Published ahead of print on December 14, 2006, doi:10.1165/rcmb.2006-0166OC
© 2007 American Thoracic Society DOI: 10.1165/rcmb.2006-0166OC Toll-Like Receptor 2 Does Not Contribute to Host Response during Postinfluenza Pneumococcal PneumoniaCenter for Infection and Immunity Amsterdam (CINIMA); Center for Experimental and Molecular Medicine; Laboratory of Experimental Immunology; Department of Pulmonology; and Department of Pathology, Academic Medical Center, University of Amsterdam, The Netherlands; and Exploratory Research for Advanced Technology, Japan Science and Technology Agency, Osaka, Japan Correspondence and requests for reprints should be addressed to Mark C. Dessing, Academic Medical Center, Laboratory of Experimental Internal Medicine, Room F0-117, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. E-mail: m.c.dessing{at}amc.uva.nl
Influenza A can be complicated by secondary bacterial pneumonia, which is most frequently caused by Streptococcus pneumoniae and associated with uncontrolled pulmonary inflammation. Evidence points to Toll-like receptor (TLR) 2 as a possible mediator of this exaggerated lung inflammation: (1) TLR2 is the most important "sensor" for gram-positive stimuli, (2) TLR2 contributes to S. pneumoniaeinduced inflammation, and (3) influenza A enhances TLR2 expression in various cell types. Therefore, the objective of this study was to determine the role of TLR2 in the host response to postinfluenza pneumococcal pneumonia. TLR2 knockout (KO) and wild-type (WT) mice were infected intranasally with influenza A virus. Fourteen days later they were administered with S. pneumoniae intranasally. Influenza was associated with a similar transient weight loss in TLR2 KO and WT mice. Both mouse strains were fully recovered and had completely cleared the virus at Day 14. Importantly, no differences between TLR2 KO and WT mice were detected during postinfluenza pneumococcal pneumonia with respect to bacterial growth, lung inflammation, or cytokine/chemokine concentrations, with the exception of lower pulmonary levels of cytokine-induced neutrophil chemoattractant in TLR2 KO mice. Toll-like receptor 2 does not contribute to host defense during murine postinfluenza pneumococcal pneumonia.
Key Words: influenza A Streptococcus pneumoniae pneumonia Toll-like receptor mice
Secondary bacterial pneumonia is a feared complication of respiratory tract infection by influenza A, responsible for at least 20,000 deaths annually in the United States alone (1). The most important pathogens causing postinfluenza pneumonia are Staphylococcus aureus, Haemophilus influenzae, and in particular Streptococcus pneumoniae (2). Although S. pneumoniae is the most common pathogen isolated from previously healthy patients with community-acquired pneumonia (3), such primary pulmonary infections with the pneumococcus are usually less severe than secondary infections following influenza A (4). Thus far, knowledge about the precise mechanism by which influenza modulates the innate immune response to facilitate secondary bacterial infection in the lung is limited. Our laboratory recently developed a model of postinfluenza pneumococcal pneumonia to obtain more insight into the pathogenetic mechanisms contributing the adverse outcome of secondary bacterial pneumonia (57). In this model mice are intranasally infected with a mouse-adapted strain of influenza A, causing a mild illness characterized by transient weight loss and a complete recovery together with viral clearance by Day 14. At this time point mice are infected with S. pneumoniae, which, in comparison with mice with primary pneumococcal pneumonia, results in an exaggerated pulmonary inflammatory response, a strongly enhanced bacterial outgrowth, and a reduced survival (57). S. pneumoniae can activate the innate immune system by an interaction with so-called pattern recognition receptors, among which Toll-like receptors (TLRs) prominently feature. Previous investigations have pointed to TLR2 as the key pattern recognition receptor in the immune response against gram-positive bacteria (810). In line, both in vitro and in vivo studies have indicated that S. pneumoniae activates the immune system at least in part via TLR2, although other TLRs, in particular TLR4, may also be involved (1014). Moreover, our laboratory recently demonstrated that TLR2 contributes to the inflammatory response after primary pneumococcal pneumonia (15). We hypothesized that signaling of S. pneumoniae via TLR2 is an important mechanism by which this pathogen causes exaggerated lung inflammation during infection following influenza A. This hypothesis, which was also put forward in a recent review on the induction of immune responses by S. pneumoniae (16), was supported by the fact that the expression of TLR2 has been found enhanced in mouse macrophages, human neutrophils, and in human epithelial cells infected with influenza A (17, 18, 19). Thus, in the present study we sought to determine the role of TLR2 during postinfluenza pneumococcal pneumonia.
Animals Specific pathogenfree 8- to 10-wk-old female C57BL/6 mice (wild-type [WT]) were purchased from Charles River (Maastricht, The Netherlands). TLR2 knockout (KO) mice were generated as described previously (8) and backcrossed to C57BL/6 background six times; these mice were bred in the animal facility of the Academic Medical Center in Amsterdam. Age- and sex-matched mice were used in all experiments. All experiments were approved by the Animal Care and Use Committee of the University of Amsterdam (Amsterdam, The Netherlands).
Postinfluenza Pneumonia
Histopathological Analysis
Cytokine and Chemokine Measurement
Statistical Analysis
Body Weight and Viral Clearance during Primary Influenza Infection The primary goal of our study was to determine the possible contribution of TLR2 signaling in the exaggerated inflammatory response during S. pneumoniae pneumonia following influenza A infection. In order to adequately address this issue, we first established whether influenza has a different course in TLR2 KO mice than in WT micethat is, in case TLR2 KO mice would handle influenza infection in a different way, the "baseline condition" upon which pneumococcal pneumonia is superimposed would differ between the two mouse strains, hampering an adequate comparison between TLR2 KO and WT mice during postinfluenza pneumonia. Thus, TLR2 KO and WT mice were intranasally infected with influenza virus and followed for 14 d. As reported earlier by our and other laboratories (57, 22), influenza virus infection resulted in a transient loss of bodyweight in WT mice. This decrease in body weight, which reached a nadir at 8 d after infection and had completely recovered at 14 d, was similar in TLR2 KO mice (Figure 1A). Next, we determined viral loads in whole lung homogenates prepared on Days 2, 8, and 14 after influenza infection using real-time quantitative PCR. No differences in viral load were found in the lungs of WT and TLR2 KO mice at any time point. At 14 d after inoculation of the virus, influenza could not be detected anymore in lungs of either group, indicating that the virus had been cleared from the lungs of both TLR2 KO and WT mice (Figure 1B).
Lung Inflammation during Primary Influenza Infection To determine whether TLR2 deficiency influences the pulmonary cytokine and chemokine response during influenza, we measured the concentrations of TNF- , IL-1 , IL-10, KC, MIP-2, and IFN- in lung homogenates obtained from TLR2 KO and WT mice at Days 2, 8, and 14 after infection with influenza (Figure 2). Although overall the levels of these mediators were relatively low, especially when compared to the levels measured after bacterial infection (see below; for reasons of clarity these latter data are also presented in Figure 2), some differences were found between TLR2 KO and WT mice. In particular, the pulmonary levels of the anti-inflammatory cytokine IL-10 were higher in TLR2 KO mice at 8 and 14 d after infection (both P < 0.05 versus WT mice), whereas lung KC concentrations were lower in TLR2 KO mice 2 and 8 d after infection (both P < 0.05 versus WT mice). IFN- production tended to be higher in TLR2 KO mice 8 and 14 d after influenza inoculation, although this was not significant (P = 0.05 and P = 0.12, respectively). Lung TNF- , IL-1 , or MIP-2 levels did not differ between TLR2 KO and WT mice.
Body Weight and Bacterial Outgrowth during Postinfluenza Pneumonia At 14 d after infection with influenza, when all mice had completely recovered and the virus was no longer detectable in lungs, TLR2 KO and WT mice were intranasally infected with S. pneumoniae. Bacterial pneumonia resulted in a marked body weight loss 48 h after infection; however, no differences were observed between TLR2 KO and WT mice (Figure 1A). To determine whether TLR2 deficiency influences bacterial outgrowth during postinfluenza pneumonia we measured the number of S. pneumoniae cfu in the lungs of TLR2 KO and WT mice 6 and 48 h after the bacterial inoculation. The 6-h time point was chosen because TLR2 plays a role in early inflammatory response in murine pneumococcal pneumonia (15). The 48-h time point was chosen because it is suitable to compare bacterial growth in this pneumonia model (2325). At neither time point did the pulmonary bacterial loads differ between the two mouse strains (Figure 3). In addition, bacteremia occurred similarly in WT and TLR2 KO mice: whereas 6 h after inoculation of S. pneumoniae neither WT nor TLR2 KO mice had positive blood cultures, 48 h after bacterial infection all mice were bacteremic.
Lung Inflammation during Postinfluenza Pneumonia Our laboratory previously showed that the lung inflammatory response to secondary S. pneumoniae infection of mice that have just recovered from influenza infection is strongly enhanced when compared to the inflammatory reaction in lungs of mice with primary S. pneumoniae pneumonia (57). Having established that TLR2 does not contribute to an effective antibacterial defense during postinfluenza pneumococcal pneumonia, we next wished to determine the possible role of TLR2 in the induction of lung inflammation after secondary bacterial respiratory tract infection. For this we semiquantitatively scored lung tissue slides obtained from TLR2 KO and WT mice 6 and 48 h after infection. No difference in pulmonary inflammation between TLR2 KO and WT mice were observed at either 6 or 48 h after inoculation of S. pneumoniae (Figure 4). To determine whether TLR2 KO mice had an altered cytokine/chemokine response to postinfluenza pneumonia, we measured several cytokines and chemokines in lung homogenates 6 and 48 h after inoculation with S. pneumoniae (Figure 2). Lung concentrations of TNF- , IL-1 , IL-10, KC, MIP-2, and IFN- did not differ between TLR2 KO and WT mice at either time point, with the exception of KC levels 48 h after bacterial infection, which were lower in the former mouse strain.
Induction of S. pneumoniae Pneumonia 8 d after Inoculation with Influenza To determine whether TLR2 plays a role in the inflammatory response to pneumococcal pneumonia superimposed on influenza induced 8 d earlier, we compared bacterial loads and cytokine/chemokine levels in lung homogenates prepared 6 h after intranasal inoculation with S. pneumoniae in TLR2 KO and WT mice infected with influenza 8 d earlier. Of note, at 8 d after inoculation with influenza, pulmonary viral loads were high and infected mice were severely ill, as illustrated by their loss of weight (see Figure 1). During the first 6 h after superinfection with S. pneumoniae, three of seven WT mice and three of seven TLR2 KO mice died. In the remaining mice, no differences were detected between TLR2 KO and WT mice with respect to bacterial loads or cytokine/chemokine levels in lungs (Table 1).
Postinfluenza pneumococcal pneumonia is associated with a much stronger inflammatory response in the lungs than primary pneumonia caused by S. pneumoniae. We here tested the hypothesis that TLR2 signaling contributes to this exaggerated pulmonary inflammation during S. pneumoniae pneumonia following influenza A infection. This hypothesis was based on the following lines of evidence: (1) TLR2 has been implicated as the most important TLR for sensing gram-positive bacteria (26); (2) TLR2 has been found important for the induction of inflammation upon infection with S. pneumoniae in vivo (11, 12, 15); and (3) TLR2 expression increased in macrophages, neutrophils, and epithelial cells upon infection with influenza A (1719). However, the main finding of this study is that, in contrast to our expectation, TLR2 does not play a role of importance in postinfluenza pneumococcal pneumonia. In a first series of experiments we established that TLR2 is not involved in the host response to influenza A infection to a significant extent. Indeed, the transient body weight loss and viral clearance were unaltered in TLR2 KO mice when compared with normal WT mice. Importantly, both mouse strains had completely cleared influenza virus at the time infection with S. pneumoniae was accomplished (that is, 2 wk after intranasal inoculation of the virus). We used this time interval in this and our previous studies on postinfluenza pneumococcal pneumonia (57) because we wished to exclude a direct interaction between influenza virus and S. pneumoniae in the lungs and because clinical data indicate that 2 wk is a common interval between influenza infection and the occurrence of secondary bacterial complications (2, 27). Notably, modest differences in pulmonary cytokine and chemokine levels were detected in TLR2 KO and WT mice infected with influenza A. In particular, TLR2 KO displayed higher pulmonary IL-10 concentrations during influenza, contrasting with findings in infections caused by other pathogens (Yersinia enterocolitica and Candida albicans) that have suggested that TLR2 stimulation results in a type 2biased immune response characterized by increased IL-10 release (28). It is unlikely that the modestly elevated IL-10 levels in TLR2 KO mice at the time S. pneumoniae was administered biased our results: higher IL-10 concentrations in theory would have reduced lung inflammation during postinfluenza pneumonia (29), and thus would have made the expected diminished lung inflammation in TLR2 KO mice more profound; clearly this was not what we found in the current investigation. The same holds true for the slightly lower KC levels in TLR2 KO mice during the initial phase of influenza. We do not have a clear explanation for these small differences between the two mouse strains, especially since there is no evidence that TLR2 contributes to cellular responsiveness to influenza virus (30, 31). TLR2 does contribute to immune responses triggered by cytomegalovirus, varicella-zoster virus, and herpes simplex (3235). Within the TLR family in particular, TLR3 is important for the innate recognition of double-stranded viral RNA (31). Influenza A virus is a negative sense single-stranded RNA virus with double-stranded replication intermediates, which are likely to be TLR3 ligands. Recently Le Goffic and coworkers showed a significant contribution of TLR3 during pulmonary infection with influenza (36). They reported that TLR3 is up-regulated during viral infection and that mice deficient of this receptor displayed significantly reduced inflammatory mediators and a lower number of CD8+ T lymphocytes in the bronchoalveolar airspace. Surprisingly, TLR3 KO mice had a survival advantage, despite a higher viral load in the lungs (36). In light of the minor differences between TLR2 KO and WT mice during influenza, we considered it feasible to use TLR2 KO mice to establish the role of this receptor in the host response to postinfluenza pneumonia. The impact of TLR2 deficiency on lung inflammation during postinfluenza pneumococcal pneumonia was evaluated at 6 and 48 h after bacterial infection. These time points were chosen in light of our previous investigation on the role of TLR2 in primary S. pneumoniae pneumonia (15). In that study, TLR2 KO mice were found to have lower pulmonary cytokine concentrations early after infection (6 h), whereas at 48 h after infection TLR2 KO mice displayed reduced lung inflammation upon semiquantitative histologic analysis (15). Such differences were not observed in the current study, although TLR2 KO mice did show reduced lung KC concentrations 48 h after inoculation with S. pneumoniae. This latter finding presumably reflects the relatively strong TLR2 dependence of KC release induced by gram-positive stimuli, including S. pneumoniae, as indicated by profoundly diminished KC production by TLR2 KO alveolar macrophages in vitro and whole lungs from TLR2 KO mice in vivo upon exposure to S. pneumoniae (15). In line with our earlier study (15), TLR2 KO mice displayed similar bacterial loads in their lungs as WT mice, and the occurrence of bacteremia was identical in both mouse strains. Together, these data indicate that the role of TLR2 in the host response to respiratory tract infection caused by S. pneumoniae is modest during primary infection and insignificant during postinfluenza pneumonia. Moreover, in additional experiments no difference in bacterial outgrowth and immune response were observed when WT and TLR2 KO mice were infected for 6 h with S. pneumoniae, 8 d after infection with influenza. Apparently, other TLRs are capable of compensating for the absence of the "gram-positive sensor" TLR2 during pneumococcal infection. Indeed, mice with a functional loss of TLR4, and in particular mice with a targeted deletion of the gene encoding the TLR9 or common TLR adaptor MyD88, demonstrated an increased susceptibility to primary pneumococcal pneumonia (13, 14, 37, 38). Further studies are warranted to establish the role of these molecules in postinfluenza pneumonia. It has been well established that influenza renders the host more susceptible to secondary infection with S. pneumoniae, which is associated with an uncontrolled inflammatory reaction in the lungs. We here investigated the potential role of TLR2 in the deregulated host response to pneumococcal pneumonia following influenza. In contrast to our expectation, TLR2 deficiency had no impact on lung inflammation or bacterial growth, suggesting that other pattern recognition receptors can compensate for the loss of TLR2 in the innate recognition of S. pneumoniae during respiratory tract infection superimposed on influenza.
The authors thank Joost Daalhuisen and Marieke ten Brink for technical assistance during the animal experiments, and Regina de Beer for preparations of lung sections.
M.C.D. is supported by an institutional grant of the Academic Medical Center. Originally Published in Press as DOI: 10.1165/rcmb.2006-0166OC on December 14, 2006 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form May 9, 2006 Accepted in final form December 4, 2006
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||