubes were ex- pression in myotubes from dysferlinopathy patients but posed to EB1089, oprozomib and the combination of increased dysferlin expression in muscle cultures from oprozomib or ixazomib and EB1089. The decrease in carriers of one mutation in DYSF [8]. Although EB1089 is TSP-1 levels could be the consequence of the trend to- 50–200 times more potent than vitamin D3 [23], when wards an increased expression of dysferlin in these con- added to cultured myotubes from a dysferlinopathy pa- ditions. However, TSP-1 levels were significantly lower tient it only produces a trend towards an increase in dys- when a combination of oprozomib and EB1089 was ferlin expression. Vitamin D3 acts through the Vitamin D added to the culture than when oprozomib was added receptor and p38 MAPK and participates in differentiation alone, suggesting that proteasome inhibitors and EB1089 events in skeletal muscle cells [32]. Whether this effect is act independently of dysferlin expression to reduce TSP- dependent on dysferlin expression remains to be eluci- 1 release. dated. Moreover, EB1089 also presented a trend towards The reduced expression of TSP-1 observed after treat- decreased expression of myogenin when was combined ment with EB1089 could be explained by mechanisms with proteasome inhibitors compared to the results of independent of dysferlin expression. Amarasekera AT treatment with a proteasome inhibitor alone. Although re- et al. suggested that vitamin D supplementation reduces sults were not statistically significant, this negative effect TSP-1 levels in healthy individuals [38]. Moreover, it has could be explained by a negative regulation of myogenin been studied that calcitrol, a vitamin D metabolite, after EB1089 treatment. In previous studies, authors downregulates TSP-1 via the activation of the mitogen showed that Vitamin D3 downregulates the expression of activated protein kinase (MAPK) [39]. Therefore, EB1089 myogenin both in vitro and in vivo [33–35]. These previ- treatment could activate the MAPK pathway leading to ous findings suggest that negative vitamin D response ele- downregulation of TSP-1. This activation could constitute ments may be present in the promoter region of an alternative mechanism to reduce TSP-1 levels that is myogenin, however those have not been identified [36]. not linked to dysferlin expression. We also analyzed the expression of TSP-1 in cell cul- The effect of proteasome inhibitors on TSP-1 could tures after treatment with proteasome inhibitors. TSP-1 also be explained by mechanisms non-related to dysfer- is secreted in response to inflammation and multiple lin expression, such as the inhibition of canonical NF-κB Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 Page 8 of 10 Fig. 5 Membrane repair and fusion index using oprozomib, ixazomib and EB1089 treatment. a Representative images of PI-positive nuclei and MyHC positive myotubes after SDS treatment at 90 s and at 10 min. The first 2 rows correspond to healthy myotubes and the last 2 rows correspond to dysf mut/mut myotubes treated with oprozomib. b Quantification of IP-positive nuclei in myotubes from healthy controls and dysf mut/mut myotubes treated with EB1089, oprozomib, and the combination of oprozomib 10 nM with EB1089at at 24 h. c Quantification of IP- positive nuclei in myotubes from healthy controls and dysf mut/mut myotubes treated with EB1089, ixazomib, and the combination of ixazomib 100 nM with EB1089at at 8 h. Differences were not observed with any treatment. d Quantification of fusion index showed a statistically significant decrease of MyHC positive myotubes compared to healthy myotubes. e Fusion index in dysf mut/mut myotubes treated with EB1089, oprozomib at 10 nM and the combination of oprozomib 10 nM with EB1089 at 8 h did not show any difference compared to non-treated myotubes. f Fusion index in dysf mut/mut myotubes treated with EB1089, ixazomib at 100 nM and the combination of ixazomib 100 nM with EB1089 at 24 h did not show any difference compared to non-treated myotubes. Data are represented as mean of 3 replicates ± standard deviation. Results were statistically analyzed using Mann-Whitney test to compare two groups and one-way Anova followed by Tukey post hoc test for multiple comparisons. Statistical significance was set at p < 0.05. *p < 0.05; **p < 0.01; ***p < 0.0001 signaling. NF-κB forms a dimer with IκB, an inhibitory response. Proteasome inhibitors seem to decrease the protein, which keeps the complex inactive in the cyto- nuclear levels of NF-κB by avoiding the degradation of plasm. When signals activate this pathway, IκB is phos- IκB, hence decreasing NF-κB activity [40]. In effect, in phorylated and degraded by the proteasome, releasing dysferlinopathy patients, damage-associated molecular the NF-κB to the nucleus where it regulates the tran- patterns (DAMPs) are released from damaged muscle scription of genes involved in the inflammatory fibers, triggering inflammatory responses via activation Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 Page 9 of 10 of NF-κB [41]. We suggest that decreased NF-κB activity the manuscript and obtained funding for the study. The author(s) read and by proteasome inhibitors could be another mechanism approved the final manuscript. involved in TSP-1 reduction and thereby reduce the in- Funding flammatory response. This project has been funded by projects from the Fundación Isabel Gemio We did not observe any sarcolemma repair when pro- to II, EG and JDM and by Fundación Ramón Areces (CIVP18A3903) to NdL. XSC was supported by a “Sara Borrell” post- doctoral fellowship project teasome inhibitors and EB1089 were added to the cul- "CD18/00195", funded by Instituto de Salud Carlos III and co-funded by Euro- ture. We have previously reported that dysferlin mut/ pean Union (ERDF/ESF, "A way to make Europe"/"Investing in your fu- mut cells fuse less efficiently than wild-type myoblasts. ture"). CERCA Programme / Generalitat de Catalunya. We also demonstrated that dysferlin mut/mut myotubes Availability of data and materials showed a reduced expression of myogenin in the nucleus The datasets used and/or analyzed during the current study are available [14]. The experiments performed in the present study from the corresponding author on reasonable request. confirm that fusion of myoblasts is impaired in dysferlin Ethics approval and consent to participate mut/mut. Lack of dysferlin impaires sarcolemmal repair The study was approved by the Ethics Committee at Hospital de la Santa both in myoblasts and myotubes therefore, we believe Creu i Sant Pau (HSCSP) in Barcelona. All patients and controls gave that reduced fusion of myoblasts has no effect on the informed consent to use their muscle biopsy for research purposes. membrane repair experiments. The trend towards in- Consent for publication creased expression of dysferlin after treatment was not Not applicable. strong enough to restore sarcolemma repair of damaged Competing interests myotubes. The authors declare that they have no competing interests. Conclusions Author details 1 Neuromuscular Diseases group. Institut de Recerca Hospital de Sant Pau, In conclusion, we did not observe a significant increase of Biomedical Research Institute Sant Pau (IIB Sant Pau), Universitat Autónoma dysferlin or myogenin in myotubes from a dysferlinopathy de Barcelona, c/Sant Antoni Mª Claret 167, 08025 Barcelona, Spain. 2Centro patient when treated with EB1089, oprozomib and ixazo- de Investigación Biomédica en Red de Enfermedades Raras (CIBERER), Valencia, Spain. 3Department of Neurology, Neuromuscular Diseases Unit, mib. However, we did observe a significant reduction in Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, TSP-1 levels. Targeting TSP-1 may constitute a thera- Barcelona, Spain. peutic approach in patients carrying mutations affecting Received: 9 December 2019 Accepted: 30 October 2020 proper folding of dysferlin since it would reduce inflam- mation and avoid the undesirable side-effects reported in patients with dysferlinopathy treated with prednisone. References 1. Bashir R, Strachan T, Keers S, Stephenson A, Mahjneh I, Marconi G, et al. A gene for autosomal recessive limb-girdle muscular dystrophy maps to Supplementary Information chromosome 2p. Hum Mol Genet. 1994;3(3):455–7. The online version contains supplementary material available at https://doi. 2. Bashir R, Keers S, Strachan T, Passos-Bueno R, Zatz M, Weissenbach J, et al. org/10.1186/s12891-020-03756-7. Genetic and physical mapping at the limb-girdle muscular dystrophy locus (LGMD2B) on chromosome 2p. Genomics. 1996;33(1):46–52. 3. Bejaoui K, Hirabayashi K, Hentati F, Haines JL, Hamida C. Ben, Belal S, et al. Additional file 1: Supplementary Figure 1.` Dysferlin and myogenin linkage of Miyoshi myopathy (distal autosomal recessive muscular expression in dysf mut/mut muscle cells. A) Representative full blot of dystrophy) locus to chromosome 2p12-14. Neurology. 1995;45(4):768–72. dysferlin with vitamin D3 and EB1089 treatment. Desmin was used as a 4. Illa I, Serrano-Munuera C, Gallardo E, Lasa A, Rojas-Garca R, Palmer J, et al. loading control. B) Representative full blot of myogenin and dysferlin Distal anterior compartment myopathy: a dysferlin mutation causing a new with oprozomib and EB1089 treatment. C) Representative full blot of muscular dystrophy phenotype. Ann Neurol. 2001;49(1):130–4. myogenin and dysferlin with ixazomib and EB1089 treatment. 5. Malcher J, Heidt L, Goyenvalle A, Escobar H, Marg A, Beley C, et al. Exon Skipping in a Dysf-Missense Mutant Mouse Model. Mol Ther Nucleic Acids. 2018;13:198–207. Abbreviations 6. Lee JJA, Maruyama R, Duddy W, Sakurai H, Yokota T. Identification of novel DYSF: Dysferlin; C2: Ca2+-binding; Dysf mut/mut: Dysferlin deficient; TSP- antisense-mediated exon skipping targets in DYSF for therapeutic treatment 1: Thrombospondin 1; UPS: Ubiquitin-proteasome system; C-L: Caspase-like; of Dysferlinopathy. Mol Ther Nucleic Acids. 2018;13:596–604. T-L: Trypsin-like; CT-L: Chemotrypsin-like; DGC: Dystrophin-glycoprotein 7. Brown RH. Dysferlinopathies. Handb Clin Neurol. 2011;101:111–8. complex; PBM: Peripheral blood monocytes; VDRE: Vitamin D3 response 8. De Luna N, Díaz-Manera J, Paradas C, Iturriaga C, Rojas-García R, Araque J, element; PI: Propidium iodide; DAMP: Damage-associated molecular pattern et al. 1α,25(OH)(2)-vitamin D3 increases dysferlin expression in vitro and in a human clinical trial. Mol Ther. 2012;20(10):1988–97. Acknowledgements 9. Woolger N, Bournazos A, Sophocleous RA, Evesson FJ, Lek A, Driemer B, Not applicable. et al. Proteolysis reveals dysferlin conformation in situ limited proteolysis as a tool to probe the tertiary conformation of dysferlin and structural Authors’ contributions consequences of patient missense variant L344P. J Biol Chem. 2017;292(45): EFS performed the experiments, analysed of the results, and prepared the 18577–91. manuscript. CLl performed the experiments. XSC analysed the results. JDM 10. Anderson LV, Davison K, Moss JA, Young C, Cullen MJ, Walsh J, et al. reviewed the manuscript and obtained funding. II reviewed the manuscript Dysferlin is a plasma membrane protein and is expressed early in human and obtained funding. EG performed the experimental design, analysed of development. Hum Mol Genet. 1999;8(5):855–61. the results, prepared the manuscript and obtained funding for the study. 11. Kerr JP, Ward CW, Bloch RJ. Dysferlin at transverse tubules regulates Ca (2+) NdL performed the experimental design, analyzed of the results, prepared homeostasis in skeletal muscle. Front Physiol. 2014;5:89. Fernández-Simón et al. BMC Musculoskeletal Disorders (2020) 21:784 Page 10 of 10 12. Abdullah N, Padmanarayana M, Marty NJ, Johnson CP. Quantitation of the 35. Endo I, Inoue D, Mitsui T, Umaki Y, Akaike M, Yoshizawa T, et al. Deletion of calcium and membrane binding properties of the C2 domains of Dysferlin. Vitamin D Receptor Gene in Mice Results in Abnormal Skeletal Muscle Biophysj. 2014;106:382–9. Development with Deregulated Expression of Myoregulatory Transcription 13. Lukyanenko V, Muriel JM, Bloch RJ. Coupling of excitation to Ca2+ release is Factors. Endocrinology. 2003;144(12):5138–44. modulated by dysferlin. J Physiol. 2017;595(15):5191–207. 36. Wagatsuma A, Sakuma K. Vitamin D signaling in Myogenesis: potential for 14. De Luna N, Gallardo E, Soriano M, Dominguez-Perles R, De La Torre C, Rojas- treatment of sarcopenia. Biomed Res Int. 2014;2014:121254. García R, et al. Absence of dysferlin alters myogenin expression and delays 37. Lopez-Dee Z, Pidcock K, Gutierrez LS. Thrombospondin-1: multiple paths to human muscle differentiation “in vitro.”. J Biol Chem. 2006;281(25):17092–8. inflammation. Mediat Inflamm. 2011;2011:296069. 15. De Luna N, Gallardo E, Sonnet C, Chazaud B, Dominguez-Perles R, Suarez- 38. Amarasekera AT, Assadi-Khansari B, Liu S, Black M, Dymmott G, Rogers NM, Calvet X, et al. Role of Thrombospondin 1 in macrophage inflammation in et al. Vitamin D supplementation lowers thrombospondin-1 levels and Dysferlin myopathy. J Neuropathol Exp Neurol. 2010;69(6):643–53. blood pressure in healthy adults. PLoS One. 2017;12(5):e0174435. 16. Urao N, Mirza RE, Corbiere TF, Hollander Z, Borchers CH, Koh TJ. 39. García-Quiroz J, Rivas-Suárez M, García-Becerra R, Barrera D, Martínez-Reza I, Thrombospondin-1 and disease progression in dysferlinopathy. Hum Mol Ordaz-Rosado D, et al. Calcitriol reduces thrombospondin-1 and increases Genet. 2017;26(24):4951–60. vascular endothelial growth factor in breast cancer cells: implications for 17. Konstantinova IM, Tsimokha AS, Mittenberg AG. Role of Proteasomes in tumor angiogenesis. J Steroid Biochem Mol Biol. 2014;144:215–22. Cellular Regulation. Int Rev Cell Mol Biol. 2008;267:59–124 Academic Press. 40. Verbrugge SE, Scheper RJ, Lems WF, de Gruijl TD, Jansen G. Proteasome 18. Ciechanover A. EMBO MEMBER’S REVIEW The ubiquitin-proteasome inhibitors as experimental therapeutics of autoimmune diseases. Arthritis pathway: on protein death and cell life. EMBO J. 1998;17(24):7151–60. Res Ther. 2015;17(1):17. 19. Glickman MH, Ciechanover A. The ubiquitin-proteasome proteolytic 41. Mariano A, Henning A, Han R. Dysferlin-deficient muscular dystrophy and pathway: Destruction for the sake of construction. Physiol Rev. 2002;82:373– innate immune activation. FEBS J. 2013;280(17):4165–76. 428 American Physiological Society. 20. Bonuccelli G, Sotgia F, Schubert W, Park DS, Frank PG, Woodman SE, et al. Publisher’s Note Proteasome inhibitor (MG-132) treatment of mdx mice rescues the Springer Nature remains neutral with regard to jurisdictional claims in expression and membrane localization of dystrophin and dystrophin- published maps and institutional affiliations. associated proteins. Am J Pathol. 2003;163(4):1663–75. 21. Assereto S, Stringara S, Sotgia F, Bonuccelli G, Broccolini A, Pedemonte M, et al. Pharmacological rescue of the dystrophin-glycoprotein complex in Duchenne and Becker skeletal muscle explants by proteasome inhibitor treatment. Am J Physiol Physiol. 2006;290(2):C577–82. 22. Gazzerro E, Assereto S, Bonetto A, Sotgia F, Scarfì S, Pistorio A, et al. Therapeutic potential of proteasome inhibition in Duchenne and Becker muscular dystrophies. Am J Pathol. 2010;176(4):1863–77. 23. Lu L, Qiu J, Liu S, Luo W. Vitamin D3 analogue EB1089 inhibits the proliferation of human laryngeal squamous carcinoma cells via p57. Mol Cancer Ther. 2008;7(5):1268–74. 24. Gallardo E, de Luna N, Diaz-Manera J, Rojas-García R, Gonzalez-Quereda L, Flix B, et al. Comparison of dysferlin expression in human skeletal muscle with that in monocytes for the diagnosis of dysferlin myopathy. PLoS One. 2011;6(12):e29061. 25. Mamchaoui K, Trollet C, Bigot A, Negroni E, Chaouch S, Wolff A, et al. Immortalized pathological human myoblasts: towards a universal tool for the study of neuromuscular disorders. Skelet Muscle. 2011;1:1(1). 26. De Luna N, Suarez-Calvet X, Garicano M, Fernandez-Simon E, Rojas-García R, Diaz-Manera J, et al. Effect of MAPK inhibition on the differentiation of a Rhabdomyosarcoma cell line combined with CRISPR/Cas9 technology: an in vitro model of human muscle diseases. J Neuropathol Exp Neurol. 2018; 77(10):964–72. 27. Malik JK, Schwarz LR, Wiebel FJ. Assessment of membrane damage in continuous cultures of mammalian cells. Chem Biol Interact. 1983;45(1):29–42. 28. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9(7):676–82. 29. Selcen D, Stilling G, Engel AG. The earliest pathologic alterations in dysferlinopathy. Neurology. 2001;56(11):1472–81. 30. Han R, Campbell KP. Dysferlin and muscle membrane repair. Curr Opin Cell Biol. 2007;19:409–16 NIH Public Access. 31. Fujita E, Kouroku Y, Isoai A, Kumagai H, Misutani A, Matsuda C, et al. Two endoplasmic reticulum-associated degradation (ERAD) systems for the novel variant of the mutant dysferlin: ubiquitin/proteasome ERAD(I) and autophagy/lysosome ERAD (II). Hum Mol Genet. 2007;16(6):618–29. 32. Irazoqui AP, Heim NB, Boland RL, Buitrago CG. 1α,25 dihydroxi-vitamin D3 modulates CDK4 and CDK6 expression and localization. Biochem Biophys Res Commun. 2015;459(1):137–42. 33. Van Der Meijden K, Bravenboer N, Dirks NF, Heijboer AC, Den Heijer M, De Wit GMJ, et al. Effects of 1,25(OH) 2 D 3 and 25(OH) D 3 on C2C12 myoblast proliferation, differentiation, and Myotube hypertrophy. J Cell Physiol. 2016; 231:2517–28. 34. Sydney JEG, South Wales N. Westmead Hospital, Sydney, New South Wales 2145, Australia; and St Vincent’s Clinical School. Aust Endocrinology. 2010; 2052:347–57.