IL-17A–Producing γδT Cells Promote CTL Responses against Listeria monocytogenes Infection by Enhancing Dendritic Cell Cross-Presentation

Interleukin-17 is a proinflammatory cytokine mainly produced by a unique subset of CD4⁺ Th cells (Th17) that are distinct from Th1 and Th2 cells (1, 2). Overproduction of IL-17A and Th17 cells has been associated with a number of autoimmune diseases, including multiple sclerosis, rheumatoid arthritis, inflammatory bowel disease, and systemic lupus erythematosus (3–6). IL-17A also promotes host defense against a number of pathogens (7, 8). It has been well accepted that the IL-17A pathway has evolved as protective mechanism in host defense against extracellular bacterial infections, including Klebsiella pneumoniae (9), Citrobacter rodentium (10), and Pseudomonas aeruginosa (11), mainly by inducing production of CXC chemokines (CXCL8, CXCL1, and CXCL10) and cytokines (IL-6, GM-CSF, and G-CSF) from endothelial/epithelial cells, fibroblasts, osteoblasts, and macrophages. These cytokines and chemokines sequentially induce the expansion and accumulation of neutrophils, which are much efficient in killing extracellular bacteria (7, 8). In contrast, IL-17A appears to be dispensable for protection against most studied intracellular bacterial infections like Mycobacterium tuberculosis (12) and Salmonella typhimurium (13). However, the involvement of IL-17A in host defense against primary Francisella tularensis infection (14) and secondary M. tuberculosis challenge were also reported (15). Therefore, the role of IL-17A in intracellular bacterial infection is controversial and needs further investigation. Listeria monocytogenes is another intracellular bacterium, which mainly infects macrophages in vivo, and needs effective cytotoxic T cells for complete clearance from the body. Although IL-17A was shown to be indispensable in the innate protection against L. monocytogenes (16), the precise mechanisms are not fully demonstrated.

IL-17A exerts its functions mainly by orchestrating innate immune responses. Recent evidence, however, strongly implicated that IL-17A extends its function in adaptive immunity. For example, a study in IL-17A–deficient mice demonstrated the importance of IL-17A in delayed type hypersensitivities (17), suggesting IL-17A may play a role in CD4⁺ Th cell response. IL-17A has been shown to suppress Th1 cell differentiation directly through IL-17RA on CD4⁺ T cells and can protect the mice from T cell-mediated intestinal inflammation (18). However, the influence of IL-17A on CD8⁺ cytotoxic T cell responses against L. monocytogenes infection and the cell sources of IL-17A in the process need to be fully understood.

γδT cells are a small subset of T cells that recognize a wide range of microbial products in a pattern recognition model and are CD4, CD8 double negative (19, 20). They express a high level of cytolytic effector molecules, such as granzyme A, granzyme B, and Fas ligand, and may help defense against pathogens via targeted killing (19). γδT cells may also modulate adaptive immune responses by secreting both proinflammatory (e.g., IFN-γ) and anti-inflammatory cytokines (e.g., IL-4 and IL-10) (21, 22). Recently, γδT cells but not Th17 cells have been shown to be the primary source of IL-17A production in the early phase of Mycobacterium tuberculosis (23) and Escherichia coli infection (24).

The purpose of this study is to identify the potential role of IL-17A in L. monocytogenes infection, especially in the regulation of adaptive CTL responses. Our data demonstrate that γδT cells are the major source of IL-17A production in vivo in response to L. monocytogenes infection, and IL-17A is required not only for innate response but also for optimal adaptive CD8⁺ T cell response against bacterial infection. IL-17A acts on dendritic cells (DCs) to upregulate their expression of MHC class I (MHC-I) molecules and makes DCs produce more IL-12, IL-6, and IL-1β, thus facilitating cross-priming and proliferation of CD8⁺ T cells. Therefore, our data suggest that γδT cell-derived IL-17A is critical for optimal induction of CTL responses and protection against primary intracellular L. monocytogenes infection. These findings may be helpful to look for new strategies that improve the efficacy of vaccination in the treatment of infection and tumor.

C57BL/6J mice were obtained from Sipper BK Experimental Animals (Shanghai, China). Ifng^−/−, CD45.1, OT-I, and OT-II mice were from The Jackson Laboratory (Bar Harbor, ME). Il17a^−/− mice were kindly provided by Dr. Chen Dong (M.D. Anderson Cancer Center, Houston, TX), and backcrossed to the C57BL/6J background for 10 generations. Il17a^+/+ and their Il17a^−/− littermates from Il17a^+/− heterozygous mating were used in the experiments. All animals were maintained in a specific pathogen-free facility and used at 6–10 wk of age.

Virulent and attenuated recombinant L. monocytogenes that secret OVA protein (LM-OVA and ΔactA LM-OVA) were kindly provided by Dr. Hao Shen (University of Pennsylvania School of Medicine, Philadelphia, PA). Mice were inoculated i.v. with 1 × 10⁶ ΔactA LM-OVA (25), unless specified otherwise. In some experiments, 100 μg recombinant murine (rm) IL-17A (Biolegend, San Diego, CA) or 200 μg anti-mouse IL-17A mAb (TC11-18H10.1, Biolegend) was injected i.p. from day 1 postinfection every other day until the end of the experiments.

Anti-CD4 (L3T4), anti-CD8a (Ly-2), anti-CD11b (M1/70), anti-CD25 (3C7), anti-CD40 (3/23), anti-CD44 (IM7), anti-CD45.1 (A20), anti-CD62L (MEL-14), anti-CD80 (16-10A1), anti-CD86 (GL1), anti–H-2K^b (AF6-88.5), anti–I-A/I-E (2G9), anti-NK1.1 (PK136), anti-Gr1, anti-γδTCR (GL3), anti–IFN-γ (XMG1.2), and anti–IL-17A (TC11-18H10) mAbs were from BD Pharmingen (San Diego, CA). Anti-CCR7 (4B12) was purchased from eBioscience (San Diego, CA). Surface staining was performed as described previously (26). For intracellular cytokine staining, cells were stimulated with 25 ng/ml PMA and 500 ng/ml ionomycin (Sigma-Aldrich, St. Louis, MO) or Ag peptide OVA_257–264 (SIINFEKL) or listeriolysin O_190–201 (NEKYAQAYPNVS) as indicated for 6 h at 37°C. Brefeldin A (10 μg/ml; eBioscience) was included for the last 4 h of incubation. Cells were stained with the Cytofix/Cytoperm kit according to the manufacturer’s instructions (eBioscience).

Bone marrow-derived DCs (BMDCs) were generated as described previously (26, 27). Briefly, bone marrow progenitors were cultured in a six-well plate at 5 × 10⁶ cells/well with RPMI 1640 medium containing 10 ng/ml GM-CSF and 1 ng/ml IL-4 (Biolegend). On day 3, nonadherent cells were gently washed out. The remaining cells were cultured for an additional 4 to 5 d and used as BMDCs.

Naive OT-I T cells were sorted by MoFlo XDP (DakoCytomation, Glostrup, Denmark) as CD44^lo and CD62L^hi CD8⁺ cells from OT-I mice. DCs were purified first from pooled spleens (typically three) by density gradient centrifugation as described previously (28), and then cells from the low-density fraction were further sorted with DakoCytomation MoFlo XDP (DakoCytomation) for CD8⁺CD11b⁻CD11c⁺ DCs. Cell purity was >97%.

Murine BMDCs were pulsed for 4 h with OVA protein (2 μg/ml) in the presence or absence of 100 ng/ml IL-17A. Cells were washed and cocultured with OVA-specific OT-I T cells at a ratio of 1:10. Fresh IL-17A was added to the experiment group. BMDCs loaded with BSA (2 μg/ml) were set as negative control. For ex vivo experiments, different DC preparations from wild-type (WT) or Il17a^−/− mice were used as stimulators for naive OT-I T cells, and the DC/T ratio was 1:1.

CD8⁺ CTL cell cytotoxicity was analyzed using flow cytometry as described previously (29, 30). Briefly, splenocytes were harvested at 7 d post L. monocytogenes infection, and CD8⁺ T cells enriched by MACS were used as effector cells. EL-4 cells were pulsed with 2 μM OVA_257–264 peptide at 37°C for 1 h. Peptide-pulsed or unpulsed EL-4 cells (2 × 10⁴/well), as target cells, were cocultured with effector cells in 96 U-bottom plates at series ratios. After incubation for 6 h at 37°C, cells were collected and stained with fluorescein isothiocyanate-conjugated Annexin V, allophycocyanin-conjugated anti-CD8 and 7-aminoactinomycin D (7-AAD). Live EL-4 target cells were identified as CD8⁻Annexin V⁻7-AAD⁻ using the FACS LSRII (BD Biosciences). Specific lysis (%) was calculated as: [(CD8⁻ cells − CD8⁻Annexin V⁻7-AAD⁻ cells)/(CD8⁻ cells)] × 100. Spontaneous apoptosis was <10% for all experiments.

Freshly isolated naive OT-I T cells were stained with 5 μM fluorescent dye CFSE for 15 min at 37°C and washed in 1% BSA three times. A total number of 2 × 10⁶ OT-I T cells in 200 μl PBS were transferred i.v. To deplete autologous γδT cells or CD4⁺ T cells in vivo, mice were given 200 μg/mouse anti-γδTCR (UC7-13D5, Biolegend) or anti-CD4 mAb (GK1.5), respectively, 3 d before infection. Depleting rate was confirmed by flow cytometry >90%. γδT cells were sorted by MoFlo XDP with anti-γδTCR (GL3, Biolegend) mAb from splenocytes of WT or Il17a^−/− mice, and the cells were used for adoptive transfer (5 × 10⁵/mouse). Purity of γδT cells was determined as >98% by FACS.

The levels of IL-17A, IFN-γ, IL-12p70, IL-6, and IL-1β were determined using ELISA kits (R&D Systems, Minneapolis, MN).

RNA was extracted from spleen cells using TRIzol (Invitrogen, Carlsbad, CA) and reverse transcribed. Gene expression was examined with a Bio-Rad iCycler Optical system (Bio-Rad, Hercules, CA). Primers used in this study were: Il17a (forward, 5′-GGCTGACCCCTAAGAAACC-3′; reverse, 5′-CTGAAAATCAATAGCACGAAC-3′); and Actb (forward, 5′-GCTGCGT TTTACACCCTTTC-3′; reverse, 5′-GCTGTCGCCTTCACCGTTC-3′).

Data are shown as mean ± SD, and the significance of differences between two means was analyzed using the Student t test. Differences were considered as statistically significant if p < 0.05.

To determine whether IL-17A is involved in the protective immune responses against L. monocytogenes infection, we first analyzed the expression of IL-17A in WT mice postinfection. We found that both IL-17A mRNA in the spleen and IL-17A in the serum increased significantly from day 1, peaked on day 5, declined by day 7, and returned to normal level after 9 d (Fig. 1A, 1B). We then infected Il17a^−/− mice to analyze the role of IL-17A in L. monocytogenes infection. About 10 times higher bacterial burden in the spleen and liver were found in Il17a^−/− mice on day 3 compared with that in WT mice (Fig. 1C), suggesting a protective function of IL-17A in the innate immune response against LM-OVA infection. Previous studies have found that IL-17A can induce neutrophils through the induction of G-CSF and CXC chemokines and protect against certain pathogens, especially extracellular bacteria (7, 8). We calculated Gr1⁺CD11b⁺ neutrophils in the blood (Fig. 1D) and the organs including spleen and liver (Supplemental Fig. 1A). A lower percentage of neutrophils was found accumulated in Il17a^−/− mice compared with their WT control at early times postinfection, which may account for their early susceptibility to L. monocytogenes infection. We further noticed the bacterial count discrepancy between Il17a^−/− mice and WT mice exacerbated to >1000 times on day 7 (Fig. 1C), when innate responses had declined and T cell responses began to dominate, indicating IL-17A may also function in the late adaptive immune responses.

More bacterial load or longer duration of infection may result in more significant CTL response (31, 32). To determine the precise role of IL-17A in adaptive immune response, we should overcome the different bacterial burden during the early phase between Il17a^−/− and WT mice. Interestingly, ΔactA LM-OVA, an attenuated strain, could also induce the production of IL-17A in WT mice (Fig. 1A, 1B) and were rapidly cleared from the spleen and liver of WT and Il17a^−/− mice at similar rates (Fig. 1E). Nonetheless, Il17a^−/− mice expanded and recruited the same percentage of neutrophils in the blood (Fig. 1F), spleen, and liver (Supplemental Fig. 1B), compared with WT mice. Therefore, the attenuated LM-OVA infection serves as an ideal in vivo model to investigate the adaptive immune responses against bacterial infection (33).

An effective cytotoxic T cell response is needed in the complete clearance of L. monocytogenes. Il17a^−/− mice displayed much more severe bacterial burden than WT mice, with more pronounced effect on day 7 than day 3 (1200:1 versus 8:1), suggesting the CTL response was impaired in the absence of IL-17A. After ΔactA LM-OVA infection, a lower percentage of CD8⁺ T cells exhibited the activated phenotype (CD44^hi or CD62L^lo) during the entire course in Il17a^−/− mice (Fig. 2A, Supplemental Fig. 2A). No significant difference was found in activated CD4⁺ T cells (Supplemental Fig. 2A). On day 7 postinfection, splenic CD8⁺ T cell-mediated cytotoxicity against OVA-pulsed EL4 target cells was greatly reduced in Il17a^−/− mice compared with that in WT mice (Fig. 2B). We further observed that both the frequency and absolute number of OVA-specific effector CD8⁺ T cells in the spleen were significantly lower in Il17a^−/− mice on day 7 (Fig. 2C, 2D). Similar results were also obtained in mice immunized with OVA protein (Supplemental Fig. 3A) and infected with different dosages of ΔactA LM-OVA (Supplemental Fig. 3B). However, the number of listeriolysin O-specific CD4⁺ T cells was similar between WT and Il17a^−/− mice (Supplemental Fig. 2B).

It is possible that induction of protective CTLs delayed in the presence of IL-17A. To address this possibility, we observed Ag-specific CTLs in WT and and Il17a^−/− mice every other day from day 1–12 postinfection. The expansion of CTLs in both mice reached their peak on day 7, but the peak expansion of CTLs was less in Il17a^−/− mice as compared with that in WT mice (Fig. 2D). Il17a^−/− mice infected with virulent LM-OVA showed consistent results with those infected with attenuated LM-OVA (Supplemental Fig. 5). These results suggest that the induction of protective CTLs was declined in the absence of IL-17A.

Furthermore, blockade of IL-17A signal via anti–IL-17A mAb decreased effector CD8⁺ T cell numbers in WT mice, whereas exogenous IL-17A partially rescued defective CD8⁺ T cell response in Il17a^−/− mice (Supplemental Fig. 2B, 2C). However, the CD8⁺ effector T cells in WT and Il17a^−/− mice had a similar phenotype and produced the same level of IL-2, TNF-α, granzyme B, and perforin (Supplemental Fig. 4A). Taken together, these results clearly show that IL-17A deficiency selectively impairs the generation of CD8⁺ effector T cells against primary infection, but does not influence their phenotype and potential function.

IL-17A is reported to be involved in the memory T cell response against secondary M. tuberculosis challenge (15). We then went on to identify the role of IL-17A in memory CTL formation after L. monocytogenes infection. On day 30 postinfection, the amount of memory CD8⁺ T cells was comparable between WT and Il17a^−/− mice (Fig. 2C). Considering a defective expansion during the effector phase, CD8⁺ T cells exhibited a lower degree of contraction in Il17a^−/− mice (Fig. 2D). In addition, memory CD8⁺ T cells in both kinds of mice expressed the same level of CD127, CD44, and CD62L and secreted same level of IFN-γ, TNF-α, and IL-2 (Supplemental Fig. 4B).

Mice were then challenged with virulent LM-OVA to access their secondary recall responses of memory T cells. No difference in bacterial burden was observed between WT and Il17a^−/− mice postchallenge (Fig. 2E). The rapid activation and expansion of memory T cells were also similar (Fig. 2C, 2D), suggesting that memory T cell response was IL-17A independent. These data demonstrate that Il17a^−/− mice generate the same level of fully functional memory T cells against secondary L. monocytogenes challenge, although they have a decreased number of effector CD8⁺ T cells during the primary infection.

IL-17A signals may act on CD8⁺ T cells directly or via an intermediate cell type. In vitro study revealed that IL-17A did not affect the proliferation and apoptosis of naive CD8⁺ T cells when T cells were activated with CD3 plus CD28 mAbs (Fig. 3A, 3B). However, proliferation of CD8⁺ T cells was significantly decreased in Il17a^−/− mice as well as in WT mice receiving the neutralizing anti–IL-17A Ab (Fig. 3C). Exogenous IL-17A partially rescued the defective proliferation of CD8⁺ T cells in Il17a^−/− mice. These results demonstrated that IL-17A indirectly exerts its function on CD8⁺ T cells to induce their optimal proliferation.

To investigate whether regulatory CD4⁺ T cells are involved in the suppression of CD8⁺ T cell proliferation in Il17a^−/− mice, we detected the regulatory T cells postinfection. However, the percentage of regulatory CD25⁺Foxp3⁺CD4⁺ T cells in Il17a^−/− mice was the same as in WT mice after L. monocytogenes infection (Supplemental Fig. 6), thus eliminating the possibility that IL-17A enhances CD8⁺ T cell proliferation by decreasing regulatory T cells.

Professional APCs, especially DCs, can process and present exogenous Ags on MHC-I molecules, thus playing an important role in priming naive CD8⁺ T cells. So, DCs may most likely serve as the intermediate cells between IL-17A and CD8⁺ T cells. In fact, IL-17A stimulated OT-I T cells to proliferate and produce IFN-γ more significantly when they were cocultured with BMDCs (Fig. 4A). The endocytosis ability of BMDCs was not changed after IL-17A stimulation (Fig. 4B). However, IL-17A could upregulate the expression of MHC-I molecule H2-K^b (Fig. 4C) without affecting the expression of CD40, CD80, CD86, and MHC class II (data not shown). IL-17A but not IL-17F stimulated BMDCs to secrete more IL-12, IL-6, and IL-1β and was more pronounced when stimulated with LPS (Fig. 4D). These data indicate IL-17A may enhance cross-presentation of BMDCs by inducing upregulation of MHC-I molecules and production of cytokines. Furthermore, the enhancement of BMDCs cross-presentation by IL-17A could be largely blocked by anti–IL-12 mAb (Fig. 4E).

We then tried to analyze cross-presentation of DCs in vivo in the absence of a IL-17A signal after L. monocytogenes infection. The absolute number of DCs and their subsets in the spleen were the same between WT and Il17a^−/− mice (Fig. 5A). CD8α⁺ DCs, the major cells contributing to cross-presentation (34–36), showed no significant differences in their expression of membrane molecules between all kinds of mice (Fig. 5B). However, CD8α⁺ DCs isolated from Il17a^−/− mice or WT mice neutralized of a IL-17A signal had significantly reduced capability to produce IL-12 (Fig. 5C) and were less potent to promote OT-I T cell activation and proliferation (Fig. 5D). Therefore, IL-17A enhances DCs cross-presentation both in vitro and in vivo.

Serum IL-17A reached its peak at day 5 postinfection (Fig. 1A). At that time, γδT cells produced abundant IL-17A under stimulation, even without stimuli, albeit at a lower level (Fig. 6A). Low but detectable IL-17A–producing CD4⁺ T and NKT cells were also found. The IL-17A response in γδT cells was both faster and stronger than that in other cells, like Th17 cells (Fig. 6B, 6C), suggesting a key role of γδT cells contributing to serum IL-17A. Relatively abundant γδT cells were also found accumulating in the lymph node and liver, but not in the thymus or bone marrow (Fig. 6D).

To further confirm our hypothesis that γδT cells but not Th17 cells were the predominant cell sources for the increased serum IL-17A, we depleted γδT cells and CD4⁺ T cells in vivo, respectively. Mice depleted of γδT cells but not CD4⁺ T cells showed a remarkable defect in serum IL-17A elevation postinfection (Fig. 6E). Additionally, adoptive transfer of γδT cells, but not CD4⁺ T cells or Il17a^−/− γδT cells, partially recovered serum IL-17A and CD8⁺ T cell priming in Il17a^−/− mice (Fig. 6F, 6G).

We further tried to characterize these IL-17A–producing γδT cells. They expressed high CD44 and CD11a and low CD27 and CD25, just like IL-17A nonsecreting γδT cells postinfection (data not shown). Intracellular cytokine staining revealed that IL-17A–producing γδT cells were distinct from IFN-γ–producing γδT cells (Fig. 6H). Under L. monocytogenes infection, only a minor percent of γδT cells produced IFN-γ, whereas a much more abundant number of these cells produced IL-17A. Furthermore, IL-17A production by γδT cells was not affected by the absence of IFN-γ. As a matter of fact, Ifng^−/− mice even showed more IL-17–producing γδT cells postinfection (Fig. 6H). Therefore, IL-17A–producing γδT cells represent a distinct subset and are the predominant source of early IL-17A after L. monocytogenes infection.

To determine whether γδT cells could promote DCs cross-priming by IL-17A, we stimulated BMDCs with LPS in the presence or absence of IL-17A–producing γδT cells. IL-17A–producing γδT cells augmented IL-12 production by BMDCs (Fig. 7A) and consequently promoted the proliferation and IFN-γ production of OT-I T cells triggered by such BMDCs (Fig. 7B, 7C). Addition of anti–IL-17A mAb significantly reduced this efficiency. Furthermore, in vivo depletion of γδT cells significantly reduced IL-12 production from CD8α⁺ DCs and suppressed effector CD8⁺ T cell differentiation, whereas depleting CD4⁺ T cells did not have such effect (Fig. 7D, 7E). We also transferred Il17a^+/+ or Il17a^−/− γδT cells to WT mice depleted of endogenous γδT cells. Only adoptive transfer of Il17a^+/+ γδT cells recovered the production of IL-12 from CD8α⁺ DCs and thus rescued the generation of optimal effector CD8⁺ T cells (Fig. 7D, 7E). These data suggested a nonredundant role of IL-17A from γδT cells in promoting CD8⁺ T cell proliferation after L. monocytogenes infection.

Th17 cells, differentiated from naive CD4⁺ T cells in the presence of TGF-β and IL-6, are recognized as the major cell type that produces IL-17A (37–39). The importance of IL-17A is well accepted in innate immunity, especially in activation and recruitment of neutrophils via chemokines and inflammatory cytokines (3). Detailed information regarding the regulation of adaptive immune responses by IL-17A remains scarce. In this study, we demonstrate that IL-17A, effectively induced after L. monocytogenes infection, could regulate both innate and adaptive immune responses against systemic L. monocytogenes infection. IL-17A modulates adaptive T cell responses by indirectly promoting optimal CD8⁺ T cell proliferation. It is intriguing to find that the CD4⁺ T cell responses were largely unchanged if any in the absence of IL-17A. Additionally, IL-17A was excluded to interfere with CD4⁺ regulatory T cells. These results strongly implicate that IL-17A could selectively and indirectly promote cross-priming of CD8⁺ T cells.

More recently, Martin-Orozco et al. (40) suggested a protective and therapeutic role for Th17 cells in antitumor immunity by enhanced CTL responses. However, how Th17 cells promote CTL response is still not fully understood. IL-17A recruits and activates neutrophils to combat infection (41). A previous study indicated that neutrophils could also function as APCs in L. monocytogenes infection to cross-prime CD8⁺ T cells (42). We found that neutrophil recruitment was not defective, but only delayed in Il17a^−/− mice against virulent L. monocytogenes infection, which may partially explain the early protective function of IL-17A (16, 43). However, we failed to find significant difference in neutrophil counts between WT and Il17a^−/− mice infected with an attenuated strain of L. monocytogenes. Immunization with OVA and CFA, in which neutrophils are not involved in Ag capture and presentation, also confirmed defective CD8⁺ T cell responses in Il17a^−/− mice. Thus, the promoting function of IL-17A in CTL response could not be attributed to neutrophils. Most recently, IL-17A was shown to promote Th1 responses in Francisella tularensis infection (14). However, in our study, CD4 Th1 cells showed no significant difference during listeriosis.

DCs are the most efficient APCs in priming both CD4⁺ and CD8⁺ T cells (44, 45). After maturation, DCs could upregulate the expression of IL-17RA and respond to IL-17A to produce more cytokines (46). In our experiments, DCs stimulated with IL-17A could promote the proliferation and activation of naive OT-I T cells through the upregulation of MHC-I molecule H2K^b and enhanced secretion of cytokines (IL-12, IL-6, and IL-1β). IL-17F, coexpressed with IL-17A in most cases, showed no stimulatory effect on DCs, as murine IL-17F cannot bind IL-17RA (23). This reveals that IL-17A promotes CTL proliferation via enhancing DC cross-presentation in vitro. CD8α⁺ DCs are responsible for the cross-priming of CD8⁺ T cells in vivo (35, 47) and control the entry of L. monocytogenes into the spleen (48). Indeed, CD8α⁺ DCs from Il17a^−/− mice produce less IL-12 and are also less potent to activate naive OT-I T cells.

Th17 cells are believed to produce the predominant IL-17A in vivo. But other cells, especially γδT cells, could also produce IL-17A under certain environments (15, 23). We found that a large amount of γδT cells start to accumulate in the lymph organs very shortly after L. monocytogenes infection and begin to produce IL-17A simultaneously. In contrast, Th17 cells emerge slowly after 3 d and are less abundant. Additionally, in vivo depletion of γδT cells greatly reduced serum IL-17A production, whereas depletion of CD4⁺ T cells had no effect, suggesting the major role of γδT cells in the production of IL-17A in vivo. Only adoptive transfer of Il17a^+/+ γδT cells could recover the optimal CD8⁺ T cell responses in WT mice depleted of autologous γδT cells, further confirming our hypothesis that IL-17A produced from γδT cells promote optimal CD8⁺ T cell response.

Importantly, these IL-17–producing γδT cells are a subset of γδT cells that do not coincide with IFN-γ–secreting γδT cells. IFN-γ deficiency does not impair, and to some extent even enhances, the ability of γδT cells to produce IL-17A. Others (46, 49) also recently characterized these IL-17A–producing γδT cells as CCR6 positive and as expressing retinoic acid-related orphan receptor γt, a specific transcriptional factor for Th17 (50). They also secrete Th17-associated cytokines IL-17F and IL-21. These data and ours indicated these cells are a unique subset of γδT cells that much resemble Th17 cells in CD4⁺ T cells.

How these γδT cells are activated to secrete IL-17A still remains unanswered. Previous studies suggested that, by γδTCR receptors, γδT cells could expand and release cytokines after recognizing certain nonpeptide compounds from a spectrum of microbial pathogens, without requirement for Ag procession and presentation (20). A recent study demonstrated that triggered through TCR, naive γδT cells make IL-17A, but Ag-experienced cells make IFN-γ (51). Furthermore, γδT cells are shown to express certain TLRs like TLR2 (49) and thus may be quickly activated by ligands from L. monocytogenes lipoproteins. Also, IL-17A could be induced from γδT cells directly by IL-1 and IL-23 derived from activated DCs (46). Further experiments will be required to identify the actual mechanisms for the generation of IL-17A–producing γδT cells after L. monocytogenes infection.

Depletion of γδT cells leads to defective CD8⁺ effector T cells, but depletion of CD4⁺ T cells does not, which is consistent with previous studies showing a redundant role of CD4⁺ T cells in the induction of effector CD8⁺ T cells (52, 53). Nomura et al. (54) also reported that γδT cells participate in the priming of bacterial Ag-specific CD8⁺ cytotoxic T cells. In addition, adoptive transfer of Il17a^+/+ γδT cells but not CD4⁺ T cells partially recovered CD8⁺ T cell priming in Il17a^−/− mice, suggesting IL-17 produced by γδT cells alone is sufficient for optimal CD8⁺ T cell priming. So, we propose that early production of IL-17A by innate γδT cells contributes to the adaptive immune response that clears the invading pathogens, whereas uncontrolled IL-17A production by Th17 cells at late stages of infection may risk exaggerated immune responses (55) and autoimmune diseases (56).

In summary, our data provide insight into the mechanism by which IL-17A helps host defense against intracellular bacterial infections, not only by expansion and accumulation of innate neutrophils, but also by promoting adaptive CTL responses by enhancing DCs cross-presentation. As IL-17A and IL-17A–producing γδT cells have shown a beneficial effect in our current study against intracellular L. monocytogenes infection, it is possible to explore IL-17A or IL-17A–producing γδT cells in vaccination against both infectious diseases and tumors.

We thank Dr. Chen Dong for kindly providing IL-17A–deficient mice, Dr. Hao Shen for kindly providing LM-OVA and ΔactA^−/− LM-OVA, and Jianqiu Long and Xiaoting Zuo for technical assistance.

Disclosures The authors have no financial conflicts of interest.