Increased frequency of circulating Th22 in addition to Th17 and Th2 lymphocytes in systemic sclerosis: association with interstitial lung disease

The peripheral blood of 33 consecutive SSc individuals that satisfied the criteria by LeRoy et al. [30] and who were not receiving disease-modifying drugs, cytokine blocking reagents or immunosuppressant agents were prospectively recruited when presenting at the Rheumatology A Unit of the Cochin hospital during a nine- month period. Their clinical characteristics are detailed in Table 1. ILD was identified by the presence of typical features on high-resolution computerized tomography (HRCT) of the chest and confirmed by total lung capacity (TLC) lower than 80% of the predicted value. Pulmonary artery hypertension (PAH) was suggested by an echocardiographic systolic pulmonary arterial pressure > 40 mmHg, or a DLCO < 50% predicted in the absence of pulmonary fibrosis or unexplained dyspnea and confirmed by right heart catheterization and was defined as a mean resting pulmonary artery pressure > 25 mmHg in the presence of a pulmonary capillary wedge pressure £15 mmHg at right heart catheterization. Skin scores were not available and cutaneous involvement was expressed as limited (lSSc) or diffuse (dSSc) [30]. None of the recruited individuals was under treatment with immunosuppressant agents at the time of blood sampling. Peripheral blood from 29 age- and sex-matched HD was provided by the Blood Transfusion Center (Geneva University Hospital, Switzerland) (Table 1). This study was approved by the ethical committees of the institutions involved and was conducted according to the Declaration of Helsinki. Written informed consent was obtained from all individuals.

Anti-CD3 (clone OKT3) monoclonal antibody (mAb) was obtained from American Tissue Culture Collection (Manassas, VA, USA). Anti-CD4-APC-Cy7 (clone IVT114), anti-CD45-RA-FITC (clone HI100), anti-CCR6-PerCP-Cy5.5 (clone 11A9), anti-CCR4-PE-Cy7 (clone 1G1), anti-CXCR3-APC (clone 1C6/CXCR3), anti-CD161-APC (clone DX12) and anti-CD28 (clone CD28.2) mAbs came from BD Biosciences (San Jose, CA, USA); anti-IL-4-APC (clone 8D4-8 anti-IFN-γ-PE-Cy7 (clone 4S.B3) and anti-IL-17A-FITC (clone BL168) mAbs were from Biolegend (San Diego, CA, USA); anti-IL-22-PE (clone 142928), anti-CCR10-PE (clone 314305) came from R&D Systems (Abingdon, UK). The Cytofix/Cytoperm fixation/permeabilization solution kit was from Becton Dickinson (San Diego, CA, USA), Ficoll-Paque Plus from GE Healthcare (Uppsala, Sweden), RPMI 1640, phosphate buffered saline (PBS), glutamine, penicillin, streptomycin, trypsin, and fetal calf serum (FCS) from Gibco (Paisley, UK), and phorbol myristate acetate (PMA), β-mercaptoethanol and brefeldin A from Sigma (St. Louis, MO, USA). Human T-activator CD3/CD28 beads were obtained from Invitrogen (Oslo, Norway) and rhIL-2 from Biogen (Cambridge, MA, USA).

PBMC were cryopreserved in liquid nitrogen until use, then thawed and maintained at 37°C for 16 h in RPMI 1640 supplemented with 1% nonessential amino acids, 1% L-glutamine, 1% sodium pyruvate, 50 U/ml penicillin, 50 μg/ml streptomycin, 5% pooled human AB serum, 5% FCS and 50 μM β-mercaptoethanol (complete medium) before use. For intracellular cytokine determination, PBMC were either cultured for 24 hours or 7 days upon activation by CD3/CD28 cross-linking. In a 24-hour assay, brefeldin A was added for the last 20 hours. In 7-day cultures IL-2 (20 U/ml) was added at 48 hours and the cells were re-stimulated for FACS analysis with PMA/ionomycin for the last 4.5 hours, in the presence of brefeldin A.

Surface staining was performed using fluorochrome-conjugated anti-CD4, anti-CD45RA, anti-CD161, anti-CCR6, anti-CCR4, anti-CXCR3 and anti-CCR10. In intracellular cytokine determination, the cells were stained with anti-CD4 mAb, fixed and incubated with anti-IL-17A, IL-22, IFNγ and IL-4 mAbs using a BD Cytofix/Cytoperm kit according to the manufacturer's instruction. FACS analysis was performed on FACSCanto flow cytometer using FACSDiva (Becton Dickinson) and FlowJo softwares (Tree Star Inc. Ashland, OR, USA). Irrelevant isotype-matched control mAb were used to determine specific staining.

All populations satisfied the Kolmogorov-Smirnov normality test according to GraphPad Prism version 4.00 (Graphpad Software, La Jolla, CA, USA). The significant difference between samples was computed using the Student's t test and correlation between variables using the Pearson correlation coefficient. A P-value < 0.05 was considered statistically significant. Box plots were automatically generated using GraphPad. The box represents values between 25^th and 75^th percentile with a line at the median (50^th percentile). The whiskers extend above and below the box to show the values at the10^th and 90^th percentiles.

Several observations support the hypothesis that T cells are responsible for the altered phenotype of fibroblast and endothelial cells, which ultimately leads to fibrosis in SSc individuals [4, 5, 31]. We aimed at assessing whether in the peripheral blood of a cohort of individuals affected by SSc not receiving disease-modifying drugs, cytokine blocking reagents or immunosuppressant agents, we could detect a peculiar Th cell subset profile when compared to age- and sex-matched HD. To assess the functional repertoire of Th cells, PBMC from individuals affected by SSc and from HD were activated by CD3/CD28 crosslinking overnight. We then detected the intracellular production of IL-17A, IL-22, IFN-γ and IL-4 in the CD4+ T cell fraction by multiparametric flow cytometry. Confirming previous data [32, 33], the frequency of CD4+ T cells was higher in SSc than in HD (data not shown). SSc individuals had significantly increased numbers of CD4+ T cells producing IL-4 in their lymphocyte population compared to controls as previously shown [34], while having similar numbers of IFN-γ producing cells (Figure 1A, B). In addition, we observed an increased number of IL-17A and, for the first time, of IL-22 producing cells in SSc (Figure 1B). Interestingly, subgroup analysis showed that IL-17A-producing cells were increased in both lSSc and dSSc, while the increase in IL-22-producing cells was observed in dSSc only (Figure 1B). These data indicate that SSc individuals compared to HD have an imbalance in their Th cells populations.

Since Th1, Th17 and Th22 cells can all produce IL-22 [35], we investigated whether in SSc individuals IL-22 was produced by the same or by distinct Th subsets. To better discriminate these low frequency subsets, PBMC were expanded in vitro for seven days before intracellular cytokine determination. Consistently with the results obtained with overnight cultures, the frequency of cells producing IL-22 was higher in SSc compared to HD after seven-day culture. This was also true for IL-17A (Figure 2A, B). Among all possible cytokine combinations, SSc individuals had a distinct increased frequency of IL-22 single positive (IL-17A-IL-22+IFNγ-IL4-cells) which identify the Th22 subset, and of IL-17A+IL-22+ double positive cells characteristic of Th17 cells. In addition, IL-17A single positive cells were also increased in SSc compared to HD (Figure 2A, C). To substantiate such findings, we tested whether the frequency of CD4+ T cells bearing the CD161 antigen was increased in SSc, since it has been shown that CD161 is preferentially expressed in Th17 cells [21]. This was indeed the case (Figure 3A) and, interestingly, CD161 expression correlated with IL-17A (Pearson r = 0.47, P = 0.046) but not IL-22 (Pearson r = 0.34, P = 0.17) production in SSc (Figure 3Band data not shown). Of note, CD161 was also expressed in non CD4+ cells; however, these cells were not increased in SSc and were not correlated with the production of IL-17A (not shown).

Altogether our data indicate a preferential expansion of IL-22-single, IL-17A-single and IL-17A/IL-22 double producing CD4+ T cells in SSc, thus supporting the contention that Th22 cells in addition to Th17 cells are expanded in SSc.

We then investigated whether SSc individuals had a skewed pattern of lymphocytes expressing chemokine-receptors involved in skin- (CCR6, CCR4 and CCR10) [10, 12] and lung-homing (CCR6) [11, 13, 36]. CXCR3, preferentially expressed on Th1 cells, served as control. The expression of CCR6, CCR4, CCR10 as well as CXCR3 on resting peripheral blood memory CD4+ T cells was similar in SSc and HD (Figure 4A). However, CCR6 expression in CD4+ T cells was positively correlated with the frequency of IL-17A+ and IL-22+ CD4+ T cells assessed after overnight activation (not shown). Of note, this positive association was observed in SSc but not in HD and was stronger with the CCR6+CCR10- CD4+ T cell subset for both IL-17A+ (r = 0.56, P = 0.013) and IL-22+ (r = 0.54, P = 0.024) CD4+ T cells (Figure 4B). No significant positive correlations were observed for the other Th subsets with the chemokine receptors studied. Together, these data suggest that IL-17A and IL-22-producing cells in the peripheral blood of SSc but not HD have skin and lung-homing properties.

We finally tested whether the production of specific T cell cytokines after overnight activation was associated with SSc clinical characteristics. SSc individuals presenting with ILD, as detected by HRCT and decreased TLC, had higher frequencies of IL-22 (P = 0.001) and IL-17 producing T-cells (P = 0.05) (Figure 5) when compared with those without ILD. Moreover, IL-4 producing T-cells showed a trend to increase in SSc individuals having ILD (P = 0.08), while IFN-γ producing T-cells did not (Figure 5). Of interest, no differences were observed in the frequencies of cytokine producing cells when considering the gender, lSSc vs. dSSc, disease duration, presence of digital ulcers or PAH, and ANA positivity thus reinforcing the specificity of our finding.