SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Objective

The binding of abatacept (CTLA-4Ig) to the B7 ligands CD80 and CD86 prevents the engagement of CD28 on T cells and thereby prevents effector T cell activation. In addition, a direct effect of CTLA-4Ig on antigen-presenting cells (APCs) could contribute to the therapeutic effect. To further elucidate the mechanism of CTLA-4Ig, we performed phenotype and functional analyses of APCs in patients with rheumatoid arthritis (RA) before and after the initiation of CTLA-4Ig therapy.

Methods

Peripheral blood mononuclear cells were analyzed before and at 2 and 4 weeks after the initiation of CTLA-4Ig therapy. Proportions of APCs were determined by flow cytometry. CD14+ monocytes were further analyzed for the expression of costimulatory and adhesion molecules and for their transendothelial migratory capacity in vitro. In addition, CD14+ monocytes from healthy controls were analyzed for their migratory and spreading capacity.

Results

Proportions and absolute numbers of monocytes were significantly increased in RA patients treated with CTLA-4Ig. The expression of several adhesion molecules was significantly diminished. In addition, monocytes displayed a significant reduction in their endothelial adhesion and transendothelial migratory capacity upon treatment with CTLA-4Ig. Likewise, isolated monocytes from healthy controls revealed a significant reduction in their migratory and spreading activity after preincubation with CTLA-4Ig or anti-CD80 and anti-CD86 antibodies.

Conclusion

We describe direct effects of CTLA-4Ig therapy on phenotype and functional characteristics of monocytes in RA patients that might interfere with the migration of monocytes to the synovial tissue. This additional mechanism of CTLA-4Ig might contribute to the beneficial effects of CTLA-4Ig treatment in RA patients.

Rheumatoid arthritis (RA) is a chronic debilitating systemic autoimmune disease characterized by inflammation and destruction of the joints. While activation of fibroblast-like synoviocytes, macrophages, dendritic cells (DCs), and B cells, including production of various cytokines and autoantibodies, is essential in RA, activated T cells play a central role in upstream parts of the inflammatory cascade (1, 2). The growing understanding of the pathophysiologic mechanisms underlying RA at the molecular and cellular levels has triggered an avalanche of new and highly efficient biologic therapies targeting proinflammatory cytokines or their receptors (e.g., infliximab, etanercept, adalimumab, tocilizumab, anakinra) as well as cell surface or costimulatory molecules (e.g., rituximab, abatacept).

Abatacept is a soluble recombinant human CTLA-4Ig fusion protein comprising the extracellular domain of human CTLA-4 and a fragment of the Fc domain of human IgG1. It has been proposed that CTLA-4Ig decreases T cell responses by competing for B7 ligand (CD80/CD86) access to CD28 and limiting the CD28 signaling that is required for T cell activation (3). In addition, however, CTLA-4Ig therapy might exert effects beyond T cells. For example, reverse signaling to antigen-presenting cells (APCs) upon binding of CTLA-4Ig to CD80/CD86 has been described that might interfere with APC activation and function (4, 5). Whether such signaling or other CTLA-4Ig–mediated effects contribute to a beneficial or adverse outcome, in particular in the therapeutic setting of RA patients, is not entirely clear so far. We therefore assessed the effects of CTLA-4Ig treatment on APCs in RA patients, and indeed we observed a profound effect of CTLA-4Ig on the phenotype and function of monocytes.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Patients and healthy volunteers.

Twelve patients with RA according to the 1987 revised criteria of the American College of Rheumatology (ACR) (6) who were eligible for CTLA-4Ig therapy were consecutively selected from our outpatient clinic. All patients were treated with 10 mg/kg CTLA-4Ig at baseline, week 2, and week 4. No premedication with glucocorticoids or antihistamines was given before the patients received CTLA-4Ig infusions. Heparinized whole blood samples were obtained at baseline and at weeks 2 and 4 during treatment with CTLA-4Ig.

In addition, heparinized whole blood samples were obtained from healthy volunteers. We obtained Ethics Committee approval for this study as well as informed consent from the patients. Detailed demographic and clinical characteristics of the RA patients are provided in Table 1. We assessed clinical disease activity with the Simplified Disease Activity Index (SDAI) and the Clinical Disease Activity Index (CDAI) (7) as well as with the Disease Activity Score in 28 joints using the C-reactive protein level (DAS28-CRP) (8).

Table 1. Demographic and clinical characteristics of the patients with rheumatoid arthritis at baseline and during CTLA-4Ig treatment*
CharacteristicBaseline (week 0)Week 2Week 4
  • *

    Except where indicated otherwise, values are the mean ± SD (range). MTX = methotrexate; CRP = C-reactive protein; SDAI = Simplified Disease Activity Index; CDAI = Clinical Disease Activity Index; DAS28-CRP = Disease Activity Score in 28 joints using the CRP level.

Age, mean years56
Prednisolone, %58
Prednisolone, mg/day5.3 ± 6.8
MTX, %58
Leflunomide, %25
CRP, mg/dl1 ± 1.2 (0.02–4.5)1.4 ± 1.5 (0.01–5.4)4.6 ± 0.7 (0.01–7.5)
SDAI score (range 0–86)25 ± 8.5 (10.1–36.4)24 ± 8.9 (9.4–36.9)24 ± 7.2 (13.5–35)
CDAI score (range 0–17)26 ± 8.9 (9.7–40.4)24 ± 8.3 (7.7–35.9)21 ± 7.5 (9.2–35.4)
DAS28-CRP3.7 ± 0.6 (2.4–4.5)3.7 ± 0.6 (2.7–4.7)3.8 ± 0.6 (2.9–4.8)

Antibodies.

Monoclonal antibodies (mAb) targeting a variety of molecules were used unlabeled or as fluorescein isothiocyanate (FITC), phycoerythrin (PE), PerCP, allophycocyanin, PE–Cy5.5, PE–Cy7, or allophycocyanin–Cy7 conjugates. Monoclonal antibodies against CD80 (L307.4), CD86 (2331 [FUN-1]), CD19 (SJ25C1), and HLA–DR (B8.12.2) were obtained from Becton Dickinson, mAb against CD14 (RMO52) was obtained from Beckman Coulter, mAb against CD40 (L0D7/6), CD50 (MEM-171), and CD54 (84H10) were obtained from Serotec, and mAb against CD15 (HI98), CD58 (HCD58), CD62E (HCD62E), CD106 (STA), CD80 (16-10A1), and CD86 (GL-1) were obtained from BioLegend. Anti-CD1c (AD5-8E-7) and anti-CD303 (AC144) were obtained from Miltenyi Biotec. Alexa Fluor 488 phalloidin was obtained from Invitrogen, and antivinculin mAb was obtained from Sigma. In all experiments, appropriate Ig isotype-matched control mAb served as negative controls.

Phenotype analysis.

After informed consent was obtained, blood was withdrawn during routine screening laboratory testing. Peripheral blood mononuclear cells (PBMCs) were isolated from heparinized blood by LSM (PAA Laboratories) density-gradient centrifugation. PBMCs were resuspended in phosphate buffered saline (PBS)/3% human IgG (Baxter International) in order to block Fc receptors and prevent nonspecific antibody binding and were incubated for 15 minutes at 4°C in the dark with combinations of FITC-, PE-, PE–Cy5.5–, PE–Cy7–, allophycocyanin-, and allophycocyanin–Cy7–conjugated mAb. Afterward, the cells were washed with PBS/1% bovine serum albumin. Background fluorescence was assessed using appropriate isotype- and fluorochrome-matched control mAb. After staining with the indicated antibodies, cells were analyzed by flow cytometry (FACSCanto II, FACSDiva software; BD Biosciences). Absolute numbers of cells were calculated from whole blood counts obtained from routine laboratory testing.

Isolation of monocytes.

PBMCs were isolated as described above. CD14+ monocytes were purified by fluorescence-activated cell sorting (FACS) on a FACSAria (BD Immunocytometry Systems). Forward scatter area versus pulse width discrimination of doublets and cell aggregates was logically combined with the identification of monocytes by means of a mouse mAb against human CD14 to trigger a positive sort decision. Post-sort reanalyses revealed the achievement of purities >98%.

Transendothelial migration assay.

Human umbilical vein endothelial cells (HUVECs) were isolated from individual human umbilical cord veins by collagenase digestion as described previously (9) and maintained in Clonetics EBM-2 Medium (CC-3156; Lonza) supplemented with SingleQuots Supplements and Growth Factors (CC-4147; Lonza) until grown to confluence. Monocyte interaction with HUVECs was examined by their capacity to migrate into a hydrated, neutralized bovine collagen gel (PureCol; Inamed) coated with a confluent monolayer of endothelial cells prepared 24 hours prior to the analyses, as described previously (10). Monocytes were applied to the gels at 2 × 105 per well and incubated for 36 hours at 37°C in an atmosphere of 5% CO2, 95% relative humidity. Nonadherent cells were collected by rinsing 3 times with warm EBM-2 Medium. Monocytes that had actively migrated into the collagen gels were recovered by gently mincing and digesting the gels with 0.1% collagenase solution in Hanks' balanced salt solution for 1 hour at 37°C. Each fraction was sedimented at 200g, resuspended in medium, and enumerated using Neubauer hemocytometer slides. Trypan blue exclusion provided information on viability of monocytes, which could be distinguished from HUVECs in the collagenase-digested fraction by size and morphology.

Spreading assay.

CD14+ monocytes from the PBMCs of healthy volunteers were purified by high-gradient magnetic cell sorting using the MACS Monocyte Isolation Kit II (Miltenyi Biotec) according to the manufacturer's instructions. Monocytes were incubated overnight with 100 μg/ml CTLA-4Ig or medium alone. Next, CD14+ monocytes were placed onto fibronectin-coated coverslips. After 90 minutes, the cells were fixed and permeabilized with 2% paraformaldehyde and 0.3% Triton X-100 (Sigma) in PBS, respectively. In pilot experiments, 100 μg/ml of CTLA-4Ig and 90 minutes of incubation were found to be optimal for the detection of differences in cell morphology (data not shown).

Subsequently, cells were processed for confocal microscopy using phalloidin to label F-actin and an antibody to vinculin to label focal adhesions. Nuclear counterstaining was performed using DAPI (Invitrogen). For confocal microscopy, images were collected on an upright Leica TCS SP5 microscope (Leica Microsystems) with a 63× oil immersion objective (Leica Microsystems), zoom 2. Fluorochromes were excited with a Multi-Photon MaiTai Ti:Sapphire laser (Spectra-Physics; Newport Corporation) tuned at 800 nm for DAPI, an argon laser at 488 nm for phalloidin, and a HeNe laser at 568 nm and at 633 nm for vinculin detection. Detector slits were configured to minimize any cross-talk between the channels. Z-stacks (optical sections) of the images were collected with an optical thickness of 0.2 μm. Images were processed using the Leica LAS software and Adobe Photoshop 8.0.1 (Adobe). Cell dimension and cell morphology were analyzed using the TissueQuest software program (TissueGnostics).

Statistical analysis.

Values are shown throughout as the mean ± SEM, except for patient characteristics, where mean ± SD values are shown. All values from the FACS analyses of week 0 were normalized to 100% to calculate percent increase or decrease after treatment with CTLA-4Ig. Proportions of lymphocyte subpopulations were compared using Student's t-test for normally distributed values and the Mann-Whitney test for values without Gaussian distribution. Relationships between different results were examined using Pearson's correlation coefficient and Spearman's rank correlation tests, as appropriate. P values less than 0.05 were considered significant in all statistical tests. All statistical analyses were performed using GraphPad Prism software, version 4.0 and SPSS software, version 12.0.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Baseline demographic and clinical characteristics.

Demographic and clinical characteristics at baseline and during treatment of RA patients with CTLA-4Ig are shown in Table 1. After 2 weeks and 4 weeks of treatment with CTLA-4Ig, no significant changes were observed in the CRP, SDAI, CDAI, or DAS28-CRP values compared with baseline.

CTLA-4Ig treatment increases proportions of CD14+ monocytes in RA patients.

We analyzed freshly isolated PBMCs from patients with RA to determine the proportions of CD14+ monocytes, CD19+ B cells, CD1c+ myeloid DCs, and CD303+ plasmacytoid DCs before and after treatment with CTLA-4Ig. As shown in Figure 1a, proportions of CD14+ cells increased significantly during treatment with CTLA-4Ig (P = 0.02 at week 2, P = 0.01 at week 4). In contrast, no significant change was observed in the proportions of B cells or DCs. Consistent with this, analysis of total cell numbers from PBMCs revealed a significant increase only for CD14+ monocytes (P = 0.03 at week 2, P = 0.04 at week 4). No difference was observed for CD19+ B cells, CD1c+ myeloid DCs, or CD303+ plasmacytoid DCs (Figure 1b).

thumbnail image

Figure 1. CTLA-4Ig treatment increases proportions and absolute numbers of CD14+ monocytes in patients with rheumatoid arthritis (RA). Proportions of antigen-presenting cells were analyzed by flow cytometry before and at different time points during CTLA-4Ig therapy. Absolute numbers of cells were calculated from whole blood counts obtained from routine laboratory testing. Values are the mean ± SEM. Significant differences are marked with an asterisk. a, A significant increase in the percentage of CD14+ cells was observed in RA patients (n = 12) 2 weeks (P = 0.02) and 4 weeks (P = 0.01) after the initiation of CTLA-4Ig therapy. b, Likewise, a significant increase was observed for absolute numbers of CD14+ monocytes at week 2 (P = 0.03) and week 4 (P = 0.04) of CTLA-4Ig therapy. G/l = ×109/liter.

Download figure to PowerPoint

Down-regulation of adhesion molecules on monocytes upon treatment with CTLA-4Ig.

The observed effect of CTLA-4Ig on monocyte numbers encouraged us to further study the effect of CTLA-4Ig on this cell population.

CTLA-4Ig treatment does not affect the expression of costimulatory molecules on CD14+ monocytes.

For this purpose, CD14+ cells were analyzed for the expression of the costimulatory molecules CD80, CD86, HLA–DR, and CD40. As shown in Figure 2, however, no significant changes were observed in the percentage or in the mean fluorescence intensity (MFI) of costimulatory molecules expressed on CD14+ cells after treatment with CTLA-4Ig.

thumbnail image

Figure 2. Influence of CTLA-4Ig treatment on the expression of costimulatory molecules on CD14+ monocytes during CTLA-4Ig treatment. CD14+ monocytes were analyzed by flow cytometry for costimulatory molecules before and at different time points during CTLA-4Ig therapy. Values are the mean ± SEM (n = 12 patients). No significant difference was observed in the percentage or mean fluorescence intensity (MFI) of costimulatory molecules expressed on CD14+ cells.

Download figure to PowerPoint

CTLA-4Ig treatment decreases the expression of adhesion molecules on CD14+ monocytes.

Next, we analyzed the expression of the adhesion molecules CD15, CD50 (intercellular adhesion molecule 3 [ICAM-3]), CD54 (ICAM-1), CD58 (lymphocyte function–associated antigen 3), CD62E (E-selectin), and CD106 (vascular cell adhesion molecule 1) on CD14+ cells in patients treated with CTLA-4Ig. Interestingly, a significant decrease was observed in the percentage of CD106+ cells (P = 0.03 after 4 weeks), CD62E+ cells (P = 0.002 after 2 weeks, P = 0.04 after 4 weeks), and CD15+ cells (P = 0.001 after 2 weeks, P = 0.04 after 4 weeks) among CD14+ monocytes. In addition, a significant decrease was observed in the MFI for expression of CD50 (P = 0.04), CD54 (P = 0.03), CD58 (P = 0.03), and CD15 (P = 0.04) on CD14+ monocytes after 2 weeks and for expression of CD58 (P = 0.02) after 4 weeks of treatment with CTLA-4Ig (Figure 3).

thumbnail image

Figure 3. Down-regulation of adhesion molecules on CD14+ monocytes during CTLA-4Ig treatment. CD14+ cells were analyzed by flow cytometry for adhesion molecules before and at different time points during CTLA-4Ig therapy. Values are the mean ± SEM (n = 5 patients). Significant differences are marked with an asterisk. A significant decrease was observed in the percentage of CD15+ cells (P = 0.001 after 2 weeks of treatment with CTLA-4Ig, P = 0.04 after 4 weeks), CD62E+ cells (P = 0.002 after 2 weeks, P = 0.04 after 4 weeks), and CD106+ cells (P = 0.03 after 4 weeks) among CD14+ monocytes. A significant decrease was observed in the mean fluorescence intensity (MFI) for the expression of CD15 (P = 0.04), CD50 (P = 0.04), CD54 (P = 0.03), and CD58 (P = 0.03) on CD14+ monocytes after 2 weeks of treatment with CTLA-4Ig, and for the expression of CD58 (P = 0.02) after 4 weeks of treatment with CTLA-4Ig.

Download figure to PowerPoint

CD80 and CD86 expression on CD14+ monocytes.

In order to test the expression levels of CD80 and CD86, CD14+ monocytes from healthy controls were isolated by cell sorting and analyzed for the expression of CD80 and CD86 by flow cytometry after overnight culture. Overnight culture of monocytes in medium alone was found to be sufficient to induce the expression of CD80 and CD86. (Representative histograms for CD80 and CD86 expression on human monocytes are available online at http://www.meduniwien.ac.at/user/clemens.scheinecker/.)

Treatment with CTLA-4Ig reduces the migratory capacity of CD14+ monocytes from patients with RA.

To further examine whether the observed down-regulation of adhesion molecules on monocytes after CTLA-4Ig treatment also affects the migratory behavior of monocytes, we isolated CD14+ cells by FACS before and after treatment with CTLA-4Ig and analyzed the cells using an in vitro assay for transendothelial migration. In fact, counting the migrated CD14+ cells in 24-well plates after transmigration revealed that monocytes from RA patients had a decreased capacity to migrate through the transendothelial cell layer after 2 weeks of treatment with CTLA-4Ig (P = 0.03), and this decreased capacity was even more pronounced after 4 weeks of treatment with CTLA-4Ig (P = 0.001). Likewise, a significant increase in the nonadherent cell fraction was observed after 2 weeks (P = 0.01) and 4 weeks (P = 0.001) of treatment with CTLA-4Ig (Figure 4a). This indicates that monocytes from RA patients exhibit a significantly decreased capacity to adhere to and to migrate through the endothelial cell layer after treatment with CTLA-4Ig, as compared to before treatment with CTLA-4Ig. Thus, the rise in circulating monocytes may be due to a decreased capacity to migrate into the tissues.

thumbnail image

Figure 4. Influence of CTLA-4Ig on adhesion and transendothelial migration of CD14+ cells in patients with rheumatoid arthritis (RA) and in healthy controls. Values are the mean ± SEM. Significant differences are marked with an asterisk. a, The adhesive and migratory capacity of monocytes from RA patients was assessed by a transendothelial migration assay before and at different time points during CTLA-4Ig treatment. The nonadherent cell fraction was significantly increased after 2 weeks (P = 0.01) and 4 weeks (P = 0.001) of treatment with CTLA-4Ig. In contrast, the percentage of migrated CD14+ cells was significantly decreased after 2 weeks (P = 0.03) and 4 weeks (P = 0.001) of treatment with CTLA-4Ig (n = 5 patients). b, Incubation of fluorescence-activated cell sorting–isolated CD14+ cells from healthy controls with 50 μg/ml and 100 μg/ml CTLA-4Ig led to a significant increment of the nonadherent cell fraction (P = 0.02 for both). The migrated cell population was significantly decreased after incubation with 50 μg/ml (P = 0.02) and 100 μg/ml (P = 0.001) CTLA-4Ig (n = 3 experiments). c, CD14+ cells from healthy controls were isolated and incubated with anti-CD80 plus anti-CD86 antibodies or 100 μg/ml CTLA-4Ig overnight. Incubation with 100 μg/ml CTLA-4Ig led to a significant increase of the nonadherent cell fraction (P = 0.02). A similar result was observed after incubation with anti-CD80 and anti-CD86 as compared to appropriate isotype control monoclonal antibodies (P = 0.008). The migrated cell population was significantly decreased after incubation with 100 μg/ml CTLA-4Ig (P = 0.0003) as well as after incubation with anti-CD80 and anti-CD86 (P = 0.02) (n = 3 experiments).

Download figure to PowerPoint

In vitro incubation with CTLA-4Ig reduces the migratory capacity of CD14+ monocytes from healthy controls.

To see whether the results of the migration assay from RA patients could also be reproduced in vitro, CD14+ cells from healthy controls were isolated and incubated with various concentrations of CTLA-4Ig overnight. Indeed, preincubated monocytes showed a dose-dependent decrease of transendothelial migration after incubation with 50 μg/ml CTLA-4Ig (P = 0.02) and 100 μg/ml CTLA-4Ig (P = 0.001). Corresponding results were observed for the nonadherent cell fraction. Incubation of monocytes from healthy controls with 50 μg/ml and 100 μg/ml CTLA-4Ig led to a significant increase in the nonadherent cell fraction (P = 0.02 for both) (Figure 4b).

CD80/CD86-dependent reduced migratory capacity of CTLA-4Ig–treated monocytes.

To further elucidate whether the observed effects of CTLA-4Ig were CD80/CD86 dependent, CD14+ cells from healthy controls were isolated and incubated with anti-CD80 plus anti-CD86 antibodies or CTLA-4Ig (100 μg/ml) overnight. Indeed, a significant and comparable decrease in transendothelial migration was observed after preincubation of monocytes with anti-CD80 plus anti-CD86 mAb as compared to appropriate isotype control mAb (P = 0.02) as well as with CTLA-4Ig as compared to medium control (P = 0.0003). Similar results were observed for the nonadherent cell fraction. Incubation of monocytes from healthy controls with anti-CD80 plus anti-CD86 mAb (P = 0.008) as well as with CTLA-4Ig (P = 0.02) led to a significant increase in the nonadherent cell fraction as compared to isotype control mAb or medium control, respectively (Figure 4c).

CD14+ monocytes from healthy controls treated with CTLA-4Ig display diminished spreading capacity.

In order to analyze the influence of CTLA-4Ig on actin cytoskeletal dynamics of cell-to-matrix adhesion in vitro, cell spreading assays were performed. We isolated CD14+ cells from the PBMCs of healthy controls by magnetic-activated cell sorting and incubated the cells overnight with or without 100 μg/ml CTLA-4Ig. Fibronectin was used as an adhesion substrate. The cells were placed on fibronectin-coated coverslips for various periods of time; their capacity to spread as a function of actin reorganization was assessed by staining of polymerized actin with phalloidin. Focal adhesions were visualized with an antibody to vinculin.

Fluorescence imaging showed that cells cultured with CTLA-4Ig exhibited diminished spreading when compared to untreated cells. CTLA-4Ig–treated CD14+ monocytes were mostly round shaped with few cellular processes and poor assembly of focal adhesions. In contrast, the untreated monocytes displayed a more polygonal cell shape with a rim of cortical actin, prominent cellular processes emanating from the cell body, and multiple focal adhesions at the periphery of the cells (Figure 5a). Cell dimensions were further analyzed using TissueQuest software. The total cell area for CD14+ cells that were cultured with CTLA-4Ig was significantly reduced compared to the total cell area for cells cultured without CTLA-4Ig (mean ± SEM 397 ± 3 μm2 versus 424 ± 2 μm2; P < 0.0001) (Figure 5b). (Representative examples of images that were analyzed using the TissueQuest program are available online at http://www.meduniwien.ac.at/user/clemens.scheinecker/.)

thumbnail image

Figure 5. CTLA-4Ig leads to decreased spreading of monocytes. Magnetic-activated cell sorting–isolated CD14+ cells from healthy controls were incubated with 100 μg/ml CTLA-4Ig and then seeded onto fibronectin-coated coverslips. Cell spreading was analyzed after 90 minutes. Polymerized actin was stained with phalloidin; focal adhesions were visualized with an antibody to vinculin. Top, Immunofluorescence analysis revealed reduced cell spreading as well as impaired formation of focal adhesions in CTLA-4Ig–treated as compared to untreated monocytes. Bottom, A significant difference (∗ = P < 0.0001) was observed for the mean cell area in CTLA-4Ig–treated monocytes (n = 1,545) as compared to untreated monocytes (n = 3,093). Values are the mean ± SEM. One of 2 similar experiments is shown.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

We show that treatment of RA patients with CTLA-4Ig increased the proportions of peripheral blood CD14+ monocytes and diminished their expression profile of adhesion molecules. The phenotype changes were accompanied by a reduced adhesion and transendothelial migratory capacity of isolated monocytes in vitro. These results were strengthened by similar in vitro observations with CD14+ monocytes from healthy controls, suggesting that CTLA-4Ig treatment exerts important effects beyond the inhibition of T cell activation.

CTLA-4Ig is a selective costimulation modulator that avidly binds to the CD80/CD86 ligands on an APC. This results in the inability of these ligands to engage the CD28 receptor on the T cell that is required for T cell activation. CTLA-4Ig was shown to dose-dependently reduce T cell proliferation, serum concentrations of acute-phase reactants, and other markers of inflammation, including rheumatoid factor produced by B cells (11).

Besides its effect on T cells, CTLA-4Ig has been suspected to affect other cell types, in particular, APCs. Several studies have shown that CTLA-4Ig induces reverse signaling in APCs, although partially conflicting results have been reported so far (4, 5).

For example, intracellular signaling events have been described that are induced by ligation of CD80 and CD86 in a B cell lymphoma, and signaling via CD86 in B cells has been reported to increase immunoglobulin production (12). Alternatively, CD80/CD86 engagement has been suggested to activate B7 molecules and intracellular signaling events, including the recruitment of p38 MAPK and NF-κB. This has been shown to induce an increase in the production of the enzyme indoleamine 2,3-dioxygenase and induction of tryptophan catabolism in murine (13) and human (5) DCs, ultimately leading to the inhibition of T cell proliferation. On the other hand, no or only minimal changes in gene expression have been observed upon in vitro treatment of APCs with abatacept in vitro (14). However, effects on the gene expression levels of CD14+ monocytes have not been analyzed so far.

We did not observe significant changes in clinical disease activity scores upon treatment with CTLA-4Ig. This, however, is not surprising since our study was not powered for clinical efficacy. Moreover, even in large clinical trials, a 20% improvement in disease activity according to the ACR criteria (an ACR20 response) (15) or an ACR50 response was only observed beginning on day 30 or day 60, respectively, of CTLA-4Ig treatment. It is therefore conceivable that early detectable CTLA-4Ig–mediated effects on monocytes might well contribute to the overall effect of CTLA-4Ig. The main mechanism of action, however, is still the modulation of CD28 signaling on T cells. This might explain the longer time period that is required for clinical effects to become apparent.

Our data show a significant increase in the percentage as well as absolute numbers of CD14+ monocytes in RA patients upon CTLA-4Ig treatment, an observation that, to our knowledge, has not previously been reported. Whether these changes affect recently identified subsets of monocytes (16, 17) in different ways was not the aim of the present study, but might justify future analysis. Our subsequent analyses did not reveal substantial changes in the expression profile of costimulatory molecules on monocytes, but rather—and most striking—a reduction in the expression of certain adhesion molecules that are required for the adhesion to, and active transmigration of, monocytes through endothelial barriers. Moreover, and consistent with the phenotype analysis, the functional assessment of isolated CD14+ monocytes revealed that abatacept treatment led to a reduced adhesion of monocytes to endothelial cells and a reduced capacity of monocytes to migrate through an endothelial cell layer in vitro. In addition, we were able to show that the observed effect of CTLA-4Ig on monocyte migration and adhesion was CD80/CD86 dependent, since in vitro experiments with antibodies against CD80 and CD86 led to similar results in migration assays as compared to CTLA-4Ig.

The migratory capacity of cells depends on several mechanisms, including the protrusion in the front of the cell, adhesion at focal contacts, concentration of the cytoplasm, and breaking of older adhesions at the rear of the cell. The reorganization of actin fibers, however, represents the main mechanism operative in this process (18). Indeed, spreading assays revealed a substantial influence of CTLA-4Ig on actin reorganization as well as formation of focal contacts. Thus, the down-regulation of adhesion molecules by abatacept appears to mediate its effects via reduced actin dynamics. The decreased adhesion and migratory capacity of the monocytes is likely responsible for the increase in CD14+ monocytes in the peripheral blood of abatacept-treated RA patients.

One of the key elements of inflammation is the accumulation of cells at sites of inflammation, and monocytes are a major proinflammatory cell population that accumulates at sites such as the RA synovial membrane, contributing to the dramatic amplification of the inflammatory process. A reduction in the migratory capacity of monocytes might contribute importantly to the decrease in inflammation and thereby help to ameliorate the disease (19). Indeed, besides efforts to interfere with lymphocyte activation, modulation of the recruitment of leukocytes to sites of inflammation is a promising evolving concept of targeted therapies for RA (20).

Moreover, putting a brake on monocyte migration may also reduce the number of osteoclast precursors, a cell population that is importantly involved in joint destruction, in the synovial tissue (19, 21). Indeed, it is conceivable that the direct inhibitory effects of CTLA-4Ig on osteoclastogenesis and its benefit on joint damage in a tumor necrosis factor–dependent arthritis model (22) are consequences of the changes described here.

Thus, CTLA-4Ig seems to possess the capacity to interfere with RA pathogenesis at 2 important pathogenetic sites, namely, at the site of the T cells, by inhibiting their activation, and at the site of the monocytes, by interfering with their migration into the joint. Accordingly, CTLA-4Ig appears to target RA pathogenesis in a dual way.

There are several limitations to our study. First, the number of patients assessed was relatively small; however, the results obtained were consistent and were seen ex vivo in RA patients as well as in vitro in healthy controls. Second, these data have not been reported previously despite many attempts to characterize additional pathways of the mode of action of CTLA-4Ig; one reason might be that previous studies on CTLA-4Ig–mediated effects have so far mainly focused on T cells and T cell subsets where no substantial shifts in the distribution were found (23), whereas detailed analysis of monocyte phenotype and function was not previously addressed. Finally, we performed our analyses within the very first weeks of abatacept therapy, and we therefore did not relate them to clinical outcomes. In this context, however, it is of lesser importance whether the effect on monocytes is accompanied by a clinical effect or not, since clinical responses are quite variable, and many patients respond differently to various agents despite a clearcut pharmacologic effect of these drugs, thus highlighting the heterogeneity of the pathways leading to RA.

It is noteworthy that most of the changes in monocyte phenotype and function were already seen 2 weeks after the first infusion of CTLA-4Ig and continued to be seen at week 4 (2 weeks after the second infusion). This indicated that the effect of CTLA-4Ig on monocytes came about quite rapidly and was maintained.

In summary, our data indicate that CTLA-4Ig treatment exerts effects on cells beyond T cells in RA patients. CTLA-4Ig was found to modulate monocytes with regard to phenotype characteristics and their migratory capacity, which might exert another beneficial effect in the treatment of RA.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Scheinecker had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Bonelli, Ferner, Göschl, Blüml, Hladik, Kiener, Byrne, Niederreiter, Bergmann, Smolen, Scheinecker.

Acquisition of data. Bonelli, Ferner, Göschl, Blüml, Hladik, Karonitsch, Niederreiter, Steiner, Rath, Scheinecker.

Analysis and interpretation of data. Bonelli, Ferner, Göschl, Blüml, Hladik, Niederreiter, Steiner, Scheinecker.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

We would like to thank the Cell Sorting Core Unit of the Medical University of Vienna and Mr. Günther Hofbauer for expert flow cytometry assistance, as well as Mrs. Anneliese Nigisch and Karolina von Dalwigk for the preparation and maintenance of cells for the migration and spreading assays.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES