Circulating regulatory Tfh cells are enriched in patients with chronic hepatitis B infection and induce the differentiation of regulatory B cells

Circulating regulatory Tfh cells are enriched in patients with chronic hepatitis B infection and induce the differentiation of regulatory B cells

Experimental Cell Research xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Experimental Cell Research journal homepage: www.elsevier.co...

NAN Sizes 0 Downloads 21 Views

Experimental Cell Research xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Experimental Cell Research journal homepage: www.elsevier.com/locate/yexcr

Circulating regulatory Tfh cells are enriched in patients with chronic hepatitis B infection and induce the differentiation of regulatory B cells Rongxin Wanga, Ruiling Xieb, Zongchang Songc,



a

Department of Infection, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, Henan Province, China Department of Respiration, Yellow River Central Hospital, Zhengzhou, Henan Province, China c Department of Hematology & Oncology, The 155th Central Hospital of PLA/The Key Laboratory of Hematology of Kaifeng City, Kaifeng, Henan Province, China b

A R T I C LE I N FO

A B S T R A C T

Keywords: HBV Tfh CXCR5 B cell Breg

Chronic hepatitis B virus (HBV) infection is a complex disease with dysregulations in the immune system. Follicular helper T (Tfh) cells are professional B helper cells that are crucial to the development of antibody responses and are involved in a variety of diseases. In this study, we examined the circulating Tfh cells in patients with chronic HBV infection. We observed that CD3+CD4+CXCR5+ circulating Tfh cells contained a CD25+Foxp3+ Treg-like subset that was significantly enriched in patients with chronic HBV infections. The CD25+ Tfh subset presented distinctive cytokine secretion profile, such as lower interferon (IFN)-γ and interleukin (IL)−17, and higher transforming growth factor (TGF)-β secretion, compared to the CD25- Tfh subset. When incubated with autologous naive CD10-CD27-CD19+ B cells, the CD25+ Tfh subset was less capable of mediating CD20-/loCD38+ plasmablast differentiation than the CD25- Tfh subset. In terms of Ig production, CD25+ Tfh cells were more potent at inducing IgM but less potent at inducing IgG and IgA than CD25- Tfh cells. Interestingly, B cells following incubation with CD25+ Tfh cells presented elevated regulatory function, with higher production of IL-10 and enhanced capacity of suppressing autologous CD8+ T cell inflammation. In the chronic HBV-infected patients, the frequency of IL-10+ B cells and the HBV viral load were positively correlated with the frequency of CD25+Foxp3+ CD4+CXCR5+ Tfh cells. Together, this study presented that CD25+Foxp3+ Treg-like Tfh cells were enriched in chronic HBV-infected patients and could promote regulatory B cell functions.

1. Introduction Over two billion individuals worldwide are currently exposed to hepatitis B virus (HBV), and approximately 400 million of these individuals are living with chronic HBV infection [1]. Carriers of HBV are predisposed to a number of hepatic pathologies, such as liver cirrhosis and hepatocellular carcinoma. However, eradication of HBV has not been achievable in most chronically infected individuals. Multiple lines of evidence suggest that some individuals, especially those that encounter the virus at birth or during childhood, develop an inadequate immune response against HBV and fail to clear the virus during the acute phase, thus resulting in chronic infections [2,3]. Hence, in order to develop an effective virus clearance strategy in early infected individuals, the immunoregulatory mechanisms should be investigated in further detail. Regulatory B (Breg) cells are critical regulators of immune responses in cancers, autoimmune diseases, and chronic virus infections [4,5]. By

expressing regulatory cytokines interleukin (IL)−10 and transforming growth factor (TGF)-β, regulatory B cells could suppress effector T cellmediated secretion of interferon (IFN)-γ and tumor necrosis factor (TNF)-α, and promote Foxp3+ Treg cell differentiation [6,7]. Breg cells could also suppress the function of antigen-presenting dendritic cells [8]. In chronic hepatitis B infections, Das et al. found that B cells contributed to the IL-10 upregulation in patient sera and the frequency of IL-10-expressing immature CD19+CD24hiCD38hi B cells correlated with disease flares [9]. Gong et al. presented that the frequency of IL10-producing Breg cells was negatively correlated with the frequency of Th1 cells in chronic HBV infected individuals [10]. Liu et al. demonstrated that CD19+CD24hiCD38hi B cells suppressed T effector cells but promoted Treg cells, and that depletion of Breg cells resulted in lower Treg, CTLA-4, IL-10, and TGF-β levels [11]. These studies together established that Breg cells presented critical immunosuppressive roles in chronic hepatitis B infections. However, it remains unclear how regulatory B cells were induced in HBV-infected individuals.

⁎ Correspondence to: Department of Hematology & Oncology, The 155th Central Hospital of PLA/The Key Laboratory of Hematology of Kaifeng City, No.3 Yi Yuan Qian Road, Kaifeng, Henan Province 475003, China. E-mail address: [email protected] (Z. Song).

https://doi.org/10.1016/j.yexcr.2018.02.031 Received 16 November 2017; Received in revised form 11 February 2018; Accepted 24 February 2018 0014-4827/ © 2018 Elsevier Inc. All rights reserved.

Please cite this article as: Wang, R., Experimental Cell Research (2018), https://doi.org/10.1016/j.yexcr.2018.02.031

Experimental Cell Research xxx (xxxx) xxx–xxx

R. Wang et al.

2.4. Luminex

The follicular helper T (Tfh) cells are recently identified as the main B cell helpers among all CD4+ T cell subsets [12]. By their characteristic CXCR5 expression on the surface, Tfh cells preferentially localize to the CXCL13-enriched B cell follicles, where the Tfh cells maintains B cell germinal center reaction via the expression of CD40L, IL-4, and/or IL-21 [12]. Interestingly, circulating CXCR5+CD4+ T cells resemble germinal center (GC) Tfh cells phenotypically and functionally and can promote plasmablast differentiation and antibody production in vitro [13,14]. Given the various impacts of Tfh cells on B cell development, the question of whether Tfh cells are involved in the development of regulatory B cells is raised. Recently, a distinctive subset of Foxp3+ Treg cells with Bcl-6 and CXCR5+ expression is discovered [15]. This subset has been observed in the GC and in the peripheral blood, and is termed follicular regulatory T (Tfr) cells [15,16]. Tfr cells present features from both Tfh cells and Treg cells by expressing PD-1, ICOS and CTlA-4 [17]. Studies in mice have shown that Tfr cells can suppress the activation and germinal center reaction of B cells and Tfh cells [17]. The role of Tfr cells in infectious diseases remains unclear. In this study, we examined the possibility that Tfh cells and Tfr cells affected the development of Breg cells in chronic HBV-infected patients.

CD25+ and CD25Tfh cells were sorted as + + + CD3 CD4 CXCR5 CD25+ or CD3+CD4+CXCR5+CD25- lymphocytes, respectively, in FACSAria cytometer, and were incubated with Human T Activator (anti-CD3/CD28 beads; Thermo Fisher) for 24 h. Commercial assay kits for the following analytes, including IFN-γ, IL10, IL-17, IL-21, and TGF-β, were purchased from Thermo Fisher Scientific and the luminex assays were carried out on the harvested supernatants according to the manufacturer's protocol. 2.5. Ig ELISA

2. Methods

5 × 104 cells mature naive B cells were sorted as CD10-CD27-CD19+ B cells in FACSAria cytometer and were added to 5 × 104 sorted CD25+ or CD25- Tfh cells in 200 µL culture medium in a 96-well plate. 2 µg/mL SEB (Sigma) was added. A separate mature naive B cell only culture was maintained. Supernatant was collected from all culture wells on Day 6. The concentration of IgM, IgG and IgA was performed using corresponding commercial kits from Abcam. The concentration in the B cell only control was subtracted from the concentration in the Tfh-containing experiments to obtain the final Ig secretion levels in the presence of Tfh cells.

2.1. Participants

2.6. B cell analyses

Nineteen patients with confirmed chronic HBV infection and nineteen uninfected healthy control volunteers were recruited at the 155th Central Hospital of PLA. The patients included 12 males and 7 females between 20 and 55 years of age, with viral loads between 1.2 × 102 IU/ mL and 3.5 × 106 IU/mL and ALT levels between 21 U/L and 770 U/L. Healthy controls included 12 males and 7 females between 23 and 49 years of age with ALT levels between 10 U/L and 30 U/L. In addition, all patients and controls were tested negative for HCV-specific antibodies. All patients provided written informed consent.

Following incubation with Tfh cells, B cells were sorted as CD3-CD19lo/+ cells in FACSAria cytometer. A portion of the cells were lysed and the RNA was harvested using the RNeasy mini kit (Qiagen) and reverse-transcribed using the iScript cDNA Synthesis kit (Bio-Rad). TaqMan gene expression assays of IL10 and ACTB were then purchased from Thermo Fisher and performed according to the manufacturer's instructions. Another portion of the cells were incubated alone for 24 h and the supernatant was harvested for IL-10 Luminex assay. The rest of the cells were incubated with autologous CD8+ T cells, sorted as CD3+CD8+ lymphocytes in FACSAria cytometer, for 72 h in the presence of 2 µg/mL SEB (Sigma). The percentage of IFN-γ+ cells in CD8+ T cells was examined using flow cytometry.

2.2. Sample collection After receiving approval from the Ethics Board of the 155th Central Hospital of PLA, peripheral blood was collected from all patients and controls into anticoagulant-containing tubes. Standard Ficoll (Gibco) gradient centrifugation was then performed to obtain peripheral blood mononuclear cells (PBMCs).

2.7. Statistics Differences between two unmatched groups were assessed using Mann-Whitney test and between two matched groups were assessed using Wilcoxon matched-pairs test. Correlation was examined using Pearson correlation test. P values smaller than 0.05 were needed for significance. All tests were performed using Prism.

2.3. Cell sorting and flow cytometry 3. Results Anti-human CD3, CD4, CD8, CD10, CD19, CD20, CD25, CD27, CD38, CXCR5, IgD, and IL-10 monoclonal antibodies were purchased from BioLegend. Anti-human Foxp3 and IFN-γ were purchased from BD. All incubation steps were performed on ice in a dark and sterile condition. For the live sorting of cells in certain experiments, PBMCs were incubated with monoclonal antibodies against surface antigens for 30 min, washed twice, and then sorted via positive selection in a FACSAria cytometer (BD). For all other flow cytometry experiments, PBMCs were first incubated sequentially with Fixable Violet Dead Cell Stain (Thermo Fisher) for 30 min, washed twice, and then with monoclonal antibodies against surface antigens for 30 min. After removal of excess antibodies against surface antigens, cells were incubated in CytoFix/CytoPerm (BD) for 15 min, and washed once in Perm Wash solution (BD). Monoclonal antibodies against intracellular antigens, including anti-human Foxp3, IL-10, and/or IFN-γ, were dissolved in Perm Wash solution and added to cells at 5 µg per 105 cells for 30 min. Cells were then washed twice and fixed in 2% formaldehyde. Sample acquisition was performed in LSR cytometer (BD).

3.1. Frequency of circulating Tfh cells in chronic HBV-infected individuals It was previously demonstrated that circulating CD4+CXCR5+ T cells were significantly more potent in stimulating the plasmablast differentiation of naive B cells than circulating CD4+CXCR5- T cells [13,18]. Here, we identified circulating Tfh cells as CD3+CD4+CXCR5+ live lymphocytes (Fig. 1A). The frequency of circulating Tfh cells in CD4+ T cells was 9.60% ± 1.08% (mean ± standard deviation) in chronic HBV-infected individuals and was slightly higher on average than that in healthy controls at 8.69% ± 0.86% (Fig. 1B). 3.2. Circulating Tfh cells presented a CD25+Foxp3+ Treg-like subset that was upregulated in chronic HBV-infected individuals Based on chemokine receptor expression and cytokine secretion, it was shown that circulating Tfh cells could be subcategorized into Th1-, 2

Experimental Cell Research xxx (xxxx) xxx–xxx

R. Wang et al.

Fig. 1. Characterization of circulating Tfh cells in chronic HBV-infected and healthy subjects. (A) Gating strategy of CD3+CD4+CXCR5+ live lymphocytes in one representative chronic HBV-infected subject. Numbers represent the percentage of total cells in each gate. (B) The frequency of CD4+CXCR5+ circulating Tfh cells in total CD4+ T cells in chronic HBV-infected patients and in healthy controls. Mann-Whitney test. (C) The expression of CD25 and Foxp3 in CD4+CXCR5+ circulating Tfh cells in one representative healthy individual and one representative chronic HBV-infected patient. Both plots were pre-gated on CD3+CD4+CXCR5+ live lymphocytes as shown in panel (A). Numbers represent the percentage of cells in the top right quadrant. (D) The frequency of CD25+Foxp3+ cells in CD4+CXCR5+ circulating Tfh cells in chronic HBV-infected individuals and in healthy controls. Mann-Whitney test. *P < 0.05. ***P < 0.001.

Th2-, and Th17-like subsets with distinctive functions and regulatory mechanisms [13,19]. However, it is yet unclear whether a Treg-like subset could be found in the circulating Tfh population. We investigated this by staining circulating Tfh cells with CD25 and Foxp3, two markers upregulated in canonical Treg cells (Fig. 1C). The frequency of CD25+Foxp3+ Treg-like cells in circulating Tfh cells in chronic HBVinfected individuals was 16.22% ± 3.43%, which was significantly higher than that in healthy individuals at 8.84% ± 2.44% (Fig. 1D). 3.3. CD25+ Tfh cells presented different cytokine production profile from CD25- Tfh cells Although CD25+Foxp3+ Treg-like cells were discovered in the circulating Tfh cells, it remains unclear whether these cells resemble canonical Treg cells functionally. Hence, we investigated the cytokine production in CD25+ and CD25- Tfh cells. The CD25+ circulating Tfh cells presented lower capacity of IFN-γ and IL-17 secretion, comparable capacity of IL-10 and IL-21 secretion, and higher capacity of TGF-β secretion, compared to CD25- circulating Tfh cells (Fig. 2).

Fig. 2. Cytokine secretion by CD25+ and CD25- circulating Tfh cells in chronic HBVinfected subjects. The cytokine secretion by CD25+ and CD25- circulating Tfh cells from chronic HBV-infected subjects. CD25+ and CD25- circulating Tfh cells were sorted by flow cytometry and stimulated with anti-CD3/CD28 beads for 24 h. Cytokine concentration was measured by Luminex. N = 19 in each group. Mean ± SD. Mann-Whitney test. *P < 0.05.

CD20-/loCD38+ plasmablasts was found (Fig. 3A, lower right). The frequency of CD19lo/+CD3- B cells in the B-T coculture was significantly higher in Day 6 than in Day 0 for all CD25+ or CD25- Tfh cocultures, suggesting a proliferation of B cells relative to Tfh cells (Fig. 3B). The increase was significantly higher in the CD25- Tfh cocultures than in the CD25+ cocultures. Similarly, the increase in the frequency of CD20-/loCD38+ plasmablasts was significantly higher in CD25- Tfh cocultures than in CD25+ cocultures (Fig. 3C).

3.4. CD25+ Tfh cells presented different B cell help capacity from CD25Tfh cells Next, we investigated the capacity of B cell help mediated by CD25+ and CD25- Tfh cells. We incubated isolated CD25+ and CD25Tfh cells with autologous mature naive CD10-CD27-CD19+ B cells at a ratio of 1-to-1 and with stimulation using superantigen SEB. The plasmablast differentiation was examined on Day 6 of the coincubation by flow cytometry staining (Fig. 3A). At Day 0, the ratio of mature naive B cells and CD25+ or CD25- Tfh cells was confirmed by staining of CD19+CD3- B cells and CD19-CD3+ T cells (Fig. 3A, upper left). At Day 6, B cells were mainly identified based on CD3- expression, and were regated as CD19lo/+CD3- cells due to a downregulation of CD19 (Fig. 3A, upper right). At Day 0, few CD20-/loCD38+ plasmablasts could be identified (Fig. 3A, lower left), while at Day 6, a sizable population of

3.5. CD25+ Tfh cells promoted the production of different isotypes of antibodies Though it is known that circulating Tfh cells are critical in mediating antibody production from B cells, it remains unclear whether CD25+ circulating Tfh cells present this functionality. Hence, we 3

Experimental Cell Research xxx (xxxx) xxx–xxx

R. Wang et al.

Fig. 3. B cell help provided by CD25+ and CD25- circulating Tfh cells in chronic HBV-infected subjects. CD25+ or CD25- circulating Tfh cells and mature naive CD10-CD27-CD19+ B cells were incubated at 5 × 104 cells per 200 µL culture medium for each cell type, with stimulation from 2 µg/mL SEB. (A) Representative gating strategy of CD19+CD3- B cells on Day 0 and CD19lo/+CD3- B cells on Day 6 and of CD20-/loCD38+ plasmablasts on Day 0 and Day 6 of the coculture. Numbers represent the percentage of total cells in each gate. (B) The increase in B cells following 6-day coculture, calculated by the frequency of CD19lo/+CD3- cells on Day 6 minus the frequency of CD19+CD3- cells on Day 0. (C) The increase in plasmablasts following 6-day coculture, calculated by the frequency of CD20-/loCD38+ B cells on Day 6 minus that on Day 0. (D) The secretion of IgM, IgG, and IgA at the end of the coculture. Mann-Whitney test. *P < 0.05. **P < 0.01. ***P < 0.001.

3.7. CD8+ T cell inflammation was reduced in the presence of B cells incubated with CD25+ Tfh cells

examined the concentration of secreted antibodies in Day 6 supernatant of the mature naive B cell-circulating Tfh cell coculture. We found that both CD25+ and CD25- circulating Tfh cells were capable of mediating antibody secretion (Fig. 3D), but compared to CD25- Tfh cells, the CD25+ Tfh cells were more potent at promoting IgM secretion and less potent at promoting IgG and IgA secretion.

IL-10 is pivotal to B cell-mediated immune regulation and the frequency of regulatory B cells was positively associated with disease flares in chronic HBV-infected individuals [9]. Here, we investigated whether the B cells following incubation with CD25+ or CD25- Tfh cells presented regulatory activity, by incubating these B cells with autologous CD8+ T cells at 1-to-1 ratio. The frequencies of IFN-γ+ CD8+ T cells were then examined by intracellular staining (Fig. 4C). We found that CD8+ T cells incubated with B cells from CD25+ Tfh cocultures presented lower IFN-γ expression than CD8+ T cells incubated with B cells from CD25- Tfh cocultures (Fig. 4D).

3.6. B cells incubated with CD25+ Tfh cells presented stronger IL-10 secretion than B cells incubated with CD25- Tfh cells Next, we characterized the functions of B cells following incubation with CD25+ and CD25- Tfh cells. CD19lo/+CD3- B cells were isolated from coculture and incubated alone for an additional 24 h, and the IL10 secretion was examined. We found that supernatant from B cells from CD25+ Tfh cocultures contained higher levels of secreted IL-10 than supernatant from B cells from CD25- Tfh cocultures (Fig. 4A). One explanation was that B cells from CD25+ Tfh cocultures presented higher IL-10 secretion capacity than B cells from CD25- Tfh cocultures. However, it was also possible that more IL-10 were consumed by B cells from CD25- Tfh cocultures, which could also result in lower IL-10 concentration. Hence, we examined the IL-10 mRNA transcription levels in these B cells. B cells from CD25+ Tfh cocultures presented significantly higher IL-10 transcription than B cells from CD25- Tfh cocultures (Fig. 4B).

3.8. The frequency of CD25+Foxp3+ circulating Tfh cells was correlated with the frequency of IL-10+ circulating B cells and the viral load Having observed that CD25+ Tfh cells promoted the development of IL-10-producing B cells with capacity to reduce IFN-γ+ CD8+ T cells, we examined the frequency of IL-10+ B cells in the peripheral blood of chronic HBV-infected individuals. We observed that, in this cohort of chronic HBV-infected patients, the frequencies of CD25+Foxp3+ circulating Tfh cells was significantly correlated with the frequencies of IL10+ circulating B cells (Fig. 5A). Subsequently, we examined whether the frequency of CD25+Foxp3+ circulating Tfh cells was associated with disease severity in chronic HBV-infected patients. The HBV viral load was positively correlated with the frequency of CD25+Foxp3+ Tfh cells 4

Experimental Cell Research xxx (xxxx) xxx–xxx

R. Wang et al.

Fig. 4. B cell regulatory function after incubation with CD25+ and CD25- circulating Tfh cells in chronic HBV-infected subjects. (A) Following coculture with CD25+ or CD25- Tfh cells, CD19lo/ + CD3- B cells were sorted by flow cytometry and incubated alone for 24 h, and the level of secreted IL-10 was examined by Luminex. Mann-Whitney test. (B) IL-10 mRNA transcript levels were examined in CD19lo/+CD3- B cells immediately after isolation from coculture with CD25+ or CD25- Tfh cells. MannWhitney test. (C) CD8+ T cells and CD19lo/+CD3- B cells were incubated at a ratio of 1-to-1 and stimulated with 2 µg/mL SEB for 72 h. The representative gating of IFN-γ+ CD8+ T cells following incubation with CD19lo/+CD3- B cells from CD25+ or CD25- Tfh cell cocultures were shown. Numbers represent the percentage of total cells in each gate. (D) The frequency of IFN-γ+ CD8+ T cells following incubation with CD19lo/+CD3- B cells from CD25+ or CD25- Tfh cell cocultures. Mann-Whitney test. *P < 0.05. **P < 0.01.

(Fig. 5B), while no association between the frequency of CD25+Foxp3+ Tfh cells and the level of alanine aminotransferase (ALT) was observed (Fig. 5C).

and because circulating Tfh cells functionally resembled GC Tfh cells in promoting antibody production [13,18,20]. For the above reasons, we investigated the circulating CD4+CXCR5+ T cells, identified as circulating Tfh cells by other studies, in chronic HBV-infected patients. We presented two main discoveries. First, the CD4+CXCR5+ circulating Tfh cells contained a CD25+Foxp3+ subset that was significantly enriched in patients with chronic HBV infections and the CD25+ Tfh cells and CD25- Tfh cells presented distinctive characteristics, and second, the CD25+ circulating Tfh cells promoted B cell-mediated IL-10 production, as well as the development of B cells that suppressed IFN-γ inflammation from autologous CD8+ T cells. The positive associations between CD25+Foxp3+ Tfh frequency and IL-10+ B cell frequency and HBV viral load suggest that the CD25+Foxp3+ Tfh cells likely present deleterious functions in chronic HBV infection and may serve as a prognostic factor and a treatment target. In vivo experiments should be

4. Discussion The critical role of Tfh cells in modulating the course of inflammation during disease is increasingly recognized. As professional B cell helpers, Tfh cells are crucial to the development of humoral antibody response [12]. In addition, Tfh cells secretes a variety of cytokines, one of which is IL-21 that was implicated in the development of T cell cytotoxicity and shown to promote anti-HBV immunity [3]. Currently, the circulating Tfh cells are the most commonly examined Tfh population in patients with human diseases because peripheral blood samples present better accessibility than lymph node and GC samples,

Fig. 5. The correlation between the frequencies of CD25+Foxp3+ circulating Tfh cells and characteristics of the chronic HBV-infected patients. (A) The correlation between CD25+Foxp3+ circulating Tfh cell frequency and IL-10+ B cell frequency. The frequency of IL-10+ B cells in the peripheral blood was examined by intracellular staining 24 h after stimulation with PMA and ionomycin. Line represents best fit in a linear model. (B) The correlation between CD25+Foxp3+ circulating Tfh cell frequency and HBV viral load in the serum. (C) The correlation between CD25+Foxp3+ circulating Tfh cell frequency and ALT level in patients. The HBV viral load and the ALT level were measured at the time of sample collection. P represents the Pearson correlation coefficient.

5

Experimental Cell Research xxx (xxxx) xxx–xxx

R. Wang et al.

performed to investigate the role of CD25+Foxp3+ Tfh cells in infectious diseases. In previous examinations, the Th1-, Th2-, and Th17-like cells were identified in circulating Tfh cells. We found that a CD25+Foxp3+ subset could be recovered in circulating Tfh cells, and this subset was significantly enriched in chronic HBV-infected patients. Together with a previous examination by Wang et al. [21], we found an enrichment of CD4+CXCR5+Foxp3+ Treg-like Tfh cells and IL-10+ Breg cells in chronic HBV-infected patients, but our investigation presented an additional discovery that the CD25+Foxp3+ Tfh cells could directly stimulate the development of IL-10+ Breg cells. We do not yet know which mechanisms are responsible for the upregulation of CD25+Foxp3+ Tfh cells in chronic HBV infection. In GCs, a population of CXCR5+Foxp3+ follicular regulatory T (Tfr) cells has been identified and has been shown to moderate GC responses by suppressing conventional GC Tfh cells and B cells [15,22–24]. Studies have suggested that GC Tfr cells originate from non-CXCR5-expressing canonical Treg cells and likely require CD28, ICOS, and Tfh cell-mediated support, but the precise conditions of Tfr development is unclear [17]. At the moment, it is unclear whether the CXCR5+Foxp3+ circulating Tfh cells in our study were the counterparts of Tfr cells in GCs. Samples of secondary lymphoid structures from chronic HBV-infected individuals are required to examine the phenotype and activity of GC Tfr cells in chronic HBV infection, and to compare and contrast GC Tfr cells with circulating Treg-like Tfh cells. Functionally, the CXCR5+Foxp3+ circulating Tfh cells in our study presented several similarities with the GC Tfr cells in mice. For example, the IgG response was suppressed in the presence of both of these populations [15,23]. Functional differences between the GC Tfr cells and our CXCR5+Foxp3+ circulating Tfh cells could also be observed. For example, IL-21 expression was not detected in GC Tfr cells but was detected in our CD25+ circulating Tfh cells, with no quantitative difference from the IL-21 expression by CD25- circulating Tfh cells. However, due to technical difficulties, including the lack of sufficient cells for intracellular staining and the inability of detecting Foxp3 expression without killing the cells, our investigation of cytokine expression was performed using CD25+ cells instead of CD25+Foxp3+ cells. There was also a CD25+Foxp3- population present and it remains unclear whether the IL-21, or the expression of other cytokines, was contributed by the CD25+Foxp3- population. This caveat should be addressed in future experiments using animal models. Another main discovery was that B cells following incubation with CD25+ circulating Tfh cells presented higher IL-10 expression and capacity to suppress IFN-γ expression by autologous CD8+ T cells in vitro. Subsequently, we demonstrated that the frequency of CD25+Foxp3+ circulating Tfh cells was correlated with the frequency of IL-10+ circulating B cells in chronic HBV-infected individuals. These results seem to suggest that CD25+Foxp3+ Tfh cells had the capacity of promoting the differentiation of Breg cells. However, correlation does not prove causation, and more in vivo experiments using conditional knockouts of CD25+Foxp3+ Tfh cells are required to establish whether the development of Breg cells are dependent on CD25+Foxp3+ Tfh cells or GC Tfr cells. Furthermore, Breg cells are a heterogeneous group with a variety of phenotypical and functional differences among the distinctive subsets [25]. Here, we focused on the B cells’ capacity to producing IL-10 and suppress CD8+ T cell inflammation because these features were previously associated with a lack of robust immune response in chronic HBV infections [9,11]. It remains to be examined whether CD25+Foxp3+ Tfh cells or GC Tfr cells are involved in one or more of these subsets.

References [1] C.-M. Tang, T.O. Yau, J. Yu, Management of chronic hepatitis B infection: current treatment guidelines, challenges, and new developments, World J. Gastroenterol. 20 (2014) 6262–6278, http://dx.doi.org/10.3748/wjg.v20.i20.6262. [2] A. Bertoletti, C. Ferrari, Innate and adaptive immune responses in chronic hepatitis B virus infections: towards restoration of immune control of viral infection, Gut 61 (2012) 1754–1764, http://dx.doi.org/10.1136/gutjnl-2011-301073. [3] J. Publicover, A. Goodsell, S. Nishimura, S. Vilarinho, Z. Wang, L. Avanesyan, et al., IL-21 is pivotal in determining age-dependent effectiveness of immune responses in a mouse model of human hepatitis B, J. Clin. Investig. 121 (2011) 1154–1162, http://dx.doi.org/10.1172/JCI44198.1154. [4] K. Yanaba, J.-D. Bouaziz, K.M. Haas, J.C. Poe, M. Fujimoto, T.F. Tedder, A regulatory B cell subset with a unique CD1dhiCD5+ phenotype controls T cell-dependent inflammatory responses, Immunity 28 (2008) 639–650 〈http://www. hubmed.org/display.cgi?uids=18482568〉. [5] E.C. Rosser, P.A. Blair, C. Mauri, Cellular targets of regulatory B cell-mediated suppression, Mol. Immunol. 62 (2014) 296–304, http://dx.doi.org/10.1016/j. molimm.2014.01.014. [6] K.M. Lee, R.T. Stott, G. Zhao, J. SooHoo, W. Xiong, M.M. Lian, et al., TGF-β-producing regulatory B cells induce regulatory T cells and promote transplantation tolerance, Eur. J. Immunol. 44 (2014) 1728–1736, http://dx.doi.org/10.1002/eji. 201344062. [7] P.A. Blair, L.Y. Noreña, F. Flores-Borja, D.J. Rawlings, D.A. Isenberg, M.R. Ehrenstein, et al., CD19(+)CD24(hi)CD38(hi) B cells exhibit regulatory capacity in healthy individuals but are functionally impaired in systemic Lupus Erythematosus patients, Immunity 32 (2010) 129–140, http://dx.doi.org/10.1016/ j.immuni.2009.11.009. [8] M. Matsumoto, A. Baba, T. Yokota, H. Nishikawa, Y. Ohkawa, H. Kayama, et al., Interleukin-10-producing plasmablasts exert regulatory function in autoimmune inflammation, Immunity 41 (2014) 1040–1051, http://dx.doi.org/10.1016/j. immuni.2014.10.016. [9] A. Das, G. Ellis, C. Pallant, A.R.R. Lopes, P. Khanna, D. Peppa, et al., IL-10-producing regulatory B cells in the pathogenesis of chronic hepatitis B virus infection, J. Immunol. 189 (2012) 3925–3935, http://dx.doi.org/10.4049/jimmunol.1103139. [10] Y. Gong, C. Zhao, P. Zhao, M. Wang, G. Zhou, F. Han, et al., Role of IL-10-producing regulatory B cells in chronic hepatitis B virus infection, Dig. Dis. Sci. 60 (2015) 1308–1314, http://dx.doi.org/10.1007/s10620-014-3358-1. [11] Y. Liu, L.-s. Cheng, S.-d. Wu, S.-q. Wang, L. Li, W.-m. She, et al., IL-10-producing regulatory B-cells suppressed effector T-cells but enhanced regulatory T-cells in chronic HBV infection, Clin. Sci. 130 (2016) 907–919, http://dx.doi.org/10.1042/ CS20160069. [12] S. Crotty, Follicular helper CD4 T cells (TFH), Annu. Rev. Immunol. 29 (2011) 621–663, http://dx.doi.org/10.1146/annurev-immunol-031210-101400. [13] R. Morita, N. Schmitt, S.-E.E. Bentebibel, R. Ranganathan, L. Bourdery, G. Zurawski, et al., Human blood CXCR5+CD4+ T cells are counterparts of T follicular cells and contain specific subsets that differentially support antibody secretion, Immunity 34 (2011) 108–121, http://dx.doi.org/10.1016/j.immuni.2010.12.012. [14] H. Ueno, J. Banchereau, C.G. Vinuesa, Pathophysiology of T follicular helper cells in humans and mice, Nat. Immunol. 16 (2015) 142–152, http://dx.doi.org/10.1038/ ni.3054. [15] Y. Chung, S. Tanaka, F. Chu, R.I. Nurieva, G.J. Martinez, S. Rawal, et al., Follicular regulatory T cells expressing Foxp3 and Bcl-6 suppress germinal center reactions, Nat. Med. 17 (2011) 983–988, http://dx.doi.org/10.1038/nm.2426. [16] P.T. Sage, D. Alvarez, J. Godec, U.H. Von Andrian, A.H. Sharpe, Circulating T follicular regulatory and helper cells have memory-like properties, J. Clin. Investig. 124 (2014) 5191–5204, http://dx.doi.org/10.1172/JCI76861. [17] P.T. Sage, A.H. Sharpe, T follicular regulatory cells, Immunol. Rev. 271 (2016) 246–259, http://dx.doi.org/10.1111/imr.12411. [18] D. Liu, J. Liu, J. Wang, C. Liu, S. Yang, Y. Jiang, Distinct phenotypic subpopulations of circulating CD4(+)CXCR5(+) follicular helper T cells in children with active IgA vasculitis, BMC Immunol. 17 (2016) 40, http://dx.doi.org/10.1186/s12865-0160176-6. [19] S. Crotty, T follicular helper cell differentiation, function, and roles in disease, Immunity 41 (2014) 529–542, http://dx.doi.org/10.1016/j.immuni.2014.10.004. [20] M. Locci, C. Havenar-Daughton, E. Landais, J. Wu, M.A. Kroenke, C.L. Arlehamn, et al., Human circulating PD-1+CXCR3-CXCR5+ memory Tfh cells are highly functional and correlate with broadly neutralizing HIV antibody responses, Immunity 39 (2013) 758–769, http://dx.doi.org/10.1016/j.immuni.2013.08.031. [21] L. Wang, J. Qiu, L. Yu, X. Hu, P. Zhao, Y. Jiang, Increased numbers of CD5 + CD19 + CD1dhighIL-10 + Bregs, CD4 + Foxp3+ Tregs, CD4 + CXCR5 + Foxp3+ follicular regulatory T (TFR) cells in CHB or CHC patients, J. Transl. Med. 12 (2014) 251, http://dx.doi.org/10.1186/s12967-014-0251-9. [22] P.T. Sage, A.H. Sharpe, T follicular regulatory cells in the regulation of B cell responses, Trends Immunol. 36 (2015) 410–418, http://dx.doi.org/10.1016/j.it. 2015.05.005. [23] I. Wollenberg, A. Agua-Doce, A. Hernandez, C. Almeida, V.G. Oliveira, J. Faro, et al., Regulation of the germinal center reaction by Foxp3+ follicular regulatory T cells, J. Immunol. 187 (2011) 4553–4560, http://dx.doi.org/10.4049/jimmunol. 1101328. [24] M.A. Linterman, W. Pierson, S.K. Lee, A. Kallies, S. Kawamoto, T.F. Rayner, et al., Foxp3+ follicular regulatory T cells control the germinal center response, Nat. Med. 17 (2011) 975–982, http://dx.doi.org/10.1038/nm.2425. [25] E.C. Rosser, C. Mauri, Regulatory B cells: origin, phenotype, and function, Immunity 42 (2015) 607–612, http://dx.doi.org/10.1016/j.immuni.2015.04.005.

Conflict of interests The authors declare no conflict of interests.

6