Introduction
B-cell maturation in adults occurs in steps. First, in the bone marrow, stem cells undergo a series of precursor stages during which they rearrange their immunoglobulin (Ig) genes to generate a wide range of unique antigen-binding specificities to develop into immature/transitional B cells. Then, in the periphery, they mature from transitional to fully mature naïve B cells. Each developmental step is tightly controlled by the expression and function of the B-cell receptor (BCR) [
1]. In mice, transitional B cells can be subdivided into two developmental subsets, T1 and T2, based on expression of CD21 and IgD. CD24
hiCD21
loIgD
lo T1 and CD24
hiCD21
hiIgD
hi T2 cells appear to have different population dynamics, and require different maturation signals [
2]. This multistep development process during the maturation from transitional B cells into naïve B cells has also been identified recently in humans. Based on CD38 expression levels, human peripheral blood immature B cells could be subdivided into CD27
−CD38
hiIgD
+ transitional B cells and CD27
−CD38
intIgD
+ pre-naïve B cells [
3,
4]. The comprehensive phenotyping and initial functional analysis clearly demonstrated that pre-naïve B cells were a maturation intermediate between transitional and naïve B cells with unique properties and functions. Notably, human peripheral maturational B-cell subsets, including pre-naïve B cells, express CD5, whereas in mice, CD5 is expressed on specialized B-cell subset B-1 B cells [
3,
5].
The essential role of mature B cells is the production of antigen (Ag)-specific antibodies (Abs) during humoral immunity by differentiating into plasma cells [
6]. B cells also mediate many other functions essential for immune homeostasis. B cells are required for initiation of T-cell immune responses by presenting Ags, providing co-stimulation, and producing cytokines to activate and expand effectors and memory T-cell populations [
7]. In addition, B cells can negatively regulate immune responses by directly inhibiting CD4
+ T cells and by inducing regulatory T cells (Tregs) through production of the cytokine interleukin (IL)-10 [
8]. These effector and regulatory B-cell functions contribute to both normal immune regulation and also immunopathology [
7,
9]. Though immature, peripheral B cells during development have a distinguished role in immune responses apart from the mature B cells. They elicit T cell-independent rapid antibody responses to polysaccharides, lipids, and other non-protein antigens which cannot bind to major histocompatibility complex (MHC) molecules [
10]. In mice, immature B cells with specialized functions were identified. Marginal zone (MZ) B cells and B-1 B cells known to elicit T cell-independent responses to antigens of microbes in mucosal tissues and microbes that enter peritoneum have been reported [
11,
12]. Distinct IL-10-producing regulatory B cells (Bregs) with immature phenotype also have been recently identified in mice and also in humans [
13,
14]. However, functions of peripheral immature B cells during normal immune responses are less well characterized and remain to be delineated in humans.
In this respect, pre-naïve B cells are an interesting human peripheral immature B-cell population worthy of further investigation. Pre-naïve B cells were phenotypically distinct from transitional and naïve B cells, expressing intermediate levels of CD38, CD10, CD9, and the ABCB1 transporter, and were also shown to be capable of differentiating into naïve B cells [
3]. Pre-naïve B cells manifested a unique set of functional characteristics [
3]. These cells had typical characteristics of immature B cells with shorter life span and defective responses to BCR stimulation. However, pre-naïve B cells were comparable to adult B cells in their capacity to respond to signaling through CD40. Moreover, collaboration with activated CD4
+ T cells resulted in their differentiation into plasma cells with the secretion of Abs. Uniquely, pre-naïve B cells expressed CD5, a marker of the B-1 subset of murine B cells, which can function as a negative regulator of BCR signaling and promote maintenance of tolerance to auto-Ag [
15‐
17]. Because of these unique functional characteristics, pre-naïve B cells have the potential to perform specialized functions during normal immune responses, not fully explored previously. In this study, we set out to determine the functional role of pre-naïve B cells during normal immune responses by activating pre-naïve B cells through CD40. In addition, we carried out experiments by using pre-naïve B cells from patients with systemic lupus erythematosus (SLE) to determine the role of these cells during pathologic immunity.
Methods
Isolation of human B and T cells
Blood samples were obtained from 50 healthy adult donors and from nine patients with SLE. Healthy individuals included 21 men and 29 women with a mean age of 29 years (range of 25–35 years). Individuals who were taking immunosuppressive drugs or who had a disease potentially affecting the immune system were excluded. Patients with SLE included one man and eight women with a mean age of 40 years (range of 25–64 years). All patients with SLE had active disease at the time of the blood sample collection (an SLE Disease Activity Index score of 12–26) and received low-dose glucocorticoid and hydroxychloroquine. One patient was taking mycofenolate mofetil in addition to the above treatment. All patients and healthy volunteers provided written informed consent prior to sample collection. This work was approved by the institutional review board of Ewha Womans University Hospital and the institutional review board of Seoul National University Hospital. Peripheral blood mononuclear cells were isolated by centrifugation on Biocoll Separating solution (Biochrom, Cambridge, UK), and B cells were enriched by using Rosette Sep human B-cell enrichment cocktails (StemCell Technology, Vancouver, BC, Canada). Enriched B cells were stained with a combination of anti-CD20-PerCP, anti-CD27-FITC, and anti-CD38-PE-Cy7 (all from BD Biosciences, Franklin Lakes, NJ, USA). Cells were sorted into CD20+CD27−CD38lo naïve, CD20+CD27−CD38Int pre-naïve, and CD20+CD27+ memory B cells by using a BD FACSAriaIII™ and very narrow gates centered around the median CD38 fluorescence intensity of each B-cell subset. CD4+CD25− T cells were sorted after staining with a combination of anti-CD4-APC-Cy7 and anti-CD25-PE.
Cell culture
Human CD154 expressing L cells (CD154-L cells, kindly provided by Rizgar A. Mageed, Queen Mary University of London) were cultured in Dulbecco’s modified Eagle’s medium containing 5 % fetal bovine serum (FBS) (Life Technologies, Carlsbad, CA, USA). The expression of CD154 was routinely checked by flow cytometry. Purified human T and B cells were co-cultured in RPMI 1640 containing 10 % FBS (Life Technologies) in 96-well plates at 3 × 104 cells/200 μl. For stimulating B cells, CD154-L cells were cultured for 18 hours and were irradiated with 60 Gy and then co-cultured with sorted B cells.
Cell functional assays
To measure IL-10 production by B cells, CD154-L cells were co-cultured with sorted B cells for 3 days and with 50 ng/ml phorbol-12-myristate-13-acetate (PMA) (Sigma-Aldrich, St. Louis, MO, USA), 1 μg/ml ionomycin (Sigma-Aldrich), and Golgistop (BD Biosciences) in the last 6 hours of culture. Cells were fixed, permeabilized, and stained with anti-human CD19-PE, IL-10-Alexa647, interferon-gamma (IFN-γ)-FITC monoclonal Abs or isotype controls. Intracellular cytokine was assayed by using a flow cytometer. Alternatively, the amount of IL-10 was measured in the culture supernatants before PMA, ionomycin, and Golgistop by enzyme-linked immunosorbent assay (ELISA) (R&D Systems, Minneapolis, MN, USA). To measure the influence of activated B cells on CD4+ T-cell activation, CD154-L cells were co-cultured with sorted CD4+CD25− T cells and B cells at a ratio of 1:10:10 (CD154-L cells: B cells: effector T cells) in the presence of suboptimal stimulation with soluble anti-CD3 (5 μg/ml) and anti-CD28 (10 μg/ml) for 5 days. Sorted autologous effector T cells were labeled with 3 μM carboxyfluorescein diacetate (Life Technologies) before co-culture with B cells. Proliferation of T cells by CFSE (carboxyfluorescein succinimidyl ester) dilution was assayed by flow cytometry. The same culture condition was used to measure T-cell cytokine production. Intracellular IFN-γ and tumor necrosis factor-alpha (TNF-α) and the cytokines in the culture supernatants were measured by same methods employed in IL-10 measurement. For identification of CD4+ T cells, T cells were stained with anti-CD3-APC, as CD4 is downregulated after activation. To measure the influence of activated B cells on CD4+FoxP3+ T-cell differentiation, CD154-L cells were co-cultured with sorted CD4+CD25− T cells and B cells at a ratio of 1:10:10 (CD154-L cells: B cells: effector T cells) in the presence of optimal stimulation with anti-CD3/CD28-coated beads at a ratio of 1:320 (beads: T cells). For identification of CD4+FoxP3+ T cells, T cells were stained with anti-CD3-APC and FoxP3-FITC. To induce in vitro differentiation of B cells into plasma cells, sorted B cells were co-cultured with 5 Gy-irradiated autologous effector T cells at a ratio of 1:1 in a 96-well round-bottom tissue culture plate pre-coated with 10 μg/ml anti-CD3 Abs (clone OKT3, eBioscience, San Diego, CA, USA) in the presence of soluble anti-CD28 Abs (5 μg/ml, clone CD28.2, eBioscience) in 96-well round-bottom tissue culture plates for 11 days. In some experiments, B cells were stimulated with 200 ng/ml IL-21 (Life Technologies) and 2 μg/ml anti-CD40 (R&D Systems) to induce differentiation in the presence or absence of 10 μg/ml anti-IL-10 (clone JES3-9D7, BD Biosciences). For analysis of accessory molecule expression, B cells were cultured in the presence of CD154-L cells for up to 2 days, and CD80, CD86, CD54, human leukocyte antigen (HLA)-DR expression was determined by flow cytometry after staining the cells with monoclonal Ab anti-CD80-PE, anti-CD86-PE, anti-HLA-DR-V450, or anti-CD54-PE. Recombinant human IL-10 (100 ng/ml, R&D Systems), anti-IL-10 (10 μg/ml, BD Biosciences), and anti-IL-10Rα (100 ng/ml, R&D Systems) blocking Abs were added where indicated. The amounts of IL-6 and TNF-α in the culture supernatants were analyzed by multiplex assay with Luminex (Bio-Rad Laboratories, Hercules, CA, USA).
Flow cytometry
Cells were acquired on an LSRII flow cytometer (BD Biosciences), and data were analyzed by using FlowJo 9.6 software (Tree Star Inc., now part of FlowJo, LLC, Ashland, OR, USA).
Measurement of antibody reactivity
Cell culture supernatants were screened for Ab reactivity to single-stranded DNA (ssDNA), double-stranded DNA (dsDNA), and histone by ELISA. All samples were diluted (1/2) with dilution buffer (2 % bovine serum albumin, 3 mM EDTA, 0.05 % Tween 20, and 0.1 % gelatin, all from Sigma-Aldrich) and incubated overnight. Microwell plates were washed and incubated with a secondary alkaline phosphatase-labeled goat anti-human IgM Ab (Southern Biotechnology, Birmingham, AL, USA) overnight, followed by development with para-nitrophenyl phosphate (Sigma-Aldrich). Serially diluted sera from patients with SLE were used for preparing standard curves. All samples were analyzed in duplicate.
Examination of immunoglobulin heavy chain complementary region 3
Single-cell polymerase chain reaction was used to amplify rearranged heavy chain Ig genes from genomic DNA as previously described [
4]. Sequences were compared with germline Ig genes by using the web-based analysis program JoinSolver [
18,
19] to examine IgH CDR3 (IgH CDR3) characteristics.
Statistical analysis
All values are expressed as mean ± standard error of mean. Mean fluorescence intensity values were calculated as the geometric mean. Data were compared by using the paired two-tailed Student’s t test. P values of less than 0.05 were considered significant. All statistical analyses were performed by using GraphPad Prism 5.0 (GraphPad Software, Inc., La Jolla, CA, USA).
Discussion
Pre-naïve B cells are a unique intermediate maturational subset with some immunocompetent functions, but their role in immune regulation has not been previously defined. In this study, we demonstrated two distinct functions of human peripheral pre-naïve B cells during normal immune responses: production of autoreactive Abs and maintenance of peripheral tolerance by auto-regulation of their accessory cell function.
B cells with autoreactive BCRs are constantly generated by random Ig gene rearrangement during B-cell ontogeny. It was reported in humans that approximately half of all BCRs expressed by immature B cells are autoreactive [
21]. Since pre-naïve B cells are developmentally not fully mature, it was possible that Abs produced by the pre-naïve B cells were autoreactive. In accordance with previous findings, we found that IgM Abs produced by pre-naïve B cells from 50 % of the normal donors tested were reactive to ssDNA and had longer IgH CDR3 length compared with transitional and naïve B cells suggestive of self-reactivity [
22,
23]. Though paradoxical to B-cell tolerance, auto-Abs are known to perform some useful functions during immune responses. An example of such beneficial auto-Abs are natural Abs, which are shown to be an essential first-line defense against pathogens [
24] and are shown to participate in the clearances of apoptotic cells [
25]. In mice, natural Abs are secreted mainly by a distinct B-cell subset, B-1 cells [
15], but in humans, existence of a distinct B-cell subset similar to mouse B-1 cells is less clear, and natural Abs are produced mainly by CD5
+ B cells not committed to a distinct lineage. Fetal and neonatal B cells [
26,
27] as well as CD5
+ B cells from healthy adults and from patients with chronic lymphocytic leukemia and autoimmune diseases such as rheumatoid arthritis and SLE were shown to produce natural IgM autoreactive Abs [
28‐
30]. Autoreactivity of Abs produced by the pre-naïve B cells and the fact that these cells are CD5
+ suggest that pre-naïve B cells may be one source of natural Abs in adult humans functioning to clear apoptotic material and other auto-Ags or to provide primary defense against certain bacterial and viral pathogens during humoral immune response.
To maintain tolerance, most of the autoreactive immature B cells in the periphery are culled during development before they enter the mature B-cell compartment by a mechanism of deletion, since they are highly susceptible to deletion by receptor cross-linking [
31]. Because pre-naïve B cells may contribute to protective autoreactive Ab production during normal immune response, it would be a loss if these cells were completely deleted from the repertoire. Their presence poses a risk, however, if they could function as Ag-presenting cells (APCs). The presence of a BCR with specificity for auto-Ags might permit them to present auto-Ags to CD4
+ T cells. We found unique mechanisms to limit this possibility in healthy individuals. The first mechanism employed by the pre-naïve B cells is regulation of co-stimulatory molecules upon receiving CD40 signal. We found that CD40-activated pre-naïve B cells inherently displayed lower levels of CD80 and that the more widely and predominantly displayed CD86 was displayed only transiently but that naïve and memory B cells continued to upregulate both CD80 and CD86. CD80 and CD86 are well-defined potent co-stimulators that play a major role in the activation of T cells by binding to CD28 [
32]. It has been suggested that CD86 is more important than CD80 in mediating resting T-cell activation by APCs [
33] and is essential for autoreactive T-cell activation and development for autoimmunity [
34]. Early downregulation of CD86 by pre-naïve B cells may be one of the mechanisms to block the autoimmune response from occurring by chance. Of note, this mechanism appeared to be independent of IL-10 secretion, even though IL-10 has been shown to regulate co-stimulatory molecule expression in dendritic cells (DCs) and macrophages by means of the E3 ubiquitin ligase membrane-associated ring-CH 1 (MARCH 1) [
35,
36].
A second mechanism to limit stimulation of autoreactive CD4
+ T cells is through regulation of cytokine secretion. IL-10 is one of the most important anti-inflammatory cytokines regulating T-cell responses produced by macrophages, DCs, B cells, and various subsets of CD4
+ and CD8
+ T cells [
37]. IL-10 produced by the B cells is known to mediate immunosuppression and to regulate T-cell responses directly or by induction of Tregs [
20,
38‐
40]. However, the capacity of IL-10 produced by pre-naïve B cells to downregulate the ability of these cells to augment CD4
+ T-cell activation has not been previously reported. We found that pre-naïve B cells produced IL-10 upon receiving a CD40 signal and that IL-10 prevented these cells from maximally supporting the stimulation of CD4
+ T-cell proliferation and biased these cells away from supporting T-cell proliferation toward participation in the differentiation of CD4
+ T cells into CD4
+FoxP3
+ Tregs. We are unable to fully explain the endocrine effect of secreted IL-10 on pre-naïve B-cell co-stimulatory function. In DCs and monocytes, downregulation of CD86 [
41,
42] and MHC class II molecules [
43,
44] was reported to be involved in IL-10-mediated poor support of T-cell responses. In contrast, it was also reported that IL-10 activates B cells and upregulates MHC class II molecule [
45,
46]. We also found that IL-10 mediated modest upregulation of CD80 and CD86, suggesting that other mechanisms employed by IL-10 to mediate poor support of T-cell activation exist which counteract B-cell activation by IL-10. In addition to production of immunoregulatory cytokine IL-10, pre-naïve B cells secreted significantly lower amounts of T-cell activating cytokines IL-6 and TNF-α, which can further prevent these cells from promoting T-cell activation.
Thus far, IL-10-producing B cells exhibiting regulatory functions independent of secreted IgG have been classified as Bregs in both mouse and humans [
8]. Stimulation with a combination of Ag, CD154, and Toll-like receptor ligands was reported to induce production of IL-10 by the Bregs [
8]. Phenotypically, two subsets of human IL-10-producing Bregs were described: CD19
+CD24
highCD38
high immature transitional B cells and CD19
+CD24
hiCD27
+ memory B cells [
14,
47]. Pre-naïve B cells also produce IL-10 after CD40 stimulation but are not classic Bregs phenotypically. They are immature B cells and express intermediate levels of CD38 and CD24 compared with the CD19
+CD24
highCD38
high Breg cells described previously [
3]. In addition, pre-naïve B cells appear to regulate immune responses through limiting their own Ag-presenting function, not by directly suppressing CD4
+ T cells as with Bregs. Therefore, IL-10 produced by the pre-naïve B cells may contribute to maintenance of self-tolerance by limiting the capacity to activate CD4
+ T cells rather than by directly suppressing their function.
Patients with SLE have been suggested to have a defect in early B-cell tolerance checkpoints leading to the accumulation of a large number of naïve B cells that express BCRs that recognize self-Ags, even in the inactive phase of the disease [
48]. Consistent with this finding, Yurasov et al. have demonstrated that 25–50 % of the naïve B cells in patients with SLE produced self-reactive Abs even before they participate in immune responses [
49]. However, the precise mechanisms contributing to early tolerance defects have not been reported until recently. In a recent study, CD19
+CD24
hiCD38
hi Bregs from patients with SLE were found to be refractory to CD40 engagement and produced less IL-10 and lacked CD4
+ T-cell suppressive capacity compared with the healthy controls, suggesting that functional defects of Bregs could lead to autoimmunity [
14]. We found that pre-naïve B cells from patients with SLE also showed a defect in producing IL-10 following CD40 stimulation and that their limited co-stimulatory molecule expression upon CD40 signaling was also impaired. SLE pre-naïve B cells expressed higher levels of CD80, CD86, and HLA-DR after CD40 stimulation and effectively promoted CD4
+ T-cell activation compared with the healthy controls. Since IL-10 and deficient co-stimulatory molecule expression appear to be a novel mechanism of self-tolerance employed by the pre-naïve B cells, a defect in pre-naïve B-cell functions could lead to a break in self-tolerance, especially in the SLE milieu, where expansion of the pre-naïve B cells has been found [
3,
4], and the immune system is intensely activated as evidenced by elevated B cell-activating factor (BAFF) levels, hyperexpression of CD40L on T cells, and/or increased levels of certain cytokines such as IL-21 [
50‐
53]. In mice, BAFF has been shown to increase IL-10 production from MZ B cells [
54]. The relevance of this to human B cells is unknown since pre-naïve B cells respond poorly to BAFF [
3], and many patients with SLE have increased levels of BAFF but decreased IL-10 production [
55,
56]. It is intriguing to note that SLE pre-naïve B cells may play a unique role in activating autoimmune CD4
+T cells since they have enhanced co-stimulatory capacity and also more frequently express autoreactive BCR. This combination of abnormalities may make them uniquely suited to take up and process auto-Ags and present the relevant peptide to autoreactive CD4
+ T cells, thus contributing to the break in T-cell tolerance and the upregulation of autoimmunity in patients with SLE.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
JHS, H-RK, S-HC, and IJK contributed to acquisition of the data. PEL contributed to study conception and design and edited the manuscript. JL contributed to study conception and design and to acquisition of the data and edited the manuscript. All authors were involved in analysis and interpretation of the data, drafting the article, or revising it critically for important intellectual content and read and approved the final version of the article.