Lung CD4−CD8− Double-Negative T Cells Are Prominent Producers of IL-17A and IFN-γ during Primary Respiratory Murine Infection with Francisella tularensis Live Vaccine Strain

Siobhán C. Cowley, Anda I. Meierovics, Jeffrey A. Frelinger, Yoichiro Iwakura and Karen L. Elkins
J Immunol May 15, 2010, 184 (10) 5791-5801; DOI: https://doi.org/10.4049/jimmunol.1000362

Abstract

For several intracellular infections, pulmonary vaccination provides measurably better protection against pulmonary challenge. The unique factors that contribute to pulmonary immune responses are not well characterized. In this study, we show that CD4−CD8− double negative (DN) T cells are a major responding T cell subset in the lungs of mice during pulmonary Francisella tularensis live vaccine strain (LVS) infection. DN T cells were a minor (<2%) subset in spleens and lungs of mice during sublethal intradermal infection with LVS. In contrast, they were a major responding T cell subset in lungs during pulmonary LVS infection, producing large quantities of IFN-γ and IL-17A. The numbers of IL-17A+ DN T cells in the lungs exceeded that of CD4+ and CD8+ T cells on day 7 postinfection; by day 14 postinfection, all three IL-17A–producing T cell subsets were present in equivalent numbers. CD4+, CD8+, and DN T cell production of IL-17A was not observed in the spleens of pulmonary-infected mice or the lungs and spleens of intradermally infected mice. Correspondingly, IL-17A knockout mice were more susceptible to respiratory than intradermal LVS infection, with delayed clearance 1–3 wk postinfection. Finally, in vitro treatment of LVS-infected macrophages and alveolar type II epithelial cells with IFN-γ and IL-17A affected significantly greater LVS growth control than treatment with either cytokine alone. The data presented in this study demonstrate that DN cells contribute to production of IL-17A and IFN-γ in the lungs during inhalational Francisella infection and that these cytokines additively activate host cells to control LVS intracellular growth.

Improved vaccines against respiratory infections such as Mycobacterium tuberculosis, which is a major public health problem, and potential airborne bioterrorism pathogens such as Francisella tularensis are sorely needed. In these cases and others, respiratory vaccination appears to provide superior protection against pulmonary exposure than more traditional parenteral vaccination (1). Although compartmentalization of mucosal immune responses is well recognized, the mechanisms that underlie these differences remain incompletely understood. Such understanding is important not only to evaluate and manage potential vaccination risks, but to improve vaccine efficacy and derive clinically meaningful correlates of protection.

To address such questions, we have taken advantage of a murine experimental model using the live vaccine strain (LVS) of the intracellular bacterium F. tularensis (2). The severity of human tularemia, caused by virulent F. tularensis, is highly dependent upon the entry route. Infection acquired through the skin results in a relatively mild disease with an untreated mortality rate of ∼5%, whereas inhalational tularemia has an abrupt onset that rapidly progresses to an acute lethal disease in 30–60% of untreated cases (3). LVS has shown potential as a protective tularemia vaccine in animal studies and is currently an investigational product in the United States (4). Optimal vaccination can be route-dependent: intranasal LVS vaccination is more effective than vaccination via dermal routes at reducing symptoms and/or mortality against virulent Francisella pulmonary challenge in humans, monkeys, guinea pigs, and mice (5, 6). Thus, the requirements for induction of pulmonary immunity are clearly different from those that combat infections acquired at other sites, but the unique components at work are unknown.

Although LVS is avirulent for humans, the outcome of a primary LVS infection in inbred mice is also dependent on the route of inoculation: the LD50 for infections initiated via the intradermal (i.d.) route is ∼106 bacteria, but ∼103–104 bacteria for intranasal (i.n.) infections (2). Bacterial burdens in mice given a sublethal primary LVS i.d. or i.n. dose increase for ∼1 wk, then decline between weeks 1–3 and are cleared. Further, such LVS-vaccinated mice survive secondary challenge with virulent Francisella and with lethal LVS doses as high as 100,000 LD50 (2). Consequently, LVS murine infection is a convenient model in which to analyze the immune response to vaccination. Further, the immune response to Francisella appears to be similar to other more clinically important intracellular pathogens, such as M. tuberculosis (2, 7). Thus, studies to determine mechanisms of protection for tularemia may further our understanding of immunity to intracellular pathogens in general.

For intracellular bacteria, T cell-mediated immune responses are paramount for control of both primary infection and adaptive secondary responses. Considerable attention has been devoted to understanding effector mechanisms provided by traditional TCRα/β+, CD3+CD4+, and CD8+ T cell subsets. We are equally interested in a third TCRα/β+, CD3+ T cell subpopulation that lacks both CD4 and CD8, as well as NK-related markers, and thus has been dubbed double-negative (DN) T cells. Our previous studies demonstrated that all three distinct αβ T cell subsets expanded in vivo during primary aerosol M. tuberculosis and i.d. Francisella infections, acquired memory cell markers, controlled bacterial growth both in vitro and in vivo, and adoptively transferred resistance to secondary challenges (8, 9). However, in mice infected with LVS i.d., DN T cells were only a minor fraction of splenic T cell subset in vivo (∼1–2% of total splenocytes) and exhibited a unique in vivo phenotype: mice depleted of all T cells except the DN T cell subset developed a long-term, chronic LVS infection. Thus, unlike CD4+ and CD8+ T cells, DN T cells lacked, or suppressed, an important in vivo function required for clearance of the infection, and their primary in vivo role remained unclear.

In this study, we sought to explore the function of these unusual T cells further. We found that purified LVS-immune DN T cells required MHC class I, but not class II, receptors to control LVS intramacrophage growth in vitro. We further found that purified LVS-immune DN T cells expressed high levels of IFN-γ and IL-17A mRNA upon stimulation with LVS-infected macrophages. Next, we explored the capacity of these cells to produce IL-17A in vivo. During LVS i.n. infection, DN T cells were recruited to infected lungs in numbers that approached that of CD8+ T cells. Interestingly, pulmonary DN T cells exhibited a unique temporal cytokine production profile. Unlike CD8+ T cells, DN T cells produced large amounts of IFN-γ during the acute phase of respiratory LVS infection. Further, DN T cells produced significantly higher levels of IL-17A than CD4+ and CD8+ T cells on day 7 after LVS i.n. infection; however, by day 14 postinfection, all three T cell subsets were producing approximately equivalent amounts of IL-17A. CD4, CD8, and DN T cell IL-17A production was evident only in the lungs, but not the spleens, during i.n. infection and was undetectable in both lungs and spleens during i.d. infection. Correspondingly, IL-17A–deficient mice were more susceptible to i.n. than i.d. LVS infection, due to an inability to control bacterial numbers between 1–3 wk postexposure; however, the mice ultimately cleared the infection. Thus, IL-17A is essential during an intermediate phase of sublethal respiratory Francisella infection: after the innate acute inflammatory phase and prior to the final clearance phase. Further, we show that IL-17A and IFN-γ together augmented control of LVS intracellular growth in two types of host cells in vitro: macrophages and alveolar type II epithelial cells. Thus, the data presented in this study demonstrate that DN cells, in addition to CD4+ and CD8+ T cells, produced IL-17A and IFN-γ in the lungs during inhalational LVS infection and that one of the roles for IL-17A may be to act in concert with IFN-γ to activate LVS-infected host cells such as macrophages and alveolar type II epithelial cells (ATII) to control LVS intracellular growth.

Materials and Methods

Bacteria

F. tularensis LVS (ATCC 29684; American Type Culture Collection, Rockville, MD) was grown and frozen as previously described (10). Viable bacteria were quantified by plating serial dilutions on supplemented Mueller-Hinton agar plates.

Animals, infections, and determination of bacterial organ burdens

Male specific pathogen-free C57BL/6J mice, as well as IL-12p35 and p40-deficient mice, were purchased from The Jackson Laboratory (Bar Harbor, ME). IL-17A–deficient mice were obtained from Yoichiro Iwakura (University of Tokyo, Tokyo, Japan) and bred at the Center for Biologics Evaluation and Research/U.S. Food and Drug Administration (Bethesda, MD). Animals were housed in a barrier environment at the Center for Biologics Evaluation and Research/U.S. Food and Drug Administration and procedures performed according to approved protocols under Animal Care and Use Committee guidelines. Throughout, LVS-immune refers to splenocytes obtained from mice given a sublethal dose of 104 LVS i.d. 1–3 mo earlier. Intranasal infections were performed by delivering the indicated LVS CFUs in a volume of 20 μl per nare to anesthetized mice. All materials, including bacteria, were diluted in PBS (Cambrex, Walkersville, MD) containing <0.01 ng/ml endotoxin. Numbers of CFUs in organs from infected mice were determined as previously described (9), using groups of three to five mice. CFUs in lungs of infected mice were assessed by plating organ homogenates on MHA plates containing an antibiotic mixture to suppress growth of adventitious agents (colistin sulfate salt, lincomycin hydrochloride, trimethoprim, and ampicillin) (11).

In vitro assessment of control of intracellular bacterial growth in bone marrow-derived macrophages

The in vitro culture systems used and validation of the culture system’s abilities to reflect known parameters of T cell activities during in vivo control of bacterial growth have been described in detail elsewhere (8, 9, 12). Briefly, bone marrow macrophages were prepared from femurs of healthy mice by standard techniques using DMEM (Life Technologies, Grand Island, NY) supplemented with 10% heat-inactivated FCS (HyClone, Logan, Utah), 10% L-929–conditioned medium, 0.2 mM l-glutamine (Life Technologies), 1 mM HEPES buffer (Life Technologies), and 0.1 mM nonessential amino acids (Life Technologies) (complete DMEM [cDMEM]) and plated at 2 × 106 viable cells per well in 24-well plates. After 7 d, a confluent monolayer of bone marrow-derived macrophage (BMMΦ) was estimated to be 1 × 107 cells/well. Macrophages were then infected with F. tularensis LVS at a multiplicity of infection (MOI) of 1:20 (bacterium-to-BMMΦ ratio) for 2 h, washed, incubated with 50 μg/ml gentamicin for 45 min, and washed again, and spleen cells or separated subpopulations were added to the indicated wells. Cocultures were incubated at 37°C in 5% CO2 for the remainder of the experiment. Bacterial uptake or recovery was determined after initial infection and, at indicated time points through 3 d of coculture, by lysis of infected macrophages with sterile distilled water, plating on agar, and counting.

Spleens used for coculture and elsewhere were aseptically removed and prepared by standard techniques. In all cases, 5 × 106 splenocytes (or their separated subpopulations) were added to the 24-well cultures (∼1 splenocyte to 2 BMMΦs). Lungs from LVS-infected mice were prepared using pressure disruption, followed by incubation for 1 h at 37°C in 5% CO2 in PBS containing 2% FCS and 150 U/ml collagenase. Released cells were then filtered through a Filtra-bag (Labplas, Ste-Julie, Quebec, Canada) and subjected to ACK lysis.

In vitro assessment of control of intracellular bacterial growth in the murine macrophage cell line RAW 264.7 and the murine ATII cell line TC-1

The in vitro culture and LVS infection of TC-1 cells has been previously described (13). One day prior to infection, RAW and TC-1 cells were plated at 2 × 105 viable cells per well in 24-well plates in cDMEM. The next day, cells were infected with F. tularensis LVS at an MOI of 1:10 or 1:1 (bacterium-to-BMMΦ ratio) for RAW and TC-1 cells, respectively, for 2 h, washed, incubated with 50 μg/ml gentamicin for 45 min, and washed again, and cDMEM containing the relevant recombinant cytokines was added to the indicated wells. rIFN-γ (R&D Systems, Minneapolis, MN) and rIL-17A (eBioscience, San Diego, CA) were added at a concentration of 250 ng/well. Cultures were incubated at 37°C in 5% CO2 for the remainder of the experiment. Bacterial growth was assessed after 3 d of culture by lysis of infected cells with sterile distilled water, plating on agar, and counting.

In vivo depletion of cell subpopulations and enrichment of T cell subpopulations

Abs for in vivo depletion of CD4+ T cells (clone GK1.5), CD8+ T cells (clone 2.43), TCR γ/δ+ T cells (clone GL3), NK1.1+ cells (clone PK136), and Thy1.2+ cells (clone 30-H12) were either produced as ascites in BALB/c nu/nu mice and precipitated with 50% ammonium sulfate or obtained as purified Abs from the National Cell Culture Center (Minneapolis, MN); all contained <10 EU/ml endotoxin. Abs were administered to mice i.p. as previously described (8, 9, 14) and routinely reduced the depleted cell type to less than 0.5% detectable remaining cells (CD4+, CD8+, TCR γ/δ+, and NK1.1+) in spleens of treated mice as assessed by flow cytometry. DN T cells were enriched using a combination of in vivo depletion by Ab treatment, in vitro depletion using MACS beads, and positive selection via MACS beads (MACS magnetic cell sorting system; Miltenyi Biotec, Auburn, CA) as previously described (8, 9). Briefly, LVS-immune mice were depleted in vivo of CD4+, CD8+, NK1.1+, and TCR γ/δ+ cells, as above; harvested splenocytes were further depleted in vitro using B220+ MACS cell depletion columns, and remaining DN T cells were obtained using a Thy1.2+ MACS cell enrichment column. Enriched cell populations of CD4+ or CD8+ T cells were similarly prepared from the respective depleted mice followed by enrichment with the appropriate MACS columns. The purified cells were routinely >90% of the intended cell type; the identity of the other contaminating cells was always accounted for and consisted of macrophages, neutrophils, or granulocytes that do not have antibacterial activity on the BMMΦ monolayer in the in vitro system (12, 15).

Quantitation of cytokines and NO in BMMΦ culture supernatants

Culture supernatants were assayed for IFN-γ, IL-17A, IL-22, and TNF by standard sandwich ELISAs using reagents obtained from BD Pharmingen (IFN-γ and TNF; San Diego, CA), eBioscience (IL-17A), or Antigenix America (IL-22; Huntington Station, NY) and quantified by comparison with recombinant standards. NO was detected in culture supernatants by the Griess reaction as previously described (16), using commercial Griess reagent (Sigma-Aldrich, St. Louis, MO).

Flow cytometry analyses and intracellular cytokine staining

Cells were stained for a panel of murine cell surface markers and analyzed using a Becton-Dickinson LSR II flow cytometer (BD Biosciences, San Jose, CA) and FlowJo software (Tree Star, Ashland, OR) as previously described (9, 12, 17). Clones used included XMG-6 (anti–IFN-γ), RM4-5 (anti-CD4), 53-6.7 (anti-CD8a), GL3 (anti-γ/δ TCR), NK1.1 (anti-NK1.1), 53-2.1 (anti-Thy1.2), H57-597 (anti–TCR-β–chain), 17A2 (anti-CD3), M1/70 (anti-CD11b), RA3-6B2 (anti-CD45/B220), M1/70 (anti-CD11b), NK1.1 (anti-NK1.1), DX5 (anti-CD49b), and GL3 (anti-γ/δ TCR). All Abs above and Fc block were obtained from BD Pharmingen. Clone eBio17B7 (anti–IL-17A) was obtained from eBioscience. Live/Dead Violet stain was obtained from Molecular Probes (Carlsbad, CA) and included in all staining protocols. Optimal Ab concentrations were determined in separate experiments and appropriate fluorochrome-labeled isotype control Abs used throughout. In all cases, cells were first gated on singlets (forward light scatter width versus forward light scatter area or height) and live cells (Live/Dead Violet-negative) prior to further analyses. Spleen or lung cells that were negative for CD4, CD8, NK1.1, TCR-γδ, and B220 but positive for Thy1.2 were designated DN T cells and total cell numbers calculated.

To follow the expression of IFN-γ and IL-17A, cells from the lungs and spleens of LVS-infected mice were harvested at the indicate time points as described above. Cells were then incubated in cDMEM containing 5 μg/ml brefeldin A at 37°C in 5% CO2 for 4 h (without additional exogenous stimulation, as cell preparations also contained live bacteria from the in vivo infection) and stained for extracellular markers. Intracellular staining was performed using the BD Biosciences buffer system according to the manufacturer’s instructions. As above, all singlet and live-gated cells designated as DN T cells were gated and then analyzed for expression of IFN-γ and IL-17A.

mRNA isolation and quantitative RT-PCR

CD4+, CD8+, and DN T cells were enriched from LVS-immune spleens harvested 2 to 3 mo after 104 LVS i.d. infection and cocultured with LVS-infected BMMΦs as described above. After 48 h, T cells were removed and RNA isolated using the Qiagen RNeasy Midi kit according to the manufacturer’s instructions (Valencia, CA). cDNA was synthesized using the Superscript First Strand cDNA Synthesis according to the manufacturer’s instructions (Invitrogen, Carlsbad, CA). Quantitative SYBR green-based RT-PCR was performed using the RT2 Profiler PCR Array for Th17 cells according to the manufacturer’s instructions (Bioscience, Frederick, MD).

Statistical analyses

All experiments were performed using three to five mice per experimental group and repeated at least twice to assess reproducibility. Differences between the means of experimental groups were analyzed using the two-tailed Student t test; p values ≤ 0.05 were considered significant.

Results

LVS-immune DN T cells are a unique T cell subset that are restricted via MHC class I molecules

Our previous studies demonstrated that DN T cells purified from the spleens of F. tularensis LVS i.d.-immunized mice controlled LVS growth in murine BMMΦs similar to LVS-immune CD4+ and CD8+ T cells (9). To directly compare their relative potencies, CD4+, CD8+, and DN T cells were isolated from the spleens of mice previously given a primary sublethal LVS i.d. infection, added in a range of concentrations to LVS-infected BMMΦs, and control of LVS intramacrophage growth determined. All three T cell subsets exhibited a dose-dependent inhibition of LVS growth (Fig. 1A). However, on a cell-by-cell basis, CD4+ and DN T cells were not different from each other, whereas both were significantly more effective than CD8+ T cells at all concentrations. Thus, LVS-immune DN T cells are as effective as CD4+ T cells at controlling LVS growth in vitro and more effective than conventional CD8+ T cells.

FIGURE 1.
FIGURE 1.

LVS-immune splenic DN T cells are restricted via MHC class I Db molecules but more effective than CD8+ T cells in controlling LVS intramacrophage growth in vitro. A, Titration of purified LVS-immune CD4, CD8, and DN T cells for control LVS intramacrophage growth. BMMΦs from WT mice were infected with LVS at an MOI of 1:20 (bacterium-to-macrophage ratio) and cocultured with splenocytes from either uninfected mice (x-axis 0 point) or increasing concentrations of purified splenocyte T cell subpopulations harvested from C57BL/6J mice infected i.d. with LVS 4–8 wk previously. Three days postinfection, BMMΦ were washed, lysed, and plated to determine the levels of intracellular bacteria. Values shown are the mean numbers of CFU/ml ± SD of viable bacteria (triplicate samples). Supernatants from triplicate samples obtained from LVS cocultures immediately prior to macrophage lysis from the indicated cultures were assessed by ELISA for IFN-γ (B), TNF-α (C), or by Griess reaction as an indirect measure of NO (D). For A–D, *p ≤ 0.05 as compared with DN T cell cocultures. In E, purified LVS-immune DN T cells were added to infected BMMΦs at a ratio of 1:2 (T cell/BMMΦs), and anti-MHC class I or II Abs were added to the wells at the beginning of coculture as indicated. Three days after the start of coculture, BMMΦs were washed, lysed, and plated to determine the levels of intracellular bacteria. Values shown are the mean numbers of CFU/ml ± SD of viable bacteria (triplicate samples). These data are representative of four independent experiments of similar design. *p ≤ 0.01 as compared with naive spleen cell cocultures.

We previously showed that all three T cell subsets rely, to varying extents, upon IFN-γ, TNF-α, and NO production to control LVS intracellular growth in vitro (8, 9). Analyses of cytokine production in the coculture supernatants demonstrated that CD4+ T cells produced significantly higher levels of IFN-γ as compared with the CD8+ and DN T cells at the highest cell concentrations (Fig. 1B). In contrast, the CD4+ and DN T cell cocultures contained marginally but significantly higher levels of TNF-α and NO compared with CD8+ T cell cultures (Fig. 1C, 1D). Thus, differences in production of these classic antimicrobial mechanisms did not correlate with the differential abilities of CD4+, CD8+, and DN T cells to control LVS growth in vitro or in vivo.

We also investigated whether DN T cells recognize Ag through MHC class I or class II during in vitro growth control. Initial studies that cocultured wild-type (WT) LVS-immune DN T cells with BMMΦs derived from mice lacking either Kb/Db MHC class I, β2-microglobulin, or IA/IE class II molecules found no effect on the ability of DN T cells to control LVS growth as compared with WT BMMΦs (data not shown). Because a phenotypically similar subset of DN T cells, observed in allograft rejection models, can take up and present Ag themselves (18), we instead blocked all Ag presentation in the cocultures by addition of neutralizing anti-MHC Abs. DN T cell control of bacterial intracellular growth in WT BMMΦs was not affected by addition of anti-MHC class II Abs, but was reversed by addition of anti-MHC class I (Kb/Db) Abs to levels that were not significantly different from naive splenocyte cocultures (Fig. 1E; p > 0.05). Further, the ability of DN T cells to control LVS intracellular growth was abrogated by addition of anti-Db, but not anti-Kb, Abs to the cocultures. Thus, LVS-immune splenic DN T cells required MHC class I Db molecules to control LVS intracellular growth.

To further explore the potential functions of DN T cells, LVS-immune CD4+, CD8+, and DN T cells were cocultured with LVS-infected BMMΦs, then T cells were harvested from the cocultures, and mRNA was isolated for analyses of expression of a panel of common T cell cytokines. As expected, the CD4+, CD8+, and DN T cell subsets all significantly upregulated expression of IFN-γ mRNA (212 ± 5-, 120 ± 14-, and 164 ± 20-fold, respectively, as compared with identically treated naive T cells). Surprisingly, however, the mRNA transcript that was upregulated to the highest extent in the DN T cell subset was IL-17A. IL-17A expression was more than 465 ± 5-fold higher in DN T cells (as compared with naive T cells cultured under the same conditions), whereas CD4+ and CD8+ T cells exhibited a much smaller upregulation of IL-17A mRNA (∼8.5 ± 4-fold and 10.5 ± 6-fold, respectively).

IL-17A–deficient mice are more susceptible than normal mice to respiratory, and to a lesser extent dermal, LVS infection

Because LVS-immune DN T cells produced IL-17A transcripts following LVS restimulation in vitro, we determined the role of IL-17A in LVS infection in vivo. Previous studies by Woolard et al. (19) have shown that IL-17A is produced primarily in the lungs, but not the spleens, of mice given primary sublethal i.n. LVS infection. Further, we did not detect IL-17A in murine spleen or lung homogenates following primary sublethal i.d. LVS infection (data not shown). This suggests that IL-17A might have a more prominent role in infections introduced via the pulmonary than the dermal route. We therefore examined primary LVS sublethal infections in IL-17A–deficient mice and WT mice inoculated via the i.n. or i.d. routes. When inoculated with varying doses of LVS i.d., IL-17A knockout (KO) mice succumbed in greater numbers than WT only at the highest i.d. dose tested (7 × 106; zero out of five IL-17A KO mice survived as compared with four out of five WT mice that survived). In contrast, between days 10 and 14 postinfection, IL-17A KO mice succumbed to considerably lower i.n. doses than WT mice (1 × 103: zero out of five IL-17A KO and five out of five WT mice survived; 5 × 102: two out of five IL-17A KO and five out of five WT mice survived; 1 × 102: five out of five WT and IL-17A KO mice survived).

To compare outcomes in more detail, IL-17A KO mice or WT mice were given sublethal i.n. (102 CFU) and sublethal i.d. (104 CFU) infections with doses that were ∼1/100 LD50 for WT mice, and growth of LVS in lungs, livers, and spleens was monitored over time. WT and IL-17A KO mice inoculated via the i.d route exhibited no significant differences in LVS growth in the lungs or livers (Figs. 2A, 3B; p > 0.05), and both WT and KO mice cleared the infection by day 14. Spleen CFUs at the peak of infection (day 7) were significantly higher in the IL-17A KO mice (∼1 log difference; p < 0.05). However, this difference was only significant in two out of four experiments. In contrast, all experiments that introduced LVS infection via the i.n route revealed significantly higher LVS organ burdens in the IL-17A KO mice as compared with WT mice (Fig. 2D–F). In two out of four experiments, growth of LVS in the lungs following i.n infection was significantly different between WT and IL-17A KO mice at the peak of infection on day 7 (Fig. 2D), but not at earlier time points (data not shown). By day 10 postinfection, however, LVS CFUs were consistently significantly higher in the lungs of IL-17A KO mice (∼2 log difference; p < 0.01). This difference became even more pronounced by day 14 (∼3.5 log difference; p < 0.01), by which time the WT mice had essentially cleared the infection. IL-17A KO mice exhibited notably delayed clearance of the bacteria, with a small number of CFUs still detectable in the lungs at day 21. Similar results were observed for the spleens and livers of the IL-17A KO mice, with significantly higher CFUs at the later time points and delayed time to LVS clearance (Fig. 2E, 2F).

FIGURE 2.
FIGURE 2.

The absence of IL-17A has a greater impact on pulmonary as compared with parenteral LVS infections. Bacterial burdens in the lungs (A, D), livers (B, E), and spleens (C, F) of WT and IL-17A KO mice during sublethal 104 LVS intradermal infection (A–C) or sublethal 102 LVS intranasal infection (D–F) were determined. CFUs in the organs (three to four mice per group) were assessed for bacterial burdens at the indicated time points. Values shown are the mean numbers of log CFU/organ ± SD of viable bacteria. Limit of detection is calculated as ∼25 CFU/spleen and lung and 50 CFU/liver (dotted lines). These data are representative of three independent experiments of similar design. *p < 0.01 as compared with WT mice.

FIGURE 3.
FIGURE 3.

Neutrophil recruitment to the lungs of IL-17A KO mice is not impaired during LVS intranasal infection. WT and IL-17A KO mice were given a sublethal 102 LVS i.n. infection. At the indicated time points, lung cells were harvested and assessed by flow cytometry for the percentages (A) and total numbers (B) of neutrophils (CD11b+Gr1hi cells) per lung. Values shown are the mean numbers or percentages of neutrophils ± SD in the lungs of three mice at each time point. These data are representative of two independent experiments of similar design. *p < 0.05 as compared with WT mice.

IL-17A is well known for its ability to regulate granulopoiesis and recruit neutrophils to the site of infection (20). To investigate the possibility that delayed or impaired neutrophil recruitment was responsible for the survival defect of i.n.-infected IL-17A KO mice, IL-17A KO and WT mice were given a sublethal (102 CFU) i.n. LVS infection. Lung cells were then analyzed using flow cytometry to determine neutrophil percentages and total numbers in infected lungs during the early bacterial growth phase of infection (day 4), at the peak of infection (day 7), and during the clearance phase of infection (day 14). As shown in Fig. 3, the proportion of neutrophils (identified as CD11b+Gr1hi cells) present in the lungs of IL-17A KO mice was significantly less than that of WT mice only on day 4 postinfection; however, when the total numbers of neutrophils were calculated, the neutrophil numbers in the WT and IL-17A KO mice were not significantly different at any of the time points tested (p > 0.05). Thus, reduced neutrophil recruitment to the lungs of IL-17A KO mice was only observed during the early acute phase of infection and was not evident during the later phases of LVS i.n. infection.

p40-dependent lung DN T cells produce IL-17A during respiratory LVS infection in vivo

Because IL-17A KO mice exhibited increased susceptibility to i.n. LVS infection, and DN T cells exhibited high levels of IL-17A production in vitro, we evaluated the recruitment and cytokine production profile of DN T cells to the lungs of mice during the course of an i.n. LVS infection. Our previous studies demonstrated that the DN T cell subset was a minor population in the spleens of i.d.-LVS infected mice, representing ∼1–2% of total spleen cells (8); total numbers are ∼10-fold less than conventional CD4+ and CD8+ T cells (Fig. 4A, 4B). We therefore analyzed the three major T cell populations present in the lungs of WT mice during a sublethal (102) i.n. LVS infection. A representative flow cytometry-gating scheme for quantification of DN T cells in purified WT mouse lung cells harvested on day 14 after i.n. LVS infection compared with a representative dot plot of DN T cells in the spleen of a 104 LVS i.d.-infected mouse also on day 14 is shown in Fig. 4E. CD4+ T cells were the most numerous T cells present at all time points tested: ∼30% of total lung cells or 5 to 6 × 106 cells/lung by days 10–14 after i.n. infection (Fig. 4C, 4D). However, DN T cells and CD8+ T cells were both recruited to LVS-infected lungs in almost equivalent amounts. Although DN T cells were often present in slightly lower proportions than CD8+ T cells, these two cell types constituted ∼11–15% of lung cells by day 14 postinfection (Fig. 4C, 4E) or ∼2 × 106 cells/lung (Fig. 4D). Thus, DN T cells are a substantial T cell population present in the lung following i.n. LVS infection and a much higher proportion of all T cells than after i.d. infection.

FIGURE 4.
FIGURE 4.

DN T cells are a major responding T cell subset in the lungs during LVS pulmonary infection. WT mice were given a sublethal primary 104 LVS intradermal infection (A, B) or a sublethal primary 102 LVS intranasal infection (C, D). The resulting numbers of CD4+, CD8+, and DN T cells subsets in the spleens (A and B) or lungs (C, D) were assessed by flow cytometry. The percentage of each subset that was present in total spleen cells (A) and total lung cells (C) is shown and compared with the absolute number of each subset that was calculated to be present in the spleens (B) and lungs (D) of mice at the indicated time points. In E, the gating scheme used to identify DN T cells in the lungs and spleens of LVS i.n.- and i.d.-infected mice on day 14 (the peak of T cell numbers) postinfection is shown. The FITC exclusion channel axis indicates staining that used a combination of FITC-labeled anti-CD4, anti-CD8, anti-NK1.1, anti-γδTCR, and anti-B220 Abs. Values shown are the mean numbers of cells ± SD in the lungs or spleens of three to four mice at each time point. These data are representative of three independent experiments of similar design. *p < 0.05, for which DN T cell values differed from that of CD8+ T cells.

To determine the potential functions of these three T cell subsets, we next examined the lung cell types producing IL-17A and IFN-γ. Mononuclear cells were harvested from the lungs of LVS i.n.-infected WT mice and subjected to intracellular cytokine staining. Representative flow cytometry dot plots are shown in Fig. 5. Initial examination revealed that the majority of lung cells that stained intracellularly for IL-17A at day 7 after i.n. LVS infection were also Thy1.2+ (78 ± 8.3% and the balance were largely neutrophils and CD11c+ cells). At both days 7 and 14, a much higher percentage of DN T cells produced IL-17A than CD4+ and CD8+ T cells, with as many as 7% of DN T cells expressing IL-17A on day 7 as compared with 0.2–0.4% of CD4+ or CD8+ T cells (Fig. 5A, 5B). These differences were also reflected in the total number of IL-17A+ T cells present in the lungs during LVS i.n. infection. DN T cells were clearly the most numerous T cell subset responsible for production of IL-17A in the lungs during the peak of infection, at day 7 postinoculation, with low numbers of IL-17A–producing CD4+ or CD8+ cells detected at this time point (Fig. 5F). However, by day 14, the numbers of DN T cells producing IL-17A diminished, whereas the numbers of IL-17A+ CD8+ T cells increased. Although the percentage of IL-17+ CD4+ T cells did not change significantly between days 7 and 14 (p > 0.05), the total numbers of CD4+ T cells increased significantly (p < 0.01; Fig. 4C, 4D), leading to an overall increase in the number of IL-17+ CD4+ T cells at this time point (p < 0.05; Fig. 5F). Therefore, by day 14 postinfection, approximately equal numbers of lung DN, CD8+, and CD4+ T cell subsets were producing IL-17A (Fig. 5F). Of note, we detected similar numbers of IL-17+CD4+ cells in the lungs as previous studies by Woolard et al. (19) (∼104 IL-17A+CD4+ T cells/lung on day 14 postinfection); in addition, no intracellular staining for IL-17A was detected in spleens harvested from i.n.- or i.d.-infected mice at any time point or in the lungs of i.d.-infected mice (representative day 7 dot plots are shown in Fig. 5C, 5D).

FIGURE 5.
FIGURE 5.

DN T cells produce high levels of IL-17A in the lungs of mice during LVS pulmonary infection, but IL-17A production is route and organ dependent. WT mice were given a sublethal 102 LVS i.n. infection, and production of IL-17A and IFN-γ by CD4+, CD8+, and DN T cells in the lungs of infected mice was monitored by flow cytometry over time. Representative dot plots of the following are shown: IFN-γ and IL-17A production in the lungs of i.n.-infected mice by CD4+, CD8+, and DN T cells at day 7 (A) and day 14 (B) postinfection; IFN-γ and IL-17A production by CD4+, CD8+, and DN T cells in the spleens of i.n.-infected mice on day 7 postinfection (C); DN T cell production of IL-17A in the lungs and spleens of sublethal 104 i.d.-infected mice on day 7 postinfection (D); and IL-17A and IFN-γ production by DN, CD4+, and CD8+ T cells in the lungs on day 14 after i.n. infection (E). The absolute numbers of each subset that produced IL-17A (F) and IFN-γ (G) in the lungs of i.n.-infected mice at the indicated time points postinfection are shown. Asterisks (*) indicate p values < 0.01 for which the numbers of cytokine-producing cells of one T cell subset differed from that of the other two T cell subsets. The numbers of IFN-γ–producing CD4+ T cells (H), DN T cells (I), and CD8+ T cells (J) that were present in the lungs of i.n.-infected WT and IL-17A KO mice during infection were also compared. These data are representative of three independent experiments of similar design. Values shown are the mean numbers of cells ± SD in the lungs or spleens of three to four mice at each time point. *p < 0.01 as compared with WT mice.

In contrast, the pattern of IFN-γ production during i.n. LVS infection was quite different. CD4+ T cells and DN T cells were the primary cells producing IFN-γ at day 7 with comparable numbers, whereas numbers of IFN-γ+ CD8+ T cells were significantly lower (Fig. 5A, 5G; p < 0.05). In contrast to IFN-γ+ DN T cells, IFN-γ+ CD4+ T cells continued to increase in numbers through day 14 and were the largest population through day 21 (Fig. 5G). Only very small numbers of DN T, CD8+ T, or CD4+ T cells produced both IL-17A and IFN-γ+ (Fig. 5A).

We further compared T cell production of IFN-γ in the WT and IL-17A KO mice. Significantly higher numbers of DN T cells and CD4+ T cells, but not CD8+ T cells, in the lungs of IL-17A KO mice produced IFN-γ as compared with WT mice (Fig. 5H–J; p < 0.05). These differences were evident on days 10 and 14, time points when CFUs in the lungs of IL-17 KO mice were significantly higher than those of the WT mice. These data suggest that lung cells of IL-17A KO mice may upregulate and/or overcompensate for the absence of IL-17 in response to higher bacterial burdens, but that these higher levels of T cell IFN-γ production were still insufficient to control LVS growth. Further, IFN-γ production in the lungs of WT mice following i.n. LVS infection was dominated by a later, largely CD4+-dependent phase of IFN-γ production between days 14 and 21, when bacteria are being eradicated from organs.

Because IL-12 (a heterodimer of the p40 and p35 subunits) is required for the development of IFN-γ–producing Th1 cells, and IL-23 (a heterodimer of the p40 and p19 subunits) is required for the development of IL-17A–producing Th17 cells, we next determined the role of these molecules in DN T cell production of IFN-γ and IL-17A during primary i.n. LVS. WT, p35 KO, and p40 KO mice were given a sublethal 102 i.n. LVS infection, and IFN-γ–producing and IL-17A–producing DN T cells in lungs determined on day 7 postinfection. At this time point, bacterial lung burdens were highest in infected p40 KO mice, intermediate in p35 KO mice, and lowest in WT mice (Fig. 6A). As expected, p35 and p40 KO mice had significantly lower numbers of IFN-γ–producing CD4+ T cells and DN T cells in their lungs (p < 0.05; Fig. 6B), consistent with the role of IL-12 in the development of Th1 cells. In contrast, only the p40 KO mice had significantly lower numbers of IL-17A–producing CD4+ T cells and DN T cells in lungs as compared with WT mice (Fig. 6C). Thus, DN T cell production of IL-17A was dependent upon the p40, but not the p35, subunit of IL-23/IL-12.

FIGURE 6.
FIGURE 6.

p40, but not p35, is required for DN T cell production of IL-17A during sublethal LVS i.n. infection. WT, p40 KO, and p35 KO mice were given a sublethal 102 LVS i.n. infection, and production of IL-17A and IFN-γ by CD4+, CD8+, and DN T cells in the lungs of mice was monitored by flow cytometry. A, Bacterial burdens in the lungs of infection WT, p35 KO, and p40 KO mice on day 7 are shown. Numbers of IFN-γ–producing (B) and IL-17A–producing (C) CD4+ T cells, DN T cells, and CD8+ T cells that were present in the lungs of i.n.-infected WT, p35 KO, and p40 KO mice on day 7 postinfection are also shown. Values shown are the mean numbers of cells ± SD in the lungs of three to four mice at each time point. These data are representative of two independent experiments of similar design. *p < 0.01 as compared with WT mice.

We previously showed that WT mice administered an Ab mixture to deplete all potential T cells, except the DN T cell subset did not clear a sublethal intradermal (104 CFU) LVS infection, but instead developed a long-term chronic infection (8). In contrast, mice depleted of all T cells using an anti-Thy1.2 Ab all died within days 11–14 after a sublethal i.d. infection, suggesting that DN T cells mediate this chronic infection. We next explored the ability of DN T cells in the absence of other T cell types to control i.n. LVS infection in the absence of IL-17A. Groups of IL-17A KO mice and WT control mice were administered an Ab mixture to deplete CD4+, CD8+, γδTCR+, and NK1.1+ cells, leaving only DN T cells as the major T cell subset available to respond to infection. Untreated and Ab mixture-depleted WT and IL-17A KO mice were then given a sublethal i.n. (2 × 102 CFU) LVS infection, and survival was monitored over time. As seen in Fig. 7, all untreated WT and IL-17A KO mice survived the infection. In contrast, 55% of the Ab mixture-treated IL-17A KO mice died of this i.n. LVS infection between days 10 and 15, whereas no Ab mixture-treated WT mice died. Similar to our previous i.d. infection studies, surviving Ab mixture-treated mice failed to clear the infection from livers, lungs, or spleens (both WT and IL-17A KO mice; final time point tested was 36 d postinfection). Collectively, these data indicate that when DN T cells are the only T cell subset available during an in vivo i.n. LVS infection, they require IL-17A for optimum control of LVS infection, but are unable to completely clear the infection from the tissues of either WT or IL-17A KO mice.

FIGURE 7.
FIGURE 7.

DN T cells require IL-17A to control primary sublethal LVS i.n. infection. WT and IL-17A KO mice were treated with either PBS or a combination of Abs designed to remove all T cells except DN T cells (Ab combination: anti-CD4, anti-CD8, anti-NK1.1, and anti–TCR-γ/δ). Ab mixture-treated mice were then given a sublethal 2 × 102 LVS i.n. infection, and survival was monitored for WT-Ab mixture (n = 15), IL-17A KO-Ab mixture (n = 20), and WT (n = 10) and IL-17A KO (n = 10) mice.

IL-17A cooperates with IFN-γ to inhibit LVS intracellular growth in ATII epithelial cells

We were next interested in assessing the direct impact of IL-17A on LVS intracellular growth in vitro. Because our data demonstrated that IL-17A is particularly important for LVS growth control in pulmonary infections in vivo, we examined the effects of IL-17A and IFN-γ on LVS growth within both a murine macrophage cell line (RAW 264.7) and an ATII epithelial cell line (TC-1). As seen in Fig. 8A, LVS grew exponentially in both the RAW and TC-1 cell lines over 72 h. Addition of rIL-17A significantly reduced LVS growth in both cell lines, with a 0.83 log10 and 0.51 log10 CFU reduction for macrophages (Fig. 8B) and ATII cells (Fig. 8C), respectively (p < 0.01). Addition of rIFN-γ also significantly reduced LVS intracellular growth in both cell lines (Fig. 8B, 8C); however, rIFN-γ had a much greater impact on LVS growth control in macrophages (3.8 log10 CFU reduction; p < 0.01) as compared with ATII cells (1.52 log10 CFU reduction; p < 0.01). Interestingly, addition of both rIL-17A and rIFN-γ to cultures significantly increased the ability of ATII cells to control LVS growth as compared with cultures that contained either IL-17A or IFN-γ alone; an additional 1.20 log10 CFU reduction was observed for the IL-17A/IFN-γ combination over IFN-γ alone in ATII cell cultures. Similarly, addition of both rIL-17A and rIFN-γ to cultures increased the ability of RAW cells to control LVS growth as compared with cultures that contained either IL-17A or IFN-γ alone (1.22 log10 CFU reduction; p < 0.01). Thus, IL-17A appears to have the capacity to work additively with IFN-γ to control LVS intracellular growth in host cells, including macrophages and ATII cells.

FIGURE 8.
FIGURE 8.

IL-17A synergizes with IFN-γ to augment control of LVS intracellular growth in the murine macrophage cell line RAW 264.7 and the murine ATII epithelial cell line TC-1. RAW 264.7 cells and TC-1 cells were infected with LVS at a 1:10 and 1:1 MOI, respectively (bacterium-to-cell ratio). Infected cells were washed, lysed, and plated to determine the levels of intracellular bacteria. In A, growth of LVS in TC-1 cells and RAW cells was enumerated at 0, 24, 48, and 72 h postinfection. To assess the effect of rIFN-γ and rIL-17A on LVS intracellular growth, RAW cells (B) and TC-1 cells (C) were infected with LVS and treated with the indicated cytokines for the duration of the culture (72 h). Values shown are the mean numbers of CFU/ml ± SD of viable bacteria (triplicate samples). These data are representative of three independent experiments of similar design. *p ≤ 0.01 as compared with LVS alone; ▼p ≤ 0.01 as compared with rIL-17A; *p ≤ 0.01 as compared with rIFN-γ.

Discussion

The mechanisms contributing to differences in effectiveness between pulmonary and parenteral vaccination remain incompletely understood, as does the function of an unusual DN T cell subpopulation. In this study, we show that DN T cells are a major responding T cell subset in the lungs of mice during pulmonary LVS infection. During primary sublethal i.n. LVS infection, DN T cells constituted as much as 10–15% of lung cells and were comprised of two distinct subpopulations that produced either IFN-γ or IL-17A (Figs. 4, 5). This is in sharp contrast to primary i.d. LVS infection, in which dissemination to the lungs was minimal, and DN T cells were only a minor splenic T cell population (1% to 2% of total spleen cells) with no detectable secretion of IL-17A (Fig. 5). Further, during primary i.n. LVS infection, DN T cells exhibited a unique temporal cytokine production profile, distinct from that of either CD4+ or CD8+ T cells (Figs. 4, 5, 7). Unlike CD8+ T cells, DN T cells produced large amounts of IFN-γ during the acute phase (day 7) of respiratory LVS infection. Further, DN T cells produced significantly higher levels of IL-17A than CD4+ and CD8+ T cells on day 7 after LVS i.n. infection. By day 14 postinfection, however, DN T cell production of IFN-γ and IL-17A declined, CD4+ and CD8+ T cell production of IL-17A rose, and CD4+ T cells became the dominant lung IFN-γ–producing T cell population. Finally, we found that IL-17A was a critical component of the pulmonary immune response that controlled primary sublethal i.n. LVS infection during an intermediate 1–3 wk time frame (Fig. 2) and that IL-17A can act in concert with IFN-γ to augment LVS growth control in both macrophages and ATII epithelial cells in vitro (Fig. 8).

IL-17A can be produced by a wide variety of cell types following infections, including γδ T cells (21, 22), lung CD4+ T cells (23), and lung CD8+ cells (Tc17 cells) (24). Interestingly, one study identified a subset of unconventional IL-17A–producing CD4−CD8− αβ T cells that naturally accumulated in spleens of uninfected CD18-deficient mice; they were dubbed neutrophil regulatory T cells (Tn cells) for their role in regulating neutrophil granulopoeisis (25). Similar to the DN T cell subset described in this study, these cells were NK1.1−, but they had an oligoclonal distribution of ∼60% Vβ8-TCR that was consistent with an NK T cell phenotype. In this study, the demonstration that LVS intramacrophage growth control by LVS-immune DN T cells is inhibited by anti-MHC class I Abs (Fig. 1) suggests that they differ from the classical invariant NK T cells that typically recognize Ag via CD1d. Indeed, the DN T cells observed during LVS infection appear to be phenotypically most similar to an MHC class I-restricted CD4−CD8−NK1.1− DN T cell subset that was identified in a murine allograft rejection model (18, 26).

Previous studies by Woolard et al. (19) following primary sublethal i.d. and i.n. LVS infections also indicated that IL-17A production was route- and organ-specific; following i.n. LVS inoculation of WT mice, IL-17A was detected in CD4+ T cells in mouse lungs but not spleens and neither lungs nor spleens following i.d. inoculation. Our results presented in this paper support and extend their findings regarding lung-specific IL-17A production, in that IL-17A KO mice were more susceptible than normal WT mice to primary i.n. LVS infections but less severely affected during primary i.d. LVS infection. Another recent study demonstrated increased susceptibility of IL-17 KO mice to one high dose of i.n. LVS infection (27), but in this study, all of the mice died by days 8–10 postinfection, limiting their ability to examine the role of IL-17A during later time points of infection; further, i.d. LVS infection was not compared. Although previous studies reported T cell production of IL-17A during primary i.n. LVS infection (19, 27), only conventional CD4, CD8, and γδ T cell subsets were examined. In this study, we observed that CD4+ T cells produced IL-17A during LVS i.n. infection in numbers that were previously reported [∼104 IL-17A+ CD4+ T cells/lung; Woolard et al. (19)], but we further demonstrate that an additional IL-17A–producing T cell subset, DN T cells, was present in even higher numbers during the peak of LVS i.n. infection (Fig. 5).

A number of studies have shown that IL-17A has an important role in mucosal infections. In particular, IL-17A KO mice are especially susceptible to primary infections with extracellular pathogens, such as Klebsiella pneumoniae and Mycoplasma pulmonis (28, 29). With these infections, Th17 products such as IL-22 and IL-17A were essential for upregulation of host defense genes, such as β-defensins, as well as the generation of chemokines necessary for neutrophils granulopoiesis and recruitment (30). Indeed, the important role of IL-17 in neutrophil responses is now well established (20).

To date, IL-17A studies in primary intracellular infections have been more difficult to interpret. For example, mice deficient in IL-17RA, which is required for signaling of IL-17A and IL-17F, were not more susceptible to virulent M. tuberculosis or Listeria monocytogenes infection as compared with WT mice (30, 31). In a separate study, IL-17A KO mice were more susceptible than WT mice to primary i.p. L. monocytogenes infection, exhibiting 100-fold higher levels of bacteria in livers, reduced liver granuloma formation, reduced β-defensin production, and increased liver damage accompanied by increased liver neutrophils (24). In our study, we observed only a minor decrease in recruitment of neutrophils to the lungs of sublethally i.n. LVS-infected IL-17A KO mice at only one of the time points tested (day 4), prior to the time when we see an impact on bacterial growth in the mice (Fig. 3). In a previous study of lethal pulmonary LVS infection of IL-17A KO mice, reduced recruitment of CD11b+Gr1+ cells was detected on day 6 as compared with WT mice [Lin et al. (27)]; however, these mice had extremely high organ burdens and were shortly going to succumb to the lethal infection. Although we cannot completely rule out the possibility that, in our sublethal infection model, the transient neutrophil recruitment defect we observed on day 4 did not impact bacterial CFUs on days 10 and 14, these data suggest that the major role of IL-17A in immunity to LVS sublethal infection may extend beyond its established role in granulopoeisis and neutrophil chemotaxis.

Because IL-17Rs are expressed on a wide variety of cell types, including leukocytes and epithelial cells, and many of these cell types can support the growth of Francisella species (13, 32, 33), we investigated the direct impact of IL-17A on control of LVS growth in host cells. Interestingly, we found that IL-17A can cooperate with IFN-γ to inhibit growth of LVS in alveolar epithelial cells and macrophages. The specific antibacterial macrophage and ATII cell mechanisms induced by IFN-γ and IL-17A remain to be determined. We further found that during i.n. LVS infection, IL-17A–deficient mice harbored significantly higher numbers of IFN-γ+ CD4+ T cells and IFN-γ+ DN T cells in lungs as compared with WT mice. The higher numbers of responding IFN-γ+ T cells in the lungs of IL-17A KO mice occurred in concert with significantly higher LVS CFUs in the lungs, further indicating that IFN-γ is not fully effective at controlling LVS growth in the absence of IL-17A. Curiously, another recent study using intratracheal LVS infection found increased susceptibility of IL-17A KO mice to i.n. LVS infection, but observed reduced IFN-γ production (27). This discrepancy is likely due to the fact that, in contrast to our study, this work examined a lethal LVS dose at a single early time point (day 6), rather than a sublethal dose at a range of later time points (days 7, 10, and 14) when T cells are actively attempting to clear the infection. Thus, although early/acute phase production of IFN-γ may be diminished by the absence of IL-17A, T cell production of IFN-γ increases later in sublethally infected IL-17A KO mice during the clearance/control phase of infection.

IL-23 is clearly required for the in vivo function of Th17 cells, promoting maintenance and terminal differentiation of the Th17 phenotype (34, 35). We previously found that mice deficient in the IL-23/IL-12 p40 subunit, but not the IL-12 p35subunit, were more susceptible to i.d. LVS infection than WT mice and developed a long-term chronic LVS infection (36). These data suggested that p40 has a role in clearance of LVS infection that is independent of IL-12 and thus likely a function of the IL-23 heterodimer or p40 homodimer. In this study, we found that p40, but not p35, is required for IL-17A production by DN T cells in LVS pulmonary infection, further supporting the conclusion that DN T cells develop in a manner similar to traditional Th17 cells (Fig. 6). However, in contrast to the long-term chronic infection exhibited by p40 KO mice, IL-17A KO mice resolved primary i.n. and i.d. LVS infections. Thus, the inability of p40 KO mice to clear LVS infection is not simply a result of its role in promoting IL-17A production, but possibly a function of other IL-23 or p40 homodimer-dependent activities, such as IL-22 production.

In summary, a unique DN T cell subset is an important responding T cell type in the lungs of mice after i.n. LVS infection, exhibiting a distinct temporal cytokine production profile that distinguishes them from conventional CD4+ and CD8+ T cells. The DN T cells that expand after pulmonary LVS infection segregated into two subsets that produced two essential cytokines, IFN-γ and IL-17A. Although maximal DN T cell production of IFN-γ and IL-17A peaked around day 7, CD4+ T cell production of these cytokines peaked later, at ∼day 14 postinfection. Further, an important role for IL-17A was delineated that functions during an intermediate phase of a sublethal i.n. LVS infection: IL-17A was important shortly after the peak of infection (days 7–14), but not required for the final clearance of the pathogen (∼week 3 to 4). Finally, we demonstrate that IL-17A has the capacity to act in concert with IFN-γ to enhance the ability of multiple host cell types, including macrophages and ATII epithelial cells, to control LVS intracellular growth.

Acknowledgments

We thank Dr. James Forman for generous contributions of materials and advice during the development of these experiments and our Center for Biologics Evaluation and Research colleagues Sheldon Morris and Ronald Rabin for thoughtful reviews of the manuscript.

Disclosures The authors have no financial conflicts of interest.

Footnotes

  • Abbreviations used in this paper:

    ATII
    alveolar type II epithelial cell
    BMMΦ
    bone marrow-derived macrophage
    cDMEM
    complete DMEM
    DN
    double-negative
    i.d.
    intradermal
    i.n.
    intranasal
    KO
    knockout
    LVS
    live vaccine strain
    MOI
    multiplicity of infection
    WT
    wild-type.

  • Received February 2, 2010.
  • Accepted March 9, 2010.

References

  1. ↵
    1. Baumann U.
    2008. Mucosal vaccination against bacterial respiratory infections. Expert Rev. Vaccines 7: 1257–1276.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Elkins K. L.,
    2. S. C. Cowley,
    3. C. M. Bosio
    . 2003. Innate and adaptive immune responses to an intracellular bacterium, Francisella tularensis live vaccine strain. Microbes Infect. 5: 135–142.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Ellis J.,
    2. P. C. Oyston,
    3. M. Green,
    4. R. W. Titball
    . 2002. Tularemia. Clin. Microbiol. Rev. 15: 631–646.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Dennis D. T.,
    2. T. V. Inglesby,
    3. D. A. Henderson,
    4. J. G. Bartlett,
    5. M. S. Ascher,
    6. E. Eitzen,
    7. A. D. Fine,
    8. A. M. Friedlander,
    9. J. Hauer,
    10. M. Layton,
    11. et al,
    12. Working Group on Civilian Biodefense
    . 2001. Tularemia as a biological weapon: medical and public health management. JAMA 285: 2763–2773.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Rick Lyons C.,
    2. T. H. Wu
    . 2007. Animal models of Francisella tularensis infection. Ann. N. Y. Acad. Sci. 1105: 238–265.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Wayne Conlan J.,
    2. P. C. Oyston
    . 2007. Vaccines against Francisella tularensis. Ann. N. Y. Acad. Sci. 1105: 325–350.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Tärnvik A.,
    2. M. Eriksson,
    3. G. Sandström,
    4. A. Sjöstedt
    . 1992. Francisella tularensis—a model for studies of the immune response to intracellular bacteria in man. Immunology 76: 349–354.
    OpenUrlPubMed
  8. ↵
    1. Cowley S. C.,
    2. E. Hamilton,
    3. J. A. Frelinger,
    4. J. Su,
    5. J. Forman,
    6. K. L. Elkins
    . 2005. CD4-CD8- T cells control intracellular bacterial infections both in vitro and in vivo. J. Exp. Med. 202: 309–319.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Cowley S. C.,
    2. K. L. Elkins
    . 2003. Multiple T cell subsets control Francisella tularensis LVS intracellular growth without stimulation through macrophage interferon gamma receptors. J. Exp. Med. 198: 379–389.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Fortier A. H.,
    2. M. V. Slayter,
    3. R. Ziemba,
    4. M. S. Meltzer,
    5. C. A. Nacy
    . 1991. Live vaccine strain of Francisella tularensis: infection and immunity in mice. Infect. Immun. 59: 2922–2928.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Petersen J. M.,
    2. M. E. Schriefer,
    3. K. L. Gage,
    4. J. A. Montenieri,
    5. L. G. Carter,
    6. M. Stanley,
    7. M. C. Chu
    . 2004. Methods for enhanced culture recovery of Francisella tularensis. Appl. Environ. Microbiol. 70: 3733–3735.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Bosio C. M.,
    2. K. L. Elkins
    . 2001. Susceptibility to secondary Francisella tularensis live vaccine strain infection in B-cell-deficient mice is associated with neutrophilia but not with defects in specific T-cell-mediated immunity. Infect. Immun. 69: 194–203.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Hall J. D.,
    2. R. R. Craven,
    3. J. R. Fuller,
    4. R. J. Pickles,
    5. T. H. Kawula
    . 2007. Francisella tularensis replicates within alveolar type II epithelial cells in vitro and in vivo following inhalation. Infect. Immun. 75: 1034–1039.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Yee D.,
    2. T. R. Rhinehart-Jones,
    3. K. L. Elkins
    . 1996. Loss of either CD4+ or CD8+ T cells does not affect the magnitude of protective immunity to an intracellular pathogen, Francisella tularensis strain LVS. J. Immunol. 157: 5042–5048.
    OpenUrlAbstract
  15. ↵
    1. Cowley S. C.,
    2. K. L. Elkins
    . 2003. CD4+ T cells mediate IFN-gamma-independent control of Mycobacterium tuberculosis infection both in vitro and in vivo. J. Immunol. 171: 4689–4699.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Green L. C.,
    2. D. A. Wagner,
    3. J. Glogowski,
    4. P. L. Skipper,
    5. J. S. Wishnok,
    6. S. R. Tannenbaum
    . 1982. Analysis of nitrate, nitrite, and [15N]nitrate in biological fluids. Anal. Biochem. 126: 131–138.
    OpenUrlCrossRefPubMed
  17. ↵
    1. De Pascalis R.,
    2. B. C. Taylor,
    3. K. L. Elkins
    . 2008. Diverse myeloid and lymphoid cell subpopulations produce gamma interferon during early innate immune responses to Francisella tularensis live vaccine strain. Infect. Immun. 76: 4311–4321.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Zhang Z. X.,
    2. L. Yang,
    3. K. J. Young,
    4. B. DuTemple,
    5. L. Zhang
    . 2000. Identification of a previously unknown antigen-specific regulatory T cell and its mechanism of suppression. Nat. Med. 6: 782–789.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Woolard M. D.,
    2. L. L. Hensley,
    3. T. H. Kawula,
    4. J. A. Frelinger
    . 2008. Respiratory Francisella tularensis live vaccine strain infection induces Th17 cells and prostaglandin E2, which inhibits generation of gamma interferon-positive T cells. Infect. Immun. 76: 2651–2659.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Iwakura Y.,
    2. S. Nakae,
    3. S. Saijo,
    4. H. Ishigame
    . 2008. The roles of IL-17A in inflammatory immune responses and host defense against pathogens. Immunol. Rev. 226: 57–79.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Hamada S.,
    2. M. Umemura,
    3. T. Shiono,
    4. H. Hara,
    5. K. Kishihara,
    6. K. Tanaka,
    7. H. Mayuzumi,
    8. T. Ohta,
    9. G. Matsuzaki
    . 2008. Importance of murine Vdelta1gammadelta T cells expressing interferon-gamma and interleukin-17A in innate protection against Listeria monocytogenes infection. Immunology 125: 170–177.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Lockhart E.,
    2. A. M. Green,
    3. J. L. Flynn
    . 2006. IL-17 production is dominated by gammadelta T cells rather than CD4 T cells during Mycobacterium tuberculosis infection. J. Immunol. 177: 4662–4669.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Khader S. A.,
    2. G. K. Bell,
    3. J. E. Pearl,
    4. J. J. Fountain,
    5. J. Rangel-Moreno,
    6. G. E. Cilley,
    7. F. Shen,
    8. S. M. Eaton,
    9. S. L. Gaffen,
    10. S. L. Swain,
    11. et al
    . 2007. IL-23 and IL-17 in the establishment of protective pulmonary CD4+ T cell responses after vaccination and during Mycobacterium tuberculosis challenge. Nat. Immunol. 8: 369–377.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Hamada H.,
    2. Mde. L. Garcia-Hernandez,
    3. J. B. Reome,
    4. S. K. Misra,
    5. T. M. Strutt,
    6. K. K. McKinstry,
    7. A. M. Cooper,
    8. S. L. Swain,
    9. R. W. Dutton
    . 2009. Tc17, a unique subset of CD8 T cells that can protect against lethal influenza challenge. J. Immunol. 182: 3469–3481.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Stark M. A.,
    2. Y. Huo,
    3. T. L. Burcin,
    4. M. A. Morris,
    5. T. S. Olson,
    6. K. Ley
    . 2005. Phagocytosis of apoptotic neutrophils regulates granulopoiesis via IL-23 and IL-17. Immunity 22: 285–294.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Zhang Z. X.,
    2. Y. Ma,
    3. H. Wang,
    4. J. Arp,
    5. J. Jiang,
    6. X. Huang,
    7. K. M. He,
    8. B. Garcia,
    9. J. Madrenas,
    10. R. Zhong
    . 2006. Double-negative T cells, activated by xenoantigen, lyse autologous B and T cells using a perforin/granzyme-dependent, Fas-Fas ligand-independent pathway. J. Immunol. 177: 6920–6929.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Lin Y.,
    2. S. Ritchea,
    3. A. Logar,
    4. S. Slight,
    5. M. Messmer,
    6. J. Rangel-Moreno,
    7. L. Guglani,
    8. J. F. Alcorn,
    9. H. Strawbridge,
    10. S. M. Park,
    11. et al
    . 2009. Interleukin-17 is required for T helper 1 cell immunity and host resistance to the intracellular pathogen Francisella tularensis. Immunity 31: 799–810.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Ye P.,
    2. F. H. Rodriguez,
    3. S. Kanaly,
    4. K. L. Stocking,
    5. J. Schurr,
    6. P. Schwarzenberger,
    7. P. Oliver,
    8. W. Huang,
    9. P. Zhang,
    10. J. Zhang,
    11. et al
    . 2001. Requirement of interleukin 17 receptor signaling for lung CXC chemokine and granulocyte colony-stimulating factor expression, neutrophil recruitment, and host defense. J. Exp. Med. 194: 519–527.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Wu Q.,
    2. R. J. Martin,
    3. J. G. Rino,
    4. R. Breed,
    5. R. M. Torres,
    6. H. W. Chu
    . 2007. IL-23-dependent IL-17 production is essential in neutrophil recruitment and activity in mouse lung defense against respiratory Mycoplasma pneumoniae infection. Microbes Infect. 9: 78–86.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Aujla S. J.,
    2. Y. R. Chan,
    3. M. Zheng,
    4. M. Fei,
    5. D. J. Askew,
    6. D. A. Pociask,
    7. T. A. Reinhart,
    8. F. McAllister,
    9. J. Edeal,
    10. K. Gaus,
    11. et al
    . 2008. IL-22 mediates mucosal host defense against Gram-negative bacterial pneumonia. Nat. Med. 14: 275–281.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Aujla S. J.,
    2. P. J. Dubin,
    3. J. K. Kolls
    . 2007. Th17 cells and mucosal host defense. Semin. Immunol. 19: 377–382.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Hall J. D.,
    2. M. D. Woolard,
    3. B. M. Gunn,
    4. R. R. Craven,
    5. S. Taft-Benz,
    6. J. A. Frelinger,
    7. T. H. Kawula
    . 2008. Infected-host-cell repertoire and cellular response in the lung following inhalation of Francisella tularensis Schu S4, LVS, or U112. Infect. Immun. 76: 5843–5852.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Gentry M.,
    2. J. Taormina,
    3. R. B. Pyles,
    4. L. Yeager,
    5. M. Kirtley,
    6. V. L. Popov,
    7. G. Klimpel,
    8. T. Eaves-Pyles
    . 2007. Role of primary human alveolar epithelial cells in host defense against Francisella tularensis infection. Infect. Immun. 75: 3969–3978.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Stritesky G. L.,
    2. N. Yeh,
    3. M. H. Kaplan
    . 2008. IL-23 promotes maintenance but not commitment to the Th17 lineage. J. Immunol. 181: 5948–5955.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. McGeachy M. J.,
    2. Y. Chen,
    3. C. M. Tato,
    4. A. Laurence,
    5. B. Joyce-Shaikh,
    6. W. M. Blumenschein,
    7. T. K. McClanahan,
    8. J. J. O’Shea,
    9. D. J. Cua
    . 2009. The interleukin 23 receptor is essential for the terminal differentiation of interleukin 17-producing effector T helper cells in vivo. Nat. Immunol. 10: 314–324.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Elkins K. L.,
    2. A. Cooper,
    3. S. M. Colombini,
    4. S. C. Cowley,
    5. T. L. Kieffer
    . 2002. In vivo clearance of an intracellular bacterium, Francisella tularensis LVS, is dependent on the p40 subunit of interleukin-12 (IL-12) but not on IL-12 p70. Infect. Immun. 70: 1936–1948.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section