IL-17A Produced by Neutrophils Protects against Pneumonic Plague through Orchestrating IFN-γ–Activated Macrophage Programming Free

Yersinia is a genus that comprises three human pathogen species. Y. enterocolitica and Y. pseudotuberculosis are enteropathogens that cause gastrointestinal diseases, and Y. pestis is the causative agent of plague (1, 2). In general, plague is manifested as either a bubonic or a pneumonic form. Bubonic plague occurs when Y. pestis is inoculated in the host by the bite of an infected flea, and this process is considered one of the most common presentations of the disease (3). When inhaled, Y. pestis causes primary pneumonic plague, a contagious disease that can be spread via respiratory droplets (1). The primary pneumonic plague is a rapidly progressive and aggressive pneumonia that usually develops in 2–3 d after exposure, with mortality rates approaching 100% if effective treatment is delayed (4, 5). However, little is known about innate host responses in the pathogenesis of the pneumonic plague at an early stage.

IL-17 is an important cytokine mainly produced by Th17, which has a major function in autoimmunity that results from a dysfunction of adaptive immunity (6–8). Although most data have suggested that the Th17/IL-17 pathway’s functions in mediating chronic inflammatory diseases, such as rheumatoid arthritis (8, 9), asthma (10, 11), systemic lupus erythematosus (6), multiple sclerosis (12), and allograft rejection (13), other findings have indicated that such a pathway may participate in innate immunity (14). Acute bacterial inflammation usually recruits neutrophils with a subsequent activation of macrophages, thereby providing protection against foreign bacterial infection. However, the mechanism involved in this activation remains unclear.

In pneumonic plague, a model of acute bacterial infection, IL-17 pathways appear to be activated (1). Infected lung injury causes production of IL-17–promoting cytokines, such as TGF-β, IL-6, TNF-α, and IL-1β, in the inflamed lung (1). In addition, Th1, Th17, and neutrophils are important in the pathogenesis of autoimmune inflammation (15–17). Recently, some studies have examined the function of IL-17 in innate immunity, including in infected lungs (18, 19). Our study demonstrated that IL-17 pathways are activated in pneumonic plague. We also showed that IL-17A produced by neutrophils is required for IFN-γ–activated classical M1 proinflammatory macrophage programming after Y. pestis challenge.

IL-17A^−/−mice (20, 21) (B6 background) and IL-17RA^−/− mice (22, 23) (B6 background) were obtained from The Jackson Laboratory (Bar Harbor, ME). IFN-γ^−/− mice (24) (B6 background) were obtained from the Center of Model Animal Research at Nanjing University (Nanjing, China). Rag1^−/− mice, CD45.1⁺ mice, and CD45.2⁺ C57BL/6 mice were obtained from Beijing University Experimental Animal Center (Beijing, China). IL-17A^−/−Rag1^−/− double knockout (KO) mice were generated by crossing Rag1^−/− and IL-17A^−/− mice. All of these mice were bred and maintained under specific pathogen–free conditions. Sex-matched littermate mice aged 6–8 wk were mainly used for the experiments. All of the animal experiments were performed in accordance with the approval of the Animal Ethics Committee of the Beijing Institute of Microbiology and Epidemiology, Beijing, China.

Y. pestis strain 91001 and strain 141 were used in this study. Strain 91001 belongs to a newly established Y. pestis biovar, Microtus (25), which is considered avirulent in humans but highly lethal in mice (26). Strain 141 was isolated from Marmota himalayana in the Qinghai-Tibet plateau and is highly virulent to both humans and Rhesus macaques (27–29). Y. pestis bacilli were grown overnight at 26°C with continuous shaking in Bacto heart infusion broth. After dilution to an OD of 0.1 at 620 nm, they were continuously grown for 3–4 h at 26°C, and then collected by centrifugation, washed with saline, and quantified by OD measurement. The number of bacteria in the inoculating dose was confirmed by plating. The LD₅₀ of strain 91001 and strain 141 via the intranasal route is ∼ 6 × 10³ CFU and 1.6 × 10⁴ CFU, respectively. Mice were infected intranasally with 0.5 LD₅₀ to generate survival curves, and 100 LD₅₀ was permitted to assess lung pathological changes and other parameters following Y. pestis strain challenge.

The cells were stained with Abs in PBS containing 0.1% (wt/vol) BSA and 0.1% NaN₃ to analyze surface markers by flow cytometry. The following Abs were purchased from different manufacturers: anti-CD11b (M1/70), anti-F4/80 (BM8), anti–Gr-1 (RB6-8C5), anti-CD45.1 (A20), and anti-CD45.2 (104; eBioscience, San Diego, CA); anti-CD11b (M1/70), anti-CD45 (TU116), and anti-CD11c (HL3; BD Biosciences, San Diego, CA); anti-CD45 (30-F11; BioLegend, San Diego, CA); anti-CXCR2 (clone 242216; R&D Systems, Minneapolis, MN); and anti-CD3 (145-2C11), anti-CD19 (6D5), and anti–Gr-1 (RB6-8C5; Miltenyi Biotec, Bergisch Gladbach, Germany). Anti–IL-17A mAb (clone 50104, R&D Systems; IgG2a) was used to neutralize IL-17A in vivo.

Bronchoalveolar lavage fluid (BALF) was collected and harvested as previously described (17). In brief, the trachea was exposed by producing a midline incision and then cannulated with a sterile 22-gauge Abbocath-T catheter. The bilateral BALF was collected after two 0.5-ml aliquots of sterile saline were instilled. Approximately 0.9–1.0 ml of BALF was retrieved from each mouse. To prepare single-cell suspensions from lung tissues, the lungs were perfused with saline containing heparin, minced, and digested with collagenase and DNase I.

Bone marrow was obtained as described before (16). Isolation of neutrophils from bone marrow cells was accomplished according to our previously published procedures (16, 30). Bone marrow CD11b⁺ cells were isolated using anti-CD11b magnetic beads and positive selection columns (Miltenyi Biotec). Ly6G⁺ cells were isolated using anti-Ly6G–PE mAb (1A8; eBioscience) and positive immunomagnetic separation using a selection kit (Stem Cell Technologies, Vancouver, BC, Canada) or CD11b⁺Ly6G⁺F4/80^- neutrophils sorted on a FACSAria II (Becton Dickinson, Franklin Lakes, NJ). Flow cytometry verified that all isolated cell purity and viability was > 95%.

On the indicated days after challenge infection, BALF, lungs, and livers were harvested and homogenized in saline. Serial dilutions of homogenates were plated on Hottinger agar plates and incubated at 26°C for 48 h. Bacterial CFUs were then counted.

Intracellular IL-17A (eBio17B7; eBioscience) and IL-17F (eBio18F10; eBioscience) were analyzed by flow cytometry according to the manufacturer’s instructions. Flow cytometry data were acquired on an FACSCalibur (Becton Dickinson), and these data were analyzed with FlowJo (TreeStar, San Carlos, CA). The number of cells of various populations was calculated by multiplying the total number of cells by the percentages of each individual population from the same mouse, and the result was averaged.

RNA was extracted using an RNeasy kit (QIAGEN, Germany), and cDNA was synthesized using SuperScript III reverse transcriptase (Invitrogen, Carlsbad, CA). A LightCycler 480 (Roche, Switzerland) real-time PCR system and primer and probe sets from Applied Biosystems were used for quantitative PCR. The PCR primer sequences used in this study are presented in Table I. To determine the relative expression of cytokine mRNA in response to various conditions, the mRNA expression level of each gene was normalized to the expression level of GAPDH, a housekeeping gene.

Mouse BALF IL-17A levels were calculated from a standard curve obtained from ELISA analysis of recombinant IL-17A (Bender MedSystems, Burlingame, CA).

Neutrophils (2 × 10⁶ cells per milliliter to 3 × 10⁶ cells per milliliter) were plated on wells in the presence or absence of 100 ng/ml TNF-α, as described previously, with slight modifications (31). The neutrophils were incubated for 1 h at 37°C; afterward, the cells were centrifuged and the supernatants were removed to determine the release of elastase and gelatinase. Elastase and gelatinase were determined to analyze enzyme release in the supernatant after the neutrophils were degranulated using an EnzChek Elastase Assay Kit and a Gelatinase Assay Kit (Invitrogen) according to the manufacturer’s instructions.

A total of 5 × 10⁶ donor CD45.2⁺ bone marrow cells from IL-17A^−/− or CD45.1 WT littermates were transplanted i.v. in 8-wk-old CD45.1⁺ mice that were lethally irradiated with 11 Gy to set up the complete chimeras, as described previously (32, 33).

Lung neutrophils and macrophages were sorted as described previously (34, 35). Y. pestis strains (Y. pestis–GFP) were grown overnight at 26°C. These strains were subsequently washed in PBS and counted. The Y. pestis containing a bacterial concentration of 10⁶ CFU was incubated with 1 × 10⁵ neutrophils in flat-bottom 96-well plates and in 200 μl RPMI medium at 37°C in 5% CO₂ for 3 h. For phagocytosis experiments, neutrophils or macrophages in some wells were collected at 30 min postinfection. The neutrophils or macrophages were blocked with an anti-mouse FcγR mAb (clone 2.4G2) and stained with anti-Ly6G–PE or anti-F4/80–PE. After washing with cold PBS three times, the phagocytosis percentages of the gated Ly6G⁺ cells or F4/80⁺ cells were then determined by an FACS scan. The surviving Y. pestis CFUs were determined as previously described (36), and the survival percentage was calculated.

The Clophosome (FormuMax, Palo Alto, CA) was used for macrophage deletion, as described previously (37). The dose was 0.15 ml for 20 g animal body weight via the i.v. route before infection. The empty liposome (Lipo) product was used as a negative control.

The respiratory burst was determined as previously described (38). Neutrophils isolated from the bone marrow were incubated in the presence of 1 μM dihydrorhodamine (Molecular Probes, Life Technologies, Grand Island, NY) during stimulation with PMA (Sigma-Aldrich, St. Louis, MO). The samples were incubated at 37°C for 15 min before flow cytometry analysis. The neutrophils were identified by staining with CD11b and Ly6G Ab.

Equal volumes of the culture supernatant or serum (100 μl) with Griess reagent were incubated, and the absorbance at 550 nm was determined using a microplate reader (Bio-Rad), as described previously (39).

At 2 d postinfection, the entire lungs were excised and washed. The lungs were fixed in 4% paraformaldehyde, embedded in paraffin, cut in 5-μm sections, and stained with H&E for histopathological examination.

For immunohistochemistry, formalin-fixed paraffin-embedded tissues were cut in 4-μm sections, and slides were stained to determine the neutrophils based on an immunoperoxidase method using a rat anti-mouse Gr-1 mAb (BD Biosciences).

Data are presented as mean ± SD. Statistical analyses were performed using the GraphPad Prism 5. A log-rank test was used for all the survival data. The Student unpaired t test was used to compare the means between two groups. For multiple comparisons, one-way ANOVA was used as indicated. A p value < 0.05 was considered statistically significant.

We initially investigated the effects of IL-17A on the pneumonic plague. IL-17A KO and wild-type (WT) mice were challenged intranasally with Y. pestis strain 141 or 91001, respectively, to induce the classical pneumonic plague as reported previously (2). As shown in Fig. 1, IL-17A KO mice displayed symptoms of severe infectious inflammation, whereas WT mice showed a delayed onset and markedly milder course of disease (Fig. 1A–C and data not shown). This result was supported by a significantly lower percentage of survival, higher body weight loss, and higher ratio between the lung and body weight of the IL-17A KO mice. The WT littermates showed alleviation in the symptoms of infection. The number of Y. pestis after the challenge showed that the lung and liver had higher bacterial CFUs in the IL-17A KO mice than in the WT controls (Fig. 1D and data not shown). Macroscopic and histological observation revealed a much more severe pathological inflammation in the lungs of IL-17A KO mice (Fig. 1E). These results indicated the possible role of IL-17A signals in inducing acute pneumonic plague inflammation in mice.

IL-17A mRNA and IL-17 protein levels of the WT mice gradually increased at different time points postinfection and peaked at 24–48 h postinfection with Y. pestis strain 141 or 91001, but their levels subsequently began to decrease (Fig. 1F and data not shown). This result is consistent with the relatively milder symptoms of infection in the WT mice compared with those in the IL-17A KO mice (Fig. 1A–E), and further indicated that IL-17A may play a protective role in the early stage of the pneumonic plague. Moreover, the IL-17A KO mouse, challenged by Y. pestis strain 141 or 91001, showed a tendency similar to that of the WT mouse with pneumonic plague. Because strain 91001 is less virulent to humans, and is convenient to use, this strain was part of the following study. We hypothesized that acute bacterial infection induces IL-17A production via IL-17R, its receptor, to provide protection against lung injury in mice with pneumonic plague. The exaggerated effects were confirmed by more severe lung disease and higher numbers of bacteria in the lungs of IL-17A^−/− and IL-17RA^−/− mice, compared with in the lungs of WT mice, which demonstrated that the IL-17 signaling pathway was involved in the early stage of the pneumonic plague in mice (Fig. 1E, 1G). The exaggerated effects were also observed in WT mice pretreated with neutralization of IL-17A with anti–IL-17A mAb, but the protective effects were observed in WT mice pretreated with IL-17A (Fig. 1G).

We performed a series of experiments to identify the cells that produce IL-17A in lungs infected with Y. pestis. At 24 h postinfection, CD11b⁺Ly6G⁺ neutrophils and F4/80⁺ macrophage infiltration were significantly increased over what was found in uninfected groups, as revealed by immunohistological staining (Supplemental Fig. 1A) and flow cytometry (Supplemental Fig. 1B–D). However, we could not find any significant difference in the absolute number of infiltrating immune cells (CD11b⁺Ly6G⁺ neutrophils, F4/80⁺ macrophages, CD3⁺ T cells, and CD19⁺ B cells) between WT and IL-17A KO mice challenged with Y. pestis (Supplemental Fig. 1). Importantly, we also found that the IL-17A–producing CD45⁺ cells were gradually increased and peaked at 24 to 48 h postinfection, but these cells subsequently began to decrease (Fig. 2A). However, the same result was not observed in IL-17F (data not shown). IL-17AKO→WT and WT→WT complete bone marrow chimeras were generated to examine the contribution of bone marrow–derived cells as the source of IL-17A in pneumonic plague. A higher percentage of deaths significantly correlated with higher bacterial colony numbers in the lungs and more severe pathological damage in the IL-17AKO→WT mice, which lacked IL-17A in bone marrow cells. The same result was not observed in WT→WT mice (Fig. 2B–D). Only IL-17AKO→IL-17AKO mice showed more severe pathological changes and weight loss compared with WT→IL-17A KO mice (Fig. 2D and data not shown) in the complete bone marrow chimeras. Therefore, IL-17A produced by bone marrow–derived cells contributed to protection against pneumonic plague in mice.

We then generated IL-17A-Rag1DKO→Rag1KO and Rag1KO→Rag1KO bone marrow chimeras. IL-17A-Rag1DKO→Rag1KO mice revealed a significantly higher percentage of deaths and bacterial colony numbers compared with the Rag1KO→Rag1KO mice (Fig. 2E, 2F). Therefore, IL-17A produced from non-T and non-B bone marrow–derived cells produced protection against pneumonic plague in mice. T and B cell–deficient Rag1 KO mice were pretreated with anti–IL-17A before being challenged with Y. pestis. We found that these mice exhibited severe pathological inflammatory damage (Supplemental Fig. 2A), lower percentage of survival (Supplemental Fig. 2B), and higher bacterial colony number (Supplemental Fig. 2C), similar to WT mice pretreated with anti–IL-17A mAb. These in vivo studies suggested that cells other than T, B, NKT, and γδT cells produced IL-17A at an early stage of pneumonic plague in mice.

The FACS results showed that IL-17A was secreted from the recruited neutrophils (Fig. 3A, 3B), but not from T or NKT cells (data not shown). A significant increase in IL-17A–producing neutrophils was found at 48 h postinfection in the lungs (Fig. 3C). To demonstrate the functional significance of IL-17A production by neutrophils in pneumonic plague, we transferred the neutrophils from the bone marrow of WT (WT Neu→IL-17A KO) or IL-17A KO mice (IL-17A KO Neu→IL-17A KO) to the recipient IL-17A KO mice. The percentage of survival, pathological damage, and number of bacterial colonies were reconstituted in the WT Neu→IL-17A KO mice, but not in the IL-17A KO Neu→IL-17A KO mice (Fig. 3D–F). Furthermore, the reconstituted protection in the WT Neu→IL-17A KO mice was blocked with 100 μg of anti-mouse IL-17A mAb (Fig. 3D–F). These data demonstrated that neutrophils, and not T, NKT, γδT, or B cells, are the major source of IL-17A production that protects against pneumonic plague in mice.

Proinflammatory cytokines mediate bacterial lung inflammation. For this reason, we investigated the time course changes of the cytokine expression in lung samples from the WT mice by real-time PCR (Table I). Postinfection, IL-6, TNF-α, IL-1β, CXCL1, and CXCL2 gradually increased and peaked at 24–48 h, but these cytokines subsequently started to decrease (Fig. 4A). Although IL-6 and TNF-α were decreased in pneumonic plague in IL-17A KO and IL-17R KO mice, compared with WT mice, IL-17 signaling could not significantly affect neutrophil bactericidal activities (Fig. 4B). Reactive oxygen metabolites are generated after neutrophil activation and are important to eradicate microorganisms. Bone marrow neutrophils from IL-17A–deficient mice challenged with Y. pestis showed a similar level of oxidative burst and peaked 20 min after these neutrophils were stimulated with PMA, compared with those from the WT controls (Fig. 4C). The mRNA expression of granule proteins such as elastase and gelatinase did not significantly differ in the neutrophils of the IL-17A KO mice compared with those of the WT mice (Fig. 4D). These results indicated that IL-17A is unlikely to be necessary in regulating neutrophil function to protect against pneumonic plague. Furthermore, we did not observe significant differences between IL-17A KO and WT mice in the abilities of neutrophils to phagocytose and eradicate Y. pestis in vitro. At 1 h postinfection, a large number of Y. pestis were phagocytosed by neutrophils, but no significant differences in the phagocytosis percentages of Y. pestis were observed between IL-17A KO and WT mice (Fig. 4E). Moreover, no significant differences in the release of gelatinase and elastase in the neutrophils were observed between IL-17A KO and WT mice (Fig. 4F). The survival percentage of Y. pestis in the IL-17A KO mice could be comparable to that of the WT controls (Fig. 4G). All of these results indicated that neutrophil-derived IL-17A could not affect neutrophil bactericidal activities in protecting against pneumonic plague in mice.

IFN-γ is an important cytokine that stimulates classically activated macrophages (M1), which figure greatly in providing protection against pneumonic plague infection (40, 41). However, the macrophage-mediated immune response might be affected by IL-17 signaling. To test this hypothesis, we observed the inflammatory effects of pneumonic plague in IFN-γ signaling deficiency. Bacterial colony number (Fig. 5A) and pathological damage (data not shown) were significantly aggravated at 24 h postinfection in mice treated with a neutralized anti–IFN-γ Ab, compared with WT mice treated with an IgG1 isotype control. This result was not observed in IL-17A KO mice. The increased number of bacterial colonies and the aggravated pathological damage were also observed in IFN-γ–deficient mice (data not shown). These results further suggested that the IFN-γ pathway is involved in IL-17A–mediated protective effects against pneumonic plague. However, how does IFN-γ mediate this protective effect? Macrophages have a major role in providing protection against pneumonic plague infection in early stages, and this protective role of macrophages usually requires IFN-γ signaling (42, 43). Thus, macrophages probably participated in providing protective effects against pneumonic plague in mice.

To verify this hypothesis, we performed a macrophage-depletion experiment in vivo (Fig. 5B, 5C). The number of bacterial colonies and the pathological damage were significantly aggravated at 48 h postinfection in the WT mice with depleted macrophages (treated with clodronate [CL] and Lipo), but this result was not observed in IL-17A KO and IFN-γ KO mice (Fig. 5D and data not shown). This result revealed that IL-17A– and IFNγ–mediated protection against pneumonic plague was related to the macrophage population in vivo. Our results are consistent with the possibility that IL-17A– and IFN-γ–activated macrophages mediated protection against pneumonic plague in mice, suggesting that these pathways are very important in macrophage-mediated protection against Y. pestis infection. Thus, IFN-γ and IL-17A signaling contributed to this process.

We treated the infected WT and IL-17A KO mice with murine IFN-γ to detect the interaction between the macrophages with IFN-γ and IL-17A pathways involved in providing protection against pneumonic plague. The IL-17A KO mice received recombinant murine (rm)IFN-γ (3000 U per mouse) 1 h prior to the bacterial challenge. The rmIFN-γ could almost completely provide protection against pneumonic plague in IL-17AKO mice, as observed in the changes in bacterial CFU, but not in the mice treated with CL and Lipo (Fig. 5E). These results are consistent with the possibility that IL-17A protected the mice against pneumonic plague via IFN-γ–activated M1 macrophages. The aggravated number of bacterial colonies, percentage of survival, and pathological changes could be significantly ameliorated in mice when M1 bone marrow–derived macrophages activated by LPS and IFN-γ were transferred to IL-17A KO mice before being subjected to Y. pestis challenge (Fig. 5G, Supplemental Fig. 3).

We analyzed the production of NO as well as arginase and inducible NO synthase (iNOS) expressions during the programming of macrophages in vitro to investigate the function of IL-17A/IL-17R signaling in programming M1 macrophages. As expected, NO production (Fig. 6A), as well as iNOSmRNA (Fig. 6B) and protein (Fig. 6C) expression, was significantly higher in the group treated with LPS + IFN-γ + IL-17 compared with the group treated with LPS or LPS + IFN-γ alone (p < 0.001). No significant differences were found between arginase protein (Fig. 6D) and mRNA (Fig. 6E) expression in the combined treatment group (IL-17, IFN-γ, and LPS) compared with LPS alone or LPS + IFN-γ–treated groups, as detected by real-time PCR and ELISA.

Bactericidal capacity is the gold standard used to evaluate macrophage function. We observed an increase in phagocytosis of the macrophages treated with IFN-γ and IL-17A. Y. pestis was significantly reduced in in vitro coculture systems, compared with IFN-γ or IL-17A treatment alone. Macrophages were treated with IL-17A or IFN-γ or the two together or none for 2 h before infection. More Y. pestis were phagocytized in the cytokine combined group than in that treated with IFN-γ or IL-17A alone (Fig. 6F). The survival percentage of Y. pestis in macrophages treated with IFN-γ + IL-17A was lower than that in macrophages treated with IFN-γ or IL-17A alone (Fig. 6G). These results revealed that IL-17A signaling contributed to IFN-γ–activated M1 macrophage protection against pneumonic plague.

The bactericidal capacity of the macrophages in vivo was investigated. Y. pestis–GFP (1 × 10⁸ CFU) was injected in the peritoneal cavity of mice. The phagocytosis percentage and the survival of Y. pestis in macrophages were investigated as described in 2Materials and Methods. We observed the decreased ability of IL-17A KO macrophages to phagocytose and eradicate Y. pestis in the in vivo system compared with that of the WT mice. At 12 h postinfection, IL-17A KO macrophages phagocytosed Y. pestis to a lesser extent than did those in the WT control mice (Supplemental Fig. 4A, 4B). The survival percentage of Y. pestis was significantly increased in IL-17A KO mice compared with WT control mice. Thus, IL-17A produced by neutrophils protected the mice against pneumonic plague in an IFN-γ–activated macrophage-dependent manner (Fig. 7).

Plague is one of the deadliest infectious diseases (1, 41). The causative agent, Y. pestis, is a Gram-negative facultative bacterium that is naturally transmitted from rodents to humans by fleas (2). Y. pestis bacilli typically infect the nearest skin-draining lymph nodes, which swell to produce diagnostic buboes upon transmission by a flea bite (44). The bubonic form of plague often leads to sepsis and occasionally progresses to secondary pneumonic infection. Pneumonic plague is almost always lethal in humans (5). Moreover, pneumonic plague can spread from person to person via infectious respiratory droplets (5). Substantial concerns are increasing because Y. pestis could be exploited as a biological weapon and extensively antibiotic-resistant Y. pestis strains are currently emerging (45). The question remains whether some natural intrinsic protective mechanisms are found in the pathological course of pneumonic plague. Numerous efforts have been devoted to the development of plague vaccines, such as subunit vaccines containing Y. pestis F1 and LcrV proteins that were recently used in human clinical trials (46, 47). In the current study, IL-17A was considered a natural protective cytokine that critically provides protection against early pneumonic plague infection by linking the reciprocal sequencing modulation among neutrophils, macrophage activation, and programming.

IL-17 has become a target molecule since its discovery, particularly after the Th17 lineage has been discovered and has become one of the most studied cytokines in immunology (48). Th17 cells account for only a fraction of IL-17 produced in vivo. IL-17 is mostly derived from innate lymphoid populations of neutrophils, monocytes, NK cells, and lymphoid tissue inducer-like cells and can rapidly produce IL-17A and IL-17F, particularly at mucosal sites (48–50). Emerging evidence has indicated that IL-17 is mainly protective at the lung surface, where IL-17 has a nonredundant function in controlling infection (18). Mice defective in IL-17 signaling are highly susceptible to infection by Klebsiella pneumoniae, whereas IL-4, IL-12, and IFN-γ are not necessary in the protection against such an infection (51, 52). IL-17 also controls responses to fungal pathogens because mice that lack IL-23 and receptors such as IL-17RA and IL-17RC are susceptible to infection with Candida albicans (53). Recent studies (54) showed that prime vaccination with D27-pLpxL confers better protection than prime-only vaccination, and the prime also increases pulmonary numbers of Th17. This finding suggests that IL-17 contributes to cell-mediated defense against pulmonary Y. pestis infection. However, the exact innate immediate protective mechanism, especially during early acute infectious disease, remains unclear. In our study, we initially clarified that IL-17 also exhibited a protective role in infectious disease. We demonstrated that the deficiency in IL-17A or IL-17R and mAb neutralization of IL-17A aggravated the acute pneumonic plague infection. IL-17A produced by Ly6G⁺ neutrophils, not by NKT or γδT cells, was important in providing protection against early pneumonic plague infection in mice (Figs. 2 and 3 and data not shown). This result confirmed that IL-17A, a natural cytokine produced by neutrophils, provided protection in early acute pneumonic plague infection.

Furthermore, our results showed that IL-17 could not significantly change neutrophil bactericidal activities (Fig. 4D–G), although IL-17A deficiency could significantly decrease the production of IL-6, TNF-α, and IL-1β (Fig. 4A, 4B). The results further revealed that the neutrophil-derived IL-17A could not affect neutrophil bactericidal activities that provide protection against pneumonic plague in mice.

Therapeutic administration of the immune serum from convalescent mice protects naive WT mice against lethal pulmonary Y. pestis challenge (55). This immune serum also poorly protects gene-targeted mice that lack the capacity to produce TNF-α or IFN-γ, suggesting that TNF-α and IFN-γ contribute to serum therapy–mediated protection against pneumonic plague. However, acute regulatory mechanisms by which TNF-α and IFN-γ contribute to protect the cells against pneumonic plague remains unclear. In this study, gene-deficient mice or a macrophage-depleted system demonstrated that IFN-γ is necessary for macrophage-activated protection against pneumonic plague, using a neutralized mAb (Fig. 5A–F). Macrophages are also a major producer of TNF-α during early pneumonic plague infection (data not shown). Thus, these results suggested that IFN-γ–activated M1 macrophage programming is critically involved in the removal of an established Y. pestis infection.

Neutrophils immediately respond as innate immune cells during infection, along with macrophages and other lymphocytes that are recruited to elicit coordinated effects and fight against infection. However, the mechanism by which neutrophils and macrophages are coordinated remains unclear. In the current study, IL-17A critically targeted M1 macrophage programming (Fig. 6A–E) and provided protection against bacterial infection in an IFN-γ–dependent manner (Fig. 6F, 6G), although IL-17A was unable to contribute to neutrophil bactericidal activity. We initially reported that IL-17A signaling was connected to neutrophil activity in providing protection against the pneumonic plague infections via bactericidal functions of M1 macrophages. Thus, IL-17 is an intrinsic regulator that coordinates neutrophils and macrophages, which are two important innate immune cells, in the innate immune defense against bacterial infectious diseases.

Little is known regarding the coordinating effects of the IL-17 pathway in the two important innate immune cells (neutrophils and macrophages), although IL-17 pathway functions in mediating autoimmune diseases have been described. Proinflammatory cytokines and chemokines are produced by airway epithelial and endothelial cells to promote the recruitment and activation of neutrophils and to secrete IL-17A in response to plague. IL-17 signaling subsequently occurs and strengthens IFN-γ–activated M1 macrophage programming and function (Fig. 7). Thus, our study demonstrated that IL-17A coordinated neutrophils and macrophages, which provide protection against early pneumonic plague infection.

We thank Dr. Hui Yang for review of the manuscript and Zhizhen Qi for help in animal observation.

The authors have no financial conflicts of interests.