Article Text
Abstract
Objective To identify interleukin (IL)-7Rα expression in the labial salivary gland (LSG) of patients with primary Sjögren's syndrome (pSS) and non-Sjögren's syndrome sicca (nSS-sicca) and to study its correlation with glandular inflammation and IL-7 expression.
Methods The presence of infiltrating immune cells and IL-7Rα cells in inflamed LSG of patients with pSS (n=12) and nSS-sicca controls (n=7) was studied by immunohistochemistry and fluorescence activated cell sorting analysis upon tissue digestion (n=15 and n=13, respectively). Additionally, the correlations of IL-7Rα cells with hallmark disease parameters of pSS, major infiltrating inflammatory cells and IL-7 were assessed.
Results In the LSG of patients with pSS increased numbers of IL-7Rα cells were found as compared with nSS-sicca patients. IL7Rα cells strongly correlated with the lymphocytic focus score, IL-7 expression, the decrease in percentage of IgA plasma cells and numbers of CD3 T cells, CD20 B cells, and CD1a and CD208 myeloid dendritic cells. Analysis of isolated cells from the LSG demonstrated strongly increased percentages of IL-7Rα CD3 T cells in pSS as compared with nSS, showing abundant IL-7Rα expression on both CD4 and CD8 T cells. Other CD45 leucocytes and CD45- tissue cells scarcely expressed IL-7Rα. Percentages of IL-7Rα T cells also significantly correlated with glandular inflammation.
Conclusions This study shows the presence of increased IL-7Rα T cells in the LSG of patients with pSS and their association with the severity of sialadenitis, disease parameters and IL-7 expression. Considering the immunostimulatory ability of IL-7Rα T cells and IL-7, this suggests that IL-7(R)-dependent T cell-driven immune activation plays an important role in inflammation in pSS.
Statistics from Altmetric.com
Introduction
Primary Sjögren's syndrome (pSS) is a systemic autoimmune disease characterised by lymphocytic infiltration of the exocrine (mainly salivary and lachrymal) glands, resulting in clinical symptoms of dryness. The lymphocytic infiltration consists of large numbers of activated T cells and B cells, together with smaller numbers of specialised antigen-presenting cells, like monocytes/macrophages and dendritic cells.1 It has been suggested that T cell-driven sialadenitis contributes to glandular damage and dysfunction associated with sicca complaints. The close relationship of major histocompatibility complex (MHC) class II molecules (HLA-DR/DQ) with the presence of autoantibodies in pSS suggests a crucial involvement of CD4 T cells in the immunopathology of pSS since these T cells need antigen presentation by MHC class II molecules.2 In addition, T helper cells secreting interferon γ (IFNγ) and interleukin (IL)-17 have been demonstrated in the inflamed salivary glands and are associated with tissue damage.3 4 Furthermore, autoreactive CD4 T cells induce SS-like symptoms, when injected in mice.5
The vast majority of peripheral blood (PB) CD4 T cells express the interleukin 7 (IL-7) receptor (IL-7R), which, through binding of IL-7 are potently activated. IL-7 signals through the IL-7R, which consists of the high-affinity α chain (IL-7Rα, CD127) in conjunction with a shared common γ chain. IL-7 is a pleiotropic immunostimulatory cytokine known for its crucial role in maintaining homoeostasis of the peripheral T cell pool. Activation of the IL-7/IL-7R pathway leads to survival, proliferation and differentiation of mature naive and memory T cells.6
Disturbance of the IL-7/IL-7R pathway has been implicated in the pathogenesis of several rheumatic autoimmune diseases. IL-7 levels are increased in patients with rheumatoid arthritis (RA) as compared with healthy controls (HCs).7 8 Associated with increased IL-7 levels, IL-7Rα cells are found to be increased in the synovial tissue of patients with RA in comparison with patients with osteoarthritis.9 Furthermore, in experimental arthritis models, IL-7 administration to collagen-immunised mice increased T cell activation, associated with a strong aggravation of joint inflammation and joint destruction.10
Blockade of the IL-7R and/or IL-7 leads to a reduction of arthritis severity in different arthritis models as demonstrated by reduced joint damage and decrease in proinflammatory cytokines and tissue-destructive mediators.12 13 In addition to RA, elevated expression levels of IL-7 and/or IL-7R mRNA were also found in the synovial fluid and tissue of patients with spondyloarthritis and serum and skin of patients with psoriasis.13 14
Recently, data were presented indicating that IL-7 can also contribute to the chronic inflammation in pSS. IL-7 expression is increased in glandular tissue of patients with pSS and correlates with increased markers of glandular inflammation. Furthermore, IL-7R-mediated in vitro activation of mononuclear cells from these patients by IL-7 is associated with production of T cell-attracting chemokines, T cell-differentiating cytokines and cytokines, indicative of T cell activation (IFNγ, IL-17).15 We and others have shown that IL-7Rα-expressing CD4 T cells strongly proliferate upon T cell receptor (TCR) activation and promote inflammatory activity in vitro, while CD4 T cells lacking the IL-7Rα and expressing CD25 are anergic and are regulatory in nature.16 17 At the level of the salivary gland an increased number of IL-7Rα T cells could lead to an increased and continuing inflammatory response, especially in the presence of an increased local IL-7 expression. Therefore, this study was designed to determine the expression of IL-7Rα on cells in labial salivary gland (LSG) of patients with pSS in relation to markers of inflammation and disease.
Patients and methods
Patients
LSG specimens of 12 patients with pSS were used for immunohistochemical staining. All patients were diagnosed according to American–European Consensus Group criteria.18 Seven patients not fulfilling these criteria for Sjögren's syndrome, were classified as patients with non-Sjögren's sicca (nSS-sicca), and were included as a control group for immunohistochemistry. All of the above-mentioned patients were naive to the use of immunosuppressive drugs. Demographic and clinical data of pSS and patients with nSS-sicca are shown in table 1.
Flow cytometry analysis was performed using LSG tissue biopsy specimen from an additional group of 15 patients with pSS and 13 control patients with nSS-sicca. Two patients with pSS used a daily low maintenance dose of prednisone (5 mg a day) at the time of biopsy; all other patients were naive to the use of immunosuppressive drugs. Demographic data and disease parameters of this group are also shown in table 1.
The study was performed according to the medical ethical regulations of the University Medical Centre Utrecht.
Immunohistochemistry of LSG tissue
From each LSG specimen, 6 µm sections were cut on a freezing microtome and mounted in a serial manner on glass slides (three sections a slide; Star Frost adhesive slides, Knittergläser, Braunschweig, Germany). Sections were fixed with acetone, air-dried and washed with phosphate-buffered saline. Different tissue sections on a glass slide were separately stained using a liquid blocker (PAP pen), keeping staining reagents localised on individual tissue sections. Endogenous peroxidase activity was blocked, and non-specific binding was prevented by incubating the tissue with 5% pooled-human AB serum and 10% goat serum. Serial tissue sections were incubated overnight respectively with monoclonal mouse anti-human IL-7Rα (eBioRDR5, eBioscience, San Diego, California, USA), mouse anti-human CD3 (BD Bioscience, San Jose, California, USA) and polyclonal rabbit-anti-human IL-7 (H151, Santa Cruz biotechnology, Santa Cruz, California, USA) on the first slide. Additionally, parallel slides were stained with a mouse anti-human CD19 (BD), mouse anti-human anti-Ki67 (DAKO, Glostrup, Denmark) as a marker for proliferating cells. Infiltrating dendritic cells were visualised with mouse anti-CD1a (OKT6, own production), mouse anti-human CD208 (Immunotech, Marseille, France) and mouse anti-human CD209 (RD, Minneapolis, Minnesota, USA). A mouse IgG1κ was used as isotype control (R&Dsystems, Minneapolis, Minnesota, USA) and control stainings were performed omitting the first antibody. Antigen–antibody complexes were visualised using a two-step peroxidase staining method (power vision-peroxidase-3,3′-diaminobenzidine for isotype control, IL-7Rα, CD3, CD19, IL-7, and power vision-peroxidase-new fuchsine for Ki67) or a two-step phosphatase staining method19 (AB complex-phosphatase-new fuchsine for CD1a, CD208, CD209), both according to standard protocols. The numbers of IL-7Rα and Ki67 positive cells were calculated per mm2. For the other stainings a semiquantitative scoring system was used to grade the intensity of expression, using a scale of 0–4, with 0 indicating no positive cells, 1 only a few positive cells, 2 considerable numbers of positive cells, 3 widespread expression and 4 abundant expression of positive cells. Scoring of subsets of cells and IL-7 expression was performed fully blinded to expression of the IL-7Rα.
LSG tissue digestion and flow cytometry
LSG tissue specimens were cut into small pieces and digested in 2 mg/ml collagenase type IV (Worthington, Lakewood, New Jersey, USA) supplemented with 10% fetal calf serum (Gibco BRL) (to inhibit possible trypsin activity) for 3 h at 37°C. LSG cell suspensions were then pushed through a 70 µm cell strainer to remove remaining tissue debris and used for flow cytometry. The following labelled antibodies were used: anti-CD45-APC (BD, Erembodegem-Aalst, Belgium), anti-CD3-FITC (Pharmingen, San Diego, CA, USA), anti-CD4-PERCP(Biolegend, San Diego, CA, USA), anti-CD8-PE (BD), anti-CD14-FITC (DAKO), anti-CD19-FITC (DAKO), anti-CD127-PE (Immunotech, Marseilles, France) and anti-CD25-FITC (DAKO). Anti-IgG1-FITC/IgG1-PE antibody (Immunotech) was used as isotype control. When appropriate, the mean fluorescence intensity was calculated from the fluorescence histogram and expressed in arbitrary units.
Additionally, peripheral blood mononuclear cells (PBMCs) from seven patients with pSS and age- and gender matched HCs were stained for IL-7Rα and CD25 as well. Simultaneously, intracellular FoxP3 staining was performed using an APC-conjugated rat antihuman FoxP3 staining set (eBioscience, San Diego, USA). As a control, an APC-labelled rat isotype control antibody was used (eBioscience).
Cell counting was done using a FACScan flow cytometer and data were analysed with FlowJo software, version 7.5 (Tree Star Inc, Oregon, USA).
CD4 T cell and CD14 monocyte (co-)culture and cytokine measurements
PBMCs were isolated by Ficoll–Paque density gradient centrifugation (Pharmacia, Uppsala, Sweden). CD14 and CD4 cells were isolated from PBMCs by magnetic-activated cell sorting (Miltenyi Biotec, Utrecht, The Netherlands).
Isolated CD4 T cells and CD14 monocytes/macrophages (both 5×105/ml) were co-cultured for 3 days at a ratio 1:1 in RPMI/glutamax (Gibco BRL, Life Technologies, Merelbeke, Belgium) with added penicillin (100 U/ml), streptomycin (100 µg/ml) and 10% pooled fetal calf serum in the presence or absence of 10 ng/ml IL-7 (PeproTech Inc, Rocky Hill, New Jersey, USA). Measurements of cytokine production (IFNγ (Biosource, Belgium) and IL-17A (R&D)) after short α−CD3/CD28 restimulation (18 h) were performed by commercially available ELISAs, according to the manufacturer's instructions.
Statistical analysis
The Mann–Whitney U test or independent sample t test was performed, where appropriate, to detect differences between patients with pSS and nSS-sicca or HCs. Values of p<0.05 were considered statistically significant. Spearman correlation was used to evaluate correlations between LSG IL-7Rα expression, expression of different inflammatory cells, and parameters of disease. Values of p<n 0.0048 were considered statistically significant. This p value (p<0.05/12) was calculated after application of a Bonferroni correction for multiple comparisons (n=12).
Results
Numbers of IL-7R cells are increased in pSS and correlate with markers of inflammation and glandular inflammatory cells.
LSG biopsy samples obtained from 12 patients with pSS and 7 nSS-sicca controls were analysed for the expression of IL-7Rα using immunohistochemistry. Since IL-7Rα is known to be expressed mainly by CD3 positive T cells, the co-localisation of CD3 and IL-7Rα was examined, together with proliferating (KI67-expressing) cells. Figure 1 provides representative photomicrographs of isotype control, CD3, IL-7Rα and Ki67 stainings in the LSG of a patient with pSS (figures 1A–D, respectively) and a nSS-sicca control (figures 1E–H, respectively). The increased IL-7Rα expression in patients with pSS was most prominently detected in the T cell-rich lymphocytic infiltrates throughout the tissue, and co-localised with CD3- and KI67-expressing cells.
In the pSS group, the number of IL-7Rα positive cells/mm2 had a median value of 357 (range 85–1640), compared with 85 (range 27–157) in the nSS-sicca group (p<0.01 figure 2A).In the whole sicca group, a significant positive correlation was found with the number of CD3 T cells (p<0.001, r=0.809, figure 2B). Additionally, LSG IL-7Rα expression strongly correlated with local disease parameters in the whole sicca patient group (lymphocytic focus score (LFS), p<0.001, r=0.720, figure 2D, table 2; % IgA plasma cells, p<0.005, r=−0.639, table 2). No significant correlation was found between the LSG IL-7Rα expression and peripheral inflammatory disease parameters—that is, sIgG, erythrocyte sedimentation rate and circulating leucocyte numbers (table 2).
Additionally, it was assessed whether the presence of CD19 B cells, CD1a, CD208 and CD209 myeloid dendritic cells (mDCs), as well as IL-7 expression, and KI67 proliferating cells were increased and whether these cells correlated with the number of IL-7Rα cells (table 2). The numbers of almost all inflammatory cells (all at least p<0.01; except CD209 mDCs, p=0.105), IL-7-expressing cells and KI67 proliferating cells (both p<0.001) were significantly higher in patients with pSS than in nSS-sicca controls (see van Blokland et al20 and data not shown). Significant positive correlations were found between numbers of IL-7Rα+ cells and the infiltrating and proliferating immune cells (table 2).
Isolated glandular T cells abundantly express the IL-7R and correlate with inflammation
The expression of the IL-7Rα (CD127) on cells isolated from labial biopsy specimen from 15 patients with pSS and 13 with nSS-sicca analysed by flow cytometry is shown in figure 3, demonstrating IL-7Rα expression mainly on T cells from the LSG. Analysis of the IL-7Rα expression on CD45− tissue cells, CD19 B cells and CD14 monocytes showed that, on average, less than 2%, 6% and 7%, respectively, of these cells express the IL-7Rα (representative FACS analysis and histograms from one patient with pSS; figure 3A). The IL-7Rα was expressed by a large percentage of CD3 T cells (nSS vs pSS; 67.5±4.5% vs 53.9±3.9%, NS), in particular on CD4 T cells and to a lesser extent on CD8 T cells (figure 3D). Supporting our immunohistochemistry data, an increased percentage of IL-7Rα+ CD3 T cells (expressed as percentage of the total number of tissue cells) was found for patients with pSS as compared with nSS-sicca controls (figure 3B). A similar difference was found for the percentage of IL-7Rα+ CD4 T cells and IL-7Rα+ CD8 T cells (both p<0.05, data not shown). In addition, a significant correlation was found between the LFS and the isolated and FACS-identified percentage of IL-7Rα+ CD3 T cells (p<0.05, r=0.516, figure 3C), and similarly for the percentage of IL-7Rα-expressing CD4 (p<0.01, r=0.425) and CD8 T cells (p<0.01, r=0.558) (data not shown).
In contrast to CD8 T cells, the percentage of IL-7Rα-expressing cells within the CD4 T population was reduced in patients with pSS as compared with patients with nSS-sicca (55.8±3.0% vs 74.1±3.5%, respectively, p<0.05, figure 3D). We and others have previously demonstrated that in the PB of patients with RA (and in HCs) regulatory T cells, which express CD25 (IL-2R), are characterised by a high FoxP3 expression and a lack of IL-7Rα expression. Therefore, it was analysed whether the reduction in the percentage of IL-7Rα+ cells was associated with increases in the percentage of IL-7Rα− CD25+ T cells. Indeed, the IL-7Rα− CD25+ subset was significantly increased in LSG biopsy specimens of patients with pSS as compared with patients with nSS-sicca (7.4±1.2%; n=7 vs 1.7±0.7%, n=5, p<0.05; figure 4A,B). The IL-7Rα− CD25− subset was also increased in patients with pSS as compared with nSS-sicca controls, although this difference did not reach statistical significance (nSS vs pSS; 23.1±5.6% vs 32.3±3.7%, p=0.20). Associated with this increase, the percentage IL-7Rα+ CD25− effector T cells showed a significant decrease (nSS vs pSS; 64.5±4.5% vs 46.6±6.1%, p<0.01).
Next, we assessed whether the observed changes in IL-7Rα CD25-defined T cell subsets in the LSG were reflected in the PB of patients with pSS as compared with controls. IL-7Rα CD25-defined CD4 T cell subsets from the PB of patients with pSS and HC were analysed (n=7), and were age- and gender-matched to the patients with pSS from the LSG group. Similar to the LSG, a significant increase in the IL-7Rα− CD25+ T cell subset was found for patients with pSS as compared with HC (3.0±0.6% vs 1.3%±0.2%, p<0.05, figure 4C). Furthermore, the IL-7Rα− CD25− subset was also found to be significantly increased in patients with pSS (HC vs pSS; 0.8±0.2% vs 5.0±2.4%, p<0.01). The FoxP3 expression in these subsets was simultaneously analysed and demonstrated that nearly all IL-7Rα− CD25+ CD4 T cells were FoxP3 positive (pSS vs HC; 94.8±0.8% vs 94.7±2.8%, figure 4D). No significant difference in the FoxP3 expression of these FoxP3brightIL-7Rα− CD25+ T cells between patients with pSS and HC was seen. However, IL-7Rα+ CD25− T cells showed an increased percentage of FoxP3dim-expressing cells in patients with pSS (HC vs pSS; 1.8±0.2% vs 4.2±0.5%, p<0.001).
To assess the possible consequences of the increased percentage of FoxP3bright IL-7Rα− CD25+ Treg population, CD4 T cells were cultured together with CD14 monocytes as antigen-presenting cells and stimulated with IL-7. IL-7 induced production of the two major Th1 and Th17-defining cytokines, IFNγ (HC vs pSS; 1785±352 pg/ml vs 2506±871 pg/ml) and IL-17A (683±215 pg/ml vs 461±188 pg/ml), equally effective in HCs and patients with pSS.
Discussion
This study shows an increased IL-7Rα expression in the LSG of patients with pSS as compared with nSS-sicca controls. Furthermore, it showed that the glandular IL-7Rα cells are mainly CD3 T cells, and that they strongly correlate with B cells, mDCs and proliferating cells as well as with IL-7 expression and glandular markers of inflammation (LFS, %IgA cells).
Previously, we demonstrated an increased expression of IL-7 in the LSG.15 Here we show that the number of IL-7Rα-expressing cells correlates with IL-7 expression, supporting the hypothesis that IL-7-driven immune activation is involved in the immunopathology of pSS. Further support for the role of IL-7R-driven immune activation is given by the association between the increased number of IL-7Rα+ cells and the increased LFS. In addition, the increased number of IL-7Rα cells correlates with decreased percentages of IgA plasma cells, which is a result of an increase in local IgM- and IgG-secreting plasma cells.21 Recently, we demonstrated that IL-7 can induce T cell-driven B cell activation in vitro, indicating that IL-7 might contribute to local T cell-dependent B cell activation in patients with pSS. Furthermore, this IL-7(R)-induced B cell activation was enhanced in the presence of myeloid antigen-presenting cells (Bikker et al, unpublished). In line with this observation, the increased number of IL-7Rα+ T cells also significantly correlated with the increased number of activated mDCls expressing DC-LAMP (CD208) or DC-SIGN (CD209), possibly contributing to enhanced IL-7-driven T and B cell activation.22
As demonstrated in our previous study IL-7 is expressed throughout the LSG, largely produced by fibroblast-like cells. The increased IL-7Rα expression was found co-localised with CD3 T cells in the lymphocytic infiltrates, and as confirmed by FACS analysis, mainly expressed on CD3 T cells, in particular on CD4 T cells, but also on CD8 T cells. Only minimal IL-7Rα expression was detected on CD19 B cells and CD14 macrophages/monocytes, and tissue cells lacking CD45 expression. From in vitro work by others and by us it is known that IL-7Rα+ CD4 T cells are highly proliferating cells upon TCR stimulation and IL-7-induced activation.23 24 In line with previous studies, we here show that this IL-7R-mediated T cell activation by IL-7 in vitro strongly induces Th1 (IFNγ) and Th17 (IL-17) cytokine secretion by T cells from patients with pSS, suggesting that overexpression of IL-7 in the exocrine glands may contribute to the Th1/Th17-driven immunopathology in pSS.23 Recently, it has been reported that the IL-7/IL-7Rα pathway has an important role in promoting Th1 cell survival and also Th17 cell survival and in vivo expansion in an experimental model of autoimmune encephalomyelitis.25 Furthermore, blockade of the IL-7/IL-7Rα pathway in vivo in experimental models for autoimmune arthritis reduced arthritis severity, associated with a reduction in proinflammatory cytokines and chemokines.11 12 26 Although functional analyses of T cell properties from the LSG are impeded by the small numbers of LSG cells, these above-mentioned studies suggest that the glandular IL-7Rα-expressing T cells in pSS are immunostimulatory and promote inflammation, especially in the presence of overexpressed IL-7.
In this study, we also found a significant increase in the percentage of IL-7Rα+ cells within the CD4 T cell population of patients with pSS. A mechanism that might account for the lower IL-7Rα expression on CD4 T cells is the transient downregulation of the IL-7R upon TCR triggering and/or binding of several cytokines, like IL-6, IL-15, and IL-7 itself.16 In addition, the reduction in IL-7R T cells is related to significant increases in the IL-7Rα− CD25 subset. Unfortunately, owing to the limited numbers of IL-7Rα− CD25+ CD4 T cells obtained from the LSG biopsy specimens and owing to technical limitations we were not able to simultaneously stain for FoxP3. However, two other studies that looked at FoxP3 expression in the LSG of patients with pSS using immunohistochemistry, found that the number of FoxP3+ T cells correlated with the grade of glandular infiltration (LFS).27 28 Also, in this study, we were able to demonstrate that the observed increases in IL-7Rα− CD25+ CD4 T cells in the LSG are reflected in the PB. Similar to HCs (and patients with RA) the majority of this subpopulation of CD4 T cells in patients with pSS expresses FoxP3 at a high level.16 23 29 An elevated expression of IL7Rα− CD25+ Tregs at the site of inflammation might be an attempt to control the continuing inflammation. Apparently, however, these Tregs seem to fail at this, which might be owing to the proinflammatory environment that provides signals to overcome suppression by Tregs or causes Tregs to lose their regulatory function. These proinflammatory signals, resulting from co-stimulation, such as CD28, or cytokines, such as IL-7, IL-1β and IL-6, might render effector T cells unresponsive to Treg-mediated suppression.23 29 In line with this, we demonstrate that IL-7 stimulation in vitro strongly induces IFNγ and IL-17A production by CD4 T cells from patients with pSS and thereby overcomes suppression by increased numbers of IL-7Rα− CD25+ (FoxP3bright) T cells.
Altogether, this study shows an increased presence of IL-7Rα+ T cells in the LSG of patients with pSS and their association with the severity of sialadenitis and disease parameters. The in vitro proinflammatory capacities of IL-7 and beneficial effects of IL-7/IL-7R blockade in several experimental autoimmune models suggests that inhibition of the IL-7/IL-7R pathway to suppress the continuing immune activation in pSS might be a potential therapeutic strategy.
Acknowledgments
The authors thank C G van Helden-Meeuwsen for technical support.
References
Footnotes
-
Funding Dutch Arthitis Association.
-
Competing interests None.
-
Provenance and peer review Not commissioned; externally peer reviewed.