Idiopathic inflammatory myopathy is a chronic inflammatory muscle disease characterized by mononuclear cell infiltration in the skeletal muscle. The infiltrated inflammatory cells express various cytokines and cytotoxic molecules. Chemokines are thought to contribute to the inflammatory cell migration into the muscle. We induced experimental autoimmune myositis (EAM) in SJL/J mice by immunization with rabbit myosin and CFA. In the affected muscles of EAM mice, CX3CL1 (fractalkine) was expressed on the infiltrated mononuclear cells and endothelial cells, and its corresponding receptor, CX3CR1, was expressed on the infiltrated CD4 and CD8 T cells and macrophages. Treatment of EAM mice with anti-CX3CL1 mAb significantly reduced the histopathological myositis score, the number of necrotic muscle fibers, and infiltration of CD4 and CD8 T cells and macrophages. Furthermore, treatment with anti-CX3CL1 mAb down-regulated the mRNA expression of TNF-α, IFN-γ, and perforin in the muscles. Our results suggest that CX3CL1-CX3CR1 interaction plays an important role in inflammatory cell migration into the muscle tissue of EAM mice. The results also point to the potential therapeutic usefulness of CX3CL1 inhibition and/or blockade of CX3CL1-CX3CR1 interaction in idiopathic inflammatory myopathy.
Idiopathic inflammatory myopathy (IIM),3 including polymyositis and dermatomyositis, is characterized by chronic inflammation of the voluntary muscles associated with infiltration of inflammatory cells, including CD4 and CD8 T cells and macrophages, in the skeletal muscle (1, 2, 3). Infiltrated CD4 and CD8 T cells express cytotoxic molecules, such as perforin and granzyme granules, and the T cells and macrophages express inflammatory cytokines, such as TNF-α and IFN-γ (4, 5, 6, 7, 8). Therefore, the infiltrated inflammatory cells might play an important role in the pathogenesis of IIM. The inflammatory cell migration into the muscle is thought to involve the interaction of chemokines and chemokine receptors (9, 10, 11, 12, 13, 14).
Chemokines are involved in leukocyte recruitment and activation at the site of inflammatory lesion (15). Approximately 50 chemokines have been identified to date, and they are classified into four subfamilies, C, CC, CXC, and CX3C chemokines, based on the conserved cystein motifs (16). Although the majority of chemokines are small secreted molecules, CX3CL1 (fractalkine) is expressed on the cell surface as a membrane-bound molecule (17, 18). The membrane-bound CX3CL1 is expressed on endothelial cells stimulated with TNF-α, IL-1, and IFN-γ (19, 20, 21), induces adhesion of the leukocytes, and supports leukocyte transmigration into tissue (22, 23). The soluble form of CX3CL1 is generated by proteolytic cleavage at a membrane-proximal region of the membrane-bound CX3CL1 by TNF-α-converting enzyme (a disintegrin and metalloproteinase domain 17) and a disintegrin and metalloproteinase domain 10 (24, 25), and is known to induce leukocyte migration (23). In contrast, CX3CR1, a unique receptor for CX3CL1, is expressed on peripheral blood CD4 and CD8 T cells that express cytotoxic molecules and type 1 cytokines (26, 27). CX3CR1 is also expressed on monocytes/macrophages, NK cells, and dendritic cells (28, 29).
Based on the infiltration of CTLs and macrophages into the affected muscles in patients with IIM, we speculated that the CX3CL1-CX3CR1 interaction might contribute to the inflammatory cell migration. In the present study we induced experimental autoimmune myositis (EAM) in SJL/J mice and examined CX3CL1 and CX3CR1 expression in the affected muscle of EAM mice. Furthermore, we studied the effect of CX3CL1 inhibition on EAM mice.
Materials and Methods
Induction of EAM
Male 5-wk-old SJL/J mice were purchased from Charles River Japan. Purified myosin from rabbit skeletal muscle (6.6 mg/ml; Sigma-Aldrich) was emulsified with an equal amount of CFA (Difco Laboratories) with 3.3 mg/ml Mycobacterium butyricum (Difco Laboratories). Mice were immunized intracutaneously with 100 μl of emulsion into four locations (total, 400 μl) on the back on days 0, 7, and 14. On day 21, the mice were killed, and the quadriceps femoris muscles were harvested. The muscle tissues were frozen immediately in chilled isopentane precooled in liquid nitrogen, and then 6-μm-thick cryostat sections were prepared at intervals of 200 μm. The sections were stained with H&E or used for immunohistochemistry. The experimental protocol was approved by the institutional animal care and use committee of Tokyo Medical and Dental University.
Immunohistological staining was performed as described previously (26, 30) with some modifications. Briefly, 6-μm-thick sections were air-dried and fixed in cold acetone at −20°C for 3 min. After air-drying at room temperature, the slides were rehydrated in PBS for 2 min three times, and then the endogenous peroxidase activity was blocked by incubation in 1.0% H2O2 in PBS for 10 min, followed by rinsing for 2 min three times in PBS. Nonspecific binding was blocked with 10% normal rabbit serum in PBS for 30 min. For CD4, CD8, and F4/80 staining, the sections were incubated with 5 μg/ml rat anti-mouse CD4 mAb (GK1.5; Cymbus Biotechnology), 2 μg/ml rat anti-mouse CD8a mAb (53-6.7; BD Pharmingen), 5 μg/ml rat anti-mouse F4/80 mAb (C1:A3-1; Serotec), or normal rat IgG in Ab diluent (BD Pharmingen) overnight at 4°C. The samples were then washed three times in PBS for 5 min each time and incubated with biotin-conjugated rabbit anti-rat IgG (DakoCytomation) for 30 min at room temperature with 5% normal mouse serum. To analyze a time course of cell infiltration, numbers of CD4+, CD8+, and F4/80+ cells in six randomly selected fields at ×200 were counted from three EAM mice on days 0, 7, 14, and 21.
For mouse vascular endothelial cell staining, we used a tyramide signal amplification kit (NEL700A; PerkinElmer). After blocking with 10% normal rabbit serum, the sections were incubated with 5 μg/ml rat anti-mouse vascular endothelial cadherin Ab (11D4.1; BD Pharmingen) or normal rat IgG overnight at 4°C. The samples were then washed three times in PBS for 5 min each time and incubated with biotin-conjugated rabbit anti-rat IgG for 30 min at room temperature with 5% normal mouse serum. After washing three times in PBS for 5 min each time, the sections were incubated with streptavidin-HRP for 30 min at room temperature and washed in PBS three times for 5 min each time. The samples were incubated with biotinyl tyramide amplification reagent at room temperature for 5 min, then washed three times in PBS for 5 min each time, and incubated again with streptavidin-HRP for 30 min. After washing three times in PBS for 5 min each time, diaminobenzidine tablets (Sigma-Aldrich) were used for visualization. The sections were counterstained in hematoxylin for 30 s and washed in tap water for 5 min.
For mouse CX3CL1 staining, the endogenous peroxidase activity was blocked by incubation in 1.0% H2O2 in methanol, and then the sections were incubated overnight at 4°C with goat anti-mouse CX3CL1 Ab (sc-7227; Santa Cruz Biotechnology) or normal goat IgG in Ab diluent at 5 μg/ml. The samples were then washed three times in PBS for 5 min each time and incubated with biotin-conjugated rabbit anti-goat IgG (DakoCytomation) for 30 min at room temperature with 5% normal mouse serum. After washing three times in PBS for 5 min each time, the sections were incubated with peroxidase-conjugated streptavidin (DakoCytomation) for 30 min at room temperature and washed three times for 5 min each time. For enhancing the expression of CX3CL1 on endothelial cells, a tyramide signal amplification kit was used as described above. Diaminobenzidine tablets were used for visualization. The sections were counterstained in hematoxylin for 30 s and washed in tap water for 5 min.
For CD4, CD8 or F4/80, and CX3CR1 double staining, the sections were incubated overnight at 4°C with 5 μg/ml rat anti-mouse CD4 mAb (GK1.5), 5 μg/ml rat anti-mouse CD8 mAb (53-6.7), 5 μg/ml rat anti-mouse F4/80 mAb (C1:A3-1), or normal rat IgG in Ab diluent. Subsequently, the samples were washed three times for 5 min each time in PBS and incubated with Alexa Fluor 488-conjugated goat anti-rat IgG (Molecular Probes) at 5 μg/ml for 1 h at room temperature. For CX3CR1 staining, the sections were washed three times in PBS for 5 min each time and then incubated with rabbit anti-mouse CX3CR1 Ab (30) or normal rabbit IgG at 5 μg/ml in Ab diluent for 2 h at room temperature. Next, the samples were washed three times for 5 min each time in PBS and incubated with Alexa Fluor 568-conjugated goat anti-rabbit IgG (Molecular Probes) at 5 μg/ml for 1 h at room temperature. The slides were examined using fluorescent microscopy (BZ-Analyzer; Keyence).
Treatment with anti-mouse CX3CL1 mAb
A mAb against murine CX3CL1 was generated from Armenian hamsters immunized with recombinant murine CX3CL1 by a standard method. One mAb, 5H8-4, was selected for additional studies. The specificity was examined by ELISA using a panel of murine CXC (MIP-2, keratinocyte-derived chemokine, and CXCL9, 10, 12, and 13), CC (CCL1–7, 9–12, 17, 19–22, 25, 27, and 28), C (XCL1), and CX3C (CX3CL1) chemokines. The mAb reacted specifically with murine CX3CL1. Five hundred micrograms of hamster anti-mouse CX3CL1 mAb (5H8-4) or control Ab (hamster IgG; ICN Pharmaceuticals) was injected into the mouse peritoneal cavity three times per week from day 0 for 3 wk. The injection of anti-CX3CL1 mAb did not affect the number of PBMC (data not shown).
The severity of inflammatory changes was classified using five grades according to the classification of Kojima et al. (31) with some modification: score 0, no inflammation; score 1, mild endomysial inflammatory changes; score 2, severe endomysial inflammatory changes; score 3, perimysial inflammatory changes in addition to score 2; and score 4, diffuse extensive lesion. If multiple lesions were found in one muscle specimen, 0.5 point was added to the indicated score. To evaluate the severity of inflammation using a different aspect, we counted the number of necrotic muscle fibers, and CD4+, CD8+, and F4/80+ cells in continuous three sections. Each section examined six random fields at ×400. The evaluation of histopathological inflammatory changes was performed in a blind fashion for the experimental group identity.
Total RNA was prepared from a 100 mg muscle block using RNA extraction solution, Isogen (Nippon Gene), and treated with DNase I (Invitrogen Life Technologies). The first-strand cDNA was synthesized using oligo(dT)12–18 primers (Pharmacia Biotech) and SuperScript II reverse transcriptase (Invitrogen Life Technologies).
The relative quantitative real-time PCR was performed using SYBR Green I on ABI PRISM 7000 (Applied Biosystems) according to the instructions provided by the manufacturer. The cDNA was amplified with primers for TNF-α (5′, GTA CCT TGT CTA CTC CCA GGT TCT CT; 3′, GTG TGG GTG AGG AGC ACG TA), IFN-γ (5′, CCT GCG GCC TAG CTC TGA; 3′, CCA TGA GGA AGA GCT GCA AAG), perforin (5′, CCA CGG CAG GGT GAA ATT C; 3′, GGC AGG TCC CTC CAG TGA), and GAPDH (5′, ATG CAT CCT GCA CCA CCA A; 3′, GTC ATG AGC CCT TCC ACA ATG). These primers were designed using the ABI Primer Express Software program (Applied Biosystems). The reaction buffer contained the following components: 25 μl of SYBR Green PCR Master Mix (Applied Biosystems), 300 nM forward and reverse primers, 50 ng cDNA template, and RNA-free distilled water up to 50 μl of total volume. The PCR was conducted using the following parameters: 50°C for 2 min, 95°C for 10 min, and 40 cycles of denaturation at 95°C for 15 s and annealing/extension at 60°C for 1 min. GAPDH mRNA was used as an internal control to standardize the amount of sample mRNA. A validation experiment demonstrated approximately equal efficiencies of the target and reference. Thus, the relative expression of real-time PCR products was determined using the ΔΔCt method that compares the mRNA expression levels of the target gene and the housekeeping gene (32, 33). One of the control samples was chosen as a calibrator sample.
Differences in the score of tissue inflammation, number of necrotic muscle fibers, number of migrated cells, and relative expression levels of TNF-α, IFN-γ, and perforin between control Ab- and anti-mouse CX3CL1 mAb-treated EAM mice, and the relative expression levels of TNF-α, IFN-γ, and perforin between normal and EAM mice were examined for statistical significance using Mann-Whitney’s U test. All data were expressed as the mean ± SEM. The difference between two groups of mice was considered significant at p < 0.05.
Development of EAM
SJL/J mice were immunized with purified rabbit myosin fraction and CFA on days 0, 7, and 14. On days 0, 7, 14, and 21, the quadriceps femoris muscles of these mice were histologically examined with H&E staining. All muscle specimens of normal SJL/J mice and immunized mice on day 7 showed normal appearance with no inflammatory changes (Fig. 1, A and B, respectively), whereas those of mice immunized with rabbit myosin fraction showed mild mononuclear cell infiltration at day 14 (Fig. 1,C). On day 21, a significant number of mononuclear cells were infiltrated among the muscle fibers (endomysium; Fig. 1,D), at perivascular areas (perimysium; Fig. 1,E), and epimysium (Fig. 1,F). Scattered lesions with aggregates of infiltrated mononuclear cells were formed, in which atrophic or necrotic muscle fibers were noted (arrow in Fig. 1 D). Injection of PBS and CFA into SJL/J mice did not show infiltration of inflammatory cells in the quadriceps femoris muscles (data not shown).
To determine the subsets of infiltrating mononuclear cells in the quadriceps femoris muscles of EAM mice, we performed immunohistochemical analysis using mAbs against CD4, CD8, and F4/80. CD4+ T cells were mainly located in the perimysium and some were found in the endomysium (Fig. 2,A). CD8+ T cells were predominantly detected in the endomysium and surrounded nonnecrotic muscle fibers (Fig. 2,B). F4/80+ macrophages were located in the endomysium as well and were especially present around the necrotic muscle fibers (Fig. 2 C). Because these histological findings of inflammatory cell infiltration patterns resembled those of affected muscle lesions in IIM patients (34, 35, 36), we decided to use the EAM mice as an experimental model of IIM.
To evaluate a time course of cellular infiltration into the muscles, we counted the numbers of infiltrated CD4+, CD8+, and F4/80+ cells on days 0, 7, 14, and 21 by immunohistochemical method. The majority of the infiltrating cells on day 14 were F4/80+ macrophage (Fig. 3). In contrast, the number of CD4+ and CD8+ T cells was not increased until day 14, and they had significantly migrated into the muscles on day 21. These results were similar to previously reported data (37).
CX3CL1 and CX3CR1 expression in the muscle of EAM mice
We examined the expression of CX3CL1 in the muscle of normal SJL/J mice and EAM mice by immunohistochemistry. In the quadriceps femoris muscles of normal mice, no CX3CL1 expression was detected (Fig. 4, A and G). In contrast, CX3CL1 was expressed on infiltrated mononuclear cells predominantly in the endomysium and vascular endothelial cells of EAM mice on day 14 (Fig. 4, B and H, respectively) and day 21 (Fig. 4, C and I, respectively).
We next examined the expression of CX3CR1 on the infiltrated mononuclear cells in the quadriceps femoris muscle of EAM mice by double immunohistochemical staining. Some CD4+ T cells expressed CX3CR1 (Fig. 5, A–C). The majority of CD8+ T cells and most of the F4/80+ macrophages expressed CX3CR1 (Fig. 5, D–F and G–I, respectively).
Effect of anti-mouse CX3CL1 mAb on EAM mice
To analyze the effect of anti-CX3CL1 mAb administration on EAM mice, we evaluated the histological changes in quadriceps femoris muscle using H&E staining. The incidence of inflammatory cell infiltration in control Ab-treated mice was 100% (n = 10). Treatment with anti-CX3CL1 mAb did not change the incidence of cellular infiltration (100%; n = 10). EAM mice treated with control Ab showed mononuclear cell infiltration with atrophy and necrosis of muscle fibers (Fig. 6,A). In comparison, anti-CX3CL1 mAb-treated EAM mice showed milder histological changes (Fig. 6,B). Analysis of histological scores of inflammatory changes in the quadriceps femoris muscles indicated that treatment with anti-CX3CL1 mAb significantly reduced inflammatory cell infiltration in the muscles of EAM mice compared with treatment with control Ab (Fig. 6,C). Moreover, anti-CX3CL1 mAb treatment reduced the number of necrotic muscle fibers in muscles (Fig. 6 D). A similar result was obtained in another independent set of experiments.
We next examined the effect of anti-CX3CL1 mAb treatment on the numbers of each subset of infiltrating cells. The numbers of CD4+, CD8+, and F4/80+ cells in quadriceps femoris muscles were counted and compared between mice treated with control Ab and those with anti-CX3CL1 mAb. Anti-CX3CL1 mAb treatment significantly reduced the number of infiltrated CD4+ T cells by ∼30% (Fig. 7,A), CD8+ T cells by ∼50%, and F4/80+ macrophages by up to 50% (Fig. 7, B and C).
We finally examined the effects of anti-CX3CL1 mAb treatment on the expression of cytokines and cytotoxic molecule in the quadriceps femoris muscle of EAM mice by quantitative RT-PCR. Although the relative quantities of TNF-α, IFN-γ, and perforin mRNA were very low in normal SJL/J mice, they were significantly up-regulated in EAM mice that received control Ab treatment (p < 0.05). Furthermore, treatment with anti-CX3CL1 mAb strikingly reduced mRNA expression (Fig. 8).
Considered together, the above results indicate that treatment with anti-CX3CL1 mAb reduced infiltration of CD4 and CD8 T cells and macrophages and reduced the expression of various inflammatory cytokines and cytotoxic molecule in muscles.
The major findings of the present study were the following. 1) CX3CL1 was expressed on infiltrated mononuclear cells and vascular endothelial cells, and its corresponding receptor, CX3CR1, was expressed on infiltrated inflammatory cells in the muscles of EAM. 2) Treatment with anti-CX3CL1 mAb ameliorated histological inflammatory changes in EAM mice, reduced the numbers of infiltrated CD4 and CD8 T cells and macrophages, and reduced the expression of TNF-α, IFN-γ, and perforin in the muscles. These results suggest that CX3CL1-CX3CR1 interaction seems to play an important role in inflammatory cell migration into the muscles of EAM mice.
Development of EAM in SJL/J mice by immunization with rabbit purified skeletal myosin fraction and CFA was previously reported (37, 38, 39, 40). We modified the method by increasing the amount of immunized myosin and CFA and the addition of Mycobacterium butyricum. This modification shortened the period required for the development of myositis from 5 wk, which was thought to be appropriate for the induction (38), to 3 wk. Moreover, although pertussis toxin (PTX) injection into the peritoneal cavity increased the severity of inflammatory changes in the muscle (31), and thus, PTX was administered in the previous models (31, 36, 38), our modified method induces significant myositis without PTX injection. The EAM mice showed inflammatory cell infiltration in the endomysium, perimysium, and epimysium with muscle fiber necrosis. Immunohistochemical analysis showed that the invading cells surrounding nonnecrotic muscle fibers in the endomysium were mainly CD8 T cells, whereas macrophages were predominantly detected in necrotic fibers, and CD4 T cells were located in perimysium. Moreover, quantitative RT-PCR showed up-regulation of expression of TNF-α, IFN-γ, and perforin mRNA in the muscle of EAM mice. These findings in EAM mice are similar to those reported in IIM patients (4, 5, 6, 7, 8, 34, 35, 36).
Inflammatory cell migration into the affected muscle of IIM is thought to involve chemokine-chemokine receptor interaction (9, 10, 11, 12, 13, 14). In the present study we focused on the role of CX3CL1-CX3CR1 interaction in the inflammatory cell migration. We showed the expression of CX3CR1 on some CD4 T cells and most CD8 T cells in EAM mice. It has been reported that CTLs including both CD4+ and CD8+ T cells invade the muscle fibers in IIM patients (3). These cells possess cytotoxic molecules, such as perforin and granzyme B, which are released into muscle cells (4, 5). Furthermore, type 1 cytokines, such as TNF-α and IFN-γ, were expressed in the inflammatory lesions of IIM patients (6, 7, 8). These findings suggest that the cytotoxic molecules and type 1 cytokines play important roles in the inflammatory lesions in IIM patients. In contrast, we reported previously that peripheral blood CX3CR1+ T cells express cytotoxic molecules and type 1 cytokines (26, 27). Therefore, the interaction of CX3CL1 and CX3CR1 could induce the migration of T cells, which express cytotoxic molecules and type 1 cytokines, into the affected muscles.
The infiltrated macrophages into the affected muscle also express inflammatory cytokines (9, 41). They express TNF-α and IL-1β, which could stimulate T cells, macrophages, and endothelial cells to produce various inflammatory cytokines, chemokines, and adhesion molecules. Moreover, these cytokines might have myocytotoxic effects (42, 43, 44). Our results showed that the majority of the F4/80+ macrophages expressed CX3CR1 in the muscle of EAM mice. Thus, the CX3CL1-CX3CR1 interaction might also play an important role in macrophage migration into the affected muscle in addition to T cell migration.
CX3CL1 was expressed on infiltrated mononuclear cells in the affected muscles of EAM mice. Because CX3CL1 expression was located in the endomysium, infiltrated macrophages and/or CD8 T cells may express CX3CL1 in the muscles. Furthermore, we showed that CX3CL1 was also expressed on vascular endothelial cells in the EAM muscle tissue on days 14 and 21, but not in normal mice. It was reported that CX3CL1 was expressed on endothelial cells activated with TNF-α and IFN-γ in vitro (19, 20, 21). Expressed CX3CL1 on endothelial cells might recruit CX3CR1+ cells, including macrophages and T cells, into muscle. These cells, in turn, express TNF-α and IFN-γ, which induce additional CX3CL1 expression on endothelial cells and also on recruited inflammatory cells. The enhanced expression of CX3CL1 may induce additional inflammatory cell migration. Consequently, these amplification cascades could contribute to the expansion of pathological changes in EAM mice. In fact, inhibition of CX3CL1 reduced the numbers of migrated CD4 and CD8 T cells and macrophages in the affected muscles of EAM mice and also reduced the expression of TNF-α, IFN-γ, and perforin. These results suggest that CX3CL1 blockade reduces the migration of inflammatory cells, which express cytotoxic molecules and cytokines, into the muscles. Thus, inhibition of CX3CL1-CX3CR1 interaction might be a potentially suitable therapeutic strategy for treatment of IIM.
Our data showed that mRNA expression of TNF-α, IFN-γ, and perforin was almost totally inhibited by anti-CX3CL1 mAb treatment, although the numbers of infiltrated monocytes were decreased by up to 50%. Recently it was reported that stimulation with CX3CL1 enhanced production of proinflammatory cytokines such as IFN-γ as well as the release of cytolytic granules by T cells (45). Thus, blockade of CX3CL1 might inhibit not only cellular migration, but also cytokine and cytotoxic molecule expression, by stimulation with CX3CL1 in the EAM muscle. Alternatively, because CX3CR1+ T cells express type 1 cytokine and cytotoxic molecules (23, 26, 27), and CX3CR1high positive monocytes greatly produce inflammatory cytokines compared with CX3CR1low positive monocytes (46, 47, 48), treatment with anti-CX3CL1 mAb may selectively inhibit the migration of such specific T cells and macrophages. Therefore, anti-CX3CL1 mAb might be able to inhibit the expression of cytokine and cytotoxic molecules effectively in muscles, but additional study is required.
We recently reported that inhibition of CX3CL1 ameliorated collagen-induced arthritis in mice, probably by suppression of inflammatory cell migration into the synovium (30). Others reported that anti-CX3CR1 Ab treatment blocked inflammatory cell infiltration in the glomeruli, prevented crescent formation, and improved renal function in the Wistar-Kyoto crescentic glomerulonephritis model (49). Furthermore, the gene deletion of CX3CR1 resulted in an ∼50% decrease in the formation of atherosclerotic lesions and the number of infiltrated macrophages in the lesion in experimental atherosclerosis mice (50, 51). These results together with our findings suggest that blockade of CX3CL1-CX3CR1 interaction might be therapeutically useful for several diseases associated with inflammatory cell infiltration. In this study we propose that such treatment is also suitable for IIM. To our knowledge, this is the first report demonstrating that a chemokine inhibitor could reduce the severity of myositis.
In conclusion, we demonstrated in the present study that inhibition of CX3CL1 significantly improved histopathological changes in the muscles of EAM mice, suggesting that blockade of CX3CL1 might be therapeutically beneficial for IIM.
We thank Dr. Hiroshi Nemoto (Toho University School of Medicine) for providing critical suggestions for the development of EAM mice. We also thank Miyuki Nishimura, Keiko Mizuno, and Yoko Inoue for their excellent technical support.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported in part by a grant-in-aid from the Japan Intractable Diseases Research Foundation; the Fund for Intractable Diseases Research by Atsuko Ouchi from Tokyo Medical and Dental University; the Ministry of Health, Labor, and Welfare; and the Ministry of Education, Science, Sports, and Culture, Japan.
Abbreviations used in this paper: IIM, idiopathic inflammatory myopathy; EAM, experimental autoimmune myositis; PTX, pertussis toxin.