Regulation of the interferon-╨Ю┬▒ production induced by RNA-containing immune complexes in plasmacytoid dendritic cells.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 60, No. 8, August 2009, pp 2418–2427 DOI 10.1002/art.24686 © 2009, American College of Rheumatology Regulation of the Interferon-␣ Production Induced by RNA-Containing Immune Complexes in Plasmacytoid Dendritic Cells Maija-Leena Eloranta,1 Tanja Lövgren,1 Doreen Finke,1 Linda Mathsson,1 Johan Rönnelid,1 Berthold Kastner,2 Gunnar V. Alm,3 and Lars Rönnblom1 Objective. Interferon-␣ (IFN␣) is produced in several autoimmune diseases, including systemic lupus erythematosus (SLE), and may be important in their pathogenesis. We undertook this study to investigate how IFN␣ production induced by RNA-containing immune complexes (ICs) in plasmacytoid dendritic cells (PDCs) is regulated. Methods. Normal PDCs purified from peripheral blood mononuclear cells (PBMCs) were cocultivated with other cell populations isolated from healthy individuals or SLE patients. IFN␣ production was induced by RNA-containing ICs, which consisted of anti-RNP autoantibodies and U1 small nuclear RNP particles, and the effects of prostaglandin E2 (PGE2), reactive oxygen species (ROS), or the cytokines IFN␣2b, granulocyte–macrophage colony-stimulating factor (GM-CSF), interleukin-10 (IL-10), or tumor necrosis factor ␣ (TNF␣) were explored. Results. Monocytes inhibited IFN␣ production by PDCs in PBMC cultures, while natural killer (NK) cells were stimulatory. The monocytes had little effect on IFN␣ production by pure PDCs but inhibited its stimulation by NK cells. Monocytes from SLE patients were less inhibitory. Exposure of PBMCs or PDCs to IFN␣2b/GM-CSF increased their IFN␣ production. RNA-containing ICs caused production of ROS, PGE2, and TNF␣, especially in monocytes. These mediators and IL-10 suppressed IFN␣ production in PBMC cultures, with ROS and PGE2 also inhibiting IFN␣ production by purified PDCs. Inhibition by all of these agents, except for ROS, was abolished by IFN␣2b/GMCSF. The inhibitory effect of monocytes was significantly counteracted by the ROS scavengers serotonin and catalase. Conclusion. IFN␣ production induced by RNAcontaining ICs in PDCs is regulated by a network of interactions between monocytes, NK cells, and PDCs, involving several pro- and antiinflammatory molecules. This should be considered when designing and applying new therapies. Supported in part by grants from the Dana Foundation, the Swedish Research Council, the Swedish Society of Medicine, the Swedish Rheumatism Foundation, the Uppsala University Hospital Development Foundation, the Agnes and Mac Rudberg Foundation, the Swedish Fund for Research Without Animal Experiments, the Nilsson Foundation, King Gustaf V’s 80-Year Foundation, and COMBINE. 1 Maija-Leena Eloranta, PhD, Tanja Lövgren, PhD, Doreen Finke, PhD, Linda Mathsson, PhD, Johan Rönnelid, MD, PhD, Lars Rönnblom, MD, PhD: Uppsala University, Uppsala, Sweden; 2 Berthold Kastner, PhD: Max Planck Institute of Biophysical Chemistry, Göttingen, Germany; 3Gunnar V. Alm, MD, PhD: Swedish University of Agricultural Sciences, Uppsala, Sweden. Address correspondence and reprint requests to Maija-Leena Eloranta, PhD, Department of Medical Sciences, Clinical Research Department 3, Systemic Autoimmunity Group, Entrance 85, 3rd Floor, Uppsala University Hospital, S-75185 Uppsala, Sweden. E-mail: [email protected] Submitted for publication November 20, 2008; accepted in revised form April 18, 2009. Systemic lupus erythematosus (SLE) is characterized by the occurrence of autoantibodies against nucleic acids and associated proteins, formation of immune complexes (ICs), and inflammation in many organs and tissues. Many SLE patients have a continuous production of type I interferon (IFN), which is detected as increased levels of IFN␣ and expression of type I IFN–inducible genes (1–3). Type I IFNs have many important immunoregulatory actions that could initiate an autoimmune process (1,4), which is further promoted by IFN␣-induced increase in autoantigen expression (5,6). A pivotal role of type I IFN in the etiology and pathogenesis of SLE has therefore been suggested and is supported by many further observations (1,2). Thus, IFN␣ levels and type I IFN–inducible gene expression 2418 REGULATION OF TYPE I IFN PRODUCTION correlate with disease activity and severity as well as with the levels of autoantibodies against double-stranded DNA (dsDNA) and RNA binding proteins (7,8). In addition, single-nucleotide polymorphisms in several genes encoding key molecules involved in IFN production and action are strongly associated with SLE (9–11). Furthermore, IFN␣ therapy in patients with malignancies or infections frequently results in autoimmune adverse effects, including SLE-like disease (12,13). Finally, administration of antibodies neutralizing IFN␣ in a phase I trial displayed promising efficacy in SLE patients, indicating that IFN␣ is indeed involved in pathogenesis and is a promising therapeutic target (14). In SLE and several other systemic autoimmune diseases, DNA– and RNA–protein complexes released by apoptotic or necrotic cells form ICs with autoantibodies, and these ICs may serve as endogenous IFN inducers (1,15–18). They activate type I IFN production in plasmacytoid dendritic cells (PDCs) after endocytosis mediated by Fc␥ receptor IIa (Fc␥RIIa) and activation of endosomal Toll-like receptors (TLRs), TLR-7 by single-stranded RNA and TLR-9 by unmethylated CpG DNA (19–23). Because of the important role of type I IFN in SLE, it is essential to clarify the mechanisms that regulate IFN␣ production induced by endogenous IFN inducers such as RNA-containing ICs. We observed previously that IFN␣ production triggered in normal peripheral blood mononuclear cells (PBMCs) in vitro by RNA-containing ICs was enhanced in the presence of several different cytokines, such as IFN ␣ / ␤ and granulocyte–macrophage colony-stimulating factor (GM-CSF) (15). This phenomenon is termed priming and is due to the induction of the transcription factors IFN regulatory factor 5 (IRF-5) and IRF-7, which are involved in the triggering of type I IFN genes (23). In contrast, the cytokines interleukin-10 (IL-10) and tumor necrosis factor ␣ (TNF␣) decreased this IC-induced IFN␣ production (15). In the present study, we used well-defined RNA-containing ICs consisting of purified U1 small nuclear RNP (snRNP) and anti-RNP autoantibodies (21,22), and we observed initially that they were poor inducers of IFN␣ production by the PDCs present in PBMC cultures but strong inducers in highly purified PDCs. These observations suggested a pronounced regulation of IFN␣ production by PDCs, which could be important in the pathogenesis of systemic autoimmune diseases. In the present study, we therefore investigated in more detail the regulatory effects of cells and their mediators, including cytokines, on IFN␣ production 2419 induced by RNA-containing ICs. We found that monocytes from healthy individuals were potent inhibitors and that natural killer (NK) cells were potent enhancers of IFN␣ production in PBMC cultures. Because PDCs are the actual IFN␣ producers, we investigated the extent to which PDCs and NK cells were the target cells for the suppressive monocytes and their potential mediators, specifically reactive oxygen species (ROS), the prostanoid prostaglandin E2 (PGE2), and the cytokines TNF␣ and IL-10. We also studied the stimulatory (“priming”) effects of cytokines IFN␣2b and GM-CSF on RNAcontaining IC–induced IFN␣ production in PDCs, as well as how such priming especially affected monocytemediated inhibition. PATIENTS AND METHODS Patients and controls. Six SLE patients attending the outpatient clinic at the Rheumatology Unit of Uppsala University Hospital were included in the study. All patients fulfilled at least 4 (mean 5; range 4–6) of the American College of Rheumatology criteria for SLE (24). The mean ⫾ SD age of the patients was 40 ⫾ 11 years, and their mean ⫾ SD disease duration was 18 ⫾ 12 years. All patients had low disease activity (score of ⬍4 on the modified SLE Disease Activity Index ). Healthy blood donors served as controls. The study was approved by the local ethics committee, and informed consent was obtained from all patients and controls. IFN inducers. U1 snRNP particles were purified from HeLa cells as described previously (26,27). The U1 snRNP preparations had a purity of at least 90%. Sera from a healthy blood donor or an SLE patient containing autoantibodies to SmB, SmD, U1 RNP A, and U1 RNP C proteins, ribosomal P antigen, histone, and dsDNA were used to prepare normal IgG and SLE IgG, respectively, by protein G chromatography. The U1 snRNP particles were used at a final concentration of 2.5 g/ml in the cell cultures, together with a final concentration of 1 mg/ml of normal IgG or SLE IgG. The normal IgG was used either untreated or aggregated by heating at 63°C for 60 minutes at a concentration of 50 mg/ml. Oligodeoxynucleotide (ODN) 2216 (3 g/ml; CyberGene, Huddinge, Sweden) was used as TLR-9 agonist. Herpes simplex virus type 1 (HSV-1), propagated in WISH cells and inactivated by ultraviolet light (17), was used as a control IFN␣ inducer. Cytokines, PGE2, and ROS. When indicated, the cultures were supplemented with IFN␣2b (500 units/ml, IntronA; Schering-Plough, Bloomfield, NJ), GM-CSF (2 ng/ml, Leukine; Berlex, Montville, NJ), IL-10 (10 ng/ml; Genzyme, Cambridge, MA), TNF␣ (100 units/ml; Genzyme), or PGE2 (100 ng/ml; Sigma-Aldrich, St. Louis, MO). H2O2 (Merck, Darmstadt, Germany) was used at indicated concentrations. ROS inhibitors. Serotonin hydrochloride, bovine liver catalase, and human erythrocyte superoxide dismutase (SOD) (all from Sigma-Aldrich) were used at final optimal concentrations of 200 M, 100 units/ml, and 100 units/ml, respectively, in the cultures. 2420 Cell preparation and culture conditions. PBMCs were prepared by Ficoll-Hypaque (GE Healthcare, Uppsala, Sweden) density-gradient centrifugation of buffy coats from healthy blood donors (Department of Transfusion Medicine, Uppsala University Hospital) or of whole blood from SLE patients and corresponding healthy controls. Pure cell populations were prepared from PBMCs by using the MACS magnetic labeling system (Miltenyi Biotec, Bergisch Gladbach, Germany). Briefly, PDCs were purified (ⱖ95% purity) by negative depletion using a PDC Isolation kit and MACS LS columns (both from Miltenyi Biotec). CD56⫹ NK cells were isolated using an NK Cell Isolation kit (Miltenyi Biotec) (ⱖ96% purity), and CD14⫹ monocytes were prepared using CD14 Microbeads (Miltenyi Biotec) (ⱖ97% purity). The purity of the isolated cell populations was determined using a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA) after staining with fluorescein isothiocyanate–labeled anti– blood dendritic cell antigen 2 (anti–BDCA-2; Miltenyi Biotec), anti-CD56 (Miltenyi Biotec), or anti-CD14 (Serotec, Oxford, UK) monoclonal antibodies (mAb). Cells were cultured in 0.1-ml volumes in 96-well flatbottomed plates (Nunclon; Nunc, Roskilde, Denmark) using Macrophage-SFM medium (Invitrogen, San Diego, CA) supplemented with HEPES (20 mM), penicillin (60 g/ml), and streptomycin (100 g/ml). Cell concentrations were 2 ⫻ 106/ml for PBMCs and 0.25 ⫻ 106/ml or 0.5 ⫻ 106/ml for isolated PDCs, while monocytes and NK cells were cultured at 0.5–2 ⫻ 106/ml as indicated. All cell cultures were set up as duplicates in each experiment. Immunoassays and ROS measurement. The IFN␣ concentrations in 20-hour culture supernatants were determined by a dissociation-enhanced lanthanide fluoroimmunoassay (DELFIA) as previously described (28). The anti-IFN␣ mAb LT27:293, which recognizes most IFN␣ subtypes except IFN␣2b, was used as capture antibody, and the europiumlabeled anti-IFN␣ mAb LT27:297 was used as detection antibody. The IFN␣ standard was calibrated against the National Institutes of Health reference leukocyte IFN␣ GA-23-02-530. This assay has a detection limit of 2 units/ml. No IFN␣ production was detected in cell cultures without RNA-containing ICs or HSV-1. Concentrations of TNF␣ and IL-10 were determined by enzyme-linked immunosorbent assay as previously described (29). Briefly, mAb to TNF␣ (MAB610; R&D Systems, Minneapolis, MN) or IL-10 (9D7; PharMingen, San Diego, CA) were used as capture antibodies, and biotin-conjugated polyclonal antibodies to TNF␣ (BAF210; R&D Systems) or biotin-conjugated mAb to IL-10 (12G8; PharMingen) were used as secondary antibodies. Horseradish peroxidase– conjugated streptavidin (R&D Systems) was used for detection. The lower detection limit is 2 pg/ml for IL-10 and 1 pg/ml for TNF␣. The levels of PGE 2 were determined using a Prostaglandin EIA kit (Sigma-Aldrich) according to the manufacturer’s instructions. The lower detection limit for this assay is 39 pg/ml. ROS were assayed by a luminol-enhanced chemiluminescence system. The chemiluminescence activity was measured in a 6-channel Biolumat LB 9505 (Berthold, Wildbad, Germany) using 105 PBMCs, CD14-depleted PBMCs, PDCs, or monocytes in 0.1 ml of culture medium. Chemiluminescence was measured continuously for 35 minutes at 37°C, beginning ELORANTA ET AL immediately after addition of luminol (5 mg/ml) and RNAcontaining ICs. Statistical analysis. The significance of differences between groups was calculated by t-test using GraphPad Prism 4.0 software (GraphPad Software, San Diego, CA). P values less than or equal to 0.05 were considered significant. RESULTS IFN␣ production by PDCs in PBMC cultures stimulated with RNA-containing ICs is deficient but is enhanced by IFN␣2b/GM-CSF. We initially evaluated the effect of IFN␣2b/GM-CSF priming on IFN␣ production by PBMCs or purified PDCs from healthy individuals after stimulation with RNA-containing ICs, consisting of U1 snRNP particles and anti-RNP– containing IgG. For comparison, we used HSV-1, which, like RNA-containing ICs, selectively induces IFN␣ production in PDCs (22). In addition to measuring levels of IFN␣ (in units/ml) in the cultures, we calculated the mean quantity of IFN␣ produced per PDC (units/PDC) by dividing the total amount of IFN␣ in the cultures (determined by the DELFIA) by the number of BDCA2–positive PDCs (determined by flow cytometry). Total PBMCs stimulated with RNA-containing ICs (Figure 1A) produced only very low amounts of IFN␣ (corresponding to a mean of 0.01 units/PDC), and these levels were increased by IFN␣2b/GM-CSF priming (to a level corresponding to a mean of 0.05 units/PDC) (P ⫽ 0.006) (Table 1). In contrast, IFN␣ production induced by HSV-1 in PBMCs was considerably higher and was not increased by priming (Figure 1A and Table 1). Compared with PBMCs, purified PDCs stimulated with RNA-containing ICs or HSV-1 produced much higher levels of IFN␣ (Figure 1A), corresponding to 0.09 units/PDC (Table 1). Priming the purified PDCs with IFN␣2b/GM-CSF significantly increased IFN␣ levels induced by RNA-containing ICs (P ⫽ 0.006) and HSV-1 (P ⫽ 0.019) (Figure 1A), resulting in 0.28 units/PDC and 0.21 units/PDC, respectively (Table 1). Thus, RNA-containing ICs and HSV-1 are both potent IFN␣ inducers in purified PDCs, but RNAcontaining ICs are very poor inducers of IFN␣ production by PDCs in PBMC cultures. With both IFN inducers, IFN␣ production increased with IFN␣2b/GM-CSF priming, strikingly so in the case of RNA-containing ICs. Monocytes inhibit IFN␣ production induced by RNA-containing ICs. The reason for the low IFN␣ production by PDCs in PBMCs from healthy individuals stimulated with RNA-containing ICs could be partly related to a lack of costimulation, since the impaired IFN␣ response was increased by priming with IFN␣2b/ GM-CSF. Another explanation could be a suppressive REGULATION OF TYPE I IFN PRODUCTION 2421 sponsible for the low IFN␣ production by PBMCs in response to RNA-containing ICs. Indeed, RNA-containing ICs induced higher levels of IFN␣ in CD14-depleted PBMCs compared with whole PBMC cultures, both with (P ⫽ 0.032) and without (P ⫽ 0.013) IFN␣2b/GM-CSF priming (Figure 1B). However, the HSV-1–induced IFN␣ production by unprimed CD14-depleted PBMCs and intact PBMCs did not differ (P ⫽ 0.09) and was only moderately increased in CD14-depleted PBMCs compared with intact PBMCs (P ⫽ 0.034) when primed with IFN␣2b/ GM-CSF (Figure 1C). Purified monocytes produced no IFN␣ (⬍10 units/ml) when stimulated with either IFN inducer. We also investigated whether CD14⫹ monocytes actually inhibited IFN␣ production by purified PDCs stimulated with RNA-containing ICs (Figure 2A). However, addition of even high numbers of purified CD14⫹ monocytes had little or no inhibitory effect (P ⫽ 0.75). In contrast, IFN␣ production by PDCs was markedly increased in cocultures with CD14-depleted PBMCs (P ⫽ 0.007; n ⫽ 6). This increase was prevented by the addition of CD14⫹ monocytes to such cultures (Figure 2B). Thus, monocytes had a profound negative impact on IFN␣ production induced by RNA-containing ICs in PBMC cultures. However, this was not a direct effect on PDCs, the actual IFN␣ producers. Instead, the monocytes appeared to inhibit the enhancing effect of other cells among the PBMCs on IFN␣ production by PDCs. IFN␣ production by PDCs stimulated with RNAcontaining ICs is enhanced by CD56ⴙ NK cells. Because NK cells can promote CpG-induced PDC activaFigure 1. Decreased interferon-␣ (IFN␣) production by plasmacytoid dendritic cells (PDCs) in cultures of peripheral blood mononuclear cells (PBMCs) stimulated with RNA-containing immune complexes (RNAIC), consisting of U1 small nuclear RNP particles and anti-RNP– containing IgG, is associated with the presence of monocytes and is increased by priming with IFN␣2b/granulocyte–macrophage colonystimulating factor (GM-CSF). A, IFN␣ production in cultures of PBMCs or purified PDCs (0.5 ⫻ 105 per well) stimulated with RNA-containing ICs or herpes simplex virus type 1 (HSV-1), with or without priming with IFN␣2b/GM-CSF. B and C, IFN␣ production by PBMCs, monocyte-depleted PBMCs (CD14⫺), or purified monocytes (CD14⫹). The cells (0.2 ⫻ 106 per culture) were stimulated with RNA-containing ICs (B) or HSV-1 (C), with or without IFN␣2b/GMCSF priming. The levels of IFN␣ in cell culture medium were determined by immunoassay after 20 hours. Horizontal lines show the median of 5 experiments. NS ⫽ not significant. action of other PBMC types on the PDCs. We therefore investigated whether the CD14⫹ monocytes were re- Table 1. Quantity of IFN␣ produced per PDC in cultures of PBMCs or purified PDCs stimulated with RNA-containing ICs or HSV-1* IFN␣ produced per PDC, units Inducer, cell population RNA-containing ICs PBMCs PDCs HSV-1 PBMCs PDCs No priming IFN␣2b/GM-CSF priming P† 0.01 ⫾ 0.006 0.09 ⫾ 0.08 0.05 ⫾ 0.02 0.28 ⫾ 0.14 0.006 0.006 0.16 ⫾ 0.060 0.09 ⫾ 0.04 0.19 ⫾ 0.16 0.21 ⫾ 0.11 NS 0.019 * Values are the mean ⫾ SD of 5 experiments. IFN␣ ⫽ interferon-␣; PDC ⫽ plasmacytoid dendritic cell; PBMCs ⫽ peripheral blood mononuclear cells; RNA-containing ICs ⫽ immune complexes consisting of U1 small nuclear RNP particles and anti-RNP–containing IgG; HSV-1 ⫽ herpes simplex virus type 1; GM-CSF ⫽ granulocyte– macrophage colony-stimulating factor; NS ⫽ not significant. † By paired t-test. P values less than or equal to 0.05 were considered significant. 2422 ELORANTA ET AL Figure 2. Effect of monocytes, monocyte-depleted PBMCs, or natural killer (NK) cells on IFN␣ production by PDCs stimulated with RNA-containing ICs (A–C) or oligodeoxynucleotide (ODN) 2216 (D). A, IFN␣ production by PDCs alone (0.25 ⫻ 105 per well), CD14-depleted PBMCs (CD14–) alone, or PDCs together with indicated numbers of either purified monocytes (CD14⫹) or CD14-depleted PBMCs. B, IFN␣ production by PDCs alone (0.25 ⫻ 105 per well) or by PDCs cocultivated either with CD14-depleted PBMCs or with CD56⫹ NK cells (CD56⫹) in the absence or presence of CD14⫹ monocytes. C, IFN␣ production by CD56⫹ NK cells alone or by PDCs alone (0.25 ⫻ 105 per well) or cocultivated with indicated numbers of either CD14⫺CD56⫺ PBMCs or CD56⫹ NK cells. D, ODN2216-triggered IFN␣ production by PDCs alone (0.25 ⫻ 105 per well), cocultivated with CD56⫹ NK cells, cocultivated with CD56⫹ NK cells and CD14⫹ monocytes in the absence or presence of heat-aggregated normal IgG. Values are the mean. Results are from 1 representative experiment of 3 performed. See Figure 1 for other definitions. tion, we investigated whether NK cells were responsible for the observed ability of CD14-depleted PBMCs to enhance RNA-containing IC–induced IFN␣ production by PDCs (30). Although the purified PDCs produced variable and often low levels of IFN␣ in response to RNA-containing ICs, their IFN␣ production was dramatically increased (P ⫽ 0.004) by cocultivation with purified NK cells (Figures 2B and C). Similar effects of NK cells were seen using HSV-1 as IFN␣ inducer (results not shown). In contrast, only small stimulatory effects were seen with NK-depleted CD14⫺ PBMCs (Figure 2C). The purified CD56⫹ NK cells alone produced no IFN␣ when stimulated with RNA-containing ICs (Figure 2C). We also investigated whether autologous CD14⫹ monocytes from healthy individuals inhibited the stimulatory effect of NK cells. We found that monocytes strongly reduced RNA-containing IC–induced IFN␣ production when added to cocultures of PDCs with CD56⫹ NK cells (Figure 2B). Consequently, NK cells can enhance RNA-containing IC–induced IFN␣ production by PDCs, but this effect is inhibited by monocytes. Next, we investigated whether the inhibitory effect of monocytes was due to TLR ligation or required Fc␥R activation (Figure 2D). Cultures of PDCs, CD56⫹ NK cells, and monocytes from healthy blood donors were stimulated with the TLR-9 agonist ODN2216, and heataggregated normal IgG was added to the cells. We found that IFN␣ production was down-regulated only in the Figure 3. Less suppression of IFN␣ production by monocytes from patients with systemic lupus erythematosus (SLE). PDCs and CD56⫹ natural killer cells were cocultivated with CD14⫹ monocytes isolated from SLE patients (n ⫽ 6) and healthy blood donors (n ⫽ 5), and IFN␣ levels in supernatants were determined by immunoassay after 20 hours of culture in all experiments. Horizontal lines show the median. See Figure 1 for other definitions. REGULATION OF TYPE I IFN PRODUCTION 2423 presence of heat-aggregated normal IgG. This indicated that a signal through Fc␥R was required for triggering the inhibitory effect of monocytes. We also compared the ability of monocytes from SLE patients and healthy controls to inhibit NK cell– enhanced IFN␣ production in PDCs induced by RNAcontaining ICs. Monocytes from SLE patients appeared less suppressive (Figure 3). Suppression of RNA-containing IC–induced IFN␣ production by cytokines, PGE2, and ROS. Cytokines or other inflammatory mediators could be responsible for the suppressive effect of monocytes on RNAcontaining IC–induced IFN␣ production. We therefore measured the levels of TNF␣, IL-10, PGE2, and ROS in Figure 4. Tumor necrosis factor ␣ (TNF␣), prostaglandin E2 (PGE2), and reactive oxygen species (ROS) production induced by RNAcontaining ICs. A, PGE2 or TNF␣ production by 0.2 ⫻ 106 PBMCs, CD14-depleted PBMCs (CD14⫺), or purified CD14⫹ monocytes (CD14⫹) stimulated with RNA-containing ICs or medium only (Mock). The levels of PGE2 or TNF␣ in 20-hour culture medium were determined by immunoassay. B, ROS production by 0.1 ⫻ 106 PBMCs, PDCs, CD14-depleted PBMCs, or purified CD14⫹ monocytes. ROS production was measured with a luminol-enhanced chemiluminescence method for 30 minutes beginning immediately after addition of RNA-containing ICs. Values are the mean. Results are from 1 representative experiment of 3 performed. See Figure 1 for other definitions. Figure 5. Inhibitory effects of prostaglandin E2 (PGE2), interleukin-10 (IL-10), tumor necrosis factor ␣ (TNF␣), and reactive oxygen species (H2O2) on IFN␣ production induced by RNA-containing ICs. A and B, IFN␣ production by PBMCs (A) or PDCs (B), with or without IFN␣2b/GM-CSF priming, stimulated with RNA-containing ICs in the presence of the indicated concentrations of PGE2, IL-10, or TNF␣. C, Effect of H2O2 on PDCs stimulated with RNA-containing ICs. IFN␣ levels in supernatants were determined by immunoassay after 20 hours of culture. Values are the mean. Results are from 1 of 3 experiments, each of which showed similar results. See Figure 1 for other definitions. the cell cultures and found that RNA-containing ICs induced production of PGE2 and TNF␣ (Figure 4A) as well as ROS (Figure 4B) by PBMCs and purified monocytes from healthy individuals. Monocyte-depleted PBMCs produced TNF␣ but little or no PGE2 and ROS (Figure 4), as did purified PDCs (results not shown for TNF␣ and PGE2). No ROS production was detected in PBMCs, PDCs, monocyte-depleted PBMCs, or monocytes without RNA-containing ICs (not shown). The 2424 ELORANTA ET AL Figure 6. Effects of reactive oxygen species scavengers on IFN␣ production induced by RNA-containing ICs. PBMCs and CD14-depleted PBMCs were stimulated with RNA-containing ICs in the presence of 200 M serotonin or 100 units/ml catalase (A) and 100 units/ml superoxide dismutase (SOD) (B), using medium only as control. The levels of IFN␣ after 20 hours of culture were determined by immunoassay. Values are the mean and SD. Results are from 1 of 3 experiments, each of which showed similar results. See Figure 1 for other definitions. levels of IL-10 were low in all cultures (⬍20 pg/ml). In general, similar levels of mediators were seen when cells were primed with IFN␣2b/GM-CSF (results not shown). The RNA-containing IC is therefore a potent activator of monocytes, inducing production of at least PGE2, TNF␣, and ROS. We further investigated whether PGE2 and the cytokines IL-10 and TNF␣ actually affected IFN␣ production by PDCs. In unprimed PBMC cultures, all these mediators were found to inhibit RNA-containing IC– induced IFN␣ production, but no inhibition was seen with IFN␣2b/GM-CSF–primed cells (Figure 5A). In cultures of purified PDCs, PGE2 caused a strong inhibition (88%) of IFN␣ production induced by RNAcontaining ICs, but not in the presence of IFN␣2b/ GM-CSF priming (Figure 5B). In contrast, TNF␣ and IL-10 had no significant inhibitory effects on the purified PDCs. With regard to the effect of ROS administered as H2O2, higher concentrations had an inhibitory effect on both purified PDCs (Figure 5C) and PBMCs (results not shown), regardless of IFN␣2b/GM-CSF priming. Because it is difficult to mimic the different ROS in vitro, we determined whether blocking of ROS production with the scavengers serotonin, catalase (Figure 6A), or SOD (Figure 6B) could restore IFN␣ production in PBMC cultures stimulated with RNA-containing ICs. All 3 agents were found to increase IFN␣ production by PBMCs, but none of them had any effect on CD14depleted PBMCs. This suggests that monocytes were responsible for ROS production and down-regulation of the type I IFN response. DISCUSSION In the present study, we examined the regulation of IFN␣ production induced in normal PDCs by RNAcontaining ICs consisting of U1 snRNP and SLE IgG, which are representative of the endogenous IFN inducers that are considered to be responsible for the activation of type I IFN production in several systemic autoimmune diseases (1,2). We initially observed exceptionally low IFN␣ production induced by RNAcontaining ICs in cultures of PBMCs from healthy individuals, while RNA-containing ICs and HSV-1 were both potent stimulators of purified PDCs. Costimulation (priming) with IFN␣2b/GM-CSF increased the amount of IFN␣ produced by both PBMCs and PDCs. However, in the case of PBMCs stimulated with RNA-containing ICs, the amount of IFN␣ produced per PDC in the PBMC cultures was still far below estimates of the IFN␣-producing capacity of PDCs in previous studies (23) and also for purified PDCs in the present study. The specific removal of the CD14⫹ monocytes from PBMCs restored the IFN␣ production stimulated with RNA-containing ICs to the same level as that stimulated with HSV-1. Furthermore, addition of monocytes to monocyte-depleted PBMCs inhibited their RNA-containing IC–induced IFN␣ production. Consequently, monocytes suppressed IFN␣ production of PBMCs in response to RNA-containing ICs. Unexpectedly, they had little effect on the purified PDCs. Because we observed that monocyte-depleted PBMCs actually strongly increased IFN␣ production by purified PDCs, we explored the possibility that the monocytes actually REGULATION OF TYPE I IFN PRODUCTION were suppressive of one or several types of PBMCs that enhanced IFN␣ production by PDCs. Indeed, purified NK (CD56⫹) cells from healthy individuals were found to be highly efficient enhancers of IFN␣ production by PDCs upon stimulation with either RNA-containing ICs or HSV-1. Conversely, elimination of NK cells essentially abrogated the enhancing effects of the monocyte-depleted PBMCs. This NK stimulation was especially pronounced at low PDC responses, with ⬃20-fold increases in the IFN␣ levels. Consequently, NK cell function is important in determining the PDC response to physiologic IFN inducers (i.e., RNA-containing ICs and HSV-1), and our results are consistent with findings that NK cells increase IFN␣ production by PDCs stimulated with synthetic unmethylated CpG-containing ODNs (30,31). Because PDCs and NK cells coexist in tissues, including secondary lymphoid organs (32), NK cells could clearly be a major physiologic regulator of IFN␣ production by PDCs, which is possibly also relevant in the activation of type I IFN production in SLE patients. However, the mechanism of the enhancing action of NK cells needs to be further investigated, including the molecular basis of PDC–NK cell interactions. The observation that CD14⫹ monocytes strongly inhibited the NK cell–mediated enhancement of IFN␣ production by PDCs induced by RNA-containing ICs, via interaction with Fc␥R, raised the question of whether RNA-containing ICs could activate the production of inflammatory mediators in monocytes, as described for other ICs (29,33,34). Such mediators include cytokines, prostanoids, and ROS, which could suppress IFN␣ production by direct inhibitory effects on PDCs or on stimulatory NK cells. In addition, ROS could reduce the proinflammatory properties of high mobility group box chromosomal protein 1 (35). Indeed, we demonstrated that RNA-containing ICs caused production of ROS, TNF␣, and PGE2, but not IL-10, in monocytes from healthy individuals. However, the latter cytokine has been reported to be produced by monocytes activated by other types of ICs (36). All 4 mediators we examined (PGE2, TNF␣, IL-10, ROS) were found to inhibit IFN␣ production by normal human PBMCs that had been stimulated with RNA-containing ICs, but only ROS (as H2O2) and PGE2 were inhibitory of purified PDCs. Interestingly, all agents except H2O2 lost all or most inhibitory activity when PBMCs or PDCs were exposed to IFN␣2b/GM-CSF. Using the scavenging inhibitors catalase, serotonin, and SOD, we confirmed that the inhibitory activity of monocytes was due to ROS to a significant extent. However, because monocytes had 2425 little impact on IFN␣ production by purified PDCs in cocultures, PDCs may also be relatively resistant to ROS and other monocyte-derived mediators. These new findings suggest that the primary target cells for monocytes and their mediators are not PDCs. Instead, NK cells are a particularly interesting candidate target, because PGE2 (37,38), IL-10 (39), and TNF␣ (40) have all been reported to impair NK cell functions. Furthermore, NK cells are very sensitive to ROS and are rendered apoptotic by ROS produced by monocyte/macrophages or delivered as exogenous H2O2 (41,42). In addition, we noticed that monocytes from SLE patients were less efficient inhibitors of IFN␣ production when added to cocultures with PDCs and NK cells from healthy individuals. Monocytes may therefore be part of an important feedback mechanism that limits the autoimmune reaction driven by IFN␣, and it is conceivable that this function is deficient in SLE. This conclusion is also supported by the observation that monocytes from SLE patients have a deficient production of ROS (43) and suggests that normalization of the monocyte/ROS system in SLE could ameliorate the type I IFN–driven disease process. The observation that IFN␣2b/GM-CSF increased the otherwise poor ability of normal PBMCs to produce IFN␣ upon stimulation with RNA-containing ICs is consistent with previous data showing such priming by IFN␣ on responses to several different IFN␣ inducers. These IFN␣ inducers include interferogenic ICs formed by apoptotic or necrotic cell material mixed with autoantibodies (16,17,44). The priming phenomenon is commonly thought to be due to direct effects of IFN␣ on PDCs, resulting in increased expression of proteins involved in type I IFN gene expression, including the IFN-inducible transcription factors IRF-5 and IRF-7 (23). Increased survival of PDCs is another possibility, because we noted that IFN␣2b/GM-CSF prevented PGE2-mediated inhibition of the IFN␣ production in purified PDCs, and PGE2 is known to cause apoptosis and inhibition of IFN␣ production in these cells (45). However, IFN␣2b/GM-CSF may also act on other cells in the PBMC population that inhibit or stimulate PDC functions (e.g., monocytes and NK cells). This possibility is supported by our findings that IFN␣2b/GM-CSF prevented the inhibitory effects of the cytokines IL-10 and TNF␣. Because SLE patients have signs of both type I IFN (1,2) and GM-CSF (46) activation, the enhancing effects of IFN␣2b/GM-CSF on IFN␣ production may be relevant in vivo by sustaining the disease process. We conclude that IFN␣ production induced by 2426 ELORANTA ET AL RNA-containing ICs in PDCs is suppressed by monocytes and that mediators such as ROS, PGE2, and the cytokines TNF␣ and IL-10 can be involved. However, the production of IFN␣ by PDCs is enhanced by NK cells and also by PDCs themselves via production of type I IFN. Consequently, there is a network of interactions between at least 3 types of cells that regulate IFN␣ production by PDCs, the details of which should be further elucidated. This is important, considering the crucial role of the type I IFN system in several systemic autoimmune diseases, and may both increase our understanding of their pathogenesis and reveal new therapeutic targets. ACKNOWLEDGMENTS 9. 10. 11. 12. 13. 14. We thank Ms Anne Trönnberg for excellent technical assistance and Ms Annika Hult for help with the ROS analysis. We also want to thank Dr. Keith Elkon for helpful discussions. 15. AUTHOR CONTRIBUTIONS 16. 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. Eloranta 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. Eloranta, Lövgren, Alm, Rönnblom. Acquisition of data. Eloranta, Lövgren, Finke, Mathsson, Rönnelid, Kastner. Analysis and interpretation of data. Eloranta, Lövgren, Finke, Rönnelid, Alm, Rönnblom. REFERENCES 1. Ronnblom L, Eloranta ML, Alm GV. The type I interferon system in systemic lupus erythematosus [review]. Arthritis Rheum 2006; 54:408–20. 2. Crow MK. Type I interferon in systemic lupus erythematosus. Curr Top Microbiol Immunol 2007;316:359–86. 3. Ronnblom L, Pascual V. The innate immune system in SLE: type I interferons and dendritic cells. Lupus 2008;17:394–9. 4. Theofilopoulos AN, Baccala R, Beutler B, Kono DH. Type I interferons (␣/␤) in immunity and autoimmunity. Annu Rev Immunol 2005;23:307–36. 5. Hueber W, Zeng D, Strober S, Utz PJ. Interferon-␣–inducible proteins are novel autoantigens in murine lupus. Arthritis Rheum 2004;50:3239–49. 6. Strandberg L, Ambrosi A, Espinosa A, Ottosson L, Eloranta ML, Zhou W, et al. Interferon-␣ induces up-regulation and nuclear translocation of the Ro52 autoantigen as detected by a panel of novel Ro52-specific monoclonal antibodies. J Clin Immunol 2008; 28:220–31. 7. Bengtsson A, Sturfelt G, Truedsson L, Blomberg J, Alm G, Vallin H, et al. Activation of type I interferon system in systemic lupus erythematosus correlates with disease activity but not antiretroviral antibodies. Lupus 2000;9:664–71. 8. Baechler EC, Batliwalla FM, Karypis G, Gaffney PM, Ortmann WA, Espe KJ, et al. Interferon-inducible gene expression signa- 17. 18. 19. 20. 21. 22. 23. 24. 25. ture in peripheral blood cells of patients with severe lupus. Proc Natl Acad Sci U S A 2003;100:2610–5. 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. Remmers EF, Plenge RM, Lee AT, Graham RR, Hom G, Behrens TW, et al. STAT4 and the risk of rheumatoid arthritis and systemic lupus erythematosus. N Engl J Med 2007;357:977–86. Sigurdsson S, Nordmark G, Garnier S, Grundberg E, Kwan T, Nilsson O, et al. A risk haplotype of STAT4 for systemic lupus erythematosus is over-expressed, correlates with anti-dsDNA and shows additive effects with two risk alleles of IRF5. Hum Mol Genet 2008;17:2868–76. Ronnblom LE, Alm GV, Oberg KE. Possible induction of systemic lupus erythematosus by interferon-␣ treatment in a patient with a malignant carcinoid tumour. J Intern Med 1990;227:207–10. Borg FA, Isenberg DA. Syndromes and complications of interferon therapy. Curr Opin Rheumatol 2007;19:61–6. Wallace DJ, Petri P, Olsen N, Kirou K, Dennis G, Yao Y, et al. MEDI-545, an anti-interferon alpha monoclonal antibody, shows evidence of clinical activity in systemic lupus erythematosus [abstract]. Arthritis Rheum 2007;56 Suppl 9:S526–7. Bave U, Vallin H, Alm GV, Ronnblom L. Activation of natural interferon-␣ producing cells by apoptotic U937 cells combined with lupus IgG and its regulation by cytokines. J Autoimmun 2001;17:71–80. Bave U, Nordmark G, Lovgren T, Ronnelid J, Cajander S, Eloranta ML, et al. Activation of the type I interferon system in primary Sjögren’s syndrome: a possible etiopathogenic mechanism. Arthritis Rheum 2005;52:1185–95. Lovgren T, Eloranta ML, Bave U, Alm GV, Ronnblom L. Induction of interferon-␣ production in plasmacytoid dendritic cells by immune complexes containing nucleic acid released by necrotic or late apoptotic cells and lupus IgG. Arthritis Rheum 2004;50:1861–72. Eloranta ML, Helmers SB, Ulfgren AK, Ronnblom L, Alm GV, Lundberg IE. A possible mechanism for endogenous activation of the type I interferon system in myositis patients with anti–Jo-1 or anti–Ro 52/anti–Ro 60 autoantibodies. Arthritis Rheum 2007;56: 3112–24. Bave U, Magnusson M, Eloranta ML, Perers A, Alm GV, Ronnblom L. Fc␥RIIa is expressed on natural IFN-␣ producing cells (plasmacytoid dendritic cells) and is required for the IFN-␣ production induced by apoptotic cells combined with lupus IgG. J Immunol 2003;171:3296–302. 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. Vollmer J, Tluk S, Schmitz C, Hamm S, Jurk M, Forsbach A, et al. Immune stimulation mediated by autoantigen binding sites within small nuclear RNAs involves Toll-like receptors 7 and 8. J Exp Med 2005;202:1575–85. Lovgren T, Eloranta ML, Kastner B, Wahren-Herlenius M, Alm GV, Ronnblom L. Induction of interferon-␣ by immune complexes or liposomes containing systemic lupus erythematosus autoantigen– and Sjögren’s syndrome autoantigen–associated RNA. Arthritis Rheum 2006;54:1917–27. Fitzgerald-Bocarsly P, Feng D. The role of type I interferon production by dendritic cells in host defense. Biochimie 2007;89: 843–55. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271–7. Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang DH, and the Committee on Prognosis Studies in SLE. Derivation of the REGULATION OF TYPE I IFN PRODUCTION 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. SLEDAI: a disease activity index for lupus patients. Arthritis Rheum 1992;35:630–40. Bach M, Krol A, Luhrmann R. Structure-probing of U1 snRNPs gradually depleted of the U1-specific proteins A, C and 70k: evidence that A interacts differentially with developmentally regulated mouse U1 snRNA variants. Nucleic Acids Res 1990;18: 449–57. Kastner B, Luhrmann R. Electron microscopy of U1 small nuclear ribonucleoprotein particles: shape of the particle and position of the 5’ RNA terminus. EMBO J 1989;8:277–86. Cederblad B, Blomberg S, Vallin H, Perers A, Alm GV, Ronnblom L. Patients with systemic lupus erythematosus have reduced numbers of circulating natural interferon-␣-producing cells. J Autoimmun 1998;11:465–70. Mathsson L, Tejde A, Carlson K, Hoglund M, Nilsson B, NilssonEkdahl K, et al. Cryoglobulin-induced cytokine production via Fc␥RIIa: inverse effects of complement blockade on the production of TNF-␣ and IL-10. Implications for the growth of malignant B-cell clones. Br J Haematol 2005;129:830–8. Della Chiesa M, Romagnani C, Thiel A, Moretta L, Moretta A. Multidirectional interactions are bridging human NK cells with plasmacytoid and monocyte-derived dendritic cells during innate immune responses. Blood 2006;108:3851–8. Gerosa F, Gobbi A, Zorzi P, Burg S, Briere F, Carra G, et al. The reciprocal interaction of NK cells with plasmacytoid or myeloid dendritic cells profoundly affects innate resistance functions. J Immunol 2005;174:727–34. Ferlazzo G, Pack M, Thomas D, Paludan C, Schmid D, Strowig T, et al. Distinct roles of IL-12 and IL-15 in human natural killer cell activation by dendritic cells from secondary lymphoid organs. Proc Natl Acad Sci U S A 2004;101:16606–11. Radstake TR, van Lent PL, Pesman GJ, Blom AB, Sweep FG, Ronnelid J, et al. High production of proinflammatory and Th1 cytokines by dendritic cells from patients with rheumatoid arthritis, and down regulation upon Fc␥R triggering. Ann Rheum Dis 2004;63:696–702. Mullazehi M, Mathsson L, Lampa J, Ronnelid J. Surface-bound anti–type II collagen–containing immune complexes induce production of tumor necrosis factor ␣, interleukin-1␤, and interleukin-8 from peripheral blood monocytes via Fc␥ receptor IIa: a potential pathophysiologic mechanism for humoral anti–type II collagen immunity in arthritis. Arthritis Rheum 2006;54:1759–71. Kazama H, Ricci JE, Herndon JM, Hoppe G, Green DR, Fergu- 2427 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. son TA. Induction of immunological tolerance by apoptotic cells requires caspase-dependent oxidation of high-mobility group box-1 protein. Immunity 2008;29:21–32. Ronnelid J, Tejde A, Mathsson L, Nilsson-Ekdahl K, Nilsson B. Immune complexes from SLE sera induce IL10 production from normal peripheral blood mononuclear cells by an Fc␥RII dependent mechanism: implications for a possible vicious cycle maintaining B cell hyperactivity in SLE. Ann Rheum Dis 2003;62: 37–42. Malygin AM, Meri S, Timonen T. Regulation of natural killer cell activity by transforming growth factor-␤ and prostaglandin E2. Scand J Immunol 1993;37:71–6. Joshi PC, Zhou X, Cuchens M, Jones Q. Prostaglandin E2 suppressed IL-15-mediated human NK cell function through down-regulation of common ␥-chain. J Immunol 2001;166:885–91. Tu Z, Bozorgzadeh A, Pierce RH, Kurtis J, Crispe N, Orloff MS. TLR-dependent cross talk between human Kupffer cells and NK cells. J Exp Med 2008;205:233–44. Romero-Reyes M, Head C, Cacalano NA, Jewett A. Potent induction of TNF-␣ during interaction of immune effectors with oral tumors as a potential mechanism for the loss of NK cell viability and function. Apoptosis 2007;12:2063–75. Betten A, Dahlgren C, Hermodsson S, Hellstrand K. Serotonin protects NK cells against oxidatively induced functional inhibition and apoptosis. J Leukoc Biol 2001;70:65–72. Betten A, Dahlgren C, Mellqvist UH, Hermodsson S, Hellstrand K. Oxygen radical induced natural killer cell dysfunction: role of myeloperoxidase and regulation by serotonin. J Leukoc Biol 2004;75:1111–5. Oates JC, Farrelly LW, Hofbauer AF, Wang W, Gilkeson GS. Association of reactive oxygen and nitrogen intermediate and complement levels with apoptosis of peripheral blood mononuclear cells in lupus patients. Arthritis Rheum 2007;56:3738–47. Bave U, Alm GV, Ronnblom L. The combination of apoptotic U937 cells and lupus IgG is a potent IFN-␣ inducer. J Immunol 2000;165:3519–26. Son Y, Ito T, Ozaki Y, Tanijiri T, Yokoi T, Nakamura K, et al. Prostaglandin E2 is a negative regulator on human plasmacytoid dendritic cells. Immunology 2006;119:36–42. Willeke P, Schluter B, Schotte H, Erren M, Mickholz E, Domschke W, et al. Increased frequency of GM-CSF secreting PBMC in patients with active systemic lupus erythematosus can be reduced by immunoadsorption. Lupus 2004;13:257–62.