γδ T Cells Are Required for Pulmonary IL-17A Expression after Ozone Exposure in Mice: Role of TNFα

Ozone is an air pollutant that causes pulmonary symptoms. In mice, ozone exposure causes pulmonary injury and increases bronchoalveolar lavage macrophages and neutrophils. We have shown that IL-17A is important in the recruitment of neutrophils after subacute ozone exposure (0.3 ppm for 24–72 h). We hypothesized that γδ T cells are the main producers of IL-17A after subacute ozone. To explore this hypothesis we exposed wildtype mice and mice deficient in γδ T cells (TCRδ−/−) to ozone or room air. Ozone-induced increases in BAL macrophages and neutrophils were attenuated in TCRδ−/− mice. Ozone increased the number of γδ T cells in the lungs and increased pulmonary Il17a mRNA expression and the number of IL-17A+ CD45+ cells in the lungs and these effects were abolished in TCRδ−/− mice. Ozone-induced increases in factors downstream of IL-17A signaling, including G-CSF, IL-6, IP-10 and KC were also decreased in TCRδ−/− versus wildtype mice. Neutralization of IL-17A during ozone exposure in wildtype mice mimicked the effects of γδ T cell deficiency. TNFR2 deficiency and etanercept, a TNFα antagonist, also reduced ozone-induced increases in Il17a mRNA, IL-17A+ CD45+ cells and BAL G-CSF as well as BAL neutrophils. TNFR2 deficient mice also had decreased ozone-induced increases in Ccl20, a chemoattractant for IL-17A+ γδ T cells. Il17a mRNA and IL-17A+ γδ T cells were also lower in obese Cpefat versus lean WT mice exposed to subacute ozone, consistent with the reduced neutrophil recruitment observed in the obese mice. Taken together, our data indicate that pulmonary inflammation induced by subacute ozone requires γδ T cells and TNFα-dependent recruitment of IL-17A+ γδ T cells to the lung.

Introduction cd T cells are a key component of the innate immune response, especially at mucosal surfaces. These cells are found throughout the lung, particularly in the subepithelial region, where they may regulate other immune cells including macrophages and dendritic cells [1]. cd T cells are an important source of IL-17A, a key cytokine involved in neutrophilic inflammation [2]. In mice, the number of pulmonary cd T cells increases following infection with certain bacteria [3]. Mice deficient in cd T cells (TCRd 2/2 mice) have attenuated pulmonary clearance of these bacteria, likely as a result of loss of IL-17A production by cd T cells and consequent reduced neutrophil recruitment [4]. The number of cd T cells in the lung also increases under conditions associated with oxidative stress, including smoking, bleomycin instillation, and allergen challenge [5][6][7][8]. Moreover, the pulmonary inflammation induced by such agents requires cd T cells.
Inhalation of ozone (O 3 ), a common air pollutant, has a significant impact on human health. O 3 causes respiratory symptoms and reductions in lung function [9][10][11][12][13]. O 3 also increases the risk of respiratory infections and is a trigger for asthma [14][15][16]. Exposure to O 3 induces oxidative stress in the lung, damages lung epithelial cells, and causes the release of numerous cytokines and chemokines that recruit neutrophils and macrophages to the lung [9,17]. We have reported increased Il17a mRNA expression and increased numbers of IL-17A + cd T cells in the lungs after subacute O 3 exposure (0.3 ppm O 3 for 24-72 h) [18]. Hence, we tested the hypothesis that cd T cells, via their ability to produce IL-17A, are involved in orchestrating the inflammatory response to subacute O 3 exposure. We examined IL-17A expression in WT and TCRd 2/2 mice after exposure to air or to O 3 (0.3 ppm for 24-72 h). We also examined the effect of IL-17A neutralizing antibodies on O 3 -induced inflammation. Our results indicate an important role for IL-17A + cd T cells in the inflammatory cell recruitment induced by subacute O 3 exposure.
TNFa a pleiotropic pro-inflammatory cytokine, enhances the recruitment of neutrophils to the lungs in response to a variety of noxious stimuli, including LPS [19], cigarette smoke [20], and enterobacteria [21]. TNFais also required for neutrophil recruitment after subacute O 3 exposure [22,23]. However, TNFa does not have direct chemoattractant activity for neutrophils [24]. Instead, TNFa recruits neutrophils in part by inducing expression of other cytokines and chemokines [24,25]. In several pathological states, TNFa induces the expression of IL-17A [26,27]. Hence, we hypothesized that TNFa contributes to neutrophil recruitment following subacute O 3 exposure by promoting recruitment to or activation of IL-17A + cd T cells in the lungs. We used two methods to test this hypothesis. First, we assessed the effect of O 3 exposure on pulmonary Il17a expression and recruitment of IL-17A + cd T cells in WT mice and in mice deficient in TNFR2 (TNFR2 2/2 mice). Others have established that either TNFR1 or TNFR2 deficiency reduces the inflammatory response to subacute O 3 , and there is no further impact of combined TNFR1/TNFR2 deficiency [22]. Second, we examined the impact of the TNFa antagonist, etanercept, on Il17a expression. Our data suggest that TNFa is required for the recruitment of IL-17A + cd T cells to the lung after subacute O 3 exposure.
Approximately one third of the US population is obese and another third is overweight, but our understanding of how obesity impacts pulmonary responses to O 3 is still rudimentary. Such an understanding may have broad reaching implications since oxidative stress also contributes to responses to a variety of other noxious stimuli [5][6][7][8], many of which are affected by obesity [28,29]. In mice, the impact of obesity on responses to O 3 depends on the nature of the exposure: the pulmonary inflammation induced by acute O 3 exposure (2 ppm for 3 h) is augmented in all types of obese mice examined to date [30][31][32][33], whereas the pulmonary inflammation induced by subacute O 3 exposure (0.3 ppm for 24-72 h) is reduced [34]. Given our findings of the requirement for TNFa-recruitment of IL-17A producing cd T cells in the induction of pulmonary inflammation after subacute O 3, we sought to determine if changes in the activation of cd T cells might explain the reduced responses to subacute O 3 we observed in obese Cpe(carboxypeptidase E) fat mice. Data described below indicate that the reduced O 3 -induced neutrophil recruitment observed in obese mice is likely the result of reduced Il23 expression leading to reduced IL-17A + cd T cells. Given the importance of IL-17 + cd T cells for responses to viral and bacterial pathogens (see above), these observations might explain the altered response of the obese to bacteria and virus (see review by Peter Mancuso [35]).

Animals
This study was approved by the Harvard Medical Area Standing Committee on Animals. Male age-matched WT and TCRd 2/2 mice were either purchased from The Jackson Laboratory (Bar Harbor, ME) and acclimated for 4 weeks, or bred in house. Cpe fat mice are deficient in carboxypeptidase E, an enzyme involved in processing neuropeptides involved in eating behaviors [36]. The breeding strategy used to generate Cpe fat / TNFR2 2/2 mice from Cpe fat and TNFR2 2/2 mice (also originally purchased from The Jackson Laboratory) was previously described [37]. All mice were on a C57BL/6J background, fed a standard mouse chow diet, and were 10-13 weeks old at the time of study.

Protocol
For comparisons of WT and TCRd 2/2 mice, mice were exposed to O 3 (0.3 ppm) or to air, for 24-72 hours, as previously described [18]. Mice were exposed in normal cages without the microisolator top, but with free access to water and food throughout exposure. Mice were checked daily. At least two mice were placed in each cage to limit stress. After exposure, mice were euthanized with an overdose of sodium pentobarbital. The trachea was cannulated and bronchoalveolar lavage (BAL) was performed. After BAL, the lungs were flushed of blood by injecting 10 ml of cold PBS through the right ventricle, after creating a large excision in the left ventricle. One lung was excised and used for flow cytometry. The other was excised and placed in RNAlater (Qiagen, Germantown, MD) for preparation of RNA for real time PCR. In another cohort, WT mice were injected i.p. with 100 mg of anti-IL-17A neutralizing monoclonal antibody (Ab) (Rat IgG2A, clone 50104, MAB421; R&D Systems, Minneapolis, MN) or isotype control Ab (clone 54447, MAB006; R&D Systems) in 100 ml of sterile saline 24 hours before O 3 exposure. Mice were exposed to O 3 for 72 hours, euthanized, and tissues were harvested as described above. In a separate series of experiments, WT, TNFR2 2/2 , Cpe fat , and Cpe fat /TNFR2 2/2 mice were exposed to room air or O 3

Bronchoalveolar Lavage
BAL was performed and cells counted as previously described [18]. BAL supernatant was stored at 280uC until assayed. BAL KC, IL-6, MCP-1, IP-10 and G-CSF were measured by ELISA (R&D Systems). In mice treated with anti-IL-17A, BAL cytokines and chemokines were measured by multiplex assay (Eve Technologies, Calgary, Alberta). Total BAL protein was measured by Bradford assay (Bio-Rad, Hercules, CA).

Real-time PCR
RNA was extracted from lung tissue and prepared for qPCR using the SYBR method as previously described [18]. All expression values were normalized to 36B4 expression using the DDCt method. The primers for Il17a and 36B4 were previously described [37]. Primers for Ccl20, Il23 (p19) and Il12Rb1 are described in Table 1. For each set of primers, melt curve analysis yielded a single peak. Il12Rb1 expression was measured at baseline in order to tease apart the effects of genotype (deficiency of TNFa signaling versus sufficient signaling) versus O 3 exposure.

Statistical Analysis
Data were analyzed by factorial ANOVA using STATISTICA software (Statistica, StatSoft; Tulsa, OK) with mouse genotype and exposure as main effects. Fisher's least significant difference test was used as a post-hoc test. BAL cells and flow cytometry data were normalized by log transformed prior to analysis. A p value , 0.05 was considered significant.

IL-17A + cd T Cells are Increased by O 3 Exposure
Flow cytometry indicated that the number of IL-17A + CD45 + cells was significantly increased by O 3 in WT mice. This effect was ablated in TCRd 2/2 mice ( Fig. 2A). Further analysis indicated that in WT mice, the numbers of IL-17A + cd T cells as well as the total number of cd T cells were increased by O 3 (Fig. 2B, C), as reported previously using a similar gating strategy [18].

Effect of Anti-IL-17A Treatment
Compared to isotype control, anti-IL-17A treatment of WT mice caused a significant reduction in BAL neutrophils and macrophages (Fig. 3A). Anti-IL-17A treatment also significantly decreased BAL protein (Fig. 3B) and BAL G-CSF (Fig. 3C). Given this key role for IL-17A, these data indicate that the decreased inflammatory response observed in the TCRd 2/2 mice was likely due to the lack of Il17a expression (Fig. 1D) and demonstrate that G-CSF likely contributes to the effect of IL-17A on neutrophil recruitment.

Role of TNFa
BAL neutrophils were significantly lower in TNFR2 2/2 versus WT mice exposed to O 3 for 48 h (Fig. 4A), consistent with the results of Cho et al [22]. Similar results were obtained in WT mice treated with etanercept versus vehicle (Fig. 4D). O 3 exposure caused a significant increase in pulmonary Il17a expression in WT mice (Fig. 4B), consistent with results described above (Fig. 1D). However in TNFR2 2/2 mice, no such increase in Il17a mRNA abundance was observed (Fig. 4B). Similar results were obtained in mice treated with etanercept (Fig. 4E). Flow cytometry also indicated a decrease in IL-17A + CD45 + cells in O 3 -exposed TNFR2 2/2 versus WT mice (Fig. 5A). This change was due to decreased numbers of IL-17A + cd T cells (Fig. 5B). BAL G-CSF was also significantly lower in O 3 -exposed TNFR2 2/2 versus WT mice (Fig. 4C) and in etanercept treated versus vehicle treated WT mice (Fig. 4F).
The requirement of IL-23 and IL-6 for IL-17A expression in cd T cells [45,46], suggested that reductions in IL-17A + cd T cells in TNFR2 2/2 mice might be the result of loss of TNFa-induced expression of IL-23 or IL-6. O 3 increased BAL IL-6 in WT mice (Fig. 1F) and O 3 also increased pulmonary Il23 (p19) mRNA abundance (Fig. 6B), but neither IL-6 nor IL-23 were affected by TNFR2 deficiency or etanercept treatment (Fig. 6A, C). In contrast, TNFR2 2/2 mice had reduced expression at baseline of Il12Rb1 (Fig. 6H), a component of the IL-23 receptor. A similar trend was observed in etanercept treated mice (data not shown). O 3 exposure had no effect on Il12Rb1 (data not shown). Expression of the other component of the IL-23 receptor, Il23R, was not affected by TNFR2 deficiency (data not shown). To determine if TNFa was having direct effects on Il12Rb1expression on cd T cells, we isolated total lung cells from WT mice, stimulated them overnight with TNFa and examined IL-12Rb1 expression on cd T cells by flow cytometry (Fig. 6I,J). TNFa had no effect on the levels of IL-12Rb1 on cd T cells as measured by MFI and did not affect the percentage of cd T cells expressing IL-12Rb1, suggesting that other cells in the lung accounted for differences in Il12Rb1 mRNA expression.
We also considered the possibility that TNFa might impact the recruitment of cd T cells to the lung. In WT mice, O 3 exposure caused an increase in pulmonary mRNA expression of Ccl20 (Fig. 6E), a chemoattractant for IL-17A + cells [47,48], whereas no such increase was observed in mice treated with etanercept (Fig. 6F), suggesting that the role of TNFa is in the CCL20 dependent recruitment of IL-17 + cd T cells to the lungs. Similarly, there was a trend towards reduced Ccl20 mRNA abundance in O 3 -exposed TNFR2 2/2 versus WT mice (Fig. 6G), although the effect did not reach statistical significance.

Response to O 3 in Obese Mice
Cpe fat mice, regardless of their TNFR2 genotype or exposure, weighed almost twice as much as controls (data not shown). BAL neutrophils were significantly lower in Cpe fat versus WT mice exposed to O 3 (Fig. 4A,D), consistent with our previous observations using this exposure regimen [34]. In contrast to the  substantial reduction in BAL neutrophils observed in TNFR2 2/2 versus WT mice, TNFR2 deficiency had no significant effect on BAL neutrophils in O 3 -exposed Cpe fat mice (Fig. 4A). Similar results were obtained in etanercept treated WT mice (Fig. 4D). Cpe genotype had no impact on the number of BAL or lung macrophages (data not shown).
Il17a expression was significantly lower in O 3 exposed Cpe fat versus WT mice (Fig. 4B,E). The number of IL-17A + CD45 + cells was also significantly lower in O 3 -exposed Cpe fat than WT mice (Fig. 5A). The total number of cd T cells and the number of IL-17A + cd T cells was also reduced in the lungs of Cpe fat versus WT mice (Fig. 5B,C). O 3 -induced increases in BAL G-CSF were also  (C) total lung cd T cells. Results are mean6SEM for 3-6 air-exposed and 4-11 O 3 -exposed mice. *p,0.05 versus genotype-matched air-exposed mice. # p,0.05 versus WT mice with same exposure. doi:10.1371/journal.pone.0097707.g002 IL-17A, cd T Cells, TNFa, and Ozone PLOS ONE | www.plosone.org lower in Cpe fat versus WT mice (Fig. 4C, E) consistent with the reductions in IL-17A expression. Both BAL IL-6 and pulmonary Il23 mRNA expression were lower in Cpe fat versus WT mice (Fig. 6A, C,D). Reductions in these cytokines would be expected to reduce IL-17A expression, as observed (Fig. 4B, E). Whereas TNFR2 deficiency and etanercept reduced Il17a mRNA, IL-17A + cd T cells, and BAL G-CSF in lean WT mice, neither TNFR2 deficiency or etanercept affected these outcomes in obese Cpe fat mice (Fig. 4B,C and 5A-C).

Discussion
Our data indicate a key role for IL-17A + cd T cells in the pulmonary inflammation induced by subacute O 3 . Our data also indicate that TNFa promotes pulmonary inflammation after subacute O 3 by inducing recruitment of IL-17A + cd T cells, likely via Ccl20 expression. Finally, our data suggest that the attenuated pulmonary inflammation observed in obese mice after subacute O 3 is the result of reduced pulmonary IL-17A + cd T cells, consequent to reduced IL-23 and IL-6 expression.
Inflammatory cell recruitment to the lungs after subacute O 3 exposure required cd T cells (Fig. 1A,B). cd T cells have also been shown to be required for the pulmonary inflammation observed 24 but not 8 hours after acute exposure to much higher O 3 concentrations (2 ppm) [49,50], consistent with the time needed for recruitment and activation of cd T cells. However, in those studies, the precise role of these cd T cells was not assessed. Our data indicate that after exposure to lower concentrations of O 3 for much longer periods of time, the role of cd T cells involved IL-17A expression. Both lung Il17a mRNA and lung IL-17A + cd T cells increased after subacute O 3 exposure with a time course similar to that of neutrophil recruitment (Figs. 1A, 1D, 2B). Furthermore, O 3 -induced increases in Il17a mRNA abundance were abolished in TCRd 2/2 mice (Fig. 1D). In addition, both BAL neutrophils  and macrophages were reduced in mice treated with anti-IL-17A versus isotype control antibody (Fig. 3A). This ability of IL-17A + cd T cells to control the influx of macrophages and neutrophils is consistent with the findings in other models of lung infection and injury [4,[51][52][53][54]. While our data indicate that IL-17 + cd T cells are required for O 3 -induced inflammatory cell recruitment, they are not sufficient. For example, O 3 is highly reactive and macrophages and epithelial cells are the initial targets of its action. These cells are the likely source of TNFa which is required for neutrophil recruitment (Fig. 4) perhaps via induction of CCL20 and consequent recruitment IL-17A+ cd T cells (Figs. 5,6). Epithelial cells are also the likely source of CCL20. Furthermore, macro-phages also produce IL-17A after O 3 exposure [18], and the role of cd T cells may be to promote these effects. Macrophages and epithelial cells are also the likely source of other chemokines that interact with IL-17A (see below) to promote neutrophil recruitment.
IL-17A has direct chemoattractant effects on macrophages [55], which likely explains the ability of anti-IL-17A to attenuate O 3induced increases in BAL macrophages (Fig. 3A). In contrast, IL-17A induces neutrophil recruitment to the lungs by inducing expression of other neutrophil chemotactic and survival factors. With subacute O 3 exposure, G-CSF appears to be one of these factors. In WT mice, the time courses of induction of BAL G-CSF  and Il17a expression were similar (Fig. 1D,E). Importantly, anti-IL-17A and cd T cell deficiency each caused a marked and significant reduction in BAL G-CSF in O 3 exposed mice (Fig. 1E,  3C). The data are also consistent with our previous observations showing reductions in BAL G-CSF in O 3 -exposed adiponectindeficient mice treated with anti-IL-17A [18]. The observed role of IL-17A in G-CSF expression is in agreement with previous reports indicating that IL-17A signaling increases the transcription and stability of the Gcsf mRNA [56,57], via effects on ERK1/2 activation [58]. G-CSF causes neutrophil release from bone marrow and promotes neutrophil survival [59]. Since serum G-CSF did not increase after subacute O 3 exposure (data not shown), G-CSF is unlikely to act via effects on bone marrow in this model. Instead, G-CSF likely contributes by increasing the survival of neutrophils recruited to the lungs in response to other factors such as IP-10 (Fig. 1G).
TNFa is not directly chemotactic for neutrophils [24]. However, in lean WT mice, TNFR2 deficiency or the TNFa antagonist, etanercept, reduced the O 3 -induced increase in BAL neutrophils (Fig. 4A,D) consistent with previous reports [22,23,60] indicating a role for TNFa in neutrophil recruitment induced by subacute O 3 . TNFa also contributes to neutrophil recruitment in other conditions (reviewed in [61]), though the mechanism is not well understood. Our data suggest that at least in the setting of O 3 exposure, the ability of TNFa to recruit neutrophils involves IL-17A and that the source of this IL-17A is cd T cells (Fig. 5). O 3induced increases in pulmonary Il17a expression were attenuated in TNFR2 2/2 versus WT mice (Fig. 4B) and in etanercept versus vehicle treated WT mice (Fig. 4E). The number of IL-17A + cd T cells in the lung was also lower in TNFR2 2/2 versus WT mice exposed to O 3 (Fig. 5A,B). The ability of TNFa to promote pulmonary IL-17A expression after O 3 exposure is consistent with the role of TNFa in other pathogenic states. For example, etanercept reduces the elevated blood and skin Th17 cells observed in patients with psoriasis [26]. Similarly, another anti-TNFa therapy, infliximab, reduces IL-17A in ocular fluid from uveitis patients with Behcet's disease [27].
To better understand the role of TNFa, we examined IL-6 and IL-23 expression. Both these cytokines can contribute to induction of IL-17A in cd T cells [45,62]. Both IL-6 and IL-23 were induced in the lungs after O 3 exposure, but were not affected by TNFR2 deficiency or by etanercept (Fig. 6A,C,D), indicating that TNFa is not required for their expression. We did observe that mRNA expression of one of the two subunits of the IL-23 receptor, Il12Rb1, was decreased (Fig. 6H) in unexposed lungs from TNFR2 2/2 mice. Similar trends were observe after etanercept treatment (data not shown). Since others have reported that TNFa can act directly on cd T cells [40,63], we considered the possibility that TNFa was acting to increase Il12Rb1 expression on cd T cells, thus increasing their ability to respond to IL-23. However, culture of lung cells with TNFa resulted in no change in surface bound IL-12Rb1 on cd T cells (Fig. 6I,J). Instead, our data, suggest that effects of TNFa on Ccl20 expression (Fig. 6F,G) account for the observed effects of TNFa/TNFR blockade on IL-17A + cd T cells. Ccl20 acts via CCR6, a receptor expressed by IL-17A + cd T cells that promotes chemotaxis of these cells [64]. TNFa is also required for pulmonary Ccl20 expression after acute O 3 exposure (2 ppm for 3 h) [37]. A role for TNFa in Ccl20 expression has also been demonstrated in dermal lesions of psoriasis patients based on treatment with the TNFa antagonist infliximab [65].
We observed fewer neutrophils in BAL fluid of obese Cpe fat versus lean WT mice after subacute O 3 exposure (Fig. 4A,D), consistent with previous observations [34]. Reduced responses are observed in Cpe fat mice not only after 48 h exposure (Fig. 4A,D), but also after 24 or 72 h exposures [34]. Pulmonary Il17a expression and IL-17A + cd T cells were also reduced in the obese mice, as was the total number of cd T cells (Fig. 4). BAL G-CSF was also lower in Cpe fat versus lean WT mice (Fig. 4C,F). Moreover, O 3 -induced increases in BAL IL-6 and pulmonary Il23 expression were also reduced in Cpe fat versus WT mice (Fig. 6C,D). TNFR2 deficiency or etancercept treatment in Cpe fat mice did not further reduce BAL neutrophils or pulmonary Il17a expression, in contrast to what was observed in WT mice (Fig. 4B,E). Given the already reduced numbers of total cd T cells in Cpe fat mice exposed to O 3 (Fig. 5C), and our observations indicating the key role for IL-17A + cd T cells in the effects of TNFa on neutrophil recruitment, it is not surprising that TNFa had no further effect on the response to O 3 in obese mice. Taken together, the data suggest that obesity-related reductions in neutrophil recruitment induced by subacute O 3 exposure are the result of reduced IL-17A-dependent G-CSF release, consequent to reduced IL-6 and IL-23 expression. However, we cannot rule out the possibility that other factors also contributed. For example, neutrophils from obese mice exhibit reduced chemotactic activity towards CXCR2 ligands [66]. Such defects in neutrophil chemotaxis would also be expected to reduce O 3 -induced neutrophil recruitment in Cpe fat mice.
In addition to affecting responses to O 3 , obesity also impacts responses to bacterial and viral infections [67][68][69][70][71]. As described above, IL-17 + cd T cells contribute to neutrophil recruitment and pathogen clearance after certain bacterial infections [3,4]. IL-17 + cd T cells are also required for clearance of secondary infections after influenza [72]. Hence, obesity-related changes in IL-17 + cd T cells (Figs. 4b, 5a,b) may contribute not only to obesity-related alterations in responses to O 3 , but may have broader implications for effects of obesity on host defense. In support of this, obese mice compared to lean mice have fewer skin cd T cells number and the few cd T cells they have are dysfunctional [73], which leads to impairment in wound healing. These decreases in cd T cells numbers and impairment in function of the skin in obese mice are due to altered STAT5 signaling and chronic TNFa signaling [74].
In summary, our data indicate that cd T cells are required for the pulmonary inflammation that occurs after subacute O 3 exposure in mice via their ability to produce IL-17A. IL-17A then leads to G-CSF expression. Our data also indicate that TNFa is required for recruitment IL-17A + cd T cells to the lungs likely through its ability to induce Ccl20. These results emphasize the importance of cd T cells not only for pathogen clearance, but also for responses to other insults that induce oxidative stress, and describe a new role for TNFa in these events. Finally, our data indicate that obesity-related reductions in the ability of subacute O 3 to promote neutrophil recruitment to the lungs are the result of reduced IL-17A + cd T cells. These results suggest that other conditions that impact cd T cell recruitment or activation will also impact responses to this common pollutant.