SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The lipopolysaccharide (LPS) and fimbriae of Porphyromonas gingivalis play important roles in periodontal inflammation and pathogenesis. We investigated fimbriae and LPS from several P. gingivalis strains in terms of relative dependence on Toll-like receptor (TLR) signalling partners or accessory pattern-recognition molecules mediating ligand transfer to TLRs, and determined induced assembly of receptor complexes in lipid rafts. Fimbriae could utilize TLR1 or TLR6 for cooperative TLR2-dependent activation of transfected cell lines, in contrast to LPS and a mutant version of fimbriae which displayed preference for TLR1. Whether used to activate human cell lines or mouse macrophages, fimbriae exhibited strong dependence on membrane-expressed CD14 (mCD14), which could not be substituted for by soluble CD14 (sCD14). In contrast, sCD14 efficiently substituted for mCD14 in LPS-induced cellular activation. LPS-binding protein was more important for LPS- than for fimbria-induced cell activation, whereas the converse was true for CD11b/CD18. Cell activation by LPS or fimbriae required lipid raft function and formation of heterotypic receptor complexes (TLR1-2/CD14/CD11b/CD18), although wild-type fimbriae additionally recruited TLR6. In summary, TLR2 activation by P. gingivalis LPS or fimbriae involves differential dependence on accessory signalling or ligand-binding receptors, which may differentially influence innate immune responses.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Porphyromonas gingivalis is an anaerobic Gram-negative oral bacterium which is strongly associated with periodontal disease and implicated as a contributory factor in the development of atherosclerosis (Zambon et al., 1994; Chun et al., 2005; Desvarieux et al., 2005). This pathogen expresses adhesive filamentous appendages on its cell surface, known as fimbriae, which constitute a major P. gingivalis virulence factor on the basis of studies in animal models of periodontitis or atherosclerosis (Malek et al., 1994; Gibson et al., 2004). Another widely studied surface structure of P. gingivalis is its lipopolysaccharide (LPS), which, however, displays significant structural and biological differences from the prototypical LPS of enteric bacteria (Dixon and Darveau, 2005). Both P. gingivalis LPS (PgLPS) and fimbriae are detected by pattern-recognition receptors (PRRs) of the innate immune system resulting in host cell activation (Hajishengallis et al., 2002; Ogawa et al., 2002; Darveau et al., 2004; Hajishengallis et al., 2005a; Zhou et al., 2005).

Lipopolysaccharide is a classical example of a pathogen-associated molecular pattern (PAMP), i.e. the recognition substrate of PRRs such as the Toll-like receptors (TLRs) (Medzhitov, 2001; Akira and Takeda, 2004). The relatively conserved structure of PAMPs renders them ideal targets for detection by the similarly conserved PRRs (Medzhitov, 2001). On the other hand, virulence factors (e.g. protein adhesins) are responsible for microbial adaptation within a particular host environment and their relative mutability renders them unlikely targets of pattern recognition during the course of evolution (Medzhitov, 2001). However, the converse notion that virulence protein adhesins may have evolved to interact with and possibly exploit certain PRRs cannot be ruled out. Using peptide mapping of fimbrial epitopes, we have recently shown that P. gingivalis fimbriae display a modular structure capable of interacting with several PRRs (Hajishengallis et al., 2005a). Specifically, distinct fimbrial epitopes interact with CD14 or the β2 integrin CD11b/CD18, while TLR2 mediates cell signalling and activation of a predominantly proinflammatory response, including induction of tumour necrosis factor-α (TNF-α) (Hajishengallis et al., 2005a; Harokopakis et al., 2006). The interaction of CD11b/CD18 with fimbriae contributes to TNF-α induction but also results in specific downregulation of IL-12 p70, suggesting a possible microbial evasion strategy (Hajishengallis et al., 2005a).

The PgLPS mode of cell activation involves complex interactions with several PRRs. PgLPS binds to serum-derived LPS-binding protein (LBP) and is transferred to CD14, although the process involved is considerably slower than in the case of enterobacterial LPS (Cunningham et al., 1996). Subsequently, PgLPS can activate TLR2 or TLR4, although activation of the latter additionally requires the MD2 accessory molecule (Darveau et al., 2004). These experiments used a highly purified monophosphorylated, tetra-acylated lipid A preparation devoid of any detectable protein contamination (Darveau et al., 2004). The presence, however, of two related but distinct structural isoforms of lipid A (molecular mass ions of m/z 1435 and m/z 1450) (Darveau et al., 2004) leaves open the possibility that TLR2 and TLR4 may be activated by distinct molecular forms of PgLPS. The heterogeneity that characterizes P. gingivalis lipid A moieties may be environmentally regulated and the resulting changes in PgLPS structure may determine the type of host response (Darveau et al., 2004; Dixon and Darveau, 2005). In an early study by our group, a highly purified PgLPS preparation, displaying a predominant TLR2 agonistic activity at concentrations ≤ 1 µg ml−1, was surprisingly found to antagonize cytokine induction by enterobacterial LPS in monocytic cells (Hajishengallis et al., 2002). This was tentatively attributed to possible competition at the level of TLR4 (Hajishengallis et al., 2002), although recent elegant studies by Coats et al. (2005) have conclusively identified TLR4-bound MD2 as the principal molecular target of this antagonistic activity. These and other pioneering studies on the biological properties of PgLPS (Ogawa et al., 1994; Kirikae et al., 1999) underscore the ‘unusual’ nature of PgLPS and contrast it with the enterobacterial LPS in terms of TLR utilization and biological outcome, perhaps reflecting different adaptation pressure encountered by oral and enteric pathogens.

The ability of LPS and fimbriae from P. gingivalis to both activate TLR2 prompted us to investigate possible similarities or differences in their mode of interaction with the TLR2-centred pattern recognition apparatus of innate immune cells. The TLR2-mediated response to distinct microbial molecules may be influenced by differential dependence on accessory receptors that cooperate with TLR2 for recognition and/or signalling. In this context, different enterobacterial LPS analogues trigger the recruitment of different TLR4-associated coreceptors to lipid raft microdomains resulting in differential host response outcome (Triantafilou et al., 2004). With regard to TLR2, the unusual property of this TLR to use signalling partners, either TLR1 or TLR6 (Ozinsky et al., 2000), presents additional complexity which may result in differential signalling outcomes. To determine the repertoire of accessory molecules that cooperate with TLR2 in the recognition of and signalling in response to PgLPS or fimbriae, we used the monocyte/macrophage model for innate recognition because of the relevance of this cell type in periodontal disease (Muthukuru et al., 2005; Ren et al., 2005). The questions asked involved the relative dependence of PgLPS and fimbriae on binding or signalling receptors that play accessory roles in TLR2 activation and form multireceptor signalling complexes in membrane lipid rafts. The information obtained from these studies is important for better understanding molecular recognition mechanisms at the host–microbial interface, and thus gleaning an insight into innate immune responses and associated pathophysiology.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

TLR6 participates in TLR2-dependent cell activation by native fimbriae but not by PgLPS

Toll-like receptor 2 activation by PgLPS requires the participation of TLR1 as a signalling partner (Darveau et al., 2004). However, it is currently unknown whether TLR2-mediated cell activation in response to fimbriae involves the participation of TLR1 or TLR6. We first confirmed that the TLR2/1 pair is important for PgLPS-induced cell activation and, moreover, included the testing of a potential role for TLR6. PgLPS from different strains was used in case of possible strain-dependent differences. Despite differences in potency for cell activation, PgLPS from strains 381, HG1691 and W50, representing serotypes O1–O3, respectively (Sims et al., 2001), invariably required the presence of TLR1 (but not of TLR6) for TLR2-dependent activation of SW620 epithelial cells (Fig. 1A). These results were verified in similarly transfected human kidney embryonic (HEK) 293T cells, which were additionally stimulated with PgLPS from strain 33277 (Fig. 1B). In the absence of TLR2 co-transfection, transfection of TLR1 alone, TLR6 alone, or TLR1 plus TLR6 did not support cell activation by any of the agonists (not shown). Pam3Cys-Ser-Lys4 lipopeptide (Pam3Cys) and macrophage-activating lipopeptide-2 (MALP-2), prototypical TLR2/1 and TLR2/6 agonists, respectively (Takeuchi et al., 2002), were used as controls for validating the assay and were found to behave as expected (Fig. 1).

image

Figure 1. LPS from different P. gingivalis strains activate TLR2/1 but not TLR2/6. SW620 cells (A) or HEK293T cells (B) were co-transfected with various combinations of TLRs or empty vector (EMV), firefly luciferase reporter genes driven by the IL-8 promoter (SW620 cells) or an NF-κB-dependent promoter (HEK293T), and a Renilla transfection control. After 48 h, the cells were stimulated for 6 h with medium only (no-agonist control), Pam3Cys (20 ng ml−1), MALP-2 (20 ng ml−1) and LPS from the indicated strains of P. gingivalis (Pg) at 10 µg ml−1 (A) or 4 µg ml−1 (B). Cellular activation is reported as relative luciferase activity, normalized to that of unstimulated (no-agonist) cells transfected with reporter and empty vectors. Bars represent the means (A, n = 2; B, n = 3) and the error bars indicate standard deviations (SD), from one set of experiments that was performed twice with similar findings. Asterisks indicate statistically significant (P < 0.05) cellular activation in the presence of TLR2 and co-transfected TLR compared with transfection with TLR2 alone.

Download figure to PowerPoint

When P. gingivalis fimbriae were tested in similarly transfected SW620 cells, they failed to induce substantial cell activation even when TLR2 was co-transfected with TLR1 or TLR6 (not shown). Because P. gingivalis fimbriae require CD14 for TLR2 activation in mouse macrophages (Hajishengallis et al., 2005a), we thought that the absence of cell surface-expressed CD14 from SW620 cells was responsible for the observed inactivity of fimbriae and that the presence of soluble CD14 in the serum-supplemented medium could not effectively replace surface-expressed CD14. Indeed, in a similar experiment which additionally included SW620 cells co-transfected with CD14, the ability of fimbriae to induce TLR2-mediated cell activation was restored (Fig. 2A). On the other hand, the presence of CD14 was not essential for the abilities of Pam3Cys and MALP-2 to activate TLR2/1- and TLR2/6-dependent cell activation respectively (Fig. 2A). Surprisingly, fimbriae from P. gingivalis strain 33277 (but not from strain OZ5001C) did not display a single preference for TLR1 or TLR6 in CD14/TLR2-dependent cell activation. Indeed, the presence of either TLR1 or TLR6 resulted in significant cell activation by fimbriae 33277 compared with that seen in similarly stimulated cells transfected with CD14 and TLR2 alone, i.e. without TLR1 or TLR6 (P < 0.05; Fig. 2A). In contrast, only the presence of TLR1 enhanced the ability of fimbriae OZ5001C to induce significant cell activation compared with background levels observed in cells transfected with CD14 and TLR2 alone (P < 0.05; Fig. 2A). These findings were verified using similarly transfected HEK293T cells, which were additionally stimulated with fimbriae from strain 381 (Fig. 2B). Fimbriae 381 activated both TLR2/1 and TLR2/6, similarly to fimbriae 33277 (Fig. 2B).

image

Figure 2. Porphyromonas gingivalis fimbriae activate TLR2 in association with TLR1 or TLR6, and CD14 is an essential co-receptor. SW620 cells (A) or HEK293T cells (B and C) were co-transfected with various combinations of TLRs, CD14 or empty vector (EMV), firefly luciferase reporter genes driven by the IL-8 promoter (SW620 cells) or an NF-κB-dependent promoter (HEK293T), and a Renilla transfection control. After 48 h, the cells were stimulated for 6 h with medium only (no-agonist control), Pam3Cys (20 ng ml−1), MALP-2 (20 ng ml−1) and fimbriae (F) from the indicated P. gingivalis strains at 4 µg ml−1. Cellular activation is reported as relative luciferase activity, normalized to that of unstimulated (no-agonist) cells transfected with reporter and empty vectors. Results are presented as means ± SD (n = 3) from one of two sets of experiments with similar findings. Asterisks indicate statistically significant (P < 0.05) cellular activation in the presence of TLR2 and co-transfected TLR (with or without CD14) compared with transfection with TLR2 alone (with or without CD14).

Download figure to PowerPoint

Strain OZ5001C is deficient in the expression of the 50 kDa PG2134 protein which is thought to be an accessory component of the fimbrial structure (Yoshimura et al., 1993; Watanabe et al., 1996). If PG2134 is associated with the TLR2/6 agonistic activity of native fimbriae as suggested by the data (Fig. 2A and B), recombinantly expressed fimbrillin subunit (rFimA) should display TLR2/1 but not TLR2/6 activity. Indeed, that was found to be the case as rFimA behaved similarly to mutant fimbriae OZ5001C (Fig. 2C). In the same experiment we included native fimbriae from strain SMF1 which is deficient in the expression of the minor fimbrial Mfa1 protein (Lamont et al., 2002). Fimbriae SMF1 were used for control purposes to eliminate any possibility that traces of undetected Mfa1 protein may have contributed to the TLR2/6 agonistic activity of the native fimbrial preparation. We found that fimbriae SMF1 behaved similarly to fimbriae 33277, as both could readily induce cell activation regardless of whether TLR1 or TLR6 was used to complement TLR2 signalling in CD14-co-transfected HEK293T cells (Fig. 2C). These data indicate that, unlike PgLPS which activates TLR2/1, native fimbriae can activate TLR2 using either TLR1 or TLR6 as signalling partners.

Cell membrane-expressed CD14 is essential for efficient cell activation by fimbriae

The above results from transfected cell lines suggested that serum-derived soluble CD14 (sCD14) may not efficiently replace cell membrane-expressed CD14 (mCD14) for cell activation by fimbriae. Furthermore, our results from Fig. 1A are consistent with earlier reports that sCD14 or mCD14 can efficiently transfer PgLPS to TLR2/1 for induction of cell activation (Darveau et al., 2004). To specifically address the requirement of mCD14 in fimbria-induced cell activation (as monitored by induction of TNF-α release), we used macrophages from wild-type or CD14-deficient mice exposed to increasing concentrations of fimbriae or PgLPS. Although TNF-α induction was significantly higher in wild-type than in CD14-deficient macrophages in response to either PgLPS or fimbriae, the effect of CD14 deficiency had a considerably greater impact on fimbria-induced TNF-α release (Fig. 3A). Specifically, when the agonists were used at 1 µg ml−1, fimbria-stimulated CD14-deficient cells displayed less than 1% of the cytokine-inducing activity of wild-type cells, whereas PgLPS-stimulated CD14-deficient cells displayed 21% of the cytokine-inducing activity of wild-type cells (Fig. 3A). When the agonists were used at 10 µg ml−1, fimbria- and PgLPS-stimulated CD14-deficient macrophages retained less than 10% and 66%, respectively, of the cytokine-inducing activity of corresponding wild-type cells. These experiments were performed in the presence of autologous mouse serum and thus CD14 was completely absent (in either soluble or membrane-bound form) from cultures of CD14-deficient cells. We thus concluded that PgLPS-stimulated cytokine release is to a certain degree independent of CD14, whereas mCD14 is essential for fimbria-stimulated cytokine release. A similar experiment was repeated using PgLPS and fimbriae from several strains at a single concentration (1 µg ml−1), and this time wild-type and CD14-deficient macrophages were stimulated in the presence of either autologous mouse serum or fetal bovine serum (FBS). In the presence of FBS, we observed no significant differences between wild-type or CD14-deficient cells with regard to PgLPS-induced TNF-α release (Fig. 3B). We thus concluded that FBS-derived sCD14 could effectively substitute for mCD14 in PgLPS-stimulated cytokine release. In great contrast, the presence of FBS only poorly supported cytokine induction in fimbria-stimulated CD14-deficient macrophages (Fig. 3B). Similarly, fimbriae (but not PgLPS) from strain 381 failed to induce TNF-α release in CD14-deficient macrophage cultures supplemented with heterologous, wild-type mouse serum (not shown). Thus, unlike mCD14, sCD14 does not facilitate cellular activation by fimbriae.

image

Figure 3. Role of soluble or membrane-expressed CD14 in cytokine induction by P. gingivalis LPS (PgLPS) or fimbriae. Mouse macrophages (mac) from wild-type (wt) mice or mice deficient in CD14 (CD14–/–) were incubated with medium only, PgLPS, or fimbriae from the indicated P. gingivalis strains and at the concentrations shown (a single concentration of 1 µg ml−1 was selected for testing in B). After 16 h, culture supernatants were assayed for induction of TNF-α release by ELISA. FBS was used as a source of soluble CD14 in experiment (A). In experiment (B), FBS or autologous mouse serum (AMS) from wild-type or CD14–/– mice was used. Results are shown as means ± SD (n = 3) from one set of experiments, performed twice (A) or thrice (B) with similar findings. Asterisks indicate statistically significant (P < 0.05) differences in cytokine induction by the same agonist in CD14-deficient macrophages versus corresponding wild-type controls.

Download figure to PowerPoint

Differential requirement for LBP in cell activation by PgLPS and fimbriae

We then compared PgLPS and fimbriae in terms of relative dependence on LBP, an additional factor involved in ligand transfer for TLR activation. Serum-derived LBP was previously shown to greatly facilitate the binding of PgLPS to CD14 (Cunningham et al., 1996). Specifically, we examined induction of TNF-α release in human monocytes stimulated with increasing concentrations of PgLPS or fimbriae from several P. gingivalis strains, in the presence or absence of serum (FBS); the latter condition was tested with or without exogenously added LBP. When monocytes were stimulated with ≤ 1 µg ml−1 PgLPS, the presence of LBP (in FBS or exogenously added to FBS-depleted cultures) was associated with significantly higher TNF-α responses compared with those seen in LBP-deprived cell cultures (P < 0.05; Fig. 4A–C). However, the presence of LBP had no significant effect when higher concentrations (≥ 1 µg ml−1) of PgLPS were used (Fig. 4A–C). Interestingly, LBP appeared to exert converse effects on fimbria-induced cytokine release (Fig. 4D–F). Specifically, LBP-supplemented monocyte cultures induced significantly higher TNF-α release compared with LBP-deprived controls, when fimbriae were tested at the highest concentration (10 µg ml−1) but generally not at lower concentrations (Fig. 4D–F). Therefore, LBP is required for maximal TNF-α responses by low concentrations of PgLPS or high concentrations of fimbriae.

image

Figure 4. Role of LBP in cytokine induction by P. gingivalis LPS (PgLPS) or fimbriae. Human monocytes were incubated with medium only, PgLPS (A–C) or fimbriae (D–E) from the indicated P. gingivalis strains and at the concentrations shown. After 16 h, culture supernatants were assayed for induction of TNF-α by ELISA. The cells were incubated in the presence or absence of serum, with or without exogenously added LBP (125 ng ml−1). Results are shown as means ± SD (n = 3) from one of two sets of experiments that yielded similar findings. Asterisks indicate statistically significant (P < 0.05) differences in cytokine induction in the presence of LBP (i.e. ‘no serum + LBP’ or ‘serum’) compared with its absence (‘no serum’).

Download figure to PowerPoint

Role of CD11b/CD18 in cell activation by PgLPS and fimbriae and comparison with TLR2

To determine the role of CD11b/CD18 in cell activation by PgLPS or fimbriae, we examined the ability of wild-type or CD11b-deficient mouse macrophages to induce TNF-α in response to increasing concentrations of PgLPS or fimbriae from P. gingivalis strains 381 and 33277. To better assess the relative impact of CD11b deficiency, TLR2-deficient macrophages were stimulated in parallel with the same agonists. We found that, in contrast to TLR2, CD11b/CD18 played a modest role in TNF-α induction by PgLPS (Fig. 5A and B). Indeed, CD11b-deficient macrophages elicited almost comparable TNF-α responses relative to wild-type controls, and only at the highest PgLPS concentration tested (10 µg ml−1) differences reached statistical significance (Fig. 5A and B). In contrast, CD11b-deficient macrophages induced significantly lower TNF-α responses to various concentrations of fimbriae compared with wild-type controls (P < 0.05; Fig. 5A and B). The TNF-α response to fimbriae was reduced by 21–31% in the absence of CD11b/CD18, but was almost eliminated in the absence of TLR2 (Fig. 5A and B).

image

Figure 5. Role of CD11b/CD18 in cytokine induction by P. gingivalis LPS or fimbriae. Mouse macrophages (mac) from wild-type (wt) mice or mice deficient in CD11b or TLR2 were incubated with medium only, PgLPS or fimbriae from P. gingivalis strains 381 (A) or 33277 (B), and at the indicated concentrations. After 16 h, culture supernatants were assayed for induction of TNF-α by ELISA. TLR2-deficient macrophages were included to assess the effect of CD11b deficiency relative to that of TLR2. Results are shown as means ± SD (n = 3) from one of two sets of experiments that yielded similar findings. Asterisks indicate statistically significant (P < 0.05) differences in cytokine induction between PRR-deficient macrophages and corresponding wild-type controls activated by the same agonist.

Download figure to PowerPoint

It thus appears that TLR2 is a major signalling receptor for TNF-α induction by fimbriae and that CD11b/CD18 may contribute to TNF-α induction in a TLR2-dependent way. To determine whether these observations also apply to whole cells of P. gingivalis, we stimulated macrophages from wild-type mice (C57/BL6 or C3H background) or mice deficient in TLR2, TLR4, CD11b (C57/BL6) or both TLR2 and TLR4 (C3H) with wild-type P. gingivalis 381 or non-fimbriated isogenic mutants (DPG3 and JH1004). We found that macrophages deficient in TLR2 or in both TLR2 and TLR4 elicited diminished TNF-α responses to P. gingivalis regardless of its fimbriation state, compared with wild-type control cells (P < 0.05; Fig. 6A). The effect of CD11b deficiency was less dramatic but it reduced TNF-α release by > 50% in P. gingivalis-stimulated cells (Fig. 6A). In contrast, TLR4-deficient macrophages responded normally to P. gingivalis like wild-type controls (Fig. 6A). The TLR specificity in these experiments was validated using established TLR2 (Pam3Cys), TLR4 (Escherichia coli LPS) and TLR9 (CpG ODN) agonists, which behaved as positive or negative controls as expected (Fig. 6A). It was, however, surprising that CpG ODN not only activated macrophages deficient in both TLR2 and TLR4 (as it was supposed to do) but it did so in a dramatically more potent way compared with wild-type cells (P < 0.05; Fig. 6A). We speculate that TLR9 expression may be upregulated in these double TLR-deficient cells as a compensatory mechanism. PgLPS was reported to function as a weak TLR4 agonist, in addition to its TLR2 activity (Darveau et al., 2004; Coats et al., 2005). In our macrophage activation system, PgLPS is a predominantly TLR2 agonist as it poorly activates TLR2-deficient cells (Fig. 5). However, when tested at high concentrations (10 µg ml−1), PgLPS (but not fimbriae) from strain 381 displayed both TLR2- and TLR4-dependent induction of TNF-α release, the latter accounting for about 28% of the total production (Fig. 6B). The results from Figs 5 and 6 collectively indicate that TLR2 is a major macrophage signalling receptor for TNF-α induction by whole cells of P. gingivalis or purified LPS and fimbriae, whereas CD11b/CD18 contributes substantially to TNF-α induction by whole cells of P. gingivalis or purified fimbriae.

image

Figure 6. TLR2 and CD11b/CD18 mediate induction of TNF-α responses to whole cells of P. gingivalis. A. Mouse macrophages from wild-type (wt) mice or mice deficient in TLR2, TLR4, CD11b, or both TLR2 and TLR4 (TLR2/4) were stimulated for 16 h with heat-inactivated P. gingivalis 381 or isogenic mutants (DPG3 and JH1004) deficient in expression of fimbriae (tested at a multiplicity of infection of 10:1). The cells were also stimulated with control TLR agonists [0.2 µg ml−1 Pam3Cys, TLR2; 0.1 µg ml−1E. coli (Ec)-LPS, TLR4; 5 µm CpG ODN 1826, TLR9]. Induction of TNF-α release in culture supernatants was assayed by ELISA. (B) shows a similar experiment in which wild-type and TLR-deficient macrophages were stimulated with a high concentration (10 µg ml−1) of PgLPS or fimbriae purified from P. gingivalis 381. Results are presented as means ± SD (n = 3) from one of two independent sets of experiments yielding similar findings. Asterisks indicate statistically significant (P < 0.05) inhibition of cytokine release in PRR-deficient cells compared with corresponding wild-type controls.

Download figure to PowerPoint

Lipid raft function is required for cell activation by PgLPS or fimbriae

Membrane lipid rafts are enriched in cholesterol, which plays an important role in maintaining them in a liquid-ordered phase, and thus depletion of cholesterol by methyl-β-cyclodextrin (MCD) results in disruption of lipid raft organization (Christian et al., 1997; Cherukuri et al., 2001; Triantafilou et al., 2002). As lipid rafts can serve as signalling platforms (Cherukuri et al., 2001; Triantafilou et al., 2002), we determined whether cytokine induction by PgLPS or fimbriae requires lipid raft function. For this purpose, monocytes were pre-treated at 37°C for 30 min with 10 mM MCD to deplete the cells of cholesterol. We found that this treatment prior to cell activation significantly inhibited the ability of PgLPS or fimbriae from several P. gingivalis strains to induce release of TNF-α (P < 0.05; Fig. 7). The MCD-mediated inhibition of TNF-α release ranged from 61% to 70% in PgLPS-stimulated monocytes and from 73% to 79% in fimbria-stimulated cells (Fig. 7). However, cholesterol repletion of MCD-treated monocytes effectively reversed the inhibitory effects of MCD (Fig. 7). This finding indicates that the MCD effect was dependent on cholesterol sequestration and cannot be attributed to non-specific toxic effects. Overall, these data suggest that cellular activation by PgLPS or fimbriae depends on functional lipid rafts.

image

Figure 7. Cholesterol-dependent inhibitory effect of methyl-β-cyclodextrin (MCD) on cytokine induction by P. gingivalis LPS (PgLPS) or fimbriae. Monocytes were pre-treated for 30 min with 10 mM MCD to deplete cholesterol, or were pre-treated for 30 min with 10 mM MCD followed by addition of 150 µm cholesterol for an additional 30 min. The MCD- and MCD/cholesterol-pre-treated monocytes, as well as cells pre-treated with medium only, were subsequently stimulated for 16 h with PgLPS or fimbriae from the indicated P. gingivalis strains, or were maintained unstimulated with medium only. Induction of TNF-α release in culture supernatants was determined by ELISA. Results are shown as means ± SD (n = 3) from one of two sets of experiments that yielded similar findings. Asterisks indicate statistically significant (P < 0.05) differences in cytokine induction compared with medium-only pre-treated cells.

Download figure to PowerPoint

PgLPS and fimbriae induce assembly of TLR2-centred heterotypic receptor complexes in lipid rafts

Although CD14 and other glycosyl-phospatidyl-inositol (GPI)-anchored proteins are constitutively found in lipid raft microdomains, TLRs and integrins are recruited there upon activation with appropriate ligands (Pfeiffer et al., 2001; Triantafilou et al., 2002). We hypothesized that PgLPS and fimbriae induce formation of a multireceptor complex, comprising TLR2, CD14 and CD11b/CD18, involved in their innate recognition. Using fluorescence resonance energy transfer (FRET), a biophysical technique which can determine sterical co-association of molecules (Triantafilou et al., 2001; 2002), we examined whether TLR2 co-associates with CD14 and CD11b/CD18, upon cell activation with PgLPS (from strain 381) or fimbriae (from strains 381 and OZ5001C). FRET measures non-radiative transfer of energy from the excited state of a donor molecule to an appropriate acceptor and can occur only when the molecules are located within close proximity (1–10 nm) (Triantafilou et al., 2001; 2002). Energy transfer was measured between TLR2 (donor; labelled with Cy3) and CD14, CD11b/CD18, or MHC Class I (acceptors; labelled with Cy5) in both unstimulated and stimulated human monocytes. There was minimal energy transfer between TLR2 and any of the acceptor receptors in unstimulated cells (Fig. 8A). However, there was significant energy transfer from TLR2 to CD14 or CD11b/CD18 (but not to MHC Class I; negative control) in cells stimulated with PgLPS or fimbriae, compared with unstimulated cells (P < 0.05; Fig. 8A). PgLPS and fimbriae appeared to induce co-association of CD14 and CD11b/CD18 with TLR2 with comparable potency, although PgLPS was used at a higher concentration (2 versus 1 µg ml−1). Taken together with data presented earlier in this manuscript, these results suggest that PgLPS and fimbriae induce the assembly of a TLR2-centred recognition apparatus which mediates cellular activation in response to these agonists in lipid raft microdomains.

image

Figure 8. TLR2 heterotypic associations in response to P. gingivalis LPS (PgLPS) or fimbriae. Human monocytes were stimulated for 10 min with PgLPS (2 µg ml−1) or fimbriae (1 µg ml−1) from P. gingivalis 33277 or OZ5001C. Energy transfer between TLR2 (Cy3) and MHC Class I, CD14, CD11b, TLR1 or TLR6 (Cy5) was measured from the increase in donor (Cy3) fluorescence after acceptor (Cy5) photobleaching. The insert in (A) displays the maximum and minimum energy transfer efficiencies in the system, determined as the energy transfer between two different epitopes on the same molecule (CD14), or between molecules that do not engage in heterotypic associations (CD14 and MHC Class I). The maximum (max) and minimum (min) energy transfer efficiencies in (B) are denoted as horizontal discontinuous lines. Results are shown as means of percentage energy transfer ± SD, calculated from three independent experiments. In (A), asterisks indicate statistically significant (P < 0.05) increase in energy transfer between TLR2 (donor) and indicated receptor (acceptor) upon cell activation compared with energy transfer between the same donor–acceptor pair in unstimulated cells.

Download figure to PowerPoint

Fimbriae 33277 activate TLR2/1 and TLR2/6 whereas fimbriae OZ5001C and PgLPS activate only TLR2/1 (Fig. 2). We thus designed an additional FRET study to address the hypothesis that fimbriae 33277 induce association of TLR2 with either TLR1 or TLR6 in lipid rafts, whereas fimbriae OZ5001C and PgLPS induce association of TLR2 with TLR1 but not with TLR6. Association of TLR2 with CD14 was used as a positive control and as an indication of TLR2 recruitment to lipid rafts, as CD14 is a lipid raft resident molecule (Triantafilou et al., 2002). We found that both native (33277) and mutant (OZ5001C) fimbriae induced association of TLR2 with CD14 and TLR1 (Fig. 8B). Fimbriae 33277, but not fimbriae OZ5001C or PgLPS, additionally induced association of TLR2 with TLR6 (Fig. 8B), thus confirming our hypothesis. We further confirmed that TLR2 is recruited to lipid rafts of fimbria-stimulated cells by demonstrating TLR2 association with another lipid raft marker, the ganglioside GM1 (Triantafilou et al., 2002). Specifically, the percentage energy transfer between TLR2 and GM1 was increased from 6 ± 0.5 in unstimulated cells to 37 ± 1.5 or 36 ± 1.0 in cells stimulated with fimbriae 33277 or OZ5001C respectively.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

A better understanding of the interactions between the host and a pathogen requires a good knowledge of both and is facilitated by integrated approaches at the molecular and cellular level. In this regard, the study of the interactions of LPS and fimbriae from P. gingivalis with innate immune cells may not only contribute to elucidating periodontal inflammation and disease pathogenesis, but also offers excellent molecular tools for the study of pattern-recognition mechanisms of innate immunity.

In this article, we have investigated mechanistic aspects of PgLPS and fimbria recognition by TLR2 and functionally associated PRRs. Perhaps the most striking identified difference between PgLPS and fimbriae was the relative promiscuity of fimbriae in interactions with TLR2 signalling partners. SW620 or HEK293T cells transfected with CD14 and TLR2 required an additional TLR for optimal activation by P. gingivalis native fimbriae and this role was served by either TLR1 or TLR6 (Figs 2 and 8B). This unusual property for a TLR2 agonist displayed by fimbriae stood in sharp contrast to Pam3Cys and PgLPS which preferentially required TLR1 and to MALP-2 which preferentially required TLR6 (Figs 1 and 2). However, there is precedence for a TLR2 agonist (the diacylated lipopeptide Pam2Cys-Ser-Lys4) that exhibits comparable activities towards both the human TLR2/1 and TLR2/6 pairs (Omueti et al., 2005).

The dual utilization of TLR1 or TLR6 by P. gingivalis fimbriae could be accounted for in modular terms. Native fimbriae from strains 33277, 381 and SMF1 could utilize either TLR1 or TLR6 for TLR2-dependent cell activation, in contrast to mutant fimbriae from strain OZ5001C which displayed an absolute preference for TLR1. These findings on cell activation through specific TLRs were correlated with FRET data indicating recruitment and co-association of the same TLRs to lipid rafts. Specifically, native fimbriae (33277) induced association of TLR2 with TLR1 or TLR6, whereas mutant fimbriae (OZ5001C) induced association of TLR2 with TLR1 but not with TLR6. Strain OZ5001C is an isogenic mutant of P. gingivalis 33277 generated by inactivation of the pg2134 gene (Hongo et al., 1999). The pg2134 gene is located downstream of and is regulated by fimA, i.e. the gene encoding for the major structural subunit (fimbrillin) of P. gingivalis fimbriae (Nishikawa et al., 2004). Previous studies by Yoshimura et al. (1993) have suggested that protein PG2134 may be a quantitatively minor component of the native fimbriae. Although the role of PG2134 is uncertain, it might have an accessory function as is the case with CsgB, a minor component of an E. coli fimbrial structure (Soto and Hultgren, 1999). Specifically, CsgB appears to facilitate the polymerization of the major fimbrial subunit, CsgA, and is moreover distributed along the length of the fimbrial structure (Soto and Hultgren, 1999). Apparently, the TLR2/6 activity associated with native fimbriae is attributable to the functionally associated PG2134 protein. This notion is consistent with the observed loss of the TLR2/6 (but not of TLR2/1) activity by recombinantly expressed (in E. coli ) fimbrillin subunit (rFimA; Fig. 2C). Additional studies are warranted to determine whether the PG2134 protein plays a direct role in TLR2/6 activation. Alternatively, the absence of PG2134 from fimbriae OZ5001C may influences the overall fimbrial structure in a way that a TLR2/6 recognizable pattern is affected. Strain SMF1 is also a mutant derivative of P. gingivalis 33277 in which the mfa1 gene, encoding for a shorter fimbrial molecule (Mfa1), was insertionally inactivated (Lamont et al., 2002). The observation that fimbriae SMF1 behaved exactly as fimbriae 33277 indicates that the TLR2/6 activity cannot be attributed to incidental contamination with presumably undetected traces of the Mfa1 fimbrial protein.

The findings that fimbriae were more potent than PgLPS in inducing TNF-α responses in monocytes and macrophages (Figs 3 and 5) were not surprising and the relatively low activity of PgLPS lipid A in inducing certain proinflammatory cytokines, including TNF-α, was previously acknowledged (Ogawa et al., 1994). What was surprising was that the mutant fimbriae (OZ5001C) appeared more proinflammatory than native fimbriae in the TNF-α induction assays (Figs 3 and 7). However, no similar significant differences were observed in NF-κB-dependent activation of the recombinant human receptor system (Fig. 2). In this regard, induction of TNF-α may be more complex than induction of NF-κB-dependent transcription in the recombinant HEK293 cell system. Indeed, besides NF-κB, there are additional transcription factors controlling TNF-α gene expression, such as the LPS-induced TNF-alpha factor (LITAF) (Tang et al., 2003). Therefore, two agonists may have comparable activities for NF-κB activation, although not necessarily for TNF-α induction. At the moment it is uncertain whether differences in TNF-α induction could be attributed to differential utilization of TLR2 signalling partners by the two molecules. Furthermore, if the PG2134 protein, which is missing from fimbriae OZ5001C, promotes the polymerization of fimbriae, then the mutant fimbriae may form oligomeric structures with different physical properties than native fimbriae. The mutant fimbriae also appeared to induce a stronger association of TLR2 with CD11b/CD18 than native fimbriae (Fig. 8). Although CD11b/CD18 contributes to TNF-α induction by fimbriae in a TLR2-dependent way (see below), it is currently uncertain whether a stronger TLR2-CD11b/CD18 co-association can be mechanistically linked to increased cell activation. In fact, this may not be the case with PgLPS which induced a strong association of TLR2 with CD11b/CD18, although the latter receptor did not contribute substantially to PgLPS-induced TNF-α release. These observations warrant further investigation for developing more accurate models of how distinct PRRs cooperate for up- or downregulating the host response. The functional association of TLR2 with CD11b/CD18 in response to different bacterial ligands may not necessarily result in similar cytokine responses if these ligands recruit additional but different coreceptors to the TLR2-CD11b/CD18 complex.

The strong requirement for mCD14 in P. gingivalis fimbria-induced cell activation and the inability of sCD14 to effectively substitute for mCD14 in this respect suggest that fimbriae may be restricted with regard to the cell types they can efficiently activate. Thus, although monocytes/macrophages may be potently activated by fimbriae, other cell types like epithelial cells, which do not express mCD14 (Uehara et al., 2003; Guillot et al., 2004), may be activated relatively weakly by fimbriae. Indeed, that was the case with the SW620 epithelial cells unless they were co-transfected with CD14 (Fig. 2A). The first physical barrier faced by P. gingivalis prior to invasion into gingival connective tissue is the epithelial (Lamont and Yilmaz, 2002). Whether P. gingivalis can come in close contact with epithelial cells through contacts with its fimbriae without activating a robust innate defence response is uncertain and warrants further investigation.

It has been suggested that LBP promotes cellular activation by disaggregating and delivering LPS in monomeric form to CD14 (Tobias et al., 1995). We found that LBP did not significantly promote fimbria-induced cell activation unless very high concentrations (10 µg ml−1) of fimbriae were used. As fimbriae are polymeric molecules some degree of aggregation may occur with increasing concentrations, and thus LBP may facilitate fimbria-induced cell activation in a mode similar to that seen in LPS-activated cells (Tobias et al., 1995).

Although the macrophage TNF-α responses to P. gingivalis fimbriae were almost eliminated by TLR2 deficiency, they were only modestly inhibited (21–31%) by CD11b deficiency. This implied that CD11b/CD18 contributes to TNF-α induction by fimbriae in a TLR2-dependent way. One possibility is that CD11b/CD18 acts as a TLR2 co-receptor, in a mode analogous to CD14 which presents ligands for TLR activation. Another, not necessarily mutually exclusive possibility is that TLR2 signalling is required for activating the high-affinity conformation of CD11b/CD18 (Harokopakis and Hajishengallis, 2005) and thus resulting in efficient binding of fimbriae.

Our findings indicate that cell activation by PgLPS or fimbriae occurs in lipid rafts, which are cholesterol-rich membrane microdomains that partition receptors for various cellular signalling and trafficking processes (Cherukuri et al., 2001; Triantafilou et al., 2002; 2004). Indeed, a cholesterol-sequestering agent (MCD) inhibited the ability of both PgLPS and fimbriae to induce TNF-α release (Fig. 7), a function that requires TLR2 signalling (Figs 2, 5 and 6). Moreover, cell activation by PgLPS or fimbriae induced recruitment and association of TLR2 with CD14 (Fig. 8A), a GPI-anchored molecule that constitutively resides in lipid rafts (Triantafilou et al., 2002). Both bacterial molecules additionally recruited CD11b/CD18 which associated with TLR2 (Fig. 8A). Not long before the discovery of TLRs, Ingalls et al. (1997) showed that the cytoplasmic tails of CD11b/CD18, although indispensable for transducing phagocytic signals, are not required for NF-κB activation in response to binding E. coli LPS. In light of what we now know about co-associations of TLR and accessory PRRs in lipid rafts, it is possible that CD11b/CD18 presents ligands for TLR stimulation, including activation of TLR2. Our findings are consistent with a model according to which induced recruitment of TLR2 and CD11b/CD18 to lipid rafts and formation of CD14-TLR2-CD11b/CD18 receptor complexes co-ordinates cellular activation in response to P. gingivalis.

In summary, despite certain similarities such as dependence on lipid rafts, TLR2 activation by P. gingivalis fimbriae or LPS involves significant differences regarding utilization or dependence on accessory signalling or ligand-binding receptors. Further studies may help correlate these differences with differential induction of the innate immune response. Knowledge on the specific molecular mechanisms of bacterially induced host cell activation is useful for understanding disease pathogenesis and for designing intervention strategies to manipulate the innate response to benefit the host.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Reagents

Recombinant human LBP was purchased from R&D Systems (Minneapolis, MN). MALP-2 was from Axxora LLC (San Diego, CA) and Pam3Cys from EMC Microcollections (Tuebingen, Germany). E. coli LPS and CpG ODN 1826 were from InVivogen (San Diego, CA). MCD and cholesterol were purchased from Sigma-Aldrich (St Louis, MO). LPS from P. gingivalis strains 381, Hg1691, WD50 and 33277 was highly purified by phenol-water extraction and subsequent treatment with DNase I, RNase A and proteinase K, followed by chromatographic purification using a column of Sephacryl S-400 HR (2.5 by 40 cm; Pharmacia Fine Chemicals, Piscataway, NJ) (Hajishengallis et al., 2002). The purity of the preparation was confirmed by immunodiffusion analysis and sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) with silver staining. Moreover, colloidal gold staining revealed no detectable protein contamination in any of the preparations (5 µg of purified LPS applied per lane). Fimbriae were purified from P. gingivalis 381 as previously described (Harokopakis and Hajishengallis, 2005) or from strains 33277, SMF1 and OZ5001C as described by Yoshimura et al. (1984). rFimA was purified from E. coli BL21(DE3) transformed with the fimA gene of P. gingivalis 381 as previously described (Amano et al., 1996) with an additional step involving chromatography through agarose-immobilized polymyxin B to remove residual endotoxin. The final fimbrial preparations were free of any contaminating substances on silver-stained SDS-PAGE, and tested negative for endotoxin (< 6 EU per mg of protein) according to quantitative Limulus amebocyte lysate assay (BioWhittaker, Walkersville, MD). Whole cells of P. gingivalis 381 or isogenic non-fimbriated mutants [DPG3 (Malek et al., 1994) or JH1004, kindly donated by Dr H.K. Kuramitsu, University at Buffalo, Buffalo, NY] were used in cytokine induction assays (below). All P. gingivalis strains were grown anaerobically at 37°C in brain–heart infusion broth supplemented with haemin (5 µg ml−1) and menadione (1 µg ml−1).

Cell culture

Monocytes were purified from human peripheral blood upon centrifugation over NycoPrep™1.068 (Axis-Shield, Oslo, Norway) (Harokopakis and Hajishengallis, 2005). Incidental non-monocytes were removed by magnetic depletion using a cocktail of biotin-conjugated monoclonal antibodies (mAbs) and magnetic microbeads coupled to anti-biotin mAb (Monocyte isolation kit II; Miltenyi Biotec, Auburn, CA). Purified monocytes were cultured at 37°C and 5% CO2 atmosphere, in RPMI 1640 (InVitrogen/Gibco, Carlsbad, CA) supplemented with 10% heat-inactivated FBS, 2 mM l-glutamine, 10 mM Hepes, 100 U ml−1 penicillin G, 100 µg ml−1 streptomycin and 0.05 mM 2-mercaptoethanol (complete RPMI). HEK293T cells were cultured in Dulbecco’s modified Eagle’s medium containing 10% heat-inactivated FBS, 100 units ml−1 penicillin and 100 µg ml−1 streptomycin (InVitrogen/Gibco). The human colonic epithelial cell line SW620 (ATCC CCL-227) was cultured in RPMI 1640 containing 10% heat-inactivated FBS, 2 mM l-glutamine, 100 U ml−1 penicillin G and 100 µg ml−1 streptomycin. Thioglycollate-elicited macrophages were isolated from the peritoneal cavity of mice deficient in CD14 (Moore et al., 2000), CD11b (Coxon et al., 1996), TLR2 (Wooten et al., 2002), TLR4(Hoshino et al., 1999), both TLR2 and TLR4 (Vabulas et al., 2002), or from wild-type control mice (C57BL/6 or C3H/HeOuJ; The Jackson Laboratory, Bar Harbor, Maine), as previously described (Hajishengallis et al., 2005a,b). The mice harbouring homozygous TLR2 and TLR4 mutations were ninefold backcrossed towards the C3H genetic background (kindly donated by Dr Carsten Kirschning, Technical University of Munich, Germany). Mouse macrophages were cultured in complete RPMI, except for the use of autologous mouse serum in assays involving comparison of CD14-deficient macrophages against wild-type controls. Cell viability was monitored using the CellTiter-Blue™ assay kit (Promega, Madison, WI). None of the experimental treatments (cell stimulation with P. gingivalis, purified molecules thereof, or other agonists, cell treatment with MCD; see below) affected cell viability compared with medium-only control treatments.

Reporter gene assays

Reporter assays based on inducible luciferase activity were performed essentially as previously described (Hajishengallis et al., 2005b; Omueti et al., 2005). Briefly, SW620 or HEK293T cells were transiently transfected with various combinations of plasmids encoding human PRRs (CD14, TLR1, TLR2, TLR6), which are not endogenously expressed by these cell lines or are expressed at very low, essentially non-functional levels. Specifically, SW620 cells do not express detectable TLR1 or TLR2, whereas TLR6 message is hardly detectable even after extended PCR cycles (Omueti et al., 2005). Moreover, SW620 cells do not express CD14 (Uehara et al., 2003). HEK293 cells express only low endogenous levels of TLR1 and TLR6, whereas TLR2 is completely absent (Massari et al., 2006). Any endogenous levels of TLR1 or TLR6 in these cell lines were apparently insufficient to confer substantial responsiveness to TLR2 agonists in TLR2-only transfected HEK293 cells (see Results; Figs 1B and 2B and C). To monitor cellular activation, the cells were co-transfected with a firefly luciferase reporter gene controlled by the IL-8 promoter (Omueti et al., 2005) (SW620) or by five tandem repeats of NF-κB consensus sequence cloned upstream of a basic promoter (pNF-κB-Luc; Stratagene, La Jolla, CA) (HEK293T), along with pRL-null, a Renilla luciferase transfection control (Promega, Madison, WI). Transfections were performed using the FuGene 6 transfection reagent (Roche Applied Science, Indianapolis, IN) at a reagent to DNA ratio of 4:1 (SW620) or 3:1 (HEK293T), according to the manufacturer’s instructions. The total amount of transfected plasmids was kept constant by supplementing with corresponding empty control vectors. Two days after transfection, the cells were stimulated with agonists as per experimental protocol. After 6 h of stimulation, the cells were lysed and the Renilla and firefly luciferase activities (in relative light units) were measured using the Dual-Glo™ luciferase reporter assay system (Promega) and a Clarity™ luminescence microplate reader (Bio-Tek, Winooski, VT). Luciferase activity was calculated as a ratio of firefly luciferase activity to Renilla luciferase activity, to correct for transfection efficiency. The results were then normalized to those of unstimulated control cells transfected with reporter and empty vectors, the value of which was taken as 1.

Cytokine induction assay

Human monocytes or mouse macrophages (2 × 105 well−1) were stimulated with PgLPS or fimbriae (0.1–10 µg ml−1) or heat-inactivated cells of P. gingivalis (from strain 381 or its isogenic fimbria-deficient mutants DPG3 (Malek et al., 1994) or JH1004, kindly provided by Dr H.K. Kuramitsu, University at Buffalo (NY) for 16 h at 37°C. Induction of release of TNF-α in culture supernatants was measured by ELISA using kits obtained from eBioscience (San Diego, CA) (Hajishengallis et al., 2005a).

MCD treatment and cholesterol reconstitution

To deplete human monocytes of cholesterol using MCD as well as to reconstitute cellular cholesterol in MCD-treated cells, we used a modification of previously published methodology (Christian et al., 1997; Lawrence et al., 2003). Briefly, human monocytes were incubated in the presence of 10 mM MCD for 30 min at 37°C to deplete the cells of cholesterol. The cells were washed and incubated for an additional 30 min with medium only or with 150 µm cholesterol. Subsequently, the cells (MCD-treated, MCD-treated and cholesterol-reconstituted, and cells treated with medium only) were incubated for 16 h with agonists for cytokine induction assays.

Fluorescence resonance energy transfer (FRET)

The procedures for measuring the efficiency of energy transfer between cell surface receptors have been previously described in detail (Triantafilou et al., 2001; 2002; 2004). Briefly, human monocytes were cultured on microchamber culture slides (Laboratory-tek, InVitrogen/Gibco). Following treatment with medium only, PgLPS or fimbriae for 10 min at 37°C, the cells were labelled with 100 µl of a mixture of Cy3-conjugated mAb to TLR2 (donor) and Cy5-conjugated mAb to CD14, CD11b, TLR1 or TLR6 (acceptors). MAb to MHC Class I conjugated to Cy5 was used for control purposes. Upon labelling, the cells were rinsed twice with phosphate-buffered saline containing 0.02% bovine serum albumin and then fixed with 4% paraformaldehyde for 15 min. Cell fixation was necessary to prevent potential reorganization of the proteins during the course of the experiment and energy transfer determinations. Energy transfer between different receptor pairs was calculated from the increase in donor fluorescence after acceptor photobleaching. The following mAbs (clones) were used for FRET: Anti-TLR2 (TL2.1) and anti-TLR1 (GD2.F4) from HyCult, Denmark; anti-TLR6 (86B1153.2) from Imgenex, San Diego, CA; anti-CD14 (Tük 4) and anti-MHC class I (W6/32) from Abcam, Cambridge, UK; anti-CD11b (CBL145) from Chemicon, Hampshire, UK. Anti-CD14 mAb, clone 26ic, was purified from hybridoma supernatant (ATCC HB246; Manassas, VA). The conjugation of antibodies to Cy3 or Cy5 was performed using kits from Amersham Biosciences (Piscataway, NJ).

Statistical analysis

Data were evaluated by analysis of variance and the Dunnett multiple-comparison test using the InStat program (GraphPad Software, San Diego, CA). Where appropriate (comparison of two groups only), two-tailed t-tests were also performed. Statistical differences were considered significant at the level of P < 0.05. Experiments were performed using triplicate samples and were performed twice or more to verify the results.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

This work was supported by U.S. Public Health Service Grants DE015254 (to G.H.), AI052344 (to R.I.T.) from the National Institutes of Health; by the Wellcome Trust and the Heart Research Fund, UK (to K.T.); and by Grants-in-Aid for Scientific Research (15591957 to F.Y. and 17791318 to S.N.) from the Japan Society for the Promotion of Science and by the AUG High-Tech Research Center Project from the Ministry of Education, Culture, Sports, Science and Technology, Japan (F.Y.).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Akira, S., and Takeda, K. (2004) Toll-like receptor signalling. Nat Rev Immunol 4: 499511.
  • Amano, A., Sharma, A., Lee, J.Y., Sojar, H.T., Raj, P.A., and Genco, R.J. (1996) Structural domains of Porphyromonas gingivalis recombinant fimbrillin that mediate binding to salivary proline-rich protein and statherin. Infect Immun 64: 16311637.
  • Cherukuri, A., Dykstra, M., and Pierce, S.K. (2001) Floating the raft hypothesis: lipid rafts play a role in immune cell activation. Immunity 14: 657660.
  • Christian, A.E., Haynes, M.P., Phillips, M.C., and Rothblat, G.H. (1997) Use of cyclodextrins for manipulating cellular cholesterol content. J Lipid Res 38: 22642272.
  • Chun, Y.H., Chun, K.R., Olguin, D., and Wang, H.L. (2005) Biological foundation for periodontitis as a potential risk factor for atherosclerosis. J Periodontal Res 40: 8795.
  • Coats, S.R., Pham, T.T., Bainbridge, B.W., Reife, R.A., and Darveau, R.P. (2005) MD-2 mediates the ability of tetra-acylated and penta-acylated lipopolysaccharides to antagonize Escherichia coli lipopolysaccharide at the TLR4 signaling complex. J Immunol 175: 44904498.
  • Coxon, A., Rieu, P., Barkalow, F.J., Askari, S., Sharpe, A.H., Von Andrian, U.H., et al. (1996) A novel role for the β2 integrin CD11b/CD18 in neutrophil apoptosis: a homeostatic mechanism in inflammation. Immunity 5: 653666.
  • Cunningham, M.D., Seachord, C., Ratcliffe, K., Bainbridge, B., Aruffo, A., and Darveau, R.P. (1996) Helicobacter pylori and Porphyromonas gingivalis lipopolysaccharides are poorly transferred to recombinant soluble CD14. Infect Immun 64: 36013608.
  • Darveau, R.P., Pham, T.T., Lemley, K., Reife, R.A., Bainbridge, B.W., Coats, S.R., et al. (2004) Porphyromonas gingivalis lipopolysaccharide contains multiple lipid A species that functionally interact with both toll-like receptors 2 and 4. Infect Immun 72: 50415051.
  • Desvarieux, M., Demmer, R.T., Rundek, T., Boden-Albala, B., Jacobs, D.R., Jr, Sacco, R.L., and Papapanou, P.N. (2005) Periodontal microbiota and carotid intima-media thickness: the Oral Infections and Vascular Disease Epidemiology Study (INVEST). Circulation 111: 576582.
  • Dixon, D.R., and Darveau, R.P. (2005) Lipopolysaccharide heterogeneity: innate host responses to bacterial modification of lipid a structure. J Dent Res 84: 584595.
  • Gibson, F.C., 3rd, Hong, C., Chou, H.H., Yumoto, H., Chen, J., Lien, E., et al. (2004) Innate immune recognition of invasive bacteria accelerates atherosclerosis in apolipoprotein E-deficient mice. Circulation 109: 28012806.
  • Guillot, L., Medjane, S., Le-Barillec, K., Balloy, V., Danel, C., Chignard, M., and Si-Tahar, M. (2004) Response of human pulmonary epithelial cells to lipopolysaccharide involves Toll-like receptor 4 (TLR4)-dependent signaling pathways: evidence for an intracellular compartmentalization of TLR4. J Biol Chem 279: 27122718.
  • Hajishengallis, G., Martin, M., Schifferle, R.E., and Genco, R.J. (2002) Counteracting interactions between lipopolysaccharide molecules with differential activation of Toll-like receptors. Infect Immun 70: 66586664.
  • Hajishengallis, G., Ratti, P., and Harokopakis, E. (2005a) Peptide mapping of bacterial fimbrial epitopes interacting with pattern recognition receptors. J Biol Chem 280: 3890238913.
  • Hajishengallis, G., Tapping, R.I., Martin, M.H., Nawar, H., Lyle, E.A., Russell, M.W., and Connell, T.D. (2005b) Toll-like receptor 2 mediates cellular activation by the B subunits of type II heat-labile enterotoxins. Infect Immun 73: 13431349.
  • Harokopakis, E., and Hajishengallis, G. (2005) Integrin activation by bacterial fimbriae through a pathway involving CD14, Toll-like receptor 2, and phosphatidylinositol-3-kinase. Eur J Immunol 35: 12011210.
  • Harokopakis, E., Albzreh, M.H., Haase, E.M., Scannapieco, F.A., and Hajishengallis, G. (2006) Inhibition of proinflammatory activities of major periodontal pathogens by aqueous extracts from elder flower (Sambucus nigra). J Periodontol 77: 271279.
  • Hongo, H., Osano, E., Ozeki, M., Onoe, T., Watanabe, K., Honda, O., et al. (1999) Characterization of an outer membrane protein gene, pgmA, and its gene product from Porphyromonas gingivalis. Microbiol Immunol 43: 937946.
  • Hoshino, K., Takeuchi, O., Kawai, T., Sanjo, H., Ogawa, T., Takeda, Y., et al. (1999) Cutting edge: toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J Immunol 162: 37493752.
  • Ingalls, R.R., Arnaout, M.A., and Golenbock, D.T. (1997) Outside-in signaling by lipopolysaccharide through a tailless integrin. J Immunol 159: 433438.
  • Kirikae, T., Nitta, T., Kirikae, F., Suda, Y., Kusumoto, S., Qureshi, N., and Nakano, M. (1999) Lipopolysaccharides (LPS) of oral black-pigmented bacteria induce tumor necrosis factor production by LPS-refractory C3H/HeJ macrophages in a way different from that of Salmonella LPS. Infect Immun 67: 17361742.
  • Lamont, R.J., and Yilmaz, O. (2002) In or out: the invasiveness of oral bacteria. Periodontology 30: 6169.
  • Lamont, R.J., El-Sabaeny, A., Park, Y., Cook, G.S., Costerton, J.W., and Demuth, D.R. (2002) Role of the Streptococcus gordonii SspB protein in the development of Porphyromonas gingivalis biofilms on streptococcal substrates. Microbiology 148: 16271636.
  • Lawrence, J.C., Saslowsky, D.E., Edwardson, J.M., and Henderson, R.M. (2003) Real-time analysis of the effects of cholesterol on lipid raft behavior using atomic force microscopy. Biophys J 84: 18271832.
  • Malek, R., Fisher, J.G., Caleca, A., Stinson, M., Van Oss, C.J., Lee, J.Y., et al. (1994) Inactivation of Porphyromonas gingivalis fimA gene blocks periodontal damage in gnotobiotic rats. J Bacteriol 176: 10521059.
  • Massari, P., Visintin, A., Gunawardana, J., Halmen, K.A., King, C.A., Golenbock, D.T., and Wetzler, L.M. (2006) Meningococcal porin PorB binds to TLR2 and requires TLR1 for signaling. J Immunol 176: 23732380.
  • Medzhitov, R. (2001) Toll-like receptors and innate immunity. Nat Rev Immunol 1: 135145.
  • Moore, K.J., Andersson, L.P., Ingalls, R.R., Monks, B.G., Li, R., Arnaout, M.A., et al. (2000) Divergent response to LPS and bacteria in CD14-deficient murine macrophages. J Immunol 165: 42724280.
  • Muthukuru, M., Jotwani, R., and Cutler, C.W. (2005) Oral mucosal endotoxin tolerance induction in chronic periodontitis. Infect Immun 73: 687694.
  • Nishikawa, K., Yoshimura, F., and Duncan, M.J. (2004) A regulation cascade controls expression of Porphyromonas gingivalis fimbriae via the FimR response regulator. Mol Microbiol 54: 546560.
  • Ogawa, T., Uchida, H., and Amino, K. (1994) Immunobiological activities of chemically defined lipid A from lipopolysaccharides of Porphyromonas gingivalis. Microbiology 140 (Pt 5): 12091216.
  • Ogawa, T., Asai, Y., Hashimoto, M., and Uchida, H. (2002) Bacterial fimbriae activate human peripheral blood monocytes utilizing TLR2, CD14 and CD11a/CD18 as cellular receptors. Eur J Immunol 32: 25432550.
  • Omueti, K.O., Beyer, J.M., Johnson, C.M., Lyle, E.A., and Tapping, R.I. (2005) Domain exchange between human Toll-like receptors 1 and 6 reveals a region required for lipopeptide discrimination. J Biol Chem 280: 3661636625.
  • Ozinsky, A., Underhill, D.M., Fontenot, J.D., Hajjar, A.M., Smith, K.D., Wilson, C.B., et al. (2000) The repertoire for pattern recognition of pathogens by the innate immune system is defined by cooperation between toll-like receptors. Proc Natl Acad Sci USA 97: 1376613771.
  • Pfeiffer, A., Bottcher, A., Orso, E., Kapinsky, M., Nagy, P., Bodnar, A., et al. (2001) Lipopolysaccharide and ceramide docking to CD14 provokes ligand-specific receptor clustering in rafts. Eur J Immunol 31: 31533164.
  • Ren, L., Leung, W.K., Darveau, R.P., and Jin, L. (2005) The expression profile of lipopolysaccharide-binding protein, membrane-bound CD14, and toll-like receptors 2 and 4 in chronic periodontitis. J Periodontol 76: 19501959.
  • Sims, T.J., Schifferle, R.E., Ali, R.W., Skaug, N., and Page, R.C. (2001) Immunoglobulin G response of periodontitis patients to Porphyromonas gingivalis capsular carbohydrate and lipopolysaccharide antigens. Oral Microbiol Immunol 16: 193201.
  • Soto, G.E., and Hultgren, S.J. (1999) Bacterial adhesins: common themes and variations in architecture and assembly. J Bacteriol 181: 10591071.
  • Takeuchi, O., Sato, S., Horiuchi, T., Hoshino, K., Takeda, K., Dong, Z., et al. (2002) Cutting edge: role of Toll-like receptor 1 in mediating immune response to microbial lipoproteins. J Immunol 169: 1014.
  • Tang, X., Fenton, M.J., and Amar, S. (2003) Identification and functional characterization of a novel binding site on TNF-α promoter. Proc Natl Acad Sci USA 100: 40964101.
  • Tobias, P.S., Soldau, K., Gegner, J.A., Mintz, D., and Ulevitch, R.J. (1995) Lipopolysaccharide binding protein-mediated complexation of lipopolysaccharide with soluble CD14. J Biol Chem 270: 1048210488.
  • Triantafilou, K., Triantafilou, M., and Dedrick, R.L. (2001) A CD14-independent LPS receptor cluster. Nat Immunol 2: 338345.
  • Triantafilou, M., Miyake, K., Golenbock, D.T., and Triantafilou, K. (2002) Mediators of innate immune recognition of bacteria concentrate in lipid rafts and facilitate lipopolysaccharide-induced cell activation. J Cell Sci 115: 26032611.
  • Triantafilou, M., Brandenburg, K., Kusumoto, S., Fukase, K., Mackie, A., Seydel, U., and Triantafilou, K. (2004) Combinational clustering of receptors following stimulation by bacterial products determines lipopolysaccharide responses. Biochem J 381: 527536.
  • Uehara, A., Sugawara, S., Watanabe, K., Echigo, S., Sato, M., Yamaguchi, T., and Takada, H. (2003) Constitutive expression of a bacterial pattern recognition receptor, CD14, in human salivary glands and secretion as a soluble form in saliva. Clin Diagn Lab Immunol 10: 286292.
  • Vabulas, R.M., Ahmad-Nejad, P., Ghose, S., Kirschning, C.J., Issels, R.D., and Wagner, H. (2002) HSP70 as endogenous stimulus of the Toll/interleukin-1 receptor signal pathway. J Biol Chem 277: 1510715112.
  • Watanabe, K., Onoe, T., Ozeki, M., Shimizu, Y., Sakayori, T., Nakamura, H., and Yoshimura, F. (1996) Sequence and product analyses of the four genes downstream from the fimbrilin gene (fimA) of the oral anaerobe Porphyromonas gingivalis. Microbiol Immunol 40: 725734.
  • Wooten, R.M., Ma, Y., Yoder, R.A., Brown, J.P., Weis, J.H., Zachary, J.F., et al. (2002) Toll-like receptor 2 is required for innate, but not acquired, host defense to Borrelia burgdorferi. J Immunol 168: 348355.
  • Yoshimura, F., Takahashi, K., Nodasaka, Y., and Suzuki, T. (1984) Purification and characterization of a novel type of fimbriae from the oral anaerobe Bacteroides gingivalis. J Bacteriol 160: 949957.
  • Yoshimura, F., Takahashi, Y., Hibi, E., Takasawa, T., Kato, H., and Dickinson, D.P. (1993) Proteins with molecular masses of 50 and 80 kilodaltons encoded by genes downstream from the fimbrilin gene (fimA) are components associated with fimbriae in the oral anaerobe Porphyromonas gingivalis. Infect Immun 61: 51815189.
  • Zambon, J.J., Grossi, S., Dunford, R., Harazsthy, V.I., Preus, H., and Genco, R.J. (1994) Epidemiology of subgingival bacterial pathogens in periodontal diseases. In Molecular Pathogenesis of Periodontal Disease. Genco, R.J., Hamada, S., Lehrer, J.R., McGhee, J.R., and Mergenhangen, S. (eds). Washington, DC: American Society Microbiology, pp. 312.
  • Zhou, Q., Desta, T., Fenton, M., Graves, D.T., and Amar, S. (2005) Cytokine profiling of macrophages exposed to Porphyromonas gingivalis, its lipopolysaccharide, or its FimA protein. Infect Immun 73: 935943.