| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Microbiology and Infectious Diseases, Toho University School of Medicine, Tokyo 143-8540, Japan
2 Fourth Department of Internal Medicine, Nippon Medical School, Tokyo, Japan
3 Department of Microbiology, Tokyo Medical University, Tokyo, Japan
4 Denkaseiken Co. Ltd, Gosenshi, Minami-motomachi 1-2-2, Niigata, Japan
5 Pulmonary and Critical Care Medicine, University of Michigan Medical School, Ann Arbor, MI 48109-0360, USA
Correspondence
Kazuhiro Tateda
kazu{at}med.toho-u.ac.jp
Received 25 August 2006
Accepted 27 October 2006
Abbreviations: BAL, bronchoalveolar lavage; BALF, BAL fluid; BM-M
s, bone-marrow-derived macrophages; EU, endotoxin units; IFN, interferon; IL, interleukin; KC, keratinocyte-derived cytokine; MIP-2, macrophage inflammatory protein-2; TLR, Toll-like receptor.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Legionella usually infects humans via inhalation of contaminated aerosols from waterborne environmental sources. In lung tissue, the bacteria multiply in several types of host cell, including macrophages, monocytes and alveolar epithelial cells (Horwitz & Silverstein, 1980; Mody et al., 1993; Nash et al., 1984). Previous studies have demonstrated the important role of Th1 cytokines, such as interferon (IFN)-
, interleukin (IL)-12 and IL-18, in L. pneumophila infection models (Brieland et al., 1998, 2000; Gebran et al., 1994; Salins et al., 2001; Skerrett & Martin, 1994) and in human samples from pneumonia patients (Tateda et al., 1998). Recently, we reported a crucial role of neutrophils as a source of IL-12 in driving Th1-type host responses in L. pneumophila pneumonia (Tateda et al., 2001a). Influx of inflammatory cells, macrophages and neutrophils to the site of infection is believed to be critical for host defence responses and immune processes (Carratala et al., 1994). Although robust production of chemokines, including macrophage inflammatory protein-2 (MIP-2), keratinocyte-derived cytokine (KC) and LPS-induced CXC chemokine, has been demonstrated in a murine model of Legionella pneumonia (Tateda et al., 2001b), the pathogenesis of the disease, particularly the mechanisms by which this organism is sensed by host cells and by which the innate immune response is triggered, remains to be elucidated.
The first line of defence against invading bacteria is provided by the innate immune system, which recognizes pathogen-associated molecular patterns, conserved microbial patterns shared by large groups of pathogens but not found in higher eukaryotes (Medzhitov, 2001; Medzhitov & Janeway, 2000; Zhang et al., 2000). Over the last few years, it has become evident that both the recognition and the subsequent response to pathogens are mainly transferred by members of the Toll-like receptor (TLR) family (Aderem & Ulevitch, 2000; Akira & Takeda, 2004; Beutler, 2004; Janeway & Medzhitov, 2002). At least 11 TLRs have been described so far, among which TLR2 and TLR4 are the best-investigated family members. TLR4 was originally recognized as a receptor for LPS of Gram-negative bacteria, whereas TLR2 has been designated the major receptor for Gram-positive bacteria by virtue of its capacity to recognize major cell-wall constituents, such as peptidoglycan, lipoteichoic acid and lipoproteins (Medzhitov, 2001; Takeda et al., 2003). However, recent progress in this field has demonstrated that LPS from certain bacteria, such as Bacteroides, Pseudomonas, Rhizobium and Legionella, is recognized mainly by TLR2, rather than by TLR4 (Erridge et al., 2004; Girard et al., 2003). Consistent with these results, Lettinga et al. (2002) demonstrated that, in pulmonary L. pneumophila infection, C3H/HeJ mice, which display a non-functional gene encoding TLR4, were indistinguishable from control C3H/HeN mice in multiple parameters such as bacterial growth, cytokine production and histopathological changes in the infected lungs. More recently, Akamine et al. (2005) demonstrated that intracellular growth of L. pneumophila was enhanced within bone-marrow-derived macrophages (BM-Ms) from TLR2-deficient mice, but not from TLR4-deficient mice, although the lethal sensitivity of TLR2-deficient mice and an association between cellular responses to L. pneumophila LPS and pathogenesis of the disease remain to be clarified.
The present study was designed to investigate further the role(s) of TLR2 in the pathogenesis of L. pneumophila pneumonia. The data obtained from TLR-deficient mice clearly demonstrated that TLR2, but not TLR4, is a critical molecule for the elaboration of an antibacterial host response against L. pneumophila. The relative contribution and importance of TLRs for detecting live L. pneumophila and its LPS component were compared in BM-Ms from TLR2- and TLR4-deficient and wild-type mice.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bacteria. Clinical isolates of L. pneumophila Suzuki (Tateda et al., 2001a), a serogroup 1 strain originally isolated at Toho University Hospital, were grown for 3–4 days at 37 °C on buffered charcoal yeast extract (BCYE) agar supplemented with L-cysteine and ferric nitrates. A single colony was transferred to 3 ml buffered yeast extract broth and incubated overnight at 37 °C with constant shaking. The bacterial suspension was transferred to fresh buffered yeast extract broth as serial fivefold dilutions and incubated overnight under the same conditions as those described above. After confirmation of bacterial motility, the concentration of bacteria in the broth was determined by measuring the OD600. Post-exponential-phase bacteria were used as challenge organisms (Byrne & Swanson, 1998). The numbers of viable bacteria in the challenge suspensions were determined by plating and incubating organisms on BCYE agar for 4 days.
Escherichia coli LPS and preparation of L. pneumophila LPS.
Commercially available E. coli LPS (O55 : B5; Difco) was used to stimulate BM-Ms from TLR2- and TLR4-deficient and wild-type control mice. LPS of L. pneumophila Suzuki was prepared as described previously (Girard et al., 2003; Yoshizawa et al., 2005). For purification of LPS, the phenol re-extraction step was included to remove lipoprotein contaminants from the preparations. Purified LPS demonstrated an activity of 429 500 endotoxin units (EU) ml–1, which is equivalent to approximately 2x109–4x109 c.f.u. ml–1. As a weighable pellet of purified LPS was not obtained, LPS concentration was expressed in EU ml–1. Protein contamination in the purified LPS was determined to be <5 µg ml–1 in 429 500 EU ml–1 using a protein assay (Bio-Rad).
Induction of L. pneumophila pneumonia in mice. Animals were anaesthetized intramuscularly with 7 mg ketamine and 15 mg xylazine kg–1. The trachea was exposed and 30 µl bacterial suspension, or saline as a control, was injected directly into the trachea with a sterile 26-gauge needle. The skin incision was closed with surgical staple.
Lung harvesting for analysis. At designated time points, mice were sacrificed by CO2 asphyxia. Before lung removal, the pulmonary vasculature was perfused with 1 ml saline via the right ventricle. After removal, whole lungs were homogenized in 1 ml saline using a tissue homogenizer (Omni International) under a vented hood. Portions of homogenates (10 µl) were inoculated onto BCYE agar after serial 1 : 10 dilutions in saline. The remaining homogenates were incubated on ice for 30 min and then centrifuged at 1400 g for 10 min. The supernatants were collected, passed through a 0.45 µm filter (Kanto Chemical) and stored at –40 °C until use.
Bronchoalveolar lavage (BAL) and collection of BAL fluid (BALF). Mice were sacrificed on day 1 after inoculation with bacteria and BAL was performed. The trachea was exposed and intubated using a 1.7 mm outer diameter polyethylene catheter. BAL was performed by instilling 2 ml PBS containing 5 mM EDTA and BALF was pooled for each animal. Leukocyte numbers were counted with a haemocytometer. Cytospins were prepared, stained with May–Giemsa and differential cell counts were performed. The remaining BALF was stored at –40 °C until further analysis.
Preparation of BM-Ms and stimulation with L. pneumophila and LPS.
BM-Ms were prepared from bone marrow exudates of mouse femurs, as described previously (Celada et al., 1984; Swanson & Isberg, 1995). The cells were seeded overnight in 96-well tissue culture plates (104 cells per well) and used for BM-M experiments. After the medium had been changed, BM-Ms were infected with L. pneumophila at the indicated m.o.i. for 2 h. At the end of the infection period (time 0), non-phagocytosed and non-adherent bacteria were removed by two washes with medium. These cells were incubated at 37 °C in a humidified atmosphere containing 5 % CO2. At the indicated time points, culture supernatants were collected and the infected macrophages were lysed, as described previously (Yoshizawa et al., 2005). Harvested bacterial suspensions were plated on BCYE agar plates after serial tenfold dilutions and incubated at 37 °C for 4 days. In some experiments, BM-Ms were incubated with LPS of L. pneumophila or E. coli at the indicated concentrations for 24 h. Supernatants of BM-M cultures were collected and centrifuged at 10 000 r.p.m. for 5 min. Samples were stored at –40 °C until used for ELISA.
ELISAs for MIP-2, KC, IL-6 and IL-12. The levels of each cytokine in the culture supernatant of cells and lung homogenates were determined using a commercially available ELISA kit (DuoSet, ELISA development system; R&D Systems), according to the manufacturer's instructions. The ELISA consistently detected MIP-2, KC, IL-6 and IL-12 concentrations of between 20 and 4000 pg ml–1. The ELISA did not cross-react with other chemokines or cyokines.
Statistical analysis. Statistical significance was determined using an unpaired, two-tailed, alternative Student's t-test. Survival curves were constructed using the Kaplan–Meier method and analysed using a log rank test. A value of P<0.05 was considered to be significant.
|
RESULTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Wild-type mice started to die from day 4 after infection, with a survival rate of 70 % (14/20) at the end of the observation period. TLR4-deficient mice demonstrated similar kinetics in lethality to wild-type mice, with an overall survival rate of 60 % (12/20). In contrast, significantly higher mortality was observed in TLR2-deficient mice, as survival in these mice was only 35 % (7/20) (P<0.05). These data suggested a protective role for TLR2, but not TLR4, in this murine model of L. pneumophila pneumonia.
|
). Despite a significant difference in lethal sensitivity, there were no differences in bacterial number in the lungs of wild-type and TLR2- and TLR4-deficient mice (n=5). These data suggested that bacterial overgrowth in the lungs did not account for the higher lethal sensitivity observed in TLR2-deficient mice.
|
). IL-12 is required for the generation of protective Th1-type host responses against Legionella infection, whereas MIP-2 and KC are major chemokines involved in neutrophil recruitment. Although L. pneumophila infection induced an increase in the expression of the cytokines in all groups, the extent and pattern differed in the various mutant and wild-type strains. At early time points (6 and 12 h), a reduced production of MIP-2, KC and IL-6 was found in TLR2-deficient mice compared with that observed in wild-type and TLR4-deficient mice. Conversely, significantly higher levels of IL-12 were observed in TLR2-deficient mice than in the wild-type control at 18 and 24 h after infection. No significant differences in IL-12 levels were noted between wild-type and TLR4-deficient mice.
|
Leukocyte accumulation in BALF
To understand further the higher mortality in TLR2-deficient mice in the absence of changes in bacterial clearance, leukocyte accumulation in the air space was examined in infected mice on day 1 (Table 1). Compared with uninfected animals, L. pneumophila administration resulted in an almost tenfold increase in BAL cells within the lungs of wild-type mice, of which >90 % were neutrophils. In TLR4-deficient mice, no substantial differences were observed in cell numbers and differentials from those of wild-type mice. Importantly, a drastic reduction in inflammatory cells was observed in TLR2-deficient mice. Total cell number and numbers of neutrophils, macrophages and lymphocytes in TLR2-deficient mice were reduced by 24.9, 23.4, 38.0 and 46.5 %, respectively, compared with wild-type mice with L. pneumophila infection. These data suggested a reduction in cellular infiltration, especially neutrophils, in the infected lungs of TLR2-deficient mice, but not TLR4-deficient mice, which was associated with a decrease in chemokine responses and with the higher mortality observed in TLR2-deficient mice.
|
responses to L. pneumophila and its LPSs from wild-type and TLR2- and TLR4-deficient mice (5x104 c.f.u. per well) were infected with L. pneumophila (day 0) and bacterial numbers and MIP-2 production were examined at the indicated time points (n=5). As shown in Fig. 4, the bacterial number of L. pneumophila decreased with time in culture. No differences were observed among wild-type and TLR2- and TLR-4-deficient mice when an m.o.i. of 10 was applied to the BM-Ms. Next, we examined MIP-2 production in culture supernatants after stimulation with L. pneumophila, LPS isolated from L. pneumophila and E. coli LPS at various concentrations. L. pneumophila infection induced MIP-2 production from BM-Ms of wild-type and TLR4-deficient mice in an m.o.i.-dependent manner, whereas significantly lower production levels of MIP-2 were noted in BM-Ms of TLR2-deficient mice (Fig. 5a). To try to determine the molecules responsible for this effect, purified LPS from L. pneumophila Suzuki strain was used to stimulate macrophages. As shown in Fig. 5(b), high levels of MIP-2 were produced by BM-Ms isolated from wild-type and TLR4-deficient mice, whereas significantly less MIP-2 was produced from BM-Ms isolated from TLR2-deficient mice in response to L. pneumophila LPS. As expected, TLR4-deficient macrophages did not respond to E. coli LPS, whereas TLR2-deficient macrophages produced amounts of MIP-2 that were comparable to that of wild-type macrophages when similarly stimulated. These data suggested that BM-Ms from TLR2-deficient mice produced considerably less MIP-2 in response to L. pneumophila and its LPS compared with that produced in response to E. coli LPS. These data further support the importance of TLR2-mediated signalling in response to L. pneumophila LPS.
|
|
|
DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our data indicated that TLR2-deficient mice exhibited a higher mortality but no differences in lung bacterial burden. On the other hand, cytokine and chemokine responses in the lungs of TLR2-deficient mice were altered compared with those of TLR4-deficient and wild-type mice. These data suggest that the lethal sensitivity of TLR2-deficient mice to L. pneumophila pneumonia may be governed by host inflammatory responses, rather than by bacterial numbers in the lungs. In this regard, we have reported previously that neutrophil depletion by administration of granulocyte-specific monoclonal antibody RB6-8C5 sensitized mice to enhanced mortality as a result of L. pneumophila infection, which was associated with a shifting of the cytokine balance from Th1 (IFN-, IL-12) towards Th2-type (IL-4, IL-10) host responses, but with no change in bacterial burden in the lungs (Tateda et al., 2001a). Although we did not observe a reduction in the Th1 cytokine IL-12 or skewing of the Th2-type host response (data not shown), lower levels of chemokines and severely reduced inflammatory cell accumulation were demonstrated in the lungs of TLR2-deficient mice. These findings may be associated with exaggerated mortality in TLR2-deficient mice with L. pneumophila pneumonia, as a blockade of CXC chemokine receptor-2, a receptor for CXC chemokines including MIP-2 and KC, strikingly enhances the mortality of infected animals (Tateda et al., 2001b). More recently, impairment of chemokine production and inflammatory cell recruitment, especially neutrophils, was reported in the lungs of TLR2-deficient mice in an aerosolized model of L. pneumophila pneumonia (Hawn et al., 2006).
Yoshida et al. (1991) reported that peritoneal macrophages from C3H/HeJ mice, which carry a non-functional gene encoding TLR4, are permissive for L. pneumophila growth, whereas macrophages from control C3H/HeN mice are resistant. In contrast, Lettinga et al. (2002) demonstrated that, in pulmonary L. pneumophila infection, C3H/HeJ mice were indistinguishable from control C3H/HeN mice in bacterial growth in infected lungs. Others have reported that intracellular growth of L. pneumophila was enhanced within TLR2-deficient macrophages compared with wild-type and TLR4-deficient mice (Akamine et al., 2005). In a mouse pneumonia model, genetic deletion of myeloid differentiation primary response gene-88 (MyD88), a major adaptor protein in TLR signalling, resulted in severe impairment of eradication of L. pneumophila from the lungs, whereas TLR2-deficient mice displayed only a modest reduction in the clearance of bacteria (Archer & Roy, 2006; Hawn et al., 2006). In our hands, no substantial differences in growth of L. pneumophila were observed between TLR2-deficient and wild-type mice in macrophages in vitro or in the in vivo pneumonia model. These data suggest that TLR2-mediated signalling and additional MyD88-dependent, TLR2-independent pathways may be essential for full protection against the growth of L. pneumophila in the lungs.
Several lines of evidence have demonstrated that TLRs are a critical sensor for pathogen-associated molecular patterns, such as LPS, peptidoglycan and flagella (Aderem & Ulevitch, 2000; Akira & Takeda, 2004; Beutler, 2004; Janeway & Medzhitov, 2002). Although TLR4 was originally recognized as a receptor for LPS of Gram-negative bacteria, a more comprehensive analysis demonstrated an expanded repertoire of TLRs that function in LPS recognition by host immune cells. The endotoxins of several pathogens, including Leptospira, Porphyromonas, Chlamydia and Rhizobium, have been reported to stimulate host cells through activation of TLR2 (Erridge et al., 2004; Girard et al., 2003; Hirschfeld et al., 2001; Werts et al., 2001). Girard et al. (2003) reported that LPS of L. pneumophila required TLR2 rather than TLR4 to elicit the expression of CD14 in bone marrow granulocytes. In addition, Braedel-Ruoff et al. (2005) reported that purified L. pneumophila LPS, as well as L. pneumophila (either viable or formalin killed), was able to activate bone marrow dendritic cells from TLR4-deficient mice, but failed to activate bone marrow dendritic cells from TLR2-deficient mice. Our data in BM-Ms stimulated with L. pneumophila LPS are in line with these reports and further demonstrate TLR2 as a recognition molecule of L. pneumophila LPS for induction of MIP-2. The present results also demonstrate that higher concentrations of LPS are able to overcome the lack of TLR2, again suggesting the involvement of other factors. Recently, Hawn et al. (2003) reported that TLR5 plays an essential role in governing host cytokine responses and susceptibility to Legionella infection in humans through the sensing mechanism of flagellin of this organism. Moreover, several investigators, including us, have reported that Naip5- and caspase-1-dependent cytosolic recognition of flagellin and subsequent host responses may be a critical factor for restriction of L. pneumophila in macrophages and in the lungs (Molofsky et al., 2006; Ren et al., 2006; Zamboni et al., 2006). It is likely that cell-surface sensing by TLR(s) and cytosolic recognition by Naip5 may be collaborating and interacting for maximal induction of host innate immune responses, which is an exciting area for future Legionella research.
Probably associated with the above results, reduction of MIP-2, KC and IL-6 production was observed in the L. pneumophila-infected lungs of TLR2-deficient mice at early time points (6 and 12 h). These data were consistent with a recent report in which complete and partial impairment of chemokine/cytokine responses was demonstrated in the L. pneumophila-infected lungs of MyD88- and TLR2-deficient mice, respectively (Hawn et al., 2006). Conversely, we observed an increase in IL-12 at relatively later time points (18 and 24 h) after infection. Although the reason for this result is unknown, a plausible explanation is that exaggerated IL-12 production may be due to a feedback mechanism to compensate for the lack of adequate host responses. A better understanding of inter- and counter-regulation of related cytokines/chemokines in the setting of TLR deficiency is required.
Neumeister et al. (1998) demonstrated that L. pneumophila LPS was about 1000 times less potent in its ability to induce pro-inflammatory cytokines from Mono Mac 6 cells than LPS of members of the Enterobacteriaceae. Several investigators have suggested a unique structure of L. pneumophila LPS, such as lack of negatively charged groups, the presence of a hydrophobic outer core and the length of the fatty acid chains (Erridge et al., 2004; Girard et al., 2003; Hirschfeld et al., 2001; Knirel et al., 1996; Moll et al., 1997). From the viewpoint of evolutionary advantage, it will be of interest to consider the reasons for TLR2-mediated recognition and the relatively weak biological activity of L. pneumophila LPS. Future studies assessing the molecular mechanisms of receptor–ligand interaction between TLR2 and L. pneumophila LPS, and defining the disparity and functional significance of TLR2 or TLR4 signalling in the pathogenesis of a variety of infectious diseases, are warranted.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
8)-Kdo disaccharide in the inner core region and the structure of the complete core region of the Legionella pneumophila serogroup 1 lipopolysaccharide. Carbohydr Res 304, 91–95.[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  G. Hajishengallis, M. Wang, G. J. Bagby, and S. Nelson Importance of TLR2 in Early Innate Immune Response to Acute Pulmonary Infection with Porphyromonas gingivalis in Mice J. Immunol., September 15, 2008; 181(6): 4141 - 4149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bugarcic, K. Hitchens, A. G Beckhouse, C. A Wells, R. B Ashman, and H. Blanchard Human and mouse macrophage-inducible C-type lectin (Mincle) bind Candida albicans Glycobiology, September 1, 2008; 18(9): 679 - 685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Schmeck, J. Lorenz, P. D. N'Guessan, B. Opitz, V. van Laak, J. Zahlten, H. Slevogt, M. Witzenrath, A. Flieger, N. Suttorp, et al. Histone Acetylation and Flagellin Are Essential for Legionella pneumophila-Induced Cytokine Expression J. Immunol., July 15, 2008; 181(2): 940 - 947. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Bhan, G. Trujillo, K. Lyn-Kew, M. W. Newstead, X. Zeng, C. M. Hogaboam, A. M. Krieg, and T. J. Standiford Toll-Like Receptor 9 Regulates the Lung Macrophage Phenotype and Host Immunity in Murine Pneumonia Caused by Legionella pneumophila Infect. Immun., July 1, 2008; 76(7): 2895 - 2904. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
) and TLR2- (
) and TLR4- (
) deficient mice were challenged intratracheally with L. pneumophila Suzuki strain (3x106 c.f.u. per mouse) and mortality was observed twice daily for 10 days after infection. *, P<0.05 compared with the wild-type.
