CD200-FC, a novel antiarthritic biologic agent that targets proinflammatory cytokine expression in the joints of mice with collagen-induced arthritis

Authors


Abstract

Objective

The CD200 receptor (CD200R) is an inhibitory receptor expressed by myeloid cells that is postulated to play an important role in regulation of the immune system. The purpose of this study was to evaluate the efficacy of a soluble ligand of CD200R in established collagen-induced arthritis (CIA) in mice and to analyze changes in cytokine expression following therapy in order to understand its primary mechanism of action.

Methods

Arthritis was induced in DBA/1 mice, and CD200-Fc fusion protein, an isotype control monoclonal antibody, or TNFR-Fc fusion protein was administered over a period of 10 days (total of 4 doses). Cytokine expression in the joint was assessed by flow cytometry, enzyme-linked immunosorbent assay, and quantitative real-time polymerase chain reaction.

Results

CD200-Fc significantly reduced the severity of established arthritis at the clinical and histologic levels. The therapeutic effect of CD200-Fc at 1 mg/kg was comparable with that of TNFR-Fc at 4 mg/kg. CD200R was found to be expressed in arthritic synovia and in lymph nodes, yet no changes in T cell cytokine levels (interferon-γ, interleukin-5 [IL-5], IL-10, IL-17) were detected after CD200-Fc therapy. There was no evidence of an expansion of forkhead box P3–positive regulatory T cells or a change in serum anticollagen IgG1 and IgG2a levels. However, administration of CD200-Fc markedly decreased the expression of messenger RNA for tumor necrosis factor α, IL-1β, IL-10, and matrix metalloproteinase 13 in the joint to the same extent as administration of TNFR-Fc.

Conclusion

CD200-Fc is an effective therapeutic agent in established CIA that targets proinflammatory cytokine expression in the joint without any obvious systemic immunosuppressive effects. Our findings indicate that CD200-Fc has considerable potential as a novel therapeutic agent in rheumatoid arthritis in humans.

The CD200 receptor (CD200R) is a member of the immunoglobulin superfamily that is expressed predominantly by cells of myeloid lineage. Several studies have suggested that stimulation of CD200R (also known as CD200R1) by its ligand, CD200, delivers inhibitory signals to myeloid cells (1–3), and this has raised the possibility of using CD200R agonists to treat chronic inflammatory diseases, such as rheumatoid arthritis (RA). For example, targeting of CD200R with soluble CD200-Fc fusion protein or with an agonistic anti-CD200R monoclonal antibody (mAb) successfully prevented the development of collagen-induced arthritis (CIA) when given prophylactically (4, 5). Conversely, CD200–/– mice had increased numbers of activated macrophages and enhanced susceptibility to a number of autoimmune diseases, including CIA and experimental autoimmune encephalomyelitis (1). In mice, there are at least 3 gene products related to CD200R (or, CD200R1); these are known as CD200R2–4. Previous studies have shown that mouse CD200 binds exclusively to the inhibitory receptor CD200R (6), although the possibility of binding to CD200R2–4 has not yet been ruled out (7); presumably, CD200-Fc targets inhibitory CD200R.

To study the possible therapeutic potential of targeting of the CD200/CD200R pathway in RA, we investigated the effect of administering CD200-Fc to mice with established CIA, a validated model of inflammatory arthritis that resembles RA, and we analyzed changes in cytokine expression in arthritic joints as well as in draining lymph nodes following therapy. Our findings reveal that CD200-Fc is effective in reducing disease activity, and it also reduces proinflammatory cytokine expression in the synovium without altering systemic immune responses, suggesting that the CD200/CD200R pathway is a potential therapeutic target in RA.

MATERIALS AND METHODS

Reagents.

The mouse CD200-Fc fusion protein (IgG2a) was provided by Trillium Therapeutics Inc. (Toronto, Ontario, Canada). IgG2a mAb of irrelevant specificity (antiragweed) and TNFR-Fc fusion protein (IgG1), a soluble murine tumor necrosis factor receptor (p75) fusion protein, were used as controls (8).

Induction of CIA, therapeutic protocol, and scoring systems.

Arthritis was induced in 8–12-week-old male DBA/1 mice by subcutaneous immunization with bovine type II collagen (200 μg) in Freund's complete adjuvant (Difco, Detroit, MI). Treatment was started after the onset of clinical arthritis (day 1 of arthritis). CD200-Fc fusion protein (1 mg/kg), an isotype control mAb (1 mg/kg), or TNFR-Fc fusion protein (4 mg/kg) was administered intraperitoneally over a period of 10 days (3 times per week, for a total of 4 injections). All animal experiments were approved by the local Ethics Review Process Committee and the UK Home Office.

The development of clinical arthritis was assessed in each paw. A semiquantitative scoring system, giving a maximum possible score of 12 per mouse, was used (9). The thickness of the hind paws was monitored using calipers.

For histologic assessments, the hind and front paws were fixed, decalcified in EDTA, and embedded in paraffin. Tissue sections were prepared and stained with hematoxylin and eosin. Histopathologic changes were assessed using a semiquantitative scoring system for each paw (0–3 scale, where 0 = normal, 1 = minimal synovial inflammation, with cartilage and bone erosions limited to discrete foci, 2 = synovial inflammation and moderate erosion, with normal joint architecture intact, and 3 = severe inflammation and severe erosion, with joint architecture disrupted), as described previously (9).

Assessment of T cell proliferation.

Single-cell suspensions were prepared from harvested spleens and inguinal lymph nodes. Cells were resuspended at 5 × 106 cells/ml in RPMI 1640 supplemented with fetal calf serum (10%), HEPES (10 mM), sodium pyruvate (1 mM), 2-mercaptoethanol (50 mM), L-glutamine (1%), and penicillin/streptomycin (100 units/ml), and cultured for 48 hours with 50 μg/ml of bovine type II collagen or 0.1 μg/ml of anti-CD3 mAb (clone 145-2C11; eBioscience, San Diego, CA). After incubation for an additional 16 hours with 1 μCi of 3H-thymidine (Amersham, Buckinghamshire, UK), the plates were harvested, and proliferation was quantified.

Expression of CD200R in synovial tissue.

Fresh frozen synovial tissues and synovial cells from the knee joints of arthritic DBA/1 mice were analyzed by immunohistochemistry and by flow cytometry (fluorescence-activated cell sorting [FACS]), respectively.

For immunohistochemical staining, cryosections of synovial tissues were fixed with acetone for 10 minutes, blocked with 2% bovine serum albumin and 5% rabbit serum, and then stained with rat anti-mouse CD200R1 mAb (clone OX-110; AbD Serotec, Raleigh, NC) for 1 hour at room temperature. A matched concentration of rat IgG2a (AbD Serotec) served as a control. Endogenous peroxidase was depleted with 0.1% H2O2 and 0.1% NaN3. The sections were incubated for 30 minutes with biotinylated rabbit anti-rat mAb (Vector, Burlingame, CA) and were subsequently stained using the biotin–streptavidin–peroxidase method (Vector). Horseradish peroxidase (HRP) activity was detected with 3,3′-diaminobenzidine and H2O2 (Sigma, St. Louis, MO). Slides were counterstained with Mayer's hematoxylin (Dako, Glostrup, Denmark).

For FACS analysis, synovial tissues (n = 10) were pooled, and single synovial cell suspensions were prepared as described previously (10). Cells were stained with rat anti-mouse phycoerythrin (PE)–labeled F4/80 mAb (Caltag, Burlingame, CA) and fluorescein isothiocyanate (FITC)–labeled CD200R1 mAb. Isotype-matched control mAb were used to determine nonspecific staining. In addition, spleen and lymph node single-cell suspensions were first stained with PE-labeled forkhead box P3 (FoxP3) mAb (FJK-16s; eBioscience), and then with FITC-labeled CD4 (GK1.5) and allophycocyanin-labeled CD8 (53-6.7) (both from BD Biosciences, San Diego, CA). Expression of cell-surface and intracellular markers was assessed using a FACSCalibur flow cytometer (Becton Dickinson, Mountain View, CA). Data were analyzed using FlowJo software (Tree Star, Ashland, OR).

Real-time quantitative reverse transcription–polymerase chain reaction (RT-PCR).

The hind paws and front paws were removed just above the ankle/wrist on day 10 of arthritis and were then dissected, snap-frozen, and pulverized. RNA isolation and RT-PCR were performed as previously described (11). RNA transcripts were quantified by real-time quantitative RT-PCR using TaqMan chemistry and a Rotorgene 6000 real-time thermocycler (Corbett Life Science, Sydney, New South Wales, Australia). Tumor necrosis factor α (TNFα), interleukin-1β (IL-1β), IL-10, matrix metalloproteinase 13 (MMP-13), and hypoxanthine guanine phosphoribosyltransferase forward and reverse primers and TaqMan-labeled probes were used (Assays-on-Demand; Applied Biosystems, Foster City, CA). Thermal cycle conditions were set according to the manufacturer's recommendations. For standards, 5-fold dilutions of lipopolysaccharide-stimulated RAW 264.7 cell complementary DNA were included in each run. Standard curves were generated by linear regression using log(Ct) versus log(cell number). The cell equivalent number for synovial samples was calculated (11).

Measurement of serum cytokine, collagen-specific IgG, and CD200-Fc levels.

Commercial enzyme-linked immunosorbent assays (ELISAs) were used to measure serum levels of interferon-γ (IFNγ), IL-5, and IL-10 (OptEIA sets; BD Biosciences). Levels of anticollagen IgG1 and IgG2a were measured by ELISA, using HRP-conjugated goat anti-mouse IgG1 or IgG2a (BD PharMingen, San Diego, CA) (9). CD200-Fc was detected using the CD200 mAb (clone OX-90; BD Biosciences) as a capture antibody and biotin-conjugated rat anti-mouse IgG2a (BD PharMingen) followed by HRP-labeled streptavidin (Amersham) to detect CD200-Fc. The detection threshold of this assay was 2 ng/ml.

Statistical analysis.

Statistical analyses were performed by Student's t-test or analysis of variance, followed by Dunnett's multiple comparison test. P values less than 0.05 were considered significant.

RESULTS

Effect of CD200-Fc therapy on established CIA.

As shown in Figure 1A, CD200-Fc markedly reduced the clinical severity of disease (P < 0.001 versus control). TNF blockade with soluble TNFR-Fc at 4 mg/kg significantly reduced the clinical severity of arthritis, as was expected. The efficacy of TNF blockade was equivalent to that of CD200-Fc treatment. Serum levels of CD200-Fc on day 10 were 8.47 ± 0.97 μg/ml (mean ± SEM; n = 23 mice). Histologic evaluation of the ankle and wrist joints revealed significantly decreased synovial inflammation and cartilage and bone destruction after CD200-Fc therapy as compared with controls (P < 0.01) (Figures 1B and C).

Figure 1.

Suppression of collagen-induced arthritis in mice by CD200-based therapy. A, Clinical arthritis scores and paw thicknesses in mice treated with CD200-Fc fusion protein (1 mg/kg; n = 45), TNFR-Fc fusion protein (4 mg/kg; n = 15), or isotype control monoclonal antibody (mAb) (1 mg/kg; n = 27) 3 times a week over a period of 10 days (total of 4 doses). Arthritis was evaluated using a semiquantitative scoring system, and paw thickness was measured with calipers (expressed as the percentage increase in paw width relative to the width before arthritis onset). Values are the mean ± SEM of pooled data from 6 different experiments. P values are for the comparison of CD200-Fc–treated versus isotype control mAb–treated mice, as determined by Dunnett's multiple comparison test. B, Representative hematoxylin and eosin–stained sections of synovium from the hind paws and front paws of arthritic mice obtained on day 10 after treatment with CD200-Fc or isotype control mAb. Histologic changes were scored on a scale of 0–3. Scores for each section are shown at the lower left. Arrows indicate synovial hyperplasia. (Original magnification × 200.) C, Histology scores on day 10 of arthritis in mice treated with CD200-Fc (n = 8) or isotype control mAb (n = 7). Values are the mean and SEM. P value was determined by Student's t-test.

Expression of CD200R in synovial tissue.

In view of the profound therapeutic effect of CD200-Fc in vivo, we sought to determine whether the target for CD200-Fc (CD200R) is expressed in the joints of mice with CIA. Immunohistochemical analysis of synovial tissues from the knee joints of arthritic mice revealed a high level of CD200R expression, primarily in the sublining layer of the synovium, although some staining was also seen in the lining layer (Figure 2A). Flow cytometric analysis revealed that ∼40% of CD200R+ synovial cells were macrophage-like cells (Figure 2B).

Figure 2.

Suppression of gene expression in the joints of mice with collagen-induced arthritis by CD200-based therapy. A, Representative immunohistochemical staining of synovial tissue sections obtained from arthritic mice on day 10. Sections were stained with CD200 receptor (CD200R) protein (rat anti-mouse CD200R1 monoclonal antibody [mAb]) or with rat IgG2a control (original magnification × 400). B, Expression of CD200R in synovium. Fluorescence-activated cell sorter analyses were performed on synovial cells isolated and pooled from arthritic mice on day 10. Values are the mean and SEM of pooled data from 5 different experiments (n = 10 mice per group). C, Results of real-time polymerase chain reaction analysis of synovial gene expression on day 10 of arthritis following therapy with CD200-Fc (1 mg/kg; n = 6), isotype control mAb (1 mg/kg; n = 10), or TNFR-Fc (4 mg/kg; n = 5). The levels of mRNA in normal DBA/1 mice without arthritis (n = 4) are shown for comparison. Results are expressed as the ratio between the gene of interest cell equivalent and the hypoxanthine guanine phosphoribosyltransferase (HPRT) cell equivalent, yielding relative expression units (REU). Values are the mean and SEM. ∗ = P < 0.05 versus controls; ∗∗ = P < versus normal mice, by Student's t-test. << = below the threshold limit.

Reduction of synovial gene expression with CD200-Fc therapy.

Because of the high level of CD200R expression in the arthritic synovium, we characterized the effect of CD200-Fc therapy on messenger RNA (mRNA) levels of 2 key proinflammatory cytokines (IL-1β and TNFα), 1 antiinflammatory cytokine (IL-10), and 1 enzyme (MMP-13 or collagenase 3) that are known to play a significant role in inflammation and joint destruction.

As shown in Figure 2C, the expression of TNFα, IL-1β, and IL-10 mRNA on day 10 of arthritis was significantly reduced in the joints of mice treated with CD200-Fc or TNFR-Fc as compared with controls (P < 0.05). CD200-Fc therapy resulted in a significant reduction of MMP-13 mRNA (P < 0.05).

No evidence of an effect of CD200-Fc therapy on T cell or B cell immunity.

CD200R was also expressed by spleen and lymph node cells (data not shown), and to assess the effects of CD200-Fc therapy on the immune system, cells from spleens and draining lymph nodes obtained from mice treated with CD200-Fc or control mAb were cultured in the presence of type II collagen or a mitogen (anti-CD3 mAb), and T cell proliferation was measured. The stimulation indices in spleen cells (data not shown) or lymph node cells cultured with type II collagen or with anti-CD3 mAb were comparable in CD200-Fc–treated and control mice (Table 1). Similarly, no changes in IFNγ, IL-5, or IL-10 levels were detected in culture supernatants.

Table 1. Effect of treatment with CD200-Fc fusion protein, isotype control mAb, and soluble TNFR-Fc fusion protein on T cell and B cell immunity in mice with collagen-induced arthritis*
 CD200-Fc fusion proteinIsotype control monoclonal antibodySoluble TNFR-Fc fusion protein
Mean ± SEMNo. of miceMean ± SEMNo. of miceMean ± SEMNo. of mice
  • *

    There was no effective modulation of T cell or B cell immunity with CD200-based therapy compared with isotype control monoclonal antibody and with TNFR-Fc therapy in mice with collagen-induced arthritis. See Materials and Methods for a description of the treatment groups. ND = not determined; FoxP3 = forkhead box P3.

  • Stimulation indices (SIs) were calculated as the proliferation of cells cultured with type II collagen (CII) or with anti-CD3 divided by the proliferation of cells cultured with medium alone.

  • Pooled sera from several mice with collagen-induced arthritis were used as a standard, and levels of collagen-specific IgG1 and IgG2a were arbitrarily assigned a concentration of 1,000 units/ml. Test sera were then related to this value and expressed in arbitrary units (AU).

T cell proliferation ex vivo, SI      
 Lymph node cells      
  CII, 50 μg/ml1.2 ± 0.453.2 ± 2.35ND 
  Anti-CD3, 0.1 μg/ml29.8 ± 1.6432.5 ± 8.84ND 
FoxP3+ regulatory T cells, %      
 Lymph node cells      
  CD4+10.5 ± 0.3410.2 ± 0.3711.4 ± 0.55
  CD8+1.4 ± 1.041.0 ± 0.150.8 ± 0.14
 Spleen cells      
  CD4+6.9 ± 0.346.9 ± 0.276.9 ± 0.35
  CD8+1.4 ± 0.840.9 ± 0.260.8 ± 0.14
Collagen-specific antibody titers, AU      
  IgG1243.3 ± 40.926208.8 ± 38.424331.2 ± 21.69
  IgG2a37.0 ± 6.02632.4 ± 5.52452.8 ± 6.19

T cells from CD200-Fc–treated and control mice were also analyzed ex vivo for intracellular cytokine expression, as determined by flow cytometry. However, no changes in the proportions of IFNγ+, IL-4+, or IL-10+ T cells were detected (data not shown). Similarly, no differences in the proportion or absolute number of FoxP3+,CD4+ or FoxP3+,CD8+ T cells in CD200-Fc–treated mice compared with controls (Table 1).

Serum levels of collagen-specific IgG1 and IgG2a were similar in the CD200-Fc–treated and control mice, indicating a lack of effect of CD200-Fc on the humoral arm of the immune response (Table 1). It was concluded that CD200-Fc was not acting primarily on the immune response to type II collagen and certainly did not have a generalized immunosuppressive effect.

DISCUSSION

There are several new observations in this study that extend earlier observations regarding the inhibitory CD200/CD200R pathway and the therapeutic potential of CD200-Fc, a fusion protein that combines the extracellular portion of murine CD200 and the Fc portion of IgG2a. Previous studies have shown that mice receiving soluble CD200-Fc from the time of immunization were resistant to CIA, whereas CD200–/– mice or mice treated with soluble CD200R protein (which binds to CD200 and thus prevents CD200–CD200R interactions) showed increased susceptibility to CIA (1, 4). However, the targeting of a particular molecule prior to or around the time of immunization often does not predict the effect of targeting that molecule in established disease, and for this reason, we analyzed the effect of CD200-Fc in established arthritis.

The results of this study show, first, that CD200-Fc strongly suppressed the clinical and histopathologic severity of established CIA, a model of RA in humans, and second, that CD200-Fc protein down-regulates mRNA expression of TNFα and IL-1β as well as MMP-13 in mice with CIA. IL-10 gene expression was reduced below the detection limit, possibly as a consequence of the diminished expansion of TNFα and IL-1β (12). It is interesting to note that the antiinflammatory effects of IL-10 production are limited to myeloid cells (in particular, macrophages).

CD200-Fc therapy was effective at a dosage of 1 mg/kg given every 2–3 days. This dosage was chosen because it was shown in previous studies to be consistently effective in preventing CIA as well as in transplantation experiments to be effective in promoting increased skin and renal graft survival (4, 13). We also found that higher dosages of CD200-Fc did not prove to be more effective, and a lower dosage of 0.5 mg/kg showed evidence of efficacy, but the small numbers of mice treated precluded statistical analysis (data not shown).

In light of the impressive therapeutic effect of CD200-Fc in CIA, we investigated whether CD200R is expressed in the joint in mice with CIA and whether CD200-Fc reduces the synovial expression of TNFα, IL-1β, IL-10, and MMP-13. Large numbers of CD200R+ cells were detected by immunohistochemistry in the sublining layer and to a lesser extent in the lining layer of the synovium. Following therapy with CD200-Fc, levels of both TNFα and IL-1β mRNA were reduced almost to levels found in healthy mice, and we concluded that inhibition of these key proinflammatory cytokines provides, at least partly, the basis for the therapeutic action of CD200-Fc. However, based on these results, we cannot deduce whether CD200-Fc acts directly on cells in the joint, causing them to reduce their output of TNFα and IL-1β, or whether the fusion protein reduces the infiltration of inflammatory cells into the joint. MMP-13 mRNA levels were also reduced, which is consistent with the reduced degree of joint erosion in CD200-Fc–treated mice because MMP-13 is known to play a major role in bone and cartilage breakdown by enhancing the cleavage of type II collagen (14).

Of interest is a recent report showing the expression of CD200R on T cells and B cells in humans as well as in mice (15). We also found CD200R to be expressed in the spleen and draining lymph nodes, but surprisingly, no changes in the immunologic parameters were detected as a result of CD200-Fc therapy. Thus, T cell proliferation, cytokine expression, and numbers of FoxP3+ regulatory T cells were all unchanged, and levels of anticollagen IgG were similar in CD200-Fc–treated and control mice. We conclude, therefore, that CD200-Fc does not directly suppress T cell responses to type II collagen but acts primarily on the inflammatory response. Intriguingly, a recent study by Cui and colleagues (16) demonstrated that CD200–/– mice had a defect in osteoclastogenesis and increased bone mass and that exogenous CD200 rescued the differentiation of osteoclasts from CD200-deficient macrophages in vitro. In fact, these results established an important role played by the CD200/CD200R pathway in the regulation of bone mass via the formation of osteoclasts.

Taken together, the findings of this study demonstrate that CD200-Fc has a profound therapeutic effect on CIA and is as effective as TNFR-Fc. This study provides the first evidence of the ability of CD200-Fc to reduce TNFα and IL-1β expression in the synovium without affecting B cell or T cell activity. CD200-Fc therefore represents a potentially new immunotherapeutic agent for the management of arthritis as well as other autoimmune diseases.

AUTHOR CONTRIBUTIONS

Dr. Šimelyte had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Šimelyte, Criado, Uger, Feldmann, Williams.

Acquisition of data. Šimelyte, Criado, Essex, Williams.

Analysis and interpretation of data. Šimelyte, Criado, Feldmann, Williams.

Manuscript preparation. Šimelyte, Feldmann, Williams.

Statistical analysis. Šimelyte.

Acknowledgements

We thank Dr. Clare A. Notley and Dr. Fiona McCann for helpful discussions. Dr. Ivo Boudakov is acknowledged for critical revision of the manuscript.

Ancillary