Enhanced Tolerance to Autoimmune Uveitis in CD200-Deficient Mice Correlates with a Pronounced Th2 Switch in Response to Antigen Challenge
http://www.100md.com
免疫学杂志 2005年第1期
Abstract
A single exposure to inhaled Ag 10 days before immunization leads to long term, Ag-specific tolerance. Respiratory tract myeloid APCs are implicated, but how regulation is invoked, and how tolerance is sustained are unclear. This study examines the in vivo function of the myeloid regulatory molecule CD200 in the process of tolerance induction. Despite earlier onset of experimental autoimmune uveitis in sham-tolerized, CD200-deficient mice, disease incidence and subsequent severity were actually reduced compared with those in wild-type mice. Protection was more effective and long term, lasting at least 28 days. Halting disease progression and tolerance in CD200–/– mice correlated with a marked increase in Th2-associated cytokine production by Ag-challenged splenocytes. Reduced overall disease and enhanced tolerance in the CD200-deficient mice in this model system were unexpected and may be related to altered populations of MHC class IIlow APC in the respiratory tract compared with wild-type mice together with associated activation of STAT6 in draining lymph nodes of tolerized mice. These data indicate that in the absence of default inhibitory CD200 receptor signaling, alternative, powerful regulatory mechanisms are invoked. This may represent either permissive dominant Th2 activation or an altered hierarchy of negative signaling by other myeloid cell-expressed regulatory molecules.
Introduction
Regulation of immune responses to infection or injury requires both initiation signals and termination signals to restore and maintain peripheral tolerance and immunologic homeostasis. IL-10 in particular plays a key role in controlling and terminating inflammatory responses (1, 2, 3), with fine-tuning of the immune response to Ag challenge being accomplished by gene families encoding related receptors with opposing functions (4). APCs of the myeloid lineage, such as macrophages and dendritic cells (DC),3 are central to these regulatory processes (5). The importance of this modulation is demonstrated by the sometimes fatal autoimmune and lymphoproliferative disorders observed in mice with targeted disruption of inhibitory receptors for IL-10 signaling (6).
In myeloid cells, activating and inhibitory receptor pairs form a family of triggering receptors that include activating triggering receptors expressed by myeloid cells as well as signal-regulating proteins SIRP and SIRP that have reciprocal roles in myeloid cell function (7, 8, 9). Now CD200 (OX-2 Ag) (10, 11) has been identified as a negative regulator of myeloid APCs that might be manipulated to provide therapeutic effects in autoimmunity and transplantation (12, 13, 14). CD200 is reported to be expressed by a variety of cells, including neurons and microvascular endothelium, and can be induced by activated T cells (15), whereas strong expression of its structurally related inhibitory receptor (CD200R) is largely restricted to cells of the myeloid lineage, including monocyte macrophages, DC, and microglia (11, 16, 17, 18). Recently, additional members of the CD200R family have been identified in mice (19). In addition to the inhibitory CD200R, which is known to be expressed by CD4+ T cells, especially polarized Th2 cells, there are at least two potentially activating isoforms, designated mCD200RLa and mCD200Lb. The latter do not bind to CD200 and have unknown ligands, but, in common with other receptor pairs, they have potential activating function through DNAX-activating protein-12 (DAP-12) adapter protein binding. These receptors also show differential, low levels of expression on other leukocyte subsets, supporting the idea that rather than merely representing redundancy of function, receptor pairs may have specialized functions in specific cells or tissues.
CD200-deficient mice (CD200–/–) exhibit various myeloid defects. These include elevated numbers of macrophages within tissues normally expressing CD200 and increased DAP-12 expression, particularly in the marginal zone of secondary lymphoid tissues, indicating myeloid cell activation. As a consequence of this phenotype, mice lacking CD200 appear to have increased susceptibility to CD4 T cell-mediated autoimmune diseases (12). In particular, CD200/CD200R regulation of microglial activation has profound effects on neuronal tissues, accelerating the onset of experimental models of autoimmunity affecting the CNS, including experimental autoimmune encephalomyelitis (EAE) (12) and experimental autoimmune uveoretinitis (EAU) (20).
EAU is mediated by retinal Ag-specific CD4+ T cells and can be modulated using various therapeutic approaches targeting Th cell function (21, 22), including induction of Ag-specific tolerance via the nasal mucosa (23, 24, 25). Activated macrophages are also required for full expression of disease (26, 27, 28, 29), but, equally, macrophages are required for resolution of inflammation. For instance, macrophages respond to signals such as IL-4 and IL-10 and actively participate in the anti-inflammatory process (30, 31, 32), demonstrating that, alternatively, activated macrophages are important for healing and tissue remodeling. Such alternatively activated macrophages have recently been described by us in the rat model of EAU (33). Myeloid APC may also have a dual role in nasal tolerance induction in EAE and EAU, where signaling by neuronally expressed CD200 must occur during the inflammatory process. In these models effective protection is associated with an initial IFN--driven priming event in cervical lymph nodes, followed by T cell apoptosis and down-regulation of the capacity of Ag-specific T cells to proliferate in response to restimulation (34, 35). IL-10 has been shown to have a pivotal role in this process (36, 37). How are these contradictory, but complementary, activities of macrophages and DC affected by the CD200/CD200R interaction?
In this study we use a model of tolerance induction by respiratory exposure to Ag in CD200–/– mice to examine the role of CD200/CD200R interactions in vivo on tolerance induction and maintenance. We chose the moderately susceptible C57BL/6 mouse EAU model (38) because uveitogenic T cells alone are insufficient to cause target organ damage. Monocyte macrophages are also necessary and prominent in the earliest inflammatory infiltrates in the retina (26). In addition, monocyte expression of type 2 NO synthase (NOS2) is required for full expression of disease (27, 39, 41). Respiratory tract DC (RTDC) and alveolar macrophages are known to be involved in tolerance induced by respiratory exposure to Ag (37, 41). T cell activation and proliferation in the draining cervical lymph are followed by systemic generation of regulatory cells in the spleen (24, 35, 42), providing a model to study myeloid APC function and T cell regulation in vivo. The data presented show that despite the accelerated disease onset that is expected in the absence of inhibitory CD200R signaling, overall disease incidence and severity were significantly reduced over time in CD200–/– mice. This result was unpredicted, but correlated with elevated numbers of regulatory T cells (Treg) and the presence of high IL-10-secreting splenic myeloid cells later in the disease process. Induction of tolerance to retinal Ag was also enhanced in CD200-deficient mice, which demonstrated an altered phenotype of APC in the respiratory tract and an enhanced Th2 switch compared with wild type (WT).
Materials and Methods
Animals and induction of EAU
CD200-deficient mice (CD200–/–) on the C57BL/6 background were generated at DNAX (Palo Alto, CA), and an isolator-reared, specific pathogen-free breeding colony was established within the biological services unit of Aberdeen University. Specific pathogen-free C57BL/6 wild type(CD200+/+) mice were purchased from Harlan Olac. Groups of three to eight mice were used as detailed in the text or figure legends for each experiment. For immunization, groups of sex and age matched mice were used at 6–8 wk of age. Mice were immunized by s.c. injection of 500 μg of peptide 1–20 of interphotoreceptor retinoid binding protein (Sigma-Genosys) in CFA (2.5 mg/ml Mycobacterium tuberculosis) plus 1 μg/ml Bordetella pertussis toxin i.p. as an additional adjuvant (38). At specified times animals were killed by CO2 asphyxiation, and eyes were enucleated for resin histology (H&E staining for histological scoring system) or for immunocytochemistry. Lymphoid tissue was also sampled to determine peptide-specific proliferative and cytokine responses. The severity of disease was assessed using a modified version of our histological grading system for rat EAU. At least three sections from each eye were scored in a masked fashion using a semiquantitative scoring system that combines the extent of the inflammatory infiltrate with tissue damage in the posterior chamber (23). All results are expressed as the mean ± 1 SEM. Comparison of histological assessment of disease in CD200–/– and CD200+/+ were analyzed using the Mann-Whitney U test (Instat software; GraphPad), and p 0.05 was considered significant. Disease incidence was analyzed using Fisher’s exact test, and p 0.05 was considered significant.
Induction of tolerance
Mucosal tolerance was induced by intranasal administration of 50 μg of peptide in 5 μl of PBS or 5 μl of PBS alone using an Oxford pipette. This regimen, administered 10 days before immunization has previously been shown to be effective in modulating EAU (25). In some experiments mice were killed at intervals over the following 48 h to examine the effects of intranasal peptide on cells in draining cervical lymph nodes and spleens or in control submandibular and mesenteric lymph nodes. In other experiments animals were immunized 10 days later with peptide or PBS in CFA with B. pertussis toxin.
Proliferation and cytokine profile assays
Lymphocyte proliferative responses to recall Ag were measured using the BrdU Flow kit (BD Biosciences) according to the manufacturer’s instructions. This test has the advantage of more detailed analysis of viable cell cycle kinetics by combining BrdU labeling of cells in S phase with measurement of total DNA content of the population as a whole. Single-cell suspensions were obtained from individual spleens by pressing the tissue through a 250-μm pore size metal sieve, and mononuclear cells were purified by Percoll density gradient centrifugation. RBC were removed by hypotonic lysis. Cultures were set up at a density of 1 x 106 cells/ml with 10 μg/ml peptide 1–20 for 96 h. This time period was established as optimum by compiling data from 48- to 120-h incubations in preliminary experiments. Each well was then pulsed with 10 μl of 1 mM BrdU for 45 min and harvested, and cells were permeabilized with Cytofix/Cytoperm (BD Biosciences) and frozen at –80°C in FCS with 10% DMSO (Sigma-Genosys) before staining and analysis. An FITC-labeled anti-BrdU Ab was used to identify the extent of BrdU incorporation, and the DNA stain 7-aminoactinomycin D was used to quantify the total DNA content. Measurements were made by two-color flow cytometry using a FACSCalibur or a FACS LSR (both from BD Biosciences, Mountain View, CA) with flow rate adjusted to give the optimal coefficient of variation of the DNA peaks. Data were obtained from at least three individual animals at each time point. Results were expressed as the mean ± SEM. For cytokine profile analysis of responding cultures, parallel 1-ml cultures were set up with 4 x 106 cells and 10 μg/ml peptide. After 72 h, supernatants were harvested, clarified by centrifugation, aliquoted, and frozen at –30°C until assay. Negative control cultures contained PBS in place of peptide, and 2.5 μg/ml Con A was added to positive control cultures to demonstrate optimum growth conditions.
IL-10 and IL-12 were measured using optELISA kits (BD Pharmingen) exactly according to the manufacturer’s instructions. The mouse cytokine bead array (CBA kit; BD Pharmingen) was used to measure other cytokines. Results are expressed as the mean ± SEM of supernatants from at least six animals in each group, and differences were considered significant at a value of p < 0.05 using unpaired Student’s t test.
Isolation and phenotyping of respiratory tract cells
Mice were anesthetized with 0.1 ml of Sagatol (Aventis Pharma) i.p. and were perfused with Ca2+/Mg2+-free PBS containing 10 U/ml heparin and 5 mM EDTA to remove blood from the respiratory tract vessels. The lungs and trachea were removed and disrupted mechanically by chopping and passing through a 250-μm pore size mesh. The resulting suspension was collected and treated with 0.7 mg/ml collagenase and 30 U/ml DNase in RPMI 1640 for 1 h at 37°C with agitation. The cells were then washed and passed through a 100-μm pore size mesh before centrifugation over Percoll 1.075. Cells were then washed, and CD45+ cells were analyzed by flow cytometry (FACSCalibur) for myeloid phenotype (CD11b, CD11c, F4/80, and CD204) and for activation markers (MHC class II, CD40, CD80, and CD86) as described below.
To determine myeloid cell distribution within the tissue, cryostat sections of Evan’s Blue-perfused lung tissue were stained with FITC-conjugated CD11b mAb M1/70 or IgG control mAb R35-95 (BD Pharmingen), and digital images were captured with an LSM confocal microscope (Zeiss) using the same laser settings for each image in the experiment. The relative abundance of CD11b fluorescence signal in each image was analyzed by measuring pixel values within defined regions using Laserpix software (Bio-Rad). Positively staining cells were counted using the Count-Size command and the Histogram command used to measure the average intensity value (mean) of positively stained cells.
Generation of bone marrow-derived DC (BMDC)
BMDC were prepared by a modification of the procedure described by Inaba et al. (43) as previously described (44). Briefly, 7.5 x 105 precursors were set up in 1-ml cultures supplemented with GM-CSF. The cultures were fed, and nonadherent cells were removed every 2 days by swirling and aspirating 75% of medium and adding fresh growth medium with GM-CSF. Developing DC remained attached to the bed of firmly adherent macrophages. Six days after initiation of the cultures, the loosely adherent clusters of DC were harvested, and contaminating granulocytes were depleted using anti-mouse Gr-1 mAb (RB6-8C5; BD Pharmingen) and Dynabeads (Dynal Biotech). Purified BMDC were further seeded at 1 x 106/ml/well in 24-well plates supplemented with GM-CSF. Then 1 μg of LPS was added to the cultures at various time points (0–24 h). Supernatant was then collected from each well 22 h after administration of LPS for cytokine measurement. Medium alone DCs were used as control.
Immunocytochemistry
Eyes from immunized mice or lymphoid tissue were dissected, snap-frozen in OCT (Miles), and 7-μm serial sections were cut, air-dried, and fixed in 100% cold acetone for immunocytochemistry using the alkaline phosphatase anti-alkaline phosphatase technique. After rehydration in TBS, sections were blocked with TBS/1% normal rabbit serum and then avidin D block solution (Vector Laboratories) for staining with polyclonal rabbit anti-STAT 4 (C20; Santa Cruz Biotechnology) STAT 6 (M20; Santa Cruz Biotechnology) or with phosphorylated STAT6 (Tyr641; New England Biolabs) using irrelevant polyclonal Ab and blocking peptides as negative controls. Other sections were stained using rat mAbs to murine CD3 and myeloid cell markers F4/80 Ag (CI:A3-1), MOMA-1, and MOMA-2 (Serotec). Activation markers included NOS2 (clone 6; Transduction Laboratories), rat anti-mouse CD86 (GL-1; BD Pharmingen), and MHC class II (I-Ab; P7.7; Serotec). Positive staining was detected by mouse absorbed biotinylated anti-rabbit Ig-alkaline phosphatase or biotinylated anti-rat Ig-alkaline phosphatase conjugate, followed by streptavidin:avidin-biotin peroxidase-alkaline phosphatase complex and Fast Red substrate (DakoCytomation) lightly counterstained with hematoxylin. Sections for image analysis were cut and stained in batches to ensure uniform labeling conditions for each Ab. Sections were then analyzed using the Aphelion Active X image analysis program from ADCIS. The program was adapted using Visual basic to allow analysis of immunostaining in user-defined regions of the image. An average value (percentage of tissue stained per x20 field) for each section was obtained from four to six fields. Results are expressed as the mean ± SEM of tissue from a minimum of three animals per treatment group. Differences between groups were analyzed using unpaired Student’s t test, and p < 0.05 was considered significant.
Flow cytometry
A FACSCalibur or FACS LSR (BD Biosciences) was used for data acquisition, and CellQuest software (BD Biosciences) was used for data analysis. APCs and lymphocytes isolated from lymph nodes and spleens were evaluated by multicolor immunofluorescent staining with mAbs to the following cell surface markers: CD11b (M1/70), CD11c (HL3), CD4 (RM4-4) CD45RB (16A), CD45R/B220 (RA3-6B2), CD40 (3/23), CD86 (GL-1), CD152 (BN13), MHC class II (I-Ab; AF6-120.1), CD25 (PC61), CD38 (92), CD8a (53-6B2), CD62L (MEL-14), and CD3 (145-2C11) were all from BD Pharmingen. For intracellular detection of IL-10, quadruple staining was performed. Cells were treated with Cytofix-Cytoperm and stained with PE-conjugated anti-IL-10 (JES5-16E3) following the manufacturer’s instructions (BD Pharmingen). Anti-CD204 (2F8), F4/80 (CI:A3-1), metallophillic macrophages (MOMA-1), and CD80 (RMMP-1) were obtained from Serotec. These were conjugated to FITC, PE, allophycocyanin, PerCP, PerCP-Cy5.5, or biotin as required. Biotin-labeled Abs were detected by addition of streptavidin-allophycocyanin (1/400; BD Pharmingen). Negative isotype controls and single positive controls were performed to allow accurate breakthrough compensation. Gates and instrument settings were set according to forward and side scatter characteristics, and populations were gated to exclude dead or clumped cells. A total of 10,000 events were collected. Fluorescence analysis was performed after additional backgating to exclude nonspecific or background staining. Data were collected from at least four individual animals at each time point in each experiment; the cells were prepared and stained for analysis on the same day. Histogram results are expressed as the mean ± SEM. Results shown are representative of at least four separate experiments.
Immunoprecipitation of STAT6
Activation of STAT6 expression was also demonstrated by immunoblotting of immunoprecipitates prepared from cytosolic and nuclear extracts from lymphoid tissues as previously described (43). Briefly, cytosolic extracts were prepared from 10 (7, 8) cells in 20 μl of lysis buffer A. Nuclei were removed by centrifugation, and supernatants were clarified and collected as the cytoplasmic fractions. Nuclei were then incubated in 20 μl of lysis buffer B (containing 0.42 M NaCl) and centrifuged, and supernatants were collected as nucleic fractions. STAT6-associated proteins were immunoprecipitated by anti-STAT6 Ab (M20; Santa Cruz Biotechnology), collected on protein A-Sepharose beads (Amersham Biosciences, Arlington Heights, IL), and resolved by SDS-PAGE. Protein was transferred onto a Hybond polyvinylidene difluoride membrane (Amersham Biosciences) by electroblotting. Initial probing was performed for phosphotyrosine (mAb 4G10; Upstate Biotechnology) detected by ECL. Membranes were then stripped and reprobed with STAT6 Ab to confirm the specificity of the signal.
Results
Tolerance induction in CD200–/– mice is robust, and disease incidence and severity are significantly reduced compared with those in WT mice
Mice were given either peptide 1–20 (tolerized group) or control PBS (sham-tolerized group) intranasally 10 days before immunization. Fig. 1 shows the pattern of disease progression in tolerized and control WT and CD200–/– mice over time. In WT mice, disease progressed rapidly, with all mice showing moderate to severe retinal inflammation and photoreceptor damage by 21 days postimmunization, persisting to day 28. In accordance with our previous observations (20), disease onset was accelerated in sham-tolerized CD200–/– mice, with expression of NOS2 in cells of the ciliary body, retinal vascular endothelium, and inner plexiform layer (IPL) of the retina being the earliest signs of disease on day 10 (Fig. 2, a and b) and significantly elevated disease scores by day 16 (p < 0.03). However, as in our previous study, no overall increase in tissue damage compared with WT mice was observed, and no additional significant disease progression occurred in sham-tolerized CD200–/– mice. By day 28 postimmunization, disease scores were, in fact, significantly lower than those in sham-tolerized WT mice (p < 0.05), with 17% mice still showing no signs of disease. No significant increase in histological scores of either sham-tolerized or tolerized CD200–/– mice from days 16–28 postimmunization was found. Significant reductions in the overall histological scores in both tolerized groups compared with their sham-tolerized controls were also observed on day 16 (WT, p < 0.01; CD200–/–, p < 0.03; Fig. 2, c–f). No significant difference in overall histological scores between tolerized WT and tolerized CD200–/– mice were found, but on day16, disease incidence in tolerized CD200–/– mice was significantly higher than that in WT tolerized mice (p < 0.05), reflecting an earlier onset of disease in these mice. By day 28, however, the situation was reversed, and disease incidence in the CD200–/– tolerized group was very significantly reduced compared with that in tolerized WT mice (p < 0.006) as disease incidence continued to increase in the WT groups.
Despite severe disease in WT mice, protection was maintained in the tolerized groups at all time points (p < 0.01). In the tolerized CD200–/– group, protection was also effective. Disease scores were significantly reduced on day 16 (p < 0.03) and day 28 (p<0.01), with up to 50% of the animals still showing no signs of disease.
In vitro proliferation of splenocytes does not correlate closely with in vivo inflammation during early disease, but is reduced in tolerized mice in late disease
In a previous study we showed that T cell proliferative responses by splenocytes to immunizing peptide 1–20 were equivalent in both WT and CD200–/– mice (20). We confirmed this result using a sensitive flow cytometric assay that combines BrdU labeling of cells in S phase with measurement of total DNA content of the population as a whole (Fig. 3a). It was also found that intranasal exposure to Ag had no inhibitory effect on peak proliferation of splenocytes on day 16 or day 21 despite evidence of a significant reduction in disease in tolerized animals at these time points (Fig. 1); however, proliferation by splenocytes from CD200–/– tolerized mice was significantly reduced compared with that in tolerized WT mice (p < 0.02). By day 28 postimmunization, the numbers of cells in S phase were reduced in both tolerized groups, and this was highly significant in the CD200–/– tolerized group compared with the sham-tolerized group (p < 0.01). Analysis of DNA content of cells suggested that this reduction of cells in S phase was due to reduced turnover of proliferating cells rather than a block in cell cycle progression or an increase in apoptotic or necrotic cells within the cultures (Fig. 3, b–d).
Protection from disease in CD200–/– mice correlates with a switch to IL-4 and IL-5, followed by IL-10 production by restimulated splenocytes
Splenocyte cultures set up in parallel with the proliferation assays were analyzed for cytokine production in response to restimulation with peptide (Fig. 4). Significantly higher levels of IL-4 were found in the tolerized CD200–/– groups on days 16 and 21 (p < 0.02, and p < 0.04 compared with sham-tolerized CD200–/– mice; p<0.01 compared with tolerized WT mice at the same time point). IL-5 was also elevated in CD200–/– tolerized mice on day 21 (p < 0.04 compared with sham-tolerized CD200–/–; p<0.02 compared with tolerized WT). This pattern was replaced by enhanced levels of IL-10 in the CD200–/– tolerized mice by day 28 (p < 0.05 compared with sham-tolerized; p < 0.02 compared with WT tolerized). IL-2 and TNF were elevated in sham-tolerized WT mice on days 16 and 21 and were significantly reduced in tolerized WT on day 16 (p < 0.05). Levels of IL-12 were low in all the cultures. Large quantities of IFN- (5–15 ng/ml) were also generated in all the cultures at all time points, but no significant differences between groups were observed (data not shown). Changes in IgG2a/IgG1 ratios can be indicative of changes in the Th1/Th2 balance of an immune response. However, due to the short half-life of mouse Ig isotypes (6–8 days for both IgG1 and IgG2a), these were inconclusive given the time span of these experiments (2–4 wk; data not shown).
STAT-6 is up-regulated in draining lymph nodes and spleen of CD200 deficient mice
The cytokine secretion assays indicated that tolerance in the CD200–/– mice appeared to be accompanied by an earlier and more pronounced Th2 switch in response to restimulation by Ag. To test whether this represented a different response to the tolerization regimen in CD200–/– mice we examined the effect of intranasal administration of Ag on cells of the draining lymph nodes and spleen. Expression of phenotypic myeloid cell markers, MHC class II, APC costimulatory molecules CD80 and CD86 as well as STAT 4 and STAT 6 were measured by image analysis of serial sections of immunostained tissue. Cervical lymph nodes that drain the nasal mucosa, lymphoid areas (white pulp) and macrophage areas (red pulp) of the spleen were examined up to 48 h post intranasal application of Ag. The submandibular lymph nodes were analyzed in parallel to serve as control tissue. Fig. 5, a–d, and Table I show that STAT6, which controls the Th2 differentiation pathway (45, 46), was markedly up-regulated in the spleen and cervical lymph nodes of tolerized CD200–/– mice 24 h after treatment. The effect was transient in the cervical nodes, levels returning to control levels by 48 h, but sustained in both lymphoid and macrophage areas of the spleen. Expression of STAT6 was also increased in the spleens, but not lymph nodes of WT mice, where levels actually declined. Immunostaining of sections with an Ab to phosphorylated STAT6 showed activation of the STAT6 pathway in the cervical lymph nodes (Fig. 5, e and f), and this was confirmed by Western blotting of cytosolic and nuclear extracts, which showed translocation of STAT6 from cytosol to nucleus in CD200–/–, but not WT spleens (Fig. 5, g and h). STAT4, which is activated after IL-12 signaling to drive Th1 responses (47, 48), was also transiently increased in draining lymph nodes and spleen of CD200–/– mice 24 h post-treatment (Table I). This was consistent with the initial Th1 priming previously observed in our rat tolerance model (24) and was accompanied by increased CD86 expression in cervical lymph nodes. However, in contrast, no increase in STAT4 expression was observed in the WT mice, levels actually falling during the sampling period. It was noted that overall expression STAT4 was low compared with STAT6 in this strain of mouse and may be consistent with the low levels of IL-12 p70 generated by Ag-restimulated splenocytes (Fig. 4).
With the exception of a significant increase in STAT6, and a transient increase in STAT4 expression in CD200–/– mice, no significant differences between control and test tissues were observed in the control submandibular tissue. Compared with percentages for cervical tissue, increases were comparatively small. This might reflect changes induced locally through ingestion of very small quantities of Ag during intranasal administration, the effects of disseminated Ag (24), or the systemic effects of changes in the spleen in these animals. No substantial or consistent changes in the percentage of F4/80- or MOMA-2-positive cells after treatment or between groups was observed to account for increased or decreased STAT expression in the tissues. Similarly, although some fluctuation in MHC class II expression was apparent, no significant pattern was found (data not shown).
DC from CD200–/– mice do not have a defect in IL-12 production
Low production of IL-12 in the Ag-restimulated spleen cells (Fig. 4) together with the low levels of STAT4 Ag detected in the lymph nodes and spleens (Table I) may represent a fundamental deficiency in CD200–/– mice that reduced their ability to mount a normal Th1 response, allowing a Th2 response to emerge by default. To determine the IL-12 vs IL-10 secretion potential of APC from CD200–/– mice, we used the model developed by Jiang et al. (44) that allows BMDC to be matured to secrete either IL-10 or IL-12. BMDC from WT C57BL/6 and CD200–/– C57BL/6 mice secreted both IL-10 and IL12 at high levels (>700 pg/ml) and with similar kinetics, indicating that there was no fundamental defect in the mice to account for the low IL-12 responses observed in our model (data not shown).
Altered phenotype of respiratory tract APC in CD200–/– compared with WT mice
RTDC have been shown to preferentially stimulate Th2 responses (41) and are implicated in the induction of tolerance to nasally administered Ags (37). Because altered populations of myeloid cells are a major feature of CD200–/– mice (12), we compared the phenotype of CD45+ cells from the respiratory tracts of WT and CD200–/– mice isolated and purified using collagenase/DNase digestion and density gradient centrifugation. In WT mice, the major population isolated were CD11c+ DC (<80%), with few (10–15%) CD11b+ cells. In contrast, in CD200–/– mice, CD11b+ cells formed a far larger proportion (<40%) of the purified cell population, with fewer CD11c+ DC (35–40%; Fig. 6a). The cells from both WT and CD200–/– exhibited low levels of activation markers, as expected for respiratory tract APC, but it was also apparent that both CD11b+ and CD11c+ cells from CD200–/– respiratory tract expressed lower levels of MHC class II Ag than WT (Fig. 6b). Confocal analysis of respiratory tract tissue showed that this increased proportion of CD11b+ cells present in CD200–/– mice was, in fact, due to increased numbers of CD11b+ cells within the tissue (p < 0.001; Fig. 6c). Increased detection was not due to higher levels of CD11b expression by positively stained cells, because there was no difference in average pixel intensity between the groups (Fig. 6d).
CD25+ Treg and IL-10high APC are present in the spleens of tolerized CD200–/– mice
Tolerance induced by respiratory tract exposure to Ag has been shown to be dependent on IL-10 and may be mediated by Treg (3, 36, 37). Regulatory cells are also known to be induced by an increase in Ag-specific Th2 activity together with decreased CNS infiltration in a C57BL/6 mouse model of EAE (49). We therefore used flow cytometry to examine the spleens of tolerized, immunized mice for the presence of Treg cells and to identify the phenotype of IL-10-secreting cells. Permeabilization of cells to detect intracellular Ag can reduce or interfere with the fluorescence of surface markers such as CD25, so we used triple fluorescence on unpermeabilized cells to establish the numbers of CD3+CD4+CD25+ present in the spleens of control and treated mice (Fig. 7a). In unimmunized control mice, 5–8% of CD3+ T cells from the spleen displayed the CD3+CD4+CD25+ (contact-dependent) regulatory cell phenotype (data not shown). This was typically increased to 10–15% in immunized mice. A trend toward higher numbers of CD4+CD25+ cells in the CD3+ gated population in CD200–/– mice was noticed, and the number was significantly elevated in tolerized CD200–/– mice compared with tolerized WT mice at all time points (day 16, p<0.01; days 21 and 28, p < 0.05). All CD3+CD4+CD25+ cells were also CTLA-4+ (data not shown).
IL-10 may be produced by CD25+ Treg cells (50), but Fig. 7b shows that CD3+ T cells did not express levels of intracellular IL-10 above background. IL-10-producing cells were found within the spleen myeloid cell population. Two main populations could be identified from day 21 onward. These were predominantly IL-10lowCD11b+ or small numbers of IL-10highCD11clow/– (Fig. 7b). IL-10low-secreting CD11b+ were also observed at later time points (day 28) in all mice (data not shown), but no significant differences between the groups were found.
It should be noted that the levels of IL-10 detected in these cells should reflect the in vivo situation, because cells were isolated and analyzed directly ex vivo. The cells were not given additional activation stimuli, and Golgi function was not blocked to artificially increase intracellular cytokine levels.
Discussion
In this study we have examined the role of CD200, an inhibitory ligand for myeloid cells, in a model of tolerance induction to a retinal autoantigen. Unexpectedly, we found that although disease onset in CD200–/– mice was accelerated, in agreement with previous studies using these mice (12, 20), over the longer term, overall disease incidence and severity compared with WT were significantly reduced. Furthermore, tolerance induction in CD200–/– was efficient, with up to 50% of eyes still protected from disease 28 days postimmunization (Fig. 1). Given our current understanding of the CD200/CD200R axis (15, 17, 20) and the results of other studies showing enhanced immunosuppression in autoimmune or transplant models where CD200 expression is elevated or signaling is potentiated by infusion of CD200 immunoadhesions (14, 51), these findings are counterintuitive.
Protection in tolerized CD200–/– mice appeared to correlate with increased Th2-associated cytokines and IL-10 in response to recall Ag, and overall reduced disease in sham-tolerized CD200–/– mice may be associated with the emergence of IL-10-secreting cells in the spleen later in the disease process. Based on the apparent redundancy in regulatory mechanisms controlling potentially tissue damaging inflammation, one possibility is that in a normal individual, the CD200/CD200R axis alone is not critical for the maintenance of immunological homeostasis (although it may contribute to it), but during inflammation, in the absence of default and necessary CD200/CD200R signaling, other regulatory mechanisms are invoked to convert the classically activated myeloid phenotype to an alternatively activated, healing, and immunomodulatory phenotype.
In support of this hypothesis is the emerging principle of dominant inhibitory signaling mediated by self-receptors that governs the maintenance of self-tolerance. Triggering receptors expressed by tissue-trafficking APC such as macrophages and DC or other leukocytes are key molecules in this process (4, 52, 53). In addition to these myeloid cell receptors, active self-regulation to control autoreactive T cell expansion through naturally occurring CD4+CD25+ T cell populations is now well accepted (54, 55). These, together with induced regulatory mechanisms driving pathogen-specific Th1/Th2 responses through differential cytokine expression (reviewed by in Ref. 56), induction of type 1 Treg cells (reviewed in Ref. 57), and alternative macrophage activation and scavenger receptor expression (29, 58) are invoked to avoid excessive immune pathology. How these diverse regulatory mechanisms interact in complex in vivo models, and the rules governing functional redundancy and hierarchy of responses have yet to be fully understood.
Recently, it has been shown that the CD200R gene family also has receptor isoforms with potential inhibitory and activating functions (19). In addition to the inhibitory isoform that binds CD200 and regulates myeloid activity in the mouse, at least two activating isoforms (CD200RLa and CD200RLb) were identified. These isoforms probably have alternative ligands and are able to pair with the adaptor protein DAP-12, supporting the idea that the CD200R family is also involved in fine-tuning of myeloid cell responses (15). Pertinent to the present study, Wright et al. (19) also showed CD200R expression by CD4+T cells and, at least at the mRNA level, preferentially by polarized Th2 cells. Thus, in Ag-specific in vivo models using reagents that modulate CD200R signaling, effects may be expected at the level of both myeloid APC and Th2 cell effector function. Enhancing inhibitory signaling received by myeloid cells may provide an overriding "off" signal to the innate and adaptive immune response and prolong graft survival or reduce autoimmune inflammation. Conversely, reagents that temporarily blockade theCD200/CD200R signaling pathway remove the dominant inhibitory effect of the inhibitory isoform of the receptor, thus promoting cellular activation and expression of NOS2 by myeloid cells. Whether Th2 cell function is also controlled through CD200R expression is currently unknown, although the present data suggest that it may be.
What happens in the long term absence of default CD200R signaling, as occurs in the CD200–/– mouse? The apparently mild phenotype suggests that the CD200/CD200R axis has a specialized function that is not generally required in the normal individual. The classically activated myeloid cell phenotype, including constitutive expression of NOS2 by retinal microglia, has the short term effect of accelerating EAU disease onset, but this was not associated with increased tissue damage (20), and we now show that accelerated disease is followed by a halt in disease progress and a down-regulation of inflammation (Fig. 1). The mechanisms are not fully defined, but significantly increased numbers of CD3+CD4+CD25+ were found in tolerized CD200–/– mice from day 16, with IL-10-secreting CD11b+ and CD11clow/– cells present in the spleens of both sham-tolerized and tolerized CD200–/– mice from day 21 (Fig. 7). The origin and lineage of these cells are not yet understood, but because they appear 3–4 wk after disease onset, prolonged stimulation, sequential differentiation of cell subsets, and/or several rounds of cell division must be involved. Several subsets of Treg have been described, and although some controversy surrounds the use of particular phenotypic or cytokine secretion profiles to define them, it is clear that more than one subset of Treg exists and that even small numbers can have profound immunosuppressive effects (50, 59, 60, 61, 62), either directly through inhibition of IL-2 transcription in the responder cell or indirectly through IL-10. These have been variously described as natural (typically cytokine-independent CD4+CD25+), adaptive (typically cytokine-dependent type 1 Treg) (61), or Ag-induced (60), and there are data suggesting that several rounds of stimulation are required for the induction of adaptive Treg during an inflammatory response (2, 63). Although the induction of IL-10 and the suppression of IL-2 in all groups on day 28 are certainly consistent with the induction of functional Treg cells during the disease process (54, 55), and there are data linking nasal administration of Ag with induction of Treg1 (36, 37), we found no direct evidence for IL-10-secreting CD4+ Treg in spleens of protected mice.
What effect might lack of signaling through CD200R expressed by myeloid APC and Th2 cells have on mucosal tolerance induction? Significantly elevated levels of Th2 cytokines and IL-10 were prominent in the cytokine responses of tolerized CD200-deficient mice. Pulmonary DC play a critical role in the selective immune response to inhaled Ag, inducing T cell hyporesponsiveness to innocuous Ags or preferential activation and expansion of Th2-biased responses (37, 41, 42, 64). This has been attributed to the mucosal microenvironment and immature phenotype of these cells. IL-10 has also been shown to have a critical role both in the induction of nasal tolerance and in limiting inflammation later in disease (3, 34). IL-10 would then be able to augment tolerance in an Ag-specific manner, because a single exposure to IL-10 can convert DC to a tolerogenic phenotype. Examination of the draining lymph nodes and spleen of tolerized mice 24–48 h after intranasal administration of Ag revealed a pronounced induction of STAT6 expression in CD200–/– mice. STAT6 is required for IL-4 production (46), suggesting that an enhanced Th2 switch in response to Ag presentation could be one mechanism underlying more rapid resolution of inflammation and induction of more effective tolerance to immunizing Ag in CD200–/– mice. This could be due to the altered phenotype and reduced proportion of CD11c+ to CD11b+ RTDC present in CD200–/– mice presenting the nasally administered Ag in a tolerogenic fashion. However, given the endogenous state of tonic activation expected in these cells from the CD200–/– background, an alternative mechanism, involving active signaling, could be involved. Although Th1 responses seem to be driven by pathogen-activated innate immune responses, the signals that initiate Th2 responses are still uncertain. It has been proposed that Th2 responses emerge by default in the absence of the activating innate signals, IFN- and IL-12, that would drive Th1 responses, and that alternative signals from DC might trigger the initial production of IL-4 (56). Our data clearly show that CD200–/– BMDC were able to secrete equally high levels of both IL-12 and IL-10 in response to an innate maturation signal, suggesting that the cytokine bias observed in our model was induced. Because the inhibitory form of CD200R is highly expressed on polarized Th2 cells and not on Th1 cells, the inference is that the CD200/CD200R axis can regulate Th2-mediated responses.
In conclusion, we have shown that in the absence of default inhibitory CD200/CD200R signaling, accelerated EAU onset is observed in CD200–/– mice immunized with retinal autoantigenic peptide in CFA, consistent with an increased level of classically activated microglia and myeloid APCs in these mice. However, this response is effectively down-regulated, with significantly reduced overall disease evident at mid and late disease stages. This is consistent with induction of alternative IL-10-dependent regulatory pathways, possibly induced through dominant activating CD200RLa signaling expected in Th2 cells that are not receiving an inhibitory CD200/CD200R signal. The enhanced tolerance to EAU induction observed in CD200–/– mice correlated directly with up-regulated Th2-associated STAT6 activation in cervical lymph nodes of nasally tolerized mice and with elevated Th2-associated cytokine levels in response to recall Ag, again implicating active enhancement of Th2 cell activation. Additional experiments are required to confirm that dominant CD200RLa signaling preferentially activates Th2 cells in CD200–/– mice, leading to predominant IL-4 secretion and alternative activation of myeloid APC as a regulatory mechanism to control excessive inflammation.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by project grants from the Wellcome Trust, Tenovus Scotland, Grampian Universities Hospitals National Health Service Trust Endowments, and the National Eye Center, Bristol. DNAX is supported by Schering Plough Corp. (Kenilworth, NJ).
2 Address correspondence and reprint requests to Dr. Janet Liversidge, College of Medicine and Medical Sciences, Institute of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen, U.K. AB25 2ZD. E-mail address: j.liversidge@abdn.ac.uk
3 Abbreviations used in this paper: DC, dendritic cell; BMDC, bone marrow-derived DC; DAP-12, DNAX-activating protein-12; EAE, experimental autoimmune encephalomyelitis; EAU, experimental autoimmune uveoretinitis; NOS2, type 2 NO synthase; RTDC, respiratory tract DC; Treg, regulatory T cell; WT, wild type.
Received for publication June 29, 2004. Accepted for publication October 13, 2004.
References
Moore, K. W., M. R. de Waal, R. L. Coffman, A. O’Garra. 2001. Interleukin-10 and the interleukin-10 receptor. Annu. Rev. Immunol. 19:683.
McGuirk, P., C. McCann, K. H. Mills. 2002. Pathogen-specific T regulatory 1 cells induced in the respiratory tract by a bacterial molecule that stimulates interleukin 10 production by dendritic cells: a novel strategy for evasion of protective T helper type 1 responses by Bordetella pertussis. J. Exp. Med. 195:221.
Kaya, Z., K. M. Dohmen, Y. Wang, J. Schlichting, M. Afanasyeva, F. Leuschner, N. R. Rose. 2002. Cutting edge: a critical role for IL-10 in induction of nasal tolerance in experimental autoimmune myocarditis. J. Immunol. 168:1552.
Lanier, L. L.. 2001. Face off-the interplay between activating and inhibitory immune receptors. Curr. Opin. Immunol. 13:326.
Goerdt, S., C. E. Orfanos. 1999. Other functions, other genes: alternative activation of antigen-presenting cells. Immunity 10:137.
Kuhn, R., J. Lohler, D. Rennick, K. Rajewsky, W. Muller. 1993. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75:263.
Cant, C. A., A. Ullrich. 2001. Signal regulation by family conspiracy. Cell. Mol. Life Sci. 58:117.
Dietrich, J., M. Cella, M. Seiffert, H. J. Buhring, M. Colonna. 2000. Cutting edge: signal-regulatory protein 1 is a DAP12-associated activating receptor expressed in myeloid cells. J. Immunol. 164:9.
Colonna, M.. 2003. TREMs in the immune system and beyond. Nat. Rev. Immunol. 3:445.
Barclay, A. N., H. A. Ward. 1982. Purification and chemical characterisation of membrane glycoproteins from rat thymocytes and brain, recognised by monoclonal antibody MRC OX 2. Eur.J. Biochem. 129:447.
Wright, G. J., M. Jones, M. J. Puklavec, M. H. Brown, A. N. Barclay. 2001. The unusual distribution of the neuronal/lymphoid cell surface CD200 (OX2) glycoprotein is conserved in humans. Immunology 102:173.
Hoek, R. M., S. R. Ruuls, C. A. Murphy, G. J. Wright, R. Goddard, S. M. Zurawski, B. Blom, M. E. Homola, W. J. Streit, M. H. Brown, et al 2000. Down-regulation of the macrophage lineage through interaction with OX2 (CD200). Science 290:1768.
Gorczynski, R. M., K. Yu, D. Clark. 2000. Receptor engagement on cells expressing a ligand for the tolerance-inducing molecule OX2 induces an immunoregulatory population that inhibits alloreactivity in vitro and in vivo. J. Immunol. 165:4854.
Gorczynski, R. M., Z. Chen, Y. Kai, J. Lei. 2000. Evidence for persistent expression of OX2 as a necessary component of prolonged renal allograft survival following portal vein immunization. Clin. Immunol. 97:69.
Barclay, A. N., G. J. Wright, G. Brooke, M. H. Brown. 2002. CD200 and membrane protein interactions in the control of myeloid cells. Trends Immunol. 23:285.
Preston, S., G. J. Wright, K. Starr, A. N. Barclay, M. H. Brown. 1997. The leukocyte/neuron cell surface antigen OX2 binds to a ligand on macrophages. Eur. J. Immunol. 27:1911.
Wright, G. J., M. J. Puklavec, A. C. Willis, R. M. Hoek, J. D. Sedgwick, M. H. Brown, A. N. Barclay. 2000. Lymphoid/neuronal cell surface OX2 glycoprotein recognizes a novel receptor on macrophages implicated in the control of their function. Immunity 13:233.
Dick, A. D., C. Broderick, J. V. Forrester, G. J. Wright. 2001. Distribution of OX2 antigen and OX2 receptor within retina. Invest Ophthalmol. Vis. Sci. 42:170.
Wright, G. J., H. Cherwinski, M. Foster-Cuevas, G. Brooke, M. J. Puklavec, M. Bigler, Y. Song, M. Jenmalm, D. Gorman, T. McClanahan, et al 2003. Characterization of the CD200 receptor family in mice and humans and their interactions with CD200. J. Immunol. 171:3034.
Broderick, C., R. M. Hoek, J. V. Forrester, J. Liversidge, J. D. Sedgwick, A. D. Dick. 2002. Constitutive retinal CD200 expression regulates resident microglia and activation state of inflammatory cells during experimental autoimmune uveoretinitis. Am. J. Pathol. 161:166.
Dick, A. D.. 2000. Immune mechanisms of uveitis: insights into disease pathogenesis and treatment. Int. Ophthalmol. Clin. 40:1.
Dick, A. D.. 1999. Immune regulation of uveoretinal inflammation. Dev. Ophthalmol. 30:187.
Dick, A. D., Y. F. Cheng, J. Liversidge, J. V. Forrester. 1994. Immunomodulation of experimental autoimmune uveoretinitis: a model of tolerance induction with retinal antigens. Eye 8:52.
Dick, A. D., V. Sharma, J. Liversidge. 2001. Single dose intranasal administration of retinal autoantigen generates a rapid accumulation and cell activation in draining lymph node and spleen: implications for tolerance therapy. Br. J Ophthalmol. 85:1001.
Jiang, H. R., N. Taylor, L. Duncan, A. D. Dick, J. V. Forrester. 2001. Total dose and frequency of administration critically affect success of nasal mucosal tolerance induction. Br. J. Ophthalmol. 85:739.
Jiang, H. R., L. Lumsden, J. V. Forrester. 1999. Macrophages and dendritic cells in IRBP-induced experimental autoimmune uveoretinitis in B10RIII mice. Invest. Ophthalmol. Vis. Sci. 40:3177.
Dick, A. D., P. G. McMenamin, H. Korner, B. J. Scallon, J. Ghrayeb, J. V. Forrester, J. D. Sedgwick. 1996. Inhibition of tumor necrosis factor activity minimizes target organ damage in experimental autoimmune uveoretinitis despite quantitatively normal activated T cell traffic to the retina. Eur. J. Immunol. 26:1018.
Liversidge, J., A. Dick, S. Gordon. 2002. Nitric oxide mediates apoptosis through formation of peroxynitrite and fas/fas-ligand interactions in experimental autoimmune uveitis. Am. J.(Neil Taylor, Karen McConn)
A single exposure to inhaled Ag 10 days before immunization leads to long term, Ag-specific tolerance. Respiratory tract myeloid APCs are implicated, but how regulation is invoked, and how tolerance is sustained are unclear. This study examines the in vivo function of the myeloid regulatory molecule CD200 in the process of tolerance induction. Despite earlier onset of experimental autoimmune uveitis in sham-tolerized, CD200-deficient mice, disease incidence and subsequent severity were actually reduced compared with those in wild-type mice. Protection was more effective and long term, lasting at least 28 days. Halting disease progression and tolerance in CD200–/– mice correlated with a marked increase in Th2-associated cytokine production by Ag-challenged splenocytes. Reduced overall disease and enhanced tolerance in the CD200-deficient mice in this model system were unexpected and may be related to altered populations of MHC class IIlow APC in the respiratory tract compared with wild-type mice together with associated activation of STAT6 in draining lymph nodes of tolerized mice. These data indicate that in the absence of default inhibitory CD200 receptor signaling, alternative, powerful regulatory mechanisms are invoked. This may represent either permissive dominant Th2 activation or an altered hierarchy of negative signaling by other myeloid cell-expressed regulatory molecules.
Introduction
Regulation of immune responses to infection or injury requires both initiation signals and termination signals to restore and maintain peripheral tolerance and immunologic homeostasis. IL-10 in particular plays a key role in controlling and terminating inflammatory responses (1, 2, 3), with fine-tuning of the immune response to Ag challenge being accomplished by gene families encoding related receptors with opposing functions (4). APCs of the myeloid lineage, such as macrophages and dendritic cells (DC),3 are central to these regulatory processes (5). The importance of this modulation is demonstrated by the sometimes fatal autoimmune and lymphoproliferative disorders observed in mice with targeted disruption of inhibitory receptors for IL-10 signaling (6).
In myeloid cells, activating and inhibitory receptor pairs form a family of triggering receptors that include activating triggering receptors expressed by myeloid cells as well as signal-regulating proteins SIRP and SIRP that have reciprocal roles in myeloid cell function (7, 8, 9). Now CD200 (OX-2 Ag) (10, 11) has been identified as a negative regulator of myeloid APCs that might be manipulated to provide therapeutic effects in autoimmunity and transplantation (12, 13, 14). CD200 is reported to be expressed by a variety of cells, including neurons and microvascular endothelium, and can be induced by activated T cells (15), whereas strong expression of its structurally related inhibitory receptor (CD200R) is largely restricted to cells of the myeloid lineage, including monocyte macrophages, DC, and microglia (11, 16, 17, 18). Recently, additional members of the CD200R family have been identified in mice (19). In addition to the inhibitory CD200R, which is known to be expressed by CD4+ T cells, especially polarized Th2 cells, there are at least two potentially activating isoforms, designated mCD200RLa and mCD200Lb. The latter do not bind to CD200 and have unknown ligands, but, in common with other receptor pairs, they have potential activating function through DNAX-activating protein-12 (DAP-12) adapter protein binding. These receptors also show differential, low levels of expression on other leukocyte subsets, supporting the idea that rather than merely representing redundancy of function, receptor pairs may have specialized functions in specific cells or tissues.
CD200-deficient mice (CD200–/–) exhibit various myeloid defects. These include elevated numbers of macrophages within tissues normally expressing CD200 and increased DAP-12 expression, particularly in the marginal zone of secondary lymphoid tissues, indicating myeloid cell activation. As a consequence of this phenotype, mice lacking CD200 appear to have increased susceptibility to CD4 T cell-mediated autoimmune diseases (12). In particular, CD200/CD200R regulation of microglial activation has profound effects on neuronal tissues, accelerating the onset of experimental models of autoimmunity affecting the CNS, including experimental autoimmune encephalomyelitis (EAE) (12) and experimental autoimmune uveoretinitis (EAU) (20).
EAU is mediated by retinal Ag-specific CD4+ T cells and can be modulated using various therapeutic approaches targeting Th cell function (21, 22), including induction of Ag-specific tolerance via the nasal mucosa (23, 24, 25). Activated macrophages are also required for full expression of disease (26, 27, 28, 29), but, equally, macrophages are required for resolution of inflammation. For instance, macrophages respond to signals such as IL-4 and IL-10 and actively participate in the anti-inflammatory process (30, 31, 32), demonstrating that, alternatively, activated macrophages are important for healing and tissue remodeling. Such alternatively activated macrophages have recently been described by us in the rat model of EAU (33). Myeloid APC may also have a dual role in nasal tolerance induction in EAE and EAU, where signaling by neuronally expressed CD200 must occur during the inflammatory process. In these models effective protection is associated with an initial IFN--driven priming event in cervical lymph nodes, followed by T cell apoptosis and down-regulation of the capacity of Ag-specific T cells to proliferate in response to restimulation (34, 35). IL-10 has been shown to have a pivotal role in this process (36, 37). How are these contradictory, but complementary, activities of macrophages and DC affected by the CD200/CD200R interaction?
In this study we use a model of tolerance induction by respiratory exposure to Ag in CD200–/– mice to examine the role of CD200/CD200R interactions in vivo on tolerance induction and maintenance. We chose the moderately susceptible C57BL/6 mouse EAU model (38) because uveitogenic T cells alone are insufficient to cause target organ damage. Monocyte macrophages are also necessary and prominent in the earliest inflammatory infiltrates in the retina (26). In addition, monocyte expression of type 2 NO synthase (NOS2) is required for full expression of disease (27, 39, 41). Respiratory tract DC (RTDC) and alveolar macrophages are known to be involved in tolerance induced by respiratory exposure to Ag (37, 41). T cell activation and proliferation in the draining cervical lymph are followed by systemic generation of regulatory cells in the spleen (24, 35, 42), providing a model to study myeloid APC function and T cell regulation in vivo. The data presented show that despite the accelerated disease onset that is expected in the absence of inhibitory CD200R signaling, overall disease incidence and severity were significantly reduced over time in CD200–/– mice. This result was unpredicted, but correlated with elevated numbers of regulatory T cells (Treg) and the presence of high IL-10-secreting splenic myeloid cells later in the disease process. Induction of tolerance to retinal Ag was also enhanced in CD200-deficient mice, which demonstrated an altered phenotype of APC in the respiratory tract and an enhanced Th2 switch compared with wild type (WT).
Materials and Methods
Animals and induction of EAU
CD200-deficient mice (CD200–/–) on the C57BL/6 background were generated at DNAX (Palo Alto, CA), and an isolator-reared, specific pathogen-free breeding colony was established within the biological services unit of Aberdeen University. Specific pathogen-free C57BL/6 wild type(CD200+/+) mice were purchased from Harlan Olac. Groups of three to eight mice were used as detailed in the text or figure legends for each experiment. For immunization, groups of sex and age matched mice were used at 6–8 wk of age. Mice were immunized by s.c. injection of 500 μg of peptide 1–20 of interphotoreceptor retinoid binding protein (Sigma-Genosys) in CFA (2.5 mg/ml Mycobacterium tuberculosis) plus 1 μg/ml Bordetella pertussis toxin i.p. as an additional adjuvant (38). At specified times animals were killed by CO2 asphyxiation, and eyes were enucleated for resin histology (H&E staining for histological scoring system) or for immunocytochemistry. Lymphoid tissue was also sampled to determine peptide-specific proliferative and cytokine responses. The severity of disease was assessed using a modified version of our histological grading system for rat EAU. At least three sections from each eye were scored in a masked fashion using a semiquantitative scoring system that combines the extent of the inflammatory infiltrate with tissue damage in the posterior chamber (23). All results are expressed as the mean ± 1 SEM. Comparison of histological assessment of disease in CD200–/– and CD200+/+ were analyzed using the Mann-Whitney U test (Instat software; GraphPad), and p 0.05 was considered significant. Disease incidence was analyzed using Fisher’s exact test, and p 0.05 was considered significant.
Induction of tolerance
Mucosal tolerance was induced by intranasal administration of 50 μg of peptide in 5 μl of PBS or 5 μl of PBS alone using an Oxford pipette. This regimen, administered 10 days before immunization has previously been shown to be effective in modulating EAU (25). In some experiments mice were killed at intervals over the following 48 h to examine the effects of intranasal peptide on cells in draining cervical lymph nodes and spleens or in control submandibular and mesenteric lymph nodes. In other experiments animals were immunized 10 days later with peptide or PBS in CFA with B. pertussis toxin.
Proliferation and cytokine profile assays
Lymphocyte proliferative responses to recall Ag were measured using the BrdU Flow kit (BD Biosciences) according to the manufacturer’s instructions. This test has the advantage of more detailed analysis of viable cell cycle kinetics by combining BrdU labeling of cells in S phase with measurement of total DNA content of the population as a whole. Single-cell suspensions were obtained from individual spleens by pressing the tissue through a 250-μm pore size metal sieve, and mononuclear cells were purified by Percoll density gradient centrifugation. RBC were removed by hypotonic lysis. Cultures were set up at a density of 1 x 106 cells/ml with 10 μg/ml peptide 1–20 for 96 h. This time period was established as optimum by compiling data from 48- to 120-h incubations in preliminary experiments. Each well was then pulsed with 10 μl of 1 mM BrdU for 45 min and harvested, and cells were permeabilized with Cytofix/Cytoperm (BD Biosciences) and frozen at –80°C in FCS with 10% DMSO (Sigma-Genosys) before staining and analysis. An FITC-labeled anti-BrdU Ab was used to identify the extent of BrdU incorporation, and the DNA stain 7-aminoactinomycin D was used to quantify the total DNA content. Measurements were made by two-color flow cytometry using a FACSCalibur or a FACS LSR (both from BD Biosciences, Mountain View, CA) with flow rate adjusted to give the optimal coefficient of variation of the DNA peaks. Data were obtained from at least three individual animals at each time point. Results were expressed as the mean ± SEM. For cytokine profile analysis of responding cultures, parallel 1-ml cultures were set up with 4 x 106 cells and 10 μg/ml peptide. After 72 h, supernatants were harvested, clarified by centrifugation, aliquoted, and frozen at –30°C until assay. Negative control cultures contained PBS in place of peptide, and 2.5 μg/ml Con A was added to positive control cultures to demonstrate optimum growth conditions.
IL-10 and IL-12 were measured using optELISA kits (BD Pharmingen) exactly according to the manufacturer’s instructions. The mouse cytokine bead array (CBA kit; BD Pharmingen) was used to measure other cytokines. Results are expressed as the mean ± SEM of supernatants from at least six animals in each group, and differences were considered significant at a value of p < 0.05 using unpaired Student’s t test.
Isolation and phenotyping of respiratory tract cells
Mice were anesthetized with 0.1 ml of Sagatol (Aventis Pharma) i.p. and were perfused with Ca2+/Mg2+-free PBS containing 10 U/ml heparin and 5 mM EDTA to remove blood from the respiratory tract vessels. The lungs and trachea were removed and disrupted mechanically by chopping and passing through a 250-μm pore size mesh. The resulting suspension was collected and treated with 0.7 mg/ml collagenase and 30 U/ml DNase in RPMI 1640 for 1 h at 37°C with agitation. The cells were then washed and passed through a 100-μm pore size mesh before centrifugation over Percoll 1.075. Cells were then washed, and CD45+ cells were analyzed by flow cytometry (FACSCalibur) for myeloid phenotype (CD11b, CD11c, F4/80, and CD204) and for activation markers (MHC class II, CD40, CD80, and CD86) as described below.
To determine myeloid cell distribution within the tissue, cryostat sections of Evan’s Blue-perfused lung tissue were stained with FITC-conjugated CD11b mAb M1/70 or IgG control mAb R35-95 (BD Pharmingen), and digital images were captured with an LSM confocal microscope (Zeiss) using the same laser settings for each image in the experiment. The relative abundance of CD11b fluorescence signal in each image was analyzed by measuring pixel values within defined regions using Laserpix software (Bio-Rad). Positively staining cells were counted using the Count-Size command and the Histogram command used to measure the average intensity value (mean) of positively stained cells.
Generation of bone marrow-derived DC (BMDC)
BMDC were prepared by a modification of the procedure described by Inaba et al. (43) as previously described (44). Briefly, 7.5 x 105 precursors were set up in 1-ml cultures supplemented with GM-CSF. The cultures were fed, and nonadherent cells were removed every 2 days by swirling and aspirating 75% of medium and adding fresh growth medium with GM-CSF. Developing DC remained attached to the bed of firmly adherent macrophages. Six days after initiation of the cultures, the loosely adherent clusters of DC were harvested, and contaminating granulocytes were depleted using anti-mouse Gr-1 mAb (RB6-8C5; BD Pharmingen) and Dynabeads (Dynal Biotech). Purified BMDC were further seeded at 1 x 106/ml/well in 24-well plates supplemented with GM-CSF. Then 1 μg of LPS was added to the cultures at various time points (0–24 h). Supernatant was then collected from each well 22 h after administration of LPS for cytokine measurement. Medium alone DCs were used as control.
Immunocytochemistry
Eyes from immunized mice or lymphoid tissue were dissected, snap-frozen in OCT (Miles), and 7-μm serial sections were cut, air-dried, and fixed in 100% cold acetone for immunocytochemistry using the alkaline phosphatase anti-alkaline phosphatase technique. After rehydration in TBS, sections were blocked with TBS/1% normal rabbit serum and then avidin D block solution (Vector Laboratories) for staining with polyclonal rabbit anti-STAT 4 (C20; Santa Cruz Biotechnology) STAT 6 (M20; Santa Cruz Biotechnology) or with phosphorylated STAT6 (Tyr641; New England Biolabs) using irrelevant polyclonal Ab and blocking peptides as negative controls. Other sections were stained using rat mAbs to murine CD3 and myeloid cell markers F4/80 Ag (CI:A3-1), MOMA-1, and MOMA-2 (Serotec). Activation markers included NOS2 (clone 6; Transduction Laboratories), rat anti-mouse CD86 (GL-1; BD Pharmingen), and MHC class II (I-Ab; P7.7; Serotec). Positive staining was detected by mouse absorbed biotinylated anti-rabbit Ig-alkaline phosphatase or biotinylated anti-rat Ig-alkaline phosphatase conjugate, followed by streptavidin:avidin-biotin peroxidase-alkaline phosphatase complex and Fast Red substrate (DakoCytomation) lightly counterstained with hematoxylin. Sections for image analysis were cut and stained in batches to ensure uniform labeling conditions for each Ab. Sections were then analyzed using the Aphelion Active X image analysis program from ADCIS. The program was adapted using Visual basic to allow analysis of immunostaining in user-defined regions of the image. An average value (percentage of tissue stained per x20 field) for each section was obtained from four to six fields. Results are expressed as the mean ± SEM of tissue from a minimum of three animals per treatment group. Differences between groups were analyzed using unpaired Student’s t test, and p < 0.05 was considered significant.
Flow cytometry
A FACSCalibur or FACS LSR (BD Biosciences) was used for data acquisition, and CellQuest software (BD Biosciences) was used for data analysis. APCs and lymphocytes isolated from lymph nodes and spleens were evaluated by multicolor immunofluorescent staining with mAbs to the following cell surface markers: CD11b (M1/70), CD11c (HL3), CD4 (RM4-4) CD45RB (16A), CD45R/B220 (RA3-6B2), CD40 (3/23), CD86 (GL-1), CD152 (BN13), MHC class II (I-Ab; AF6-120.1), CD25 (PC61), CD38 (92), CD8a (53-6B2), CD62L (MEL-14), and CD3 (145-2C11) were all from BD Pharmingen. For intracellular detection of IL-10, quadruple staining was performed. Cells were treated with Cytofix-Cytoperm and stained with PE-conjugated anti-IL-10 (JES5-16E3) following the manufacturer’s instructions (BD Pharmingen). Anti-CD204 (2F8), F4/80 (CI:A3-1), metallophillic macrophages (MOMA-1), and CD80 (RMMP-1) were obtained from Serotec. These were conjugated to FITC, PE, allophycocyanin, PerCP, PerCP-Cy5.5, or biotin as required. Biotin-labeled Abs were detected by addition of streptavidin-allophycocyanin (1/400; BD Pharmingen). Negative isotype controls and single positive controls were performed to allow accurate breakthrough compensation. Gates and instrument settings were set according to forward and side scatter characteristics, and populations were gated to exclude dead or clumped cells. A total of 10,000 events were collected. Fluorescence analysis was performed after additional backgating to exclude nonspecific or background staining. Data were collected from at least four individual animals at each time point in each experiment; the cells were prepared and stained for analysis on the same day. Histogram results are expressed as the mean ± SEM. Results shown are representative of at least four separate experiments.
Immunoprecipitation of STAT6
Activation of STAT6 expression was also demonstrated by immunoblotting of immunoprecipitates prepared from cytosolic and nuclear extracts from lymphoid tissues as previously described (43). Briefly, cytosolic extracts were prepared from 10 (7, 8) cells in 20 μl of lysis buffer A. Nuclei were removed by centrifugation, and supernatants were clarified and collected as the cytoplasmic fractions. Nuclei were then incubated in 20 μl of lysis buffer B (containing 0.42 M NaCl) and centrifuged, and supernatants were collected as nucleic fractions. STAT6-associated proteins were immunoprecipitated by anti-STAT6 Ab (M20; Santa Cruz Biotechnology), collected on protein A-Sepharose beads (Amersham Biosciences, Arlington Heights, IL), and resolved by SDS-PAGE. Protein was transferred onto a Hybond polyvinylidene difluoride membrane (Amersham Biosciences) by electroblotting. Initial probing was performed for phosphotyrosine (mAb 4G10; Upstate Biotechnology) detected by ECL. Membranes were then stripped and reprobed with STAT6 Ab to confirm the specificity of the signal.
Results
Tolerance induction in CD200–/– mice is robust, and disease incidence and severity are significantly reduced compared with those in WT mice
Mice were given either peptide 1–20 (tolerized group) or control PBS (sham-tolerized group) intranasally 10 days before immunization. Fig. 1 shows the pattern of disease progression in tolerized and control WT and CD200–/– mice over time. In WT mice, disease progressed rapidly, with all mice showing moderate to severe retinal inflammation and photoreceptor damage by 21 days postimmunization, persisting to day 28. In accordance with our previous observations (20), disease onset was accelerated in sham-tolerized CD200–/– mice, with expression of NOS2 in cells of the ciliary body, retinal vascular endothelium, and inner plexiform layer (IPL) of the retina being the earliest signs of disease on day 10 (Fig. 2, a and b) and significantly elevated disease scores by day 16 (p < 0.03). However, as in our previous study, no overall increase in tissue damage compared with WT mice was observed, and no additional significant disease progression occurred in sham-tolerized CD200–/– mice. By day 28 postimmunization, disease scores were, in fact, significantly lower than those in sham-tolerized WT mice (p < 0.05), with 17% mice still showing no signs of disease. No significant increase in histological scores of either sham-tolerized or tolerized CD200–/– mice from days 16–28 postimmunization was found. Significant reductions in the overall histological scores in both tolerized groups compared with their sham-tolerized controls were also observed on day 16 (WT, p < 0.01; CD200–/–, p < 0.03; Fig. 2, c–f). No significant difference in overall histological scores between tolerized WT and tolerized CD200–/– mice were found, but on day16, disease incidence in tolerized CD200–/– mice was significantly higher than that in WT tolerized mice (p < 0.05), reflecting an earlier onset of disease in these mice. By day 28, however, the situation was reversed, and disease incidence in the CD200–/– tolerized group was very significantly reduced compared with that in tolerized WT mice (p < 0.006) as disease incidence continued to increase in the WT groups.
Despite severe disease in WT mice, protection was maintained in the tolerized groups at all time points (p < 0.01). In the tolerized CD200–/– group, protection was also effective. Disease scores were significantly reduced on day 16 (p < 0.03) and day 28 (p<0.01), with up to 50% of the animals still showing no signs of disease.
In vitro proliferation of splenocytes does not correlate closely with in vivo inflammation during early disease, but is reduced in tolerized mice in late disease
In a previous study we showed that T cell proliferative responses by splenocytes to immunizing peptide 1–20 were equivalent in both WT and CD200–/– mice (20). We confirmed this result using a sensitive flow cytometric assay that combines BrdU labeling of cells in S phase with measurement of total DNA content of the population as a whole (Fig. 3a). It was also found that intranasal exposure to Ag had no inhibitory effect on peak proliferation of splenocytes on day 16 or day 21 despite evidence of a significant reduction in disease in tolerized animals at these time points (Fig. 1); however, proliferation by splenocytes from CD200–/– tolerized mice was significantly reduced compared with that in tolerized WT mice (p < 0.02). By day 28 postimmunization, the numbers of cells in S phase were reduced in both tolerized groups, and this was highly significant in the CD200–/– tolerized group compared with the sham-tolerized group (p < 0.01). Analysis of DNA content of cells suggested that this reduction of cells in S phase was due to reduced turnover of proliferating cells rather than a block in cell cycle progression or an increase in apoptotic or necrotic cells within the cultures (Fig. 3, b–d).
Protection from disease in CD200–/– mice correlates with a switch to IL-4 and IL-5, followed by IL-10 production by restimulated splenocytes
Splenocyte cultures set up in parallel with the proliferation assays were analyzed for cytokine production in response to restimulation with peptide (Fig. 4). Significantly higher levels of IL-4 were found in the tolerized CD200–/– groups on days 16 and 21 (p < 0.02, and p < 0.04 compared with sham-tolerized CD200–/– mice; p<0.01 compared with tolerized WT mice at the same time point). IL-5 was also elevated in CD200–/– tolerized mice on day 21 (p < 0.04 compared with sham-tolerized CD200–/–; p<0.02 compared with tolerized WT). This pattern was replaced by enhanced levels of IL-10 in the CD200–/– tolerized mice by day 28 (p < 0.05 compared with sham-tolerized; p < 0.02 compared with WT tolerized). IL-2 and TNF were elevated in sham-tolerized WT mice on days 16 and 21 and were significantly reduced in tolerized WT on day 16 (p < 0.05). Levels of IL-12 were low in all the cultures. Large quantities of IFN- (5–15 ng/ml) were also generated in all the cultures at all time points, but no significant differences between groups were observed (data not shown). Changes in IgG2a/IgG1 ratios can be indicative of changes in the Th1/Th2 balance of an immune response. However, due to the short half-life of mouse Ig isotypes (6–8 days for both IgG1 and IgG2a), these were inconclusive given the time span of these experiments (2–4 wk; data not shown).
STAT-6 is up-regulated in draining lymph nodes and spleen of CD200 deficient mice
The cytokine secretion assays indicated that tolerance in the CD200–/– mice appeared to be accompanied by an earlier and more pronounced Th2 switch in response to restimulation by Ag. To test whether this represented a different response to the tolerization regimen in CD200–/– mice we examined the effect of intranasal administration of Ag on cells of the draining lymph nodes and spleen. Expression of phenotypic myeloid cell markers, MHC class II, APC costimulatory molecules CD80 and CD86 as well as STAT 4 and STAT 6 were measured by image analysis of serial sections of immunostained tissue. Cervical lymph nodes that drain the nasal mucosa, lymphoid areas (white pulp) and macrophage areas (red pulp) of the spleen were examined up to 48 h post intranasal application of Ag. The submandibular lymph nodes were analyzed in parallel to serve as control tissue. Fig. 5, a–d, and Table I show that STAT6, which controls the Th2 differentiation pathway (45, 46), was markedly up-regulated in the spleen and cervical lymph nodes of tolerized CD200–/– mice 24 h after treatment. The effect was transient in the cervical nodes, levels returning to control levels by 48 h, but sustained in both lymphoid and macrophage areas of the spleen. Expression of STAT6 was also increased in the spleens, but not lymph nodes of WT mice, where levels actually declined. Immunostaining of sections with an Ab to phosphorylated STAT6 showed activation of the STAT6 pathway in the cervical lymph nodes (Fig. 5, e and f), and this was confirmed by Western blotting of cytosolic and nuclear extracts, which showed translocation of STAT6 from cytosol to nucleus in CD200–/–, but not WT spleens (Fig. 5, g and h). STAT4, which is activated after IL-12 signaling to drive Th1 responses (47, 48), was also transiently increased in draining lymph nodes and spleen of CD200–/– mice 24 h post-treatment (Table I). This was consistent with the initial Th1 priming previously observed in our rat tolerance model (24) and was accompanied by increased CD86 expression in cervical lymph nodes. However, in contrast, no increase in STAT4 expression was observed in the WT mice, levels actually falling during the sampling period. It was noted that overall expression STAT4 was low compared with STAT6 in this strain of mouse and may be consistent with the low levels of IL-12 p70 generated by Ag-restimulated splenocytes (Fig. 4).
With the exception of a significant increase in STAT6, and a transient increase in STAT4 expression in CD200–/– mice, no significant differences between control and test tissues were observed in the control submandibular tissue. Compared with percentages for cervical tissue, increases were comparatively small. This might reflect changes induced locally through ingestion of very small quantities of Ag during intranasal administration, the effects of disseminated Ag (24), or the systemic effects of changes in the spleen in these animals. No substantial or consistent changes in the percentage of F4/80- or MOMA-2-positive cells after treatment or between groups was observed to account for increased or decreased STAT expression in the tissues. Similarly, although some fluctuation in MHC class II expression was apparent, no significant pattern was found (data not shown).
DC from CD200–/– mice do not have a defect in IL-12 production
Low production of IL-12 in the Ag-restimulated spleen cells (Fig. 4) together with the low levels of STAT4 Ag detected in the lymph nodes and spleens (Table I) may represent a fundamental deficiency in CD200–/– mice that reduced their ability to mount a normal Th1 response, allowing a Th2 response to emerge by default. To determine the IL-12 vs IL-10 secretion potential of APC from CD200–/– mice, we used the model developed by Jiang et al. (44) that allows BMDC to be matured to secrete either IL-10 or IL-12. BMDC from WT C57BL/6 and CD200–/– C57BL/6 mice secreted both IL-10 and IL12 at high levels (>700 pg/ml) and with similar kinetics, indicating that there was no fundamental defect in the mice to account for the low IL-12 responses observed in our model (data not shown).
Altered phenotype of respiratory tract APC in CD200–/– compared with WT mice
RTDC have been shown to preferentially stimulate Th2 responses (41) and are implicated in the induction of tolerance to nasally administered Ags (37). Because altered populations of myeloid cells are a major feature of CD200–/– mice (12), we compared the phenotype of CD45+ cells from the respiratory tracts of WT and CD200–/– mice isolated and purified using collagenase/DNase digestion and density gradient centrifugation. In WT mice, the major population isolated were CD11c+ DC (<80%), with few (10–15%) CD11b+ cells. In contrast, in CD200–/– mice, CD11b+ cells formed a far larger proportion (<40%) of the purified cell population, with fewer CD11c+ DC (35–40%; Fig. 6a). The cells from both WT and CD200–/– exhibited low levels of activation markers, as expected for respiratory tract APC, but it was also apparent that both CD11b+ and CD11c+ cells from CD200–/– respiratory tract expressed lower levels of MHC class II Ag than WT (Fig. 6b). Confocal analysis of respiratory tract tissue showed that this increased proportion of CD11b+ cells present in CD200–/– mice was, in fact, due to increased numbers of CD11b+ cells within the tissue (p < 0.001; Fig. 6c). Increased detection was not due to higher levels of CD11b expression by positively stained cells, because there was no difference in average pixel intensity between the groups (Fig. 6d).
CD25+ Treg and IL-10high APC are present in the spleens of tolerized CD200–/– mice
Tolerance induced by respiratory tract exposure to Ag has been shown to be dependent on IL-10 and may be mediated by Treg (3, 36, 37). Regulatory cells are also known to be induced by an increase in Ag-specific Th2 activity together with decreased CNS infiltration in a C57BL/6 mouse model of EAE (49). We therefore used flow cytometry to examine the spleens of tolerized, immunized mice for the presence of Treg cells and to identify the phenotype of IL-10-secreting cells. Permeabilization of cells to detect intracellular Ag can reduce or interfere with the fluorescence of surface markers such as CD25, so we used triple fluorescence on unpermeabilized cells to establish the numbers of CD3+CD4+CD25+ present in the spleens of control and treated mice (Fig. 7a). In unimmunized control mice, 5–8% of CD3+ T cells from the spleen displayed the CD3+CD4+CD25+ (contact-dependent) regulatory cell phenotype (data not shown). This was typically increased to 10–15% in immunized mice. A trend toward higher numbers of CD4+CD25+ cells in the CD3+ gated population in CD200–/– mice was noticed, and the number was significantly elevated in tolerized CD200–/– mice compared with tolerized WT mice at all time points (day 16, p<0.01; days 21 and 28, p < 0.05). All CD3+CD4+CD25+ cells were also CTLA-4+ (data not shown).
IL-10 may be produced by CD25+ Treg cells (50), but Fig. 7b shows that CD3+ T cells did not express levels of intracellular IL-10 above background. IL-10-producing cells were found within the spleen myeloid cell population. Two main populations could be identified from day 21 onward. These were predominantly IL-10lowCD11b+ or small numbers of IL-10highCD11clow/– (Fig. 7b). IL-10low-secreting CD11b+ were also observed at later time points (day 28) in all mice (data not shown), but no significant differences between the groups were found.
It should be noted that the levels of IL-10 detected in these cells should reflect the in vivo situation, because cells were isolated and analyzed directly ex vivo. The cells were not given additional activation stimuli, and Golgi function was not blocked to artificially increase intracellular cytokine levels.
Discussion
In this study we have examined the role of CD200, an inhibitory ligand for myeloid cells, in a model of tolerance induction to a retinal autoantigen. Unexpectedly, we found that although disease onset in CD200–/– mice was accelerated, in agreement with previous studies using these mice (12, 20), over the longer term, overall disease incidence and severity compared with WT were significantly reduced. Furthermore, tolerance induction in CD200–/– was efficient, with up to 50% of eyes still protected from disease 28 days postimmunization (Fig. 1). Given our current understanding of the CD200/CD200R axis (15, 17, 20) and the results of other studies showing enhanced immunosuppression in autoimmune or transplant models where CD200 expression is elevated or signaling is potentiated by infusion of CD200 immunoadhesions (14, 51), these findings are counterintuitive.
Protection in tolerized CD200–/– mice appeared to correlate with increased Th2-associated cytokines and IL-10 in response to recall Ag, and overall reduced disease in sham-tolerized CD200–/– mice may be associated with the emergence of IL-10-secreting cells in the spleen later in the disease process. Based on the apparent redundancy in regulatory mechanisms controlling potentially tissue damaging inflammation, one possibility is that in a normal individual, the CD200/CD200R axis alone is not critical for the maintenance of immunological homeostasis (although it may contribute to it), but during inflammation, in the absence of default and necessary CD200/CD200R signaling, other regulatory mechanisms are invoked to convert the classically activated myeloid phenotype to an alternatively activated, healing, and immunomodulatory phenotype.
In support of this hypothesis is the emerging principle of dominant inhibitory signaling mediated by self-receptors that governs the maintenance of self-tolerance. Triggering receptors expressed by tissue-trafficking APC such as macrophages and DC or other leukocytes are key molecules in this process (4, 52, 53). In addition to these myeloid cell receptors, active self-regulation to control autoreactive T cell expansion through naturally occurring CD4+CD25+ T cell populations is now well accepted (54, 55). These, together with induced regulatory mechanisms driving pathogen-specific Th1/Th2 responses through differential cytokine expression (reviewed by in Ref. 56), induction of type 1 Treg cells (reviewed in Ref. 57), and alternative macrophage activation and scavenger receptor expression (29, 58) are invoked to avoid excessive immune pathology. How these diverse regulatory mechanisms interact in complex in vivo models, and the rules governing functional redundancy and hierarchy of responses have yet to be fully understood.
Recently, it has been shown that the CD200R gene family also has receptor isoforms with potential inhibitory and activating functions (19). In addition to the inhibitory isoform that binds CD200 and regulates myeloid activity in the mouse, at least two activating isoforms (CD200RLa and CD200RLb) were identified. These isoforms probably have alternative ligands and are able to pair with the adaptor protein DAP-12, supporting the idea that the CD200R family is also involved in fine-tuning of myeloid cell responses (15). Pertinent to the present study, Wright et al. (19) also showed CD200R expression by CD4+T cells and, at least at the mRNA level, preferentially by polarized Th2 cells. Thus, in Ag-specific in vivo models using reagents that modulate CD200R signaling, effects may be expected at the level of both myeloid APC and Th2 cell effector function. Enhancing inhibitory signaling received by myeloid cells may provide an overriding "off" signal to the innate and adaptive immune response and prolong graft survival or reduce autoimmune inflammation. Conversely, reagents that temporarily blockade theCD200/CD200R signaling pathway remove the dominant inhibitory effect of the inhibitory isoform of the receptor, thus promoting cellular activation and expression of NOS2 by myeloid cells. Whether Th2 cell function is also controlled through CD200R expression is currently unknown, although the present data suggest that it may be.
What happens in the long term absence of default CD200R signaling, as occurs in the CD200–/– mouse? The apparently mild phenotype suggests that the CD200/CD200R axis has a specialized function that is not generally required in the normal individual. The classically activated myeloid cell phenotype, including constitutive expression of NOS2 by retinal microglia, has the short term effect of accelerating EAU disease onset, but this was not associated with increased tissue damage (20), and we now show that accelerated disease is followed by a halt in disease progress and a down-regulation of inflammation (Fig. 1). The mechanisms are not fully defined, but significantly increased numbers of CD3+CD4+CD25+ were found in tolerized CD200–/– mice from day 16, with IL-10-secreting CD11b+ and CD11clow/– cells present in the spleens of both sham-tolerized and tolerized CD200–/– mice from day 21 (Fig. 7). The origin and lineage of these cells are not yet understood, but because they appear 3–4 wk after disease onset, prolonged stimulation, sequential differentiation of cell subsets, and/or several rounds of cell division must be involved. Several subsets of Treg have been described, and although some controversy surrounds the use of particular phenotypic or cytokine secretion profiles to define them, it is clear that more than one subset of Treg exists and that even small numbers can have profound immunosuppressive effects (50, 59, 60, 61, 62), either directly through inhibition of IL-2 transcription in the responder cell or indirectly through IL-10. These have been variously described as natural (typically cytokine-independent CD4+CD25+), adaptive (typically cytokine-dependent type 1 Treg) (61), or Ag-induced (60), and there are data suggesting that several rounds of stimulation are required for the induction of adaptive Treg during an inflammatory response (2, 63). Although the induction of IL-10 and the suppression of IL-2 in all groups on day 28 are certainly consistent with the induction of functional Treg cells during the disease process (54, 55), and there are data linking nasal administration of Ag with induction of Treg1 (36, 37), we found no direct evidence for IL-10-secreting CD4+ Treg in spleens of protected mice.
What effect might lack of signaling through CD200R expressed by myeloid APC and Th2 cells have on mucosal tolerance induction? Significantly elevated levels of Th2 cytokines and IL-10 were prominent in the cytokine responses of tolerized CD200-deficient mice. Pulmonary DC play a critical role in the selective immune response to inhaled Ag, inducing T cell hyporesponsiveness to innocuous Ags or preferential activation and expansion of Th2-biased responses (37, 41, 42, 64). This has been attributed to the mucosal microenvironment and immature phenotype of these cells. IL-10 has also been shown to have a critical role both in the induction of nasal tolerance and in limiting inflammation later in disease (3, 34). IL-10 would then be able to augment tolerance in an Ag-specific manner, because a single exposure to IL-10 can convert DC to a tolerogenic phenotype. Examination of the draining lymph nodes and spleen of tolerized mice 24–48 h after intranasal administration of Ag revealed a pronounced induction of STAT6 expression in CD200–/– mice. STAT6 is required for IL-4 production (46), suggesting that an enhanced Th2 switch in response to Ag presentation could be one mechanism underlying more rapid resolution of inflammation and induction of more effective tolerance to immunizing Ag in CD200–/– mice. This could be due to the altered phenotype and reduced proportion of CD11c+ to CD11b+ RTDC present in CD200–/– mice presenting the nasally administered Ag in a tolerogenic fashion. However, given the endogenous state of tonic activation expected in these cells from the CD200–/– background, an alternative mechanism, involving active signaling, could be involved. Although Th1 responses seem to be driven by pathogen-activated innate immune responses, the signals that initiate Th2 responses are still uncertain. It has been proposed that Th2 responses emerge by default in the absence of the activating innate signals, IFN- and IL-12, that would drive Th1 responses, and that alternative signals from DC might trigger the initial production of IL-4 (56). Our data clearly show that CD200–/– BMDC were able to secrete equally high levels of both IL-12 and IL-10 in response to an innate maturation signal, suggesting that the cytokine bias observed in our model was induced. Because the inhibitory form of CD200R is highly expressed on polarized Th2 cells and not on Th1 cells, the inference is that the CD200/CD200R axis can regulate Th2-mediated responses.
In conclusion, we have shown that in the absence of default inhibitory CD200/CD200R signaling, accelerated EAU onset is observed in CD200–/– mice immunized with retinal autoantigenic peptide in CFA, consistent with an increased level of classically activated microglia and myeloid APCs in these mice. However, this response is effectively down-regulated, with significantly reduced overall disease evident at mid and late disease stages. This is consistent with induction of alternative IL-10-dependent regulatory pathways, possibly induced through dominant activating CD200RLa signaling expected in Th2 cells that are not receiving an inhibitory CD200/CD200R signal. The enhanced tolerance to EAU induction observed in CD200–/– mice correlated directly with up-regulated Th2-associated STAT6 activation in cervical lymph nodes of nasally tolerized mice and with elevated Th2-associated cytokine levels in response to recall Ag, again implicating active enhancement of Th2 cell activation. Additional experiments are required to confirm that dominant CD200RLa signaling preferentially activates Th2 cells in CD200–/– mice, leading to predominant IL-4 secretion and alternative activation of myeloid APC as a regulatory mechanism to control excessive inflammation.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by project grants from the Wellcome Trust, Tenovus Scotland, Grampian Universities Hospitals National Health Service Trust Endowments, and the National Eye Center, Bristol. DNAX is supported by Schering Plough Corp. (Kenilworth, NJ).
2 Address correspondence and reprint requests to Dr. Janet Liversidge, College of Medicine and Medical Sciences, Institute of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen, U.K. AB25 2ZD. E-mail address: j.liversidge@abdn.ac.uk
3 Abbreviations used in this paper: DC, dendritic cell; BMDC, bone marrow-derived DC; DAP-12, DNAX-activating protein-12; EAE, experimental autoimmune encephalomyelitis; EAU, experimental autoimmune uveoretinitis; NOS2, type 2 NO synthase; RTDC, respiratory tract DC; Treg, regulatory T cell; WT, wild type.
Received for publication June 29, 2004. Accepted for publication October 13, 2004.
References
Moore, K. W., M. R. de Waal, R. L. Coffman, A. O’Garra. 2001. Interleukin-10 and the interleukin-10 receptor. Annu. Rev. Immunol. 19:683.
McGuirk, P., C. McCann, K. H. Mills. 2002. Pathogen-specific T regulatory 1 cells induced in the respiratory tract by a bacterial molecule that stimulates interleukin 10 production by dendritic cells: a novel strategy for evasion of protective T helper type 1 responses by Bordetella pertussis. J. Exp. Med. 195:221.
Kaya, Z., K. M. Dohmen, Y. Wang, J. Schlichting, M. Afanasyeva, F. Leuschner, N. R. Rose. 2002. Cutting edge: a critical role for IL-10 in induction of nasal tolerance in experimental autoimmune myocarditis. J. Immunol. 168:1552.
Lanier, L. L.. 2001. Face off-the interplay between activating and inhibitory immune receptors. Curr. Opin. Immunol. 13:326.
Goerdt, S., C. E. Orfanos. 1999. Other functions, other genes: alternative activation of antigen-presenting cells. Immunity 10:137.
Kuhn, R., J. Lohler, D. Rennick, K. Rajewsky, W. Muller. 1993. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75:263.
Cant, C. A., A. Ullrich. 2001. Signal regulation by family conspiracy. Cell. Mol. Life Sci. 58:117.
Dietrich, J., M. Cella, M. Seiffert, H. J. Buhring, M. Colonna. 2000. Cutting edge: signal-regulatory protein 1 is a DAP12-associated activating receptor expressed in myeloid cells. J. Immunol. 164:9.
Colonna, M.. 2003. TREMs in the immune system and beyond. Nat. Rev. Immunol. 3:445.
Barclay, A. N., H. A. Ward. 1982. Purification and chemical characterisation of membrane glycoproteins from rat thymocytes and brain, recognised by monoclonal antibody MRC OX 2. Eur.J. Biochem. 129:447.
Wright, G. J., M. Jones, M. J. Puklavec, M. H. Brown, A. N. Barclay. 2001. The unusual distribution of the neuronal/lymphoid cell surface CD200 (OX2) glycoprotein is conserved in humans. Immunology 102:173.
Hoek, R. M., S. R. Ruuls, C. A. Murphy, G. J. Wright, R. Goddard, S. M. Zurawski, B. Blom, M. E. Homola, W. J. Streit, M. H. Brown, et al 2000. Down-regulation of the macrophage lineage through interaction with OX2 (CD200). Science 290:1768.
Gorczynski, R. M., K. Yu, D. Clark. 2000. Receptor engagement on cells expressing a ligand for the tolerance-inducing molecule OX2 induces an immunoregulatory population that inhibits alloreactivity in vitro and in vivo. J. Immunol. 165:4854.
Gorczynski, R. M., Z. Chen, Y. Kai, J. Lei. 2000. Evidence for persistent expression of OX2 as a necessary component of prolonged renal allograft survival following portal vein immunization. Clin. Immunol. 97:69.
Barclay, A. N., G. J. Wright, G. Brooke, M. H. Brown. 2002. CD200 and membrane protein interactions in the control of myeloid cells. Trends Immunol. 23:285.
Preston, S., G. J. Wright, K. Starr, A. N. Barclay, M. H. Brown. 1997. The leukocyte/neuron cell surface antigen OX2 binds to a ligand on macrophages. Eur. J. Immunol. 27:1911.
Wright, G. J., M. J. Puklavec, A. C. Willis, R. M. Hoek, J. D. Sedgwick, M. H. Brown, A. N. Barclay. 2000. Lymphoid/neuronal cell surface OX2 glycoprotein recognizes a novel receptor on macrophages implicated in the control of their function. Immunity 13:233.
Dick, A. D., C. Broderick, J. V. Forrester, G. J. Wright. 2001. Distribution of OX2 antigen and OX2 receptor within retina. Invest Ophthalmol. Vis. Sci. 42:170.
Wright, G. J., H. Cherwinski, M. Foster-Cuevas, G. Brooke, M. J. Puklavec, M. Bigler, Y. Song, M. Jenmalm, D. Gorman, T. McClanahan, et al 2003. Characterization of the CD200 receptor family in mice and humans and their interactions with CD200. J. Immunol. 171:3034.
Broderick, C., R. M. Hoek, J. V. Forrester, J. Liversidge, J. D. Sedgwick, A. D. Dick. 2002. Constitutive retinal CD200 expression regulates resident microglia and activation state of inflammatory cells during experimental autoimmune uveoretinitis. Am. J. Pathol. 161:166.
Dick, A. D.. 2000. Immune mechanisms of uveitis: insights into disease pathogenesis and treatment. Int. Ophthalmol. Clin. 40:1.
Dick, A. D.. 1999. Immune regulation of uveoretinal inflammation. Dev. Ophthalmol. 30:187.
Dick, A. D., Y. F. Cheng, J. Liversidge, J. V. Forrester. 1994. Immunomodulation of experimental autoimmune uveoretinitis: a model of tolerance induction with retinal antigens. Eye 8:52.
Dick, A. D., V. Sharma, J. Liversidge. 2001. Single dose intranasal administration of retinal autoantigen generates a rapid accumulation and cell activation in draining lymph node and spleen: implications for tolerance therapy. Br. J Ophthalmol. 85:1001.
Jiang, H. R., N. Taylor, L. Duncan, A. D. Dick, J. V. Forrester. 2001. Total dose and frequency of administration critically affect success of nasal mucosal tolerance induction. Br. J. Ophthalmol. 85:739.
Jiang, H. R., L. Lumsden, J. V. Forrester. 1999. Macrophages and dendritic cells in IRBP-induced experimental autoimmune uveoretinitis in B10RIII mice. Invest. Ophthalmol. Vis. Sci. 40:3177.
Dick, A. D., P. G. McMenamin, H. Korner, B. J. Scallon, J. Ghrayeb, J. V. Forrester, J. D. Sedgwick. 1996. Inhibition of tumor necrosis factor activity minimizes target organ damage in experimental autoimmune uveoretinitis despite quantitatively normal activated T cell traffic to the retina. Eur. J. Immunol. 26:1018.
Liversidge, J., A. Dick, S. Gordon. 2002. Nitric oxide mediates apoptosis through formation of peroxynitrite and fas/fas-ligand interactions in experimental autoimmune uveitis. Am. J.(Neil Taylor, Karen McConn)