Inflammatory response in human skeletal muscle cells: CXCL10 as a potential therapeutic target

European Journal of Cell Biology 91 (2012) 139–149

Contents lists available at SciVerse ScienceDirect

European Journal of Cell Biology journal homepage: www.elsevier.de/ejcb

Inflammatory response in human skeletal muscle cells: CXCL10 as a potential therapeutic target Clara Crescioli a,∗ , Mariangela Sottili b , Paolo Bonini b , Lorenzo Cosmi b , Paola Chiarugi c , Paola Romagnani b , Gabriella B. Vannelli d , Marta Colletti a , Andrea M. Isidori e , Mario Serio b , Andrea Lenzi e , Luigi Di Luigi a a

Department of Health Sciences, University of Rome “Foro Italico”, 00135 Rome, Italy Excellence Center for Research, Transfer and High Education (DENOthe), University of Florence, 50139 Florence, Italy c Department of Biochemical Science, University of Florence, 50134 Florence, Italy d Department of Anatomy, Histology and Forensic Medicine, University of Florence, 50139 Florence, Italy e Department of Experimental Medicine, Section of Medical Pathophysiology and Endocrinology, Sapienza University of Rome, 00185 Rome, Italy b

a r t i c l e

i n f o

Article history: Received 9 June 2011 Received in revised form 9 August 2011 Accepted 20 September 2011 This work is dedicated to the memory of Prof. Mario Serio. Keywords: Human skeletal muscle cells Inflammation CXCL10 Biomolecular target(s) Immunosuppressants Inflammatory myopathies Th1 polarization Autoimmunity

a b s t r a c t Inflammatory myopathies (IMs) are systemic diseases characterized by a T helper (Th) 1 type inflammatory response and cell infiltrates within skeletal muscles. The mainstay of treatment is drugs aimed at suppressing the immune system – corticosteroids and immunosuppressants. About 25% of patients are non-responders. Skeletal muscle cells seem actively involved in the immune-inflammatory response and not only a target; understanding the molecular bases of IMs might help drug development strategies. Within muscles the interaction between the chemokine interferon (IFN)␥ inducible 10 kDa protein, CXCL10 or IP-10, and its specific receptor CXCR3, present on Th1 type infiltrating cells, likely plays a pivotal role, potentially offering the opportunity for therapeutic intervention. We aimed to clarify the involvement of human skeletal muscle cells in inflammatory processes in terms of CXCL10 secretion, to elucidate the engaged molecular mechanism(s) and, finally, to evaluate muscular cell responses, if any, to some immunosuppressants routinely used in IM treatment, such as methylprednisolone, methotrexate, cyclosporin A and Infliximab. We first isolated and characterized human fetal skeletal muscle cells (Hfsmc), which expressed the specific lineage markers and showed the competence to react in the context of an in vitro alloresponse. CXCL10 protein secretion by Hfsmc was similarly induced by the inflammatory cytokines interferon (IFN)␥ and tumor necrosis factor (TNF)␣, above undetectable control levels, through the activation of Stat1 and NF-kB pathways, respectively; CXCL10 secretion was significantly magnified by cytokine combination, and this synergy was associated to a significant up-regulation of TNF␣RII; cytokine-induced CXCL10 secretion was considerably affected only by Infliximab. Our data suggested that human skeletal muscle cells might actively self-promote muscular inflammation by eliciting CXCL10 secretion, which is known to amplify Th1 cell tissue infiltration in vivo. In conclusion, we sustain that pharmacological targeting of CXCL10 within muscular cells might contribute to keep in control pro-Th1 polarization of the immune/inflammatory response. © 2011 Elsevier GmbH. All rights reserved.

Introduction Inflammatory myopathies (IMs) are chronic autoimmune diseases characterized by decreased muscle endurance and symmetrical muscle weakness (Grundtman et al., 2007). On the basis of their different immunopathologies, clinical features and response to therapies, distinct main subgroups have been categorized,

∗ Corresponding author. Tel.: +39 0636733543; fax: +39 0636733231. E-mail addresses: [email protected], [email protected] (C. Crescioli). 0171-9335/$ – see front matter © 2011 Elsevier GmbH. All rights reserved. doi:10.1016/j.ejcb.2011.09.011

such as idiopathic dermatomyositis, polymyositis, inclusion body myositis, necrotizing autoimmune myositis or myositis associated with systemic disorders (Dalakas, 1991, 2011; Mantegazza et al., 1997). Nevertheless, all IM subtypes share some common features such as inflammation, fibrosis and muscle loss together with T helper (Th) 1 immune reaction predominance and the presence of lymphocyte infiltrate into the damaged muscle (De Paepe et al., 2005, 2007, 2009; Grundtman et al., 2007). The mainstay treatment for IMs are immunomodulating agents specifically designed to target immune cells, such as corticosteroids, often administered with second-line immunosuppressants to reduce the corticosteroid dose and related side effects (Tournadre et al., 2010; Wiendl,

140

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149

2008). Approximately 25% of the patients are non-responders to or cannot tolerate these drugs, continue to experience clinical relapses and are left with disability (Dalakas, 2001; Wiendl, 2008). The need for more specific and safer therapies has been driving the search for a deeper understanding of the pivotal mediators and mechanisms associated with muscular disease also at the molecular level. In IMs, a cascade of inflammatory mediators, such as cytokines, chemokines and adhesion molecules, modulates biological functions and downstream signaling pathways (FigarellaBranger et al., 2003; Szodoray et al., 2010). Several proinflammatory cytokines, i.e. tumor necrosis factor (TNF)␣, interferons (IFNs), interleukines (IL)-1␣ and ␤, have been described as being highly present in IM patients, both in muscle tissues and in serum (De Paepe et al., 2009; Lindberg et al., 1995; Lundberg et al., 1995; Nyberg et al., 2000; Schmidt et al., 2008; Szodoray et al., 2010; Wolf and Baethge, 1990). In particular, IFN␥ and TNF␣, both with strong Th1 association, have been found to be upregulated in IMs (Dalakas, 2001; De Paepe et al., 2009; Lundberg et al., 1995). During inflammation, muscle cells behave as immunoactive cells, secreting molecules of the monocyte-macrophage lineage (i.e. IL-1␣ and ␤, IL-6, TNF␣) and chemokines (De Rossi et al., 2000). The latter ones are a class of small chemotactic cytokines which drives leukocyte migration from blood to inflammation sites (Lazzeri and Romagnani, 2005; Zlotnik and Yoshie, 2000) and amplifies inflammatory responses also in IMs (De Paepe et al., 2008; Tournadre and Miossec, 2009). Besides the known involvement of ␤ (or CC-) chemokines, a role for ␣ (or CXC-) chemokines inducible by IFN␥ has been recently pointed out in IMs (De Paepe et al., 2007, 2009). In particular, IFN␥ inducible 10 kDa protein CXCL10 or IP-10, known to play a predominant role during Th1-mediated responses (Lee et al., 2009; Rotondi et al., 2007), has been recently found raised in IM patient serum (Lee et al., 2009; Szodoray et al., 2010). In IM tissue extracts CXCL10 protein and gene expression have been shown to be significantly increased together with its specific receptor CXCR3, expressed by Th1 infiltrating cells (De Paepe et al., 2005, 2009; Fall et al., 2005; Lee et al., 2009). Indeed, CXCL10-CXCR3 interaction has been hypothesized as a potential target for novel therapeutic interventions (De Paepe et al., 2005, 2007; Lee et al., 2009). Our in vitro investigation aimed to clarify the contribution of human skeletal muscle cells to inflammatory processes in terms of CXCL10 secretion and to elucidate the involved molecular mechanism(s). We also tested human skeletal muscle cell response, if any, to some drugs routinely used in IM treatment – methylprednisolone (MeP), methotrexate (MTX), cyclosporin A (CsA), and the TNF␣ blocking agent Infliximab. To perform this study we first established and characterized in vitro cultures of human fetal skeletal muscle cells (Hfsmc). In Hfsmc we analyzed the response to the prototypic inflammatory stimuli IFN␥ and TNF␣, in terms of CXCL10 protein secretion along with some of the involved molecular pathways, and the effect of the above mentioned immunosuppressants; IL-6 and IL-8 were also tested, since both are generic parameters of muscular damage and inflammation (Dieli-Conwright et al., 2009). We also verified Hfscm competence to react in the context of an in vitro alloresponse.

Materials and methods Chemicals Dulbecco Modified Eagle Medium (DMEM)/Ham’s F-12 medium (1:1) with and without phenol red, RPMI 1640, phosphate buffered saline Ca2+ /Mg2+ -free (PBS), bovine serum albumin (BSA) fraction V, antibiotics, NaOH, absolute ethanol, EDTA–trypsin solution,

Ficoll-Hypaque, Bradford reagent and all reagents for Western blot, methotrexate (MTX), methylprednisolone (MeP), cyclosporin A (CsA) were from Sigma–Aldrich Corp. (St. Louis, MO, USA). The chimeric monoclonal antibody Infliximab (Remicade® ) was from Centocor B.V. (Leiden, The Netherlands). Fetal bovine serum (FBS) and fetal calf serum (FCS) were purchased from Hyclone (Logan, UT, USA). l-Glutamine, nonessential amino acids, pyruvate, and 2-mercaptoethanol were from Gibco Laboratories (Grand Island, NY). The Coomassie Bio-Rad protein assay kit was obtained from Bio-Rad Laboratories, Inc. (Hercules, CA, USA). Collagenase type IV was from Worthington (Lakewood, NJ, USA). IFN␥, TNF␣ and ELISA kits for CXCL10, IFN␥, TNF␣, IL-6 and IL-8 measurement were from R&D Systems (Minneapolis, MN, USA). For flow cytometry analysis, PE-conjugated anti-CD119 monoclonal antibody (mAb) (GIR-208, mouse IgG1) was from BD Biosciences (Mountain View, CA, USA); PE-conjugated anti-TNFRII mAb (22235.311, mouse IgG2a) was from R&D Systems; conjugated isotype-matched control Abs (mouse IgG1: clone 15H6, mouse IgG2a: clone HOPC-1) were from Southern Biotechnology Associated Inc. (Birmingham, AL, USA). For RNA extraction, the RNeasy Mini reagent kit was purchased from Quiagen Italy (Milan, Italy). The TaqMan Reverse Transcription Reagents kit, all primer/probe mixes (Taqman Gene® Expression Assays), CXCL10 (ID number Hs00171042-ml), IFN␥R (ID number Hs00166223-m1), TNF␣RII (ID number Hs00153550-m1), Pax-3 (ID number Hs00240950-m1), Pax-7 (ID number Hs00242962m1), MyoD (ID number Hs00323851-m1), myogenin (ID number Hs01072232-m1) and 1× Universal Master Mix were from Applied Biosystems (Forster City, CA, USA). Quantitative PCR human reference total RNA was purchased from Stratagene (La Jolla, CA, USA). The polyvinylidene difluoride membranes (Hybond-P) were from Amersham Bioscience (Little Chalfont, UK). For Western blot, immunocytochemical and/or immunohistochemical analysis, primary Abs: goat polyclonal Ab (pAb) anti-tropomyosin, mouse anti-␤ actin (␤ actin) mAb, mouse antimyogenin mAb, rabbit anti-myosin heavy chain (MYH) pAb, rabbit anti-NF-kB p65 (NF-kB, C-20) pAb were from Santa Cruz Biotechnology (Santa Cruz, CA, USA); rabbit anti-phospho Tyr701 Stat1 (pStat1) pAb was from Cell Signaling (Danvers, MA, USA); mouse anti-sarcomeric actin mAb (Clone Alpha-Sr-1) was from DakoCytomation (Glostrup, Denmark); rabbit anti-myosin 1␤ pAb was from Sigma–Aldrich; rabbit anti-myostatin pAb was from Chemicon International Inc. (Temecula, CA, USA). Mouse anti-CD3 mAb (clone HIT3a) was from BD Biosciences. Alexa Fluor 488 conjugate goat anti-rabbit and goat anti-mouse Abs were from Molecular Probes (Eugene, OR, USA); FITC conjugate rabbit anti-goat and peroxidase-secondary Abs were from Sigma–Aldrich. The ultravision large volume detection system anti-polyvalent was from Lab-Vision (Fremont, CA, USA). 3 ,3 Diaminobenzidine tetrahydrochloride was from Sigma–Aldrich. All reagents for Sodium Dodecyl Sulphate-PolyAcrylamide Gel Electrophoresis (SDS-PAGE) were from GE Healthcare. Trypan blue was from Euroclone (Pavia, Italy). Plasticware for cell cultures and disposable filtration units for growth media preparation were purchased from Corning (Milan, Italy). Cell cultures and tissues Hfsmc were isolated from 11 fetal skeletal male muscles (four upper and seven lower limbs) obtained after voluntary abortion (10–12 weeks of gestation). Legal abortions were performed in authorized hospitals, and written certificates of consent were obtained. The use of human fetal tissue for research purposes was approved by the Committee for investigation in humans of the Azienda Ospedaliero-Universitaria Careggi, Florence, Italy (protocol n◦ 6783-04). All samples have been handled in the same way and maintained in ice-cold PBS until processed for culture preparation,

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149

as described elsewhere (Crescioli et al., 2008). Cell cultures showed a high degree of myogenic purity (virtually 100% positive cells). Cell expansion (population doublings, PDs) was assessed by counting the cell number after every harvest in a hemocytometer chamber. The formula applied for PD calculation was PD = log 2(Ni /No ), where Ni is the number of cells yielded and No the number of cells plated. Cell viability was assessed by trypan blue exclusion. Confluent cell cultures were split into a 1:2–1:4 ratio using EDTA–trypsin solution (0.2–0.5%), and used within the 10th passage. Adult tissue was obtained from the limb skeletal muscle of an 18 year old male healthy subject after a written certificate of consent; the used procedures were in accordance with the Regional Ethics Committee on human experimentation. Immunocytochemical and immunohistochemical studies For structural protein detection: 104 cells (p3–p7) were seeded on glass coverslips in growth medium and processed after 24 h; cryostat sections, obtained from 12 week old fetal limb skeletal muscles and from an 18 year old male healthy subject, were used for comparison. For Stat1 or NF-kB evaluation, cells, seeded as above, were incubated with serum-free medium overnight, and then treated for 30 min with IFN␥ (1000 U/ml) or TNF␣ (10 ng/ml). Cells in phenol red- and serum-free medium 0.1% BSA and vehicle (Hepes or PBS, 0.05%, vol/vol) were used as control. Immunostaining on cells and tissues was performed as previously described (Crescioli et al., 2008; Fibbi et al., 2009). Briefly, slides were incubated overnight at 4 ◦ C with primary Abs against sarcomeric actin (1:100), myosin 1␤ (1:100), myostatin (1:100), tropomyosin (1:50), NF-kB p65 (1:100) or pStat1 (1:100) and then rinsed in PBS. For immunofluorescence: slides were incubated with Alexa Fluor 488 conjugate goat anti-rabbit and goat anti-mouse (1:200) or FITC conjugate rabbit anti-goat (1:50) secondary Abs. For immunoperoxidase histochemistry: slides were incubated with biotinylated secondary Abs and then with a streptavidin–biotin peroxidase complex (Ultravision large volume detection system anti-polyvalent). The reaction product was developed with 3 ,3 -diaminobenzidine tetrahydrochloride as chromogen. For method specificity slides lacking the primary Abs or stained with the corresponding nonimmune serum were processed. Slides were examined with a phase contrast microscope (Nikon Microphot-FX/FXA microscope, Nikon, Tokyo, Japan). For quantification of pStat1 and NF-kB nuclear staining: the percentage of pStat1 and NF-kB nuclear positive cells was calculated by counting the number of stained cells over the total cells in at least 15 separate fields per slide; experiments were performed five times with different cell preparations. Western blot analysis For protein analysis, Hfsmc (600,000 in 100 mm dishes, p5/6) were maintained in phenol red- and serum-free medium for 24 h, then harvested before any treatment (T0) or incubated for 5–7 days in phenol red- and serum-free medium containing 0.1% BSA (starvation) with or without insulin (100 nM). Cells were then processed as previously described (Crescioli et al., 2008). Briefly, after protein concentration measurement with a Coomassie protein assay kit (Bio-Rad Laboratories), protein aliquots (20 ␮g), processed and loaded onto 10% SDS-PAGE, were transferred on nitrocellulose membranes. Thereafter, membranes were incubated with primary Abs appropriately diluted in Tween Tris-buffered saline (TTBS) (for anti-myogenin and anti-MYH 1:1000; for anti-␤ actin 1:10,000), followed by peroxidase-conjugated secondary IgG (1:3000). Proteins were revealed by the enhanced chemiluminescence system (ECL plus; Amersham Bioscience). Image acquisition and densitometric analysis were performed with Quantity One software on a ChemiDoc XRS instrument (BIO-RAD Labs). Western blot

141

analysis was performed for at least three independent experiments with different cell preparations. To confirm equal protein loading, membranes were stripped (Pierce) and re-probed with the appropriate primary Ab. RNA extraction For mRNA analysis, Hfsmc, plated and maintained as previously described (Crescioli et al., 2008), were either harvested and processed for skeletal muscle specific transcripts (p1 and p6/7) or incubated for 24 h with TNF␣ (10 ng/ml) alone or combined with IFN␥ (1000 U/ml). Cells in phenol red- and serum-free medium containing 0.1% BSA and vehicle (Hepes or PBS, 0.05%, vol/vol) were used as the control. After trypsinization cells were processed as reported elsewhere (Crescioli et al., 2008). Experiments were performed six times with different cell preparations. Total RNA from the cells was extracted with the RNeasy Mini reagent kit according to the manufacturer’s recommendations. RNA concentration and quality were measured by a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). Real time PCR Total RNA (400 ng) was reverse transcribed using the TaqMan Reverse Transcription Reagents kit, measurement of gene expression was performed by quantitative real time PCR (TaqManTM ) and samples were processed as previously reported (Crescioli et al., 2007). The amount of target, normalized to an endogenous reference (GAPDH, pre-developed TaqMan® Assay Reagents) and relative to a calibrator (Quantitative PCR human reference total RNA) was given by 2−Ct calculation (Crescioli et al., 2007). Experiments were performed in three different preparations for each gene expression analysis. Preparation and isolation of PBMCs Buffy coats from healthy adult anonymous donors were obtained in accordance with local ethical committee approval. PBMCs were isolated by centrifugation of heparinized blood on the Ficoll-Hypaque gradient and maintained in RPMI 1640 supplemented with 2 mmol/l l-glutamine, 1% nonessential amino acids, 1% pyruvate, 2 × 10−5 M 2-mercaptoethanol, 10% FCS. Cytokine secretion assays For CXCL10, IL-6, IL-8 secretion assays, 4000 cells/well were seeded in 96-well flat bottom plates and maintained as previously described (Crescioli et al., 2008). After 24 h, different stimuli were added in phenol red- and serum-free medium with 0.1% BSA (200 ␮l/well). Cells in phenol red- and serum-free medium containing 0.1% BSA and vehicle (Hepes or PBS, 0.05%, vol/vol) were used as the control. Cells were incubated for 24 h with TNF␣ at 0.1, 1, 10, 100 and 500 ng/ml or IFN␥ at 10, 100, 1000, 5000 and 10,000 U/ml, alone or combined with 10 ng/ml of TNF␣; or with a combination of IFN␥ (1000 U/ml) + TNF␣ (10 ng/ml), in the absence or in the presence of Infliximab (10 ␮g/ml), MTX (0.4 ␮M), MeP (250 ng/ml) or CsA (250 ng/ml). The supernatant was harvested and kept frozen at −20 ◦ C until performing ELISA assays. Experiments were performed in at least triplicate in three to five different cell preparations. For CXCL10, IFN␥, TNF␣ secretion assays in co-cultures with PBMCs, 5000 Hfsmc/well were seeded in 96-well plates; after 24 h, 100,000 PBMCs/well were seeded alone or co-cultured with adherent Hfsmc in a 1:1 mix of RPMI 1640 with 10% FCS and serumfree (DMEM)/Ham’s F-12 medium with 0.1% BSA (total volume 200 ␮l/well), in the absence or in the presence of anti-CD3 mAb (1 ␮g/ml). After 24 h the supernatant was harvested, centrifuged to

142

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149

remove cells, and stored at −20 ◦ C until performing ELISA assays. Experiments were performed in triplicate in three different cell preparations. ELISA assays CXCL10, IL-6, IL-8, IFN␥ and TNF␣ levels were measured in cell culture supernatants using commercially available kits, according to the manufacturer’s recommendations. The sensitivity ranged from 0.41 to 4.46 pg/ml for CXCL10, less than 0.70 pg/ml for IL6, from 1.5 to 7.5 pg/ml for IL-8, less than 8.0 pg/ml for IFN␥ and from 0.5 to 5.5 pg/ml for TNF␣. The intra- and interassay coefficients of variation were: 3.1% and 6.7% for CXCL10, 4.4% and 3.7% for IL-6, 4.6% and 8.1% for IL-8, 2.6% and 6.4% for IFN␥, 5.3% and 6.8% for TNF␣. Five to nine experiments were performed in triplicate with different cell preparations. Quality control pools of low, normal, or high concentrations for all parameters were included in each assay. The amount of secreted cytokines was expressed as pg/␮g total protein amount or as percent of IFN␥ + TNF␣-induced secretion, as appropriate, in Hfsmc; as pg/ml in PBMCs + Hfsmc cocultures. Protein extraction and measurement to normalize Hfsmc secretion were performed as reported elsewhere (Crescioli et al., 2007). Flow cytometry analysis For flow cytometry analysis, cells were seeded and maintained in the same conditions detailed elsewhere (Crescioli et al., 2007). Cells were stimulated for 24 h with IFN␥ (1000 U/ml) or TNF␣ (10 ng/ml) in phenol red- and serum-free medium with 0.1% BSA. Cells in phenol red- and serum-free medium with 0.1% BSA and vehicle (Hepes and PBS, 0.05%, vol/vol) were used as the control. Flow cytometric analysis was performed as detailed elsewhere (Cosmi et al., 2000). Cells were analyzed on a BDLSRII cytometer, using Diva software (BD Biosciences). The area of positivity was determined using an isotype-matched control mAb. Ten thousand events for each sample were acquired. The FASER technique (Crescioli et al., 2008) has been applied following the manufacturer’s instructions. Experiments were performed four times with five different cell preparations. Results are expressed as fold increase vs. control. Statistical analysis The statistical analysis was performed using the SPSS 12.0 software package (SPSS for Windows 12.0, SPSS Inc, Chicago, IL, USA). The Kolmogorov–Smirnov test was used to test for normal distribution of the data. One-way analysis of variance (ANOVA) was applied. A P value less than 0.05 was considered significant and was corrected for comparisons using the Dunnett’s or Bonferroni’s post hoc test where appropriate. Data were expressed as mean ± SE. Results Characteristics of isolated Hfsmc Immunofluorocytochemistry performed in Hfsmc between passage (p)3–7 showed positive staining for tropomyosin, myosin 1␤, sarcomeric actin – all components of the sarcomere multi-protein complex – and myostatin – considered a negative regulator of muscle growth – virtually in all the cells (Fig. 1A, upper panels). Fetal and adult skeletal muscle tissues were used as positive controls (Fig. 1A, middle and lower panels); negative controls as cells (left upper panel) or fetal tissue slides (left lower panel) stained with the corresponding nonimmune serum are depicted in Fig. 1. The cell culture exhibited a high degree of purity as shown by

Table 1 Gene analysis of myogenic markers. mRNA expression at p6/7 (fold increase vs. p1) Pax-3 Pax-7 MyoD Myogenin

1.55 0.25 1.64 2.77

± ± ± ±

0.41 0.06** 1.41 0.52*

Gene analysis of myogenesis regulator transcription factors in Hfsmc at p1 and p6 revealed that the specific transcript for Pax-7 diminished, while myogenin gene expression increased along with passages, documenting cell maturation; Pax-3 and MyoD mRNA expression did not significantly change. * P < 0.05, p6 vs. p1, taken as 1. ** P < 0.01, p6 vs. p1, taken as 1.

positive staining for all markers and, in particular, for sarcomeric actin virtually in all the cells. Western blot analysis in total cell lysates of Hfsmc at p5/6 showed that 5–7 day-starvation or insulin treatment similarly increased the expression of myosin heavy chain (MYH) and myogenic factor 4 (myogenin) as compared to T0 (before differentiating treatment), when myogenin expression was not detectable (Fig. 1B, upper and middle panels), ␤-actin was used as the loading control (Fig. 1B, lower panel). The Hfsmc lifespan was assessed by continuous passage. Cells showed steadily increasing doubling time, without numerical or structural chromosomal abnormalities within p48/50 (not shown), and finally underwent growth arrest after about 80 population doublings (PDs) (Fig. 1C). Quantitative RT-PCR analysis (Table 1) showed that Hfsmc expressed specific transcripts to Paired box (Pax)-3, Pax-7, myogenic determination factor 1 (MyoD) and myogenin, all skeletal muscle lineage transcription factors, with integrated roles in developmental and adult regenerative myogenesis. Myogenin expression significantly increased at p6/7 vs. p1 (P < 0.05), while Pax-7 significantly decreased (P < 0.01). Effect of proinflammatory cytokines on CXCL10 secretion in Hfsmc Treatment for 24 h with increasing concentrations of IFN␥ (10–1000, 5000 and 10,000 U/ml) or TNF␣ (0.1–100 and 500 ng/ml) dose-dependently increased CXCL10 protein secretion by Hfsmc over undetectable levels in control cells (Fig. 2A and B, P < 0.01 vs. control). Both cytokines evoked a similar response in terms of the CXCL10 protein amount secreted (mean value at highest dose of: IFN␥ 160 ± 23.99 pg/␮g protein; TNF␣ 137.38 ± 45.35 pg/␮g protein). IFN␥-induced protein secretion did not achieve plateau (Fig. 2A; P < 0.05 highest dose vs. preceding doses). Thirty minute stimulation of Hfsmc with IFN␥ (1000 U/ml), or TNF␣ (10 ng/ml), alone (Fig. 2C and D, lower left panels) or combined (Fig. 2C and D, lower right panels), resulted in a compact increase in phosphorylation and nuclear translocation of Stat1 (nuclear staining with: IFN␥ 70.12 ± 2.96%; with IFN␥ + TNF␣: 69.12 ± 3.5% vs. 0.45 ± 0.3% in control, P < 0.01, Fig. 2E) and NFkB (nuclear staining with TNF␣: 68.71 ± 1.88%; with IFN␥ + TNF␣: 63.55 ± 1.79% vs. 0.54 ± 0.38% in control, P < 0.01, Fig. 2F) respectively, as assessed by immunofluorescence microscopy. No effect onto Stat-1 or NF-kB nuclear translocation in cells treated respectively with TNF␣ or IFN␥ (Fig. 2C and D, upper right panels) as compared to control cells (Fig. 2C and D, upper left panels). Synergy between IFN and TNF˛ in Hfsmc The addition of a fixed dose of TNF␣ (10 ng/ml) to scalar doses of IFN␥ (10–1000, 5000 and 10,000 U/ml) amplified CXCL10 protein secretion at each tested dose (Fig. 3A, 2318.98 ± 144.6 pg/␮g protein, mean value at highest dose, P < 0.01 vs. control and vs.

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149

143

Fig. 1. Features, lineage marker expression and cell population doublings (PDs), in Hfscm. (A) All Hfsmc are positive for tropomyosin, myosin 1, sarcomeric actin and myostatin (cell purity virtually 100%). Cryostat sections of fetal and adult tissues were reported for comparison. Slides stained with the corresponding non immune serum were depicted as the negative control. Scale bars: 100 ␮m. (B) As revealed by Western blot analysis, 7 day-starvation (starv.) and insulin treatment (ins.) increased MYH and myogenin protein expression as compared to T0 (before any treatment), documenting cell differentiation; MYH was expressed at T0. ␤-Actin was used as the loading control. (C) Hfsmc lifespan, assessed by continuous passages, showed that cells could be expanded in culture for about 80 PDs in a 5 month period. D, ordinate: number of PDs plotted as a function of time in culture.

IFN␥ alone corresponding dose already depicted in Fig. 2A). As for other cell types (Antonelli et al., 2006b; Crescioli et al., 2007, 2008; Rotondi et al., 2005), the CXCL10 protein secretion plateau in Hfsmc was reached starting from IFN␥ (1000 U/ml) combined with TNF␣ (2229.48 ± 192.9 pg/␮g protein, P < 0.05 vs. the preceding dose); thus, this dose combination has been utilized in the following experiments.

Taqman real time PCR analysis showed that this cytokine combination resulted in about a three-fold increase in CXCL10 specific mRNA, as compared to IFN␥-induced expression (Fig. 3B, P < 0.05). We described in other cell types that the synergy between IFN␥ and TNF␣ was associated with IFN␥R up-regulation induced by TNF␣, without TNF␣RII involvement (Crescioli et al., 2007, 2008).

144

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149

Fig. 2. The effect of IFN␥ and TNF␣ in Hfscm. (A) Increasing concentrations of IFN␥ (10–10,000 U/ml) significantly induced CXCL10 protein secretion (**P < 0.01 vs. control) and did not achieve plateau (16.24 ± 2.16 to 160 ± 23.99 pg/␮g protein, lowest to highest dose, ◦ P < 0.05 vs. the preceding dose). (B) The treatment with increasing concentration of TNF␣ (0.1–100 and 500 ng/ml) significantly enhanced CXCL10 secretion (**P < 0.01 vs. control) to a similar extent as IFN␥ (22.97 ± 4.29 to 137.38 ± 45.35 pg/␮g protein, lowest to highest dose). (C) Nuclear accumulation of phosphorylated Stat1 was observed in Hfsmc stimulated with IFN␥ alone (1000 U/ml, lower left panel) or combined with TNF␣ (10 ng/ml, lower right panel) as compared to control (upper left panel) or to TNF␣ (upper right panel). Scale bars: 100 ␮m. (D) The treatment of Hfsmc with TNF␣ alone (10 ng/ml, lower left panel) or with IFN␥ (1000 U/ml, lower right panel) induced a significant NF-kB p65 subunit translocation from the cytoplasmic to the nuclear compartment, as compared to control (upper left panel). IFN␥ did not exert any effect on NF-kB nuclear translocation (upper right panel). Scale bars: 100 ␮m. (E) Cells were scored as either positive or negative for pStat1 nuclear accumulation. Nuclear stained cells were virtually absent in control cells (0.45 ± 0.3%); after stimulation with IFN␥ or IFN␥ + TNF␣ the percentage of nuclear positive cells was 70.12 ± 2.96% and 69.12 ± 3.5%, respectively (**P < 0.01 vs. control). (F) Cells were scored as either positive or negative for NF-kB nuclear accumulation. The percentage of cell nuclear staining after stimulation with TNF␣ or IFN␥ + TNF␣ was 68.71 ± 1.88% and 63.55 ± 1.79%, respectively (**P < 0.01 vs. 0.54 ± 0.38% in control cells). A, B ordinate: CXCL10 secretion expressed as pg/␮g protein. E, F ordinate: pStat1 or NF-kB nuclear positive cells expressed as the percentage of positive cells over the total.

We performed flow cytometry and mRNA analysis of the IFN␥R and TNF␣RII in Hfsmc stimulated with TNF␣ (10 ng/ml) or IFN␥ (1000 U/ml). IFN␥R membrane protein expression did not significantly change with TNF␣ as compared to control cells, and it was significantly reduced by IFN␥ (Fig. 3C, P < 0.05 vs. control). Gene analysis showed that IFN␥R is modulated by both cytokines although with opposite effects (Fig. 3D, P < 0.05 vs. control). The expression of TNF␣RII membrane protein (Fig. 3E, P < 0.05 vs. control) and of its specific transcript (Fig. 3F, P < 0.01 or P < 0.05 vs. control) significantly increased both with IFN␥ and TNF␣.

CXCL10 secretion in Hfsmc co-cultured with PBMCs To evaluate whether CXCL10 secretion occurs in the context of an in vitro alloresponse we co-cultured Hfsmc with PBMCs, with or without anti-CD3 mAb (1 ␮g/ml). In the absence of anti-CD3 mAb, CXCL10 protein secretion was virtually absent in both control cell types, alone or cocultured (Fig. 4A). The addition of anti-CD3 mAb triggered CXCL10 secretion in PBMCs alone (161.35 ± 47.15 pg/ml, P < 0.05 vs. control). The addition of Hfsmc to anti-CD3-treated PBMCs notably enhanced CXCL10 protein secretion (769.56 ± 109.5 pg/ml, P < 0.05

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149

145

Fig. 3. Synergy between TNF␣ and IFN␥. (A) A fixed dose of TNF␣ (10 ng/ml) amplified CXCL10 secretion induced by IFN␥ (10–10,000 U/ml) (367.24 ± 20.9 to 2318.98 ± 144.6 pg/␮g protein, lowest to highest dose; **P < 0.01 vs. control cells, ◦ P < 0.05 vs. the preceding dose, #P < 0.05 vs. IFN␥ alone corresponding dose). (B) CXCL10 mRNA expression significantly increased in the presence of IFN␥ (1000 U/ml) or TNF␣ (10 ng/ml); cytokine combination significantly magnified single cytokine-induced CXCL10 gene expression (*P < 0.05 vs. control, taken as 1, # P < 0.05 vs. IFN␥- or TNF␣-induced expression). (C) Flow cytometry analysis revealed that IFN␥ (1000 U/ml) significantly reduced IFN␥R membrane protein expression (*P < 0.05 vs. control, taken as 1), while TNF␣ (10 ng/ml) did not exert any significant effect. (D) IFN␥R mRNA expression was significantly reduced by IFN␥ (1000 U/ml) and significantly increased by TNF␣ (10 ng/ml) (*P < 0.05 vs. control, taken as 1). (E) TNF␣RII membrane protein expression was significantly up-regulated both by IFN␥ (1000 U/ml) and TNF␣ (10 ng/ml) (*P < 0.05 vs. control, taken as 1). (F) The treatment with IFN␥ (1000 U/ml) and TNF␣ (10 ng/ml) significantly increased TNF␣RII mRNA expression (**P < 0.01, *P < 0.05 vs. control, taken as 1). A, ordinate: CXCL10 secretion expressed as pg/␮g protein. B, ordinate: CXCL10 mRNA expression depicted as fold increase vs. control, taken as 1. C–F, ordinate: IFN␥R or TNF␣RII protein or mRNA expression depicted as fold increase vs. control, taken as 1.

vs. control PBMCs, vs. control Hfsmc + PBMCs, vs. anti-CD3-treated PBMCs). As we measured IFN␥ and TNF␣ secretion in PBMCs, we observed that both cytokine secretion was significantly induced by anti-CD3 mAb (1 ␮g/ml) as compared to control cells (2473.04 ± 11.05 pg/ml and 1397.6 ± 147.4 pg/ml, respectively, P < 0.05 vs. their control, Fig. 4B and C). Only TNF␣ was detectable in control PBMCs (526.69 ± 258.8 pg/ml, Fig. 4C). As anti-CD3-activated PBMCs were co-cultured with Hfsmc, the TNF␣ level in the supernatant significantly decreased

(508.52 ± 111.3 pg/ml, P < 0.05 vs. anti-CD3-activated PBMCs, Fig. 4C) whereas the IFN␥ secreted level remained unchanged (2376.10 ± 289.70 pg/ml, Fig. 4B). Response of Hfsmc to immunosuppressants To investigate the response of cytokine-stimulated Hfsmc to immunosuppressive drugs, cells were simultaneously incubated with IFN␥ (1000 U/ml) + TNF␣ (10 ng/ml) with or without Infliximab (10 ␮g/ml), MTX (0.4 ␮M), MeP (250 ng/ml) and CsA

146

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149

(250 ng/ml). The drug concentrations were selected on the basis of their near therapy doses, according to their pharmacokinetics (Cmax and area under the time-concentration curve, AUC). Cytokine-induced CXCL10 secretion (Fig. 5A) was inhibited to the highest degree by Infliximab (inhibition: 57%, P < 0.01 vs. IFN␥ + TNF␣-treated cells, expressed as 100%, and 0.05 vs. each other drug treatment); little inhibition, although still statistically significant, by MTX, MeP and CsA occurred (inhibition range 21–30%, P < 0.05 vs. IFN␥ + TNF␣-treated cells). We also tested IL-8 and IL-6, both indices of skeletal muscle inflammation, in the same experimental conditions. Treatment of Hfsmc with pro-inflammatory cytokines significantly increased both IL-8 and IL-6 secretion over the control (P < 0.01, not shown). The simultaneous presence of MTX, MeP and CsA reduced the IL-8 level more than 40% (inhibition range: 42–50% Fig. 5B, P < 0.05 vs. IFN␥ + TNF␣-treated cells, expressed as 100%); Infliximab almost abrogated IL-8 secretion (inhibition: 89%, P < 0.01 vs. IFN␥ + TNF␣-treated cells, expressed as 100%, P < 0.01 vs. every other drug treatment). Cytokine-induced IL-6 secretion, taken as 100%, was similarly inhibited by Infliximab and MeP (Fig. 5C, inhibition range: 86–90%, P < 0.01 vs. IFN␥ + TNF␣-treated cells, P < 0.01 vs. MTX); MTX also significantly reduced IL-6 secretion (inhibition: 37%, P < 0.05 vs. IFN␥ + TNF␣-treated cells), while CsA seemed to exert no effect.

Discussion

Fig. 4. In vitro alloresponse in Hfsmc co-cultured with PBMCs. (A) CXCL10 protein secretion was absent in control Hfsmc and PBMCs, alone or co-cultured. Anti-CD3 mAb (1 ␮g/ml) significantly induced CXCL10 protein secretion by PBMCs (161.35 ± 47.15 pg/ml, ◦ P < 0.05, *P < 0.05 vs. control PBMCs and Hfsmc, respectively); co-culture of anti-CD3-activated PBMCs with Hfsmc significantly enhanced CXCL10 protein secretion (769.56 ± 109.5 pg/ml, # P < 0.05 vs. anti-CD3-activated PBMCs and Hfsmc; ◦ P < 0.05, *P < 0.05 vs. control PBMCs and Hfsmc, respectively). (B) IFN␥ protein secretion, undetectable in both control cell types, was significantly induced by anti-CD3 mAb (1 ␮g/ml) in PBMCs (2473.04 ± 11.05 pg/ml, ◦ P < 0.05 vs. control PBMCs); co-culture of PBMCs with Hfsmc did not exert any further significant effect on this secretion (2376.10 ± 289.70 pg/ml). (C) TNF␣ protein secretion, present only in control PBMCs (526.69 ± 258.8 pg/ml), was significantly increased by anti-CD3 mAb (1 ␮g/ml) in PBMCs (1397.6 ± 147.4 pg/ml, ◦ P < 0.05, *P < 0.05 vs. control PBMCs and Hfsmc, respectively); co-culture with Hfsmc significantly reduced TNF␣ secretion by anti-CD3-activated PBMCs (508.52 ± 111.3 pg/ml, # P < 0.05 vs. anti-CD3-activated PBMCs, *P < 0.05 vs. control Hfsmc).

IMs are Th1-driven autoimmune processes characterized by significant inflammatory cell infiltrates and ending in muscle injury (De Paepe et al., 2005, 2007, 2009; Grundtman et al., 2007). About 25% of patients cannot tolerate or are unresponsive to the classic therapies (Dalakas, 2001; Wiendl, 2008) and there are not yet defined guidelines or best treatment protocols agreed on internationally (Tournadre et al., 2010; Wiendl, 2008). Skeletal muscles cells actively contribute to the inflammatory response and should be considered dynamic participants rather than passive targets of immune reaction (De Rossi et al., 2000; Wiendl et al., 2005). A better decode of the molecular events associated with inflammation at the muscular level may help to identify new potential biomolecular target(s) for pharmacological intervention. The majority of in vitro studies in human skeletal muscle cells refers to myogenic satellite cells derived from muscle biopsies of IM patients (Wiendl et al., 2005). The usefulness of stem cells seems limited for understanding the immunobiology of human skeletal muscle diseases (Wiendl et al., 2005) since those cells respond to various stimuli, such as cytokines, differently from mature cells. Indeed, “mature” myoblasts which orchestrate muscle repair, represent the cell population pivotal in immune processes of myositis (Schwab et al., 2008). To perform this investigation we first established in vitro cultures of human skeletal muscle cells, Hfsmc, on the basis of our previous experience with human fetal striatal cells (cardiomyocytes) (Crescioli et al., 2008). Hfsmc expressed motor and structural proteins typical of a mature cell phenotype, such as tropomyosin, myosin 1␤, sarcomeric actin and myostatin. In serum deprivation, Hfsmc spontaneously shifted towards myotubes, as documented by myogenin and MYH increased protein expression (Andrès and Walsh, 1996). In compliance with the cell maturation stage, myogenin gene expression significantly increased following cell passages, with a simultaneous significant decrease in gene expression of Pax 7 – a transcription factor required for myogenic commitment, together with Pax 3 (Brzóska et al., 2009).

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149

Fig. 5. The response of Hfsmc to immunomodulatory drugs. (A) Infliximab (10 ␮g/ml) reduced to the highest degree cytokine-induced CXCL10 secretion by Hfsmc (**P < 0.01 vs. IFN␥ + TNF␣ treated cells, # P < 0.05 vs. every other drug treatment), while MTX (0.4 M), MeP (250 ng/ml) and CsA (250 ng/ml) exerted a little, although still statistically significant, effect (*P < 0.05 vs. IFN␥ + TNF␣-treated cells). (B) Cytokine-induced IL-8 secretion was significantly decreased by all the tested drugs; the maximal inhibitory effect was exerted by Infliximab (**P < 0.01, *P < 0.05 vs. IFN␥ + TNF␣-treated cells; Infliximab: ## P < 0.01 vs. every other drug treatment). (C) Cytokine-induced IL-6 secretion was similarly inhibited by Infliximab and MeP (**P < 0.01 vs. IFN␥ + TNF␣-treated cells, ## P < 0.01 vs. MTX); MTX-induced IL-6 reduction was weaker although still statistically significant (*P < 0.05 vs. IFN␥ + TNF␣-treated cells); CsA exerted no significant effect. A–C, ordinate: CXCL10, IL-8 or IL-6 secretion expressed as percent variation over the maximal IFN␥ + TNF␣induced secretion.

147

Gene expression of MyoD, a regulatory factor necessary for cell differentiation and myotube maturation (Cao et al., 2006), has been detected. A Pax 7/MyoD/Myog positive cell population has been reported as a transitional stage within recently differentiated myoblasts and, occasionally, in newly formed myotubes as well (Yablonka-Reuveni et al., 2008). Hence, we documented first that Hfsmc spontaneously retained the functional competence of phenotypically mature cells along with proliferation capability, albeit limited, as shown by their lifespan. Next, we showed that Hfsmc have the capability to react in the context of an alloresponse, although fetal cells are reported to display a reduced immunogenicity (Dekel and Reisner, 2004; Foglia et al., 1986). Current findings suggest that skeletal muscle cells during inflammation are actively involved in the immune response secreting immunoactive molecules under the paracrine or autocrine effect of cytokines released by endothelial cells, infiltrating lymphocytes or muscle cells themselves (De Paepe et al., 2009; De Rossi et al., 2000; Loell and Lundberg, 2011; Nielsen and Pedersen, 2008). We documented the ability of Hfsmc to secrete CXCL10, virtually absent in control cells, in the context of an in vitro alloresponse or after the addition of exogenous IFN␥ and TNF␣; those latter ones enhanced IL-8 and IL-6 production, both constitutively secreted (not shown). Thus far, Hfsmc are reliable and reproducible cultures, and represent in our opinion a useful tool to focus on some molecular pathways engaged in the immune-mediated inflammatory response of human skeletal muscles. Consistency between experiments with a fetal cell line and the change observed in adult biopsies has been previously reported (Nagaraju et al., 1998). Anyhow, we also verified that the inflammatory response induced by IFN␥ and TNF␣ was similar between human myoblasts and myotubes, either fetal or adult (personal observation). During muscle inflammation there is an upregulation of several proinflammatory chemokines, such as CC (␤)- or CXC (␣)-chemokines, and cytokines, i.e. IL-1␤ or TNF␣, as extensively documented (De Paepe et al., 2005, 2007, 2008, 2009; De Rossi et al., 2000; Schmidt et al., 2008; Szodoray et al., 2010; Tournadre and Miossec, 2009). Out of the others, we focused on CXCL10/IP10 – functionally categorized as the Th1 chemokine – since it is known to critically drive early T cell responses towards Th1 polarization and to modify endothelial and smooth muscle vascular cell functions in several allo and autoimmune or chronic inflammatory processes (Antonelli et al., 2006a; Crescioli et al., 2007, 2008; Hancock et al., 2001; Lande et al., 2004; Lazzeri and Romagnani, 2005; Romagnani et al., 2002, 2004; Rotondi et al., 2007; Sagrinati et al., 2010; Wang et al., 1996). Notably, CXCL10 seems responsible for initiating the immune response following antigenic challenge (Campbell et al., 2004). CXCL10 has been shown to play a pivotal role in the Th1-driven immune response in myopathies (De Paepe et al., 2008, 2009; Fall et al., 2005; Feferman et al., 2005; Lee et al., 2009) as well; in particular it seems involved in the early inflammatory signals (Raju et al., 2003). Therefore, a deeper understanding of the molecular mechanisms underlying CXCL10 action(s) could help to keep in control its downstream effect(s). So far, CXCL10 has been described as being triggered mainly or exclusively by IFN␥ in cultured myotubes, myoblasts and in other cell types (Antonelli et al., 2006b; Crescioli et al., 2007, 2008; Raju et al., 2003; Rotondi et al., 2005). We demonstrated that in Hfsmc exogenous TNF␣ elicited CXCL10 protein secretion to a similar extent as IFN␥. Each cytokine promoted its specific pathway activation, throughout NF-kB or Stat1 nuclear translocation, respectively. We observed that the simultaneous presence of both cytokines exerted a synergistic effect on CXCL10 secretion, in association with

148

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149

a significant upregulation of TNF␣RII, the receptor subtype mainly involved in immune response regulation (Hehlgans and Pfeffer, 2005; Suvannavejh et al., 2000). We previously showed that synergy between IFN␥ and TNF␣ engaged, in other cell types, only an IFN␥R protein expression increase to potentiate the IFN␥-induced inflammatory response, without TNF␣R involvement (Crescioli et al., 2007, 2008). In Hfsmc, IFN␥ and TNF␣ synergy appears to be exerted essentially throughout TNF␣RII upregulation, which likely accounts for the raised magnitude of the cell response to TNF␣, in terms of CXCL10 secretion. In this light, we can speculate that the TNF␣ decreased level in the co-culture supernatant may be due to cytokine capture by TNF␣Rs of Hfsmc. Hence, in Hfsmc TNF␣ seems the pivotal cytokine in promoting muscular inflammation at the cellular level, although the Th1 polarized response has been almost exclusively referred to as an IFN␥-driven one (Figarella-Branger et al., 2003; Frasca et al., 2008). Indirect evidence is that TNF␣ blockade with Infliximab exerted the strongest inhibition on cytokine-induced CXCL10 protein secretion by Hfsmc, although the effect of the other drugs still exhibited statistical significance. In fact, TNF␣ has been widely described as being elevated in the blood and tissues of IM patients (Lindberg et al., 1995; Szodoray et al., 2010), while IFN␥ has been often questioned as very little of its expression has been reported in muscle tissues of myositis patients (Lundberg et al., 1995). However, a possible explanation of IFN␥ restricted expression may be that this cytokine is rapidly degraded. Our data, while confirming the critical role of TNF␣ in IM pathogenesis (Lundberg et al., 1995; Tournadre and Miossec, 2009) added, in our opinion, new insights into the molecular pathways engaged in CXCL10 secretion by skeletal muscle cells, highlighting potential molecular targets for therapeutic intervention at the muscular level. Although immunosuppressants specifically target immune cells, we previously reported that some immunomodulators impaired CXCL10 secretion also by organ resident cells (Crescioli et al., 2007, 2008; Sagrinati et al., 2010). In Hfsmc none of the current immunosuppressants caused a really consistent – although significant – impairment of cytokine-induced CXCL10 or IL-6 and IL-8 secretion, except for Infliximab or MeP, the latter effective only on IL-6. Once more, this result drives the attention on the actual need for a therapeutic intervention not only at the systemic but also at the tissue/cellular level. Given the complexity of the cytokine network involved in IM pathogenesis, our study is limited, since it only deals with a narrow view linked to our in vitro observations. Focusing just on CXCL10 could seem undoubtedly restrictive, however, we could hypothesize that interfering also with local tissue chemokine production might be helpful to control inflammatory processes in IM. In particular, neutralizing the in vivo activity of CXCL10 can alter the Th1/Th2 balance due, at least in part, to a direct effect of CXCL10 on T-cell polarization towards the Th1 cell subset (Salomon et al., 2002). Accordingly, targeted CXCL10 vaccines have been intended to avoid pro-Th1 cell polarization, as in the case of trials using an anti-CXCL10 human mAb in rheumatoid arthritis or ulcerative colitis patients (see ClinicalTrials.gov identifier: NCT01017367; NCT00295282; NCT00656890). In conclusion, we can speculate that human skeletal muscle cells immunoactively contribute to Th1 immune cascade in vivo throughout local CXCL10 secretion, which likely elicits and perpetuates a self-promoting inflammatory loop. Ongoing in vivo and further in vitro studies would hopefully help us to deepen our understanding of molecular mechanisms underlying IM, to better clarify the chemokine/cytokine cross-talk between

immune and tissue cells, and, finally, to provide the basis for rational drug development to aid IM patient management. Conflict of interest The authors declare that no conflict of interest has prejudiced the impartiality of this research. Acknowledgements This research was supported by MIUR (Ministero dell’Istruzione, dell’Università e della Ricerca). The authors would like to thank Dr. Andrea Matucci, Dr. Alessandra Vultaggio (Department of Biomedicine, Immunoallergology Unit, Azienda Ospedaliera Universitaria Careggi, Florence, Italy) and Dr. Marta Nesi (Department of Health Sciences, University of Rome “Foro Italico”, Italy) for their technical support. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ejcb.2011.09.011. References Andrès, V., Walsh, K., 1996. Myogenin expression, cell cycle withdrawal, and phenotypic differentiation are temporally separable events that precede cell fusion upon myogenesis. J. Cell Biol. 132, 657–666. Antonelli, A., Fallahi, P., Rotondi, M., Ferrari, S.M., Romagnani, P., Grosso, M., Ferrannini, E., Serio, M., 2006a. Increased serum CXCL10 in Graves’ disease or autoimmune thyroiditis is associated with hyper- or hypothyroidism per se, but is specifically sustained by the autoimmune, inflammatory process. Eur. J. Endocrinol. 154, 651–658. Antonelli, A., Rotondi, M., Ferrari, S.M., Fallahi, P., Romagnani, P., Franceschini, S.S., Serio, M., Ferrannini, E., 2006b. Interferon-gamma-inducible alpha-chemokine CXCL10 involvement in Graves’ ophthalmopathy: modulation by peroxisome proliferator-activated receptor-gamma agonists. J. Clin. Endocrinol. Metab. 91, 614–620. ´ ´ Brzóska, E., Przewozniak, M., Grabowska, I., Janczyk-Ilach, K., Moraczewski, J., 2009. Pax3 and Pax7 expression during myoblast differentiation in vitro and fast and slow muscle regeneration in vivo. Cell Biol. Int. 33, 483–492. Campbell, J.D., Gangur, V., Simons, F.E., HayGlass, K.T., 2004. Allergic humans are hypo-responsive to a CXCR3 ligand-mediated Th1 immunity-promoting loop. FASEB J. 18, 329–331. Cao, Y., Kumar, R., Penn, B., Berkes, C.A., Kooperberg, C., Boyer, L.A., Young, R.A., Tapscott, S.J., 2006. Global and gene-specific analyses show distinct roles for Myod and Myog at a common set of promoters. EMBO J. 25, 502–511. Cosmi, L., Annunziato, F., Galli, G., Maggi, E., Nagata, K., Romagnani, S., 2000. CRTH2 is the most reliable marker for the detection of circulating human type 2 Th and type 2 T cytotoxic cells in health and disease. Eur. J. Immunol. 30, 2972–2979. Crescioli, C., Cosmi, L., Borgogni, E., Santarlasci, V., Gelmini, S., Sottili, M., Sarchielli, E., Mazzinghi, B., Francalanci, M., Pezzatini, A., Perigli, G., Vannelli, G.B., Annunziato, F., Serio, M., 2007. Methimazole inhibits CXC chemokine ligand 10 secretion in human thyrocytes. J. Endocrinol. 195, 145–155. Crescioli, C., Squecco, R., Cosmi, L., Sottili, M., Gelmini, S., Borgogni, E., Sarchielli, E., Scolletta, S., Francini, F., Annunziato, F., Vannelli, G.B., Serio, M., 2008. Immunosuppression in cardiac graft rejection: a human in vitro model to study the potential use of new immunomodulatory drugs. Exp. Cell Res. 314, 1337–1350. Dalakas, M.C., 1991. Polymyositis, dermatomyositis and inclusion-body myositis. N. Engl. J. Med. 325, 1487–1498. Dalakas, M.C., 2001. The molecular and cellular pathology of inflammatory muscle diseases. Curr. Opin. Pharmacol. 1, 300–306. Dalakas, M.C., 2011. Pathophysiology of inflammatory and autoimmune myopathies. Presse Med. 40, e237–e247. De Paepe, B., De Keyzer, K., Martin, J.J., De Bleecker, J.L., 2005. Alpha-chemokine receptors CXCR1-3 and their ligands in idiopathic inflammatory myopathies. Acta Neuropathol. 109, 576–582. De Paepe, B., Creus, K.K., De Bleecker, J.L., 2007. Chemokine profile of different inflammatory myopathies reflects humoral versus cytotoxic immune responses. Ann. N. Y. Acad. Sci. 1109, 441–453. De Paepe, B., Creus, K.K., De Bleecker, J.L., 2008. Chemokines in idiopathic inflammatory myopathies. Front. Biosci. 13, 2548–2577. De Paepe, B., Creus, K.K., De Bleecker, J.L., 2009. Role of cytokines and chemokines in idiopathic inflammatory myopathies. Curr. Opin. Rheumatol. 21, 610–616. De Rossi, M., Bernasconi, P., Baggi, F., de Waal Malefyt, R., Mantegazza, R., 2000. Cytokines and chemokines are both expressed by human myoblasts: possible

C. Crescioli et al. / European Journal of Cell Biology 91 (2012) 139–149 relevance for the immune pathogenesis of muscle inflammation. Int. Immunol. 12, 1329–1335. Dekel, B., Reisner, Y., 2004. Embryonic committed stem cells as a solution to kidney donor shortage. Expert Opin. Biol. Ther. 4, 443–454. Dieli-Conwright, C.M., Spektor, T.M., Rice, J.C., Schroeder, E.T., 2009. Hormone therapy attenuates exercise-induced skeletal muscle damage in postmenopausal women. J. Appl. Physiol. 107, 853–858. Fall, N., Bove, K.E., Stringer, K., Lovell, D.J., Brunner, H.I., Weiss, J., Higgins, G.C., Bowyer, S.L., Graham, T.B., Thornton, S., Grom, A.A., 2005. Association between lack of angiogenic response in muscle tissue and high expression of angiostatic ELR-negative CXC chemokines in patients with juvenile dermatomyositis: possible link to vasculopathy. Arthritis Rheum. 52, 3175–3180. Feferman, T., Maiti, P.K., Berrih-Aknin, S., Bismuth, J., Bidault, J., Fuchs, S., Souroujon, M.C., 2005. Overexpression of IFN-induced protein 10 and its receptor CXCR3 in myasthenia gravis. J. Immunol. 174, 5324–5331. Fibbi, B., Filippi, S., Morelli, A., Vignozzi, L., Silvestrini, E., Chavalmane, A., De Vita, G., Marini, M., Gacci, M., Manieri, C., Vannelli, G.B., Maggi, M., 2009. Estrogens regulate humans and rabbit epididymal contractility through the RhoA/Rhokinase pathway. J. Sex. Med. 6, 2173–2186. Figarella-Branger, D., Civatte, M., Bartoli, C., Pellissier, J.F., 2003. Cytokines, chemokines, and cell adhesion molecules in inflammatory myopathies. Muscle Nerve 28, 659–682. Foglia, R.P., DiPreta, J., Statter, M.B., Donahoe, P.K., 1986. Fetal allograft survival in immunocompetent recipients is age dependent and organ specific. Ann. Surg. 204, 402–410. Frasca, L., Nasso, M., Spensieri, F., Fedele, G., Palazzo, R., Malavasi, F., Ausiello, C.M., 2008. IFN-gamma arms human dendritic cells to perform multiple effector functions. J. Immunol. 180, 1471–1481. Grundtman, C., Malmstrom, V., Lundberg, I.E., 2007. Immune mechanisms in the pathogenesis of idiopathic inflammatory myopathies. Arthritis Res. Ther. 9, 208–220. Hancock, W.W., Gao, W., Csizmadia, V., Faia, K.L., Shemmeri, N., Luster, A.D., 2001. Donor-derived IP-10 initiates development of acute allograft rejection. J. Exp. Med. 193, 975–980. Hehlgans, T., Pfeffer, K., 2005. The intriguing biology of the tumour necrosis factor/tumour necrosis factor receptor superfamily: players, rules and the games. Immunology 115, 1–20. Lande, R., Giacomini, E., Serafini, B., Rosicarelli, B., Sebastiani, G.D., Minisola, G., Tarantino, U., Riccieri, V., Valesini, G., Coccia, E.M., 2004. Characterization and recruitment of plasmacytoid dendritic cells in synovial fluid and tissue of patients with chronic inflammatory arthritis. J. Immunol. 173, 2815–2824. Lazzeri, E., Romagnani, P., 2005. CXCR3-binding chemokines: novel multifunctional therapeutic targets. Curr. Drug Targets Immune Endocr. Metabol. Disord. 5, 109–118. Lee, E.Y., Lee, Z.H., Song, Y.W., 2009. CXCL10 and autoimmune diseases. Autoimmun. Rev. 8, 379–383. Lindberg, C., Oldfors, A., Tarkowski, A., 1995. Local T-cell proliferation and differentiation in inflammatory myopathies. Scand. J. Immunol. 41, 421–426. Loell, I., Lundberg, I.E., 2011. Can muscle regeneration fail in chronic inflammation: a weakness in inflammatory myopathies? J. Intern. Med. 269, 243–257. Lundberg, I., Brengman, J.M., Engel, A.G., 1995. Analysis of cytokine expression in muscle in inflammatory myopathies, Duchenne dystrophy, and non-weak controls. J. Neuroimmunol. 63, 9–16. Mantegazza, R., Bernasconi, P., Confalonieri, P., Cornelio, F., 1997. Inflammatory myopathies and systemic disorders: a review of immunopathogenetic mechanisms and clinical features. J. Neurol. 244, 277–287. Nagaraju, K., Raben, N., Merritt, G., Loeffler, L., Kirk, K., 1998. A variety of cytokines and immunologically relevant surface are expressed by normal human skeletal muscle cells under proinflammatory stimuli. Clin. Exp. Immunol. 113, 407–414. Nielsen, S., Pedersen, B.K., 2008. Skeletal muscle as an immunogenic organ. Curr. Opin. Pharmacol. 8, 346–351.

149

Nyberg, P., Wikman, A.L., Nennesmo, I., Lundberg, I., 2000. Increased expression of interleukin 1alpha and MHC class I in muscle tissue of patients with chronic, inactive polymyositis and dermatomyositis. J. Rheumatol. 27, 940–948. Raju, R., Vasconcelos, O., Granger, R., Dalakas, M.C., 2003. Expression of IFNgamma-inducible chemokines in inclusion body myositis. J. Neuroimmunol. 141, 125–131. Romagnani, P., Lasagni, L., Annunziato, F., Serio, M., Romagnani, S., 2004. CXC chemokines the regulatory link between inflammation and angiogenesis. Trends Immunol. 25, 201–209. Romagnani, P., Rotondi, M., Lazzeri, E., Lasagni, L., Francalanci, M., Buonamano, A., Milani, S., Vitti, P., Chiovato, L., Tonacchera, M., Bellastella, A., Serio, M., 2002. Expression of IP-10/CXCL10 and MIG/CXCL9 in the thyroid and increased serum levels of IP-10/CXCL10 in the serum of subjects with recent onset Graves’disease. Am. J. Physiol. 161, 195–206. Rotondi, M., Falorni, A., De Bellis, A., Laureti, S., Ferruzzi, P., Romagnani, P., Buonamano, A., Lazzeri, E., Crescioli, C., Mannelli, M., Santeusanio, F., Bellastella, A., Serio, M., 2005. Elevated serum interferon-gamma-inducible chemokine10/CXC chemokine ligand-10 in autoimmune primary adrenal insufficiency and in vitro expression in human adrenal cells primary cultures after stimulation with proinflammatory cytokines. J. Clin. Endocrinol. Metab. 90, 2357–2363. Rotondi, M., Chiovato, L., Romagnani, S., Serio, M., Romagnani, P., 2007. Role of chemokines in endocrine autoimmune diseases. Endocr. Rev. 28, 492–520. Sagrinati, C., Sottili, M., Mazzinghi, B., Borgogni, E., Adorini, L., Serio, M., Romagnani, P., Crescioli, C., 2010. Comparison between VDR analogs and current immunosuppressive drugs in relation to CXCL10 secretion by human renal tubular cells. Transpl. Int. 23, 914–923. Salomon, I., Netzer, N., Wildbaum, G., Schif-Zuck, S., Maor, G., Karin, N., 2002. Targeting the function of IFN-␥-inducible protein 10 suppresses ongoing adjuvant arthritis. J. Immunol. 169, 2685–2693. Schmidt, J., Barthel, K., Wrede, A., Salajegheh, M., Bähr, M., Dalakas, M.C., 2008. Interrelation of inflammation and APP in sIBM: IL-1 beta induces accumulation of beta-amyloid in skeletal muscle. Brain 131, 1228–1240. Schwab, N., Waschbisch, A., Wrobel, B., Lochmüller, H., Sommer, C., Wiendl, H., 2008. Human myoblasts modulate the function of antigen-presenting cells. J. Neuroimmunol. 200, 62–70. Suvannavejh, G.C., Lee, H.O., Padilla, J., Dal Canto, M.C., Barrett, T.A., Miller, S.D., 2000. Divergent roles for p55 and p75 tumor necrosis factor receptors in the pathogenesis of MOG(35-55)-induced experimental autoimmune encephalomyelitis. Cell. Immunol. 205, 24–33. Szodoray, P., Alex, P., Knowlton, N., Centola, M., Dozmorov, I., Csipo, I., Nagy, A.T., Constantin, T., Ponyi, A., Nakken, B., Danko, K., 2010. Idiopathic inflammatory myopathies, signified by distinctive peripheral cytokines, chemokines and the TNF family members B-cell activating factor and a proliferation inducing ligand. Rheumatology (Oxford, U.K.) 49, 1867–1877. Tournadre, A., Miossec, P., 2009. Chemokines and dendritic cells in inflammatory myopathies, Ann. Rheum. Dis. 68, 300–304. Tournadre, A., Dubost, J.J., Soubrier, M., 2010. Treatment of inflammatory muscle disease in adults. Joint Bone Spine 77, 390–394. Wang, X., Yue, T.L., Ohlstein, E.H., Sung, C.P., Feuerstein, G.Z., 1996. Interferon␥inducible protein-10 involves vascular smooth muscle cell migration, proliferation, and inflammatory response. J. Biol. Chem. 271, 24286–24293. Wiendl, H., Hohlfeld, R., Kieseier, B.C., 2005. Immunobiology of muscle: advances in understanding an immunological microenvironment. Trends Immunol. 26, 373–380. Wiendl, H., 2008. Idiopathic inflammatory myopathies: current and future therapeutic options. Neurotherapeutics 5, 548–557. Wolf, R.E., Baethge, B.A., 1990. Interleukin-1 alpha, interleukin-2, and soluble interleukin-2 receptors in polymyositis. Arthritis Rheum. 33, 1007–1014. Yablonka-Reuveni, Z., Day, K., Vine, A., Shefer, G., 2008. Defining the transcriptional signature of skeletal muscle stem cells. J. Anim. Sci. 86, E207–E216. Zlotnik, A., Yoshie, O., 2000. Chemokines: a new classification system and their role in immunity. Immunity 12, 121–127.