B Cells Inhibit CD4⁺ T Cell-Mediated Immunity to Brucella Infection in a Major Histocompatibility Complex Class II-Dependent Manner

Brucella spp. are facultative intracellular bacteria notorious for their ability to induce a chronic, and often lifelong, infection known as brucellosis. To date, no licensed vaccine exists for prevention of human disease, and mechanisms underlying chronic illness and immune evasion remain elusive. We and others have observed that B cell-deficient mice challenged with Brucella display reduced bacterial burden following infection, but the underlying mechanism has not been clearly defined.

ABSTRACT

Brucella spp. are facultative intracellular bacteria notorious for their ability to induce a chronic, and often lifelong, infection known as brucellosis. To date, no licensed vaccine exists for prevention of human disease, and mechanisms underlying chronic illness and immune evasion remain elusive. We and others have observed that B cell-deficient mice challenged with Brucella display reduced bacterial burden following infection, but the underlying mechanism has not been clearly defined. Here, we show that at 1 month postinfection, B cell deficiency alone enhanced resistance to splenic infection ∼100-fold; however, combined B and T cell deficiency did not impact bacterial burden, indicating that B cells only enhance susceptibility to infection when T cells are present. Therefore, we investigated whether B cells inhibit T cell-mediated protection against Brucella. Using B and T cell-deficient Rag1^−/− animals as recipients, we demonstrate that adoptive transfer of CD4⁺ T cells alone confers marked protection against Brucella melitensis that is abrogated by cotransfer of B cells. Interestingly, depletion of CD4⁺ T cells from B cell-deficient, but not wild-type, mice enhanced susceptibility to infection, further confirming that CD4⁺ T cell-mediated immunity against Brucella is inhibited by B cells. In addition, we found that the ability of B cells to suppress CD4⁺ T cell-mediated immunity and modulate CD4⁺ T cell effector responses during infection was major histocompatibility complex class II (MHCII)-dependent. Collectively, these findings indicate that B cells modulate CD4⁺ T cell function through an MHCII-dependent mechanism which enhances susceptibility to Brucella infection.

INTRODUCTION

Evidence suggests that Brucella spp. infected hominids over 2.4 million years ago, but it is likely that these pathogens began to pose a significant risk to human health shortly after the domestication of agricultural species such as sheep, cattle, and goats (1, 2). Despite this, brucellosis is still counted among the most common zoonotic diseases according to the World Health Organization (3). Roughly 500,000 new brucellosis cases are reported annually; however, the nonspecific presentation of this illness results in widespread misdiagnosis and underreporting, and the true incidence of disease is thought to range between 5 and 12.5 million cases annually (4, 5). Caused by the facultative intracellular bacteria of the Brucella genus, humans most often contract brucellosis via direct contact with infected animals or consumption of unpasteurized dairy products from these animals (6–9). Although several Brucella species can cause disease in humans and ruminants alike, the vast majority of human cases occur following exposure to B. melitensis, B. abortus, or B. suis (10, 11). In humans, brucellosis presents with a range of nonspecific clinical symptoms, including myalgia, arthralgia, and undulating fever, which likely contribute to the disparity between confirmed cases and estimates of global prevalence (3, 10, 12, 13). Yet the most devastating aspect of this disease stems from Brucella’s proclivity for triggering chronic, often lifelong, disease even in cases where aggressive antibiotic therapy is employed (14, 15). The propensity for Brucella to induce chronic infection remains a major obstacle in developing better therapeutics for infected individuals and underscores the fact that the mechanisms underlying lifelong infection continue to elude us.

Although Brucella spp. are adept at host immune evasion, components of the host immune response can control, if not clear, Brucella infection. Interferon gamma (IFN-γ), for instance, activates macrophages to kill internalized Brucella (16), and IFN-γ^−/− mice succumb to Brucella infection regardless of the route of challenge (17, 18), demonstrating the critical role of IFN-γ in the control of Brucella infection. While the protective effects of IFN-γ are well known within the field, the relative contribution of CD4⁺ and CD8⁺ T cells to the control of Brucella infection in mice remains unclear (11, 19–23). Various reports have indicated at least some level of protection conferred by both Th1 and Th17 type effector T cell responses, while others report a dominant role for cytotoxic CD8⁺ T cell function (11, 19, 21–24). Nevertheless, Brucella’s success at inducing chronic infections demonstrates that T cell responses are inefficient at clearing infection. Clearly, investigating mechanisms underlying the inefficiency of the T cell response to Brucella is vital to our understanding of chronic brucellosis.

B cell-deficient mice have been shown to be resistant to infection with B. abortus or B. melitensis (22, 25), and reconstitution of B cell-deficient animals with immune sera does not alter susceptibility to B. abortus, indicating that the deleterious effect of B cells is not antibody driven (25). Rather, resistance in B cell-deficient mice was correlated with an increased frequency of IFN-γ-expressing T cells, suggesting a regulatory role for B cells during brucellosis, though this was not directly tested (25). Here, we investigated how B cells alter the T cell response to Brucella and show that B cells mediate enhanced susceptibility to brucellosis via modulation of CD4⁺ T cell responses in a major histocompatibility complex class II (MHCII)-dependent manner.

RESULTS

B cell-deficient mice display enhanced resistance to Brucella melitensis infection.

Previous reports have demonstrated enhanced resistance to infection with B. abortus and B. melitensis infection in B cell-deficient (μMT^−/−) animals (22, 25). To confirm these findings, wild-type (WT) or μMT^−/− animals were infected intraperitoneally (i.p.) with B. melitensis 16M, and bacterial burdens in spleens were assessed. Similar to previous studies, μMT^−/− mice displayed significantly reduced bacterial burden in the spleen, particularly at 4 weeks postinfection when bacterial levels were ∼100-fold lower than in WT controls (Fig. 1A and C). μMT^−/− spleen weights were significantly lower than WT at 4 weeks postinfection (Fig. 1D). However, at 2 weeks postinfection, the weight of spleens from μMT^−/− mice did not reproducibly differ from the weight of WT spleens (Fig. 1B). This is surprising given the difference in spleen size in naive μMT^−/− compared to WT mice (see Fig. S1A in the supplemental material) and could possibly result from an enhanced inflammatory response in mice lacking B cells. Indeed, we did observe that spleens from B cell-deficient mice had a higher proportion of neutrophils in the first 2 weeks after Brucella infection (Fig. S1B). Previous studies also established the ability of B. abortus to infect and survive within B cells both in vitro and in vivo (26–28). To determine if B. melitensis can survive within B cells in vivo, we sorted splenic CD19⁺ B cells from infected WT mice and plated them onto agar. Similar to what has been shown with B. abortus, we recovered B. melitensis from within the CD19⁺ splenic B cell fraction at 1, 2, and 4 weeks postinfection (data not shown and Fig. S1C).

FIG 1.

B cell-deficient mice display enhanced resistance to B. melitensis infection. (A to D) WT and μMT^−/− mice were infected i.p. with 1 × 10⁵ CFU of B. melitensis 16M. Mice were sacrificed at 2 (A, B) or 4 (C, D) weeks postinfection, and spleen weights and bacterial burdens were measured. Data are representative of 2 to 3 independent experiments (n = 4 to 5 mice/group). *, P < 0.05 compared to WT; NS, not significant. Error bars represent the standard deviation (SD) of the mean.

B cell-deficient mice display an accelerated inflammatory response during Brucella infection.

While our data and that of others show that μMT^−/− mice are resistant to Brucella infection (22, 25), the mechanism underlying this finding has not been clearly defined. To determine how B cell deficiency alters the immune response to Brucella, we compared cytokine levels in the spleens of WT and μMT^−/− mice at 2 and 4 weeks postinfection. We observed that μMT^−/− mice exhibited significantly increased levels of IFN-γ, interleukin-12p70 (IL-12p70), and tumor necrosis factor alpha (TNF-α) compared to WT controls at 2 weeks postinfection (Fig. 2A to C), indicating accelerated production of proinflammatory, Th1 type effector cytokines (29–31). Similarly, in B cell-deficient animals at 2 weeks postinfection, we observed significantly increased levels of IL-6, IL-17A, and IL-1β (Fig. 2D to F), which promote Th17 cell differentiation (32). TGF-β, which is important for both T regulatory and Th17 effector development (32, 33), was produced at similar levels in both strains throughout the study (data not shown). By 4 weeks postchallenge, cytokine levels declined in both experimental groups, consistent with an overall decline in Brucella numbers from 2 to 4 weeks postinfection (Fig. 1, Fig. 2A to F). Additionally, the enhanced levels of inflammatory cytokines observed in μMT^−/− mice compared to WT mice at 2 weeks postinfection had largely subsided by 4 weeks postinfection. However, this result was unsurprising given that μMT^−/− mice exhibit 100-fold lower splenic bacterial burdens than WT animals by 4 weeks postinfection (Fig. 1C). Collectively, these data indicate that μMT^−/− animals have an accelerated inflammatory response to B. melitensis associated with increased levels of Th1 and Th17 cytokines.

FIG 2.

B cell-deficient mice display an accelerated inflammatory response during B. melitensis infection. (A to F) WT or μMT^−/− animals were infected i.p. with 1 × 10⁵ CFU of B. melitensis 16M. Mice were sacrificed 2 (D14) or 4 (D28) weeks postinfection when spleens were collected for measurement of cytokines (n = 4 to 5 mice/group). Cytokine levels were also measured in spleens harvested from naive WT and μMT^−/− mice (n = 4 to 5 mice/group). *, P < 0.05 compared to WT animals at the same time point; NS, not significant. Error bars represent the SD of the mean.

B cell-mediated susceptibility to infection depends on the presence of T cells.

Our cytokine profiling of μMT^−/− and WT mice suggested an accelerated Th1 and Th17 response in mice lacking B cells (Fig. 2). We and others have reported that Th1 and Th17 responses can confer some level of protection against Brucella infection (18, 19, 34, 35), which could account for enhanced resistance in μMT^−/− animals. Thus, we queried whether B cell-mediated susceptibility to B. melitensis depends on the presence of T cells. To test this, we first challenged Rag1^−/− mice, which lack both B and T cells, and compared bacterial loads to WT animals 4 weeks postchallenge, when resistance to B. melitensis infection in μMT^−/− mice was most pronounced (Fig. 1C). Similar to what others have reported when infecting Rag1^−/− mice with B. abortus (36), we observed no marked effect of combined B and T cell deficiency on splenic bacterial titer (Fig. 3A). Taken together with our finding that B cell deficiency (Fig. 1) enhances resistance to infection, the lack of an effect of combined T and B cell deficiency on control of infection suggests that T cells can confer protection against B. melitensis when B cells are absent, and it suggests that the deleterious effect of B cells may require the presence of T cells. Alternatively, in Rag1^−/− mice, the concomitant loss of a potentially protective T cell response could mask the protective effect of B cell deficiency and explain the similar levels of Brucella we observed in WT and Rag1^−/− mice. However, when we challenged TCRα^−/− mice, which have an intact B cell compartment but lack CD4⁺ and CD8⁺ T cells, we observed no difference in splenic bacterial burden compared to WT animals 4 weeks postinfection (Fig. 3B). Collectively, our finding that B cell-deficient, but not B and T cell-deficient, mice are resistant to infection, coupled with our observation that the absence of αβ T cells does not impact susceptibility to infection, indicates that T cells may be more efficient at control of Brucella infection when B cells are absent.

FIG 3.

B cell-mediated susceptibility to B. melitensis infection requires the presence of T cells. (A, B) Animals were challenged with 1 × 10⁵ CFU of B. melitensis 16M i.p. and sacrificed 4 weeks postinfection. Splenic Brucella burdens for Rag1^−/− (A) and TCRα^−/− (B) mice relative to WT mice were measured. n = 5 mice/group. NS, not significant. Error bars represent the SD of the mean.

B cells inhibit CD4⁺ T cell-mediated protection against B. melitensis infection.

Based on our findings shown in Fig. 3, we hypothesized that αβ T cells are more efficient at mediating protection against B. melitensis when B cells are absent. To test this, we purified splenic CD8⁺ or CD4⁺ T cells from naive animals and performed adoptive transfers of each cell type into Rag1^−/− mice prior to challenge with B. melitensis. Interestingly, we observed no marked effect on control of infection within the spleens of animals which received CD8⁺ T cells (Fig. S2A). In contrast, we found that transfer of CD4⁺ T cells reduced splenic bacterial burden 50- to 100-fold compared to phosphate-buffered saline (PBS)-treated animals (Fig. 4A and B), indicating that CD4⁺ T cells are capable of mediating protection against B. melitensis.

FIG 4.

B cells inhibit CD4⁺ T cell-mediated protection against B. melitensis infection. (A) Rag1^−/− mice (n = 4 to 5/group) were transferred 1.5 × 10⁷ WT CD4⁺ T cells or PBS alone 1 day prior to i.p. challenge with 1 × 10⁵ CFU of B. melitensis 16M. Four weeks postinfection, animals were euthanized and spleens were harvested for determination of Brucella burden. *, P < 0.05 compared to mice receiving PBS. (B) Rag1^−/− mice (n = 4 to 5 group) were transferred red blood cell-depleted whole splenocytes (1 × 10⁸/mouse), CD4⁺ T cells (1.5 × 10⁷/mouse), CD4⁺ T cells (1.5 × 10⁷/mouse) and B cells (5 × 10⁷/mouse), or PBS (vehicle) intravenously. One day after transfer, mice were challenged i.p. with 1 × 10⁵ CFU of B. melitensis 16M. Four weeks postinfection, mice were euthanized and splenic bacterial burdens were measured. (C) WT or μMT^−/− mice (n = 5/group) were treated with anti-CD4 depleting antibody or IgG. Animals were challenged i.p. with 1 × 10⁵ CFU of B. melitensis, and spleens were harvested 4 weeks postinfection for determination of Brucella burden. Means with the same letter do not differ significantly (P < 0.05) from each other as determined by ANOVA followed by Tukey’s test in panels B and C. Error bars represent the SD of the mean. Data in panels A and B are representative of 2 independent experiments.

While transfer of CD4⁺ T cells into Rag1^−/− mice mediated protection (Fig. 4A), others have reported that genetic or antibody-mediated CD4⁺ T cell deficiency does not impair, and in some cases may enhance, resistance to Brucella infection (11, 21, 23, 36). Thus, we next queried whether the protective effects of CD4⁺ T cells are impaired by the presence of B cells by performing additional adoptive transfer studies. Rag1^−/− recipients received PBS, purified CD4⁺ T cells alone, or cotransfer of CD4⁺ T cells and purified B cells to determine if CD4⁺ T cell-mediated protection is altered when B cells are present. We also harvested total splenic leukocytes to determine the effect of adoptive transfer of CD4⁺ and CD8⁺ T cells, B cells, and other resident spleen cells (e.g., myeloid and dendritic) from the spleens of WT donors into Rag1^−/− animals. In agreement with our previous experiments, transfer of CD4⁺ T cells alone significantly reduced B. melitensis levels compared to administration of PBS at 4 weeks postinfection (Fig. 4B). In contrast, we observed no effect on splenic bacterial burden between recipients of whole splenocytes or PBS (Fig. 4B). These results suggest that the ability of CD4⁺ T cells to restrict Brucella infection is inhibited by cotransfer of other splenic cell types. Intriguingly, when we evaluated the impact of the cotransfer of B cells on splenic burdens, we observed that the presence of B cells abrogated the protection conferred by CD4⁺ T cells (Fig. 4B). Importantly, transfer efficiency of CD4⁺ T cells did not differ significantly between recipients of CD4⁺ T cells alone compared to those receiving both CD4⁺ T cells and B cells (26.2% and 27.18%, respectively, of recipient splenic leukocytes were CD4⁺), confirming that abrogation of protection did not result from decreased numbers of CD4⁺ T cells in cotransfer groups. To control for the possibility that the inhibition of CD4⁺ T cell-mediated immunity by B cells (Fig. 4B) was an artifact of our adoptive transfer model, we also employed CD4⁺ T cell depletion studies. We treated WT and μMT^−/− mice with anti-CD4 antibodies to deplete CD4⁺ T cells or IgG as an isotype control and assessed splenic bacterial loads 4 weeks post-B. melitensis challenge. Remarkably, depletion of CD4⁺ T cells from WT animals reduced splenic bacterial numbers compared to isotype-treated WT mice (Fig. 4C), which suggested that CD4⁺ T cells may be deleterious to control of Brucella infection when B cells are present. In contrast, treatment of B cell-deficient mice with CD4-depleting antibody enhanced splenic bacterial levels compared to isotype-treated μMT^−/− mice (Fig. 4C), demonstrating that, in the absence of B cells, CD4⁺ T cells mediate protection during Brucella infection. In addition, while isotype-treated μMT^−/− mice exhibited a >100-fold decrease in bacterial burdens compared to isotype-treated WT mice, CD4⁺ T cell-depleted WT and μMT^−/− mice displayed similar bacterial titers (Fig. 4C). These data indicate that B cells are not deleterious to the control of Brucella infection when CD4⁺ T cells are also absent. Collectively, our results employing both adoptive transfer (Fig. 4B) and T cell depletion (Fig. 4C) demonstrate that B cells inhibit CD4⁺ T cell-mediated protection during B. melitensis infection.

B cell-deficient mice have altered CD4⁺ T cell function during brucellosis.

Having confirmed that B cells inhibit CD4⁺ T cell-derived protection against B. melitensis, we next investigated the impact of B cells on CD4⁺ T cell function following challenge with B. melitensis. Our previous experiments demonstrated that differences in the overall splenic inflammatory responses between WT and μMT^−/− mice were more pronounced at 2 than at 4 weeks postchallenge (Fig. 2). Therefore, we evaluated T cell function within the first 2 weeks of infection. WT and μMT^−/− mice were challenged with B. melitensis and sacrificed 1 or 2 weeks postinfection. CD4⁺ T cells were isolated from the spleens of infected animals and restimulated with plate-bound anti-CD3 before measurement of cytokine production (37–40). Naive animals were also sacrificed and the CD4⁺ T cells purified to determine basal levels of cytokine expression upon CD3 stimulation. Cytokine production by CD4⁺ T cells from naive mice was minimal, though IFN-γ production by CD4⁺ T cells from WT mice was slightly lower than IFN-γ production by CD4⁺ T cells from μMT^−/− mice (Fig. 5A and B). In contrast, at 1 week postchallenge, WT CD4⁺ T cells actually produced significantly more IFN-γ than μMT^−/− CD4⁺ T cells (Fig. 5A), and by 2 weeks postinfection, IFN-γ production was similar between groups. Thus, inhibition of CD4⁺ T cell IFN-γ production is likely not the mechanism by which B cells promote susceptibility to Brucella infection. However, T cell intrinsic IL-10 production has been reported to have a deleterious impact on bacterial control during brucellosis (41); therefore, we also investigated whether IL-10 production was altered in CD4⁺ T cells from WT versus μMT^−/− mice. Indeed, we observed a 5-fold decrease in IL-10 production by μMT^−/− CD4⁺ T cells compared to that of WT CD4⁺ T cells 1 week postinfection (Fig. 5B). Moreover, we observed that impaired IL-10 production by CD4⁺ T cells harvested from μMT^−/− mice was sustained at 2 weeks after challenge (Fig. 5B). Therefore, enhanced susceptibility in WT animals may derive in part from enhanced IL-10 production by CD4⁺ T cells following B. melitensis challenge. Together, these data indicate that the function of CD4⁺ T cells from μMT^−/− and WT animals differs in their response to B. melitensis infection, suggesting that the presence of B cells alters CD4⁺ T cell function.

FIG 5.

B cell-deficient mice display altered CD4⁺ T cell function during brucellosis. CD4⁺ T cells were purified from naive WT and μMT^−/− mice and from WT and μMT^−/− mice 1 (D7) or 2 (D14) weeks after i.p. infection with 1 × 10⁵ CFU of B. melitensis 16M. Purified CD4⁺ T cells (5 to 6 wells/group) were cultured with anti-CD3 for 66 hours when supernatants were collected, and production of IFN-γ (A) and IL-10 (B) were measured. Data are representative of two independent experiments. *, P < 0.05 as determined by t test; NS, not significant.

B cell inhibition of CD4⁺ T cell-mediated protection against Brucella is MHCII-dependent.

B cells are capable of modulating T cell behavior both indirectly via cytokine secretion and through direct interaction with T cells (42). A major source of direct B and T cell interaction within the spleen occurs during B cell antigen presentation to CD4⁺ T cells in which the T cell receptor (TCR) is engaged by peptide:MHCII moieties on B cells. Therefore, we performed adoptive transfer experiments to determine whether B cell inhibition of CD4⁺ T cell-mediated protection was dependent on MHCII expression by B cells. Similar to our previous experiments, mice received CD4⁺ T cells alone, CD4⁺ T cells and B cells from WT animals, or CD4⁺ T cells from WT donors coupled with B cells from MHCII-deficient (MHCII^−/−) mice. As we previously described, mice that received both CD4⁺ T cells and WT B cells had higher B. melitensis burdens than mice that received CD4⁺ T cells alone (Fig. 6A). Interestingly, cotransfer of MHCII^−/− B cells with CD4⁺ T cells did not alter control of B. melitensis (Fig. 6A), indicating that MHCII expression is required for B cells to promote susceptibility to infection. Of note, as in previous transfer experiments, CD4⁺ T cell transfer efficiency did not differ between CD4⁺ T cell single transfers and those receiving cotransfer of CD4⁺ T and WT B cells (Fig. S2B). Despite lower bacterial burdens, however, a moderate but significant reduction in CD4⁺ T cell transfer efficiency was observed in recipients that were cotransferred MHCII^−/− B cells compared to recipients of CD4⁺ T cells alone or CD4⁺ T cells and WT B cells (Fig. S2B). The transfer efficiency of WT B cells and MHCII^−/− B cells did not significantly differ (Fig. S2C).

FIG 6.

B cell inhibition of CD4⁺ T cell-mediated protection is MHCII-dependent. (A to C) Rag1^−/− mice were intravenously transferred CD4⁺ T cells (1.5 × 10⁷/mouse) or cotransferred CD4⁺ T cells (1.5 × 10⁷/mouse) with either WT B cells (5 × 10⁷/mouse) or MHCII^−/− B cells (5 × 10⁷/mouse). (A) Animals (n = 4 to 5 mice/group) were sacrificed 4 weeks postinfection, and spleens were collected for enumeration of bacterial burdens. (B, C) Mice were euthanized 2 weeks postinfection, and purified splenic CD4⁺ T cells (5 to 6 wells/group) were cultured with anti-CD3 for 66 hours, at which time supernatants were collected and production of IFN-γ (B) and IL-10 (C) was quantified. Data in panel A are representative of 2 independent experiments. Means with the same letter do not differ significantly (P < 0.05) from each other as determined by ANOVA followed by Tukey’s test.

In our previous experiments, we found altered function in CD4⁺ T cells derived from WT animals compared to those collected from μMT^−/− mice (Fig. 5) and observed that cotransfer of WT, but not MHCII^−/−, B cells abrogated CD4⁺ T cell-mediated protection against B. melitensis (Fig. 4B and 6A). To determine how B cell MHCII expression affects CD4⁺ T cell function during brucellosis, we repeated our cotransfer experiments of CD4⁺ T cells with WT or MHCII^−/− B cells into Rag1^−/− mice and assessed CD4⁺ T cell cytokine production 2 weeks postinfection. Analogous to our previous observations (Fig. 5), IFN-γ production was significantly increased by CD4⁺ T cells purified from the CD4⁺/WT B cell cotransfer group compared to CD4⁺ T cells purified from animals that received CD4⁺ T cells alone (Fig. 6B), suggesting that B cells do not impede, and may actually promote, IFN-γ production by CD4⁺ T cells. Likewise, CD4⁺ T cells from mice that received CD4⁺/WT B cell cotransfers produced significantly increased levels of IL-10 compared to CD4⁺ T cells purified from mice that received CD4⁺ T cells alone (Fig. 6C). Therefore, the presence of B cells during B. melitensis infection promotes CD4⁺ T cell production of IL-10. Intriguingly, while cotransfer of WT B cells promotes IL-10 production by CD4⁺ T cells, cotransfer of MHCII^−/− B cells actually reduced IL-10 production by CD4⁺ T cells (Fig. 6C). Thus, B cells alter CD4⁺ T cell function via an MHCII-dependent mechanism that drives increased IL-10 production during Brucella infection.

DISCUSSION

Previous reports, along with our findings here, demonstrate an inherent resistance to Brucella infection in B cell-deficient mice following challenge with B. abortus or B. melitensis (22, 25). Here, we investigated mechanisms by which B cells enhance susceptibility to brucellosis. Our findings that μMT^−/− (Fig. 1C), but not Rag1^−/−, mice (Fig. 3A) display enhanced resistance to B. melitensis infection indicate that T cells confer protection in B cell-deficient animals. However, the absence of αβ T cells did not impact susceptibility to splenic infection at the same time point (Fig. 3B). Taken together, these data suggest that the presence of B cells can suppress the potentially protective effect of T cells during Brucella infection. Alternatively, deficiency of γδ T cells, rather than αβ T cells, in Rag1^−/− mice could explain why μMT^−/− animals are protected from infection while Rag1^−/− mice are not. However, we previously found that the protective effect of γδ T cells during brucellosis appears to be limited to the first week of infection (43) and therefore focused our future studies on interactions between B cells and αβ T cells.

Despite strong agreement regarding the protective effects of IFN-γ and the presumed role of CD4⁺ and CD8⁺ T cells in protection against Brucella, the relative contribution of T cell subsets to the control of brucellosis remains unclear. In particular, the protective capacity of CD4⁺ versus CD8⁺ T cells during brucellosis varies between studies. As reviewed by others, these discrepant findings may be due to the use of different Brucella spp. and strains for infection, variability in the infection route employed (i.e., intranasal versus intraperitoneal etc.), as well as the dose of Brucella and time postinfection when bacterial loads were assessed in animals (11). Additional discrepancies occur when the effects of the same T cell population are evaluated using different mouse strains. For example, the relative role and importance of CD4⁺ T cells can differ depending on whether the investigators deplete CD4⁺ T cells or utilize CD4^−/− or MHCII^−/− mice (19, 23, 35, 36), potentially contributing to conflicting results between reports. What can be inferred from the literature currently is that endogenous Th1, Th17, and CD8⁺ T cell responses can, in the right context, offer some degree of protection against infection (11, 19, 21, 23, 24, 34, 35). However, the fact that Brucella can cause a lifelong infection indicates that T cell responses to brucellosis are inefficient at clearing infection. Our data suggest that B cells might suppress the protective effect of T cells, and therefore we directly investigated whether B cells alter the function and protective capability of T cells during brucellosis.

T cell responses appear to be more efficient in μMT^−/− mice, as we observed an accelerated inflammatory response indicative of enhanced Th1 and Th17 type cytokine production in the absence of B cells (Fig. 2). In addition, adoptive transfer of CD4⁺ T cells into B and T cell-deficient Rag1^−/− mice revealed that CD4⁺ T cells can confer protection against Brucella infection (Fig. 4A). It is worth noting that, as a consequence of lymphopenia, transfer of CD4⁺ T cells into Rag1^−/− mice has been reported to induce homeostatic proliferation, skewed T cell responses, and accelerated graft rejection in transplant models (44). This effect could possibly explain the enhanced resistance against B. melitensis infection in Rag1^−/− mice that received CD4⁺ T cells. Nevertheless, cotransfer of B cells abrogated the protective effect against infection mediated by transfer of CD4⁺ T cells. In addition, the transfer efficiency of CD4⁺ T cells was not affected by the presence or absence of B cells in Rag1^−/− recipients, signifying that mitigation of protection was not a consequence of reduced CD4⁺ T cell number (Fig. S2B). Additionally, depletion of CD4⁺ T cells from μMT^−/−, but not WT, animals enhances susceptibility to Brucella infection. Therefore, using both adoptive transfers and antibody depletion, we have demonstrated that B cells alter the protective capability of CD4⁺ T cells. Thus, a major determinant for the seemingly inefficient T cell response during Brucella infection arises from B cell inhibition of CD4⁺ T cell responses.

Using both μMT^−/− mice and cotransfer models, we found that the presence of B cells did not impair CD4⁺ T cell IFN-γ production upon restimulation following B. melitensis challenge (Fig. 5 and 6). This result was surprising given that B. abortus infection in μMT^−/− mice is reported to result in an increased proportion of IFN-γ⁺CD4⁺ and IFN-γ⁺CD8⁺ T cells among splenocytes compared to WT animals 4 weeks postinfection. In the same study, however, total IFN-γ⁺ cell numbers did not significantly differ reproducibly between the two groups, but an ∼2-fold enhancement in the proportion of CD4⁺ and CD8⁺ T cells among splenocytes was found in μMT^−/− compared to WT mice (25). Therefore, the increased proportion of IFN-γ⁺ T cells within μMT^−/− spleens reported previously may have resulted in part to due to a general increase in the proportion of T cells. In addition, Goenka et al. (25) utilized whole-splenocyte restimulation with phorbol myristate acetate (PMA) and ionomycin to evaluate T cell function by flow cytometry, whereas our experiments employed purified CD4⁺ T cells stimulated via the TCR using anti-CD3. The variances in T cell stimulation techniques and time points assessed could explain the discrepancy between our findings and those of others. Regardless, these data suggest that inhibition of CD4⁺ T cell IFN-γ production is likely not a key mechanism involved in B cell-driven susceptibility to Brucella infection, as we also did not observe enhanced IFN-γ production in CD4⁺ T cell single transfer versus CD4⁺ T and WT B cell cotransfer group cells upon restimulation.

B. abortus infection in μMT^−/− mice is associated with an overall reduction in IL-10⁺ splenic cells by 4 weeks postinfection (25). We also observed a substantial impairment in IL-10 production upon restimulation of CD4⁺ T cells from μMT^−/− animals (Fig. 5), which suggests that the presence of B cells enhances T cell IL-10 production and possibly accounts for reduced IL-10⁺ splenic cells in Brucella-infected μMT^−/− mice. Importantly, this effect was not specific to μMT^−/− animal-derived CD4⁺ T cells, as our adoptive transfer experiments recapitulated these results (Fig. 6). We also found that MHCII expression was required for B cells both to promote CD4⁺ T cell IL-10 production and to inhibit CD4⁺ T cell-mediated protection (Fig. 6). T cell IL-10 production is known to have deleterious effects on the control of Brucella burden by acting upon myeloid cells (41). Thus, it is possible that B-cell-mediated promotion of CD4⁺ T cell IL-10 production hampers control of Brucella in vivo.

Considering the similar resistance observed when B cells are absent during B. abortus and B. melitensis infection (22, 25) and Brucella’s proclivity for associating with human B lymphocytes (45), it is plausible that targeting of B cells and subsequent alteration of T cell responses could be a conserved mechanism for establishment of chronic Brucella infection. For example, Brucella possesses the virulence protein PrPA, which elicits polyclonal activation and IL-10 production by B cells and enhances the capacity for development of chronic infection in mice (46, 47). Various studies report that B-cell-derived cytokines can tailor T cell differentiation during infection (39, 48, 49). This function may be exploited by Brucella, as we observed that enhanced IL-10 production by CD4⁺ T cells was at least partly dependent on MHCII expression by B cells.

Here, we demonstrate that enhanced resistance to B. melitensis infection in μMT^−/− mice stems directly from B cell inhibition of CD4⁺ T cell-mediated protection. Using both adoptive transfer and depletion experiments, we confirmed that the protective potential of CD4⁺ T cells is inhibited by the presence of B cells. Finally, we discovered that the ability of B cells to enhance susceptibility to infection and modulate CD4⁺ T cell cytokine expression was mitigated when B cells lacked MHCII. Thus, B cell-associated susceptibility during brucellosis is likely an antigen-specific process that is at least partly dependent upon direct interaction with CD4⁺ T cells. Clearly, further investigation of this conundrum is paramount to our understanding of chronic brucellosis and advancement of novel therapeutic and vaccine efforts.

MATERIALS AND METHODS

Growth conditions and bacterial strains.

All experiments with Brucella melitensis were performed in biosafety level 3 (BSL-3) facilities. B. melitensis 16M obtained from Montana State University (Bozeman, MT) was used for all studies and was grown on brucella agar (Becton, Dickinson) at 37°C/5% CO₂. Colonies were picked from agar plates, and B. melitensis was were cultured in Brucella broth overnight at 37°C in an orbital shaker. Bacterial titers were approximated via measurement of optical density at 600 nm, with inoculum diluted using sterile phosphate-buffered saline (PBS). The challenge dose was confirmed by serial dilution and plating of inoculum onto Brucella agar plates.

Mice.

All animal studies were conducted in compliance with the University of Missouri Animal Care and Use Committee and utilized 6- to 12-week-old and sex-matched animals on a C57BL/6J background. μMT^−/−, TCRα^−/−, Rag1^−/−, and MHCII^−/− animals were purchased from Jackson Laboratory (Bar Harbor, ME). Mice were administered 1 × 10⁵ CFU of B. melitensis 16M intraperitoneally (i.p.) in 200 μl of PBS. Following infection, animals were maintained in individually ventilated caging under high-efficiency particulate air-filtered barrier conditions with 12-hour light and dark cycles within animal biosafety level 3 (ABSL-3) facilities at the University of Missouri. Mice were provided food and water ad libitum.

Bacterial burden and tissue cytokine quantification.

Following infection, animals were euthanized at various time points, and spleens were aseptically collected. Tissues were homogenized mechanically (50). Samples were then serially diluted, and aliquots were plated in triplicate onto brucella agar. Plates were incubated for 3 to 4 days at 37°C/5% CO₂, colonies were enumerated, and the number of CFU/tissue was calculated as previously described (51). For spleen homogenate cytokine quantification, tissues were centrifuged at 8,000 × g for 5 minutes before addition of Halt protease inhibitor (Thermo Scientific) and penicillin/streptomycin (Gibco). Supernatants were sterile-filtered and stored at –70°C until analysis using a Luminex MagPix instrument. Cytokine measurement was carried out using Milliplex magnetic reagents according to the manufacturer’s instructions (Millipore). Data were analyzed using Milliplex Analyst software (Millipore).

Assessment of neutrophil infiltration.

Spleens were collected and single suspensions were generated as previously described (50). All samples were stained following Fc block (clone number 2.4G2, Leinco) with anti-CD11b-FITC (clone number M1/70, Leinco) and anti-Ly6G-APC-Cy7 (clone number 1A8, BioLegend). Flow cytometry was performed using a BD LSR Fortessa X-20 instrument, and neutrophils were gated as (Ly6G⁺CD11b⁺). FlowJo software (Tree Star) was used for analysis.

Adoptive transfers.

Spleens were harvested aseptically from naive mice and homogenized. CD4⁺ T cells and B cells were purified magnetically using CD4 and CD19 magnetic bead isolation kits, respectively (Miltenyi Biotec). Following purification, cells were resuspended in sterile PBS and injected via the tail vein into recipient animals 1 day prior to infection with B. melitensis. Sorted fractions were evaluated via flow cytometry on a BD LSR Fortessa X-20 instrument using anti-CD4-PE-Cy7 (clone number GK1.5, BioLegend), anti-CD3-PE (clone number 145-2011, BioLegend), anti-CD8-APC-eflour780 (clone number 53-6.7, eBioscience), anti-CD19-APC or anti-CD19-FITC (clone number 1D3, Leinco), and/or anti-B220-FITC (clone number RA3-6B2, BioLegend) to confirm purity of >90%. Following sacrifice, adoptive transfer efficacy was assessed via flow cytometry in all recipient animals.

In vivo CD4 T cell depletions.

WT or μMT^−/− animals were depleted of CD4⁺ T cells in vivo as previously described (52). Briefly, animals were treated with 0.5 mg of rat anti-CD4 monoclonal antibody (MAb) GK1.5 (Leinco) i.p. 1 day prior to challenge with 1 × 10⁵ CFU of B. melitensis 16M and then once weekly for the remainder of the study. Control animals received rat IgG (Southern Biotech). Upon euthanasia, depletion of splenic CD4⁺ T cell populations was confirmed via flow cytometry for all animals in the study.

Ex vivo T cell restimulation and cytokine quantification.

Sterile 48-well plates were coated overnight with sterile PBS containing 5 μg/ml anti-CD3 (clone number 145-2C11, Leinco) at 4°C as previously described (37–40). Prior to addition of T cells, plates were washed with sterile PBS. CD4⁺ T cells from infected animals were purified as described above and seeded at 1 × 10⁶ cells/ml in complete medium (CM; RPMI 1640, 0.1 HEPES, 1 mM sodium pyruvate, 1 mM nonessential amino acids, and 10% fetal bovine serum [FBS]) containing 5 μg/ml gentamicin and allowed to incubate at 37°C/5% CO₂. Cells were cultured for 66 hours prior to addition of Halt protease inhibitor (Thermo Scientific) and filter sterilization. Supernatants were stored at –20°C until analysis. Cell culture supernatants were then assayed for cytokine production using enzyme-linked immunosorbent assay (ELISA) kits: mouse IFN-γ (eBioscience) and mouse IL-10 (R&D Systems) as per manufacturer’s instructions.

Statistical analysis.

All comparisons of means between two groups were assessed via Student’s t test with significance set at P ≤ 0.05. Comparisons of three or more groups were conducted using one-way analysis of variance analysis (ANOVA), followed by Tukey’s test for correction of multiple comparisons. Error bars represent the standard deviation of the sample mean.