IL-17 and TNF-α Are Key Mediators of Moraxella catarrhalis Triggered Exacerbation of Allergic Airway Inflammation

Alterations of the airway microbiome are often associated with pulmonary diseases. For example, detection of the bacterial pathogen Moraxella catarrhalis in the upper airways is linked with an increased risk to develop or exacerbate asthma. However, the mechanisms by which M. catarrhalis augments allergic airway inflammation (AAI) remain unclear. We here characterized the cellular and soluble mediators of M. catarrhalis triggered excacerbation of AAI in wt and IL-17 deficient as well as in animals treated with TNF-α and IL-6 neutralizing antibodies. We compared the type of inflammatory response in M. catarrhalis infected, house dust mite (HDM)-allergic and animals infected with M. catarrhalis at different time points of HDM sensitization. We found that airway infection of mice with M. catarrhalis triggers a strong inflammatory response with massive neutrophilic infiltrates, high amounts of IL-6 and TNF-α and moderate levels of CD4+ T-cell-derived IFN-γ and IL-17. If bacterial infection occurred during HDM allergen sensitization, the allergic airway response was exacerbated, particularly by the expansion of Th17 cells and increased TNF-α levels. Neutralization of IL-17 or TNF-α but not IL-6 resulted in accelerated clearance of M. catarrhalis and effectively prevented infection-induced exacerbation of AAI. Taken together, our data demonstrate an essential role for TNF-α and IL-17 in infection-triggered exacerbation of AAI.

Alterations of the airway microbiome are often associated with pulmonary diseases. For example, detection of the bacterial pathogen Moraxella catarrhalis in the upper airways is linked with an increased risk to develop or exacerbate asthma. However, the mechanisms by which M. catarrhalis augments allergic airway inflammation (AAI) remain unclear. We here characterized the cellular and soluble mediators of M. catarrhalis triggered excacerbation of AAI in wt and IL-17 deficient as well as in animals treated with TNF-α and IL-6 neutralizing antibodies. We compared the type of inflammatory response in M. catarrhalis infected, house dust mite (HDM)-allergic and animals infected with M. catarrhalis at different time points of HDM sensitization. We found that airway infection of mice with M. catarrhalis triggers a strong inflammatory response with massive neutrophilic infiltrates, high amounts of IL-6 and TNF-α and moderate levels of CD4 + T-cell-derived IFN-γ and IL-17. If bacterial infection occurred during HDM allergen sensitization, the allergic airway response was exacerbated, particularly by the expansion of Th17 cells and increased TNF-α levels. Neutralization of IL-17 or TNF-α but not IL-6 resulted in accelerated clearance of M. catarrhalis and effectively prevented infectioninduced exacerbation of AAI. Taken together, our data demonstrate an essential role for TNF-α and IL-17 in infection-triggered exacerbation of AAI.
Keywords: exacerbation of pulmonary inflammation, il-17, TnF-α, Moraxellaceae infections, infection and allergy, exacerbation of allergic reactions, pulmonary inflammation, microbial exacerbation of pulmonary inflammation inTrODUcTiOn Constant exposure of the respiratory mucosa to pollutants, allergens, and pathogens requires robust and at the same time strictly regulated immune responses of the airways. A breakdown of controlled airway immune responses to these environmental factors may result in acute or chronic inflammation, including asthma, and chronic obstructive pulmonary disease (COPD). Respiratory tract infections have emerged as the most frequent triggers for pulmonary inflammation in both children and adults (1). Pathogenic bacteria including Streptococcus pneumoniae, Hemophilus influenzae, and Moraxella catarrhalis are microbial colonizers of the airway mucosa inducing an inflammatory immune response that prepares the ground for increased asthma susceptibility or exacerbation of established disease M. Triggered Exacerbation Frontiers in Immunology | www.frontiersin.org November 2017 | Volume 8 | Article 1562 (2). Infection-triggered airway inflammation is often associated with increased amounts of inflammatory cytokines such as IL-6, IL-17, and TNF-α (3) and infiltrates of neutrophils, eosinophils, and different subtypes of T helper cells (4, 5) which seem to be related to the severity and pathogenesis of pulmonary inflammation (6). Paradoxically, many of these cytokines which induce pulmonary inflammation are also involved in antimicrobial host defense (7, 8), e.g., IL-17 and TNF-α mediate influx of neutrophils mediating first-line defense through uptake and killing of microbes. Although M. catarrhalis is a common pathogen known to trigger or exacerbate established pulmonary inflammation, the molecular and cellular mechanism still remains obscure (9). M. catarrhalis is able to efficiently adhere to the epithelium of distinct mucosal tissues such as lung and nasopharynx. Together with S. pneumoniae and H. influenzae, M. catarrhalis is one of the major pathogens causing otitis media (OM) in children. Furthermore, several studies have reported that M. catarrhalis contributes to the exacerbations of COPD in adults (10). During the course of infection, M. catarrhalis induces a robust inflammatory response characterized by infiltrations of macrophages, lymphocytes, and neutrophils into infected tissue, which probably causes the pathogenesis of OM and also leads to exacerbations of COPD. The investigation of the molecular interplay between M. catarrhalis and host immune system has revealed that the recognition of this pathogen by multiple toll-like receptors (TLRs) such as TLR4 and TLR9 triggers the production of pro-inflammatory cytokines IL-6 and TNF-α (11). In many asthma patients, M. catarrhalis is one of the dominant pathogen species found within the airway bacterial community (12). Although a link between infection with M. catarrhalis and exacerbation of asthma has been proposed, no functional data exist to our knowledge, indicating that this pathogen is causative for exacerbation of this disease.
Our aim was to investigate whether M. catarrhalis infection has the potential to exacerbate allergic airway inflammation (AAI) and if yes, to analyze the underlying pathomechanisms. Therefore, we established for the first time a murine model of M. catarrhalis airway infection and rigorously analyzed the mechanisms of pulmonary inflammation triggered exclusively by the pathogen or during developing or established AAI against the house dust mite (HDM) allergen.
We here show that airway infection with M. catarrhalis augments the phenotype of AAI mainly via TNF-α and T-cell derived IL-17 but not IL-6. Yet, in the absence of IL-17 and/or TNF-α, infection-triggered lung inflammation as well as exacerbation of AAI was very mild and resulted in accelerated clearance of pathogenic airway bacteria, emphasizing the inflammation promoting role of these cytokines.

characteristics of airway inflammation caused by M. catarrhalis infection
Before investigating the consequences of bacterial airway infection on AAI, we established a murine infection model to first study the immune responses triggered by M. catarrhalis. Two hours after intranasal inoculation, infection was fully established and bacterial titers of the lung dropped massively at day 2 p.i. and were below detection at day 6, revealing clearance of M. catarrhalis in normal animals ( Figure 1A). It has to be noted that approx. 50% of wt animals intranasally infected with M. catarrhalis showed moribund symptoms at day 3 p.i. (Figure 2G). Total cell numbers in bronchoaleveolar lavage (BAL) increased 1 day after intranasal inoculation, peaked at day 4, and then decreased. M. catarrhalis infection caused airway inflammation primarily through massive influx of neutrophils and lower numbers of eosinophils and macrophages reaching their maximum at day 4, followed by an increase in lymphocytes which peaked at day 7 p.i. Twelve days after infection, all BAL cells numbers returned to baseline level (Figures 1B,C). Accordingly, early and strong expression of CXCL1 and CXCL10 in the BAL was observed, both known to attract neutrophils and other inflammatory cells ( Figure 1D). As M. catarrhalis infection causes airway inflammation (9, 13), we performed a kinetic study of inflammatory cytokines in the BAL. While IL-6, TNF-α and IL-1β peaked at day 1 p.i. and rapidly decreased to low levels by day 4, the amounts of IFN-γ and IL-17 were highest at day 7 p.i., coinciding with the peak of BAL lymphocytes ( Figure 1E). Due to the late occurrence of IL-17 in the BAL, we continued to analyze the mRNA expression of this cytokine in lung homogenates. In accordance with BAL data, strong IL-17 mRNA expression was found at day 7, correlating with the rise of IL-17 + CD4 + T cells at this time point (Figures 1F,G). To identify the type of T cells producing pulmonary IL-17 after M. catarrhalis infection, lung lymphocytes were analyzed by flow cytometry. The majority of IL-17 secreting cells in the lung were conventional CD4 + T cells (70%) which increased between days 7 and 15 p.i. and only a minor fraction of TCR γ + δ + cells (14%) was positive for IL-17 ( Figure 1H). We next investigated inflammatory responses in IL-17 deficient mice. Interestingly, despite similar initial bacterial colonization of the lungs, IL-17 KO mice showed increased bacterial clearance as compared to wt animals (Figure 2A). Furthermore, these mice revealed significantly reduced total and differential BAL cell counts (Figures 2B,C) and lung tissues showed a massive reduction of cellular infiltrates as compared to wt animals ( Figure 2D). The amounts of IL-6 and TNF-α in BAL and serum of infected IL-17 KO were diminished and mice did not succumb to M. catarrhalis infection wt animals (Figures 2E-G). Interestingly, despite the lack of IL-17, the number of neutrophils in the lung at day 1 p.i. was comparable to wt animals. Similar amounts of the CXCL1, 5, and 10 chemokines in IL-17 KO mice might compensate for early neutrophil attraction, independently in IL-17 ( Figure S1 in Supplementary Material). Likewise, neutralization of TNF-α, which is also produced by epithelial cells upon M. catarrhalis infection ( Figure S2 in Supplementary Material), resulted in reduction of local and systemic IL-1β, IL-6, and IL-17 and subsequent protection against M. catarrhalis infection (Figures 3A-C). Interestingly, in contrast to IL-17 and TNF-α, we found that neutralization of IL-6 increased local and systemic TNF-α as well as early IL-1β and IL-17 levels in the BAL. Anti-IL-6 treatment resulted in slightly enhanced numbers of animals with moribund symptoms, suggesting that IL-6 exerts an anti-inflammatory function during pulmonary M. catarrhalis infection (Figures 4A-C).

Pulmonary infection with Moraxella catarrhalis exacerbates hDM-induced aai
To study the mechanisms of M. catarrhalis caused exacerbation of AAI, wt C57Bl/6 mice (white bars) were intranasally infected with M. catarrhalis after the second exposure with HDM and analyzed on day 23 as shown in Figure 5A. While HDM exposure induced airway inflammation, additional infection of HDM sensitized wt mice with M. catarrhalis led to hugely increased numbers of leukocytes in the BALF mainly through infiltrating neutrophils, eosinophils, and lymphocytes ( Figures 5B,C). This was also evident in lung histology, showing increased airway inflammation, goblet cell hyperplasia, and mucus production in M. catarrhalis infected, HDM-allergic mice (Figures 5D-F).
To further study the cytokine pattern of pulmonary CD4 + T-cell populations, we compared HDM allergic mice with those that additionally contracted an infection. The latter group revealed significantly increased frequencies of IFN-γ, IL-17A, and IL-5 + / IL-13 + double positive CD4 + T cells, although the absolute frequencies of IL-5 + /IL-13 + Th2 cells were low as compared to those of IFN-γ + and IL-17 + CD4 + T cells ( Figure 5G; Figure S5A in Supplementary Material). Accordingly, increased amounts of IL-17 and IFN-γ were found in the lungs of HDM allergic wt mice (white bars) concomitantly infected with M. catarrhalis ( Figure  S5B in Supplementary Material). Of note, cellular infiltrates and cytokine responses measured at day 23 were significantly lower than during acute infection with M. catarrhalis. We then wondered whether the time of M. catarrhalis infection (prior to, during or after complete HDM sensitization) has an influence on exacerbation of AAI. Similar to infection during HDM sensitization, bacterial infection after established allergic reactions increased total BAL leukocyte counts mainly by neutrophils and lymphocytes. Together with increased lymphocytic infiltration, airway infection enhanced the frequencies of lung IL-17, IFN-γ, and IL-5 secreting CD4 + T cells as compared to with the HDM-only group ( Figure S3 in Supplementary Material). Furthermore, we wondered whether a previously resolved

il-17 is a central Mediator of infection-Triggered exacerbation of aai
Since IL-17 has been reported to be associated with mucoepithelial infections and allergic responses (14), we next tested IL-17 deficient mice for the development of HDM-allergic airway responses. Interestingly, wt animals (white bars) and IL-17 KO mice (black bars) revealed similar numbers of neutrophils, eosinophils, and lymphocytes in the BALF (Figures 5B,C). Lung histology revealed slightly reduced pulmonary infiltrates in IL-17 KO mice as compared to with wt mice after HDM exposure ( Figure 5D). Accordingly, airway inflammation, goblet cell hyperplasia, and mucus production were marginally diminished in IL-17 KO mice as compared to with controls ( Figures 5E,F).
As early antibacterial immunity is mediated by IL-17controlled influx of neutrophils into infected tissues (15), we next investigated the role of this cytokine to trigger infection-induced exacerbation of AAI. IL-17 KO and wt mice were infected during the second HDM allergen exposure and analyzed at day 23, as shown in Figure 5A. The data demonstrate that in contrast to wt animals, the number of neutrophils, eosinophils, and lymphocytes in the BALF of infected IL-17 KO mice was comparable to the non-infected, HDM group. Furthermore, comparable numbers of mucus secreting cells and similar inflammation scores suggest IL-17 as key cytokine of M. catarrhalis-induced exacerbation of AAI (Figures 5D-F)   anti TnF-α Treatment attenuating il-17-Dependent exacerbation of infection-induced aai It has been shown that neutrophil recruitment during inflammation is mediated by synergistic effects of IL-17 and TNF-α in endothelial cells by enhancing the expression of P-and E-selectin (16). We thus wondered whether neutralization of TNF-α also attenuates infection-induced exacerbation of AAI. Briefly, HDM sensitized mice were treated only once with anti-TNF-α mAb 4 h before intranasal infection with M. catarrhalis and analyzed at day 23, as depicted in Figure 6A. Control groups included mice that received phosphate-buffered saline (PBS) or HDM alone as well as infected, HDM sensitized animals.
Differential cell counts of BAL cells revealed that predominantly neutrophils were reduced by anti-TNF-α treatment, while the numbers of macrophages were increased when compared with HDM/infected controls. The increase of BAL macrophages after TNF-α neutralization seems to compensate for the low number of neutrophils in order to eliminate M. catarrhalis. Interestingly, the number of eosinophils or lymphocytes were not significantly reduced after TNF-α treatment as shown in Figures 6B,C. Lung histology confirmed that anti-TNF-α treatment vastly reduced infection-triggered inflammatory infiltrates in HDM allergic animals which was also reflected by the decreased number of PAS + cells and the low-inflammation score (Figures 6D-F). We next wondered whether TNF-α neutralization also affects pulmonary T-cell activation. Anti-TNF-α treatment attenuated the activation of IFN-γ + and IL-17 + lymphocytes but had no influence on the low frequency of IL-5/IL-13 secreting, pulmonary CD4 + T cells ( Figure 6G; Figure S6 in Supplementary Material).
In summary, the experiments revealed that a single application of anti-TNF-α mAb is sufficient to markedly attenuate M. catarrhalis-induced exacerbation of AAI.

DiscUssiOn
We here show that infection with M. catarrhalis exacerbates HDM-triggered AAI mainly by an IL-17-and TNF-α-dependent inflammatory response. Lack of IL-17 or neutralization of TNF-α prevented infection-triggered exacerbation of allergic pulmonary disease.
Infection with M. catarrhalis has been shown to be an important risk factor for newborns or patients with chronic pulmonary diseases to either develop asthma or to exacerbate disease symptoms (17). While evidence suggests that the frequency of respiratory infections rather than the type of pathogen influences asthma development (15), the immunological mechanisms of infectiontriggered exacerbation of AAI are not entirely understood, which is reflected by the limited therapeutic options. Infection with M. catarrhalis during HDM allergen exposure strongly increased the influx of neutrophils, eosinophils, and T cells, while infection after established allergic inflammation caused exacerbation primarily by neutrophils. In contrast, infection prior to HDM exposure predominantly triggered the expansion of pulmonary T-lymphocyte populations. However, common to all three settings of infection is the induction of a mixed pulmonary Th1, Th2, and Th17 immune response. Thus, our data suggest that bacterial infection influences the development and outcome of AAI irrespective of the time-point of allergen exposure. This reflects the situation in patients where viral and/or bacterial infections in early life or during asthma have been associated with the development or exacerbation of the disease, respectively (18).
The importance of IL-17 in protection against pulmonary bacterial infection was shown for several pathogens, including Klebsiella pneumoniae, Mycoplasma pneumoniae, Bordatella pertussis, and Mycobacterium tuberculosis (19). Accordingly, IL-17 receptor KO mice revealed increased susceptibility to bacterial infections due to impaired neutrophil and macrophage recruitment to the site of bacterial entry (20). On the other hand, the expression of IL-17 in the airways of asthma patients was shown to correlate with neutrophilic lung inflammation (21)(22)(23). Despite this ambiguous role, our data revealed that IL-17 deficient animals cleared the pathogen even faster than control littermates. Unimpaired expression of CXCL1, 5, and 10 chemokines in IL-17 KO mice seems to compensate for early neutrophil influx but not for later stages of infection where the pulmonary neutrophil response was decreased in KO animals.
Infection of human respiratory tract epithelial cells with M. catarrhalis has been shown to increase various pro-inflammatory genes, e.g., TNF-α, IL-1β, and IL-17 (10). We observed a similar cytokine response in M. catarrhalis infected wt but not IL-17 KO mice which showed even increased resistance against infection. Furthermore, in the absence of IL-17, the mixed neutrophilic/eosinophilic airway inflammation of HDM challenged wt mice switched to an attenuated, almost exclusive eosinophilic response. Likewise, M. catarrhalis infection of HDM allergic mice did not lead to exacerbation of pulmonary inflammation in IL-17 deficient mice. Taken together, these findings indicate that pulmonary infection is a crucial trigger for IL-17-induced, exacerbation of neutrophilic inflammation. Besides IL-17, IL-6 has been considered as promising target to reduce pathogenesis of asthma and COPD, particularly those of eosinophilic/neutrophilic phenotype (24,25). IL-6 is known to mediate pro-inflammatory and anti-inflammatory immune functions via trans-and classical signaling, respectively (26). As anti-IL-6 treatment led to enhanced secretion of local and systemic TNF-α with high susceptibility to M. catarrhalis infection, we did not consider IL-6 as appropriate target for treatment.
The contradicting reports about efficiency and safety of TNF-α blockers in animal models and patients with severe asthma (27,28) prompted us to study the effect of TNF-α neutralization during pulmonary infection and exacerbation of AAI.
We found that neutralization of TNF-α prior to pulmonary M. catarrhalis infection resulted in a substantial decline of IL-6, Il-1β, and IL-17, together with increased survival of infected animals. Furthermore, TNF-α neutralization was able to effectively reduce infection-induced exacerbation of HDM-allergic reactions despite minor or no reduction of pulmonary Th1, Th17, and Th2 cells, respectively. Even though we found a strong decrease of neutrophils in the BAL, the mechanisms by which TNF-α acts in asthmatic airways might be manifold: Besides its effect to recruit neutrophils and eosinophils (29), it has been shown to enhance T-cell responses (30), to promote glucocorticoid resistance and airway remodeling (31).
While it has been shown that TNF-α or IL-17 alone are able to induce inflammatory responses, inflammation was enhanced in the presence of both cytokines (32,33).
In line with these observations, our data demonstrate that neutralization of IL-17 or TNF-α is similar effective in attenuating exacerbation of AAI. Whether neutralization of these cytokines is also suitable for therapy still remains to be determined. Since

Bacteria and infection
A patient isolate of M. catarrhalis was stored at −80°C in glycerol. Bacteria from frozen stocks were grown aerobically in Columbia blood agar plates (BBL; Becton Dickinson, MD, USA) with 5% sheep blood at 37°C overnight, washed off the plate and resuspended in sterile PBS for infection. Anesthetized mice were inoculated intranasally (i.n.) with 2 × 10 8 CFU M. catarrhalis resuspended in 50-µL PBS, before during or after HDM challenge.

Determination of cFU in Organs
Lungs were aseptically removed and homogenized in 1 mL of sterile PBS. Serial dilutions of BAL fluid and lung homogenates were prepared in sterile PBS, plated on Columbia blood agar plates (BBL; Becton Dickinson, MD USA) with 5% sheep blood and M. experimental Model of acute hDM aai 8-10-week-old C57Bl/6 mice were slightly anesthetized (Ketamin/2%Rompun, i.p.) and weekly challenged by intranasal (i.n.) administration of 100-µg HDM extract (GREER, Lenoir, USA) in 50-µL PBS on days 0, 7, 14, and 21. Two days after last HDM challenges mice were sacrificed and blood, bronchoalveolar lavage (BAL) fluids, lungs were collected for further analysis.

administration of neutralizing antibodies
To assess the role of IL-6 and TNF-α during the infection with M. catarrhalis, neutralizing anti-IL-6 (clone MP5-20F3) or anti-TNF-α (clone MP6-XT22) antibodies were used. Rat IgG1 isotype control was applied for treatment of control mice. Four hours before M. catarrhalis infection, mice were treated intraperitoneally once with 100 µg of neutralizing antibodies, respectively. All used antibodies were generated and purified by Manuela Staeber (Max Planck Institute for Infection Biology, Berlin, Germany).

Bronchoaleveolar lavage
Bronchoaleveolar lavage was performed 48 h after the last challenge using 1-mL cold PBS supplemented with a protease inhibitor cocktail (Roche, Mannheim, Germany). An automated Casy TT cell counter (Schaerfe System, Reutlingen, Germany) was used to determine the total leukocyte cell counts. Cells were centrifuged and the cell-free supernatant was stored at −20°C until cytokines were measured by Elisa. For differential cell counts, Cytospin preparations were fixed and then stained with Diff-Quick (Merz &Dade AG, Dudingen, Switzerland). Macrophages, lymphocytes, eosinophils, and neutrophils were identified by standard morphologic criteria and 300 cells were counted per cytospin. The measurement of cytokines IFN-γ, IL-17, and chemokines (CXCL1, CXCL5, and CXCL10) was performed by Cytometric Bead Array (BD, Heidelberg, Germany) according to manufacturer's protocol.

histology of the airways
Directly after BAL, lungs were fixed with 10% formalin via the trachea, removed, and stored in 10% formalin. Lung tissues were embedded into paraffin. Tissues were cut in 3-µm sections and stained using hematoxylin and eosin (HE) as well as periodic acid Schiff (PAS). Goblet hyperplasia was assessed by cell counting and expressed as the number of goblet cells per 100-µm basement membrane, according to Foster et al. (2000).

Measurement of hDM-specific antibodies in serum samples
The amount of HDM-specific IgG1 and IgG2c antibodies was measured by the ELISA. Maxi-Sorb plates were coated with 50-µg/ mL HDM extract in bicarbonate buffer (pH 9.6) overnight at 4°C. Subsequently, coated wells were blocked with 1% w/V BSA in PBS for 2 h at RT. After washing, serum samples (1:100) were incubated overnight at 4°C, washed, and incubated with the corresponding biotin-labeled anti-immunoglobulin antibody overnight at 4°C.
Plates were washed and incubated with streptavidin-peroxidase for 30 min at RT. After development with substrate, the reaction was stopped using H2SO4 and ODs were read at 450 nm.

Quantitative analysis of cytokine expression
Expression of IFN-γ and IL-17 mRNA was determined by quantitative real-time polymerase chain reaction (RT-PCR) analysis using LightCycler technology (Roche). RNA was extracted from lung tissues according to manufacturer's instructions. cDNA synthesis was performed after removal of contaminating genomic DNA by DNase treatment using Superscript reverse transcriptase (Invitrogen, Karlsruhe, Germany) as described by manufacturer. Quantitative PCR (QPCR) was performed by using the QuantiTect SYBR Green PCR kit (QIAgen) and primers specific for IFN-γ