Article Text
PDF
Abstract
Objectives To investigate the role of the interleukin (IL)-33–ST2 axis in the pathophysiology of primary Sjögren's syndrome (pSS).
Methods Serum levels of IL-33 and sST2 were determined by ELISA. The expression of IL-33 and ST2 was investigated in salivary glands (SG) by immunohistochemistry. PBMC were isolated and stimulated with IL-33, IL-12 and IL-23 and the cytokine profile response was examined by flow cytometry. Intracellular cytokine detection of IFNγ and IL-17 was performed by flow cytometry.
Results Serum IL-33 and sST2 levels were increased in pSS patients compared with controls and patients with systemic lupus erythematosus. Expression of IL-33 was upregulated in SG with Chisholm scores of 2 and 3 of pSS patients but comparable with controls for SG with Chisholm score of 4. ST2 expression in SG was downregulated in pSS patients. IL-33 at different concentrations did not increase the secretion of pro-inflammatory cytokines but acted synergistically with IL-12 and IL-23 to promote IFNγ production. NK and NKT cells were identified as main producers of IFNγ in vitro and were found in SG of pSS patients.
Conclusions IL-33 is released in pSS, and acts with IL-12 and IL-23 to favour the secretion of IFNγ by NK and NKT cells.
- Sjögren's Syndrome
- Cytokines
- Autoimmunity
- Inflammation
Statistics from Altmetric.com
Introduction
Sjögren's syndrome (SS) is a chronic autoimmune disease characterised by the lymphocytic infiltration of salivary and lacrimal glands entailing the sicca syndrome of xerostomia and keratoconjunctivitis sicca as well as systemic manifestations.1 Much progress has been made to decipher the pathophysiological mechanisms underlying SS, but the exact trigger of the disease stills remains to be unshrouded. Likely, the conjuncture of a genetic predisposal, environmental insults and hormonal disequilibrium may lead to the activation of the resting epithelium, the upregulation of toll-like receptors, the secretion of damage associated molecular pattern molecules (DAMPs), cell death, apoptosis and ensuing the liberation of pro-inflammatory cytokines further fostering downstream autoimmune inflammation.2 DAMPs and alarmins are present inside the living cells and secreted ensuing cell damage. Following their release (secondary either to necrosis or to defective clearance of apoptotic bodies), the organism predominantly responds by fostering a pro-inflammatory milieu wreaking further tissue damage.3 The potential contribution of alarmins to the pathogenesis of SS remains largely elusive.
Interleukin-33 (IL-33) is a recently discovered member of the IL-1 cytokine family. IL-33 was identified as a ligand for the IL-1 receptor (IL-1R) family member ST2.4 IL-33 is constitutively expressed in the nucleus of endothelial and epithelial cells.5 This cytokine acts as a dual-function protein, similar to IL-1α and high mobility group protein B1, displaying both nuclear and extracellular effects. The primary function of IL-33 is to behave as an alarmin, released, that is, after cell injury.6 Once extracellular, IL-33 binds to its receptor ST2 or ST2L and to its coreceptor IL-1R accessory protein (IL-1RacP) ensuing activation of NFκB and MAPK pathways.4 Due to alternative splicing, the ST2 receptor exists in a soluble form (sST2) that acts as a decoy receptor. IL-33 expression is upregulated in epithelial cells cultured in the presence of pro-inflammatory stimuli such as IL-1β and TNF-α or by activation of the toll-like receptors 3, 4 and 9 pathways.6
The role of IL-33–ST2 axis has been shown in autoimmune diseases such as systemic sclerosis and rheumatoid arthritis.7 ,8 In the present study, we hypothesised that the IL-33–ST2 axis could contribute to the pathogenesis of primary Sjögren's syndrome (pSS).
Patients and methods
Patients
A total of 54 patients diagnosed with pSS according to the revised American–European criteria9 and 27 patients with sicca syndrome have been included in the study. The results were compared with a population of 32 healthy non-smoking volunteers (controls) and 20 patients with systemic lupus erythematosus matched for sex and age. Local ethics committee approved the study and informed consent was obtained from patients and control subjects. Clinical data of patients were gathered including their medical history, current medications and several laboratory parameters.
Cytokine determination by ELISA
IL-3, sST2 and interferon (IFN)γ were measured by ELISA according to manufacturer's protocol (R&D Systems, Minneapolis, Minnesota, USA).
Immunohistochemistry
Paraffin-embedded sections of SG were subjected to immunohistochemistry with antihuman IL-33, ST2, IL-1RacP (R&D Systems), CD56 (Dako, Heverlee, Belgium) and NKG2D (Millipore, Overijse, Belgium) antibodies as previously described.10 ST2 recombinant protein (R&D Systems) was used to validate the specificity of anti-ST2 immunohistochemistry. Labelling of IL-33 and ST2 was quantified using Image J software by calculating the percentage of labelled cells on 6–10 fields.
Cell culture
PBMC from six pSS patients and five healthy controls were isolated and stimulated with either IL-33 alone, IL-33 with IL-12, IL-33 with IL-23, IL-33 with IL-12 and IL-23 to determine its capacity to induce cytokine production (IL-2, IL-4, IL-5, IL-6, IFNγ, IL-9, TNF-α, IL-10, IL-12, IL-13, IL-17, IL-22, IL-23) by flow cytomix (ebioscience, Vienna, Austria).
Flow cytometry
PBMC isolated from pSS and control patients were stained for fluorescence-activated cell sorter analysis (FACS) using anti-CD3, anti-CD4, anti-CD8 and anti-CD56. A FACS calibur instrument was used for gating on viable lymphocytes based on forward and side scatter light patterns. For identification of IFNγ, PE-conjugated anti-IFNγ (BD-biosciences, Erembodegem, Belgium) was used.
Statistical analysis
Significance was assessed by one-way analysis of variance followed by multiple comparisons post hoc test (Bonferroni's method) for multiple comparisons. p Value <0.05 was considered as significant.
Results
In pSS patients, sera levels of both IL-33 and sST2 were significantly increased compared with controls (figure 1A–B). Sera levels of IL-33 and sST2 did not present any significant correlations with the Chisholm score (CS),11 EULAR Sjögren's Syndrome Disease Activity Index12 and any biological parameters (online supplementary table S1).
Serum levels of interleukin (IL)-33 (A) and sST2 (B) in control, sicca, primary Sjögren's syndrome (pSS) and systemic lupus erythematosus (SLE) patients. Determination of serum IL-33 and sST2 was performed by ELISA in patients with pSS (n=54), patients with sicca symptoms but without immunological aberrations and not fulfilling the classification criteria for pSS according to the American–European classification criteria for Sjögren's syndrome (n=27) and healthy controls (n=32). The sera levels of IL-33 and sST2 were significantly increased in pSS patients compared with healthy controls, sicca patients and patients with SLE. IL-33 immunohistochemistry on labial salivary glands from sicca patients (C) and pSS patients (D-E), haematoxylin counterstained (scale bar=50 um). Immunoreactivity for IL-33 is detected in acinar cells (D), in the basal cellular layer of excretory canal (E) as well as on some cell infiltrates in the connective tissue. Representative IL-33 labelling from pSS salivary glands is shown according to Chisholm scoring (F). Quantification of IL-33 labelling on salivary glands sections shows a differential pattern in relation with the Chisholm score, reaching a maximum labelling index for Chisholm 2 (G). ST2 (H) and IL-1RacP (I) immunostaining in the salivary glands from pSS patients similarly also denotes some differential expression according to Chisholm scoring. Bars represent values of mean±SE of the mean. *p<0.05, **p<0.01, ***p<0.001.
IL-33, ST2 and IL-1RacP immunoreactivites were fainted in the acinar cells and ducts of SG from controls. We observed increased expression of IL-33 in SG of pSS group relative to controls (figure 1C–E). The upregulation of IL-33 in SG of pSS was observed in patients with CS of 2 and 3, while patients with CS of 4 had similar IL-33 expression to those with CS of 1 (figure 1F–G) (data representing IL-33 labelling index with the focus score is shown in online supplementary figure S1). ST2 expression in the SG was predominantly observed in the cytoplasmic compartment of the ducts. In pSS patients, we observed a downregulation of the expression of ST2 in the ducts of SG with CS of 3 and 4 (figure 1H). IL-1RacP expression in SG was observed in the basolateral membrane and cytoplasmic compartment of ducts and acini. Similar to that observed for ST2 expression, diminished immunostaining of IL-1RacP was predominantly observed in pSS patients with CS of 3 and 4 (figure 1I).
PBMC stimulation by increasing concentrations of IL-33 alone did not trigger the production of pro-inflammatory cytokines (data not shown). Since pSS is characterised mainly by a Th1 phenotype and IFNγ production, we hypothesised that IL-33 could act synergistically with other pro-inflammatory cytokines to trigger the secretion of IFNγ. Interestingly, we observed a 10-fold increase in IFNγ secretion when PBMC were stimulated with IL-12 and IL-33 compared with stimulation by IL-12 or IL-33 alone (figure 2A) but no significant differences were observed between PBMC from pSS patients and controls. There was also a significant increase in the production of IFNγ following stimulation of PBMC with IL-33 combined with IL-23 but the net effect was significantly less compared with the stimulation by IL-33 and IL-12. Because IL-23 activates Th17 cells to produce IL-17, we determined if the combined effect of IL-33, IL-12 and IL-23 could foster the secretion of IL-17A but no significant increase in IL-17A was observed (data not shown).
Interleukin (IL)-33 acts synergistically with IL-12 on PBMC to synthetise interferon (IFN)γ (A). PBMC were plated for 2 days in culture and stimulated with recombinant IL-12, IL-23 and IL-33 alone or in combinations. IFNγ secretion was determined by ELISA. Values represent mean±SE of the mean from three independent experiments. *p<0.05, **p<0.01. Intracytoplasmic detection of IFNγ (C) showing that the main producers of IFNγ following stimulation with IL-33/IL-12/IL-23 are the CD56+ population encompassing NK and NKT cells (B), the CD3+CD4−CD8− γδ cells (C) and the CD3+CD16+CD56+ NKT cells (D). CD3+CD16−CD56− cells (E), CD8+ and CD4+ cells (F) produced low levels of IFNγ. Immunostaining of NKG2D shows the presence of γδ, NK and NKT cells in salivary glands from primary Sjögren's syndrome (pSS) (H), while absent in control subjects (G). CD56+ (NK and NKT) cells were also found around acini (arrows, I) and in lesser amounts in inflammatory infiltrates (arrowheads, J) from pSS salivary glands.
To determine which cell type was involved in IFNγ secretion after stimulation of PBMC with IL-33, IL-12 and IL-23, intracytoplasmic detection of IFNγ was performed. NK (CD3− CD16+ CD56+), NKT (CD3+ CD16+ CD56+) and γδ cells (CD3+ CD4− CD8−) produced up to 34.69% of IFNγ intracytoplasmic total amount (figure 2B–E). Production of IFNγ by CD8+ T cells and CD4+ T cells was merely detectable (less than 1.06%) (figure 2F). Compared with controls (figure 2G), SG from pSS patients expressed a higher number of NKG2D+ cells (figure 2H) and CD56+ cells, mostly located around acini (figure 2I–J).
Discussion
In this study, we demonstrate that expression of IL-33 is increased in pSS compared with control groups. The increased expression of IL-33 in SG was highest at CS 2 and 3 thereafter decreasing with CS of 4.
A fundamental concept of the IL-33–ST2 axis is the balance between cytokine and the competing receptor antagonist influencing cytokine bioavailability.13 The sST2 protein is reported to possess anti-inflammatory properties such as the blockade of IL-1R, but more importantly serves as a decoy receptor in forestalling the effects of IL-33 signalling through the ST2L.14 The significantly increased serum levels of sST2 in pSS group relative to control groups is expected as a counter-regulatory measure to dampen the effects of IL-33.
We also investigated the effects of recombinant IL-33 on the cytokine profile expression but there were no significant alterations in the cytokine profile probably due to the overwhelming effects of the costimulation by anti-CD3/CD28 such that a potential modification of cytokine profile could not be detected. We next assessed if IL-33 could act synergistically with other cytokines to trigger an inflammatory response. Our experiments indicate that IL-33 acts in a concerted fashion with IL-12 as well as with IL-23 to induce the secretion of IFNγ. Since IL-23 activates Th17 cells to produce IL-17, we also determined if the synergistic actions of IL-33 and IL-23 could promote the production of IL-17.15 The combined effect of IL-33 and IL-23 did not result in significant production of IL-17A that could be translated by the predominant IFNγ secretion inhibiting the activation of Th17 cells.16 Intracellular cytokine detection of IFNγ revealed that NK cells and NKT cells are the main IFNγ producers. These cells, found in abundance within the salivary parenchyma of pSS patients, might play a pivotal role.10 ,17 One fundamental question that still remains to be answered is the link between the increased IL-33–ST2 axis and NK/NKT cells in the pathogenesis of pSS. It could be hypothesised that IL-33 together with IL-12/IL23 could activate NK/NKT cells to produce IFNγ thereby perpetuating cellular damage.
In conclusion, in pSS patients IL-33 expression is upregulated, acting in concerted fashion with IL-12 and IL-23 to trigger the secretion of IFNγ by NK and NKT cells.
Acknowledgments
The authors wish to thank Drs M Svoboda and C Truyens for their useful advice.
References
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:
- Data supplement 1 - Online supplement
Footnotes
Handling editor Tore K Kvien
AA and CN contributed equally to the work.
-
Contributors AA and CN designed and performed experiments, analysed data and revised the manuscript. SE performed experiments, analysed data and revised the manuscript. LS designed experiments and analysed data. JR and VG analysed data and revised the manuscript. IP performed experiments and analysed data. MSS designed experiments, analysed data and wrote the manuscript.
-
Competing interests None.
-
Patient consent Obtained.
-
Ethics approval Hôpital Erasme Local Ethics committee.
-
Provenance and peer review Not commissioned; externally peer reviewed.
-
Data sharing statement All the authors in this current work declare that there are no unpublished data and that all data have been presented in the current manuscript.