Although IL-4-producing B cells (B effector 2 cells) are found following infection and immunization, the signals regulating IL-4 production by Be2 cells are unknown. We show that culturing naive B cells with Th2 cells induces up-regulation of IL-4 in the B cells with a concomitant down-regulation of T-bet, IL-12Rβ2, and IFN-γ. Up-regulation of IL-4 in the Be2 cells is dependent on both T cells and IL-4 as IL-4Rα-deficient B cells primed with Th2 cells did not transcribe IL-4, and B cells primed in the presence of IL-4-deficient Th2 cells produced IFN-γ instead of IL-4. Likewise, the in vivo development of IL-4-expressing B cells in a nematode infection model was dependent on both T cells and IL-4Rα-mediated signals. Thus, the differentiation of naive B cells into IL-4-expressing Be2 cells is regulated by a combination of T cell-dependent signals and the cytokine environment and this process is critically dependent upon the IL-4/IL-4R signaling pathway.
Bcells, like T cells, can produce an array of cytokines when stimulated, including polarizing cytokines such as IL-4, IL-12, and IFN-γ (1). Indeed, splenic B cells (2, 3) and germinal center B cells are able to produce IL-4 (4, 5), while immature B cells migrating to the spleen (6) and mature activated B cells produce IFN-γ (2, 7, 8). Importantly, we identified B effector 1 (Be1)4 and B effector 2 (Be2) cells producing “polarized” subsets of cytokines in vitro after culture with Ag and effector T cells and in vivo following infection with type-1- and type-2-inducing pathogens (2). Although B cells produce polarizing cytokines during the course of immune responses to pathogens and autoantigens (1, 9, 10, 11, 12), little is known about the factors that control cytokine production by B cells. In this study, we determined the cellular and molecular signals required for the generation of IL-4-expressing Be2 cells and show that Th2 cells and IL-4Rα-dependent signals control the development of Be2 cells in vitro and in vivo.
Materials and Methods
Mice were bred in the Trudeau Institute Animal Facility as described (8, 13). Strains used include: hen egg lysozyme (HEL)-specific BCR transgenic (Tg) mice (B10.BR H2kH2-T18a/SgSn-Tg(IghelMD4)J), influenza hemagglutinin (HA)-specific CD4 TCR Tg mice (BALB/c-Tg(HNT)By), pigeon cytochrome c fragment (PCCF)-specific CD4 TCR Tg mice (B10.BR H2kH2-T18a/SgSn-Tg(AND)J), AND Tg mice crossed to IL-4-deficient B10.BR H2kH2-T18a/SgSn Il4tm1Cgn/J mice, bicistronic “4get” IL-4-reporter mice (BALB/c.129-Il4tm1lky/J), 4get mice crossed to IL-4Rα-deficient mice (BALB/c-Il4ratm1Sz/J), and IL-4-deficient BALB/c mice (19). All procedures involving animals were approved by the Trudeau Institute Institutional Animal Care and Use Committee.
Peptides (PCCF88–104 and HA26–138) were purchased from New England Peptide. Recombinant cytokines and CTLA4-Ig were obtained from: Genetics Institute (IL-12), DNAX (IL-4), or R&D Systems (all others). mAb were obtained from the Trudeau Institute Ab core facility (anti-IL-4; clone 11B11, anti-CD4; clone GK1.5, anti-CD40; clone 1C10; anti-CD154 blocking Ab (MR1)), from ICN Biomedicals, anti-IgM F(ab′)2), from the National Cancer Institute (anti-IFN-γ; clone XMG1.2) or BD Biosciences (all others). Secreted Ab was detected by ELISA using purified anti-Ig (H+L) and HRP-labeled anti-κ (Southern Biotechnology Associates).
Naive or effector T and B cells were purified by MACS (Miltenyi Biotec) as described (2). Cell purities were routinely 90–95% for naive B and T cells and >99% for effector B cells.
Generation of T and B cell effectors
Th1 and Th2 effectors were generated in vitro from spleens of normal mice or TCR Tg mice as described (2, 8) using plate-bound anti-CD3 + anti-CD28 and exogenously added cytokines and anti-cytokine Abs. Be1 and Be2 effectors were generated from purified naive splenic B cells (from normal mice, BCR Tg mice, or 4get mice) cultured with mitomycin C-treated syngeneic, Tg Th1 (for Be1 cultures), or Th2 (for Be2 cultures) effector cells and Ag (PCCF or HA peptide and HEL or anti-IgM) as described (2, 8). In some experiments, rIL-4 (100 U/ml), MR1 (50 μg/ml), or CTLA4-Ig (1 μg/ml) were added at the initiation of the B cell effector cultures. Cytokine production by effector B cells was induced by stimulation with PMA (5 ng/ml) and Ca2+ ionophore (A23187, 1.25 μM).
Infection with Heligmosomoides polygyrus
Mice were infected by gavage with 200 third-stage H. polygyrus larvae (2). After 14–21 days, mice were sacrificed and mesenteric lymph node (MLN) cells were isolated from either individual mice or from pooled groups of mice (n = 3). In some experiments, mice were depleted of CD4 T cells by i.v. administration of depleting anti-CD4 mAb (clone GK1.5, 250 μg) given 2 days before H. polygyrus infection and 7 days after infection.
The protocol for qPCR and sequence of primers and probes for IFN-γ and T-bet were previously described (8). BSAP (Pax-5) and Blimp-1 (prdm1) primer and probe sets were purchased from Applied Biosystems. Primer/probe sets for IL-4 and IL-12Rβ2 are: IL-4: 5′-ACAGGAGAAGGGACGCCAT-3′ (forward), 5′-GAAGCCCTACAGACGAGCTCA-3′ (reverse), 5′-TCCTCACAGCAACGAAGAACACCACA-3′ (probe). IL-12Rβ2: 5′-CAAGCATTTGCATCGCTATCA-3′ (forward), 5′-AATGCCTTTTGCCGGAAGT-3′ (reverse), 5′-ACGAATTGAGAACGTGCCCACCGT-3′ (probe).
All studies were repeated at least twice with similar results and all in vivo experiments used at least three or more mice per group. Significant differences between groups were determined by Student’s t test.
Results and Discussion
Naive B cells cultured in the presence of effector Th1 cells and Ag differentiate into Be1 cells which secrete IFN-γ upon restimulation while naive B cells cultured in the presence of effector Th2 cells and Ag differentiate into Be2 cells which produce IL-4 upon restimulation (2, 8). To determine the molecular requirements for the generation of IL-4-producing Be2 cells, we cultured naive HEL-specific Tg B cells with mitomycin C-treated day 4 effector Th2 or Th1 cells specific for PCCF. We added Ag in the form of HEL and PCCF to facilitate cognate interactions and to induce TCR- and BCR-specific signaling. At 24 h intervals, we repurified B cells from the cultures and assessed the activation and differentiation profile of the cells. To determine when transcription of the Il4 gene was initiated in the developing Be2 cells, we determined IL-4 mRNA levels by quantitative PCR (qPCR) using naive B cells as a baseline control. Because no IL-4 mRNA was detected in the cDNA prepared from the naive B cells after 40 cycles of PCR (data not shown), we arbitrarily set the baseline for the naive B cells at 40 cycles and then calculated the minimum fold-induction of IL-4 mRNA in the developing Be2 cells. As seen in Fig. 1 A, IL-4 mRNA was detected within 24 h in the developing Be2 cells and the IL-4 mRNA levels in the Be-2 cells were increased at least 270-fold over the naive B cells within 3 days. In contrast, no IL-4 transcripts were detected in the Be1 cells at any time point (not shown). Thus, IL-4 gene transcription is initiated rapidly in developing Be2 cells but not in Be1 cells.
The differentiation of naive B cells into IFN-γ-producing Be1 cells is dependent on IFN-γ and IL-12, as well as the transcription factor T-bet (8). To determine whether expression of these genes was down-regulated in the developing Be2 cells, we prepared RNA from day 2 Be1 and Be2 cells and measured IFN-γ, IL-12Rβ2, and T-bet mRNA levels. As expected (8), IFN-γ mRNA was expressed in naive B cells and increased significantly in the day 2 Be1 cells (Fig. 1,B). In contrast, IFN-γ mRNA expression was suppressed in the Be2 cells relative to naive B cells (Fig. 1,B). Similar results were observed with T-bet and IL-12Rβ2 (Fig. 1 B), indicating that the T-bet/IFN-γ/IL-12 signaling pathways are actively suppressed in developing Be2 cells.
Next, to determine whether the Be2 cells differentiate into Ab-secreting cells, we examined expression of Blimp-1, which is up-regulated in plasma cells, and BSAP, which is down-regulated in plasma cells (14). Interestingly, while expression of Blimp-1 increased and BSAP decreased by day 4 in Be1 cells, the transcript levels of these genes remained relatively unchanged in the Be2 cells during the 4 days of culture (Fig. 1,C). Likewise, while we could easily detect secreted Ab in the cultures of PMA + Ca2+ ionophore-restimulated day 3 and 4 Be1 cells, very little secreted Ab was found in the Be2 cultures at any time point (Fig. 1 D). Thus, Be2 cells develop into IL-4-expressing effector cells that down-regulate expression of IFN-γ and do not terminally differentiate into Ab-secreting plasma cells, at least over 4 days.
IL-4 regulates the development of IL-4-producing Th2 cells (15). To test whether IL-4 promotes either the generation or expansion and survival of Be2 cells, we generated day 4 IL-4-deficient and wild-type (WT) PCCF-specific Th2 cells in vitro in cultures containing exogenously added IL-4. The IL-4-deficient PCCF-specific Th2 cells (Th2-IL-4−/−) activated and expanded normally in culture, but upon restimulation with anti-CD3, the Th2-IL-4−/− cells failed to secrete IL-4, although they produced the Th2 cytokine IL-5 in large amounts (data not shown). Next, we purified naive HEL-specific B cells and cultured the cells with Ag and the PCCF-specific WT Th2 cells (Th2-WT) or with the PCCF-specific Th2-IL-4−/− cells. After 4 days, the effector B cells were purified and restimulated for analysis of cytokine production. Be2 cells generated with Th2-WT cells produced >1000 pg/ml IL-4 and secreted very modest amounts of IFN-γ (Fig. 2,A). In contrast, Be2 cells generated in the presence of Th2-IL-4−/− cells failed to produce IL-4 and furthermore produced IFN-γ at levels equivalent to that seen for Be1 cells activated in the presence of Th1 cells and Ag (Fig. 2 A). These results demonstrate that IL-4 sufficient Th2 cells are obligate for Be2 development in this in vitro culture system. Furthermore, the data suggest that B cells default toward a Be1 phenotype unless IL-4 is present.
Although IL-4 expression by the Th2 cells was clearly important for the development or expansion of IL-4-producing Be-2 cells, this could be due to a direct effect of IL-4 on the developing Be2 cells or due to a requirement for IL-4 produced by the T cell on the differentiation or effector function of the Th2 cell. To test whether IL-4 in combination with effector T cells is sufficient to promote Be2 development or expansion, we cultured naive B cells isolated from bicistronic IL-4 reporter mice (4get mice; Ref.13) with Ag and Th2-WT or Th2-IL-4−/− Th2 cells in the presence or absence of exogenously added IL-4 (Fig. 2,B). Importantly, the B cells isolated from 4get mice maintain the ability to produce IL-4 and IL-4-expressing 4get cells coexpress the GFP reporter allowing for their easy detection by FACS (13). After 4 days of culturing the 4get B cells with Th2-WT or with Th2-IL-4−/− cells, we used FACS to determine the percentage of B cells expressing the IL-4 reporter GFP. As expected, B cells cultured with Th2-IL-4−/− cells did not become GFP+ after 4 days of culture, while a small, but easily detectable, subset of B cells cultured with the Th2-WT cells expressed the IL-4 reporter GFP (Fig. 2,C). When IL-4 was added to the cultures containing the Th2-IL-4−/− cells, the B cells differentiated into IL-4-expressing cells (Fig. 2 C), suggesting that IL-4 can act directly on the B cell to promote Be2 development or survival and expansion.
To confirm that IL-4 acts directly on the developing Be2 cell, we cultured purified naive splenic B cells isolated from 4get mice or from IL-4Rα-deficient 4get mice with Th2-WT cells and Ag (Fig. 2,B). Expression of the IL-4 reporter, GFP, was analyzed in the developing Be2 cells by FACS over the next 4 days. Less than 0.5% of CD19+ 4get B cells expressed detectable levels of the IL-4 reporter GFP protein at 24 h (Fig. 2,D), however within 48 h the fraction of GFP-expressing 4get B cells had increased to 1.5% (data not shown) and by day 4 ∼7% of the 4get B cells expressed GFP (Fig. 2,D). Interestingly, although the frequency of GFP+ IL-4Rα-deficient B cells doubled between days 1 and 4 (Fig. 2,D), the frequency of GFP+ IL-4Rα-deficient B cells was reduced ∼15-fold compared with the normal 4get B cells (Fig. 2 D). These data indicate that IL-4 and IL-4Rα signals are necessary for the development or expansion and survival of IL-4-producing Be2 cells.
To test whether IL-4 is sufficient to promote Be2 cell development or expansion in the absence of T cells, we stimulated naive B cells with IL-4, anti-CD40, and anti-IgM in the presence of anti-IFN-γ Ab for 4 days and then measured cytokine production in PMA + Ca2+ ionophore-restimulated cells. The restimulated B cells did not produce detectable amounts of IL-4 (not shown), therefore we next examined whether Th2 cells provide costimulatory signals that enhance Be2 development by culturing naive 4get B cells with day 4 Th2-WT effectors and Ag in the presence and absence of Abs that block CD40/CD154 (MR1 Ab) or CD28/B7 interactions (CTLA4-Ig, Fig. 2,B). On day 4, we assessed whether these blocking reagents inhibited the development or expansion of Be2 cells by determining the percentage of B cells that expressed the IL-4 reporter GFP. As shown in Fig. 2 E, inclusion of MR1, CTLA4-Ig, or MR1 + CTLA4-Ig in the cultures inhibited Be2 development by ∼50%. Thus, Th2 cells provide both cytokines and costimulatory signals that are required for optimal Be2 cell development or expansion and survival, at least in in vitro cultures.
Based on our in vitro data showing that both IL-4 and T cell-derived signals regulate Be2 development or expansion, we tested whether these signals are also required for the in vivo generation of Be2 cells. We infected 4get mice, 4get × IL-4Rα-deficient mice or WT BALB/c mice with H. polygyrus, a nematode that induces a potent type 2 immune response (16). We isolated the draining MLN on day 21 and analyzed GFP expression by the B cells. In H. polygyrus-infected 4get mice, we consistently observed a small (<1% of total B cells), yet easily identifiable population of GFP-expressing B cells that made up ∼12% of the non-T cell GFP+ lymphocytes in the MLN (Fig. 3,A). Importantly, essentially all of these GFP+ B cells were able to secrete IL-4 when restimulated in vitro with PMA and Ca2+ ionophore (Fig. 3,B). In addition, we found that the frequency of GFP-expressing B cells was significantly reduced in H. polygyrus-infected IL-4Rα-deficient 4get mice (Fig. 3 A), indicating that IL-4Rα-mediated signals are also required for the in vivo generation of IL-4-expressing B cells.
Finally, to test whether CD4 T cells are required for the generation of IL-4-expressing B cells in vivo, we treated 4get mice with an anti-CD4 mAb to deplete the CD4 T cells before infection and then infected the mice with H. polygyrus. As a control, we infected a group of 4get mice that were not CD4-depleted. We determined the frequency of GFP+ B and T cells by FACS on day 21. Approximately 40% of the total cells present in the MLN of the infected 4get mice were CD4+ T cells and of those, 10% expressed GFP (not shown). As expected, CD4+ T cells were not detected in the MLN of the CD4-depleted 4get mice (<1% of total cells, not shown). Analysis of the B cell population revealed a distinct population of CD19+GFP+ cells in infected, non-CD4-depleted mice that made up ∼1.3% of the total B cells and 19% of the non-T cell GFP+ lymphocytes (Fig. 3,C). In contrast, the frequency of IL-4-expressing B cells identified in the infected CD4-depleted 4get mice was reduced by >98% (Fig. 3,C) and was equivalent to the frequency observed in uninfected mice (Fig. 3 C). Thus, the development of IL-4-expressing Be2 cells in vivo is dependent on IL-4 and CD4 T cells.
Naive B cells constitutively express T-bet and IFN-γ transcripts and when activated in the absence of any polarizing cytokines produce small quantities of IFN-γ but no IL-4 (1). If IFN-γ is also present during B cell priming, the B cells differentiate into Be1 cells that produce large quantities of IFN-γ (8) and secrete Ab. However, if naive B cells interact with IL-4-expressing Th2 cells during priming, the B cells divert from the “default” Be1 pathway, down-regulate transcription of IFN-γ and T-bet, and up-regulate IL-4. Interestingly, the in vitro-derived Be2 cells do not substantially up-regulate Blimp-1 or down-regulate BSAP and do not secrete Ab. Instead, these Be2 cells express costimulatory molecules (1) and secrete cytokines like IL-4 which help to promote the differentiation of naive T cells into Th2 cells (2). Therefore, we propose that IL-4-producing Be2 cells, generated in response to cognate interactions with Th2 cells, may serve to amplify the Th2 response by activating and polarizing new cohorts of naive CD4 T cells. Indeed, a number of studies demonstrated that B cells are required for the generation and/or maintenance of Th2 responses (3, 17, 18) and there is now data showing that germinal center B cells regulate Th2 development through an IL-4-dependent process (4). Taken together, these data suggest that while type 2 immunity and allergic responses and the accompanying pathology associated with these responses are most efficiently initiated by T cells and DCs, this response may be sustained and potentially amplified by an IL-4-driven feedback loop between Ag-specific T and B cells.
We thank Troy Randall for reviewing this manuscript.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by Trudeau Institute and National Institutes of Health Grants AI-50844 (to F.E.L.) and AI-45666 and AI-0465530 (to M.M.).
Abbreviations used in this paper: Be1, B effector 1 cell; Be2, B effector 2 cell; HA, hemagglutinin; PCCF, pigeon cytochrome c fragment; HEL, hen egg lysozyme; qPCR, quantitative PCR; Tg, transgenic; MLN, mesenteric lymph node; WT, wild type.