Increased frequency of circulating IL-21 producing Th-cells in patients with granulomatosis with polyangiitis (GPA)

Forty-two patients with GPA and 29 age- and sex-matched healthy controls (HC) (18 male, 11 female, mean age 56 (SD ± 13) years, range 26 to 72 years, P = 0.16) were included in this study. The diagnosis of GPA was established according to the definitions of the Chapel Hill Consensus Conference and patients fulfilled the classification criteria of the American College of Rheumatology (ACR)[25, 26]. Only patients without clinical signs and symptoms of active vasculitis and considered to be in complete remission, as indicated by a score of zero on the Birmingham Vasculitis Activity Score (BVAS), were included in the study [27]. Serostatus for ANCA was followed for several months in all patients, and patients with a stable serostatus for ANCA (positive or negative) for at least 3 months were included in this study. Based on these criteria, 23 patients were positive for PR3-ANCA, whereas 19 were ANCA-negative. There was history of generalized disease including renal involvement in 27 patients, and 15 patients had localized disease, which had been confined to the upper and/or lower respiratory tract. None of the patients and controls had infection at the time of sampling. Eleven GPA-patients (eight ANCA-positive, and three ANCA-negative) were treated with maintenance immunosuppressive therapy at the time of blood sampling. Four of them were treated with only azathioprine (25 to 100 mg/day), one patient with mycophenolate mofetil (1500 mg/day), and six patients received prednisolone (5 to 10 mg/day) in combination with azathioprine (125 mg/day). Participants in rituximab trials were excluded from the present study. The main clinical and laboratory data from the patients are summarized in Table 1. All patients and healthy individuals provided informed consent and the study was approved by the local Medical Ethics Committee of the University Medical Centre Groningen, University of Groningen (NL).

ANCA titers were measured by indirect immunofluorescence (IIF) on ethanol-fixed human granulocytes according to standard procedures as previously described [28]. ANCA titers higher than 1:40 were considered positive. ANCA antigenic specificity was determined using an in-house capture ELISA as described before [29, 30]. Briefly, a 96-well plate was coated with goat-anti-mouse Ig for 48 hours. After washing, plates were incubated with mouse monoclonal antibody against human PR3 for 2 hours. After washing, the plate was incubated overnight at 4°C with an extract of human azurophilic granules, which were isolated from neutrophils of healthy donors. Further, serial dilutions of serum (with a starting dilution of 1:100) were incubated for 1 hour. The plate was washed, and the captured antibodies were detected with purified F(ab)₂ goat-anti-human IgG conjugated to alkaline phosphatase. P-nitrophenyl-phosphate disodium was used as a substrate, and the optical density was measured at 405 nm.

The following conjugated antibodies were used in flow cytometry: allophycocyanin (APC)-conjugated anti-CD3 (clone UCHT1), peridin chlorophyll protein (PerCP)-conjugated anti-CD8 (clone SK1), phycoerythrin (PE)-conjugated anti-IL-21-receptor (clone 17A12), and PerCP-conjugated anti-CD19 (clone 4G7), all purchased from Becton & Dickinson (Amsterdam, The Netherlands); PE-conjugated anti-IL-21 (clone eBio3A3-N2), Alexa Fluor^® 488 (A488)-conjugated anti-IL-17 (clone eBio64DEC17), and A488-conjugated anti-FoxP3 (clone PCH101), all purchased from eBioscience (San Diego, CA, USA); and PE-conjugated anti-BCL-6 (clone IC5046P) obtained from R&D Systems (Minneapolis, MN, USA).

Lithium-heparinized venous blood was obtained from patients and healthy donors. Immediately after sampling, 400 μL blood was mixed with 400 μL RPMI1640 (Cambrex Bio Science, Verviers, Belgium), supplemented with 50 μg/ml gentamycin (Gibco, Scotland, UK), and aliquoted in two 5-mL polypropylene tubes (BD Biosciences, Amsterdam, The Netherlands) (400 μL per tube). Diluted blood samples were stimulated 4 hours with 50 ng/mL phorbol myristate acetate (PMA; Sigma-Aldrich, Steinheim, Germany) and 2 nM calcium ionophore (Ca-Io; Sigma-Aldrich). As a negative control, one sample was kept in medium only without stimulation. For inhibiting cytokine release from the cells, 10 μg/ml of brefeldin A (Sigma-Aldrich) was added to each sample.

After stimulation cells were washed in wash buffer (PBS, 5% fetal bovine serum (FBS), 0.1% sodium azide (Merk, Germany)] and stained with PerCP-conjugated anti-CD8 and APC-conjugated anti-CD3 for 15 minutes at room temperature. Cells were fixed with 100 μL Reagent A (Caltag/Invitrogen., Breda, The Netherlands) for 10 minutes. After washing, the pellet was resuspended in 100 μL permeabilization Reagent B (Caltag/Invitrogen) and labeled with A488-conjugated anti-IL-17 and PE-conjugated anti-IL-21 for 20 minutes in the dark. After staining, the cells were washed and immediately analyzed on the FACS-Calibur flow cytometer (Becton & Dickinson). Data were collected for 2 × 10⁵ cells, and plotted using the Win-List software package (Verity Software House Inc, ME, USA) ME, USA). Because stimulation reduces surface expression of CD4 on T-cells, CD4⁺T-cells were identified indirectly by gating on CD3-positive and CD8-negative lymphocytes. Gated CD4⁺ T-cells were further displayed as a dot plot for evaluation of intracellular cytokine production. The unstimulated control sample was used as a guide for setting the linear gates to discriminate positive and negative populations.

Peripheral blood mononuclear cells (PBMCs) from GPA patients and HCs were prepared from heparinized venous blood by density-gradient centrifugation on Lymphoprep (Axis-Shield PoC AS, Oslo, Norway). Cells recovered from the gradient interface were washed twice in PBS and stained for BCL-6 and FoxP3 according to the manufacturer's instructions (eBioscience staining set for transcription factors). Briefly, PBMCs were adjusted to 1 × 10⁶ cells in 100 μL and incubated with appropriate concentration of APC-conjugated anti-CD3 and PerCP-conjugated anti-CD8 for 30 minutes at 4°C in the dark, followed by fixation and permeabilizaion in Fix/Perm buffer (eBioscience) for 45 minutes. Cells were then washed twice with 1 × permeabilization buffer (eBioscience), and stained with PE-conjugated anti-BCL-6 and A488-conjugated anti-FoxP3. After incubation for 30 minutes in the dark, the cell suspension was washed and immediately analyzed on the FACS-Calibur flow cytometer (Becton & Dickinson). Lymphocytes were gated by forward and side scatter patterns, and plotted using the Win-List software package (Verity Software House Inc). Isotype matched control antibodies of irrelevant specificity were obtained from eBioscience and R&D systems.

Fresh blood samples from GPA patients and HCs were labeled with PE-conjugated anti-IL21R, and PerCP-conjugated anti-CD19 for 15 minutes in the dark. Cells were successively treated with 2 mL diluted FACS lysing solution (BD, Amsterdam, The Netherlands) for 10 minutes and then washed twice in wash buffer and immediately analyzed by flow cytometry.

Erythrocytes were lysed and leukocytes were fixed and washed twice in 1% BSA. RNA was isolated from total leukocytes with TRIzol reagent (Invitrogen) according to the manufacturer's instructions. DNAse treatment (Ambion, Huntingdon, Cambridgeshire, UK) was performed and subsequently cDNA was synthesized using M-MLV reverse transcriptase and oligo (dT) 14 to 18 as primer. For measurement of mRNA for BCL-6 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 1 μL of cDNA in triplicate was used for amplification by the Taqman RT-PCR system (ABI Prism 7900HT Sequence Detection System, Applied Biosystems, Foster City, CA, USA) with specific Taqman primers/probes (BCL-6 (Hs 00153368_m1) and GAPDH (Hs 99999905_m1), Applied Biosystems). Amplification was performed using standard conditions and calculations of fold induction were performed. We normalized gene expression to GAPDH and expressed values relative to control using the ^ΔΔCT (cycle threshold) method.

PBMCs recovered from the gradient interface were washed twice in PBS and adjusted to 10⁶ cells/mL in RPMI 1640 (Lonza, Basel, Switzerland) supplemented with 10% FCS (Lonza, Switzerland) and 50 μg/mL gentamicin (GIBCO, Invitrogen). Cells were cultured in the presence of 100 ng/mL rhIL-21 (ImmunoTools GmbH, Friesoythe, Germany) and/or 100 ng/mL rhBAFF (PeproTech, NJ, USA) for 12 days at 37°C with 5% CO_2. After 12 days, culture supernatants were collected and total IgG was measured using an in-house ELISA as described previously [31]. Briefly, Costar 96-well ELISA plates were coated with 2 mg/mL goat anti-human-Ig antibody (Southern Biotech, Birmingham, AL, USA) in carbonate buffer (0.01 M, pH 9.6). Plates were washed with washing buffer (0.025 M Tris-HCl, 0.15 M NaCl, 0.05% Tween-20) and blocked for 1 hour with blocking/incubation-buffer (0.05 M Tris-HCl, 0.3 M NaCl, 0.05% Tween-20, 1% BSA). Cell culture supernatants were diluted in incubation buffer. Purified human IgG with a known concentration was used as a standard sample. The bound IgG was detected with goat-anti-human-IgG antibody conjugated with alkaline phosphatase (Sigma, St Louis, MO, USA). P-nitrophenyl-phosphate disodium was used as substrate and optical density was read at 405 nm using an Emax microplate reader (Molecular Devices, Silicon Valley, CA, USA).

In vitro PR3-ANCA IgG production in PBMC culture supernatants was measured by Phadia ImmunoCAP^® 250 analyzer (Thermo Fisher Scientific, MA, USA) using ELiA™ PR3, and the levels of PR3-ANCA IgG production were expressed in response units (RU).

Data are presented as median values unless stated otherwise. The nonparametric Mann-Whitney U-test was used to compare data from patients with those of HCs. The Wilcoxon matched pairs test was used for intra-individual comparison. Correlations were assessed using Spearman's rank correlation coefficient. Two-tailed P-values lower than 0.05 were considered statistically significant.

We initially determined the frequency of IL-21-producing CD4 T-cells in the peripheral blood of GPA patients (n = 42) and HCs (n = 29) after in vitro stimulation. The percentage of circulating IL-21⁺ Th-cells was significantly higher in GPA patients compared with the control group (Figure 1B). Of note, Th17 cells may produce IL-21 in addition to their signature cytokine IL-17. Since Th17 cells are increased in GPA patients [10, 11], we next extended our analysis to investigate whether increased IL-21⁺ Th-cells in GPA patients resulted from an increase in Th17 cells. To this end, IL-17 staining was included in the analysis to determine what percentage of the total IL-21⁺ Th-cells are Th17 cells. Using this approach, we assessed the frequency of IL-21⁺IL-17^-, IL-21⁺IL-17⁺, and IL-21^-IL-17⁺ cells within the CD4 T-cells in GPA patients and HCs. As shown in Figure (1C, D and 1E), GPA patients in remission had a significantly higher percentages of circulating IL-21⁺IL-17^-, IL-21⁺IL-17⁺, and IL-21^-IL-17⁺ cells compared with the control group. However, the majority of circulating CD4⁺ T-cells that produced only IL-21 were distinct from Th17 cells, that is, negative for IL-17. To assess the possible role of IL-21⁺IL-17^- Th-cells in ANCA production, we compared their percentage between patients who were ANCA-positive (n = 23; IIF titer >1:40) or ANCA-negative (n = 19) at the time of inclusion. Significant increases in the frequencies of IL-21⁺IL-17^- Th-cells were observed in ANCA-positive patients in comparison with HCs and ANCA-negative patients, whereas no significant difference was found between ANCA-negative patients and HCs (Figure 1F). In contrast, the percentages of IL-21⁺IL-17⁺ and IL-21^-IL-17⁺ Th-cells in ANCA-positive GPA patients did not differ from those in ANCA-negative GPA-patients (Figure 1G and 1H). These results suggest that persistence of IL-21⁺IL-17^- Th-cells during remission plays a role in the ongoing humoral autoimmune response in ANCA-positive GPA patients.

To rule out the possibility that the increased proportion of IL-21⁺IL-17^- Th-cells in GPA patients was the result of current treatment, the ANCA-positive patient group was divided into treated and untreated patients, and the percentages of IL-21⁺IL-17^- Th-cells were compared. No significant differences were observed between treated and untreated patients (data not shown). We also compared the percentage of IL-21⁺IL-17^- Th-cells between currently untreated ANCA-positive patients with a history of generalized disease and those with localized disease. No difference was found between these patient groups (data not shown).

It has been reported that IL-21 is a key factor regulating the differentiation of naïve CD4⁺ T-cells into Th17 cells [32, 33]. In order to analyze this relationship, we tested correlation between percentages of IL-21⁺IL-17^- Th-cells and percentages of terminally differentiated Th17 cells (IL-21^-IL-17⁺) in GPA patients (n = 42) and HCs (n = 29). Interestingly, a significant positive correlation was observed between IL-21⁺IL-17^- Th-cells and IL-21^-IL-17⁺ Th-cells in both GPA patients and HCs (r = 0.58, P < 0.0001 and r = 0.37, P = 0.04, respectively) (Figure 2A and 2B).

Since IL-21 is not the only marker for T_FH cells, we further characterized the identity of circulating IL-21-producing cells by analyzing BCL-6 expression, which is considered a master regulator and specific transcription factor for T_FH cells [34, 35]. To this end, the expression of mRNA BCL-6 was assessed in circulating leukocytes from GPA patients and HCs by real-time RT-PCR. Restricted numbers of patients and controls were included in this analysis due to insufficient cell numbers. Patients with ANCA-positive GPA (n = 10) had a significantly higher expression of mRNA BCL-6 than ANCA-negative patients (n = 6) and HCs (n = 11) (Figure 3B). In addition, intracellular FACS-staining for BCL-6 within circulating CD4+ T cells confirmed the increased BCL-6 expression in ANCA-positive GPA patients (Figure 3A and 3C). We have also analyzed the MFI (mean fluorescence intensity) of BCL-6 expression in CD4⁺ T-cells from patients and HCs and found that the expression level of BCL-6 per Th-cell in GPA patients was similar to that in HCs (data not shown). Thus, BCL-6 expression is increased in GPA patients due to increased frequencies of circulating BCL-6⁺ CD4⁺ T-cells.

Because in recent studies a new population of FoxP3⁺ regulatory T-cells has been described that shares features with T_FH cells by expressing the transcription factor BCL-6, we also evaluated whether the increase in BCL-6⁺ T-cells in GPA patients was a result of an increase in FoxP3⁺BCL-6⁺ T-cells [36, 37]. This analysis showed that the increase in BCL-6 expression in GPA patients was restricted to T_FH cells and although a low percentage of FoxP3⁺BCL-6⁺ T-cells was found (< 0.3%), no differences in these cell frequencies were observed between GPA patients and HCs (data not shown).

Since it is well known that IL-21 acts on B-cells to support their expansion and antibody production [38‐40], we conducted further analysis to compare the expression of IL-21R on B cells from GPA patients and HCs. No differences were seen in the percentages of IL-21R⁺ B-cells either between ANCA-positive (n = 13) and ANCA-negative patients (n = 14) or between patients and HCs (n = 19) (Figure 4).

To explore the interplay between IL-21-producing Th-cells and B-cells in GPA patients, we investigated the effect of IL-21 on IgG antibody-production by B-cells from GPA patients. Restricted numbers of patients and controls were enrolled in this analysis due to insufficient cell numbers. PBMCs from GPA patients were cultured in vitro in the presence or absence of exogenous IL-21 for 12 days and total IgG was measured in supernatants by ELISA. Because IL-21 promotes B-cell differentiation by synergizing with BAFF [12, 13], we questioned whether the effect of IL-21 on IgG production could be augmented by adding BAFF to the culture. Of note, autologous T-cells in our culture system act as a natural provider of CD40 ligation for B-cells, as this ligation is required for B-cell activation, isotype switching and memory development. As shown in Figure 5, IL-21 significantly enhanced the production of IgG in vitro in stimulated PBMCs from both ANCA-positive (n = 7) and ANCA-negative (n = 6) GPA-patients, whereas stimulation with BAFF alone did not result in increased IgG production. The combination of BAFF and IL-21 tended to increase IgG production more than IL-21 alone. Next, we assessed the effect of IL-21 plus BAFF on in vitro production of PR3-ANCA. As shown in Figure 5B, spontaneous PR3-ANCA production was observed in cultured PBMCs from ANCA-positive patients (n = 16) in comparison with cells from ANCA-negative patients (n = 12). Importantly, IL-21 induces a significant enhancement in PR3-ANCA production in PBMCs isolated from ANCA-positive patients in comparison with ANCA-negative patients. So it is conceivable that autoreactive B-cells were enriched in the peripheral blood of ANCA-positive patients.