IL-10 down-regulates the APC function of many dendritic cells (DC), including human peripheral blood (PB) DC. In rheumatoid arthritis (RA), synovial fluid (SF) DC express markers of differentiation and are effective APC despite abundant synovial IL-10. The regulation of DC responsiveness to IL-10 was therefore examined by comparing the effect of IL-10 on normal PB and RA SF DC. Whereas IL-10 down-modulated APC function and MHC class II and B7 expression of PB DC, IL-10 had no such effect on SF DC. Since SF DC have differentiated in vivo in the presence of proinflammatory cytokines, PB DC were cocultured in the presence of IL-10 and either GM-CSF, IL-1β, TNF-α, IL-6, or TGF-β. GM-CSF, IL-1β, and TNF-α were all able to restore APC function. Whereas the effects of IL-10 on PB DC were shown to be mediated by IL-10R1, neither PB nor RA SF DC constitutively expressed IL-10R1 mRNA or detectable surface protein. In contrast, IL-10R1 protein was demonstrated in PB and SF DC whole cell lysates, suggestive of predominant intracellular localization of the receptor. Thus, DC responsiveness to IL-10 may be regulated through modulation of cell surface IL-10R1 expression or signaling.
Interleukin-10 is a pleiotropic cytokine produced by T cells, monocytes, macrophages, B cells, keratinocytes, and several tumor cell lines (1). It exhibits both immunosuppressive and immunostimulatory properties, in that it down-regulates the function of T cells, monocytes, macrophages, and dendritic cells (DC)3 but has immunostimulatory effects on B cells and endothelial cells (1, 2). Although the exact mechanism of the down-regulation of APC function of DC by IL-10 is not fully understood, the expression of both MHC class II molecules and CD86 by DC is down-modulated (3, 4). Furthermore, the spontaneous apoptosis of murine Langerhans cells (LC) that occurs upon in vitro incubation is enhanced in the presence of IL-10 (5).
DC are potent APC whose function is specialized for the stimulation of the primary immune response. The functional capacity of DC is tightly regulated and is closely related to the process of DC differentiation. In man, DC isolated from rheumatoid arthritis (RA) synovial fluid (SF) provide an easily accessible model of functional DC that have differentiated from precursors in vivo (6). We and others have previously demonstrated expression of both IL-10 mRNA and protein in RA synovial cells (7, 8, 9, 10). Nevertheless, DC derived from RA synovium have been demonstrated to be differentiated phenotypically and functionally, in that they express markers of maturation, as well as markers of activation, including CD80 and CD86 (6). It has previously been proposed that such activated synovial DC play a key role in the perpetuation of the rheumatoid immune response (11). In view of the high levels of IL-10 detected within the rheumatoid synovium, it appeared paradoxical that DC in this location expressed markers of activation and differentiation in vivo and were fully functional in vitro. It was therefore of interest to examine the effects of IL-10 upon rheumatoid SF DC in vitro. The data demonstrate that, in contrast to normal peripheral blood (PB) DC precursors, rheumatoid SF DC are resistant to the immunosuppressive effect of IL-10 in vitro.
Materials and Methods
SF was obtained from 12 patients with RA. Five were taking methotrexate, one sulfasalazine, one hydroxychloroquine, four penicillamine, and one i.m. gold, and two were on no disease-modifying drug. Four of the patients were also on low dose prednisone. PB was obtained from normal volunteers.
All cell cultures were conducted in medium RPMI 1640 (Sigma, St. Louis, MO) supplemented with penicillin G (200 U/ml), gentamicin (10 μg/ml), l-glutamine (0.3 mg/ml), and 10% FCS (Commonwealth Serum Laboratories, Melbourne, Australia).
Cells and reagents
Human TF1 cells expressing either an irrelevant murine (ecotropic) receptor or recombinant human IL-10R (hIL-10R) were provided by Dr. K. Moore, DNAX, Palo Alto, CA (12). TF1 cells express low levels of endogenous IL-10R at the cell surface. TF1 cell lines were maintained in culture medium supplemented with human GM-CSF (20 U/ml), and, for selection purposes, 2 μg/ml puromycin (Sigma) was added to TF1-hIL-10R. The following mAb were used: FITC-conjugated Leu-M1 (Becton Dickinson, San Jose, CA) and PE/Cy5-conjugated Tuk4 (Caltag, South San Francisco, CA), directed against CD14 on human monocytes; PE-conjugated Leu-M9, directed against CD33 on human myeloid cells; Leu-11b, directed against CD16 (FcεRIII); Leu-12, directed against CD19 (Becton Dickinson); L243, directed against monomorphic determinants of HLA-DR (American Type Culture Collection (ATCC), Manassas, VA); OKT3, directed at the CD3 complex on human T cells (ATCC); B70, directed against CD86/B7-2 (PharMingen, San Diego, CA); BB1, directed against CD80/B7-1 (Ancell, Bayport, MN); 3F9, directed against the IL-10-binding epitope of the human IL-10R (provided by Dr. K. Moore; Ref. 12); and 19F1, directed against human and viral IL-10 (ATCC). Control Abs included mouse IgG1 (Dako, Carpinteria, CA), rat Ig (Dako), FITC-IgG1 and PE-IgG1 (Becton Dickinson), and biotinylated anti-mouse Ig and biotinylated anti-rat Ig (Dako). PE-conjugated streptavidin and HRP-streptavidin were purchased from Dako. Recombinant human GM-CSF was a gift from Schering-Plough, Sydney, Australia; recombinant TNF-α and recombinant IL-1β were obtained from the World Health Organization International Laboratory for Biological Standards, (Hertfordshire, U.K.). Recombinant human IL-10 was expressed in Escherichia coli and purified (13).
DC, monocyte, monocyte-derived DC (MDDC), and T cell preparation
PBMC and SF MNC were prepared as previously described (14). Briefly, heparinized venous blood from healthy adult donors, or SF from patients with RA, was sedimented over ficoll diatrizoate (Pharmacia, Uppsala, Sweden). Cells were washed and then incubated with neuraminidase-treated sheep red blood cells (NSRBC). The rosetting and nonrosetting populations were separated on ficoll diatrizoate gradients. After lysis of the red cells with 1 M ammonium chloride, PB rosetting cells were passed over a nylon wool column. Pure populations of responder T cells were prepared by magnetic immunodepletion with anti-CD14, anti-CD16, and anti-HLA-DR, followed by GAMIg-conjugated magnetic beads (Miltenyi Biotec, San Francisco, CA) as described. Greater than 98% of the recovered T cells expressed CD3 by flow cytometric analysis.
PB non-T cells were myeloid enriched by magnetic immunodepletion using anti-CD19, anti-CD3, and anti-CD16, followed by GAMIg-conjugated magnetic beads as described. By flow cytometric analysis, 98% of these cells lacked expression of CD19, CD3, and CD16, and 80–95% expressed CD33. SF non-T cell populations contained less than 10% of T, B, and NK cells and ∼30% each of DC, monocytes, and granulocytes as described previously (14).
SF non-T cells or myeloid-enriched PB non-T cells were incubated with FITC-conjugated anti-CD14 and PE-conjugated anti-CD33 on ice for 30 min and washed. Gates were set, and DC were sorted on an Epics Elite Flow Cytometer (Coulter Electronics, Hialeah, FL) as CD33+CD14dim cells and monocytes as CD33+CD14bright cells as described previously (6). The threshold was set to abort dead cells, debris, and granulocytes as determined by light scatter. Postsort analysis showed that <3% of the cells sorted as negative were positive, and >95% of the cells sorted as positive were positive.
For generation of MDDC, myeloid-enriched PB non-T cells were cultured at 1.3 × 106 cells/ml in the presence of GM-CSF (800 U/ml) and IL-4 (400 U/ml) for 7 days as previously described (54).
Staining of non-T cells for flow cytometric analysis
Freshly isolated SF non-T cells or myeloid-enriched PB non-T cells were incubated with optimal concentrations of mAb or rat Ig on ice for 30 min, washed twice, then incubated with biotinylated anti-rat Ig for 30 min on ice, washed twice, then finally incubated with streptavidin-PE, FITC-conjugated-anti-HLA-DR, and PE/Cy5-conjugated anti-CD14 with or without PE/CY5-conjugated anti-CD19 for 30 min on ice. After staining, all cells were fixed with 1% paraformaldehyde (Sigma) and analyzed on an Epics Elite Flow Cytometer using a single argon laser. DC and monocytes were gated as HLA-DR+ CD14−/dim CD19− cells and HLA-DR+CD14bright cells, respectively. In some experiments, cultured sorted DC were stained with biotinylated anti-CD86 followed by streptavidin PE/Cy5 and HLA-DR-FITC and analyzed for two colors as above.
Induction of T cell responses in MLR and DC differentiation in vitro
DC were incubated in 96-well round-bottom polypropylene dishes (Costar, Cambridge, MA) for 18 h at 37°C in the presence or absence of various cytokines or mAb as described in the text. Cells were washed and counted and either added to the MLR or stained for flow cytometry. For the MLR, cells were resuspended at 5 × 106 cells/ml in medium containing 0.08 mg/ml mitomycin C (Sigma) to inhibit cell proliferation. Cells were incubated at 37°C in the dark for 20 min, then washed three times in Hanks buffered saline solution. Various numbers of APC were incubated in triplicate in round-bottom 96-well tissue culture plates (Costar) with 105 freshly isolated purified normal allogeneic PB T cells at 37°C for 5 days. T cell proliferation was measured by the uptake of [3H]thymidine (1 μCi/well; 6.7 Ci/mM, ICN, Costa Mesa, CA), which was added during the final 18 h of the culture period. Cells were harvested onto glass fiber filter paper with an automated 96-well harvester (Wallac, Turku, Finland), and [3H]thymidine incorporation was determined by liquid scintillation spectroscopy. The responses are reported as the mean cpm ± SEM for triplicate wells.
Acridine orange assay for apoptosis
Cell cultures were stained with acridine orange, a fluorescent nuclear binding dye that facilitates the distinction of apoptotic cells from healthy cells based on changes in nuclear morphology (15). Cell counting was undertaken within 5 min of the addition of 10 μg/ml acridine orange to the culture. Apoptotic cells were identified on the basis of cellular shrinkage and condensed and highly fluorescent chromatin.
RNA isolation and RT-PCR analysis
Total RNA was isolated using RNAzol (Biotecx, Houston, Tx) according to the manufacturer’s instructions. First strand cDNA was synthesized using oligo(dT)20 (Pharmacia Biotech) as a primer and AMV-RT (Promega, Madison, WI) as previously described (16). Complementary DNA was mixed with 25 pmol each of forward and reverse oligonucleotide primers, 200 μM each of dGTP, dATP, dCTP, and dTTP (Perkin-Elmer, Norwalk, CT) in 10 mM Tris-HCl (pH 9.0), 1.5 mM MgCl2, 50 mM KCl, and 0.1% Triton X-100 in a total volume of 25 μl. The reaction mix was heated at 95°C for 7 min before adding two units of Taq polymerase (Perkin-Elmer), followed by 35 cycles of 95°C (1 min), 63°C (1 min), 72°C (1 min 30 s), and a final extension step at 72°C for 7 min. The IL-10R primer sequences were: sense primer, 5′-GTA CCA CAG CAA TGG CTA CC-3′; and antisense primer, 5′-CAC GGT GAA ATA CTG CCT GG-3′ (17). The GAPDH primers were: sense primer, 5′-ACC ACA GTC CAT GCC ATC AC-3′; and antisense primer, 5′-CAC GGT GAA ATA CTG CCT GG-3′ (Clontech, Palo Alto, CA). PCR products were analyzed by electrophoresis on a 1.8% agarose gel and visualized by staining with ethidium bromide. To confirm product specificity, purified fragments were prepared, using a PRISM Ready Reaction DyeDeoxy Terminator Cycle sequencing kit (Applied Biosytems, Foster City, CA), for analysis on the Applied Biosystem DNA Sequencing System.
After sorting or incubation, cells were washed in ice cold PBS, and protein extracts from either whole cell lysates or purified membranes were prepared. For lysates, cells were resuspended at 2 × 107 cells/ml in Nonidet P-40 (NP-40) lysis buffer (150 mM NaCl, 1% NP-40, 50 mM Tris (pH 8.0), 1 mM PMSF (Sigma)) and incubated for 30 min on ice. The lysate was centrifuged at 600 × g for 10 min, and the supernatant was collected. Lysates were diluted 1:1 with 2× gel-loading buffer (50 mM Tris-HCl, 2% w/v lauryl sulfate, 0.1% bromophenol blue, 10% w/v glycerol, and 5% v/v 2-ME (pH 8.0)) and heat denatured. For membrane purification, cells were resuspended at 2 × 107 cells/ml in ice cold homogenization buffer (10 mM Tris (pH 7.6), 0.5 mM MgCl2, 10 μg/ml leupeptin (ICN), 10 μg/ml aprotinin (ICN), 2 μg/ml pepstatin (ICN), and 1 mM PMSF), transferred to a clean, chilled Dounce homogenizer, and homogenized by delivering 30 strokes. Efficiency of homogenization was monitored by phase contrast microscopy. The homogenate was transferred to a clean microfuge tube, and one-quarter volume of restoration buffer (10 mM Tris-HCl (pH 7.6), 0.5 mM MgCl2, 0.6 M NaCl, 10 μg/ml leupeptin, 10 μg/ml aprotinin, 2 μg/ml pepstatin, and 1 mM PMSF) was added. The nuclear fraction was removed by centrifugation at 500 × g for 5 min at 4°C. The supernatant was centrifuged at 13,500 × g for 1 h at 4°C, and pelleted membrane protein was washed in TBS supplemented with 10 μg/ml leupeptin, 10 μg/ml aprotinin, and 1 mM PMSF and resuspended in 2 × gel-loading buffer at 1 × 107 cells/ml. Lysate and membrane preparations were loaded (105 or 106 cell equivalents/lane, respectively, or in some experiments, 25 μg protein/lane) alongside prestained protein m.w. standards (5 μl/lane; Bio-Rad, Hercules, CA) on an 8% polyacrylamide gel and elect-rophoresed for 1 h at 150 V. Protein was electrophoretically transferred to nitrocellulose membranes (Amersham, Sussex, England), which were blocked in 10% skim milk powder in TBS. Membranes were incubated in either 3F9 mAb (0.5 μg/ml in TBS containing 0.5% milk powder) or irrelevant mAb, washed twice in 0.1% Tween 20 in TBS (TBS-T), then incubated in biotin-conjugated anti-rat Ig in TBS-T, washed again, and incubated with streptavidin-conjugated HRP. After washing and a 1-min chemiluminescent enhanced chemiluminescence (ECL) reaction (Amersham), membranes were exposed to Hybond x-ray film (Amersham).
IL-10 down-regulates the APC function of DC derived from normal PB but not RA SF
The first experiments compared the effect of IL-10 on the APC function of freshly isolated DC derived either from normal PB or RA SF. To specifically examine the effect of IL-10 on DC and not on T cells, DC APC function was examined after their differentiation in the presence of varying concentrations of IL-10. CD33+CD14dim DC were sorted from normal PB or RA SF and preincubated for 18 h in the presence or absence of IL-10, and the ability of the washed DC to stimulate purified allogeneic PB T cells was examined. As seen in Fig. 1,A, preincubation of PB DC in the presence of IL-10 induced a dose-dependent reduction in their APC function. To exclude the possibility that death of DC induced by IL-10 was responsible, cells from parallel 24- and 48-h MLR cultures were stained with trypan blue. Equivalent numbers of viable DC were added to the MLR after preincubation in either medium or IL-10, and equivalent numbers of viable DC remained under both conditions for 48 h (data not shown). Furthermore, 24- and 48-h cultures of DC, which had been preincubated in either medium or IL-10, were examined for apoptosis using acridine orange staining. The number of apoptotic cells was similar in DC cultures irrespective of prior IL-10 exposure (data not shown). In contrast to PB DC, there was no functional effect of preincubation of SF DC in the presence of IL-10 (Fig. 1 B). The data indicate that incubation of PB but not RA SF DC in the presence of IL-10 down-regulates the APC function of the DC.
IL-10 down-regulates the expression of HLA-DR and CD86 by PB but not SF DC
Previously IL-10 has been shown to down-regulate expression of HLA-DR and CD86 expression by PB DC. To determine whether this was also true for rheumatoid SF DC, freshly isolated normal PB DC or rheumatoid SF DC were incubated with varying concentrations of IL-10 for 18 h, and the expression of cell surface molecules was subsequently examined. As shown in Fig. 2, HLA-DR and CD86 expression was down-regulated in a dose-dependent fashion by incubation in IL-10. However, there was no effect on the expression of these cell surface molecules in the case of RA SF DC (Fig. 3). Consistent with previous reports, there was no effect of IL-10 on CD80 expression by either cell type (data not shown) (4). The data demonstrate that rheumatoid SF DC are resistant to down-modulation of MHC class II molecule and CD86 expression by IL-10 and further suggest that these DC are not susceptible to IL-10 in vitro.
The effect of IL-10 on PB DC APC function is reversed in the presence of GM-CSF, IL-1β, or TNF-α
Since differentiation of PB DC precursors in the presence of IL-10 led to suppression of their APC function, and SF DC had been exposed to IL-10 in vivo, but were still functional APC, the next experiments examined whether proinflammatory cytokines present within RA synovium might influence the differentiation of DC so as to render them resistant to IL-10. Therefore, PB DC precursors were cocultured with or without IL-10 in the presence of either GM-CSF, IL-1-β, TNF-α, IL-6, or TGF-β. Cells were then washed and used to stimulate purified allogeneic T cells in MLR. As seen in Fig. 4, coculture in the presence of IL-10 and either GM-CSF, IL-1-β, or TNF-α, but not IL-6 or TGF-β, reversed the effect of IL-10 on DC APC function. Furthermore, the effects of IL-10 on PB DC HLA-DR and CD86 expression were reversible by coculture in the presence of GM-CSF, IL-1β, and TNF-α (data not shown). Hence, the data suggest that differentiation of DC in the presence of GM-CSF, IL-1β, or TNF-α can alter the ability of these DC to respond to IL-10.
IL-10R1 is essential for IL-10 signaling of PB monocytes and DC
Two components of the IL-10 receptor (IL-10R1 and IL-10R2) have been identified. However, their expression by PB DC has yet to be characterized (17, 18). The next experiments therefore examined whether IL-10R1 was involved in IL-10-mediated suppression of PB DC and monocyte APC function. Freshly purified PB DC or monocytes were preincubated for 18 h with or without IL-10 (100 U/ml) in the presence or absence of anti-IL-10R1 mAb 3F9 (5 μg/ml) or control rat Ig. This mAb recognizes the IL-10-binding epitope of IL-10R1. Cells were then washed and used to stimulate purified allogeneic T cells in MLR. 3F9, but not rat Ig effectively blocked the effects of IL-10 on PB DC and monocyte APC function, indicating that IL-10R1 is essential for IL-10 signaling of these cell populations (Fig. 5).
Expression of surface IL-10R1
Since SF DC derived from patients with RA appeared resistant to the down-modulatory effects of IL-10 and since IL-10R1 was shown to mediate IL-10 signaling of PB DC, the next experiments examined whether this resistance was due to modified IL-10R1 expression by RA SF DC. Cell surface expression of IL-10R1 by normal PB MNC and RA SF DC and monocytes was examined by flow cytometry using mAb 3F9. Cell surface IL-10R1 was expressed at low levels by normal PB T cells as previously reported (12) and by RA SF monocytes (Fig. 6, C and D, respectively). In contrast, IL-10R1 expression was detected at low levels on normal PB monocytes in some but not all donors but was not detectable on the surface of normal PB or SF DC (Fig. 6, A, B, and F, respectively). Furthermore, cell surface IL-10R1 was not detected after incubation of DC in either medium or any of the cytokines used in the functional assays shown in Fig. 4 (data not shown). Significant levels of IL-10 are found in RA SF, and previous studies have demonstrated that murine macrophages may display significant quantities of functional cell surface IL-10 (19). Thus, to determine whether the 3F9-binding epitope was blocked by IL-10 bound in vivo, freshly isolated normal PB and RA SF monocytes and DC were stained with anti-IL-10 mAb. Whereas low levels of surface IL-10 were detected on RA SF monocytes (Fig. 6,E), no IL-10 was detected on RA SF DC (Fig. 6,G), PB DC, or monocytes (data not shown). Therefore, IL-10 did not appear to be bound to RA SF DC via cell surface IL-10R1. The data demonstrate that RA SF monocytes expressed an increased level of surface IL-10R1 as compared with normal PB monocytes, some of which is bound by IL-10. In contrast, the level of expression by normal PB monocytes of some individuals, and all PB and RA SF DC, was undetectable by flow cytometry using mAb 3F9. Recently, human PB DC were found to bind IL-10 in vitro (20). Since binding studies demonstrate ∼300 IL-10R1 per T cell and that this can be detected flow cytometrically (Refs. 21 and 22 , and Fig. 6 C), the data suggest, surprisingly, that freshly isolated DC express fewer than 300 IL-10R1 at the cell surface. Finally, since this low level of expression was not detectable in the current assay, a comparison between PB and SF DC surface Il-10R1 expression could not be made.
IL-10R1 mRNA expression
Since cell surface IL-10R1 expression was undetectable on both PB and SF DC, the next experiments examined whether IL-10R1 was transcriptionally regulated by the SF DC. Using human IL-10R1-specific oligonucleotides, RT-PCR demonstrated the expression of IL-10R1 mRNA by freshly isolated PB T cells as has been previously reported (17). In contrast, as found for IL-10R1 protein expression, IL-10R1 mRNA was inconsistently detected in freshly isolated PB or SF monocytes and not detected in freshly isolated PB or SF DC (Fig. 7,A). Previously, PB monocytes and macrophage cell lines have been reported to express IL-10R1 mRNA (17, 21). To determine whether the lack of expression by myeloid cells was due to a loss of message during cell purification or a lack of sensitivity of the RT-PCR technique, CD4+ T cells were sorted and cDNA titrated. IL-10R1 mRNA was not degraded as a result of the rigorous purification protocol required for obtaining pure monocyte and DC populations, since cDNA derived from normal PB CD4+ T cells purified by sorting generated an IL-10R1 amplicon. Complementary DNA derived from an equivalent number of monocytes and DC derived from the same PB sample, however, failed to produce an IL-10R1 PCR fragment (Fig. 7,B). Furthermore, the lack of IL-10R1 mRNA expression in DC and monocytes is unlikely to be a result of insensitivity of the RT-PCR assay or an alternately spliced mRNA. First, titration of the sorted CD4+ T cell cDNA revealed that an IL-10R1 PCR fragment could be produced from as few as 32 T cells (data not shown). Second, an alternative set of IL-10R1-specific oligonucleotides failed to improve the sensitivity of the assay (data not shown). To determine whether IL-10R1 mRNA might be expressed in either myeloid precursors or upon DC differentiation, either cultured DC or myeloid-enriched non-T cells (which contain DC, monocytes, and a small percentage of myeloid precursors) were examined (54). Whereas IL-10R1 mRNA was detectable in myeloid-enriched non-T cells (Fig. 7,A), it was not detected in cDNA derived from cultured DC (Fig. 7 C). The data therefore suggest that IL-10R1 mRNA is expressed at early stages but is undetectable at later stages of myeloid differentiation.
IL-10R1 protein expression
The previous experiments demonstrated that both cell surface IL-10R1 and IL-10R1 mRNA were undetectable in PB DC and monocytes, yet these cells are capable of responding to IL-10 through IL-10R1 in vitro. Immunoblotting using anti-IL-10R1 mAb 3F9, or control rat Ig, was therefore used to determine whether IL-10R1 were present in the cells. An immunoreactive 100-kDa protein was detected in whole cell lysates of sorted PB DC and monocytes but not of equivalent numbers of T cells. The 3F9 immunoreactive protein was of an identical size to the immunoreactive IL-10R1 detected in TF1-hIL-10R lysates (Fig. 8,A). No immunoreactive protein was detected when 3F9 was replaced with control rat Ig (data not shown). These data demonstrate that immunoblotting is insufficiently sensitive to detect the small number of T cell surface IL-10R1. Nevertheless, IL-10R1 protein was readily detected in PB monocytes and DC lysates, although the intensity of the signal was variable between donors. Therefore, in these cells, IL10-R1 must be intracellular. Moreover, IL-10R1 was even more abundant in lysates from equivalent cell numbers of SF monocytes and DC (Fig. 8,A). To determine whether the difference in IL-10R1 expression between PB and SF DC and monocytes was due to a difference in cell size and activation state, cell lysates were normalized for protein, electrophoresed, and blotted. In this case, myeloid-enriched SF protein preparations contained more IL-10R1 than an equivalent amount of myeloid-enriched PB protein (data not shown). Although the flow cytometric analysis demonstrated that at least some of the SF monocyte IL-10R1 was expressed on the membrane, the abundant signal obtained by immunoblotting suggests that a larger proportion was intracellular. The immunoreactivity of whole cell lysate protein derived from 105 cells was therefore compared with that of membrane protein derived from 106 cells. Whereas IL-10R1 was detected in cell lysates prepared from either MDDC, freshly purified PB DC or monocytes, or RA SF MNC, immunoreactive protein was detectable only in the TF1-hIL-10R cell line membrane preparations (Fig. 8 B). The control membrane protein CD31 was readily detectable in membrane preparations from each cell type (data not shown). The data therefore indicate that, in contrast to T cells, IL-10R1 is predominantly intracellular in DC and monocytes. Furthermore, these cells are likely to regulate their response to IL-10 by transporting IL-10R1 between an intracellular compartment and the cell surface.
IL-10 exerts antiinflammatory effects in a variety of ways, including suppression of monocyte cytokine production, down-regulation of monocyte and DC APC and costimulatory function with consequent inhibition of T cell proliferation, as well as a direct inhibition of T cell IL-2 and TNF-α production (23, 24, 25, 26, 27). IL-10 is present at low levels in normal serum and has been demonstrated, alongside proinflammatory cytokines, in a wide variety of inflammatory and infective situations (28, 29, 30, 31). Given that IL-10 is rapidly produced in the immune response, particularly in response to TNF-α induction in monocytes and IFN-γ in T cells (32), the mechanisms whereby DC continue to function as effective APC are poorly understood. This study examined the regulation of DC responsiveness to IL-10 using as a model the relatively mature DC that can be easily obtained from the inflammatory autoimmune source RA SF. The data demonstrate that, in contrast to normal PB DC precursors, rheumatoid SF DC are resistant to the immunosuppressive effects of IL-10. Furthermore, differentiation of PB DC in the presence of the proinflammatory cytokines TNF-α, GM-CSF, and IL-1β rendered the DC unresponsive to IL-10. Although the exact mechanism of this resistance of SF DC is still not clear, the current studies suggest that DC responsiveness to IL-10 may be regulated through modulation of cell surface IL-10R1 expression or signaling.
The hIL-10R, a member of the class II cytokine receptor family, comprises two subunits, IL-10R1 and IL-10R2 (18, 33, 34). The expression of IL-10R1, the ligand binding subunit and most widely studied of the subunits, is largely restricted to hemopoietic cells. In all cell types examined, only hundreds of IL-10R1 molecules per cell have been detectable (21, 22, 35). In contrast, IL-10R2 expression appears less restricted; however, its expression and function remain to be fully elucidated. The current studies suggest that the mechanism of regulation of surface IL-10R1 expression and IL-10 responsiveness of DC might differ from that of lymphocytes. In this regard, mature circulating T cells express abundant IL-10R1 mRNA, hundreds of IL-10R1 molecules on the cell surface, and very few or no receptors intracellularly. In vitro activation results in diminished IL-10R1 mRNA and surface protein expression (17). In contrast, constitutive expression of either IL-10R1 mRNA or surface protein was below the threshold of detection in PB DC, was inconsistently detected in PB monocytes, and was detectable in PB preparations enriched for myeloid precursors. Intracellular IL-10R1 protein was detectable in freshly isolated PB monocytes and DC. Thus, it appears likely that early myeloid precursors transiently produce IL-10R1 that is stored intracellularly for transport to the cell surface. However, there are likely to be other signals for intermittent IL-10R1 mRNA production in monocytes and DC. In this regard, constitutive IL-10R1 mRNA expression by PB monocytes purified by adherence has previously been shown (17). Preliminary data indicate that freshly purified normal PB monocytes up-regulate IL-10R1 mRNA in response to adherence for 2 h in vitro (data not shown). Furthermore, the increased IL-10R1 immunoreactivity of SF DC, as compared with PB DC, implies that IL-10R1 mRNA was expressed at some stage during the differentiation of DC precursors in the synovial environment.
The functional capacity of DC is tightly regulated and is closely related to the process of DC differentiation. This process involves the up-regulation of MHC and costimulatory molecule expression, migration of mature or maturing DC to the paracortex of secondary lymphoid organs, and down-regulation of Ag uptake and processing capacity (36, 37). Coculture of freshly isolated PB DC in the presence of IL-10 and either GM-CSF, IL-1β, or TNF-α prevented the immunosuppressive effects of IL-10 on PB DC APC function. It is of interest that GM-CSF, TNF-α, and IL-1 have been shown previously to represent key cytokines involved in the differentiation and up-regulation of function of DC precursors derived from many tissue sources, including the blood (38, 39, 40, 41). Therefore, these proinflammatory cytokines present within RA synovium, and indeed at many inflammatory sites, appear to modulate the immunosuppressive effects of IL-10 on DC. However, neither IL-6 nor TGF-β could modulate the effects of IL-10 on PB DC precursors in this in vitro system. In support of these findings, IL-6 appears to be important in the early differentiation of the DC lineage within the bone marrow but has little effect on DC function at the later differentiation stages (42). Similarly, TGF-β has been demonstrated to play a critical role in the early development of LC and PB DC; however, it also has been shown to exert inhibitory effects on DC function (43, 44, 45). Although not examined here, it is also possible that the location and kinetics of exposure of differentiating, migrating DC to various pro- and antiinflammatory cytokines might be critical to the outcome of an immune response. Taken together, the data suggest that, during the differentiation of DC within the inflammatory synovial environment, the capacity to transport intracellular IL-10R1 to the cell surface or to signal through IL-10R1 is modulated, possibly through combined or sequential exposure to IL-10 and certain proinflammatory cytokines.
Previous studies indicate that LC differentiated in the presence of keratinocytes are less susceptible than fresh LC to IL-10-mediated suppression and that addition of IL-10 to the MLR has little effect unless added early (27, 46). Although it might be interpreted from these studies that mature DC are resistant to IL-10, the current data support the notion that IL-10 resistance is not related to the “mature phenotype” alone. Rather, certain factors are required to be present and to signal DC during the process of differentiation to mediate IL-10 resistance. In support of this, in vitro differentiated PB DC that express a fully mature phenotype are also susceptible to suppression by IL-10 (4). Thus, IL-10 resistance does not strictly correlate with functional maturity but is more likely to be related to DC differentiation signals affecting expression of surface IL-10R1.
The current data reflect the complexity of interactions between cytokines present in RA synovium on the outcome of the immune response. IL-10 is likely to be important for the differentiation of RA synovial B cells into plasma cells and the production of Ig and rheumatoid factor (47, 48). IL-10 has also been shown to induce cartilage destruction, mediated both directly and through TNF-α and IL-1β (49). Several studies suggest that synovial monocytes are less susceptible than PB monocytes to the immunosuppressive effects of IL-10 (50, 51, 52). Furthermore, expression of MHC class II by SF monocytes appears to be more resistant to IL-10-mediated suppression than proinflammatory cytokine secretion (50). Taken together, these and the current studies suggest that various cell types within RA synovium are responsive to IL-10 to varying degrees. Furthermore, the overall balance between proinflammatory and antiinflammatory cytokines in the synovium is likely to be less important than the determination of specific key interactions for specific cell types.
The current data have important implications for the potential efficacy of IL-10 immunotherapy in RA. Clinical trials of IL-10 immunotherapy in RA are in progress. On balance, the available data might predict that IL-10 would ameliorate disease but not induce remission, as is the case in the collagen-induced arthritis model (53). With regard to our hypothesis that TNF-α is likely to play a significant role in the orchestration of DC escape from IL-10, it is of interest that IL-10 provides little additional benefit to that induced by anti-TNF-α Abs in the collagen-induced arthritis model (53), and it is possible that part of the mechanism of action of anti-TNF-α in murine and human disease relates to enhanced cellular responsiveness to endogenous IL-10. In summary, the data provide evidence for control of the IL-10R as an important DC escape mechanism from IL-10 and suggest that manipulations that circumvent this control could enhance responsiveness to either endogenous or exogenous IL-10.
We thank K. Moore for helpful discussions and reagents, and K. Moore and I. Frazer for critical reading of the manuscript.
This work was supported by grants from the National Health and Medical Research Council of Australia and the Arthritis Foundation of Australia. R.T. and K.P.A.M. are supported by the Arthritis Foundation of Queensland.
Abbreviations used in this paper: DC, dendritic cell; LC, Langerhans cell; RA, rheumatoid arthritis; SF, synovial fluid; hIL-10R, human IL-10R; PB, peripheral blood; MDDC, monocyte-derived DC.