You are here

Article Text

Article menu
Basic and translational research
Extended report
NKG2D stimulated T-cell autoreactivity in giant cell arteritis and polymyalgia rheumatica
  1. Christian Dejaco1,2,
  2. Christina Duftner1,3,
  3. Juman Al-Massad4,
  4. Annette D Wagner4,
  5. Joon-Keun Park4,
  6. Johannes Fessler2,
  7. Ariane Aigelsreiter5,
  8. Franz Hafner6,
  9. Sandra Vega7,
  10. William Sterlacci8,9,
  11. Beatrix Grubeck-Loebenstein7,
  12. Alexander Tzankov8,10,
  13. Thomas Ness11,
  14. Luigi Boiardi12,
  15. Carlo Salvarani12,
  16. Michael Schirmer1
  1. 1Department of Internal Medicine, Innsbruck Medical University, Innsbruck, Austria
  2. 2Department of Rheumatology and Immunology, Medical University Graz, Graz, Austria
  3. 3Department of Internal Medicine, Hospital of Kufstein, Kufstein, Austria
  4. 4Department of Internal Medicine, Division of Nephrology, Hannover Medical School, Hannover, Germany
  5. 5Institute of Pathology, Medical University Graz, Graz, Austria
  6. 6Department of Angiology, Medical University Graz, Graz, Austria
  7. 7Institute for Biomedical Aging Research, University of Innsbruck, Innsbruck, Austria
  8. 8Institute of Pathology, Innsbruck Medical University, Innsbruck, Austria
  9. 9Institute of Pathology, Academic Teaching Hospital Feldkirch, Feldkirch, Austria
  10. 10Institute of Pathology, University Hospital Basel, Basel, Switzerland
  11. 11University Clinic for Ophthalmology, Freiburg, Germany
  12. 12Rheumatology Unit, Department of Internal Medicine, Azienda Ospedaliera ASMN, Istituto di Ricovero e Cura a Carattere Scientifico, Reggio Emilia, Italy
  1. Correspondence to Professor Michael Schirmer, Department of Internal Medicine, Innsbruck Medical University, Anichstrasse 35, Innsbruck A-6020, Austria; michael.schirmer{at}i-med.ac.at

Abstract

Objective To investigate functional expression of NKG2D on CD4 and CD8 T-cells in patients with giant cell arteritis (GCA) and polymyalgia rheumatica (PMR).

Methods Peripheral blood was drawn from patients with GCA (n=16), PMR (n=78) and healthy controls (HC, n=64). Tissue samples were obtained from GCA patients and controls. Proliferation and cytokine production assays were performed using CFSE and intracellular IFN-γ or TNF-α staining, respectively, and flow cytometry analysis. Immunofluorescence and immunohistology were applied to analyse the presence of NKG2D-expressing T-cells and NKG2D-ligands in temporal arteries, respectively. mRNA levels of NKG2D-ligands were determined by RT-PCR.

Results In both GCA and PMR patients, NKG2D was preferentially expressed on senescent CD4CD28− and CD8CD28−, as well as on CD8CD28 T-cells. Frequencies of senescent T-cells were increased in GCA and PMR patients compared to HC. In GCA tissue samples, infiltrating T-cells were predominately CD28−. NKG2D expressing T-cells concentrated around the vasa vasorum of the adventitia. Antigenic stimulation induced rapid up-regulation of NKG2D on CD4CD28− and CD4CD28 T-cells, whereas TNF-α and interleukin-15 enhanced NKG2D expression on senescent CD4 and CD8 T-cells only. NKG2D cross-linkage augmented anti-CD3 triggered proliferation, IFN-γ and TNF-α production of CD8 T-cells. In CD4CD28− T-cells, NKG2D ligation resulted in increased IFN-γ production only. NKG2D ligands were expressed in temporal arteries from GCA patients, particularly in the adventitial and medial layers of affected vessels.

Conclusions NKG2D is functionally expressed on CD4CD28− and CD8 T-cells in GCA and PMR. NKG2D-ligands are present in temporal arteries and may co-stimulate NKG2D expressing T-cells.

  • Polymyalgia Rheumatica
  • Giant Cell Arteritis
  • T Cells

https://doi.org/10.1136/annrheumdis-2012-201660

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Giant cell arteritis (GCA) and polymyalgia rheumatica (PMR) are closely related clinical conditions that are currently understood as different states of the same disease.1–3 Both GCA and PMR are a result of an inadequate immune reaction against an unknown auto-antigen present in the adventitia of large vessels.3 In GCA, T-cells enter the adventitial layer via the vasa vasorum and produce large amounts of pro-inflammatory cytokines, including interferon-γ (IFN-γ).1 ,3 IFN-γ activates macrophages to release metalloproteinases (MMPs), disrupting the media and the internal elastic lamina.1 ,3 In PMR, arterial inflammation is restricted to the adventitia accompanied by an intense systemic immune reaction.4 The specific T-cell subsets contributing to inflammation in GCA and PMR as well as the micro-milieu regulating T-cell activity are still unexplored.

    We and others described a subset of senescent CD4 T-cells accumulating in patients with chronic inflammatory diseases. These T-cells are characterised by the loss of the co-stimulatory molecule CD28, are autoreactive and apoptosis resistant, and reveal up-regulated alternative (co-)stimulatory receptors.5–10

    One of these alternative receptors is NKG2D, a C-type lectin-like activating natural killer (NK)-cell receptor usually expressed on NK-cells and NK/T-cells. NKG2D interacts with major histocompatibility class I related chain (MIC)A, MICB and cytomegalovirus UL-16 binding proteins (ULBPs) 1–3 that are usually up-regulated on malignant transformation, oxidative stress, and viral or bacterial infections.11–17

    In the present study we investigated functional expression of NKG2D on CD4 and CD8 T-cell subsets in GCA/PMR. The following observations suggest a possible relevance of this pathway for the pathogenesis of the disease: (1) The prevalence of circulating CD4CD28− T-cells is increased in GCA/PMR, and in rheumatoid arthritis (RA), NKG2D was functionally expressed on CD28− T-cells.15 (2) NKG2D ligands are up-regulated on viral or bacterial infections and chronic infections have been suspected to trigger GCA/PMR for decades. In some studies, microbial antigens were detected in temporal arteries from GCA patients.18 (3) MHC class I chain-related gene A (MICA) polymorphisms were associated with the occurrence of GCA and Takayasu's arteritis in previous studies.19

    Methods

    For a detailed description of the materials and methods, see online supplementary file 1.

    Patients, cell preparation and short-term cell lines

    After written informed consent was obtained, heparinised blood samples were drawn from peripheral veins of consecutive GCA (n=16) and PMR (n=78) patients, diagnosed according to the American College of Rheumatology or Bird's classification criteria, respectively, as well as healthy controls (HC, n=64).20 ,21 Patients and controls were comparable regarding age and sex. Clinical data were retrieved by chart review and were available in 11 GCA and 67 PMR patients (see table 1 for clinical characteristics). Analysis of NKG2D expression and correlations with clinical data were performed in this subgroup only. Remission or active disease of GCA and PMR patients was retrospectively determined by a rheumatologist (MS, blinded to experimental data) based on documented history and clinical testing as well as acute phase reactants. The study protocol was approved by the institutional review board of Innsbruck Medical University.

    View this table:
    Table 1

    Patients’ characteristics

    Peripheral blood mononuclear cells (PBMCs) were isolated using magnetic bead labelled anti-CD4 and anti-CD8 monoclonal antibodies (mAbs), and short-term cell lines were established as previously described.5 ,6

    NKG2D regulation, cytokine production and proliferation assays

    CD4 and CD8 T-cells were stimulated with anti-CD3 (3 μg/ml), interleukin (IL)-15 (20 ng/ml), tumour necrosis factor-α (TNF-a; 20 ng/ml), IL-18 (100 ng/ml) or IFN-γ (10 ng/ml). NKG2D expression (quantified by the mean fluorescence intensity (MFI) using flow cytometry) was tested before and at various time points after anti-CD3 (6, 12, 24 h) or cytokine (24, 48, 72 h) exposure.

    For proliferation and cytokine production assays, CD4 and CD8 T-cells were stimulated with OKT3±anti-NKG2D or control immunoglobulin. Stimulation with anti-NKG2D mAb was shown to have comparable effects to MICA tetramers in functional assays.22

    For proliferation assays, T-cells were labelled with the fluorescent dye 5-(and -6)-carboxyfluorescein diacetate succinimidyl ester (CFSE) and the percentage of proliferating cells was determined by flow cytometry.23 For cytokine assays, cells were permeabilised and intracellularly stained with anti-IFN-γ, anti-TNF-α, anti-IL-4, anti-IL-10 mAbs or control immunoglobulin. Surface staining was done with CD4, CD8 or CD28 mAbs. The MFI was used to quantify the stimulatory effects of OKT3±anti-NKG2D on cytokine production, because the MFI reflecting the proportion of additional cytokine producing cells as well as enhanced expression in pre-stimulated cells is more sensitive to change than measuring the proportion of new cytokine positive cells only.

    Immunohistochemistry, RT-PCR and ELISA for NKG2D ligands

    For immunohistochemistry, we retrospectively identified nine paraffin-embedded cases of histologically confirmed temporal arteritis, four cases with GCA (physician's final diagnosis, no formal criteria applied) and negative temporal artery histology as well as seven cases without GCA.

    To compare mRNA expression of different NKG2D ligands (MICA, MICB and ULBP1–3) in patients with temporal arteritis, we performed  real time (RT)-PCR using fresh-frozen temporal artery specimens from five patients with histologically confirmed GCA.

    Soluble levels of MICA were tested in blinded sera in duplicates with an ELISA kit (R&D Systems) according to the manufacturer's instructions.

    Immunofluorescence analysis

    For immunofluorescence analysis of NKG2D, CD3, CD4, CD8 and CD28 in temporal arteries, biopsies from three (NKG2D/CD3 double stainings) or five (other stainings) independent individuals with histologically confirmed GCA and five controls without GCA were used. A standard protocol was applied.

    Statistical analysis

    Statistical analysis was performed using SPSS V.15.0. The χ2 test, the Student t test, the Mann–Whitney U test or the Wilcoxon test were performed as appropriate.

    Results

    Circulating CD4CD28− and CD8 T-cells express NKG2D in GCA and PMR patients

    To explore the expression of NKG2D on the cell surface of circulating T-cell subsets in GCA and PMR, flow cytometry studies were performed. In the CD4 T-cell subset, NKG2D was more frequently expressed on CD28− T-cells compared to CD28 T-cells (GCA: median 5.7% (range 0–59.1) vs 0.8% (0.1–4.2), p=0.021; PMR: 15.9% (0–81.7) vs 1.5% (0.3–13.1), p<0.001) whereas almost all CD8 T-cells expressed NKG2D (GCA: CD8CD28− 98.3% (91.3–99.4) vs CD8CD28 97.5% (92.3–99.1), NS; PMR: 97.2% (78.5–99.1) vs 95.8% (78.0–100), NS) (figure 1A).

    Figure 1
    Figure 1

    Expression of NKG2D on circulating CD4CD28− and CD8 T-cells from giant cell arteritis (GCA)/polymyalgia rheumatica (PMR) patients and presence of NKG2D+ T-cells in temporal artery biopsies. (A) Surface expression of NKG2D (% positive cells out of CD28 or CD28− subsets as indicated) on T-cell populations in PMR and GCA patients. (B) Prevalence of circulating CD3CD4CD28− and CD3CD8CD28− T-cells in healthy controls, PMR and GCA patients (percentage out of all CD3CD4 T-cells). Data were analysed by flow cytometry and are shown as box-plots. The Mann–Whitney U and Wilcoxon tests were used as appropriate; **p<0.01; ***p<0.001. (C) Representative two-colour immunofluorescence staining of a paraffin-embedded temporal artery biopsy from a patient with histologically confirmed GCA. Cells expressing NKG2D (green), CD3 (red) or both (yellow) are shown. Original magnification ×40.

    NKG2D was more commonly found on CD8CD28 T-cells from GCA and PMR patients than from HC (91.0% (84.7–97.9), p<0.05), whereas on CD8CD28− and CD4 T-cell subsets similar expression levels were observed in patients and controls.

    NKG2D expression was increased on CD4CD28− T-cells from GCA/PMR patients (pooled analysis) with active versus inactive disease, whereas no difference was found regarding CD4CD28 T-cells and CD8 T-cell subsets. NKG2D expression was independent from disease duration; however, patients on high-dose corticosteroid treatment showed higher NKG2D expression on CD8CD28 T-cells than did those on low-dose steroids (see online supplementary figure S1).

    The frequencies of CD3CD4CD28− and CD3CD8CD28− T-cells were increased in patients with GCA (4.5% out of CD4 T-cells (range 0.2–30.6), p<0.01 and 40.6% out of CD8 T-cells (5.6–75.3), p<0.01, respectively) and PMR (2.1% (0–28.4), p<0.01 and 40.4%(8.4–79.8), p<0.01) compared to HC (0.7% (0–2.9) and 19.8% (7.6–54.9)) (figure 1B). The prevalence of circulating CD3CD4CD28− and CD3CD8CD28− T-cells did not differ between GCA and PMR patients. No association was found between disease activity, disease duration, prednisone dose and prevalence of CD28− T-cells in GCA and PMR (data not shown).

    NKG2D expressing T-cells concentrate around adventitial vasa vasorum in GCA patients

    Immunofluorescence analysis indicated the presence of NKG2D expressing CD3 T-cells in the adventitia and media of temporal arteries from GCA patients as shown in figure 1C. NKG2D expressing T-cells particularly concentrated around the vasa vasorum in the adventitial layer (see online supplementary table S1 for quantification of CD3NKG2D+ cells). Most T-cells infiltrating temporal arteries from GCA patients were CD4; ≥99% of CD4 and CD8 T-cells were CD28− as shown in online supplementary figures S2 and S3. In comparison, temporal artery specimens from controls revealed no T-cells within the vessel wall.

    NKG2D expression is up-regulated by antigenic stimulation and pro-inflammatory cytokines

    To test whether T-cell antigen receptor complex stimulation leads to up-regulation of NKG2D, isolated circulating CD4 and CD8 T-cells were stimulated with cross-linked anti-CD3. Up-regulation of NKG2D on the cell surface of CD4CD28− T-cells was detected already after 6 h of stimulation and the highest NKG2D levels were found after 12 h (figure 2A). CD4CD28 T-cells showed a slight, short-lasting up-regulation of NKG2D after 6 h. Expression of NKG2D on CD8CD28− and CD8CD28 T-cells was not influenced by anti-CD3 stimulation (data not shown).

    Figure 2
    Figure 2

    Up-regulation of NKG2D by T-cell antigen receptor complex or cytokine stimulation in giant cell arteritis/polymyalgia rheumatica. Purified CD4 or CD8 T-cells were stimulated with plate-bound anti-CD3 or tumour necrosis factor-α (TNF-α) for different periods. (A) shows the stimulatory effect of anti-CD3 on CD4CD28− and CD4CD28 T-cells and (B) indicates up-regulation of NKG2D on CD4 and CD8 T-cell subsets after exposure to TNF-α. Box-plots indicate mean fluorescence intensity of NKG2D expression of each T-cell subset. Differences were tested for statistical significance with the Wilcoxon test. *p<0.05; AU, arbitrary units; h, hours; ns, not significant.

    Cytokine stimulation assays were performed using TNF-α (figure 2B), IL-15 (data not shown), IL-18 (data not shown) and IFN-γ (data not shown). Stimulation of T-cells with TNF-α or IL-15 led to an up-regulation of NKG2D on CD4CD28− but not CD4CD28 T-cells. Besides, TNF-α or IL-15 induced up-regulation of NKG2D on CD8CD28− T-cells whereas only stimulation with TNF-α  promoted NKG2D expression of CD8CD28 T-cells. The kinetics of NKG2D up-regulation was much slower by cytokine compared to anti-CD3 stimulation: NKG2D expression reached the highest levels after 24–48 h of cytokine exposure. IL-18 and IFN-γ did not influence NKG2D expression levels on CD4 or CD8 T-cells.

    NKG2D co-stimulates pro-inflammatory activity of CD4CD28− and CD8 T-cells

    To test whether NKG2D expressed on CD4 and CD8 T-cells is functional, proliferation and cytokine production assays were performed.

    Whereas NKG2D stimulation alone did not induce cell division of CD8CD28− or CD8CD28 T-cells as indicated by the CFSE assay (median percentage of cells with at least one cell division: 3.7 (range 0.8–7.0) and 1.5 (1.0–9.5), respectively), NKG2D cross-linkage augmented anti-CD3 mediated proliferation of CD8CD28− T-cells (50.0% (36.7–86.5) vs 37.0% (15.6–58.9) with anti-CD3 only; p<0.05) and CD8CD28 T-cells (69.9 (51.2–83.1) vs 40.6 (27.3–75.3); p<0.05) (figure 3). In the CD4 T-cell subset, anti-CD3 induced proliferation was not further increased by NKG2D (not shown).

    Figure 3
    Figure 3

    NKG2D co-stimulates proliferation of CD8 T-cells in giant cell arteritis/polymyalgia rheumatica. Purified CD8 T-cells were stimulated with plate-bound anti-CD3 and anti-NKG2D as indicated. After 24 h cells were harvested and proliferation was assessed by 5-(and -6)-carboxyfluorescein diacetate succinimidyl ester (CFSE) staining and flow cytometry. (A) Representative dot plot and histograms showing dilution of the fluorescent dye CFSE (filled red curves) after stimulation compared to staining without stimulation (black line). Gates were set on lymphocytes (forward and sideward scatter) as well as on CD8CD28 and CD8CD28− cells. (B) Box-plots indicate the percentage (%) of CD8CD28− and CD8CD28 T-cells with at least one cell division. Differences were tested for statistical significance with the Wilcoxon test. *p<0.05.

    In cytokine production assays intracellular accumulation of Th1-type (IFN-γ and TNF-α) and Th2-type cytokines (IL-4 and IL-10) was tested by flow cytometry. NKG2D cross-linkage augmented anti-CD3 induced IFN-γ production of CD8CD28− T-cells (median MFI 34.2 (range 15.4–159.5) vs 21.5 (16.8–104.3) with CD3 only; p<0.05), CD8CD28 T-cells (36.7 (10.9–135.4) vs 21.4 (12.2–93.0); p<0.05) and CD4CD28− T-cells (53.9 (14.0–190.6) vs 42.3 (11.7–170.6); p<0.05) (figure 4).

    Figure 4
    Figure 4

    NKG2D co-stimulates interferon-γ production of CD4CD28−, CD8CD28− and CD8CD28 T-cells in giant cell arteritis/polymyalgia rheumatica. Purified CD4 and CD8 T-cells were stimulated with plate-bound anti-CD3 and anti-NKG2D as indicated. After 24 h cells were harvested and interferon-γ (IFN-γ) production was assessed by intracellular cytokine staining and flow cytometry. Box-plots indicate mean fluorescence intensity of IFN-γ production of (A) CD8 CD28− or CD28 and (B) CD4 CD28− or CD28 T-cells. Differences were tested for statistical significance with the Wilcoxon test. *p<0.05.

    TNF-α production was increased by anti-CD3 plus anti-NKG2D compared to anti-CD3 alone in CD8CD28− T-cells (median MFI 11.9 (range 7.6–57.9) vs 9.6 (5.8–31.3); p<0.05) and CD8CD28 T-cells (9.8 (5.9–11.6) vs 7.7 (5.2–11.0); p<0.05 each) whereas the co-stimulatory effect was absent in the CD4 T-cell subset (data not shown). Cross-linking NKG2D did not influence IL-4 and IL-10 production in any CD4 or CD8 T-cell subset (data not shown).

    Expression of NKG2D ligands in temporal arteritis

    Immunohistochemistry studies indicated a higher MICA expression in temporal artery samples from patients with histologically confirmed temporal arteritis compared to patients with clinical GCA but negative histology in patients without GCA. Representative images are depicted in figure 5. The most notable finding was a higher MICA staining in the adventitial and medial layers of cases with positive versus negative histology. Moderate MICA expression of endothelial cells of intima and vasa vasorum was observed in most tissue specimens regardless of inflammation (see online supplementary table S2 for details). Inflammatory cells (lymphocytes and giant cells) of patients with temporal arteritis also stained positively for MICA.

    Figure 5
    Figure 5

    Expression of NKG2D ligands in temporal arteritis. Representative images of immunohistological detection of major histocompatibility class I chain-related gene A (MICA) in paraffin-embedded temporal artery specimens from patients with histologically confirmed temporal arteritis (A–C), patients with a clinical diagnosis of giant cell arteritis (GCA) but negative histology (D–F) and patients without GCA (G–I). (A,B) Artery section overview (4× and 10× magnifications). The intima and adventitia show moderate MICA staining whereas the smooth muscle cells of the media display a strong staining pattern. Endothelial cells of the intima and vasa vasorum also stain positively for MICA. (C) Detail view of medial layer (40× magnification). The inflammatory infiltrate (small lymphocytes and giant cells) as well as smooth muscle cells stain positively for MICA. (D,E) Artery section overview (4× and 10× magnifications). The intimal and adventitial layers show weak MICA staining whereas smooth muscle cells of media and endothelial cells of vasa vasorum display a moderate staining pattern. No inflammatory infiltrates are present. (F) Detail view of intimal, medial and adventitial layers (40× magnification). (G,H) Artery section overview (4× and 10× magnifications). (F) Detail view of intimal, medial and adventitial layers (40× magnification). The MICA staining pattern is comparable to (D–F).

    RT-PCR was performed to compare the expression of NKG2D ligands MICA, MICB and ULBP1–3 in temporal arteritis. As depicted in online supplementary figure S4, median MICA mRNA concentrations were 5.0–20.2-fold higher compared to the other ligands tested. In addition, ULBP1–3 mRNAs were inconstantly expressed (no detectable mRNA in 2/5 (ULBP1) and 1/5 (ULBP2 and ULBP3) temporal artery samples).

    Serum levels of soluble MICA in GCA, PMR and HC

    Levels of soluble MICA were tested in blinded serum samples from GCA (n=10) and PMR patients (n=70) as well as HC (n=38). The serum concentrations of soluble MICA were similar in GCA (36.3 pg/ml; range 0–341), PMR (77.3; 0–3293.3) and HC (93.2; 0–3214.4). Soluble MICA concentrations were negatively correlated with the prednisone dose (corrcoeff −0.47, p=0.002), but were not associated with disease activity or disease duration.

    Discussion

    Our results indicate that NKG2D receptor-expressing CD4CD28− and CD8 T-cells contribute to the pathogenesis of GCA and PMR. We found NKG2D expressing T-cells and their corresponding ligands in the vessel wall of GCA patients and detected circulating NKG2D+ T-cells in peripheral blood of patients with GCA and PMR. NKG2D on T-cells was functional, promoting the production of pro-inflammatory cytokines known to be involved in the pathogenesis of these diseases.

    NKG2D+ T-cells concentrated around the vasa vasorum of the adventitial layer in GCA patients, suggesting that these cells enter from peripheral blood to the site of inflammation. Alternatively, they mature locally, up-regulating NKG2D. The site of primary T-cell stimulation is located around the vasa vasora24 ,25 and we showed in our in vitro experiments that antigenic stimulation promotes NKG2D expression of CD4 T-cells. We were unable to perform CD3–CD4/8-NKG2D–CD28 multiple stains of tissue sections allowing a detailed characterisation of NKG2D expressing T-cells; however, our sequential and double stainings (see online supplementary figure S3) reveal a predominance of CD4CD28− T-cells within the inflammatory aggregates around the vasa vasora, whereas CD8CD28− T-cells occur less commonly. CD28 T-cells are rarely present in GCA lesions. Extrapolating the data from peripheral blood analysis, we hypothesise that most NKG2D+ T-cells present in temporal arteries of GCA patients are CD4CD28−. In contrast, most peripheral NKG2D+ T-cells are CD8, because almost all CD8 but only senescent CD4 T-cells express NKG2D.26

    The data from our stimulation assays indicate different functional roles of NKG2D for CD4 and CD8 T-cell subsets in GCA/PMR. NKG2D cross-linkage promoted proliferation of CD28CD8 and CD28−CD8 T-cells, whereas this effect was not observed in CD4CD28− T-cells. Besides, NKG2D augmented TNF-α production of CD8 T-cell subsets only, whereas IFN-γ production was stimulated in CD8CD28−, CD8CD28 and CD4CD28− T-cells. Our in vitro data partially contrast previous observations in RA, in which cell division as well as IFN-γ and TNF-α production of CD4CD28− T-cells was enhanced by NKG2D ligation.15 Methodical issues may account for the divergences: ELISA detection of cytokines, the (3H)thymidine proliferation assay and the utilisation of isolated CD4CD28−NKG2D+ T-cells in the previous study might have been more sensitive than our approaches. The different effects of NKG2D on IFN-γ and TNF-α production by CD4CD28− T-cells in GCA/PMR versus RA patients, however, cannot be explained by technical reasons only. Given that NKG2D+CD4CD28− T-cells (and the corresponding NKG2D ligands) are present in the adventitia of GCA lesions, these cells may preferentially secrete IFN-γ known to promote vessel wall destruction in temporal arteritis.3 In contrast, the predominant cytokine in the RA synovium is TNF-α released by CD4CD28− T-cells after NKG2D stimulation.15 ,27

    NKG2D expression of circulating CD4CD28− T-cells was higher in GCA/PMR patients with active versus inactive disease but was not influenced by disease duration or therapy. The clinical relevance of NKG2D up-regulation on CD8CD28 T-cells by corticosteroids is probably small given the high expression of this receptor on CD8CD28 T-cells in general.

    Our in vitro data indicate a preferential up-regulation of NKG2D on CD4CD28− T-cells by T-cell receptor stimulation as well as TNF-α and IL-15, supporting the observed clinical association between disease activity and NKG2D expression on this subset. IL-18 and IFN-γ had no effect in vitro, which is consistent with previous reports on NK-cells.28 Cross-linkage of CD3 resulted in a more rapid, but shorter lasting up-regulation of NKG2D compared to stimulation with cytokines. The different kinetics were explained by rapid NKG2D surface expression due to redistribution of intracellular proteins by anti-CD3 compared to delayed receptor appearance by induction of NKG2D mRNA by cytokines.15 CD8 T-cells may contribute to the regulation of NKG2D surface expression because they produce large amounts of TNF-α once they are activated by T-cell receptor plus NKG2D stimulation.

    The presence and role of NKG2D ligands in vascular lesions of GCA patients has not been investigated so far: we found higher expression of MICA in temporal arteritis specimens compared to probes with normal histology, suggesting that NKG2D ligands may perpetuate the autoimmune response in temporal arteritis. MICA appears to be the most relevant NKG2D ligand as indicated by RT-PCR results. Using immunohistochemistry, MICA was predominately detected in the adventitia and media of inflamed vessels, whereas in Takayasu's arteritis MICA was preferentially expressed on vascular smooth muscle cells.29 Interestingly, lymphocytes and giant cells also positively stained for MICA, suggesting that inflammatory cells might co-stimulate NKG2D expressing T-cells in a cell to cell contact dependent manner. Previous studies have already demonstrated an up-regulation of MICA on peripheral lymphocytes on T-cell receptor activation.30 ,31 Another notable observation is the expression of MICA on endothelial and smooth muscle cells in the majority of samples (even in those without inflammation). Apart from stress-induced up-regulation of MICA due to the surgical procedure, our results could explain the high susceptibility of temporal arteries to inflammation. Future studies (eg, using animal models of GCA) should be performed to clarify these issues.

    Serum levels of soluble MICA did not differ between GCA, PMR and HC and were much lower compared to levels previously reported in RA.15 This finding is surprising because matrix MMPs are known to induce shedding of NKG2D ligands32 ,33 and MMPs are also involved in the pathogenesis of GCA, leading to the fragmentation of the internal elastic lamina.3 ,25 This observation could be explained by the fact that most patients were treated with corticosteroids. We found a negative correlation between the prednisone dose and MICA concentrations, and corticosteroids are known to reduce MMP activity.34 ,35 In addition, genetic variants of MICA associated with GCA, PMR, RA, other autoimmune diseases and carcinoma might have different shedding susceptibilities to MMPs.36–42

    In summary, our data indicate that the activating NK-cell receptor NKG2D is functionally expressed on circulating CD4CD28− and CD8 T-cells in patients with GCA and PMR. NKG2D ligands in temporal arteries may co-stimulate NKG2D expressing T-cells in GCA and perpetuate the chronic inflammatory process.

    Acknowledgments

    Samples and/or data used in this project have been provided by Biobank Graz.

    References

    1. ↵
      1. Salvarani C,
      2. Cantini F,
      3. Boiardi L,
      4. et al
      . Polymyalgia rheumatica. Best Pract Res Clin Rheumatol 2004;18:705–22.
      OpenUrlCrossRefPubMed
    2. ↵
      1. Blockmans D,
      2. De Ceuninck L,
      3. Vanderschueren S,
      4. et al
      . Repetitive 18-fluorodeoxyglucose positron emission tomography in isolated polymyalgia rheumatica: a prospective study in 35 patients. Rheumatology 2007;46:672–7.
      OpenUrlAbstract/FREE Full Text
    3. ↵
      1. Weyand CM,
      2. Goronzy JJ
      . Giant-cell arteritis and polymyalgia rheumatica. Ann Intern Med 2003;139:505–15.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Chatelain D,
      2. Duhaut P,
      3. Loire R,
      4. et al
      . Small-vessel vasculitis surrounding an uninflamed temporal artery: a new diagnostic criterion for polymyalgia rheumatica? Arthritis Rheum 2008;58:2565–73.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Duftner C,
      2. Dejaco C,
      3. Kullich W,
      4. et al
      . Preferential type 1 chemokine receptors and cytokine production of CD28- T cells in ankylosing spondylitis. Ann Rheum Dis 2006;65:647–53.
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Duftner C,
      2. Goldberger C,
      3. Falkenbach A,
      4. et al
      . Prevalence, clinical relevance and characterization of circulating cytotoxic CD4+CD28- T cells in ankylosing spondylitis. Arthritis Res Ther 2003;5:R292–300.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Lamprecht P,
      2. Moosig F,
      3. Csernok E,
      4. et al
      . CD28 negative T cells are enriched in granulomatous lesions of the respiratory tract in Wegener's granulomatosis. Thorax 2001;56:751–7.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Namekawa T,
      2. Snyder MR,
      3. Yen JH,
      4. et al
      . Killer cell activating receptors function as costimulatory molecules on CD4+CD28null T cells clonally expanded in rheumatoid arthritis. J Immunol 2000;165:1138–45.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Vallejo AN,
      2. Schirmer M,
      3. Weyand CM,
      4. et al
      . Clonality and longevity of CD4+CD28null T cells are associated with defects in apoptotic pathways. J Immunol 2000;165:6301–7.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Schirmer M,
      2. Vallejo AN,
      3. Weyand CM,
      4. et al
      . Resistance to apoptosis and elevated expression of Bcl-2 in clonally expanded CD4+CD28- T cells from rheumatoid arthritis patients. J Immunol 1998;161:1018–25.
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Groh V,
      2. Bahram S,
      3. Bauer S,
      4. et al
      . Cell stress-regulated human major histocompatibility complex class I gene expressed in gastrointestinal epithelium. Proc Natl Acad Sci USA 1996;93:12445–50.
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Groh V,
      2. Steinle A,
      3. Bauer S,
      4. et al
      . Recognition of stress-induced MHC molecules by intestinal epithelial gammadelta T cells. Science 1998;279:1737–40.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Groh V,
      2. Rhinehart R,
      3. Secrist H,
      4. et al
      . Broad tumor-associated expression and recognition by tumor-derived gamma delta T cells of MICA and MICB. Proc Natl Acad Sci USA 1999;96:6879–84.
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Groh V,
      2. Rhinehart R,
      3. Randolph-Habecker J,
      4. et al
      . Costimulation of CD8alphabeta T cells by NKG2D via engagement by MIC induced on virus-infected cells. Nat Immunol 2001;2:255–60.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Groh V,
      2. Bruhl A,
      3. El-Gabalawy H,
      4. et al
      . Stimulation of T cell autoreactivity by anomalous expression of NKG2D and its MIC ligands in rheumatoid arthritis. Proc Natl Acad Sci USA 2003;100:9452–7.
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Das H,
      2. Groh V,
      3. Kuijl C,
      4. et al
      . MICA engagement by human Vgamma2Vdelta2 T cells enhances their antigen-dependent effector function. Immunity 2001;15:83–93.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Tieng V,
      2. Le Bouguenec C,
      3. du Merle L,
      4. et al
      . Binding of Escherichia coli adhesin AfaE to CD55 triggers cell-surface expression of the MHC class I-related molecule MICA. Proc Natl Acad Sci USA 2002;99:2977–82.
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Wagner AD,
      2. Gerard HC,
      3. Fresemann T,
      4. et al
      . Detection of Chlamydia pneumoniae in giant cell vasculitis and correlation with the topographic arrangement of tissue-infiltrating dendritic cells. Arthritis Rheum 2000;43:1543–51.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Kimura A,
      2. Kobayashi Y,
      3. Takahashi M,
      4. et al
      . MICA gene polymorphism in Takayasu's arteritis and Buerger's disease. Int J Cardiol 1998;66(Suppl 1):S107–13.
      OpenUrlCrossRefPubMed
    20. ↵
      1. Hunder GG,
      2. Bloch DA,
      3. Michel BA,
      4. et al
      . The American College of Rheumatology 1990 criteria for the classification of giant cell arteritis. Arthritis Rheum 1990;33:1122–8.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Bird HA,
      2. Esselinckx W,
      3. Dixon AS,
      4. et al
      . An evaluation of criteria for polymyalgia rheumatica. Ann Rheum Dis 1979;38:434–9.
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. Maasho K,
      2. Opoku-Anane J,
      3. Marusina AI,
      4. et al
      . NKG2D is a costimulatory receptor for human naive CD8+ T cells. J Immunol 2005;174:4480–4.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Herndler-Brandstetter D,
      2. Schwaiger S,
      3. Veel E,
      4. et al
      . CD25-expressing CD8+ T cells are potent memory cells in old age. J Immunol 2005;175:1566–74.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Restuccia G,
      2. Cavazza A,
      3. Boiardi L,
      4. et al
      . Small-vessel vasculitis surrounding an uninflamed temporal artery and isolated vasa vasorum vasculitis of the temporal artery: two subsets of giant cell arteritis. Arthritis Rheum 2012;64:549–56.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Weyand CM,
      2. Goronzy JJ
      . Pathogenic principles in giant cell arteritis. Int J Cardiol 2000;75(Suppl 1):S9–15.
      OpenUrlCrossRefPubMed
    26. ↵
      1. Champsaur M,
      2. Lanier LL
      . Effect of NKG2D ligand expression on host immune responses. Immunol Rev 2010;235:267–85.
      OpenUrlPubMedWeb of Science
    27. ↵
      1. Scott DL,
      2. Wolfe F,
      3. Huizinga TW
      . Rheumatoid arthritis. Lancet 2010;376:1094–108.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Zhang C,
      2. Zhang J,
      3. Sun R,
      4. et al
      . Opposing effect of IFNgamma and IFNalpha on expression of NKG2 receptors: negative regulation of IFNgamma on NK cells. Int Immunopharmacol 2005;5:1057–67.
      OpenUrlCrossRefPubMed
    29. ↵
      1. Seko Y,
      2. Sugishita K,
      3. Sato O,
      4. et al
      . Expression of costimulatory molecules (4-1BBL and Fas) and major histocompatibility class I chain-related A (MICA) in aortic tissue with Takayasu's arteritis. J Vasc Res 2004;41:84–90.
      OpenUrlCrossRefPubMed
    30. ↵
      1. Cerwenka A,
      2. O'Callaghan CA,
      3. Hamerman JA,
      4. et al
      . Cutting edge: the minor histocompatibility antigen H60 peptide interacts with both H-2Kb and NKG2D. J Immunol 2002;168:3131–4.
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Molinero LL,
      2. Fuertes MB,
      3. Girart MV,
      4. et al
      . NF-kappa B regulates expression of the MHC class I-related chain A gene in activated T lymphocytes. J Immunol 2004;173:5583–90.
      OpenUrlAbstract/FREE Full Text
    32. ↵
      1. Groh V,
      2. Wu J,
      3. Yee C,
      4. et al
      . Tumour-derived soluble MIC ligands impair expression of NKG2D and T-cell activation. Nature 2002;419:734–8.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Salih HR,
      2. Rammensee HG,
      3. Steinle A
      . Cutting edge: down-regulation of MICA on human tumors by proteolytic shedding. J Immunol 2002;169:4098–102.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Tchetverikov I,
      2. Lard LR,
      3. DeGroot J,
      4. et al
      . Matrix metalloproteinases-3, -8, -9 as markers of disease activity and joint damage progression in early rheumatoid arthritis. Ann Rheum Dis 2003;62:1094–9.
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Matsuyama A,
      2. Sakai N,
      3. Ishigami M,
      4. et al
      . Matrix metalloproteinases as novel disease markers in Takayasu arteritis. Circulation 2003;108:1469–73.
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Gonzalez-Gay MA,
      2. Rueda B,
      3. Vilchez JR,
      4. et al
      . Contribution of MHC class I region to genetic susceptibility for giant cell arteritis. Rheumatology 2007;46:431–4.
      OpenUrlAbstract/FREE Full Text
    37. ↵
      1. Gambelunghe G,
      2. Ghaderi M,
      3. Tortoioli C,
      4. et al
      . Two distinct MICA gene markers discriminate major autoimmune diabetes types. J Clin Endocrinol Metab 2001;86:3754–60.
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Gupta M,
      2. Nikitina-Zake L,
      3. Zarghami M,
      4. et al
      . Association between the transmembrane region polymorphism of MHC class I chain related gene-A and type 1 diabetes mellitus in Sweden. Hum Immunol 2003;64:553–61.
      OpenUrlCrossRefPubMedWeb of Science
    39. ↵
      1. Jumnainsong A,
      2. Romphruk AV,
      3. Jearanaikoon P,
      4. et al
      . Association of polymorphic extracellular domains of MICA with cervical cancer in northeastern Thai population. Tissue Antigens 2007;69:326–33.
      OpenUrlCrossRefPubMedWeb of Science
    40. ↵
      1. Martinez A,
      2. Fernandez-Arquero M,
      3. Balsa A,
      4. et al
      . Primary association of a MICA allele with protection against rheumatoid arthritis. Arthritis Rheum 2001; 44:1261–5.
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      1. Reinders J,
      2. Rozemuller EH,
      3. van der Ven KJ,
      4. et al
      . MHC class I chain-related gene a diversity in head and neck squamous cell carcinoma. Hum Immunol 2006;67:196–203.
      OpenUrlCrossRefPubMed
    42. ↵
      1. Chen JR,
      2. Lee YJ,
      3. Chen T,
      4. et al
      . MHC class I chain-related gene A (MICA) polymorphism and the different histological types of cervical cancer. Neoplasma 2005;52:369–73.
      OpenUrlPubMedWeb of Science

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • Handling editor Tore K Kvien

    • C Dejaco and C Duftner contributed equally.

    • Contributors All authors contributed to conception and design of the study, acquisition and/or interpretation of data, drafting the article or revising it critically for important intellectual content and approved the final version of the manuscript. All authors had full access to all of the data (including statistical reports and tables) in the study and can take responsibility for the integrity of the data and the accuracy of the data analysis.

    • Funding This study was supported by the Tyrolean Research Fund, Innsbruck (#0404/121 to C Duftner), Innsbruck Medical University, Innsbruck as well as the ‘Verein zur Förderung der Hämatologie, Onkologie und Immunologie’, Innsbruck, Austria.

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval Institutional Review board of Innsbruck Medical University.

    • Provenance and peer review Not commissioned; externally peer reviewed.