Active systemic lupus erythematosus is associated with a reduced cytokine production by B cells in response to TLR9 stimulation

For the analysis of cytokine production by B cells, peripheral blood was collected from 18 SLE patients (17 females/1 male) with a mean age of 34.9 ± 10.4 years and 13 healthy donors (12 females/1 male) with a mean age of 36.7 ± 14.9 years. For the analysis of activation and IL-10 expression in B cells using flow cytometry (FC), peripheral blood was collected from 6 female SLE patients with a mean age of 38.8 ± 12.9 years and 10 healthy donors (8 female/2 male) with a mean age of 32.9 ± 11.1 years. For the analysis of TLR9 expression, peripheral blood was collected from patients with SLE (12 female/1 male, 38.4 ± 18.4) and 5 female healthy donors (29.4 ± 5.0).

The study was approved by the local ethics committee of the Charité Universitätsmedizin Berlin and written consent was obtained from all donors. The consents are on file held by the principal investigator and available for review by the editor-in-chief upon request. All patients met the revised American College of Rheumatology classification criteria for SLE [23]. The disease activity was assessed using the SLE disease activity index (SLEDAI) modified according to the SELENA-trial [24]. Details of the clinical characteristics and treatment regimens of the analyzed SLE patients are provided in Table 1.

Peripheral blood mononuclear cells (PBMCs) were isolated with density gradient centrifugation using lymphocyte separation medium (PAA Laboratories, Pasching, Austria) as previously described [25]. Subsequently, B cells were negatively purified by magnetic activated cell sorting (MACS®) using the B-cell Isolation Kit II (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s instructions and B cell purity was checked by flow cytometry. The contamination with CD3⁺, CD14⁺ and dead cells was below 5% in all samples.

Purified B cells were stained at 4°C for 15 minutes with antibodies against CD14-PB (M5E2), CD3-PB (UCHT1), CD27-Cy5 (2E4), CD19-PE-Cy7 (SJ25C1), CD20-PerCP-Cy5 (L27), and IgD-FITC (IA62) to control the purity used for the subsequent analyses.

Before and after stimulation, PBMCs were stained first with antibodies against CD14- Pacific blue (PB) (M5E2), CD3-PB (UCHT1), CD27-fluorescein isothiocyanate (FITC) (L128), CD38-PercP-Cy5.5 (HIT2) and CD20-Pacific orange (PO) (H147) for 10 minutes on ice. After washing, PBMCs were incubated with 400 μl of 1 × FACS permeabilizing solution 2 (Becton Dickinson (BD) Franklin Lakes, NJ, USA) for 10 minutes at room temperature (RT). After permeabilization and washing, PBMCs were stained with anti-Ki67-PE-Cy7 (B56) and anti-IL-10-APC (JES3-9D7) and anti-TLR9-PE (eB72-1665) antibodies for 10 minutes at RT. All antibodies were purchased from BD; beside Cy5-conjugated anti-CD27 antibody (2E4) (kind gift from Andreas Thiel, Berlin Center for Regenerative Therapy, Charité Berlin) and anti-IL-10-APC antibodies purchased from Miltenyi Biotec. Stained cells were analyzed by FC using the FACSCanto™II flow-cytometer (BD). FC data were analyzed using FlowJo (Tree Star, Inc., Ashland, OR, USA).

B cells were stimulated in vitro with CpG 2006 oligonucleotide (CpG) (TIB MolBiol Synthese Labor GmbH, Berlin, Germany). The cells were resuspended in RPMI 1640 Glutamax supplemented with 10% FCS (Lonza, Köln, Germany), 5% penicillin/streptomycin, and 0.05 mM 2-mercaptoethanol (Gibco® Life Technologies GmbH, Darmstadt, Germany). B cells, 10⁵, were seeded and stimulated with 2.5 μg/mL CpG for 48 h at 37°C and 5% CO₂. After 2 days of culture, the supernatants were harvested and frozen at -70°C prior to analysis.

To analyze the IL-10 production by B cells, PBMCs (10⁶/well) were cultured with CpG 2006 in vitro as described [26],[27]. Intracellular staining of IL-10 and Ki67 was performed on PBMCs after 2 days of culture. PBMCs were re-stimulated for 4 h with 10 ng/mL PMA and 1 μM ionomycin including 2 μg/mL brefeldin A for the last 2 h (all from Sigma Munich, Germany) prior to intracellular staining. Unstimulated cells served as controls.

Cryopreserved supernatants were assessed for determination of cytokine concentration using Bio-Plex® technology (Bio-Rad Laboratories, Inc., CA, USA) according to the manufacturer’s instructions. The cytokines analyzed were IL-1β, IL-1ra, IL-2, IL-4, IL-5, IL-6, IL-7, IL-8/CXCL8 (chemokine (C-X-C motif) ligand 8), IL-9, IL-10, IL-12p70, IL-13, IL-15, IL-17A, eotaxin-1/CCL11 (chemokine (C-C motif) ligand 11), basic fibroblast growth factors (FGF), granulocyte colony-stimulating factor (G-CSF), granulocyte macrophage colony-stimulating factor (GM-CSF), IFN-α2, IFN-γ, IP-10 (IFNγ-induced protein 10)/CXCL10, monocyte chemotactic protein-1 (MCP-1)/CCL2, macrophage inflammatory protein-1α (MIP-1β)/CCL3, MIP-1β/CCL4, platelet-derived growth factor-BB (PDGF-BB), regulated on activation, normal T-cell expressed and secreted (RANTES)/CCL5, vascular endothelial growth factor (VEGF) and TNF-α. The assay sensitivity depends on the particular cytokines analyzed from 0.3 pg/mL for IL-10 to 6.4 pg/mL for IFN-γ. Although all 28 cytokines and growth factors were detectable in the supernatant of the B-cell culture; 12 of them (IL-1β, IL-2, IL-4, IL-5, IL-7, IL-12p70, IL-13, IL-15, GM-CSF, IFN-α2, eotaxin-1, and MCP-1) were produced at low levels with a mean concentration below 20 pg/mL after TLR9 stimulation (Additional file 1) and were therefore not considered for further analysis.

The statistical analysis was performed with SPSS (version 20, IBM, NY, Chicago, IL, USA). To compare data from healthy donors and SLE patients, the nonparametric Mann-Whitney U-test was used. The Wilcoxon test was used to compare results after TLR9 stimulation with unstimulated controls for the cytokine production. Multiple comparisons were performed using one way analysis of variance (ANOVA) with Dunnett’s post hoc test. To correlate cytokine levels with SLEDAI scores or with dsDNA titers, Spearman correlation analysis was performed. P-values <0.05 were considered statistically significant. The statistical tests used are indicated in each figure legend.

We first evaluated the response of B cells to TLR9 stimulation in terms of proliferation and activation. The frequency of proliferating cells (% of Ki67⁺ B cells), and the upregulation of the activation marker CD38 after 2-day culture with CpG were analyzed by FC (Figure 1A). CpG induced B cell proliferation independently of B cell receptor (BCR) engagement; however, SLE patients had a lower frequency of proliferating B cells upon TLR9 stimulation in comparison with healthy donors (P <0.05) (Figure 1B, left graph). Moreover, activation of B cells was evaluated by upregulation of CD38 expression (mean fluorescence intensity, MFI) upon TLR9 stimulation. B cells from healthy donors significantly upregulated CD38, resulting in a 3-fold increase after TLR9 stimulation, while the response of B cells from SLE patients was significantly lower (Figure 1B, right graph, P <0.05).

Subsequently, the influence of TLR9 stimulation on cytokine production was evaluated by analyzing the concentrations of 28 cytokines and growth factors in the supernatants of B cell cultures from healthy donors and SLE patients using BioPlex technology. The cytokines were grouped according to the level of their induction upon TLR9 engagement compared to unstimulated B cells. The first group included cytokines showing more than a 4-fold increase after TLR9 stimulation and comprised IL-1ra, TNF-α, IL-6, and IL-10, the chemokines IP-10, MIP-1α, -1β, and the growth factor VEGF. The second group was defined by a moderate (2- to 4-fold) increase after TLR9 stimulation, and comprised IL-8, IL-9, IL-17A and IFN-γ. A third group of growth factors and chemokines, defined by a very small increase or no increase after TLR9 stimulation (maximum 2-fold increase) comprised basic FGF, G-CSF, PDGF-BB, and RANTES (Figure 2). Overall, the profiles of cytokine secretion observed in the supernatants of cultured B cells from SLE patients and healthy donors shared very large similarities. We found that none of the cytokines from groups 1 and 2 were secreted at a higher level by B cells from SLE patients, but rather at lower levels in comparison to B cells from healthy donors. Since earlier reports found that B cells from active SLE patients were less responsive to TLR9 stimulation in terms of IL-6 and IL-10 production [21], we analyzed the relation between the cytokines from groups 1 and 2 and the disease activity (SLEDAI) by using a heat map (Figure 3A). When the patients were ordered according to their SLEDAI, it became apparent that B cells from patients with SLEDAI of 4 (n = 3) produced larger amounts of cytokines than those from patients presenting with a SLEDAI higher than 14 (n = 3). The remaining patients with a SLEDAI between 4 and 14 displayed an intermediate but clearly ranked profile (Figure 3A). In greater detail, correlation analyses of individual cytokines revealed significant inverse correlations between the SLEDAI and inducible amounts of IL-6, IL-9, IL-17A, IFN-γ, IP-10, MIP-1α, MIP-1β, TNF-α, and VEGF of TLR9-activated SLE B cells (Figure 3B). In contrast, there was no correlation between the spontaneous (unstimulated) production of these cytokines and the SLEDAI, highlighting the specificity of this association with TLR9 signaling.

This global analysis showed that TLR9-stimulated B cells from patients with active SLE produced fewer cytokines than those from patients with less active disease. In order to further compare the data to healthy donors, we divided the patients into two groups with high (≥6) and low SLEDAI (<6) [28] (Figure 4A). Notably, the production of IL-6, IL-1ra, and VEGF by B cells was significantly reduced in patients with active SLE with a SLEDAI ≥6 compared to healthy donors. The secretion of IL-10 was also reduced by trend, although not statistically significant (P = 0.051). FC analysis of IL-10-producing B cells showed a significantly increased frequency of IL-10⁺ B cells after TLR9 stimulation (Figure 4B). However, there was no difference between B cells from SLE patients and healthy donors. While the generation of IL-10-producing B cells was not reduced in culture of B cells from SLE patients, there was a clear reduction of the amount of IL-10 induced in individual IL-10⁺ B cells compared to healthy donors as shown by the MFI of intracellular IL-10 (Figure 4B).

A key serologic parameter reflecting the breakdown of tolerance and related with lupus disease activity are anti-dsDNA autoantibodies that have been linked to TLR9 stimulation [29]. We found a significant (P <0.05) inverse correlation between the serum anti-dsDNA titers and amounts of IL-1ra, IL-6, IL-9, IL-17A, IFN-γ, MIP-1α, -1β, TNF-α, and VEGF produced by TLR9-activated B cells from SLE patients (Figure 5). Thus, the higher anti-dsDNA antibody titer in serum, the lower was the level of these cytokines produced in vitro by CpG activated B cells.

In an attempt to explain the reduced response of B cells from SLE patients to TLR9 agonist compared to controls, we analyzed whether TLR9 was differentially expressed by these cells. FC analysis of TLR9 expression in B cells from healthy donors and SLE patients showed that the MFI of TLR9 was similar in B cells from healthy donors compared to those from SLE patients (242.8 ± 41.9 and 218.5 ± 53.2) (Figure 6). However, a significant reduction of TLR9 expression was found for B cells from SLE patients with a SLEDAI ≥6 (191.8 ± 19.5) (Figure 6). This result is in line with our observations on cytokine production, suggesting that the observed reduction in CpG responsiveness for B cells from patients with high disease activity could be related to a down-modulation of TLR9 expression.