Bordetella bronchiseptica is a Gram-negative bacterium equipped with several colonization factors that allow it to establish a persistent infection of the murine respiratory tract. Previous studies indicate that B. bronchiseptica adenylate cyclase toxin (ACT) and the type III secretion system (TTSS) synergize to drive dendritic cells into an altered phenotype to down-regulate the host immune response. In this study, we examined the effects of B. bronchiseptica ACT and TTSS on murine bone marrow-derived macrophages. We demonstrate that ACT and TTSS are required for the inhibition of Ag-driven CD4+ T cell proliferation by bacteria-infected macrophages. We identify PGE2 as the mediator of this inhibition, and we show that ACT and the TTSS synergize to increase macrophage production of PGE2. We further demonstrate that B. bronchiseptica can modulate normal macrophage function and drive the immune response toward a Th17 phenotype classified by the significant production of IL-17. In this study, we show that B. bronchiseptica-infected macrophages can induce IL-17 production from naive CD4+ splenocytes, and that lung tissues from B. bronchiseptica-infected mice exhibit a strong Th17 immune response. ACT inhibited surface expression of CD40 and CD86, suppressed TNF-α production, and up-regulated IL-6 production. TTSS also synergized with ACT to up-regulate IL-10 and PGE2 secretion. These findings indicate that persistent colonization by B. bronchiseptica may rely on the ability of the bacteria to differentially modulate both macrophage and dendritic cell function leading to an altered adaptive immune response and subsequent bacterial colonization.
Macrophages are an APC population that represents the first line of defense against invading pathogens. They play important roles in both innate and adaptive immunity and can both internalize bacteria through phagocytosis and recruit other cell types to the site of infection through cytokine secretion and Ag presentation (1). The importance of macrophages in regulating the early stages of the host immune response makes them an ideal target for immunomodulation by invading pathogens. Microbes capable of establishing a persistent infection use a wide array of cell-associated and secreted molecules to alter host cell function (2). Bordetella bronchiseptica is an aerobic, Gram-negative bacterium that expresses various colonization factors including adhesions, toxins, and a type III secretion system (TTSS)3 (3). Our previous studies indicate that B. bronchiseptica adenylate cyclase toxin (ACT) and TTSS modulate multiple signal transduction pathways resulting in bacterial persistence (4, 5, 6, 7).
ACT is a potent exotoxin that has pleiotropic effects on a variety of mammalian cells (8, 9). It is a member of the repeats in the structural toxin family of bacterial exotoxins and induces increased production of cellular cAMP within infected cells (8, 9). The resulting supraphysiological levels of intracellular cAMP inhibit phagocytosis, NO production, and chemotaxis, while promoting apoptosis in macrophages (8, 10, 11, 12). Type III secretion is used by various Gram-negative bacteria to inject eukaryotic host cells with bacterial effector proteins to alter host cell function (13, 14, 15). B. bronchiseptica TTSS is required for persistent colonization of the murine lower respiratory tract and modulation of the host immune response (6).
We have previously shown that B. bronchiseptica TTSS and ACT either independently or synergistically disrupt cellular host responses and alter macrophage and dendritic cell phenotype in a cell-specific manner (4, 5, 7). Characterizing the effects of these molecules on APCs will elucidate not only the mechanisms by which persistent bacteria alter host immunity but also identify potential avenues for therapeutic immunomodulation. In this report, we examine the effect of ACT and TTSS specifically on bone marrow-derived murine macrophages (BMM) and compare some of these results to our findings with bone marrow-derived dendritic cells (BMDC).
We have reported that ACT and TTSS alter cell-signaling pathways associated with the initial host-pathogen interaction. This disruption leads to the production of immunosuppressive cytokines and the systemic suppression of IFN-γ production (4, 5). We hypothesize that ACT and TTSS disrupt APC function leading to a modulated host immune response. In this report, we assess the function of B. bronchiseptica-infected APCs by coculturing infected BMMs or BMDCs with naive splenocytes. We found that wild-type (WT) bacteria-infected BMMs inhibited Ag-induced CD4+ T cell proliferation whereas infected BMDCs had no effect on proliferation. Mutants that did not express either the TTSS or ACT failed to inhibit splenocyte proliferation, indicating that both ACT and TTSS are required for this phenotype by BMMs. Furthermore, we found that significant levels of PGE2 were produced by WT-infected macrophages and that PGE2 was a primary mediator of inhibited CD4+ T cell proliferation.
APCs play a key role in shaping the adaptive immune response by releasing Th cell-polarizing cytokines that drive the development of distinct Th cell lineages. Previously, lineages of CD4+ Th cells were primarily classified as Th1 or Th2 as determined by their production of IFN-γ or IL-4, respectively (16, 17). Recently, a new lineage of Th cells has been added to the Th paradigm which are characterized by their production of IL-17 and are now referred to as the Th17 lineage (18, 19). IL-23 has been shown to be a Th17-polarizing cytokine (20, 21, 22, 23), and PGE2 has been identified as an inducer of IL-23 production in APCs (20, 24). Consequently, PGE2 may be an early component of a Th17 immune response (20). IL-17-producing T cells have been widely associated with autoimmune inflammatory diseases (18, 19, 23, 25), but the role of IL-17-producing T cells in pathogen-related diseases has recently become of interest (20, 21, 26). In this study, we show that B. bronchiseptica-infected macrophages, when cocultured with naive splenocytes, can inhibit Ag-driven CD4+ T cell proliferation and induce significant IL-17 secretion. We also show that lung tissue from mice infected with B. bronchiseptica produce high amounts of IL-17 upon restimulation, indicating the preferential development of a Th17 immune response. The production of PGE2 and various other cytokines by macrophages infected with B. bronchiseptica may be consistent with the induction of a Th17 immune response. Previously, we classified the murine immune response to B. bronchiseptica as an altered Th1 response, but our present results suggest that the altered Th1 response may be a Th17 response (7).
Materials and Methods
BMM and BMDC were prepared as described by Lutz et al. (27) with slight modification. Briefly, bone marrow was isolated from femurs of BALB/c mice and cultured in RPMI 1640 with 10% heat-inactivated FCS (HyClone), 100 IU/ml penicillin, 100 μg/ml streptomycin, 50 μM 2-ME, 2 mM l-glutamine (Invitrogen Life Technologies), and 20 ng/ml murine M-CSF (PeproTech). Cells were cultured for 10 days, and the BMMs were primed with 2 ng/ml IFN-γ for 24 h before the experiment. The cells were then washed twice in PBS and transferred into antibiotic free media with 10 ng/ml M-CSF (BMM) or GM-CSF (BMDC). Cells were seeded in 96-well plates at 4 × 105 cells/well for infection. WT, ΔbscN, ΔcyaA, or heat-killed (HK) B. bronchiseptica were then added to the cells at a multiplicity of infection (MOI) of 10 and incubated for up to 30 min. After 30 min, gentamicin was added to kill extracellular bacteria, and the cells were further incubated for specific time points. Cell death due to infection with WT B. bronchiseptica or its mutants was negligible (data not shown).
B. bronchiseptica strains RB50 (WT), WD3 (ΔbscN), and RB58 (ΔcyaA) were described previously (28, 29, 30). Live bacteria growing in the exponential phase were used for all experiments. HK bacteria were obtained by an additional incubation of RB50 at 70°C for 30 min. Bacterial concentration was assessed by OD and adjusted for MOI of 10. Bacteria were plated and counted to confirm MOI.
Female BALB/c and DO11.10 mice 6–10 wk old were obtained from The Jackson Laboratory. All mice were cared for in compliance with the Institutional Animal Care and Use Committee at the University of Pennsylvania School of Medicine animal facility. Female BALB/c mice used in the ex vivo experiment were inoculated intranasally with WT B. bronchiseptica at 1 × 105 bacteria/mouse.
Abs and ELISAs
2 ELISA kit (Oxford Biochemicals). The neutralizing anti-IL-6 (MP5-20F3) was a gift from Dr. C. A. Hunter (University of Pennsylvania, Philadelphia, PA). ELISAs of culture supernatants for cytokines were performed according to the manufacturer’s instructions (R&D System Elisa Duo Sets). ELISA plates were read using a Synergy HT microplate reader (BioTek) and analyzed using KC4 software (BioTek). In all experiments, nonspecific binding of Abs to BMMs was blocked with unlabeled FcγRIII/II.
Surface and intracellular stains
After infection and subsequent addition of gentamicin, BMMs were cultured in vitro at 37°C for an additional 18 h. BMMs were then blocked for nonspecific binding with unlabeled FcγRIII/II Ab for 15 min at 4°C. BMMs were then stained with FITC-, PE-, PerCP-, or APC-conjugated Abs. For intracellular staining, cells were pretreated with brefeldin A for 3 h and permeabilized using the BD Pharmingen Cytofix/Cytoperm kit. Stained cells were acquired with a FACSCalibur flow cytometer (BD Biosciences) and analyzed using Flow-Jo 4.0 (Tree Star).
Ag-dependent CD4+ T cell proliferation assay
The assay was conducted using methods reported by Mosser et al. (31). DO11.10 transgenic mice, expressing the DO11.10 TCR specific for MHC class II OVA peptide (OVA323–339; American Peptide Company), were sacrificed, and the spleens were processed into a single-cell suspension by maceration on a wire mesh screen. Erythrocytes were lysed, and the remaining splenocytes were washed followed by CFSE (Molecular Probes) staining. The CFSE-stained splenocytes were then washed again and counted. BMMs and BMDCs were plated at 3 × 104 APC/well in a 24-well format. They were incubated with HK, WT, ΔbscN, or ΔcyaA at an MOI of 10 with or without OVA peptide for 30 min followed by the addition of gentamicin to all wells. Indomethacin (5.0 μM; Cayman Chemical) or 2.0 nM exogenous PGE2 (Oxford Biochemicals) was added to control wells. Two hours after addition of gentamicin, the CFSE-stained splenocytes were added to the wells at 2.5 × 106 splenocytes/well. To determine whether there were any surviving bacteria in the experiment, the cultures were plated at 12-h intervals. No CFUs were detected at any sampled time point (data not shown). The cells were cocultured for 5 days with medium exchanges on days 2 and 4. On day 5, cells were collected, surface stained, and examined by flow cytometry.
Purification of CD4+ T cells
Spleens were processed into a single-cell suspension, and CD4+ T cells were collected by negative selection. Splenocytes were labeled with a biotin-Ab mixture and anti-biotin microbeads provided in the MACS CD4+ T cell isolation kit (Miltenyi Biotec). The labeled cells were then passed through an LS separation column (Miltenyi Biotec), and CD4+ and CD4− populations were collected. A 95% pure CD4+ population was attained.
IL-17 coculture assay
WT, ΔbscN (type III secretion defective), ΔcyaA (ACT defective), and HK B. bronchiseptica
B. bronchiseptica-infected BMMs inhibit Ag-stimulated CD4+ T cell proliferation
We examined the ability of B. bronchiseptica-pulsed macrophages to induce in vitro proliferation of CD4+ T cells after Ag stimulation. BMM from BALB/c mice were infected with HK, WT, ΔbscN (type III secretion defective), or ΔcyaA (ACT deletion) B. bronchiseptica together with OVA peptide, and then incubated with CFSE-labeled OVA TCR transgenic splenocytes (from DO11.10 mice) for 5 days. The proliferation of CD4+ splenocytes was measured by the dilution of CFSE and analyzed by flow cytometry. Although macrophages infected by the ΔbscN or ΔcyaA mutants led to significant proliferation of cocultured CD4+ splenocytes (as indicated by the decrease of fluorescent intensity), WT-infected macrophages did not (Fig. 1⇓). Therefore, B. bronchiseptica-infected macrophages appear to inhibit proliferation of CD4+ T cells; this inhibition is mediated by the TTSS and ACT. Concurrent experiments with bacterial-pulsed BMDCs using CFSE-labeled splenocytes showed that WT-infected BMDCs did not inhibit CD4+ T cell proliferation. Therefore, the ability to inhibit CD4+ T cell proliferation appears to be unique to Bordetella-infected BMMs, and this phenotype requires the expression of ACT and TTSS.
B. bronchiseptica ACT and TTSS induce PGE2 production by BMM
Previous reports indicate that BMMs are a major source of PGE2 and that PGE2 can selectively inhibit CD4+ T cell proliferation (24, 32). To assess the production of PGE2 from B. bronchiseptica-infected BMMs, the concentration of PGE2 in the culture supernatants was measured by ELISA. WT-infected BMMs produced significantly higher amounts of PGE2 compared with cells infected by the ΔbscN or ΔcyaA mutants (Fig. 2⇓). This indicates that both ACT and TTSS contribute to the increased secretion of PGE2 by BMMs infected with B. bronchiseptica. This up-regulation of PGE2 production may be responsible for the observed antiproliferative effect of WT B. bronchiseptica-infected BMMs on CD4+ T cells.
PGE2 is a primary mediator of the inhibition of CD4+ T cell proliferation
To specifically examine the effect of PGE2 on CD4+ T cell proliferation, we used an inhibitor of PGE2 expression, indomethacin. We repeated the experiments described in Fig. 1⇑ in the presence of 5.0 μM indomethacin. Fig. 3⇓A shows that in the presence of indomethacin, WT bacteria-infected BMM did not inhibit the proliferation of T cells. Therefore, the inhibition of CD4+ T cell proliferation appears to be mediated via PGE2 production. When we added exogenous PGE2 (2 nM) to uninfected BMMs with splenocytes, we observed significant inhibition of CD4+ T cell proliferation (Fig. 3⇓B). These results are consistent with the conclusion that the elevated levels of PGE2 produced by WT-infected BMMs contribute to the inhibition of Ag-driven CD4+ T cell proliferation.
B. bronchiseptica-infected BMMs induce IL-17 secretion by naive CD4+ splenocytes
Previous reports indicate that PGE2 secreted by APCs can lead to the induction of IL-17 expression by splenocytes via the IL-23/IL-17 pathway (20, 24). We used an ELISA to measure IL-17 production by naive splenocytes that have been cocultured for 5 days with BMMs that had been infected by WT or mutant B. bronchiseptica. We found that WT-infected BMMs induced higher amounts of IL-17 secretion by the splenocytes in vitro than did the ΔbscN or ΔcyaA mutants (Fig. 4⇓). This indicates that both TTSS and ACT contribute to the up-regulation of IL-17 production in this in vitro system. However, because IL-17 production was not completely abrogated, TTSS and ACT may not be the only bacterial factors contributing to the production of IL-17 in this system. We also measured the effect of PGE2 on IL-17 production by pretreating WT-infected BMMs with 5.0 μM indomethacin. We found that IL-17 production was suppressed almost 3-fold when naive splenocytes were cocultured with indomethacin, confirming the role of PGE2 as an inducer of IL-17 production in this system.
Previous reports have identified IL-23 as the cytokine responsible for the development of IL-17-producing T cells (20, 21, 22, 23). IL-23 is a dimer composed of p40 and p19 subunits. We found that the addition of a neutralizing anti-p40 Ab considerably suppressed the production of IL-17 in this assay (Fig. 5⇓). Anti-IL-6 and anti-IL-10 neutralizing Abs had no inhibitory effect on IL-17 production (Fig. 5⇓). These results indicate that p40 is required for the induction of IL-17 from naive splenocytes cocultured with B. bronchiseptica-infected BMMs. The substantial production of IL-17 in this in vitro system may indicate that B. bronchiseptica-infected BMMs may direct T cells within the splenocyte population into a Th17 phenotype.
A population of unique CD4+ Th cells distinct from Th1 and Th2 lineages has recently been identified that selectively secrete IL-17 in the absence of IFN-γ and are termed Th17 cells (18, 19, 23). To confirm that B. bronchiseptica-infected BMMs induce IL-17 production from CD4+ T cells, we purified CD4+ T cells by magnetic bead separation and cocultured both the CD4+ and CD4− populations with B. bronchiseptica-infected BMMs. We then measured the levels of IL-17 and IFN-γ in the supernatants by ELISA and found that CD4+ T cells cocultured with infected BMMs produce high amounts of IL-17 and comparably little IFN-γ (Fig. 6⇓). The CD4− population produced only baseline amounts of IL-17, indicating that IL-17 is being produced almost exclusively from CD4+ T cells (Fig. 6⇓). The substantial amount of IL-17 and low amount of IFN-γ indicate the preferential induction of a Th17 response by naive CD4+ T cells in response to B. bronchiseptica-infected BMMs.
B. bronchiseptica induces a Th17 response in lung tissues from infected mice
To determine whether a Th17 response can result from an actual in vivo infection, we removed the lungs and lung draining lymph nodes from mice infected with B. bronchiseptica for 7 days. We restimulated the lung tissue with HK B. bronchiseptica for 72 h and then measured the levels of IL-17 and IFN-γ in the supernatant by ELISA. We found that restimulated cells from the lung tissues of infected mice produced >20-fold the amount of IL-17 than did cells from uninfected mice (Fig. 7⇓). Comparatively, IFN-γ production is minimal from cells of infected animals. The preferential induction of IL-17 secretion by restimulated lung tissues from B. bronchiseptica-infected mice may indicate that B. bronchiseptica is modulating the immune response away from a Th1 and toward a Th17 phenotype.
B. bronchiseptica up-regulates IL-10 and IL-6 production while suppressing TNF-α production by BMM
In addition to having an effect on T cell proliferation and differentiation, PGE2 also plays a role in regulating cytokine production from macrophages (33). To address the production of cytokines by B. bronchiseptica-infected APCs, we measured the intracellular levels of cytokines by flow cytometry. To examine the production of immunosuppressive cytokines by B. bronchiseptica-infected BMMs and BMDCs, we measured the intracellular levels of IL-10 of infected BMMs and BMDCs by flow cytometry (Fig. 8⇓A) and observed that WT-infected BMMs produced more IL-10 than the ΔbscN and ΔcyaA-infected APCs in this assay. We also used an ELISA to measure IL-10 levels from culture supernatants of APCs that were infected with WT, ΔbscN, ΔcyaA, or HK bacteria. The IL-10 ELISA was consistent with the intracellular data showing that WT-infected APCs produced higher amounts of IL-10 than did the ΔbscN- and ΔcyaA-infected APCs (data not shown). This indicates that both ACT and TTSS contribute to the up-regulation of IL-10 secretion by WT-infected BMMs and BMDCs.
IL-6 can impair the ability of BMMs to present Ag to splenocytes and may suppress T cell proliferation (34, 35). We examined the production of IL-6 in infected APCs by intracellular staining. BMMs infected by ΔcyaA bacteria produced considerably less IL-6 than did WT- and ΔbscN-infected BMMs (Fig. 8⇑B). In contrast, WT-, ΔbscN-, and ΔcyaA-infected BMDCs produced equivalent amounts of IL-6 (Fig. 8⇑B). Therefore, ACT is required for the induction of high level IL-6 production in BMMs but has no effect on BMDCs.
Our previous studies demonstrated that B. bronchiseptica suppressed the expression of proinflammatory Th1-polarizing cytokines in BMDCs (4). To address proinflammatory cytokine production in B. bronchiseptica-infected APCs, we measured the intracellular levels of TNF-α by flow cytometry. TNF-α levels in both WT- and ΔbscN-infected BMMs were significantly lower than that from ΔcyaA-infected BMMs (Fig. 8⇑C). We also measured the cytokine levels of TNF-α, IL-12p70, and IFN-γ by ELISA. Analysis by ELISA of TNF-α production in B. bronchiseptica-infected BMMs (data not shown) also correlated with the intracellular staining data. Therefore, ACT is required for the inhibition of TNF-α in B. bronchiseptica-infected BMMs. Likewise, intracellular cytokine staining of infected BMDCs demonstrated ACT-dependent inhibition of TNF-α production (Fig. 8⇑C). The cytokine levels of IL-12p70 and IFN-γ were below the detection limits for all BMM samples (data not shown). Therefore, ACT is required for the suppression of TNF-α production in both BMMs and BMDCs infected with B. bronchiseptica.
B. bronchiseptica ACT down-regulates surface expression of costimulatory molecules CD86 and CD40
Our previous studies showed that the Bordetella TTSS mediated the increased expression of MHCII, CD86, and CD80 surface molecules in infected BMDCs, whereas ACT inhibited CD40 expression (4, 5). We assayed BMMs infected by various strains of B. bronchiseptica to determine the effect of TTSS and ACT on the surface expression of stimulatory and costimulatory molecules. BMMs were pulsed with bacteria for 30 min and then cultured for 18 h and surface stained. Fig. 9⇓ shows that WT, ΔbscN, and ΔcyaA bacteria all up-regulated MHCII and CD80 expression in the infected BMMs to similar levels. Infection by WT B. bronchiseptica or the ΔbscN mutant did not lead to significant up-regulation of the costimulatory molecule CD86 in these cells. However, BMMs infected by the ΔcyaA mutant showed a significant increase in the surface expression of CD86. The surface expression of CD40 was slightly up-regulated in BMMs infected by WT or ΔbscN bacteria, but the ΔcyaA mutant significantly up-regulated CD40 surface expression. Therefore, ACT is required for the inhibition of expression of CD86 and CD40 in infected BMMs, and the expression profile of stimulatory and costimulatory molecules is different from that of infected BMDCs.
In summary, the inhibition of Ag-driven CD4+ T cell proliferation and the induction of a Th17 immune response may be significant factors in the establishment of a persistent infection by B. bronchiseptica. The increased production of PGE2, IL-6, and IL-10 and the suppression of TNF-α production, combined with the differential up-regulation of costimulatory surface molecules in BMMs infected by B. bronchiseptica, may constitute at least some of mechanisms whereby these infected BMMs can alter the host immune response. Studies with the mutants also indicate that the Bordetella ACT and TTSS play significant roles in mediating these phenotypes and are thus critical colonization factors for the modulation of the host immune response to B. bronchiseptica infection.
B. bronchiseptica establishes a persistent asymptomatic respiratory infection in the host and maintains a balanced interaction with the host immune response. Interactions of B. bronchiseptica with dendritic cells and macrophages are likely to have profound effects on the development of innate and adaptive host immune responses that mediate the interaction between the host and the pathogen. We have previously shown that dendritic cells infected by B. bronchiseptica drive the host immune response toward an immunosuppressive phenotype; away from that of a classical Th1 response that clears the bacteria from the host (7). Macrophages, along with dendritic cells, direct the differentiation of naive CD4+ T cells into functionally distinct Th1, Th2, Th17, or regulatory T cell phenotypes (36, 37). In this study, we show that B. bronchiseptica promotes the production of immunosuppressive cytokines by infected macrophages and that these cells can induce high levels of IL-17 expression when cocultured with naive splenocytes. We also show the induction of IL-17 from lung tissues of B. bronchiseptica-infected mice, suggesting that B. bronchiseptica infection leads to the induction of a Th17 immune response.
The proliferation of CD4+ T cells upon interaction with activated APCs is a critical component of a productive adaptive immune response. The failure of APCs to stimulate CD4+ T cell proliferation could permit bacterial colonization and result in the inability of the host to clear the pathogen (38). In this report, we show B. bronchiseptica-infected BMMs can substantially inhibit Ag-dependent CD4+ T cell proliferation in vitro. A previous report has shown that the human pathogen, Bordetella pertussis, a species closely related to B. bronchiseptica, can also mediate the inhibition of Ag-dependent CD4+ T cell proliferation via macrophages (39). However, the mechanism by which B. pertussis mediates this phenotype is not clear. In this study, we show that ACT and TTSS contribute to the inhibition of CD4+ T cell proliferation in B. bronchiseptica infection, given that both the ΔbscN and ΔcyaA mutant-infected macrophages failed to inhibit proliferation. Conversely, B. bronchiseptica-infected BMDCs did not inhibit the proliferation of CD4+ T cells in this system. We have previously reported on the contributions of ACT and TTSS on BMDC signaling and phenotype (4). Therefore, the inhibition of CD4+ T cell proliferation by BMMs may be due to the differential effects of ACT and TTSS on BMMs compared with BMDCs.
We found that both TTSS and ACT are required for the up-regulation of PGE2 production by infected BMMs (32). PGE2 is produced primarily by macrophages, suggesting that PGE2 production is pivotal in driving specific immune responses during early stages of infection (40). Nataraj et al. reported that PGE2 suppresses Th1 immune responses including CD4+ T cell proliferation. Therefore, we hypothesized that the increased levels of PGE2 from WT-infected BMMs may account for the inhibition of CD4+ T cell proliferation. We found that indomethacin, an inhibitor of PGE2 production, abrogated the inhibitory effect on proliferation in our experimental system. Conversely, the addition of exogenous PGE2 to uninfected BMMs inhibited Ag-driven CD4+ T cell proliferation. These findings are consistent with the idea that PGE2 production by infected macrophages can play a significant immunomodulatory role during B. bronchiseptica infection.
Previous studies indicate that PGE2 modulates various host immune functions and plays a key role in Th cell differentiation (41). Its production by innate immune cells has recently been associated with the development of IL-17-producing T cells via the IL-23/IL-17 pathway (20). Our present studies show that B. bronchiseptica-infected BMMs can induce significant production of IL-17 from naive splenocytes (Fig. 4⇑). Similar to the induction of PGE2 expression in our experimental system, the production of IL-17 was greatest in the splenocytes cocultured with WT-pulsed BMMs compared with those infected by ΔbscN and ΔcyaA mutants, indicating that both TTSS and ACT contribute to this phenotype. The pretreatment of WT-infected BMMs with indomethacin, a PGE2 inhibitor, suppressed IL-17 secretion from naive splenocytes; thus confirming the role of PGE2 as a promoter of IL-17 expression in this system (Fig. 4⇑). Recent studies have shown that IL-23 is often required for the development of IL-17-producing T cells (20). IL-23 is composed of two subunits, p40 and p19, and we found that an anti-p40 neutralizing Ab significantly suppressed the production of IL-17 in our assay system (Fig. 5⇑). Although p40 is also a subunit of IL-12, we did not detect any IL-12 production from BMMs infected by B. bronchiseptica. Therefore, the inhibitory effect of the anti-p40 Ab is most likely directed via the inhibition of IL-23 activity.
The induction of IL-17 from naive splenocytes suggests the development of a Th17 immune response. To classify the production of IL-17 we found in this study as the development of a Th17 immune response we had to confirm that IL-17 was coming from CD4+ T cells. We purified CD4+ T cells by negative selection and found that the IL-17 in our assay was indeed coming almost exclusively from CD4+ T cells (Fig. 6⇑). The lack of IFN-γ production in these cells is further evidence of the minimal development of any Th1 responses in this system. These data provide strong evidence that B. bronchiseptica-infected BMMs can drive the development of a Th17 immune response in vitro. To address the possibility of an in vivo Th17 response to B. bronchiseptica infection, we removed the lungs and lung draining lymph nodes from infected mice and evaluated the IL-17 and IFN-γ production from these cells. We found that restimulated cells from the lung tissues of infected mice produced substantial amounts of IL-17 compared with IFN-γ, indicating that B. bronchiseptica infection leads to the induction of a Th17 immune response in vivo. The generation of a Th17 response in vivo has been shown to be important in host defense against Klebsiella pneumoniae (21). Our findings indicate that B. bronchiseptica can also induce a Th17 immune response evident by the production of IL-17 both in vitro and ex vivo.
The ability of B. bronchiseptica to establish a persistent infection in vivo may depend on its ability to generate an immunosuppressive environment early in the infection process. B. bronchiseptica ACT and TTSS are required for the up-regulated expression of IL-10 and the suppression of TNF-α production by infected BMMs. IL-10 modulates the immune response at the site of infection by suppressing the production of proinflammatory cytokines by neighboring cells, whereas TNF-α is a proinflammatory cytokine that is essential for mounting an early immune response against an invading pathogen (42, 43). IL-10 produced by macrophages may also hinder the function of dendritic cells and other neighboring APCs in vivo (42). Previous reports have shown that IL-10 expression may contribute to the generation of IL-17-producing T cells (35), but inhibition of IL-10 does not appear to have an effect on the production of IL-17 in our in vitro system (Fig. 5⇑). Our previous studies have shown that splenocytes from mice infected with WT B. bronchiseptica produce increased amounts of IL-10 and decreased amounts of IFN-γ upon restimulation with HK Bordetella (7). This is indicative of the inhibitory effect of B. bronchiseptica on the development of a Th1 immune response, and the suppression of IFN-γ can also create favorable conditions for the development of a Th17 response (23, 35). The increased levels of IL-10 produced by infected BMMs may contribute to the generation of immunosuppressive responses that shift the immune response away from a Th1 phenotype. We also examined the IL-6 production of B. bronchiseptica-infected BMMs and found that ACT is required for the up-regulation of IL-6. Van Heyningen et al. (34) demonstrated that IL-6 from macrophages infected by Mycobacterium tuberculosis can suppress the ability of macrophages to process and present Ag to T cells. Previous reports have also shown that IL-6 can stimulate the polarization of naive T cells into a Th17 phenotype (35), but IL-6 does not appear to play a significant role in the induction of IL-17 production in our system (Fig. 5⇑). This might be due to our use of a mixed population of splenocytes compared with a purified population of naive CD4+ T cells (CD25−CD44lowCD62Lhigh) (35). Overall, the cytokine expression profile of macrophages infected by B. bronchiseptica is consistent with the generation of an immunosuppressive environment.
Stimulation of T cells and other aspects of host immune responses are dependent on the surface expression of stimulatory and costimulatory molecules on APCs. Our previous studies indicate that B. bronchiseptica up-regulates surface expression of MHCII, CD86, and CD80, but not CD40, on infected BMDCs (4). Our present study shows that B. bronchiseptica up-regulated expression of MHCII and CD80 but failed to up-regulate CD86 and CD40 on infected BMMs. ACT is required for this down-regulation of CD86 and CD40. The down-regulation of CD86 is a phenotype specific to infected BMMs, and it is not observed on infected BMDCs (4). In contrast to that observed on BMDCs where TTSS up-regulates MHCII, CD80, and CD86 (4), Bordetella TTSS has no apparent effect on the expression of any BMM-stimulatory molecules. The failure of B. bronchiseptica to up-regulate CD86 expression by BMMs may promote the inhibitory effect of infected BMMs on Ag-driven CD4+ T cell proliferation due to the lack of costimulation. Therefore, inhibition of CD86 surface expression, coupled with a down-modulated expression of CD40, may prevent the activation of T cells by B. bronchiseptica-infected BMMs.
We hypothesize that the overall effects of B. bronchiseptica on APCs lead to the creation of an altered immune response that favors the pathogen. Our previous findings with dendritic cells suggest that B. bronchiseptica ACT and TTSS synergize to direct dendritic cells into a semimature phenotype (4). This may lead to the generation of an immunosuppressive regulatory T cell response (44). Supporting this conclusion, we have also reported that persistent WT colonization is associated with a systemic decrease in IFN-γ production and increased secretion of IL-10 (7). In this report, we show that B. bronchiseptica-infected macrophages may also contribute to reduced Th1 responses through an altered phenotype modulated by ACT and the TTSS. This altered phenotype can lead to the inhibition of OVA-stimulated CD4+ T cell proliferation and also redirects polarization of the immune response away from a Th1 response and toward a Th17 response. Our study shows that restimulated lung tissues from mice infected with B. bronchiseptica produce substantial amounts of IL-17. This suggests that Th17 responses are induced during an in vivo infection. Happel et al. (21) has shown that K. pneumoniae induces a Th17 response in vivo and that this response is necessary for host defense, thus providing evidence that a Th17 response may be necessary to facilitate the clearance of extracellular pathogens. However, further studies are needed to establish the consequence of a Th17 host immune response in B. bronchiseptica infection. We are also continuing to investigate the mechanism by which B. bronchiseptica-infected BMMs induce IL-17 secretion by CD4+ T cells and are also examining the effects of B. bronchiseptica-infected dendritic cells on naive T cells. The role of Th17 differentiation in a pathogen-related immune response is an exciting new paradigm, and the B. bronchiseptica infection system may be an excellent model for studying this pathway.
We thank members of the Yuk laboratory for their assistance in preparing the manuscript. We also thank Dr. Hao Shen and Dr. C. A. Hunter and their respective laboratories for scientific and technical input.
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.
↵1 This work was supported by grants from the National Institutes of Health (AI049346) and the Phillip Morris Research Management Group.
↵2 Address correspondence and reprint requests to Dr. Ming Yuk, 201C Johnson Pavillion, 3610 Hamilton Walk, Philadelphia, PA 19104. E-mail address:
↵3 Abbreviations used in this paper: TTSS, type III secretion system; ACT, adenylate cyclase toxin; BMM, bone marrow-derived macrophages; BMDC, bone marrow-derived dendritic cells; WT, wild type; HK, heat killed; MOI, multiplicity of infection.
- Received May 16, 2006.
- Accepted August 25, 2006.
- Copyright © 2006 by The American Association of Immunologists