sign in sign up
<<
Home , Streptococcus pneumoniae

Suppresses NADPH Oxidase−Dependent Phagocytic Bacterial Clearance and Enhances Susceptibility to Secondary Methicillin-Resistant This information is current as Infection of September 30, 2021. Keer Sun and Dennis W. Metzger J Immunol 2014; 192:3301-3307; Prepublished online 21 February 2014; doi: 10.4049/jimmunol.1303049 Downloaded from http://www.jimmunol.org/content/192/7/3301

Supplementary http://www.jimmunol.org/content/suppl/2014/02/21/jimmunol.130304 http://www.jimmunol.org/ Material 9.DCSupplemental References This article cites 50 articles, 20 of which you can access for free at: http://www.jimmunol.org/content/192/7/3301.full#ref-list-1

Why The JI? Submit online.

by guest on September 30, 2021 • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Influenza Infection Suppresses NADPH Oxidase–Dependent Phagocytic Bacterial Clearance and Enhances Susceptibility to Secondary Methicillin-Resistant Staphylococcus aureus Infection

Keer Sun and Dennis W. Metzger

Methicillin-resistant Staphylococcus aureus (MRSA) has emerged as a leading contributor to mortality during recent influenza pandemics. The mechanism for this influenza-induced susceptibility to secondary S. aureus infection is poorly understood. In this study, we show that innate antibacterial immunity was significantly suppressed during the recovery stage of influenza infection, even though MRSA superinfection had no significant effect on viral burdens. Compared with mice infected with alone, postinfluenza MRSA–infected mice exhibited impaired bacterial clearance, which was not due to defective recruitment, Downloaded from but rather coincided with reduced intracellular reactive oxygen species levels in alveolar macrophages and neutrophils. NADPH oxidase is responsible for reactive oxygen species production during phagocytic bacterial killing, a process also known as oxidative burst. We found that gp91phox-containing NADPH oxidase activity in macrophages and neutrophils was essential for optimal bacterial clearance during respiratory MRSA . In contrast to wild-type animals, gp91phox2/2 mice exhibited similar defects in MRSA clearance before and after influenza infection. Using gp91phox+/2 mosaic mice, we further demonstrate that http://www.jimmunol.org/ influenza infection inhibits a -intrinsic contribution of NADPH oxidase to phagocyte bactericidal activity. Taken together, our results establish that influenza infection suppresses NADPH oxidase–dependent bacterial clearance and leads to susceptibility to secondary MRSA infection. The Journal of Immunology, 2014, 192: 3301–3307.

iral influenza is often a seasonal infection that can lead to shown to exacerbate inflammation (15–18), even though primary influenza . However, secondary bac- there is no evidence linking its detrimental effect to enhanced en- V terial infections, commonly associated with zymatic activity. In contrast, during S. aureus lung infection alone,

pneumoniae, Staphylococcus aureus,andHaemophilus influenzae, phagocyte ROS generation is essential for antibacterial immunity by guest on September 30, 2021 are known to be more frequent causes of severe morbidity and (19, 20). However, the outcome of these conflicting contributions of mortality (1). Among these, methicillin-resistant S. aureus (MRSA) NADPH oxidase to immune defense against influenza and S. aureus has emerged as the leading contributor to mortality during recent coinfection has not been addressed in any reported studies. Instead, influenza pandemics and epidemics (2–7). A dysregulated immune several reports have demonstrated that synergistic influenza and defense against either influenza virus or S. aureus has been pro- S. aureus coinfection is associated with dysregulation of antibac- posed as a contributor to coinfection pathogenesis (8–11); how- terial immunity (8–11). Therefore, we sought to determine the reg- ever, an incomplete understanding of coinfection ulatory effect of influenza infection on ROS-dependent antibacterial has slowed the development of effective treatments (12–14). activity and its contribution to susceptibility to subsequent S. aureus NADPH oxidase produces superoxide, which can spontaneously infection. In our model, we found that innate antibacterial immunity form that will undergo further reactions to was significantly suppressed following influenza infection. In ad- generate reactive oxygen species (ROS). During influenza virus dition, mice were more susceptible to respiratory MRSA infection infection alone, gp91phox-containing NADPH oxidase has been in the absence of NADPH oxidase activity. Importantly, we show that influenza infection inhibited the cell-intrinsic contribution of NADPH oxidase to phagocytic bacterial killing, which led to sus- Center for Immunology and Microbial Disease, Albany Medical College, Albany, ceptibility to secondary MRSA infection. NY 12208 Received for publication November 12, 2013. Accepted for publication January 21, 2014. Materials and Methods This work was supported by a Pfizer Young Investigator award, National Institutes of Murine model of viral and bacterial infection Health/National Heart, Lung, and Blood Institute Grant R01 HL118408 (to K.S.), and National Institutes of Health/National Institute of Allergy and Infectious Diseases Specific -free, C57BL/6 wild-type (WT), gp91phox2/y, gp91phox2/2, Grants R01 AI41517 and R01 AI75312 (to D.W.M.). p47phox2/2, NOS22/2, and “Mafia” mice were initially purchased from Address correspondence and reprint requests to Dr. Keer Sun at the current address: The Jackson Laboratory (Bar Harbor, ME) and bred at Albany Medical Department of Pathology and Microbiology, University of Nebraska Medical Center, College following Institutional Animal Care and Use Committee guide- Omaha, NE 68198-5900. E-mail address: [email protected] lines. The online version of this article contains supplemental material. Viral challenge was performed with A/PR/8 or H1N1 CA4 administered Abbreviations used in this article: BAL, bronchoalveolar lavage; BALF, bronchoal- intranasally (i.n.) to anesthetized and sex- and age-matched adult mice in veolar lavage fluid; i.n., intranasal(ly); MRSA, methicillin-resistant Staphylococcus 50 ml sterile PBS. Unless otherwise specified, bacterial coinfection was aureus; ROS, reactive oxygen species; WT, wild-type. performed at 7 d after influenza infection. To induce , anesthetized mice were inoculated i.n. with 50 ml PBS containing 107–2 3 Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 108 CFU community-acquired MRSA strain BAA-1695 (an isolate from www.jimmunol.org/cgi/doi/10.4049/jimmunol.1303049 3302 INFLUENZA INFECTION INHIBITS PHAGOCYTIC BACTERIAL KILLING patient ) or MRSA BAA-1692 (an isolate from human sinus) ob- AP20187 was a gift from Ariad Pharmaceuticals. Lyophilized AP20187 tained from the American Type Culture Collection. Titers of virus stocks was dissolved in 100% ethanol at a concentration of 13.75 mg/ml. A and viral levels in the bronchoalveolar lavage fluids (BALF) and of working solution for peritoneal injections was prepared from the stock infected mice were determined by plaque assays on MDCK cell mono- diluted to 0.55 mg/ml in 4% ethanol, 10% polyethylene glycol-400, and layers (21). Bacterial burdens in the lungs and BALF were measured by 1.7% Tween 20. Mafia mice were injected i.p. for 5 consecutive days with sacrificing infected mice at the indicated time points and plating serial 10- 300 ml AP20187 or diluent control. The mice were then rested for another fold dilutions of each sample onto blood agar plates. 6 d before i.n. infection with S. aureus. The efficiency of pulmonary macrophage depletion was confirmed by flow cytometry (24). Bronchoalveolar lavage cell analysis Statistical analyses Bronchoalveolar lavage (BAL) was collected by making an incision in the Results are expressed as means 6 SD. Significant differences between and gavaging the lung twice with 0.8 ml PBS (pH 7.4). The BAL experimental groups were determined using a Student t test (to compare g cells were incubated with2.4G2mAbagainstFc RII/III and stained with PE- two samples) or an ANOVA analysis, followed by Tukey’s multiple or allophycocyanin-conjugated anti-CD11c (Caltag Laboratories, Burlingame, comparisons test (to compare multiple samples) in GraphPad Prism 6 (La CA), FITC- or PE-Cy7–conjugated anti-CD11b (BD Biosciences), FITC- Jolla, CA). Survival analyses were performed using the Kaplan–Meier log- or PE-conjugated anti-Ly6G mAb (clone 1A8; BD Biosciences), and APC- rank test. For all analyses, p , 0.05 was considered to be significant. Cy7–conjugated anti-Ly6C mAb (BD Biosciences). The stained cells were analyzed on a BD FACSCanto using BD FACSDiva software. Results Detection of gp91phox WT and deficient neutrophils Influenza infection impairs innate defense against a subsequent gp91phox expression in neutrophils was assessed with an anti-mouse S. aureus respiratory challenge gp91phox mAb (BD Biosciences) (22). Briefly, after staining for neu-

To establish a mouse model that mimics the clinical observation Downloaded from trophils with FITC-conjugated anti-mouse Ly6G mAb and PE-Cy7–con- that influenza infection predisposes individuals to S. aureus in- jugated anti-mouse CD11b mAb, cells in suspension were fixed with 2% paraformaldehyde for 20 min. Cells were then washed with PBS and fection, we followed an experimental protocol similar to our pre- permeabilized with BD Perm Buffer IV for 20 min. Cells were washed vious secondary studies (25, 26). Specifically, twice with 2% BSA in PBS and then incubated with purified mouse anti- C57BL/6 mice were infected with a sublethal dose of H1N1 CA04 phox gp91 mAb for 30 min. Cells were washed and incubated with PE- influenza virus and followed by i.n. challenge with 107 CFU MRSA conjugated rat anti-mouse IgG1 Ab (BD Biosciences) prior to analysis BAA-1692 at days 5, 7, or 9 after viral infection. Coinfected mice by flow cytometry. http://www.jimmunol.org/ were sacrificed 24 h after MRSA infection (i.e., days 6, 8, and 10 Flow cytometry analysis of in BALF cells postinfluenza infection) for determination of lung viral and bacterial Seven days after i.n. inoculation of 50 PFU PR8 influenza virus, mice were burdens. Compared with animals infected with MRSA only, influenza infected i.n. with 107 CFU MRSA, and 24 h later, BALF cells were col- and MRSA-coinfected mice exhibited decreased ability to clear bac- lected and incubated with CellROX Deep Red reagent (Invitrogen) at teria from their lungs (Fig. 1A). However, the relative defect in bac- a final concentration of 5 mM in complete RPMI 1640 medium for 30 min terial clearance was significantly exacerbated in mice superinfected at 37˚C, washed three times with PBS, and then fixed with 3.5% formalde- with MRSA at day 7 postinfluenza as compared with day 5 super- hyde for 15 min before staining for surface marker expression (Supplemental Fig. 1). BALF cells incubated with RPMI 1640 medium alone were used for infected animals (Fig. 1B), indicating greater susceptibility to sec-

background staining of individual cell subsets. ondary S. aureus infection at the beginning of the viral recovery phase by guest on September 30, 2021 (Fig. 1C). Neutrophil depletion during MRSA infection To determine the possible strain-specific effect of influenza virus Anti–Gr-1 mAb was purified from culture supernatants of the RB6-8C5 or S. aureus on coinfection outcomes, mice were next infected hybridoma using anti-rat IgG–agarose columns (Sigma-Aldrich, St. Louis, with PR8 virus and superinfected with the community-acquired MO). For neutrophil depletion, mice were injected i.p. 48 and 24 h before MRSA strain BAA-1695 on day 7 postinfluenza. Consistent with bacterial infection with 0.1 mg RB6-8C5 anti–Gr-1 mAb or with rat IgG as a control. The efficiency and specificity of neutrophil depletion in BALF of the observations above, bacterial burdens were significantly in- MRSA-infected mice was confirmed by flow cytometry analysis (Supplemental creased in PR8-infected mice (Fig. 2A). Interestingly, MRSA Fig. 2). superinfection did not appear to affect viral loads in the lung (Fig. 2B). Accordingly, coinfection was associated a high mor- Alveolar macrophage depletion during MRSA infection tality rate (.90%), after an increased (2 3 108 CFU) MRSA chal- Mafia transgenic mice have an inducible Fas suicide/apoptotic system lenge dose, which was not observed with either infectious agent driven by the mouse colony stimulating factor-1 receptor promoter (23). alone (Fig. 2C). Collectively, these results indicate that anti-MRSA The transgene insert contains a mutant human FK506 binding 1A, which preferentially binds the dimerization drug AP20187. Administration immunity is profoundly inhibited at the recovery stage of influenza of AP20187 induces apoptosis specifically in macrophages and dendritic infection, which is correlated with a lethal outcome following cells. influenza and S. aureus coinfection.

FIGURE 1. Influenza enhances susceptibility to MRSA infection during the recovery stage of viral infection. Bacterial burdens (A), influenza-induced bacterial outgrowth (B), and viral burdens (C) in the lungs at various days after 3 3 103 PFU H1N1 CA04 influenza virus infection of C57BL/6 (B6) WT mice and 24 h postinfection with 107 CFU/mouse of MRSA BAA-1692 (five to seven mice per group). In (B), the relative increases of bacterial burdens in influenza-infected mice 24 h after MRSA infection are represented as fold increases relative to mean lung bacterial CFUs in mice infected with MRSA alone (A) included at each time point. p , 0.001, ANOVA (A, C). p , 0.01, ANOVA (B). ***p , 0.001, Tukey’s multiple comparisons test. The Journal of Immunology 3303

FIGURE 2. Influenza infection impairs innate immunity against respiratory MRSA challenge. Bacterial (A) and viral burdens (B) in the lungs 24 h after inoculation of naive (MRSA) or day 7 postinfluenza–infected (Flu+MRSA) C57BL/6 mice with 107 CFU MRSA BAA-1695 (Flu+MRSA) or PBS (Flu). ***p , 0.001 compared with mice not infected with influenza virus, unpaired t test. (C) Survival of C57BL/6 mice postinfection with 2 3 108 CFU MRSA BAA-1695 alone (MRSA) or on day 7 after PR8 infection (Flu+MRSA). Also shown, survival of mice infected with 50 PFU PR8 and inoculated with PBS control, instead of MRSA, 7 d later (Flu).

Influenza infection inhibits phagocyte antibacterial function in influenza-induced susceptibility to MRSA infection is essentially the airways a result of the negative regulation of these antibacterial factors. NADPH oxidase is responsible for ROS production during phago- To determine how influenza infection leads to defective bacterial phox phox clearance, we next examined airway phagocyte recruitment after cytic killing of S. aureus (30), and both gp91 and p47 are influenza and/or MRSA infection. On the basis of the published critical subunits for activity. Consistent with findings from + other in vivo S. aureus infection studies (19, 20), we showed that reports (27, 28), alveolar macrophages were defined as CD11c Downloaded from CD11blow, neutrophils as CD11b+Ly6G+, and inflammatory mono- NADPH oxidase was necessary for optimal killing of S. aureus + 2 using in the current model, as demonstrated by significantly in- cytes as CD11b Ly6G (Supplemental Fig. 1). Pulmonary S. aureus phox2/2 phox2/2 + + creased lung bacterial burdens in gp91 and p47 ,but infection induced significant neutrophil (CD11b Ly6G ) recruitment 2/2 into the (Fig. 3A). However, comparable numbers of not NOS2 mice (Fig. 4A). neutrophils were detected in MRSA-infected and coinfected mice, In contrast, previous reports suggest that innate clearance of S. aureus only requires neutrophils but not macrophages (20, 31, 32). which were higher than neutrophil counts in mice infected with phox http://www.jimmunol.org/ influenza alone. In fact, there were increased total numbers of gp91 is the specific catalytic subunit of NADPH oxidase in in coinfected mice because of the recruitment of in- phagocytes, including both macrophages and neutrophils. Our flammatory monocytes (CD11b+Ly6ChiLy6G2) following influenza results above demonstrated that influenza infection reduces ROS infection. On the basis of this observation, we hypothesized that the levels in alveolar macrophages. To provide direct evidence that antibacterial function of these phagocytes is impaired, which then macrophages and their oxidative burst are essential for pulmo- nary clearance of S. aureus, we compared bacterial burdens in leads to defective bacterial clearance in coinfected mice. phox2/2 Given the critical role of NADPH oxidase in innate defense against neutrophil-depleted WT and gp91 mice. Pulmonary MRSA S. aureus lung infection, we next determined whether phagocyte burdens were significantly increased after neutrophil depletion

(Fig. 4B). Interestingly, there was a significant difference in bac- by guest on September 30, 2021 oxidative burst was impaired following influenza infection. Signifi- phox2/2 cantly decreased ROS levels were observed in alveolar macrophages terial burdens between neutrophil-depleted gp91 and (CD11c+CD11blow) and neutrophils (CD11b+Ly6G+) isolated from neutrophil-depleted WT mice (Fig. 4B), suggesting an important influenza or coinfected mice compared with animals infected with contribution of macrophages to NADPH oxidase–dependent con- MRSA alone (Fig. 3B, Supplemental Fig. 1). It is noteworthy that trol of bacterial outgrowth. Consistent with this, increased bacte- alveolar macrophages from normal mice exhibited increased ROS rial burdens were found in macrophage-depleted Mafia mice mean fluorescence intensity compared with neutrophils, which was (Fig. 4C). The essential role of neutrophils in host defense against not necessarily due to their greater ROS-generating capacity, but pulmonary MRSA infection was revealed by increased mortality rather, their higher autofluorescence (29). This result suggests that rates in neutrophil-depleted mice (Fig. 4D). Prolonged macro- influenza infection impairs the ability of alveolar macrophages and phage depletion compromises the health of Mafia mice (23); neutrophils to mediate ROS-dependent bactericidal function. therefore, it was not feasible to directly evaluate the contribution of macrophages to survival from MRSA infection. Nonetheless, the Influenza infection suppresses NADPH oxidase–dependent highly elevated bacterial counts in neutrophil-depleted gp91phox2/2 phagocytic bacterial killing mice were accompanied by increased mortality after a relatively We next assessed the possible innate immune components involved low-dose (1 3 107 CFU) MRSA infection (Fig. 4E). Taken together, in pulmonary S. aureus killing without influenza infection, because these results indicate that neutrophils, macrophages, and their

FIGURE 3. Influenza infection regulates the antibacterial ability of airway phagocytes. Air- way phagocyte profiles (three mice per group) (A) and intracellular ROS levels (three mice per group) (B) at 24 h postinfection with 107 CFU MRSA BAA-1695 alone (MRSA) or on day 7 after PR8 infection (Flu+MRSA). Control mice were inoculated with only PBS or infected with 50 PFU PR8 and inoculated with PBS (Flu) 7 d after PR8 infection. The data are representative of two independent experiments. (A and B) p , 0.001, ANOVA. *p , 0.05, ***p , 0.001, Tukey’s multiple comparisons test. 3304 INFLUENZA INFECTION INHIBITS PHAGOCYTIC BACTERIAL KILLING

FIGURE 4. Macrophage oxidative burst is essential for lung S. aureus clearance. Numbers of bacteria remaining in the lungs 24 h after MRSA BAA-1695 infection of C57BL/6 WT, NOS22/2, gp91phox2/2, and p47phox2/2 mice (A), a-PMN Ab-treated WT and gp91phox2/2 mice (B), and Mafia mice (C). In (B), mice were treated with RB6-8C5 mAb for neutrophil de- pletion. In (C), Mafia mice were depleted of macro- phages using AP20187 dissolved in ethanol. Control WT mice were also treated with AP20187. (D) Survival of anti-PMN Ab-treated WT mice after i.n. infection with 108 CFU MRSA. Control WT mice were treated with Rat IgG. (E) Survival of anti-PMN Ab-treated WT and gp91phox2/2 mice after i.n. infection with 107 CFU MRSA. (A–C) p , 0.001, ANOVA. ***p , 0.001, Tukey’s multiple comparisons test. associated NADPH oxidase are essential for efficient pulmonary The cell-intrinsic contribution of NADPH oxidase to the MRSA containment. immune defense against postinfluenza MRSA infection Downloaded from To determine the possible effect of influenza infection on ROS- The data above demonstrate that influenza infection inhibits NADPH dependent bacterial clearance, we next examined anti-MRSA phox2/2 oxidase–dependent bacterial clearance and predisposes mice to immunity in influenza-infected gp91 mice. We hypothe- secondary MRSA infection. Conversely, it has been reported that sized that if influenza infection impaired anti-MRSA immune gp91phox-containing NADPH oxidase promotes influenza infection– mechanisms other than NADPH oxidase-mediated bacterial kill- associated immunopathology to surrounding tissues, because of the ing, coinfected gp91phox2/2 mice should have further increased phox2/2 nonspecific toxic effect of extracellular ROS (15–18). To differentiate bacterial burdens as compared with MRSA-infected gp91 http://www.jimmunol.org/ phox2/2 these concomitant but conflicting contributions of NADPH oxidase, animals. On the contrary, we found that gp91 mice we next sought to determine whether influenza infection impairs exhibited similar defects in MRSA clearance both before and after the cell-intrinsic contribution of NADPH oxidase to phagocytic bac- influenza infection, whereas WT mice were defective only after terial killing. The gp91phox subunit is an X –linked . influenza infection (Fig. 5A). This indicates that influenza infec- Female gp91phox+/2 (mosaic) mice have both gp91phox WT and de- tion predominantly inhibits NADPH oxidase–dependent MRSA ficient circulating neutrophils because of random inactivation of the clearance. In addition, BALF cytology revealed an increase in phox2/2 X chromosome in individual cells (Fig. 7A). Therefore, the potential neutrophil counts in the coinfected gp91 mice relative to cell-intrinsic NADPH oxidase–dependent phagocytic killing could be coinfected WT mice (Fig. 5B). However, in the absence of down-

determined in such mosaic animals. Influenza infection induced an by guest on September 30, 2021 stream effector NADPH oxidase, this enhanced antibacterial im- increase in the percentage of blood neutrophils in gp91phox mosaic mune response failed to facilitate bacterial clearance in coinfected phox phox2/2 mice (Fig. 7A, 7B). However, a similar ratio of gp91 WT and gp91 mice (Fig. 5A). These results further established that deficient blood neutrophils was detected in mosaic mice before and susceptibility to postinfluenza MRSA is due to defective phagocyte after influenza infection, which was also comparable to that in air- antibacterial function. Interestingly, influenza infection resulted in 2/2 ways after influenza infection (Fig. 7B). These observations suggest defective bacterial clearance in NOS2 mice in a pattern similar that gp91phox WT and deficient neutrophil subpopulations in mosaic to WT mice (data not shown), even though NOS2 also participates mice perform similarly in terms of differentiation and migration. in phagocyte oxidative burst. Conversely, despite the abundance of gp91phox WT neutrophils in Despite of its detrimental effect on lung inflammation (15–18), mosaic mice, their lung bacterial burdens were greatly increased an overall impact of NADPH oxidase on animal mortality was not compared with WT mice after MRSA infection (Fig. 7C). Moreover, evident after a lethal dose of influenza infection alone (Fig. 6A). phox2/2 phox2/2 phox2/2 similar to gp91 mice, mosaic mice exhibited defective MRSA Conversely, both p47 and gp91 mice exhibited in- clearance both before and after influenza infection (Fig. 7C). Taken creased susceptibility to S. aureus infection alone compared with phox2/2 together, these results further demonstrated that influenza infection WT mice (Fig. 6B). Furthermore, gp91 mice were highly suppresses cell-intrinsic contributions of NADPH oxidase to phago- susceptible to postinfluenza bacterial infection, as revealed by cytic bacterial killing. rapid mortality kinetics and decreased survival rates compared with WT animals (Fig. 6C, 6D, respectively). These data suggest Discussion that NADPH oxidase–dependent anti-MRSA immunity is impaired, In the current study, we demonstrated that influenza and MRSA although not absolutely absent, in coinfected WT mice. coinfection is associated with defective antibacterial immunity.

FIGURE 5. Influenza infection inhibits NADPH oxidase–dependent bacterial killing. Numbers of bacteria (four to eight mice per group) in the lungs (A), and airway inflammatory cell accumulation (B) at 24 h postinfection WT and gp91phox2/2 mice (n = 4) with 107 CFU MRSA BAA-1695 alone (MRSA) or on day 7 after PR8 infection (Flu+MRSA). (A and B) p , 0.001, ANOVA. *p , 0.05, ***p , 0.001, Tukey’s multiple comparisons test. The Journal of Immunology 3305

FIGURE 6. Influenza infection inhibits NADPH oxidase–dependent immune defense against S. au- reus.(A) Survival of WT and gp91phox2/2 mice postinfection with 103 PFU PR8 virus alone. (B) Survival of WT and gp91phox2/2 mice postinfection with 1 3 108 CFU MRSA BAA-1695 alone. (C and D) Survival of WT and gp91phox2/2 mice postin- fection with 2 3 108 CFU (C) and 1 3 108 CFU (D) MRSA BAA-1695 on day 7 after PR8 infection.

Compared with mice challenged with bacteria alone, postinfluenza sive inflammatory infiltrates but not exacerbated viral infection. MRSA–infected animals were characterized by significantly in- These effects were observed with both PR8 and H1N1 CA04 in- creased bacterial burdens, increased numbers of inflammatory mono- fluenza virus infection (25, 26). Downloaded from cytes, reduced ROS levels in alveolar macrophages and neutrophils In recent studies, impaired antibacterial immunity has been in the airways. At the same time, no significant effect of MRSA shown to be associated with dysregulated pulmonary cytokine re- superinfection on antiviral immunity was observed. Furthermore, sponses following influenza infection. Small et al. (9) reported gp91phox2/2 phagocytes were defective in their antibacterial that influenza infection impairs TNF-a production and causes in- function. This cell-intrinsic contribution of NADPH oxidase ac- creased susceptibility to subsequent S. aureus infection. It has also tivity to phagocytic bacterial killing was established using mosaic been reported that influenza A inhibits IL-1b and Th17-mediated http://www.jimmunol.org/ mice containing both NADPH oxidase WT and deficient phagocyte host defense against S. aureus pneumonia in mice, although the subpopulations. Importantly, similar to gp91phox2/2 mice, these molecular mechanisms involved remain elusive (11, 33). Whereas mosaic mice exhibited similar defective MRSA clearance both be- those studies described an association between suppression of fore and after influenza infection. Collectively, the results presented proinflammatory responses and defective bacterial clearance dur- in this study demonstrate that influenza virus infection suppresses ing influenza infection, how decreased cytokine responses lead to NADPH oxidase–dependent phagocytic bacterial clearance and defective antibacterial immunity remains unclear. Nonetheless, it leads to susceptibility to secondary MRSA infection. has been show that TNF-a and IL-1b prime the respiratory burst Multiple mechanisms have been implicated in the adverse in phagocytes (41, 42), and their reduced production in influenza- outcome of influenza and bacterial coinfection, namely: 1) direct infected mice might be responsible for decreased ROS. Our pre- by guest on September 30, 2021 viral cytotoxicity associated with increased viral burden (8, 10); 2) vious work with secondary pneumococcal infection demonstrated cytotoxicity as a result of bacterial outgrowth (9, 10, 33–37); and that IFN-g is responsible for inhibition of innate bacterial clear- 3) inflammatory tissue damage because of excessive production of ance (25). The dynamics of susceptibility to postinfluenza MRSA proinflammatory mediators and defective resolution/tissue repair infection resemble those of postinfluenza pneumococcal infec- responses (38–40). The observed differences in the relative con- tion. Moreover, an in vitro study suggests that IFN-g can inhibit tributions of virus, bacteria, and host inflammation to the patho- macrophage-mediated of S. aureus (43). Conversely, genesis of coinfection primarily reflect variation in the timing of it is shown that T cell–derived IFN-g can actually perpetuate in- bacterial inoculation in these experimental models. When mice fection in a S. aureus wound infection model by increasing neu- were infected with MRSA at the beginning of the viral recovery trophil recruitment, cells that can serve a reservoir for bacterial phase, we found that coinfection-induced high mortality rates that growth (44). Therefore, IFN-g might have a different effect on were associated with impaired antibacterial immunity and exces- phagocyte-mediated clearance of S. aureus versus pneumococci.

FIGURE 7. Influenza infection inhibits cell-intrinsic contribution of NADPH oxidase to phagocytic bacte- rial killing. (A) Flow cytometry analysis of gp91phox expression in blood neutrophils (CD11b+Ly6G+) from female gp91phox WT, knockout (KO), and mosaic mice (n = 5). (B) Composition of gp91phox WT and deficient neutrophils in blood and airways of gp91phox+/2 mosaic mice (n = 5) at day 7 after influenza infection. (C)Num- bers of bacteria (five mice per group) in the lungs 24 h after challenge of naive (MRSA) or day 7 postinfluenza infected (Flu+MRSA) female WT and gp91phox+/2 mosaic mice with 107 CFU MRSA BAA-1695. (C) p , 0.01, ANOVA. **p , 0.01, ***p , 0.001, Tukey’s multiple comparisons test. 3306 INFLUENZA INFECTION INHIBITS PHAGOCYTIC BACTERIAL KILLING

This differential regulatory role may be related to differences in The present studies were conducted using a mouse infection the effector mechanisms that are needed to eliminate these two model, which is well accepted for studying influenza and influenza- . For example, a macrophage receptor with collagenous compromised bacterial infection (50). However, mice are not only structure is critical for alveolar macrophage–mediated clearance highly resistant to S. aureus infection but also exhibit very rapid of pneumococci but not S. aureus clearance (K. Sun, unpublished mortality kinetics after a lethal inoculum. To minimize compli- observations). Conversely, NADPH oxidase activity, which is likely cations from compensatory or secondary inflammatory responses to overcome the antioxidant action of endogenous staphylococcal because of lethal bacterial burdens, we infected mice with a rela- , is essential for pulmonary S. aureus clearance but is dis- tively low dose of MRSA to examine the influence of influenza pensable for pneumococcal eradiation (24). infection on protective antibacterial immunity and then confirmed There is controversy surrounding the relative contribution of these findings with higher inoculums for survival studies. The macrophages to pulmonary clearance of S. aureus (20, 31, 32). results of this study establish for the first time, to our knowledge, However, we found that the immunological killing of S. aureus that influenza infection suppresses cell-intrinsic contribution of requires both neutrophils and macrophages. NADPH oxidase also NADPH oxidase to phagocytic bacterial killing, and predisposes has been found to be important for antibacterial immunity (20). hosts to secondary S. aureus infection. Consistent with this, we observed increased bacterial loads and mortality after MRSA infection of NADPH oxidase-deficient mice. Acknowledgments Because 98% of leukocytes in the airways of MRSA-infected mice We thank Drs. Tammy Kielian and Victor Huber for critical review of the were composed of alveolar macrophages and neutrophils (Fig. 3A, manuscript and Kari Nelson for editorial comments.

Supplemental Fig. 1), the lack of control of MRSA infection Downloaded from phox2/2 observed in gp91 mice is likely due to defective oxidase- Disclosures dependent bacterial killing by these phagocytes. In postinfluenza The authors have no financial conflicts of interest. MRSA–infected mice, analysis of myeloid cell profiles in the airway revealed accumulation of neutrophils and inflammatory monocytes, suggesting that the defective antibacterial immunity in References 1. Morens, D. M., J. K. Taubenberger, and A. S. Fauci. 2008. Predominant role of these mice was not associated with total numbers of phagocytes http://www.jimmunol.org/ bacterial pneumonia as a cause of death in pandemic influenza: implications for but rather their reduced antimicrobial ability. It is noteworthy that pandemic influenza preparedness. J. Infect. Dis. 198: 962–970. in contrast to our observations in animal models, in vitro coculture 2. Louria, D. B., H. L. Blumenfeld, J. T. Ellis, E. D. Kilbourne, and D. E. Rogers. of influenza A virus/S. pneumonia with neutrophils increased the 1959. Studies on influenza in the pandemic of 1957-1958. II. Pulmonary com- plications of influenza. J. Clin. Invest. 38: 213–265. respiratory burst activity of these host cells (45), probably because 3. Centers for Disease Control and Prevention. 2007. Severe methicillin-resistant pneumolysin released by S. pneumonia is a potent activator of Staphylococcus aureus community-acquired pneumonia associated with influ- enza—Louisiana and Georgia, December 2006-January 2007. MMWR Morb. intracellular oxygen radical production (46). Nonetheless, unlike Mortal. Wkly. Rep. 56: 325–329. S. aureus, NADPH oxidase is not essential for phagocytic killing 4. Finelli, L., A. Fiore, R. Dhara, L. Brammer, D. K. Shay, L. Kamimoto, A. Fry, of S. pneumonia (24, 47). In fact, gp91phox2/2 mice demonstrate J. Hageman, R. Gorwitz, J. Bresee, and T. Uyeki. 2008. Influenza-associated pediatric mortality in the United States: increase of Staphylococcus aureus by guest on September 30, 2021 increased resistance to pneumococcal infection compared with coinfection. Pediatrics 122: 805–811. WT controls (48). Nonetheless, we found that influenza infection 5. Murray, R. J., J. O. Robinson, J. N. White, F. Hughes, G. W. Coombs, impaired ROS generation in both alveolar macrophages and neu- J. C. Pearson, H. L. Tan, G. Chidlow, S. Williams, K. J. Christiansen, and D. W. Smith. 2010. Community-acquired pneumonia due to pandemic A(H1N1) trophils, which, together with the observation that similar defec- 2009 influenzavirus and methicillin resistant Staphylococcus aureus co-infec- tive bacterial clearance was detected in gp91phox-deficient and tion. PLoS ONE 5: e8705. 6. Randolph, A. G., F. Vaughn, R. Sullivan, L. Rubinson, B. T. Thompson, G. Yoon, mosaic mice with or without influenza infection, suggests that E. Smoot, T. W. Rice, L. L. Loftis, M. Helfaer, et al. 2011. Critically ill children NADPH oxidase–dependent bacterial killing is the principal path- during the 2009-2010 influenza pandemic in the United States. Pediatrics 128: way by which influenza infection regulates innate antibacterial e1450–e1458. 7. Centers for Disease Control and Prevention. 2012. Severe coinfection with immunity. seasonal influenza A (H3N2) virus and Staphylococcus aureus—Maryland, Although phagocyte NADPH oxidase is critical for host an- February-March 2012. MMWR Morb. Mortal. Wkly. Rep. 61: 289–291. tibacterial immunity, it also has been shown to promote detri- 8. Tashiro, M., P. Ciborowski, H. D. Klenk, G. Pulverer, and R. Rott. 1987. Role of Staphylococcus protease in the development of influenza pneumonia. Nature mental inflammatory responses during influenza infection (15– 325: 536–537. 18). As such, alleviated lung pathologies and reduced accumu- 9. Small, C. L., C. R. Shaler, S. McCormick, M. Jeyanathan, D. Damjanovic, E. G. Brown, P. Arck, M. Jordana, C. Kaushic, A. A. Ashkar, and Z. Xing. 2010. lation of airway inflammatory cells were detected in NADPH Influenza infection leads to increased susceptibility to subsequent bacterial su- oxidase–deficient mice in response to inactivated H5N1 avian perinfection by impairing NK cell responses in the lung. J. Immunol. 184: 2048– influenza virus (18) or early after low pathogenicity H3N2 influenza 2056. 10. Iverson, A. R., K. L. Boyd, J. L. McAuley, L. R. Plano, M. E. Hart, and infection (15). Therefore, the negative regulation of phagocyte ROS J. A. McCullers. 2011. Influenza virus primes mice for pneumonia from generation at the recovery stage of influenza infection, as revealed Staphylococcus aureus. J. Infect. Dis. 203: 880–888. in the current study, has likely evolved to minimize influenza in- 11. Robinson, K. M., S. M. Choi, K. J. McHugh, S. Mandalapu, R. I. Enelow, J. K. Kolls, and J. F. Alcorn. 2013. Influenza A exacerbates Staphylococcus fection–associated lung damage. In agreement with our findings, it aureus pneumonia by attenuating IL-1b production in mice. J. Immunol. 191: has been shown that both basal and phorbol dibutyrate-stimulated 5153–5159. 12. Vardakas, K. Z., D. K. Matthaiou, and M. E. Falagas. 2009. Incidence, charac- superoxide production in airway inflammatory cells are decreased teristics and outcomes of patients with severe community acquired-MRSA from days 3 to 7 after influenza infection, which is associated with pneumonia. Eur. Respir. J. 34: 1148–1158. a suppressive effect of Nox1 oxidase in tissue cells (49). Consid- 13. Blyth, C. C., S. A. Webb, J. Kok, D. E. Dwyer, S. J. van Hal, H. Foo, A. N. Ginn, phox A. M. Kesson, I. Seppelt, J. R. Iredell, et al. 2013. The impact of bacterial and ering that gp91 -dependent generation of superoxide is not ab- viral co-infection in severe influenza. Influenza Other Respi. Viruses 7: 168‑176. sent, but rather reduced in influenza-infected WT mice, it is not 14. Metersky, M. L., R. G. Masterton, H. Lode, T. M. File, Jr., and T. Babinchak. surprising that gp91phox2/2 mice exhibit attenuated lung inflam- 2012. Epidemiology, microbiology, and treatment considerations for bacterial pneumonia complicating influenza. Int. J. Infect. Dis. 16: e321–e331. matory damage compared with WT controls and inhibition of 15. Vlahos, R., J. Stambas, S. Bozinovski, B. R. Broughton, G. R. Drummond, and NADPH oxidase reduces disease severity during influenza infec- S. Selemidis. 2011. Inhibition of Nox2 oxidase activity ameliorates influenza A virus‑induced lung inflammation. PLoS Pathog. 7: e1001271. tion alone (i.e., in the absence of secondary S. aureus infection) 16. Akaike, T., Y. Noguchi, S. Ijiri, K. Setoguchi, M. Suga, Y. M. Zheng, (15–17). B. Dietzschold, and H. Maeda. 1996. Pathogenesis of influenza virus-induced The Journal of Immunology 3307

pneumonia: involvement of both nitric oxide and oxygen radicals. Proc. Natl. 34. Lee, M. H., C. Arrecubieta, F. J. Martin, A. Prince, A. C. Borczuk, and Acad. Sci. USA 93: 2448–2453. F. D. Lowy. 2010. A postinfluenza model of Staphylococcus aureus pneumonia. 17. Snelgrove, R. J., L. Edwards, A. J. Rae, and T. Hussell. 2006. An absence of J. Infect. Dis. 201: 508–515. reactive oxygen species improves the resolution of lung influenza infection. Eur. 35. Shahangian, A., E. K. Chow, X. Tian, J. R. Kang, A. Ghaffari, S. Y. Liu, J. Immunol. 36: 1364–1373. J. A. Belperio, G. Cheng, and J. C. Deng. 2009. Type I IFNs mediate development 18. Imai, Y., K. Kuba, G. G. Neely, R. Yaghubian-Malhami, T. Perkmann, G. van of postinfluenza bacterial pneumonia in mice. J. Clin. Invest. 119: 1910–1920. Loo, M. Ermolaeva, R. Veldhuizen, Y. H. Leung, H. Wang, et al. 2008. Identi- 36. Didierlaurent, A., J. Goulding, S. Patel, R. Snelgrove, L. Low, M. Bebien, fication of oxidative stress and Toll-like receptor 4 signaling as a key pathway of T. Lawrence, L. S. van Rijt, B. N. Lambrecht, J. C. Sirard, and T. Hussell. 2008. acute lung injury. Cell 133: 235–249. Sustained desensitization to bacterial Toll-like receptor ligands after resolution 19. Jann, N. J., M. Schmaler, F. Ferracin, and R. Landmann. 2011. TLR2 enhances of respiratory influenza infection. J. Exp. Med. 205: 323–329. NADPH oxidase activity and killing of Staphylococcus aureus by PMN. 37. Navarini, A. A., M. Recher, K. S. Lang, P. Georgiev, S. Meury, A. Bergthaler, Immunol. Lett. 135: 17–23. L. Flatz, J. Bille, R. Landmann, B. Odermatt, et al. 2006. Increased susceptibility 20. Ko¨hler, J., K. Breitbach, C. Renner, A. K. Heitsch, A. Bast, N. van Rooijen, to bacterial superinfection as a consequence of innate antiviral responses. Proc. S. Vogelgesang, and I. Steinmetz. 2011. NADPH-oxidase but not inducible nitric Natl. Acad. Sci. USA 103: 15535–15539. oxide synthase contributes to resistance in a murine Staphylococcus aureus 38. Kash, J. C., K. A. Walters, A. S. Davis, A. Sandouk, L. M. Schwartzman, Newman pneumonia model. Microbes Infect. 13: 914–922. B. W. Jagger, D. S. Chertow, Q. Li, R. E. Kuestner, A. Ozinsky, and 21. Gaush, C. R., and T. F. Smith. 1968. Replication and plaque assay of influenza J. K. Taubenberger. 2011. Lethal synergism of 2009 pandemic H1N1 influenza virus in an established line of canine kidney cells. Appl. Microbiol. 16: 588–594. virus and Streptococcus pneumoniae coinfection is associated with loss of mu- 22. Chandra, R., S. Federici, Z. H. Ne´meth, B. Horva´th, P. Pacher, G. Hasko´, rine lung repair responses. MBio 2: e00172-11. E. A. Deitch, and Z. Spolarics. 2011. Female X-chromosome mosaicism for 39. Karlstro¨m, A., S. M. Heston, K. L. Boyd, E. I. Tuomanen, and J. A. McCullers. NOX2 deficiency presents unique inflammatory and improves out- 2011. Toll-like receptor 2 mediates fatal immunopathology in mice during come in polymicrobial . J. Immunol. 186: 6465–6473. treatment of secondary pneumococcal pneumonia following influenza. J. Infect. 23. Chinnery, H. R., E. C. Carlson, Y. Sun, M. Lin, S. H. Burnett, V. L. Perez, Dis. 204: 1358–1366. P. G. McMenamin, and E. Pearlman. 2009. Bone marrow chimeras and c-fms 40. Karlstro¨m, A., K. L. Boyd, B. K. English, and J. A. McCullers. 2009. Treatment conditional ablation (Mafia) mice reveal an essential role for resident myeloid with protein synthesis inhibitors improves outcomes of secondary bacterial cells in lipopolysaccharide/TLR4-induced corneal inflammation. J. Immunol. pneumonia after influenza. J. Infect. Dis. 199: 311–319. Downloaded from 182: 2738–2744. 41. Gauss, K. A., L. K. Nelson-Overton, D. W. Siemsen, Y. Gao, F. R. DeLeo, and 24. Sun, K., Y. Gan, and D. W. Metzger. 2011. Analysis of murine genetic predis- M. T. Quinn. 2007. Role of NF-kB in transcriptional regulation of the phagocyte position to pneumococcal infection reveals a critical role of alveolar macro- NADPH oxidase by tumor necrosis factor-alpha. J. Leukoc. Biol. 82: 729–741. phages in maintaining the sterility of the lower respiratory tract. Infect. Immun. 42. El-Benna, J., P. M. Dang, and M. A. Gougerot-Pocidalo. 2008. Priming of the 79: 1842–1847. neutrophil NADPH oxidase activation: role of p47phox phosphorylation and 25. Sun, K., and D. W. Metzger. 2008. Inhibition of pulmonary antibacterial defense NOX2 mobilization to the plasma membrane. Semin. Immunopathol. 30: 279– by interferon-g during recovery from influenza infection. Nat. Med. 14: 558– 289.

564. 43. Hang, T. T., E. J. Choi, J. Y. Song, S. E. Kim, J. Kwak, and Y. K. Shin. 2011. http://www.jimmunol.org/ 26. Sun, K., J. Ye, D. R. Perez, and D. W. Metzger. 2011. Seasonal FluMist vac- Differential effect of prior influenza infection on alveolar macrophage phago- cination induces cross-reactive T cell immunity against H1N1 (2009) influenza cytosis of Staphylococcus aureus and Escherichia coli: involvement of inter- and secondary bacterial infections. J. Immunol. 186: 987–993. feron-g production. Microbiol. Immunol. 55: 751–759. 27. Seo, S. U., H. J. Kwon, H. J. Ko, Y. H. Byun, B. L. Seong, S. Uematsu, S. Akira, 44. McLoughlin, R. M., J. C. Lee, D. L. Kasper, and A. O. Tzianabos. 2008. IFN-g and M. N. Kweon. 2011. Type I interferon signaling regulates Ly6Chi monocytes regulated chemokine production determines the outcome of Staphylococcus and neutrophils during acute viral pneumonia in mice. PLoS Pathog. 7: aureus infection. J. Immunol. 181: 1323–1332. e1001304. 45. Engelich, G., M. White, and K. L. Hartshorn. 2001. Neutrophil survival is 28. Lin, K. L., Y. Suzuki, H. Nakano, E. Ramsburg, and M. D. Gunn. 2008. CCR2+ markedly reduced by incubation with influenza virus and Streptococcus pneu- monocyte-derived dendritic cells and exudate macrophages produce influenza- moniae: role of respiratory burst. J. Leukoc. Biol. 69: 50–56. induced pulmonary immune pathology and mortality. J. Immunol. 180: 2562– 46. Martner, A., C. Dahlgren, J. C. Paton, and A. E. Wold. 2008. Pneumolysin re- 2572. leased during Streptococcus pneumoniae autolysis is a potent activator of in-

29. Duan, M., W. C. Li, R. Vlahos, M. J. Maxwell, G. P. Anderson, and M. L. Hibbs. tracellular oxygen radical production in neutrophils. Infect. Immun. 76: 4079– by guest on September 30, 2021 2012. Distinct macrophage subpopulations characterize acute infection and 4087. chronic inflammatory lung disease. J. Immunol. 189: 946–955. 47. Marriott, H. M., P. G. Hellewell, M. K. Whyte, and D. H. Dockrell. 2007. 30. Hampton, M. B., A. J. Kettle, and C. C. Winterbourn. 1996. Involvement of Contrasting roles for reactive oxygen species and nitric oxide in the innate re- superoxide and myeloperoxidase in oxygen-dependent killing of Staphylococcus sponse to pulmonary infection with Streptococcus pneumoniae. 25: aureus by neutrophils. Infect. Immun. 64: 3512–3517. 2485–2490. 31. Robertson, C. M., E. E. Perrone, K. W. McConnell, W. M. Dunne, B. Boody, 48. Marriott, H. M., L. E. Jackson, T. S. Wilkinson, A. J. Simpson, T. J. Mitchell, T. Brahmbhatt, M. J. Diacovo, N. Van Rooijen, L. A. Hogue, C. L. Cannon, et al. D. J. Buttle, S. S. Cross, P. G. Ince, P. G. Hellewell, M. K. Whyte, and 2008. Neutrophil depletion causes a fatal defect in murine pulmonary Staphy- D. H. Dockrell. 2008. Reactive oxygen species regulate neutrophil recruitment lococcus aureus clearance. J. Surg. Res. 150: 278–285. and survival in pneumococcal pneumonia. Am. J. Respir. Crit. Care Med. 177: 32. Rigby, K. M., and F. R. DeLeo. 2012. Neutrophils in innate host defense against 887–895. Staphylococcus aureus infections. Semin. Immunopathol. 34: 237–259. 49. Selemidis, S., H. J. Seow, B. R. Broughton, A. Vinh, S. Bozinovski, C. G. Sobey, 33. Kudva, A., E. V. Scheller, K. M. Robinson, C. R. Crowe, S. M. Choi, G. R. Drummond, and R. Vlahos. 2013. Nox1 oxidase suppresses influenza S. R. Slight, S. A. Khader, P. J. Dubin, R. I. Enelow, J. K. Kolls, and J. F. Alcorn. a virus-induced lung inflammation and oxidative stress. PLoS One 8: e60792. 2011. Influenza A inhibits Th17-mediated host defense against bacterial pneu- 50. Metzger, D. W., and K. Sun. 2013. Immune dysfunction and bacterial coin- monia in mice. J. Immunol. 186: 1666–1674. fections following influenza. J. Immunol. 191: 2047–2052.