Dysfunction of CD19+CD24hiCD27+ B regulatory cells in patients with bullous pemphigoid

Bullous pemphigoid (BP) is an autoimmune blistering skin disease characterized by the production of autoantibodies against the hemidesmosomal protein BP180. B regulatory cells (Bregs) are crucial in maintaining self-tolerance and suppressing autoantibody production. However, it is still unclear whether the dysfunctions of Bregs contributes to the autoantibody production in BP patients. In this study, we found that CD19+CD24hiCD27+ Bregs and IL-10+CD19+ Bregs were significantly increased in the peripheral blood samples of BP patients compared with that in healthy controls. Moreover, compared to Bregs from healthy individuals, we found that Bregs from BP patients fails to suppress the production of specific anti-BP180 autoantibody when co-cultured with patient-derived PBMCs. Additionally, Bregs from BP patients were defective in suppressing the CD4+ T cell proliferation and the cytokines expression (including IFN-γ, TNF-α and IL-4). Notably, we found that patient-derived Bregs produced high level of TNF-α and the TNF inhibitor etanercept could inhibit the autoantibody production in the culture system in vitro. Our results indicate that Bregs from BP patient appear phenotypically pro-inflammatory by their cytokine profile and are defective in immunosuppressive function, which suggest that Bregs play a pro-inflammatory role rather than a regulatory role in the pathogenesis of BP.


Increased Breg frequency in BP patients.
To observe the frequency of Bregs in BP patients, we identified Bregs as CD19 + CD24 hi CD27 + and IL-10 + CD19 + by flow cytometry. We found that the frequency of CD19 + CD24 hi CD27 + Bregs ( Fig. 1A and B) and IL-10 + CD19 + Bregs (Sup Fig. 2A and B) were both significantly higher in BP patients compared with that in healthy controls (p < 0.05). To confirm the CD19 + CD24 hi CD27 + Bregs were actually IL-10-producing Bregs, we analyzed the expression of IL-10 in activated Bregs from BP patients and from healthy controls. The result showed that more than 10% of CD19 + CD24 hi CD27 + Bregs produced IL-10 ( Fig. 1A and C), which was in line with the previous findings by Iwata et al. 10 . Our results indicated that the frequency of CD19 + CD24 hi CD27 + Bregs in BP patients was elevated compared with that in healthy controls.
Modified function of Bregs in suppressing autoantibody production in BP patients. To investigate the function of Bregs from BP patients in regulating immune responses, Bregs from BP patients and healthy controls were isolated and then observed for their effects on autoantibody production in vitro. In preparation for these assays, recombinant human BP180-NC16A proteins were expressed and purified, as shown in Sup Fig. 3. ELISA assays were performed to evaluate the binding activity of BP autoantibody to recombinant human BP180-NC16A. The results showed that GST-tagged NC16A bound patient autoantibodies robustly, whereas GST Representative FACS data of the frequency of CD19 + CD27 + cells gated from lymphocytes in FSC/SSC dot plots (top), CD24 hi Bregs within gated CD19 + CD27 + cells (middle) and IL-10 + Bregs within gated CD19 + CD24 hi CD27 + Bregs (bottom) of PBMCs frome BP patients and healthy controls. PerCP/Cy5.5 and PE mouse IgG1 isotype and FITC mouse IgG2a isotype were used as a negative control for immunofluorescence staining and flow cytometry assay. (B) Statistical analysis of the CD19 + CD24 hi CD27 + /CD19 + cell ratios in indicated groups (n = 25 per group). (C) Statistical analysis of the IL-10 + cells within gated CD19 + CD24 hi CD27 + Bregs in indicated groups (n = 5 per group). p < 0.05 determined by two-tailed Student's t test.
Scientific REPoRTs | (2018) 8:703 | DOI:10.1038/s41598-018-19226-z did not ( Fig. 2A). Next, we incubated PBMCs from BP patients and from healthy controls respectively with the recombinant human BP180-NC16A protein. We found high levels of specific anti-BP180 antibody in cell culture supernatants of patient-derived PBMCs, whereas barely detectable anti-BP180 antibody titer in the cell culture supernatants of PBMCs from healthy controls (Fig. 2B).
To obverse the effect of Bregs on suppressing autoantibody production in BP patients, we compared anti-BP180 antibody titers in cell culture supernatants of the patient-derived PBMCs with or without the depletion of Bregs incubated with the recombinant human BP180-NC16A protein. Patient-derived BPMCs alone were used as negative controls. Notably, we observed no significant difference in the production of specific anti-BP180 antibody between patient-derived PBMCs with Bregs depletion and without Bregs depletion (Fig. 2C). These results suggested that Bregs from BP patients failed to suppress autoantibody production. To further determine the defective function of Bregs from BP patients, we co-cultured patient-derived PBMCs with Bregs from the BP patient or healthy controls, followed by incubation with BP180-NC16A. As expected, compared to Bregs from healthy controls, patient-derived Bregs showed impairment in suppressing autoantibody production (Fig. 2D). Our results suggest that the functional deficiency of Bregs may contribute to autoantibody production in BP.
Modified function of BP Bregs to restrain CD4 + T cell activation. Activated CD4 + T cells are necessary to facilitate B cell proliferation and differentiation by producing pro-inflammatory cytokines, such as IL-4, IL-21, and TNF-α 17 . Previous studies showed that Bregs exhibit immunosuppressive function mainly by inhibiting CD4 + T cell proliferation and inflammatory cytokine production 18,19 . Given that we had found BP patient-derived Bregs have modified function in suppressing autoantibody production, we speculated that those Bregs have functional deficiency in restraining CD4 + T cell activation.
To investigate the effect of the modified Bregs function on T cell proliferation in BP patients, CFSE labeled CD4 + T cells were co-cultured with Bregs and the proliferation rate of CD4 + T cells were determined by flow ELISA analysis of the specific anti-NC16A antibody production in PBMCs from BP patients with or without Breg deletion (n = 5). (D) PBMCs from BP patients were co-cultured with CD19 + CD24 hi CD27 + Bregs or CD19 + CD24 − CD27 − non-Bregs (3:1) from the third-part BP patients and healthy controls (n = 5 per groups). ELISA analysis of the specific anti-NC16A antibody production in the co-cultures. N stands for normal, P stands for patients. *p < 0.05, **p < 0.01 and ***p < 0.001 determined by paired version of two-tailed Student's t test or one-way ANOVA followed by Bonferroni corrections for post hoc t-test.
Scientific REPoRTs | (2018) 8:703 | DOI:10.1038/s41598-018-19226-z cytometry. We found that the proliferation rate of CD4 + T cells was decreased when co-cultured with Bregs from healthy controls, compared to the non-Breg groups. In contrast, the proliferation rate of CD4 + T cells did not differ when cultured with Bregs or non-Bregs from BP patients ( Fig. 3A and B). These results indicate that healthy-derived Bregs was able to suppress CD4 + T cell proliferation, whereas BP patient-derived Bregs was lack of this function.
To further determine the effect of BP patient-derived Bregs on cytokine production of CD4 + T cells, purified CD4 + T cells were co-cultured with Bregs or non-Bregs and the expression of IFN-γ, TNF-α, and IL-4 in CD4 + T cells were assessed by flow cytometry. The resulted showed that CD4 + T cells produced significantly lower levels of these cytokines when co-cultured with Bregs than non-Bregs from healthy controls. In contrast, we observed no differences of these cytokines expressed by CD4 + T cells following the co-culture with either Bregs or non-Bregs from BP patients ( Fig. 4 and Sup Fig. 4). In conclusion, these data indicate that healthy-derived Bregs suppress cytokine expression in CD4 + T cells, whereas patient-derived Bregs lack this function.

TNF-α expression in Bregs from BP patient.
To investigate the underlying mechanism that responding to the dysfunction of Bregs in BP patients, we first observed the expression of inflammatory cytokines in the sorted CD19 + CD24 hi CD27 + Bregs from BP patients and from healthy controls. The real-time PCR results showed that TNF-α and IFN-γ were significantly up-regulated in the CD19 + CD24 hi CD27 + Bregs from patients compared with that in healthy controls (Fig. 5A). There were no significant difference of IL-10, IL-6, IL-22 and IL-23 mRNA between the two groups. In addition, we examined the expression of TNF-α in Bregs from BP patients and healthy controls by flow cytometry. The result showed that Bregs from patient produced high levels of TNF-α compared with that in healthy controls ( Fig. 5B and C). Further, we added TNF-α antagonist etanercept into the co-culture system to observe its effect on the autoantibody production. The results showed that etanercept could inhibit the autoantibody production in the supernatant of patient-derived PBMCs (Fig. 5D). These results indicate that that Bregs from BP patient might express more inflammatory cytokines TNF-α and thus result to its modified function in suppressing autoantibody production.

Discussion
In this study, we found that the frequency of circulating CD19 + CD24 hi CD27 + Bregs and IL-10 + CD19 + Bregs were increased in BP patients. Moreover, our in vitro study suggested that Bregs from BP patient were defective in suppressing the CD4 + T cell activation and the specific autoantibody production. Furthermore, we found that these Bregs aberrantly produced high level of TNF-α in BP patients. Meantime, etanercept could down-regulate the BP autoantibody production. All these result highlight that Bregs in BP appear phenotypically pro-inflammatory by their cytokine profile and defective in immunosuppressive function, suggesting that Bregs play a pro-inflammatory role rather than a regulatory role in the pathogenesis of BP.
BP is a prevalent autoimmune blistering disease caused by autoantibodies against BP180. Studies have found that several subsets of immune cells, including Th1 cells, Th2 cells and Treg cells, are involved in the production of BP autoantibodies 20,21 . Our previous study also showed that the frequency of follicular T helper cells also contribute to BP by producing IL-21 22 . However, whether Breg cells are involved in the process is still unknown.
Bregs are a small population of B cells that participates in immunomodulation and in suppression of immune responses 23 . Directly, Bregs can interact with cognate T cell and control Treg cell induction 24 . Indirectly, Breg cells suppress the differentiation of Th1 and Th17 cells by suppressing pro-inflammatory cytokine production by To date, whether Breg cells uniquely derive from a specific progenitor or originate within conventional B cell subsets is still unclear, which make it difficult to identify the exact phenotypes of Bregs 29 . However, there is a consensus that Bregs suppress the immune response are mainly through IL-10 production or contact-dependent suppression manner 23 . In humans, both CD19 + CD24 hi CD38 hi and CD19 + CD24 hi CD27 + B cells are shown to control immune responses by secreting IL-10 10,30 . Nevertheless, these two subsets show distinct effects on T cells. Bregs with the immature B cell marker CD38 are considered to be able to induce the development of T regulatory cells while limiting the differentiation of Th1 and Th17 cells 31 . And Bregs with memory B cell marker CD27 are mainly responsible for suppressing the activity of CD4 + T cells. Previously, Iwata et al. confirmed that CD19 + IL-10 + Bregs correspond to CD24 hi CD27 + B cells in humans 10 . Hence, we focus on investigating the frequencies and immune-regulatory function of CD19 + CD24 hi CD27 + Bregs in BP patients. Accumulated evidence has shown that the frequency of Bregs is aberrant in human autoimmune diseases 32 . Daien et al. found that the number of CD19 + IL-10 + Bregs was decreased in patients with rheumatoid arthritis and inversely correlated with disease activity 33 . However, the frequency of CD24 hi CD38 hi and CD24 hi CD27 + Bregs was similar in patients with rheumatoid arthritis and healthy controls 34 . Thus, it is still controversial about the frequency of Bregs in autoimmune diseases. In our study, we focused on the change of Bregs in BP patients and found that both circulating CD19 + CD24 hi CD27 + Bregs and IL-10 + CD19 + Bregs increased in BP patients compared to healthy controls. Further, we confirmed that CD19 + CD24 hi CD27 + Bregs from BP patients and healthy controls have the ability to secreting IL-10. Considering the regulatory function of Bregs on immune response, what is the role of elevated Bregs in the pathogenesis of BP? Our in vitro study provides evidence that CD19 + CD24 hi CD27 + Bregs from BP patients were defective in suppressing autoantibody production. This result were similar with the study in in patients with pemphigus that CD19 + CD24 hi CD38 hi Bregs were elevated in pemphigus patients and were defective regulatory function on T helper 1 cells 35 . Collectively, our study indicates that the modified function of Bregs may be a critical cause of BP.
Bregs are considered to suppress the activation of CD4 + T cells mainly by secreting IL-10 10 . In addition, IL-10 is an important anti-inflammatory cytokine and several studies showed that the level of IL-10 was decreased in T cell mediated autoimmune diseases, such as diabetes, psoriasis and rheumatoid arthritis, which indicates that decreased levels of IL-10 may cause activation of T cells, further leading to autoimmune diseases 36,37 . However, we found that CD19 + CD24 hi CD27 + Bregs produced comparable IL-10 between BP patients and healthy controls. Further, we showed that the number of IL-10 producing B cells were even increased in BP patients. Meanwhile, we noticed that the mRNA level of IL-10 was increased in PBMCs, while the serum level of IL-10 was comparable in BP patient compared with healthy controls (Sup Fig. 4c and d). It seems our results contradict with previous reports on T cell mediated autoimmune diseases. Nevertheless, several studies have shown that IL-10 level is elevated in some autoantibody-mediated autoimmune diseases, such as SLE and pemphigus vulgaris, and reduced IL-10 production is usually associated with remission 30,35 . In addition, IL-10 antagonists are effective in treating animal models of SLE 38 . All these suggest that modified function of Bregs that contribute to the pathogenesis of BP are independent on IL-10 production.
CD19 + CD24 hi CD27 + Bregs were mainly responsible for suppressing T cell activation, which is necessary for autoantibody production in B cells 10 . Thus, we investigated the effect of CD19 + CD24 hi CD27 + Bregs from patients on the activation of T cells. We found that BP patient-derived CD19 + CD24 hi CD27 + Bregs showed modified function in suppressing the proliferation and the pro-inflammatory cytokines secretion in CD4 + T cells. These findings suggest that Bregs contribute to the pathogenesis of BP through this shift in function to control the activation of CD4 + T cells.
Our result and other studies have indicated that Bregs showed modified function in immune-suppressive function in certain autoimmune disease 11 , but the underlining mechanism are still unclear. Recent study reported that pDCs induce the differentiation of Breg cells in an IFN-α-dependent manner in health individuals and this kind of regulation are defective in SLE patients which leading to impairment of Bregs in immune-suppressive function 39 . However, our result showed that Bregs gain the function to produce more inflammatory cytokines (mainly TNF-α, INF-γ) in BP patients. It seems controversial between the pro-inflammatory phenotype and weak immune-suppressive function of Breg in BP patients. However, recent studies proposed that immunosuppression is not the purview of a devoted Breg cell lineage with a specific phenotype but rather is the outcome of the dynamic balance between multiple B cell subsets and other cells of the immune system 26 . Functional flexibility of Breg subsets maybe responsible for the contradiction between phenotype and function in our study. In addition, more resolution of Breg phenotype will be helpful to understand the function in BP pathogenesis.
TNF-α is considered to be a critical cytokine for the initiation and perpetuation of inflammation. Studies showed that TNF-α contributes to many human autoimmune diseases by promoting the expansion and survival of T cells, including diabetes, rheumatoid arthritis, and psoriasis 13 . In healthy controls, TNF-α is mainly produced by pro-inflammatory lymphocytes and β-glucan-activated B lymphocytes could up-regulate pro-inflammatory cytokines TNF-α 14 . we further found that TNF-α blockade could suppress the autoantibody production, suggesting that defective function of Bregs is due to the production of TNF-α. Therefore, we speculated that Bregs play a pro-inflammatory role rather than a regulatory role in the pathogenesis of BP 40 .
However, there have several limitations. Firstly, as the minimal difference of increased frequency in Bregs, it just reflect a trend to an increased proportion of one particular type of Bregs in BP. The mechanistic and functional importance of this trend still need further investigation. Secondly, this study illustrated the pro-inflammatory function of Bregs and the crucial role of TNF-α in Bregs dysfunction in BP. However, it is unclear whether these findings are a general mechanisms of autoimmunity or just a phenomena of BP. Current finding would be more significance by including other antibody mediated autoimmune patients in this study. Finally, current result only uncovered the change of TNF-α in dysfunctional Bregs in BP. It would be more valuable to discuss the molecules or pathways changes with high throughput approach.
In summary, we found that frequency of circulating CD19 + CD24 hi CD27 + Bregs were increased in BP patients, and these cells were defective in suppressing autoantibody production and CD4 + T cell activation. Moreover, we provide strong evidence to suggest that the decreased suppressive capacity of Bregs is due to aberrant TNF-α production, rather than decreased IL-10 production. Our results provide new insights into the role of Bregs in the pathogenesis of BP.

Methods
Patients and healthy controls. Blood  ultrafiltration through Amicon Ultra 15-mL filters. The molecular weight and purity of the purified GST-NC16A fusion protein were estimated by SDS-PAGE electrophoresis. Proteins (10 μg) were separated by SDS-PAGE and stained by coomassie brilliant blue solution. Proteins were detected using a chemiluminescence detection kit (KPL, Gaithersburg, MD). Serial dilutions of patient-derived serum samples were added to a 24-well plate pre-coated with the GST-NC16A fusion protein. The affinity of BP180-NC16A binding to autoantibodies were analyzed by enzyme-linked immune sorbent assay (ELISA).

Function of Bregs in suppressing autoantibody production.
To identify anti-NC16A antibody specifically produced by BP patient-derived PBMCs, PBMCs from BP patients and healthy controls were cultured in 24-well plates (1.5 × 10 6 cells per well) pre-coated with purified NC16A protein (5.0 μg/mL, 300 μL per well). Next, to test the effect of Bregs on autoantibody production of BP patients, PBMCs isolated from BP patients with or without Bregs depletion were cultured in 24-well plates (1.5 × 10 6 cells per well) pre-coated with purified NC16A protein (5.0 μg/mL, 300 μL per well). In addition, patient-derived PBMCs (1 × 10 5 per well) were co-cultured with Bregs (3 × 10 4 cells per well) from healthy controls or another patients in 96-well plates coated with the NC16A protein (5.0 μg/mL, 100 μL per well). Then, the supernatants were harvested 72 h later, and specific anti-NC16A antibody levels were determined by ELISA assay.
To investigate the role of TNF-α in autoantibody production, patient-derived PBMCs were treated with etanercept (Pfizer, USA) to block the effect of TNF-α. Because the therapeutic range of residual serum concentration is between 1 and 10 μg/ml, we used 10 μg/ml of etanercept to incubate PBMCs for 72 hours. Then the supernatants were collected, the autoantibody production were determined by Elisa assay.
Enzyme-linked immune sorbent assay (ELISA) and quantitative PCR (qPCR). Levels of IL-10 in serum or culture supernatants were measured by using a human IL-10 ELISA kit according to the manufacturer's protocol (R&D Systems). The levels of anti-NC16A antibody were measured using a human ELISA kit (Medical&Biological laboratories, LTD. KDX Nagoya Sakae Bldg, Japan). The optical density (OD) was read at 450 nm in a microplate reader (Bio-Rad Model 680, CA, USA).
Total RNA from the PBMCs and sorted Bregs was extracted by using TRIzol (Takara) according to the manufacturer's instruction. cDNAs were then generated using a PrimeScript RT regent kit with 1 μg of total RNA per reaction. Quantitative real-time PCR was conducted using the SYBR Premix Ex Taq. β-Actin was used as an internal control. Samples were normalized to the independent control housekeeping gene β-actin and were reported according to the ΔΔCT method as RNA fold increase: 2 ΔΔCT = 2 ΔCT sample − 2 ΔCT reference . The sequences of each primers are given in Supplementary Table S2. Statistical analysis. Aggregate data are presented as the mean ± SD. The two-tailed Student's t test was used to analyze difference between two groups, and more than three groups were analyzed by one-way ANOVA followed by Bonferroni corrections for post hoc t-test. All analyses were performed using GraphPad Prism software, version 5 (GraphPad Software Inc, CA, USA).