Interferon regulatory factor 5 is critical for the development of lupus in MRLlpr mice.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 63, No. 3, March 2011, pp 738–748 DOI 10.1002/art.30183 © 2011, American College of Rheumatology Interferon Regulatory Factor 5 Is Critical for the Development of Lupus in MRL/lpr Mice Yoshifumi Tada,1 Seiji Kondo,2 Shigehisa Aoki,1 Syuichi Koarada,1 Hisako Inoue,1 Rie Suematsu,1 Akihide Ohta,1 Tak W. Mak,3 and Kohei Nagasawa1 Objective. Interferon regulatory factor 5 (IRF-5) is a transcription factor that mediates intracellular signals activated by engagement of Toll-like receptors (TLRs). IRF5 polymorphisms are associated with an increased or decreased risk of systemic lupus erythematosus (SLE) in various human populations, but the precise role of IRF5 in SLE development is not fully understood. This study was undertaken to examine the role of IRF5 in the development of murine lupus. Methods. We crossed gene-targeted IRF5deficient (IRF5ⴚ/ⴚ) mice with MRL/MpJ-lpr/lpr (MRL/ lpr) mice and examined the progeny for survival, glomerulonephritis, autoantibody levels, immune system cell populations, and dendritic cell function. Results. IRF5ⴚ/ⴚMRL/lpr mice survived longer than control IRF5ⴙ/ⴙMRL/lpr mice and displayed only very mild glomerulonephritis. Autoantibodies to SLErelated nuclear antigens were lower in IRF5ⴚ/ⴚMRL/lpr mouse serum, and numbers of activated CD4ⴙ T cells were reduced in the spleen. Splenic DCs from IRF5ⴚ/ⴚ MRL/lpr mice produced lower levels of inflammatory cytokines when treated in vitro with TLR-7 or TLR-9 ligands or immune complexes. Interferon-␣ production in response to CpG was also decreased. Conclusion. Our results show that IRF5 is a crucial driver of lupus development in mice, and indicate that IRF5 may be an attractive new target for therapeutic intervention to control disease in SLE patients. Systemic lupus erythematosus (SLE) is a systemic inflammatory autoimmune disease in humans that affects multiple organs and is characterized by production of autoantibodies. These autoantibodies are mainly directed against nuclear antigens such as double-stranded DNA (dsDNA), or against RNA-containing proteins such as the Sm antigen or RNP (1). Attack by these autoantibodies and immune cells results in damage to multiple organs, including the kidney, skin, joints, central nervous system (CNS), and vascular system. The damage to the kidney in SLE takes the form of glomerulonephritis characterized by the infiltration of activated inflammatory cells and deposition of immune complexes (ICs) (2). The mechanism driving SLE is not yet fully understood, but various studies have suggested that type I interferon (IFN) plays a role in its pathogenesis. Serum type I IFN levels are elevated in SLE patients (3,4), and the expression of IFN-responsive genes is elevated in peripheral blood cells from these individuals (5,6). Furthermore, the frequencies of nephritis, hematopoietic disorders, and CNS disease are highest in SLE patients whose peripheral blood cells show the greatest increases in expression of IFN-responsive genes (5,6). Recent genetic studies of SLE patients have revealed that IRF5 gene polymorphisms are associated with an increased risk of SLE (7–12). IRF5 is a member of the IRF family of transcription factors that induce the expression of IFN-responsive genes. Expression of the IRF proteins is triggered by viral infection, IFNs, cytokines, and Toll-like receptor (TLR) ligands (13). Upon stimulation of cells by viruses or TLR ligands, IRF5 is activated and translocates to the nucleus, where it Supported by the Ministry of Education, Culture, Sports, Science, and Technology of Japan (grant 22590893) and by a research project grant from the Faculty of Medicine, Saga University. 1 Yoshifumi Tada, MD, PhD, Shigehisa Aoki, MD, PhD, Syuichi Koarada, MD, PhD, Hisako Inoue, MD, PhD, Rie Suematsu, MD, Akihide Ohta, MD, PhD, Kohei Nagasawa MD, PhD: Saga University, Saga, Japan; 2Seiji Kondo, MD, PhD: National Hospital Organization Kyushu Medical Center, Fukuoka, Japan; 3Tak W. Mak, PhD: Campbell Family Institute for Breast Cancer Research at Princess Margaret Hospital and University Health Network, Toronto, Ontario, Canada. Drs. Tada and Kondo contributed equally to this work. Address correspondence to Yoshifumi Tada, MD, PhD, Department of Rheumatology, Faculty of Medicine, Saga University, 5-1-1 Nabeshima, Saga 849-8501, Japan. E-mail: email@example.com. Submitted for publication May 20, 2010; accepted in revised form November 30, 2010. 738 AMELIORATION OF LUPUS IN IRF5-DEFICIENT MRL/lpr MICE induces the expression of IFN and chemokine genes (14,15). Studies of IRF5⫺/⫺ mice have demonstrated that IRF5 is critically involved in the production of proinflammatory cytokines, particularly tumor necrosis factor ␣ (TNF␣) and interleukin-6 (IL-6), by dendritic cells (DCs) or macrophages treated with ligands of TLR-3, TLR-4, TLR-7, or TLR-9 (16,17). Although it is widely accepted that IRF5 is an important genetic factor in the development of SLE (18), its actual contribution to the development of lupus or other forms of autoimmunity is unclear. To address this issue, we investigated the role of IRF5 in a murine model of SLE, the MRL/lpr strain. These mutants spontaneously develop a systemic autoimmune disorder (murine lupus) that closely resembles SLE in humans. The clinical features of murine lupus include autoantibody production, hypergammaglobulinemia, glomerulonephritis, arthritis, vasculitis, splenomegaly, and lymphadenopathy (19–21). The molecular defect underlying this phenotype is a mutation in the gene encoding Fas, a cell surface receptor that triggers apoptosis (22,23). In this study, we generated IRF5-deficient MRL/lpr mice and analyzed their phenotype. In IRF5⫺/⫺MRL/lpr mice, autoantibody production was markedly reduced, glomerulonephritis was mild, and mouse survival was much improved. Production of proinflammatory cytokines and IFN␣ by splenic DCs from IRF5⫺/⫺MRL/lpr mice in response to treatment with TLR-7 and TLR-9 ligands or ICs was decreased. Our work demonstrates that IRF5 is required for the development of murine lupus in MRL/lpr mice, and supports the findings of genetic studies indicating that IRF5 is an important contributor to human SLE. MATERIALS AND METHODS Mice. Homozygous MRL/lpr mice were purchased from Charles River Japan. Gene-targeted IRF5-deficient mice (16) were backcrossed to MRL/lpr mice for ⬎8 generations and genotyped by polymerase chain reaction (PCR) using DNA obtained by tail biopsy. Survival analyses were performed using female mice backcrossed for 8 generations, and other experiments were performed using male and female mice backcrossed for ⬎10 generations. Littermate IRF5⫹/⫹ mice were used as controls. All mice were bred and maintained in the Saga University animal facility, and all animal care was carried out in accordance with institutional guidelines. Measurement of autoantibodies and serum Ig. Antinuclear antibodies (ANAs) were detected by indirect immunofluorescence using HEp-2 cell slides (Orgentec) and fluorescein-conjugated anti-mouse IgG (Santa Cruz Biotechnology). Sera were diluted at 1:100 and 1:500, and the fluorescence intensity of nuclear staining was scored on a scale of 0–3 by 2 independent observers in a blinded manner. 739 Anti-dsDNA antibodies were detected by enzymelinked immunosorbent assay (ELISA) as described previously (24). Briefly, ELISA plates were coated with calf dsDNA and blocked with phosphate buffered saline (PBS) containing 1% bovine serum albumin (BSA). Mouse serum samples were then serially diluted and added to the plates for a 1-hour incubation at 37°C. Plates were washed, and peroxidase-conjugated goat anti-mouse IgG1, IgG2a, IgG3, or IgM (Southern Biotechnology Associates) was added for 1 hour at 37°C. Twofold dilutions of standard serum were added to each plate as an internal control. A standard curve was constructed in which the standard serum was defined as 100 arbitrary units, and the antibody titers of experimental serum samples were calculated. For measurement of anti-Sm and anti-RNP antibodies, ELISA plates were precoated with purified Sm antigen or recombinant human RNP (MBL). Primary antibody binding was detected using peroxidase-conjugated goat anti-mouse IgG (Southern Biotechnology Associates). Mouse serum levels of IgG1, IgG2a, IgG3, and IgM antibodies were measured by ELISA using a clonotyping system and a mouse Ig standard panel according to the recommendations of the manufacturer (Southern Biotechnology Associates). Histopathologic analysis, immunofluorescence, and immunostaining. For histopathologic analysis, mouse kidneys were fixed in buffered 10% formalin, embedded in paraffin, sectioned, and stained with periodic acid–Schiff, periodic acid–methenamine–silver (PAM), or hematoxylin and eosin. Glomerular lesion severity (mesangial hypercellularity, proliferative changes, hyalinosis, exudates, necrosis, and/or crescent formation) was scored on a scale of 0–3 by a pathologist (SA) in a blinded manner as described previously (25). Perivascular inflammation and tubular lesions (atrophy, dilation, and necrosis) were also scored in the same manner. For immunofluorescence, mouse kidneys were embedded in TissueTek medium (Miles) and frozen in hexane chilled in dry ice plus acetone. Sections were cut using a cryostat, fixed in acetone, rinsed with PBS, and incubated with F(ab⬘)2 fluorescein-conjugated anti-mouse IgG (Santa Cruz Biotechnology) or with F(ab⬘)2 fluorescein-conjugated antimouse C3 (MP Biomedicals). Glomerular deposition of IgG and C3 was evaluated on a scale of 0–3 as described previously (26). For immunostaining, cryostat sections were incubated with anti-CD4, anti-CD8, and anti-B220 rat anti-mouse monoclonal antibodies (mAb; BD Biosciences) to detect T cells, and with anti-F4/80 rat anti-mouse mAb (Abcam) to detect macrophages. Antibodies were visualized using an immunoperoxidase method. Cellular infiltration in glomeruli, periglomeruli, and interstitial lesions was scored on a scale of 0–4, where 0 ⫽ none, 1 ⫽ ⬍5 cells, 2 ⫽ 5–10 cells, 3 ⫽ 11–20 cells, and 4 ⫽ ⬎21 cells per glomerulus or per 0.1 mm2. The mean ⫾ SEM scores for ⬎15 glomeruli or tissue areas were calculated. Real-time quantitative reverse transcriptase–PCR (RT-PCR). Levels of TNF␣, IL-6, IFN␥, and CCL5 (RANTES) messenger RNA (mRNA) in mouse kidneys were measured by real-time RT-PCR. Briefly, total RNA was extracted from the cortex of kidneys that were snap-frozen in liquid nitrogen. Quantitative PCR was performed using the SYBR Green method and a LightCycler (Roche Diagnostics, Penzberg, Germany). Primers were from Nihon Gene Research Labora- 740 TADA ET AL Purification of ICs. ICs were prepared from the sera of aged MRL/lpr mice. Pooled sera were passed through a 0.22-m filter and applied to an antibody purification spin column (Proteus Protein-G Kit; Prochem). Eluted IgG were desalted, and ICs were isolated and concentrated using a centrifugal filter device with a 300,000 molecular weight cutoff (Pall) and dialyzed against DMEM. Anti-dsDNA antibody Figure 1. Survival curves for female IRF5⫹/⫹MRL/lpr mice (n ⫽ 10), IRF5⫹/⫺MRL/lpr mice (n ⫽ 26), and IRF5⫺/⫺MRL/lpr mice (n ⫽ 16), determined by the Kaplan-Meier method. Loss of IRF5 prolonged the survival of MRL/lpr mice. tories. Relative mRNA expression was determined and normalized to the expression of the internal housekeeping gene ␤-actin. Flow cytometric analysis. Mouse spleen cells were suspended in PBS containing 1% BSA and 0.1% sodium azide and incubated with various combinations of conjugated antibodies as described previously (24). Anti-CD4, anti-CD8, antiCD11c, anti-CD11b, anti-CD44, and anti-CD62L mAb were purchased from BD Bioscience; anti-CD80 and anti-CD86 mAb were obtained from Caltag; anti-B220 and anti-CD3 mAb were from Beckman Coulter; and anti–PDCA-1 mAb was from Miltenyi Biotec. FoxP3 staining was performed using a kit (eBioscience). At least 3 ⫻ 105 immunostained cells per sample were collected using a FACSCalibur and analyzed with CellQuest software (BD Biosciences). Analysis of DC cytokine production. CD11c⫹ DCs were isolated from total mouse spleen cells by magnetic cell sorting using anti-CD11c microbeads (Miltenyi Biotec). Briefly, spleens from 2–3 mice/group (10–12 weeks old) were digested with collagenase D (Roche Diagnostics). Mouse spleen cells were blocked by incubation with anti-CD16/ 32 mAb, labeled with anti-CD11c beads, and applied to a magnetic separation column. After washing, the CD11c⫹ DCs were eluted from the column by flushing. The resulting cell purity was ⬎90% as determined by flow cytometry. Conventional DCs and plasmacytoid DCs (PDCs) were identified by staining with anti-CD11c plus anti-B220 or anti–PDCA-1, respectively. For DC stimulation, DCs were resuspended in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal calf serum, seeded in 96-well plates at 3 ⫻ 105/well, and stimulated for 24 hours with either TLR ligands or ICs. TLR ligands (poly[I-C], lipopolysaccharide, flagellin, loxoribine, oligodeoxynucleotide (ODN) 1585, and ODN 1826) were all purchased from InvivoGen, and ICs were isolated from mouse serum as described below. Cytokines secreted into culture supernatants were measured using a cytometric bead assay system (BD Bioscience) and a FACSCalibur instrument. IFN␣ levels were measured using an ELISA kit (PBL Biomedical). Figure 2. Decreased serum autoantibody and Ig levels in IRF5⫺/⫺ MRL/lpr mice. A, Detection of antinuclear antibodies (ANAs) by immunofluorescence. Serum samples (1:100 dilution) from 18-weekold IRF5⫹/⫹MRL/lpr or IRF5⫺/⫺MRL/lpr mice were applied to HEp-2 cell–bearing slides, and the binding of ANAs was detected by indirect immunofluorescence. Representative stainings are shown. Original magnification ⫻ 400. B, Fluorescence staining intensity scores. The slides described in A were scored for the intensity of fluorescence staining of sera (1:100 or 1:500 dilutions), as described in Materials and Methods. C and D, Levels of anti–double-stranded DNA (antidsDNA) (C) and IgG anti-Sm and IgG anti-RNP (D) antibodies (Ab) in IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mouse serum, detected by enzyme-linked immunosorbent assay. Horizontal lines in B–D show the median. E, Serum levels of total IgG1, IgG2a, IgG3, and IgM in 18-week-old IRF5⫹/⫹MRL/lpr mice (n ⫽ 12) and IRF5⫺/⫺MRL/lpr mice (n ⫽ 11). Bars show the mean ⫾ SEM. ⴱ ⫽ P ⬍ 0.01; ⴱⴱ ⫽ P ⬍ 0.005; ⴱⴱⴱ ⫽ P ⬍ 0.001, by Mann-Whitney U test in B–D and by Student’s t-test in E. AU ⫽ arbitrary units. AMELIORATION OF LUPUS IN IRF5-DEFICIENT MRL/lpr MICE 741 Figure 3. Amelioration of glomerulonephritis in IRF5⫺/⫺MRL/lpr mice. Kidneys from 18-week-old IRF5⫹/⫹ MRL/lpr mice (n ⫽ 13) and IRF5⫺/⫺MRL/lpr mice (n ⫽ 12) were examined for renal disease. A, Representative images of glomerular lesions from IRF5⫹/⫹MRL/lpr mice (i–iii) and IRF5⫺/⫺MRL/lpr mice (iv–vi), stained with periodic acid–Schiff (PAS) (i and iv), periodic acid–methenamine–silver (PAM) (ii and v), or hematoxylin and eosin (H&E) (iii and vi). Original magnification ⫻ 200 in i, ii, iv, and v; ⫻ 40 in iii and vi. B, Glomerular, vascular, and tubular disease scores. The slides described in A were scored on a scale of 0–3. C, Serum blood urea nitrogen (BUN) levels in 18-week-old IRF5⫹/⫹MRL/lpr mice (n ⫽ 12) and IRF5⫺/⫺MRL/lpr mice (n ⫽ 11). D, Deposition of IgG and C3 in the glomeruli of IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice at 18 weeks of age. Representative images are shown. Original magnification ⫻ 100. E, Staining intensity scores for IgG and C3 (n ⫽ 7 mice per group). Bars show the mean ⫾ SEM. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.005; ⴱⴱⴱ ⫽ P ⬍ 0.001; ⴱⴱⴱⴱ ⫽ P ⬍ 0.0001, by Student’s t-test. levels and IgG concentrations of purified ICs were determined by ELISA. Statistical analysis. Statistical analyses were performed using GraphPad Prism 5.01 software. In most experiments, Student’s unpaired 2-tailed t-test was used to determine the statistical significance of differences between groups. The Mann-Whitney U test was used for statistical analysis of differences in autoantibody levels, cytokine mRNA levels measured by RT-PCR, and renal immunostaining scores. Mouse survival was analyzed by the Kaplan-Meier method and the log rank test. RESULTS Prolonged survival of IRF5-deficient MRL/lpr mice. We first determined whether loss of IRF5 would have any effect on the survival of female MRL/lpr mice. IRF5⫺/⫺MRL/lpr mice lived much longer than IRF5⫹/⫹ MRL/lpr littermate controls (Figure 1). All IRF5⫹/⫹ MRL/lpr mice died by 52 weeks of age, whereas only 12.5% of the IRF5⫺/⫺MRL/lpr mice had died by that age. This effect on survival was dose-dependent, as evidenced by the finding that 50% of IRF5⫹/⫺MRL/lpr mice lived longer than 52 weeks (P ⬍ 0.0001). These results suggest that IRF5 promotes disease progression and contributes to the premature death of MRL/lpr mice. Decreased autoantibody levels in IRF5-deficient MRL/lpr mice. We measured concentrations of autoantibodies and Ig in the sera of IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice at 18 weeks of age. Whereas ANA in the sera of IRF5⫹/⫹MRL/lpr mice was readily detected by its strong fluorescence intensity, sera from IRF5⫺/⫺MRL/lpr mice showed only weak ANA staining (Figure 2A). This pattern held when sera were diluted to 1:100 or 1:500 (P ⬍ 0.001 and P ⬍ 0.005, respectively) (Figure 2B). Serum levels of the IgG subclasses of antidsDNA antibodies were also significantly decreased in IRF5⫺/⫺MRL/lpr mice compared to IRF5⫹/⫹MRL/lpr mice (P ⬍ 0.01 for IgG1; P ⬍ 0.001 for IgG2a) (Figure 2C). The IgG3 subclass of anti-dsDNA antibodies was undetectable in IRF5⫺/⫺MRL/lpr mice. In contrast, levels of anti-dsDNA antibodies of the IgM subclass were similar in IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice. Notably, neither IgG anti-Sm antibodies nor IgG anti-RNP antibodies were detectable in the sera of IRF5⫺/⫺MRL/lpr mice (Figure 2D). With respect to serum 742 TADA ET AL Figure 4. Decreased immune cell infiltration and production of inflammatory cytokines in the absence of IRF5. Cell infiltration into the kidneys of 18-week-old IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice was examined. A, Representative images of kidney tissue, showing glomerular or interstitial infiltration of CD4⫹, CD8⫹, and B220⫹ T cells and F4/80⫹ macrophages. Arrows indicate positively stained cells. Original magnification ⫻ 400. B, Scores for infiltration of CD4⫹, CD8⫹, and B220⫹ T cells and F4/80⫹ macrophages. Bars show the mean ⫾ SEM (n ⫽ 6 mice per group). C, Levels of mRNA for tumor necrosis factor ␣ (TNF␣), interleukin-6 (IL-6), CCL5, and interferon-␥ (IFN␥) in the kidney cortex of IRF5⫹/⫹MRL/lpr mice (n ⫽ 5) and IRF5⫺/⫺MRL/lpr mice (n ⫽ 4), assessed by reverse transcriptase–polymerase chain reaction. Bars show the mean ⫾ SEM ratio of cytokine mRNA to ␤-actin mRNA. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.005, by Mann-Whitney U test. Ig, IgG3 levels were significantly decreased in the absence of IRF5 (P ⬍ 0.001), but serum IgG1, IgG2a, and IgM levels were comparable to those in controls (Figure 2E). These data indicate that the production of autoantibodies, including ANA, anti-dsDNA antibody, anti-Sm antibody, and anti-RNP antibody, is dramatically attenuated in MRL/lpr mice deficient in IRF5. Amelioration of glomerulonephritis in IRF5deficient MRL/lpr mice. To assess the effect of IRF5 on autoimmune phenotypes in mouse organs, we scored the pathologic severity of glomerular, vascular, and tubular lesions in the kidneys of 18-week-old IRF5⫹/⫹MRL/ lpr and IRF5⫺/⫺MRL/lpr mice. IRF5⫹/⫹MRL/lpr mice showed severe glomerulonephritis at this age, exhibiting mesangial proliferation, necrosis, lobulation, and crescent formation in glomeruli (Figure 3A, part i). Both destruction and neoformation of basement membranes were detected in these control animals by PAM staining (Figure 3A, part ii). In contrast, glomerular lesions in IRF5⫺/⫺MRL/lpr mice were very mild, with only low to moderate mesangial proliferation and almost intact basement membranes (Figure 3A, parts iv and v). Whereas typical dense perivascular infiltration of leukocytes was apparent in IRF5⫹/⫹MRL/lpr mice (Figure 3A, part iii), IRF5⫺/⫺ MRL/lpr mice showed minimal or only mild infiltration around vessels (Figure 3A, part vi). The total pathologic scores for the glomerular, vascular, and tubular lesions in IRF5⫺/⫺MRL/lpr mice were significantly lower than those in IRF5⫹/⫹MRL/lpr controls (P ⬍ 0.001, P ⬍ 0.0001, and P ⬍ 0.05, respectively) (Figure 3B). Moreover, serum blood urea nitrogen concentration, an indicator of renal function, was significantly reduced in IRF5⫺/⫺MRL/lpr mice at 18 weeks of age compared to IRF5⫹/⫹MRL/lpr mice (mean ⫾ SEM 29.7 ⫾ 5.4 mg/dl versus 40.2 ⫾ 7.9 mg/dl; P ⬍ 0.005) (Figure 3C). Deposition of IgG and complement component C3 in glomeruli is a characteristic feature of the glomerulonephritis observed in MRL/lpr mice. However, in contrast to the intense deposition of IgG and C3 in the glomeruli of IRF5⫹/⫹MRL/lpr mice, histologic analysis of the glomeruli of IRF5⫺/⫺MRL/lpr mice revealed only mild to moderate deposition of IgG, and no or very little C3 deposition (Figure 3D). Accordingly, the glomerular deposition scores for IgG and C3 were significantly decreased in IRF5⫺/⫺MRL/lpr mice compared to IRF5⫹/⫹MRL/lpr mice (P ⬍ 0.05 and P ⬍ 0.001, respectively) (Figure 3E). AMELIORATION OF LUPUS IN IRF5-DEFICIENT MRL/lpr MICE 743 Figure 5. Altered splenic lymphocyte and dendritic cell (DC) populations in the absence of IRF5. A, Decreased splenomegaly and lymph node (LN) cellularity. Spleen weight (left) and the total cell numbers of axillary and inguinal LNs (right) in 18-week-old IRF5⫹/⫹MRL/lpr mice (n ⫽ 13) and IRF5⫺/⫺MRL/lpr mice (n ⫽ 12) are shown. Horizontal lines show the median. B, Levels of B220⫹CD3⫹, B220⫺CD3⫹, CD19⫹, CD4⫹, and CD8⫹ cells in IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice. Splenocytes were immunostained to detect the indicated markers. C, Altered subpopulations of splenic CD4⫹ T cells. CD4⫹ T cells were fractionated into naive, activated, memory, unique abnormal (B220⫹), and regulatory (FoxP3⫹) subsets. D, Levels of macrophages (CD11b⫹CD11c⫹), conventional DCs (cDCs; CD11c⫹B220⫺), and plasmacytoid DCs (PDCs; CD11c⫹PDCA1⫹) in total splenocytes from IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice (n ⫽ 8 per group), detected by immunostaining. E, Levels of CD80 and CD86 on conventional DCs and PDCs. RIgG ⫽ antibody binding control. In B–D, bars show the mean ⫾ SEM percentage of total cells. In E, bars show the mean ⫾ SEM mean fluorescence intensity (MFI). ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.01; ⴱⴱⴱ ⫽ P ⬍ 0.001; ⴱⴱⴱⴱ ⫽ P ⬍ 0.0001, by Student’s t-test. We next identified the types of cells infiltrating the mouse kidney by immunostaining with anti-CD4, anti-CD8, and anti-B220 mAb to detect T cells, and with anti-F4/80 mAb to detect macrophages. As previously reported, anti-B220–positive cells in the MRL/lpr mouse kidney are unique double-negative T cells (27). The glomeruli of IRF5⫹/⫹MRL/lpr mice showed many infiltrating cells, which marker analysis identified as mainly CD4⫹ T cells and macrophages (Figure 4A, top). In contrast, there was very little inflammatory cell infiltration in the glomeruli of IRF5⫺/⫺MRL/lpr mice (Figure 4A, bottom). Analysis of cellular invasion into the glomeruli and the interstitial area showed that the infiltration of CD4⫹ T cells and macrophages into the glomeruli was significantly decreased in the absence of IRF5 (Figure 4B). These data indicate that IRF5 promotes inflammatory cell infiltration in the kidneys of MRL/lpr mice. Previous work has shown that inflammatory cytokines and chemokines are up-regulated in the kidneys of MRL/lpr mice (28,29). We therefore compared levels of various cytokine transcripts in kidney cortices from IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice. Levels of TNF␣, IL-6, and CCL5 mRNA were significantly reduced in the kidneys of IRF5⫺/⫺MRL/lpr mice compared to controls (P ⬍ 0.05 for all), whereas IFN␥ expression was comparable in the kidneys of IRF5⫹/⫹ MRL/lpr and IRF5⫺/⫺MRL/lpr mice (Figure 4C). Thus, IRF5 is important for inflammatory cytokine and chemokine expression. Taken together, these results indicate that the progression of glomerulonephritis in MRL/ lpr mice is mitigated in the absence of IRF5. Altered lymphocyte and DC populations in the spleens of IRF5-deficient MRL/lpr mice. MRL/lpr mice exhibit autoimmune symptoms in various lymphoid organs, including splenomegaly. We therefore compared spleen weights, lymph node (LN) cell numbers, and splenic lymphocyte populations in IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice. Splenomegaly was significantly reduced in the absence of IRF5 (mean ⫾ SEM 0.59 ⫾ 0.13 gm versus 0.82 ⫾ 0.28 gm; P ⬍ 0.05) (Figure 5A, left). The total number of LN cells also tended to be lower in IRF5⫺/⫺MRL/lpr mice although the difference was not statistically significant (Figure 5A, right). Flow cytometric analysis of total mouse spleen cell populations revealed that the numbers of CD4⫹ T cells 744 TADA ET AL were significantly decreased (P ⬍ 0.01), but CD19⫹ B cells and CD8⫹ T cells were significantly increased (P ⬍ 0.0001 and P ⬍ 0.001, respectively), in IRF5⫺/⫺ MRL/lpr mice (Figure 5B). Among CD4⫹ T cells, the memory (CD62LlowCD44high; P ⬍ 0.05) and activated (CD62LhighCD44high; P ⬍ 0.01) CD4⫹ T cell populations were decreased in the absence of IRF5, as were B220⫹ activated CD4⫹ T cells (P ⬍ 0.05) (Figure 5C). In contrast, the naive (CD62LhighCD44low) and regulatory (FoxP3⫹) CD4⫹ T cell subpopulations showed no marked change (Figure 5C). In addition to altered T cell populations, the spleens of IRF5⫺/⫺MRL/lpr mice showed increases in the percentages of CD11b⫹CD11c⫹ macrophages and CD11c⫹PDCA-1⫹ PDCs (P ⬍ 0.01 and P ⬍ 0.05, respectively) (Figure 5D). However, the percentage of CD11c⫹B220⫺ conventional DCs was similar to that in spleens of IRF5⫹/⫹MRL/lpr mice. When we evaluated the expression of the costimulatory molecules CD80 and CD86 on DCs to assess their activation status, we found that CD86 could be detected on both conventional DCs and PDCs in both groups of mice whereas CD80 could not, and that the expression levels of these molecules were similar (Figure 5E). These data indicate that activated CD4⫹ T cells were decreased in the absence of IRF5 in MRL/lpr mice. Defective cytokine production by DCs from IRF5deficient MRL/lpr mice in response to TLR-7 and TLR-9 engagement. Previous studies have shown that DCs from IRF5⫺/⫺ mice produce decreased levels of IFNs and other cytokines in response to treatment with TLR ligands (16,17). We investigated whether loss of IRF5 had a similar effect on TLR-mediated cytokine production by splenic DCs from MRL/lpr mice. We first evaluated the percentages of conventional DCs and PDCs among splenic CD11c⫹ DCs purified from Figure 6. Defective cytokine production by splenic dendritic cells (DCs) in the absence of IRF5. CD11c⫹ DCs were purified from the spleens of 12–14-week-old IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice (n ⫽ 2–3 mice per experiment). DCs were stimulated in vitro for 24 hours with Toll-like receptor (TLR) ligands or immune complexes (ICs). A, Levels of tumor necrosis factor ␣ (TNF␣), interleukin-6 (IL-6), IL-10, interferon-␥ (IFN␥), and monocyte chemotactic protein 1 (MCP-1) in mouse DCs stimulated with the TLR ligands poly(I-C) (10 g/ml), lipopolysaccharide (LPS; 1 g/ml), flagellin (1 g/ml), loxoribine (1 mM), and CpG (1 M). Supernatants were collected and concentrations of the indicated cytokines were measured by cytometric bead assay. B, Levels of IFN␣ in mouse DCs stimulated with the TLR ligands poly(I-C) (10 g/ml) and CpG (1 M). Supernatants were collected and concentrations of the indicated cytokines were measured by enzyme-linked immunosorbent assay. C, Levels of TNF␣ and IL-6 in mouse DCs stimulated with ICs purified from IRF5⫹/⫹MRL/lpr mouse serum. Supernatants were collected and concentrations of the indicated cytokines were measured by cytometric bead assay. Bars show the mean ⫾ SEM cytokine concentrations from 5 experiments. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.01; ⴱⴱⴱ ⫽ P ⬍ 0.001, by Student’s t-test. AMELIORATION OF LUPUS IN IRF5-DEFICIENT MRL/lpr MICE IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice but found no difference (for IRF5⫹/⫹MRL/lpr mice, mean ⫾ SEM 61.2 ⫾ 9.8% for conventional DCs and 29.9 ⫾ 8.9% for PDCs; for IRF5⫺/⫺MRL/lpr mice, mean ⫾ SEM 58.7 ⫾ 8.3% for conventional DCs and 32.4 ⫾ 6.9% for PDCs; n ⫽ 5 mice per group). Levels of TNF␣, IL-6, and IL-10 production by splenic DCs from IRF5⫺/⫺MRL/lpr mice were markedly decreased in response to the TLR-9 ligand CpG (type B) and TLR-7 ligand loxorbine (Figure 6A). Levels of IFN␥ and monocyte chemotactic protein 1 were also reduced in response to CpG. Similarly, IFN␣ production by IRF5⫺/⫺MRL/lpr DCs in response to treatment with CpG (type A) was decreased (Figure 6B). Thus, as was the case in mice of a nonautoimmune background, IRF5 deficiency attenuated inflammatory cytokine and chemokine production in MRL/lpr mice. It has recently been demonstrated that ICs containing DNA or RNA can activate DCs and B cells via engagement of TLRs and Fc␥ receptors (30–33). To examine the impact of IRF5 loss on IC effects, we purified ICs from the sera of MRL/lpr mice, applied them to mouse splenic DCs, and measured cytokine production. Although TNF␣ and IL-6 production by splenic DCs from IRF5⫹/⫹MRL/lpr mice was enhanced by stimulation with ICs in a dose-dependent manner, IC-induced cytokine production was significantly reduced in DCs from IRF5⫺/⫺MRL/lpr mice (P ⬍ 0.05) (Figure 6C). We did not detect IFN␣ production in response to ICs from DCs of IRF5⫹/⫹MRL/lpr and IRF5⫺/⫺MRL/lpr mice. Taken together, these data suggest that IRF5 promotes the activation of DCs that are stimulated by ICs or by TLR-7 or TLR-9 engagement, and that, in the absence of IRF5, there is a deficit in proinflammatory cytokine production that decreases the autoimmune symptoms of MRL/lpr mice. DISCUSSION Since 2005, many reports have been published demonstrating an association between IRF5 polymorphisms and an increased or decreased risk of SLE in humans (7–12). In contrast to the wealth of genetic findings, little has been discovered about the actual role of the IRF-5 protein in autoimmunity and the development of autoimmune diseases. In SLE patients, certain IRF5 polymorphisms are associated with increased expression of IFN-responsive genes as well as with elevated serum IFN␣ levels (11,34). ICs from the serum of SLE patients induced the production of IL-6 and type I IFN by murine DCs, and this effect was abolished in 745 DCs lacking IRF5 or IRF7 (35). In this study, we demonstrated that IRF5 deficiency reduced the production of autoantibodies and attenuated the development of glomerulonephritis such that mouse survival was greatly improved. Thus, we propose that IRF5 plays a crucial role in the development of systemic autoimmune disease in MRL/lpr mice. IRF5 is required for the production of proinflammatory cytokines induced by TLR ligands (16). Although it was originally unexpectedly found that type I IFN production was not decreased in IRF5⫺/⫺ mouse PDCs (16), a subsequent study showed that IRF5 was necessary for type I IFN production in response to certain stimuli (17,36). In this study, we demonstrated that TLR-7– and TLR-9–induced proinflammatory cytokine production and TLR-9–induced IFN␣ production by mouse splenic DCs are regulated by IRF5. However, our in vitro data show minor discrepancies with previous reports. For example, TLR-3– and TLR-4–induced cytokine production was not affected in DCs lacking IRF5 in our study, but has been shown to be suppressed in others (16,35). This discrepancy may be due to differences in cell populations examined, i.e., splenic DCs versus bone marrow–derived DCs versus macrophages; the genetic background of the mice used; or the sources of the stimuli (17). Our results suggest that loss of IRF5 downregulates DC activation signals delivered through TLR-7 and TLR-9, leading to attenuated disease in IRF5⫺/⫺ MRL/lpr mice. Previous studies have shown that TLR-7 and TLR-9 play significant roles in the development of lupus in MRL/lpr mice (37–40). TLR-7–deficient MRL/lpr mice showed milder glomerulonephritis and decreased levels of anti-Sm and anti-RNP antibodies (39), whereas TLR-9–deficient MRL/lpr mice had exacerbated glomerulonephritis and heightened lymphocyte activation (37–39). Interestingly, however, TLR-7 and TLR-9 double-knockout MRL/lpr mice showed a marked reduction in autoantibodies and glomerulonephritis (40), and 3d mutant B6-Faslpr mice, which have defective TLR-3, TLR-7, and TLR-9 signaling, also showed reduced levels of autoantibodies and improved survival (41). These data suggest that blockade of TLR-7 or TLR-9 alone has a limited effect on autoantibody production and opposing effects in glomerulonephritis, but the simultaneous blocking of both TLR-7 and TLR-9 abolishes most of the clinical features of murine lupus. We found that disease in the IRF5-deficient MRL/lpr mice was similar to that observed in MRL/lpr mice lacking both TLR-7 and TLR-9 (40), although the glomerulonephritis in the IRF5-deficient MRL/lpr mice 746 was milder and their survival longer. Disease in IRF5deficient MRL/lpr mice was also similar to that in 3d mutant B6-Faslpr mice (41). These studies suggest that IRF5 plays a key role in the development of lupus by regulating signals through TLR-7 and TLR-9, an essential combination for full-blown lupus. It is possible that another pathway mediated by IRF5 but not triggered by TLR-7 or TLR-9 engagement contributes to the attenuated disease observed in IRF5⫺/⫺MRL/lpr mice. As described previously, signals through TLR-3 or TLR-4 are candidate routes. Further study will clarify the precise function of IRF5 in the activation of various immune system cells in MRL/lpr mice. It has been reported that ICs are important stimuli that promote the activation of the TLR-7 and TLR-9 pathways in murine lupus (30–33,35). DNAcontaining ICs activate DCs through TLR-9 and induce the production of TNF␣, IL-12, and IFN␣ (30,31), and RNA-containing ICs activate DCs through TLR-7 and result in IL-6 and IFN␣ production (33,35). Our findings confirmed that TNF␣ and IL-6 production by DCs was enhanced in response to ICs purified from MRL/lpr mice, and that this effect was diminished in DCs lacking IRF5. These data suggest that IRF5 promotes lupus by inducing these inflammatory cytokines, presumably in response to ICs composed of autoantibodies and nucleic acids. We also observed increased expression of TNF␣, IL-6, and CCL5 in the kidney. It has been shown that IL-6 plays important roles in lupus (42) and that TNF␣ promotes focal inflammation in murine lupus (43), although genetic loss of Tnfr1 accelerates the progression of disease (44). Previous studies of human SLE and mice with lupus showed conflicting results with regard to the contribution of type I IFN to disease progression. In human SLE, type I IFN is considered to be pathogenic and its expression is linked to a susceptibility haplotype of IRF5 (5,6,34), whereas it has been reported that type I IFN protects against murine lupus (45). Our data showing decreased IFN␣ production in response to CpG from IRF5-deficient DCs suggest that loss of IRF5 promotes lupus. However, the effects of IRF5 in the induction of type I IFNs vary based on cell type and/or stimulus (16,17,36), and we did not detect IFN␣ production in response to ICs. Although further study is necessary to explore the type I IFN profile in these mice, we speculate that IRF5 is a more critical regulator of proinflammatory cytokines than type I IFNs, and/or that the beneficial effects of reduced inflammatory cytokine levels overcome the effect of decreased type I IFN production. Irregardless, the pathogenic role of IRF5 in MRL/ TADA ET AL lpr mice appears to be independent of type I IFN and may not be equal to that in human SLE. Recently, phenotypes of the lupus models Fc␥RIIB⫺/⫺ Yaa and Fc␥RIIB⫺/⫺ mice (46), and mice with pristane-induced lupus (47), all of which are IRF5deficient, have been described. These mice showed diminished autoantibody production and attenuated glomerulonephritis. Many of the results of these studies are consistent with our data. However, lymphosplenomegaly and serum IgG levels were markedly reduced in the IRF5-deficient Fc␥RIIB⫺/⫺ Yaa mice (46), whereas IRF5⫺/⫺MRL/lpr mice showed only a modest decrease in splenomegaly and LN cellularity, and only serum IgG3 was decreased. Despite these differences, these models are consistent in their delineation of a critical role for IRF5 in murine lupus. In conclusion, we confirmed in vivo that IRF5 deficiency prevents the development of autoantibody production and glomerulonephritis in MRL/lpr mice, and thereby extends mouse survival. We also confirmed the findings of previous genetic studies showing that IRF5 is critically involved in the pathogenesis of SLE. Our results have important therapeutic implications; namely, that treatment with inhibitors targeting IRF5 may be beneficial for SLE patients. ACKNOWLEDGMENTS We thank Motoko Fujisaki for technical assistance and animal care, Yumiko Tsugitomi for technical help with the histology, immunofluorescence, and immunostaining studies, and Mary Saunders for scientific editing of this manuscript. 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. Tada 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. Tada, Kondo, Koarada, Inoue, Suematsu, Ohta, Nagasawa. Acquisition of data. Tada, Kondo, Aoki, Koarada. Analysis and interpretation of data. Tada, Kondo, Aoki, Mak. REFERENCES 1. Kotzin BL. Systemic lupus erythematosus. Cell 1996;85:303–6. 2. Cameron JS. Lupus nephritis. J Am Soc Nephrol 1999;10:413–24. 3. Hooks JJ, Moutsopoulos HM, Geis SA, Stahl NI, Decker JL, Notkins AL. Immune interferon in the circulation of patients with autoimmune disease. N Engl J Med 1979;301:5–8. 4. Ytterberg SR, Schnitzer TJ. Serum interferon levels in patients with systemic lupus erythematosus. Arthritis Rheum 1982;25: 401–6. 5. Baechler EC, Batliwalla FM, Karypis G, Gaffney PM, Ortmann AMELIORATION OF LUPUS IN IRF5-DEFICIENT MRL/lpr MICE 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. WA, Espe KJ, et al. Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus. Proc Natl Acad Sci U S A 2003;100:2610–5. Kirou KA, Lee C, George S, Louca K, Papagiannis IG, Peterson MG, et al. Coordinate over-expression of interferon-␣–induced genes in systemic lupus erythematosus. Arthritis Rheum 2004;50: 3958–67. Sigurdsson S, Nordmark G, Goring HH, Lindroos K, Wiman AC, Sturfelt G, et al. Polymorphisms in the tyrosine kinase 2 and interferon regulatory factor 5 genes are associated with systemic lupus erythematosus. Am J Hum Genet 2005;76:528–37. Graham RR, Kozyrev SV, Baechler EC, Reddy MV, Plenge RM, Bauer JW, et al. A common haplotype of interferon regulatory factor 5 (IRF5) regulates splicing and expression and is associated with increased risk of systemic lupus erythematosus. Nat Genet 2006;38:550–5. Graham RR, Kyogoku C, Sigurdsson S, Vlasova IA, Davies LR, Baechler EC, et al. Three functional variants of IFN regulatory factor 5 (IRF5) define risk and protective haplotypes for human lupus. Proc Natl Acad Sci U S A 2007;104:6758–63. Shin HD, Sung YK, Choi CB, Lee SO, Lee HW, Bae SC. Replication of the genetic effects of IFN regulatory factor 5 (IRF5) on systemic lupus erythematosus in a Korean population. Arthritis Res Ther 2007;9:R32. Reddy MV, Velazquez-Cruz R, Baca V, Lima G, Granados J, Orozco L, et al. Genetic association of IRF5 with SLE in Mexicans: higher frequency of the risk haplotype and its homozygozity than Europeans. Hum Genet 2007;121:721–7. Kawasaki A, Kyogoku C, Ohashi J, Miyashita R, Hikami K, Kusaoi M, et al. Association of IRF5 polymorphisms with systemic lupus erythematosus in a Japanese population. Arthritis Rheum 2008; 58:826–34. Honda K, Taniguchi T. IRFs: master regulators of signaling by Toll-like receptors and cytosolic pattern-recognition receptors. Nat Rev Immunol 2006;6:644–58. Barnes BJ, Kellum MJ, Field AE, Pitha PM. Multiple regulatory domains of IRF-5 control activation, cellular localization, and induction of chemokines that mediate recruitment of T lymphocytes. Mol Cell Biol 202;22:5721–40. Schoenemeyer A, Barnes BJ, Mancl ME, Latz E, Goutagny N, Pitha PM, et al. The interferon regulatory factor, IRF5, is a central mediator of Toll-like receptor 7 signaling. J Biol Chem 2005;280: 17005–12. Takaoka A, Yanai H, Kondo S, Duncan G, Negishi H, Mizutani T, et al. Integral role of IRF-5 in the gene induction programme activated by Toll-like receptors. Nature 2005;434:243–9. Paun A, Reinert JT, Jiang Z, Medin C, Balkhi MY, Fitzgerald KA, et al. Functional characterization of murine interferon regulatory factor 5 (IRF-5) and its role in the innate antiviral response. J Biol Chem 2008;283:14295–308. Kyogoku C, Tsuchiya N. A compass that points to lupus: genetic studies on type I interferon pathway. Genes Immun 2007;8:445–5. Cohen PL, Eisenberg RA. Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu Rev Immunol 1991;9:243–69. Theofilopoulos AN, Dixon FJ. Murine models of systemic lupus erythematosus. Adv Immunol 1985;37:269–390. Andrews BS, Eisenberg RA, Theofilopoulos AN, Izui S, Wilson CB, McConahey PJ, et al. Spontaneous murine lupus-like syndromes: clinical and immunopathological manifestations in several strains. J Exp Med 1978;148:1198–215. Watanabe-Fukunaga R, Brannan CI, Copeland NG, Jenkins NA, Nagata S. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 1992;356: 314–7. Watson ML, Rao JK, Gilkeson GS, Ruiz P, Eicher EM, Pisetsky DS, et al. Genetic analysis of MRL-lpr mice: relationship of the 747 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. Fas apoptosis gene to disease manifestation and renal diseasemodifying loci. J Exp Med 1992;176:1645–56. Tada Y, Koarada S, Tomiyoshi Y, Morito F, Mitamura M, Haruta Y, et al. Role of inducible costimulator in the development of lupus in MRL/lpr mice. Clin Immunol 2006;120:179–88. Kinoshita K, Yamagata T, Nozaki Y, Sugiyama M, Ikoma S, Funauchi M, et al. Blockade of IL-18 receptor signaling delays the onset of autoimmune disease in MRL-Faslpr mice. J Immunol 2004;173:5312–8. Tada Y, Nagasawa K, Ho A, Morito F, Koarada S, Ushiyama O, et al. Role of the costimulatory molecule CD28 in the development of lupus in MRL/lpr mice. J Immunol 1999;163:3153–9. Schwarting A, Wada T, Kinoshita K, Tesch G, Kelly VR. IFN-␥ receptor signaling is essential for the initiation, acceleration, and destruction of autoimmune kidney disease in MRL-Faslpr mice. J Immunol 1998;161:494–503. Yokoyama H, Kreft B, Kelley VR. Biphasic increase in circulating and renal TNF-␣ in MRL-lpr mice with differing regulatory mechanisms. Kidney Int 1995;47:122–30. Moore KJ, Wada T, Barbee SD, Kelley VR. Gene transfer of RANTES elicits autoimmune renal injury in MRL-Faslpr mice. Kidney Int 1998;53:1631–41. Boule MW, Broughton C, Mackay F, Akira S, Marshak-Rothstein A, Rifkin IR. Toll-like receptor 9–dependent and –independent dendritic cell activation by chromatin–immunoglobulin G complexes. J Exp Med 2004;199:1631–40. Means TK, Latz E, Hayashi F, Murali MR, Golenbock DT, Luster AD. Human lupus autoantibody-DNA complexes activate DCs through cooperation of CD32 and TLR9. J Clin Invest 2005;115: 407–17. Savarese E, Chae O, Trowitzsch S, Weber G, Kastner B, Akira S, et al. U1 small nuclear ribonucleroprotein immune complexes induce type I interferon in plasmacytoid dendritic cells through TLR7. Blood 2006;107:3229–34. Ledbetter EA, Rifkin IR, Hohlbaum AM, Beaudette BC, Schlomchik MJ, Marshak-Rothstein A. Chromatin-IgG complexes activate B cells by dual engagement of IgM and Toll-like receptors. Nature 2002;416:603–7. Niewold TB, Kelly JA, Flesch MH, Espinoza LR, Harley JB, Crow MK. Association of the IRF5 risk haplotype with high serum interferon-␣ activity in systemic lupus erythematosus patients. Arthritis Rheum 2008;58:2481–7. Yasuda K, Richez C, Maciaszek JW, Agrawal N, Akira S, Marshak-Rothstein A, et al. Murine dendritic cell type I IFN production induced by human IgG-RNA immune complexes is IFN regulatory factor (IRF)5 and IRF7 dependent and is required for IL-6 production. J Immunol 2007;178:6876–85. Yanai H, Chen H, Inuzuka T, Kondo S, Mak TW, Takaoka A, et al. Role of IFN regulatory factor 5 transcription factor in antiviral immunity and tumor suppression. Proc Natl Acad Sci U S A 2007;104:3402–7. Christensen SR, Kashgarian M, Alexopoulou L, Flavell RA, Akira S, Schlomchik MJ. Toll-like receptor 9 controls anti-DNA autoantibody production in murine lupus. J Exp Med 2005;202:321–31. Wu X, Peng SL. Toll-like receptor 9 signaling protects against murine lupus. Arthritis Rheum 2006;54:336–42. Christensen SR, Shupe J, Nickerson K, Kashgarian M, Flavell RA, Schlomchik MJ. Toll-like receptor 7 and TLR9 dictate autoantibody specificity and have opposing inflammatory and regulatory roles in a murine model of lupus. Immunity 2006;25:417–28. Nickerson KM, Christensen SR, Shupe J, Kashgarian M, Kim D, Elkon K, et al. TLR9 regulates TLR7- and MyD88-dependent autoantibody production and disease in a murine model of lupus. J Immunol 2010;184:1840–8. Kono DH, Haraldsson MK, Lawson BR, Pollard KM, Koh YT, Du X, et al. Endosomal TLR signaling is required for anti-nucleic 748 acid and rheumatoid factor autoantibodies in lupus. Proc Natl Acad Sci USA 2009;106:12061–66. 42. Tackey E, Lipsky PE, Illei GG. Rationale for interleukin-6 blockade in systemic lupus erythematosus. Lupus 2004;13:339–43. 43. McHale JF, Harari OA, Marshall D, Haskard DO. TNF-␣ and IL-1 sequentially induce endothelial ICAM-1 and VCAM-1 expression in MRL/lpr lupus-prone mice. J Immunol 1999;163:3993–4000. 44. Zhou T, Edwards CK III, Yang P, Wang Z, Bluethmann H, Mountz JD. Greatly accelerated lymphadenopathy and autoimmune disease in lpr mice lacking tumor necrosis factor receptor I. J Immunol 1996;156:2661–5. TADA ET AL 45. Hron JD, Peng SL. Type I IFN protects against murine lupus. J Immunol 2004;173:2134–42. 46. Richez C, Yasuda K, Bonegio RG, Watkins AA, Aprahamian T, Busto P, et al. IFN regulatory factor 5 is required for disease development in the Fc␥RIIB⫺/⫺Yaa and Fc␥RIIB⫺/⫺ mouse models of systemic lupus erythematosus. J Immunol 2010;184: 796–806. 47. Savitsky DA, Yanai H, Tamura T, Taniguchi T, Honda K. Contribution of IRF5 in B cells to the development of murine SLE-like disease through its transcriptional control of the IgG2a locus. Proc Natl Acad Sci U S A 2010;107:10154–9.