Retinoid ameliorates experimental autoimmune myositis with modulation of Th cell differentiation and antibody production in vivo.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 60, No. 10, October 2009, pp 3118–3127 DOI 10.1002/art.24930 © 2009, American College of Rheumatology Retinoid Ameliorates Experimental Autoimmune Myositis, With Modulation of Th Cell Differentiation and Antibody Production In Vivo Naho Ohyanagi,1 Miwako Ishido,2 Fumihito Suzuki,1 Kayoko Kaneko,1 Tetsuo Kubota,1 Nobuyuki Miyasaka,1 and Toshihiro Nanki1 myositis, orally administered Am80 significantly reduced the number of infiltrating inflammatory cells and the expression of tumor necrosis factor ␣ and interleukin-1␤ (IL-1␤) in muscle. Moreover, Am80 increased production of interferon-␥, IL-4, and IL-10, but not IL-17, by myosin-stimulated splenic T cells of mice with experimental autoimmune myositis, suggesting that it could enhance differentiation into Th1 and Th2, but not Th17, in vivo. Am80 also decreased serum levels of IgG2a and IgG2b antimyosin antibodies, but did not affect levels of IgG1 antimyosin antibodies. In addition, it suppressed chemokine expression and activator protein 1 activity in myoblasts in vitro. Conclusion. The synthetic retinoid Am80 has an inhibitory effect on experimental autoimmune myositis. It might regulate the development of Th phenotype and antibody production in vivo, in addition to its effects on cytokine and chemokine production. Objective. Polymyositis and dermatomyositis are chronic inflammatory muscle diseases. Retinoids are compounds that bind to the retinoic acid binding site of retinoic acid receptors and have biologic activities similar to those of vitamin A. Recent studies indicate that retinoids promote Th2 differentiation and suppress Th1 and Th17 differentiation in vitro. The present study was undertaken to examine the effects of a synthetic retinoid, Am80, on experimental autoimmune myositis as well as on Th phenotype development and antibody production. Methods. Experimental autoimmune myositis was induced in SJL/J mice by immunization with rabbit myosin. Am80 was administered orally once daily. Its effects were evaluated by measurement of the numbers of infiltrating inflammatory cells, production of inflammatory cytokines in muscle, production of Th-specific cytokines by myosin-stimulated splenic T cells, and production of antimyosin antibodies in serum. Results. In mice with experimental autoimmune Polymyositis (PM) and dermatomyositis (DM) are characterized by chronic inflammation of the skeletal muscles associated with infiltration by inflammatory cells. In particular, CD8⫹ T cells and macrophages infiltrate mainly into endomysial areas in patients with PM, and in contrast, CD4⫹ T cells, B cells, macrophages, and dendritic cells are largely located in the perivascular and/or perimysial areas in DM muscle (1–6). A type 1 immune response, including high production of tumor necrosis factor ␣ (TNF␣) and interferon-␥ (IFN␥), has been reported to be involved in the development of muscle inflammation, and overexpression of such cytokines has been demonstrated in mononuclear infiltrates surrounding muscle fibers, with up-regulation of adhesion molecules and chemokines (6–10). In addition, type I interferons might have an important role in disease pathogenesis, especially in DM (5,11). Supported in part by Grants-in-Aid for Scientific Research from the Ministry of Health, Labor, and Welfare and the Japanese Ministry of Education, Culture, Sports, Science, and Technology, and by grants from the Global Center of Excellence Program of the Japanese Ministry of Education to the International Research Center for Molecular Science in Tooth and Bone Diseases at Tokyo Medical and Dental University. 1 Naho Ohyanagi, MSc, Fumihito Suzuki, MD, Kayoko Kaneko, MD, Tetsuo Kubota, MD, Nobuyuki Miyasaka, MD, Toshihiro Nanki, MD: Tokyo Medical and Dental University, Tokyo, Japan; 2 Miwako Ishido, PhD: R&R, Inc., Tokyo, Japan. Dr. Miyasaka has received consulting fees, speaking fees, and/or honoraria from Mitsubishi Tanabe Pharma, Wyeth Japan, Takeda Pharmaceutical, Abbott Japan, and Eisai Company, Ltd. (less than $10,000 each). Address correspondence and reprint requests to Toshihiro Nanki, MD, Department of Medicine and Rheumatology, Graduate School, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan. E-mail: firstname.lastname@example.org. Submitted for publication June 21, 2008; accepted in revised form June 29, 2009. 3118 RETINOID AMELIORATES EXPERIMENTAL AUTOIMMUNE MYOSITIS We have developed an experimental model of autoimmune myositis in mice, which is induced by repeated immunization with rabbit myosin (12). Inflammatory cell infiltration and necrotic muscle fiber are evident in this model. In the muscle of animals with experimental autoimmune myositis, CD4⫹ T cells are mainly located in the perimysium and CD8⫹ T cells are chiefly located in the endomysium and surrounded nonnecrotic muscle fibers (12). We also showed that expression of TNF␣, IFN␥, and perforin was up-regulated in the muscles of mice with experimental autoimmune myositis (12). Moreover, expression of intercellular adhesion molecule 1 was increased in the muscles (13). Retinoid is a general term for compounds that bind to and activate retinoic acid receptors (RARs [RAR␣, RAR␤, and RAR␥]) and/or retinoid X receptors (RXRs [RXR␣, RXR␤, and RXR␥]), members of the nuclear receptor superfamily, and have biologic activities similar to those of vitamin A. The most important endogenous retinoid is all-trans-retinoic acid, which is a ligand for RAR␣, RAR␤, and RAR␥ (14). Retinoids have important roles in cell proliferation, differentiation, and morphogenesis (15,16). They also have a modulating function on inflammatory and immunocompetent cells, including T cells and macrophages (17). In addition, retinoids suppress differentiation into Th1 cells but promote Th2 differentiation in vitro (18–20). Moreover, recent studies indicate that retinoids inhibit differentiation into Th17 and increase differentiation into regulatory T cells (21,22). Therefore, retinoids may have a beneficial effect in Th1- and/or Th17-dominant diseases. In fact, retinoid treatment has been shown to be effective in experimental autoimmune encephalomyelitis and collagen-induced arthritis (23–25), both of which are thought to be Th1/Th17-related animal models (26–28). Am80, a synthetic retinoid, binds to RAR␣ and RAR␤, but not to RAR␥. It was launched in the Japanese market as a drug for acute promyelocytic leukemia (29), as was all-trans-retinoic acid. The purpose of this study was to determine the effects of Am80 on experimental autoimmune myositis, and its immunoregulatory effects in vivo. MATERIALS AND METHODS Induction of experimental autoimmune myositis and treatment with Am80. The experimental protocol was approved by the Institutional Animal Care and Use Committee of Tokyo Medical and Dental University. The method for induction of experimental autoimmune myositis has been described previously (12). Briefly, 5-week-old male SJL/J mice were purchased from Charles River Japan (Yokohama, Ja- 3119 pan). Purified myosin from rabbit skeletal muscle (6.6 mg/ml; Sigma-Aldrich, St. Louis, MO) was emulsified with an equal amount of Freund’s complete adjuvant (CFA; Difco, Detroit, MI) and 3.3 mg/ml Mycobacterium butyricum (Difco). To evaluate the prophylactic effect of Am80, mice were immunized intracutaneously with 100 l of the emulsion into 4 locations (total 400 l) on the back on days 1, 8, and 15. Am80 was suspended in carboxymethylcellulose. Am80 (0.2 mg/kg [n ⫽ 18], 2.0 mg/kg [n ⫽ 18], or 4.0 mg/kg [n ⫽ 16]) or carboxymethylcellulose alone as vehicle (n ⫽ 20) was administered orally once per day from day 1 to day 21. On day 22, the mice were killed and the quadriceps femoris muscles were harvested. To analyze the therapeutic effects of Am80, mice were immunized with myosin plus CFA on days 1, 8, 15, and 22 and treated with vehicle (n ⫽ 7) or Am80 (4.0 mg/kg [n ⫽ 8]) from day 15 to day 28. On day 29, the mice were killed and muscle tissues were harvested. The muscle tissue was immediately frozen in chilled isopentane precooled in liquid nitrogen, and then 8-m–thick cryostat sections (intervals of 320 m) were prepared. The sections were stained with hematoxylin and eosin (H&E). To evaluate the severity of muscle inflammatory changes, we counted the total number of infiltrated cells in the H&Estained sections. Three sections from each mouse were prepared, and photomicrographs of 3 randomly selected fields per section were obtained at 200⫻ magnification. The numbers of infiltrating mononuclear cells were counted by 2 evaluators who were blinded with regard to the experimental group, and the mean number from the 2 counts was used. Immunohistochemistry. Eight-micrometer–thick cryostat sections of muscle were air-dried and fixed in cold acetone for 3 minutes at ⫺20°C. The slides were rehydrated in phosphate buffered saline (PBS) 3 times for 2 minutes each time, and then endogenous peroxidase activity was blocked by incubation in 1.0% H2O2 in PBS for 10 minutes, followed by rinsing with PBS. Nonspecific binding was blocked by incubation with 10% rabbit serum in PBS for 30 minutes. The sections were incubated overnight at 4°C with 5 g/ml rat anti-mouse CD4 monoclonal antibody (mAb) (GK1.5; Cymbus Biotechnology, Hampshire, UK), 2 g/ml rat anti-mouse CD8a mAb (53-6.7; BD PharMingen, Franklin Lakes, NJ), 5 g/ml rat anti-mouse F4/80 mAb (C1:A3-1; Serotec, Planegg, Germany), or normal rat IgG in antibody diluent (BD PharMingen). The samples were washed 3 times in PBS and then incubated for 30 minutes with 2.5 g/ml biotin-conjugated rabbit anti-rat IgG (Dako Cytomation, Glostrup, Denmark) pretreated with 5% normal mouse serum to reduce nonspecific binding. After washing in PBS, the sections were incubated for 30 minutes with streptavidin–horseradish peroxidase (HRP). After washing in PBS, diaminobenzidine (Sigma-Aldrich) was used for visualization. The sections were counterstained with hematoxylin for 30 seconds and washed in tap water for 5 minutes. To analyze cell infiltration, we prepared 2 sections from each mouse for each staining, and the numbers of CD4⫹, CD8⫹, and F4/80⫹ cells were counted in 3 randomly selected fields per section at 200⫻ magnification. Real-time reverse transcriptase–polymerase chain reaction (RT-PCR). Total RNA was prepared from 50-mg muscle blocks using RNA extraction solution (Isogen; Nippon Gene, Tokyo, Japan) and treated with DNase I (Invitrogen Life Technologies, Carlsbad, CA). First-strand complementary 3120 DNA (cDNA) was synthesized using oligo(dT) primers (Pharmacia Biotech, Buckinghamshire, UK) and Superscript ⌱⌱ RT (Invitrogen Life Technologies). Real-time RT-PCR was performed in a total volume of 50 l containing 50 ng cDNA, 1⫻ TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA), and 250 nM of each TaqMan probe (TNF␣ Mm00443258, interleukin-1␤ [IL-1␤] Mm00434228, GAPDH Mm99999915), using a thermal cycler (ABI Prism 7000; Applied Biosystems). After the initial step (50°C for 2 minutes and 95°C for 10 minutes), denaturation, annealing, and amplification were performed for 40 cycles at 95°C for 15 seconds and 60°C for 1 minute. The relative expression of real-time RT-PCR products was determined using the ⌬⌬Ct method, which compares the messenger RNA (mRNA) expression levels of the target gene and the housekeeping gene, GAPDH. The threshold cycle was determined, and relative gene expression was calculated as the fold difference ⫽ 2⫺⌬⌬Ct, where ⌬Ct ⫽ Ct of the target gene ⫺ Ct of GAPDH, and ⌬⌬Ct ⫽ ⌬Ct normal ⫺ ⌬Ct experimental autoimmune myositis. Cytokine production by rabbit myosin–stimulated splenic T cells. Thy1.2⫹ splenic T cells from normal mice and mice with experimental autoimmune myositis were purified using magnetic-activated cell sorting (MACS) microbead– coupled mAb and an automatic cell separation system (auto MACS; Miltenyi Biotec, Auburn, CA). An antigen-presenting cell (APC)–enriched population was prepared from normal splenocytes depleted of Thy1.2⫹ T cells and B220⫹ B cells using MACS. Purified T cells (4 ⫻ 105) and APC-enriched splenocytes (1 ⫻ 105) were cocultured in 96-well plates in RPMI 1640 (Sigma-Aldrich) with 10% fetal calf serum (FCS; Sigma-Aldrich) supplemented (where indicated) with 50 g/ml denatured (100°C, 10 minutes) rabbit myosin. After 72 hours, concentrations of IFN␥, IL-4, IL-17, and IL-10 in the culture supernatant were measured by DuoSet enzyme-linked immunosorbent assay (ELISA; R&D Systems, Minneapolis, MN) for IFN␥, IL-4, and IL-17 or Quantikine ELISA (R&D Systems) for IL-10, according to the instructions of the manufacturer. Measurement of serum antimyosin antibody. Serum IgG1, IgG2a, and IgG2b antimyosin antibodies in 14 normal mice, 20 mice with experimental autoimmune myositis treated with vehicle, 15 mice with experimental autoimmune myositis treated with 0.2 mg/kg Am80, 13 mice with experimental autoimmune myositis treated with 2.0 mg/kg Am80, and 13 mice with experimental autoimmune myositis treated with 4.0 mg/kg Am80) were measured by ELISA. Pooled sera from the mice with experimental autoimmune myositis were used as standard. Purified rabbit myosin (2.45 g/ml) was coated on 96-well plates overnight at room temperature. The plates were washed twice with PBS containing 0.05% Tween 20 and then blocked for 3 hours with 2% bovine serum albumin (BSA) in PBS at room temperature. After the blocking solution was removed, plates were washed and 50 l of each serum sample (at a 1:50,000 dilution for IgG1 and IgG2b assay and a 1:10 dilution for IgG2a assay) in 0.05% Tween 20 and 2% BSA in PBS was incubated in the plate for 2 hours at room temperature. After the plates were washed, 50 l of HRP-conjugated rabbit anti-mouse IgG1, IgG2a, or IgG2b subclass–specific antibody (Zymed, Charlotte, NC) was added and incubated at room temperature for 30 minutes. After washing, -phenylenediamine was added, and for each well, absorbance at 450 nm was measured with a microplate reader. The anti- OHYANAGI ET AL body titer of the samples was determined from the absorbance, using a standard curve constructed for each IgG subclass. A standard serum from mice with experimental autoimmune myositis was added to each plate in serial dilutions, and a standard curve was constructed. The standard serum was arbitrarily designated as 3,000 units/ml (for IgG1 and IgG2b measurement) or 4,000 units/ml (for IgG2a measurement), and the antibody titers of serum samples were determined with the standard curves. Culture of mouse myoblasts and macrophages. Mouse myoblasts (Summit Pharmaceuticals, Tokyo, Japan) were cultured at 1 ⫻ 105 cells/ml in Dulbecco’s modified Eagle’s medium (DMEM; Sigma-Aldrich) with 10% FCS at 37°C in 96-well plates. After 24 hours, the cells were stimulated with 5 ng/ml TNF␣ (PeproTech, Rocky Hill, NJ) or IL-1␤ (PeproTech) for 24 hours at 37°C, without Am80 or in the presence of Am80 at 10⫺8, 10⫺7, or 10⫺6 moles/liter). The concentrations of CCL2/monocyte chemotactic protein 1 (MCP-1) and CCL5/RANTES in the culture supernatant were determined by ELISA according to the instructions of the manufacturer (BioSource International, Camarillo, CA). CD11b⫹ macrophages were isolated from mouse splenocytes using MACS. The cells were incubated at 1 ⫻ 105 cells/ml in RPMI 1640–10% FCS, in 96-well plates with 5 ng/ml TNF␣ or IL-1␤ in the presence or absence of 10⫺6 moles/liter Am80, for 24 hours at 37°C. The concentration of CCL5 was measured by ELISA. Since adding Am80 to the standard recombinant CCL2 or CCL5 did not alter the results of ELISAs (data not shown), it was assumed that Am80 did not influence the ELISA measurements of CCL2 and CCL5. Measurement of activator protein 1 (AP-1) activity in mouse myoblasts. Mouse myoblasts were cultured at 1 ⫻ 105 cells/ml in DMEM–10% FCS at 37°C in 60-mm dishes. After 24 hours, the cells were stimulated for 2 hours with 5 ng/ml TNF␣ or IL-1␤ in the presence or absence of 10⫺6 moles/liter Am80. Then, nucleoprotein was extracted using a nuclear extraction kit (Active Motif, Carlsbad, CA), and AP-1 activity was measured with an AP-1 transcription factor microplate assay kit according to the protocol of the manufacturer (Marligen Biosciences, Ijamsville, MD). The fold increase of AP-1 activity was calculated (AP-1 activity with cytokine stimulation and/or Am80 treatment divided by AP-1 activity without stimulation). Semiquantitative RT-PCR. Total RNA was prepared from mouse muscle tissue, myoblasts, and CD11b⫹ splenic macrophages as described above, and first-strand cDNA was synthesized using 2 g total RNA. PCR was performed in a total volume of 50 l containing 1 l cDNA, 0.2 mM dNTP, 0.02 M of each primer, 1⫻ PCR buffer (Roche Molecular Systems, Branchburg, NJ), and FastStart Taq DNA polymerase (Roche Molecular Systems), with a thermal cycler (PTC-200; MJ Research, Saint Bruno, Quebec, Canada). After the initial denaturing step (94°C for 4 minutes), amplification was performed for 40 or 45 cycles (40 cycles for GAPDH, RAR␣, and RAR␤; 45 cycles for RAR␥) at 95°C for 40 seconds, 59°C or 64°C for 30 seconds (59°C for GAPDH, RAR␣, and RAR␤; 64°C for RAR␥), and 72°C for 60 seconds. The final cycle was followed by an extension step of 5 minutes at 72°C. Sequences of the sense and antisense primers were as follows: GAPDH 5⬘-ACCCAGAAGACTGTGGATGG-3⬘ (sense), 3⬘-GTCATCATCCTTGGCAGGTT-5⬘ (antisense); RAR␣ 5⬘-CTGGG- RETINOID AMELIORATES EXPERIMENTAL AUTOIMMUNE MYOSITIS 3121 GGCGGGCACCTCAATGG-3⬘ (sense), 3⬘-CGGCAGTACTGGCAGCGGTTCC-5⬘ (antisense); RAR␤ 5⬘-CGTCCCGAGCCCACCATC-3⬘ (sense), 3⬘-TGTCCCAGAGGCCCAAGTCC-5⬘ (antisense); RAR␥ 5⬘-CCCCGCCCTCCCCTCCAGCAGTTT-3⬘ (sense), 3⬘-GAGGAGGTGGTGGGGGTGAGGGAGAGC-5⬘ (antisense). PCR products were resolved by electrophoresis on 1.2% agarose gels (Takara Bio, Otsu, Japan) containing ethidium bromide. Statistical analysis. The significance of differences in numbers of infiltrating cells, cytokine expression in the muscle, and production of cytokines, antimyosin antibodies, and chemokines was tested by Mann-Whitney U test. All data were expressed as the mean ⫾ SEM. P values less than 0.05 were considered significant. RESULTS Effects of Am80 on inflammatory changes in mice with experimental autoimmune myositis. We induced experimental autoimmune myositis in 72 SJL/J mice by immunization with rabbit myosin plus CFA on days 1, 8, and 15. Am80 was administered orally once daily from day 1 to day 21. On day 22, the quadriceps femoris muscles were harvested and stained with H&E. Muscle specimens from normal mice exhibited no inflammatory changes (Figure 1A), whereas those from mice immunized with rabbit myosin exhibited inflammatory cell infiltration in the endomysium and perimysium, and necrotic muscle fibers (Figure 1B). The incidence of inflammatory cell infiltration in vehicle-treated control mice (n ⫽ 20) was 100%. Treatment with Am80 did not change the incidence of cellular infiltration (100% incidence with Am80 at 0.2 mg/kg [n ⫽ 18], 2.0 mg/kg [n ⫽ 18], and 4.0 mg/kg [n ⫽ 16]). Treatment with Am80 at 0.2 mg/kg did not significantly alter the characteristics of the cellular infiltration (Figure 1C). In contrast, Am80 at 2.0 mg/kg and 4.0 mg/kg induced a significant decrease in inflammatory changes (Figures 1D and E, respectively). To quantitatively evaluate the effects of Am80 on muscle inflammation, we counted the infiltrating mononuclear cells. Substantial numbers of mononuclear cells were observed in mice with experimental autoimmune myositis, and treatment with Am80 at 2.0 mg/kg and 4.0 mg/kg resulted in a significant decline in the numbers of infiltrating mononuclear cells compared with controls (Figure 1F). We also counted necrotic fibers in the muscle sections; treatment with Am80 did not significantly change the numbers of necrotic fibers (data not shown). Next, to determine the therapeutic effect of Am80 on experimental autoimmune myositis, 15 mice were immunized with rabbit myosin on days 1, 8, 15, and 22. Since myositis develops by day 15 in this model (12), we treated the mice with Am80 from day 15 to day 28. Figure 1. Inhibition of inflammatory changes in the muscle of mice with experimental autoimmune myositis (EAM) by treatment with Am80. To induce experimental autoimmune myositis, mice were immunized with rabbit myosin and Freund’s complete adjuvant (CFA) on days 1, 8, and 15. Am80 was administered orally from day 1 to day 21. On day 22, quadriceps femoris muscles were harvested and stained with hematoxylin and eosin (H&E). A–E, Representative photomicrographs from a normal mouse (A), a mouse with vehicle-treated experimental autoimmune myositis (B), and mice with experimental autoimmune myositis treated with Am80 at 0.2 mg/kg (C), 2.0 mg/kg (D), or 4.0 mg/kg (E) (original magnification ⫻ 200). F and G, Number of infiltrating mononuclear cells per randomly selected field. In studies of the prophylactic effects of Am80 (F), mice were treated as described above (n ⫽ 15 normal mice, 20 mice with vehicle-treated experimental autoimmune myositis, and 18, 18, and 16 mice with experimental autoimmune myositis treated with Am80 at 0.2 mg/kg, 2.0 mg/kg, and 4.0 mg/kg, respectively). In studies of the therapeutic effects of Am80 (G), mice were immunized with rabbit myosin and CFA on days 1, 8, 15, and 22, and Am80 was administered orally from day 15 to day 28. The quadriceps femoris muscles were harvested and stained with H&E on day 29 (n ⫽ 3 normal mice, 7 mice with vehicle-treated experimental autoimmune myositis, and 8 mice with experimental autoimmune myositis treated with Am80 at 4.0 mg/kg). Values are the mean and SEM from 3 separate experiments. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.01, versus mice with vehicle-treated experimental autoimmune myositis. On day 29, the quadriceps femoris muscles were examined histologically. Treatment with 4.0 mg/kg Am80 (n ⫽ 8) resulted in a significant decrease in the numbers 3122 Figure 2. Treatment with Am80 reduces infiltration of T cells and macrophages into muscle tissue. Specimens from the quadriceps femoris muscles used in the experiments shown in Figure 1F, from normal mice, mice with vehicle-treated experimental autoimmune myositis (EAM), and mice with experimental autoimmune myositis treated with Am80 at 4.0 mg/kg, were stained for CD4 (A), CD8 (B), and F4/80 (C), and positive cells were counted. Values are the mean and SEM. of infiltrating mononuclear cells in the muscle sections compared with control (n ⫽ 7) (Figure 1G). These results indicate that Am80 attenuates the inflammatory changes not only prophylactically, but also therapeutically. To analyze the effects of Am80 on the number of each subset of infiltrating cells, the muscle tissue specimens from prophylactically treated mice with experimental autoimmune myositis shown in Figure 1F were stained for CD4, CD8, and F4/80, and the numbers of positive cells were counted. Double staining with CD3 and CD4 showed that most of the CD4⫹ cells were CD3⫹ T cells (fluorescence microscopy images available from the authors upon request). Myosin immunization increased the numbers of CD4⫹ and CD8⫹ T cells and F4/80⫹ macrophages. Treatment with Am80, however, significantly diminished the numbers of CD4⫹ and CD8⫹ T cells, as well as the numbers of macrophages (Figure 2). Effects of Am80 on expression of inflammatory cytokines. To examine the effects of Am80 on the expression of inflammatory cytokines, we analyzed TNF␣ and IL-1␤ expression in muscle tissue by quantitative RT-PCR. The levels of expression of mRNA for TNF␣ and IL-1␤, respectively, were ⬃3.5 times and ⬃7.0 times higher in mice with experimental auto- OHYANAGI ET AL immune myositis than in normal mice. Treatment with 4.0 mg/kg Am80 significantly attenuated the expression level of TNF␣ (relative expression level [mean ⫾ SEM] 3.6 ⫾ 0.7 with vehicle treatment, 2.3 ⫾ 1.1 with Am80 at 0.2 mg/kg [36.1% reduction compared with vehicle; P not significant], 2.9 ⫾ 0.9 with Am80 at 2.0 mg/kg [19.4% reduction compared with vehicle; P not significant], and 1.7 ⫾ 0.6 with Am80 at 4.0 [52.8% reduction compared with vehicle; P ⬍ 0.05]) (Figure 3A). Expression of IL-1␤ was also significantly reduced by Am80 treatment (relative expression level 7.3 ⫾ 1.7 with vehicle treatment, 3.9 ⫾ 0.8 with Am80 at 0.2 mg/kg [46.6% reduction compared with vehicle; P ⬍ 0.05], 3.4 ⫾ 0.9 with Am80 at 2.0 mg/kg [53.4% reduction compared with vehicle; P ⬍ 0.01], and 2.9 ⫾ 0.7 with Am80 at 4.0 mg/kg [60.3% reduction compared with vehicle; P ⬍ 0.01]) (Figure 3B). Effects of Am80 on Th phenotype development in vivo. We analyzed the effects of Am80 on Th cell differentiation induced by myosin immunization. Splenic T cells isolated from mice with experimental autoimmune myositis with or without Am80 treatment and APC-enriched populations prepared from normal mouse spleen were cocultured with rabbit myosin, and Figure 3. Am80 treatment suppresses tumor necrosis factor ␣ (TNF␣) and interleukin-1␤ (IL-1␤) expression in muscle. Expression of mRNA for TNF␣ (A) and IL-1␤ (B) in the quadriceps femoris muscles used in the experiments shown in Figure 1F, from normal mice, mice with vehicle-treated experimental autoimmune myositis (EAM), and mice with experimental autoimmune myositis treated with Am80, was measured by real-time polymerase chain reaction. Values are the mean and SEM from triplicate assays. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.01, versus mice with vehicle-treated experimental autoimmune myositis. RETINOID AMELIORATES EXPERIMENTAL AUTOIMMUNE MYOSITIS Figure 4. Effects of Am80 on cytokine expression in myosinstimulated splenic T cells. Isolated splenic T cells (4 ⫻ 105 cells/well) from mice with experimental autoimmune myositis (EAM) from the experiment shown in Figure 1F and antigen-presenting cell–rich splenocytes (1 ⫻ 105 cells/well) from normal mice were cocultured for 72 hours in 96-well plates with RPMI 1640 and 10% fetal calf serum supplemented with 50 g/ml rabbit myosin, and levels of interferon-␥ (IFN␥) (A), interleukin-4 (IL-4) (B), IL-17 (C), and IL-10 (D) in the supernatants were analyzed by enzyme-linked immunosorbent assay. Values are the mean and SEM from 3–4 animals per group, analyzed in duplicate. NS ⫽ not significant. concentrations of IFN␥, IL-4, and IL-17 in the supernatant were measured by ELISA. Without myosin stimulation, none of the cytokines investigated by ELISA were detected. T cells from normal mice did not express IFN␥, IL-4, or IL-17 even with myosin stimulation. In contrast, T cells from mice with experimental autoimmune myositis expressed the cytokines following incubation with myosin. Treatment with Am80 did not significantly alter production of IL-17, while Am80 increased IFN␥ and IL-4 production (Figures 4A–C). IFN␥ and IL-4 concentrations were ⬃2.7 times and ⬃5.4 times higher, respectively, in Am80-treated mice compared with those in vehicletreated mice with experimental autoimmune myositis. These results suggest that Am80 enhances differentiation into Th1 and Th2, although the treatment did not change Th17 differentiation in vivo. We also analyzed 3123 IL-10 expression by myosin-stimulated splenic T cells. Expression of IL-10, like that of IFN␥ and IL-4, was significantly higher in Am80-treated mice with experimental autoimmune myositis than in vehicle-treated mice with the disease (Figure 4D). Effects of Am80 on antimyosin antibody production. To determine the effect of Am80 on antimyosin antibody production, we measured mouse serum antimyosin antibodies by ELISA. While antimyosin antibody was not detected in normal mice, antimyosin antibodies of the IgG1, IgG2a, and IgG2b subclasses were produced in the serum of mice with experimental autoimmune myositis (Figure 5). Production of the IgG1 subclass was not altered by Am80 (Figure 5A). In contrast, production of antimyosin antibodies of the IgG2a subclass was significantly attenuated by treatment with Am80 at 2.0 mg/kg and 4.0 mg/kg (mean ⫾ SEM 793.9 ⫾ 189.0 units/ml with vehicle treatment, 616.6 ⫾ 78.1 units/ml with Am80 at 0.2 mg/kg [22.3% reduction compared with vehicle; P not significant], 576.1 ⫾ 172.3 units/ml with Am80 at 2.0 mg/kg [27.4% reduction compared with vehicle; P ⬍ 0.05], and 407.0 ⫾ 57.4 units/ml with Am80 at 4.0 mg/kg (48.7% reduction compared with vehicle; P ⬍ 0.01]) (Figure 5B). Production of the IgG2b subclass was also attenuated by Am80 (745.3 ⫾ 129.7 units/ml with vehicle treatment, 334.1 ⫾ 84.4 units/ml with Am80 at 0.2 mg/kg [55.2% reduction compared with vehicle; P ⬍ 0.001], 99.7 ⫾ 8.4 units/ml with Am80 at 2.0 mg/kg [86.6% reduction compared with vehicle; P ⬍ 0.001], and 95.8 ⫾ 9.8 units/ml with Am80 at 4.0 mg/kg [87.1% reduction compared with vehicle; P ⬍ 0.001]) (Figure 5C). Effects of Am80 on chemokine production by mouse myoblasts and splenic macrophages. To examine the effect of Am80 on chemokine production by mouse myoblasts in vitro, myoblasts from normal mice were stimulated with TNF␣ or IL-1␤ for 24 hours in the presence of Am80. Concentrations of CCL2/MCP-1 and CCL5/RANTES in the culture supernatant were then measured. While TNF␣ significantly increased the expression of CCL2 (mean ⫾ SEM 3,695.3 ⫾ 86.6 pg/ml) and CCL5 (383.0 ⫾ 26.3 pg/ml), the up-regulated expression was reduced by Am80 (with Am80 at 10⫺8 moles/liter, CCL2 3,357.3 ⫾ 263.2 pg/ml [9.1% reduction; P not significant] and CCL5 250.2 ⫾ 2.8 pg/ml [34.7% reduction; P ⬍ 0.05], with Am80 at 10⫺7 moles/liter, CCL2 3,316.5 ⫾ 67.0 pg/ml [10.3% reduction; P ⬍ 0.05] and CCL5 197.9 ⫾ 24.1 pg/ml [48.3% reduction; P ⬍ 0.05], with Am80 at 10⫺6 moles/liter, CCL2 2,596.7 ⫾ 101.5 pg/ml [29.7% reduction; P ⬍ 0.05] and CCL5 151.0 ⫾ 3124 OHYANAGI ET AL Figure 5. Effects of Am80 on serum antimyosin antibody levels. Serum samples were obtained on day 22, and levels of IgG1 (A), IgG2a (B), and IgG2b (C) antimyosin antibodies from normal mice, mice with vehicle-treated experimental autoimmune myositis (EAM), and mice with experimental autoimmune myositis treated with Am80, used in the experiments shown in Figure 1F, were measured by enzyme-linked immunosorbent assay. Values are the mean and SEM from duplicate assays. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.01; ⴱⴱⴱ ⫽ P ⬍ 0.001, versus mice with vehicle-treated experimental autoimmune myositis. 17.0 pg/ml [60.6% reduction; P ⬍ 0.05]) (Figures 6A and B). Stimulation with IL-1␤ also significantly increased expression of CCL2 (2,282.4 ⫾ 149.2 pg/ml) and CCL5 (93.9 ⫾ 4.5 pg/ml). Levels of IL-1␤–induced CCL2 and CCL5 were reduced by treatment with Am80 (with Am80 at 10⫺8 moles/liter, CCL2 2,070.8 ⫾ 25.7 pg/ml [9.2% reduction; P not significant] and CCL5 88.8 ⫾ 8.6 pg/ml [5.4% reduction; P not significant], with Am80 at 10⫺7 moles/liter, CCL2 1,614.6 ⫾ 70.5 pg/ml [29.3% reduction; P ⬍ 0.05] and CCL5 80.9 ⫾ 7.5 pg/ml [13.8% reduction; P not significant], with Am80 at 10⫺6 moles/ liter, CCL2 1,556.9 ⫾ 56.9 pg/ml [31.8% reduction; P ⬍ 0.05] and CCL5 73.2 ⫾ 2.2 pg/ml [22.0% reduction; P ⬍ 0.05]), although the effect of Am80 on IL-1␤–induced CCL2 and CCL5 production was smaller than that on TNF␣-induced CCL2 and CCL5 production (Figures 6A and B). Without cytokine stimulation, spontaneous expression of CCL2 was also slightly reduced in the presence of Am80. No similar decrease in CCL5 expression was observed. We also determined the effects of Am80 on chemokine production by macrophages. Expression of CCL2 by splenic CD11b⫹ macrophages from normal mice with or without TNF␣ or IL-1␤ stimulation was not detected by ELISA. In contrast, CCL5 was expressed by macrophages (mean ⫾ SEM 117.7 ⫾ 13.7 pg/ml), and stimulation with TNF␣ and IL-1␤ up-regulated its expression (277.6 ⫾ 5.6 pg/ml and 177.0 ⫾ 6.1 pg/ml, respectively). Treatment with Am80 significantly reduced the expression of CCL5 (unstimulated 62.1 ⫾ 2.8 pg/ml [47.2% reduction], TNF␣-stimulated 140.7 ⫾ 9.4 [49.3% reduction], IL-1␤-stimulated 92.4 ⫾ 7.5 [47.8% reduction]) (Figure 6C). Effects of Am80 on AP-1 activity. Since it has been reported that retinoids inhibit AP-1 activity (30), we examined the effect of Am80 on AP-1 activity in mouse myoblasts stimulated with TNF␣ or IL-1␤ in vitro. The AP-1 activity level in TNF␣- and IL-1␤– stimulated myoblasts was 2.6 times and 1.7 times higher, respectively, than that in unstimulated cells. However, the up-regulation of AP-1 activity was significantly suppressed by Am80 (Figure 6D). Expression of RARs in mouse muscle tissue, myoblasts, and splenic macrophages. Am80 is a selective agonist for RAR␣ and RAR␤, but not for RAR␥ (29). To determine the expression of RARs in mouse muscle tissue, myoblasts, and splenic macrophages, we examined RAR␣, RAR␤, and RAR␥ expression, by RT-PCR. Although the levels of expression of RAR␤ and RAR␥ mRNA were lower in splenic macrophages, RETINOID AMELIORATES EXPERIMENTAL AUTOIMMUNE MYOSITIS 3125 all RARs were expressed in muscle tissue, myoblasts, and macrophages (Figure 6E). DISCUSSION Figure 6. Effects of Am80 on chemokine production and activator protein 1 (AP-1) activity, and expression of retinoic acid receptors (RARs) in mouse muscle tissue, myoblasts, and splenic macrophages. A and B, Myoblasts from normal mice were cultured with or without tumor necrosis factor ␣ (TNF␣) or interleukin-1␤ (IL-1␤) for 24 hours in the presence or absence of Am80, and the production of CCL2/ monocyte chemotactic protein 1 (MCP-1) (A) and CCL5/RANTES (B) in the culture supernatants was measured by enzyme-linked immunosorbent assay. Values are the mean and SEM from 2 independent experiments analyzed in triplicate. ⴱ ⫽ P ⬍ 0.05 versus vehicle-treated controls. C, Splenic macrophages from normal mice were cultured as described above, and the production of CCL5/ RANTES in the culture supernatants was measured by enzyme-linked immunosorbent assay. Values are the mean and SEM from 2 independent experiments analyzed in triplicate. ⴱ ⫽ P ⬍ 0.05. D, Mouse myoblasts were stimulated with TNF␣ or IL-1␤ for 2 hours in the presence or absence of Am80, and AP-1 activity was measured. Values are the mean and SEM from 3 independent experiments analyzed in duplicate. ⴱ ⫽ P ⬍ 0.05. NS ⫽ not significant. E, Total RNA was extracted from mouse muscle tissue, myoblasts, and splenic macrophages. Expression of mRNA for RAR␣, RAR␤, and RAR␥ was analyzed by reverse transcriptase–polymerase chain reaction. Results shown are representative of 3 experiments with similar findings. In this study, we found that treatment with a synthetic retinoid, Am80, reduced inflammatory cell infiltration and attenuated the expression of inflammatory cytokines in the muscles of mice with experimental autoimmune myositis. Moreover, Am80 promoted differentiation into Th1 and Th2, but Th17 differentiation was not altered. Am80 reduced antimyosin antibody production in vivo. In addition, it decreased chemokine expression by mouse myoblasts and macrophages in vitro. Treatment with Am80 reduced the numbers of inflammatory cells, including CD4⫹ and CD8⫹ T cells and F4/80⫹ macrophages, in muscle. It is thought that chemokines such as CCL2 and CCL5 are involved in leukocyte recruitment and activation at the site of inflammatory lesions (31). Enhanced expression of such chemokines in muscle tissue of patients with PM and DM has been reported (32–34). Am80 reduced TNF␣and IL-1␤–induced CCL2 and CCL5 expression in mouse myoblasts and suppressed CCL5 expression by macrophages. These findings suggest that Am80 could reduce chemokine expression in muscle tissue following abrogation of inflammatory cell infiltration into the muscle tissue. Previous data showed that retinoid interferes with the AP-1 signaling pathway through RARs (30). We also observed that AP-1 activity in TNF␣- or IL-1␤–stimulated mouse myoblasts was suppressed by Am80. Since NF-B and AP-1 cooperatively up-regulate chemokine expression (35), Am80 might suppress chemokine expression via interference with the AP-1 signaling pathway. In the present study, Am80 attenuated TNF␣ and IL-1␤ mRNA expression in muscle. These inflammatory cytokines are expressed by infiltrating inflammatory cells in the muscles of patients with PM and DM (36,37). Since the number of infiltrating inflammatory cells in the muscle of mice with experimental autoimmune myositis was reduced by Am80, inflammatory cytokine production might be attenuated in the muscles. Alternatively, Am80 could inhibit cytokine production directly. Retinoids have been shown to down-regulate TNF␣ expression on lipopolysaccharide-stimulated macrophages (38). In addition, retinoids down-regulated Toll-like receptor 2 and CD14 expression on monocytes and reduced expression of cytokines, such as TNF␣ and IL-6, 3126 by Toll-like receptor ligand–stimulated monocytes (39). Thus, suppression of chemokine and cytokine production in the muscle tissue might be one of the mechanisms mediating the effect of Am80 in mice with experimental autoimmune myositis. It has been reported that retinoids inhibit differentiation into Th1 and Th17, while they enhance Th2 and regulatory T cell differentiation in vitro (18–22). Interestingly, in our experiments, treatment with Am80 significantly increased IFN␥, IL-4, and IL-10 production by myosin-stimulated splenic T cells of mice with experimental autoimmune myositis, whereas production of IL-17 was not altered by Am80. These results indicate that Am80 enhances differentiation into Th1 and Th2, and does not affect Th17 differentiation in mice with experimental autoimmune myositis in vivo. In addition, Am80 might increase IL-10–producing T cells, since IL-10 production was enhanced by myosin-stimulated splenic T cells in Am80-treated mice with experimental autoimmune myositis. Since IL-10 has been thought to be an antiinflammatory or regulatory cytokine (40), increased IL-10 production as well as enhanced Th2 might also be a mechanism by which experimental autoimmune myositis is ameliorated by Am80. Our observations on the effect of Am80 on Th phenotype development in mice with experimental autoimmune myositis differed from findings of previously reported in vitro studies (18–22). The conflicting results might be due to differences between in vitro and in vivo experimental conditions. In vitro, retinoid affects only T cells and controls Th cell differentiation directly. In contrast, complex mechanisms, such as quantity and quality of existing antigens, surrounding APCs including dendritic cells, macrophages, and B cells, and differences in cytokine environment, might influence Th cell differentiation in vivo. In this regard, it has been reported that retinoid drives the differentiation of monocytes into dendritic cells with granulocyte–macrophage colony-stimulating factor (41). These retinoid-induced dendritic cells secrete IL-12 without the need for any maturation agent and can drive T cells toward IL-12– dependent Th1 cells that produce IFN␥ (41). Treatment with Am80 also attenuated the production of serum antimyosin antibodies of the IgG2a and IgG2b subclasses, but not the IgG1 subclass. Previous studies showed that retinoid could directly regulate IgG1 production in vitro (42,43). However, the effect of retinoid in vivo and its effect on IgG2a and IgG2b production have not been reported. Although the mechanism of the effect of Am80 was not clear, reduction of serum antimyosin antibodies might also have contrib- OHYANAGI ET AL uted to the attenuation of experimental autoimmune myositis. In conclusion, we demonstrated in the present study that Am80 prophylactically and therapeutically reduced experimental autoimmune myositis. This effect was probably due to regulation of Th differentiation, reduction of antimyosin antibody production, and decreased chemokine expression. ACKNOWLEDGMENTS We thank Fumiko Inoue and Aya Sato (Tokyo Medical and Dental University) for excellent technical support, and Yousuke Murakami and Tomohiro Morio (Tokyo Medical and Dental University) for excellent advice. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Nanki had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Ishido, Suzuki, Miyasaka, Nanki. Acquisition of data. Ohyanagi, Ishido, Suzuki, Kaneko, Kubota, Nanki. Analysis and interpretation of data. Ohyanagi, Kaneko, Kubota, Miyasaka, Nanki. REFERENCES 1. Arahata K, Engel AG. Monoclonal antibody analysis of mononuclear cells in myopathies. I. Quantitation of subsets according to diagnosis and sites of accumulation and demonstration and counts of muscle fibers invaded by T cells. Ann Neurol 1984;16:193–208. 2. Engel AG, Arahata K. Monoclonal antibody analysis of mononuclear cells in myopathies. II. Phenotypes of autoinvasive cells in polymyositis and inclusion body myositis. Ann Neurol 1984;16: 209–15. 3. Dalakas MC. Polymyositis, dermatomyositis and inclusion-body myositis. N Engl J Med 1991;325:1487–98. 4. Nagaraju K, Lundberg IE. Inflammatory diseases of muscle and other myopathies. In: Firestein GS, Budd RC, Harris ED Jr, McInnes IB, Ruddy S, Sergent JS, editors. Kelley’s textbook of rheumatology. 8th ed. Philadelphia: Saunders Elsevier; 2009. p. 1353–80. 5. Greenberg SA, Pinkus JL, Pinkus GS, Burleson T, Sanoudou D, Tawil R, et al. Interferon-␣/␤-mediated innate immune mechanisms in dermatomyositis. Ann Neurol 2005;57:664–78. 6. Goebels N, Michaelis D, Engelhardt M, Huber S, Bender A, Pongratz D, et al. Differential expression of perforin in muscleinfiltrating T cells in polymyositis and dermatomyositis. J Clin Invest 1996;97:2905–10. 7. Cherin P, Herson S, Crevon MC, Hauw JJ, Cervera P, Galanaud P, et al. Mechanisms of lysis by activated cytotoxic cells expressing perforin and granzyme-B genes and the protein TIA-1 in muscle biopsies of myositis. J Rheumatol 1996;23:1135–42. 8. Tews DS, Goebel HH. Cytokine expression profile in idiopathic inflammatory myopathies. J Neuropathol Exp Neurol 1996;55: 342–7. 9. Lepidi H, Frances V, Figarella-Branger D, Bartoli C, MachadoBaeta A, Pellissier JF. Local expression of cytokines in idiopathic RETINOID AMELIORATES EXPERIMENTAL AUTOIMMUNE MYOSITIS 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. inflammatory myopathies. Neuropathol Appl Neurobiol 1998;24: 73–9. Sugiura T, Kawaguchi Y, Harigai M, Takagi K, Ohta S, Fukasawa C, et al. Increased CD40 expression on muscle cells of polymyositis and dermatomyositis: role of CD40-CD40 ligand interaction in IL-6, IL-8, IL-15, and monocyte chemoattractant protein-1 production. J Immunol 2000;164:6593–600. Walsh RJ, Kong SW, Yao Y, Jallal B, Kiener PA, Pinkus JL, et al. Type I interferon–inducible gene expression in blood is present and reflects disease activity in dermatomyositis and polymyositis. Arthritis Rheum 2007;56:3784–92. Suzuki F, Nanki T, Imai T, Kikuchi H, Hirohata S, Kohsaka H, et al. Inhibition of CX3CL1 (fractalkine) improves experimental autoimmune myositis in SJL/J mice. J Immunol 2005;175:6987–96. Nemoto H, Nemoto K, Sugimoto H, Kinoshita M. FK506 suppressed the inflammatory change of EAM in SJL/J mice. J Neurol Sci 2001;193:7–11. Noy N. Physical-chemical properties and action of retinoids. In: Nau H, Blaner WS, editors. Retinoids: the biochemical and molecular basis of vitamin A and retinoid action. New York: Springer; 1999. p. 3–30. Gudas LJ. Retinoids and vertebrate development. J Biol Chem 1994;269:15399–402. Fisher GJ, Voorhees JJ. Molecular mechanisms of retinoid actions in skin. FASEB J 1996;10:1002–13. Hayes CE, Nashold FE, Gomez FE, Hoag KA. Retinoids and immunity. In: Nau H, Blaner WS, editors. Retinoids: the biochemical and molecular basis of vitamin A and retinoid action. New York: Springer; 1999. p. 589–610. Iwata M, Eshima Y, Kagechika H. Retinoic acids exert direct effects on T cells to suppress Th1 development and enhance Th2 development via retinoic acid receptors. Int Immunol 2003;15: 1017–25. Hoag KA, Nashold FE, Goverman J, Hayes CE. Retinoic acid enhances the T helper 2 cell development that is essential for robust antibody responses through its action on antigen-presenting cells. J Nutr 2002;132:3736–9. Kang BY, Chung SW, Kim SH, Kang SN, Choe YK, Kim TS. Retinoid-mediated inhibition of interleukin-12 production in mouse macrophages suppresses Th1 cytokine profile in CD4⫹ T cells. Br J Pharmacol 2000;130:581–6. Mucida D, Park Y, Kim G, Turovskaya O, Scott I, Kronenberg M, et al. Reciprocal TH17 and regulatory T cell differentiation mediated by retinoic acid. Science 2007;317:256–60. Elias KM, Laurence A, Davidson TS, Stephens G, Kanno Y, Shevach EM, et al. Retinoic acid inhibits Th17 polarization and enhances FoxP3 expression through a Stat-3/Stat-5 independent signaling pathway. Blood 2008;111:1013–20. Massacesi L, Castigli E, Vergelli M, Olivotto J, Abbamondi AL, Sarlo F, et al. Immunosuppressive activity of 13-cis-retinoic acid and prevention of experimental autoimmune encephalomyelitis in rats. J Clin Invest 1991;88:1331–7. Racke MK, Burnett D, Pak SH, Albert PS, Cannella B, Raine CS, et al. Retinoid treatment of experimental allergic encephalomyelitis: IL-4 production correlates with improved disease course. J Immunol 1995;154:450–8. Wang T, Niwa S, Bouda K, Matsuura S, Homma T, Shudo K, et al. The effect of Am-80, one of retinoids derivatives on experimental allergic encephalomyelitis in rats. Life Sci 2000;67:1869–79. Aranami T, Yamamura T. Th17 cells and autoimmune encephalomyelitis (EAE/MS). Allergol Int 2008;57:115–20. 3127 27. Notley CA, Inglis JJ, Alzabin S, McCann FE, McNamee KE, Williams RO. Blockade of tumor necrosis factor in collageninduced arthritis reveals a novel immunoregulatory pathway for Th1 and Th17 cells. J Exp Med 2008;205:2491–7. 28. Serada S, Fujimoto M, Mihara M, Koike N, Ohsugi Y, Nomura S, et al. IL-6 blockade inhibits the induction of myelin antigenspecific Th17 cells and Th1 cells in experimental autoimmune encephalomyelitis. Proc Natl Acad Sci U S A 2008;105:9041–6. 29. Miwako I, Kagechika H. Tamibarotene. Drugs Today (Barc) 2007;43:563–8. 30. Dedieu S, Lefebvre P. Retinoids interfere with the AP1 signalling pathway in human breast cancer cells. Cell Signal 2006;18:889–98. 31. Nanki T, Nagasaka K, Hayashida K, Saita Y, Miyasaka N. Chemokines regulate IL-6 and IL-8 production by fibroblast-like synoviocytes from patients with rheumatoid arthritis. J Immunol 2001;167:5381–5. 32. Confalonieri P, Bernasconi P, Megna P, Galbiati S, Cornelio F, Mantegazza R. Increased expression of ␤-chemokines in muscle of patients with inflammatory myopathies. J Neuropathol Exp Neurol 2000;59:164–9. 33. Liprandi A, Bartoli C, Figarella-Branger D, Pellissier JF, Lepidi H. Local expression of monocyte chemoattractant protein-1 (MCP-1) in idiopathic inflammatory myopathies. Acta Neuropathol 1999;97:642–8. 34. Adams EM, Kirkley J, Eidelman G, Dohlman J, Plotz PH. The predominance of ␤ (CC) chemokine transcripts in idiopathic inflammatory muscle diseases. Proc Assoc Am Physicians 1997; 109:275–85. 35. Martin T, Cardarelli PM, Parry GC, Felts KA, Cobb RR. Cytokine induction of monocyte chemoattractant protein-1 gene expression in human endothelial cells depends on the cooperative action of NF-B and AP-1. Eur J Immunol 1997;27:1091–7. 36. Lundberg I, Ulfgren AK, Nyberg P, Andersson U, Klareskog L. Cytokine production in muscle tissue of patients with idiopathic inflammatory myopathies. Arthritis Rheum 1997;40:865–74. 37. Figarella-Branger D, Civatte M, Bartoli C, Pellissier JF. Cytokines, chemokines, and cell adhesion molecules in inflammatory myopathies. Muscle Nerve 2003;28:659–82. 38. Wang X, Allen C, Ballow M. Retinoic acid enhances the production of IL-10 while reducing the synthesis of IL-12 and TNF-␣ from LPS-stimulated monocytes/macrophages. J Clin Immunol 2007;27:193–200. 39. Liu PT, Krutzik SR, Kim J, Modlin RL. Cutting edge: all-trans retinoic acid down-regulates TLR2 expression and function. J Immunol 2005;174:2467–70. 40. Abbas AK, Lichtman AH. Cytokine. In: Abbas AK, Lichtman AH, editors. Cellular and molecular immunology. 5th ed. Philadelphia: Elsevier Saunders; 2005. p. 243–74. 41. Mohty M, Morbelli S, Isnardon D, Sainty D, Arnoulet C, Gaugler B, et al. All-trans retinoic acid skews monocyte differentiation into interleukin-12-secreting dendritic-like cells. Br J Haematol 2003; 122:829–36. 42. Tokuyama H, Tokuyama Y. The regulatory effects of all-transretinoic acid on isotype switching: retinoic acid induces IgA switch rearrangement in cooperation with IL-5 and inhibits IgG1 switching. Cell Immunol 1999;192:41–7. 43. Chen Q, Ross AC. Vitamin A and immune function: retinoic acid modulates population dynamics in antigen receptor and CD38stimulated splenic B cells. Proc Natl Acad Sci U S A 2005;102: 14142–9.