Skip to main content
NCBI home page
Infection and Immunity logoLink to Infection and Immunity
. 2003 Dec;71(12):6799–6807. doi: 10.1128/IAI.71.12.6799-6807.2003

Porphyromonas gingivalis Lipopolysaccharide Antagonizes Escherichia coli Lipopolysaccharide at Toll-Like Receptor 4 in Human Endothelial Cells

Stephen R Coats 1,*, Robert A Reife 1, Brian W Bainbridge 1, Thu-Thao T Pham 1, Richard P Darveau 1
PMCID: PMC308937  PMID: 14638766

Abstract

E. coli lipopolysaccharide (LPS) induces cytokine and adhesion molecule expression via the toll-like receptor 4 (TLR4) signaling complex in human endothelial cells. In the present study, we investigated the mechanism by which Porphyromonas gingivalis LPS antagonizes E. coli LPS-dependent activation of human endothelial cells. P. gingivalis LPS at 1 μg/ml inhibited both E. coli LPS (10 ng/ml) and Mycobacterium tuberculosis heat shock protein (HSP) 60.1 (10 μg/ml) stimulation of E-selectin mRNA expression in human umbilical vein endothelial cells (HUVEC) without inhibiting interleukin-1 beta (IL-1β) stimulation. P. gingivalis LPS (1 μg/ml) also blocked both E. coli LPS-dependent and M. tuberculosis HSP60.1-dependent but not IL-1β-dependent activation of NF-κB in human microvascular endothelial (HMEC-1) cells, consistent with antagonism occurring upstream from the TLR/IL-1 receptor adaptor protein, MyD88. Surprisingly, P. gingivalis LPS weakly but significantly activated NF-κB in HMEC-1 cells in the absence of E. coli LPS, and the P. gingivalis LPS-dependent agonism was blocked by transient expression of a dominant negative murine TLR4. Pretreatment of HUVECs with P. gingivalis LPS did not influence the ability of E. coli LPS to stimulate E-selectin mRNA expression. Taken together, these data provide the first evidence that P. gingivalis LPS-dependent antagonism of E. coli LPS in human endothelial cells likely involves the ability of P. gingivalis LPS to directly compete with E. coli LPS at the TLR4 signaling complex.

The role that Porphyromonas gingivalis plays in the development of periodontal disease likely involves its ability to invade the gingiva and modulate innate host inflammatory responses via proteinases and lipopolysaccharide (LPS) (28, 32, 46, 47). Previous studies have demonstrated that P. gingivalis disrupts the ability of gingival epithelial cells to produce interleukin-8 (IL-8) (8). These data suggest that such “chemokine paralysis” suppresses the host's ability to recruit and localize neutrophils to gingival sites of the infection via an IL-8 gradient (48). Gingival fibroblasts are likely to figure prominently in inflammatory responses to P. gingivalis. For example, P. gingivalis LPS has been shown to stimulate the production of a variety of cytokines, including IL-1, IL-6, and IL-8, in gingival fibroblasts, and it is chronic and excessive cytokine production that is believed to participate in tissue destruction during the course of periodontal disease (49). On the other hand, monocytes and human endothelial cells exhibit a low responsiveness to P. gingivalis LPS compared to E. coli LPS (7, 9, 30). In addition, in vivo studies demonstrated the low biological activity of P. gingivalis LPS in stimulating cytokine and adhesion molecule expression in mice (38).

Another key property of P. gingivalis LPS is that it not only fails to stimulate E-selectin expression or p38 mitogen-activated protein kinase activation in human umbilical endothelial cells (HUVEC), but can potently antagonize the ability of E. coli LPS to stimulate adhesion molecule expression in HUVEC (7, 9). This property suggests that P. gingivalis LPS provides a “stealth” function allowing P. gingivalis to escape innate immune system detection via the vasculature during the course of human periodontal disease initiation and progression.

The cellular signaling pathway and mechanism by which P. gingivalis LPS antagonizes E. coli LPS-dependent activation of human endothelial cells has not been identified. The toll-like receptor 4 (TLR4) and its coreceptor, MD-2, are considered to represent the authentic LPS signal transducers in many cell types and therefore are valid candidates for the site of P. gingivalis LPS antagonism (2, 21, 36, 41, 44). Data supporting a role for TLR4 in mediating the ability of P. gingivalis LPS to act as an antagonist for E. coli LPS were recently presented. In one study, P. gingivalis LPS was unable to activate p38 mitogen-activated protein kinase in either human endothelial cells or CHO cells stably expressing human TLR4 and mCD14. In both cell types, P. gingivalis LPS effectively blocked E. coli LPS-dependent activation of p38 mitogen-activated protein kinase (7). Similarly, P. gingivalis LPS does not activate NF-κB in CHO cells stably expressing human TLR4 and mCD14, but is able to antagonize E. coli LPS-dependent NF-κB in these cells (16, 52). However, P. gingivalis LPS is able to activate NF-κB in CHO cells expressing human TLR2, consistent with previous reports indicating that P. gingivalis LPS can function as an agonist through TLR2 (1, 20).

Interestingly, P. gingivalis LPS can act as both an agonist and an antagonist for cytokine release in human monocytes or for adhesion molecule expression in human gingival fibroblasts, and each of these cell types express both TLR4 and TLR2 (16, 52). Therefore, identification of the specific TLR that is responsible for P. gingivalis LPS-dependent agonism or antagonism in a given cell type may be critical to elucidating whether or not distinct or similar molecular mechanisms underlie P. gingivalis LPS-dependent antagonism for different cell types.

Currently, at least two distinct mechanisms have been proposed to account for the ability of P. gingivalis LPS to functionally antagonize E. coli LPS-dependent cell activation. P. gingivalis LPS might bind directly at TLR4 to block E. coli LPS binding and activation at this receptor complex in CHO cells expressing human TLR4 or in human THP-1 monocytes (7, 16). Alternatively, it has been suggested that P. gingivalis LPS might abrogate E. coli LPS-dependent activation indirectly via an LPS-dependent tolerance mechanism involving down-regulation and uncoupling of key TLR signaling components following extended exposure of murine macrophages to LPS (12).

In this study, we present novel evidence that P. gingivalis LPS-mediated antagonism of E. coli LPS in human endothelial cells occurs via a TLR4-mediated mechanism. Although P. gingivalis LPS fails to induce significant E-selectin expression in human endothelial cells, it functions as a weak agonist to elicit TLR4-dependent NF-κB activation in the human microvascular endothelial cell line HMEC-1. Brief pretreatment of HUVEC with P. gingivalis LPS suggests that P. gingivalis LPS does not antagonize E. coli LPS through indirect mechanisms such as sustained negative regulation of TLR4 signaling components. These findings have important implications with regard to elucidating the mechanism of P. gingivalis LPS-dependent antagonism of E. coli LPS in human endothelial cells.

MATERIALS AND METHODS

LPS preparations and cytokines.

E. coli O111:B4 LPS was obtained from Sigma (St. Louis, Mo.). This preparation was further purified by the method of Manthey and Vogel to remove trace contaminating proteins (29) and was used in all experiments performed in this study. The repurified LPS was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and stained for protein by the enhanced colloidal gold procedure to verify purity (29). Colloidal gold staining of reextracted LPS preparations revealed no detectable protein bands compared to primary LPS preparations (data not shown). The repurified E. coli LPS potently stimulates NF-κB activation in HEK cells cotransfected with TLR4 (data not shown). However, the repurified E. coli LPS is unable to activate NF-κB in HEK293 cells cotransfected with TLR2 and TLR6, thus functionally verifying the removal of contaminating lipoproteins (data not shown).

P. gingivalis strain ATCC 33277 was obtained from the American Type Culture Collection. P. gingivalis LPS was prepared by the cold MgCl2-ethanol procedure (10) followed by lipid extraction (14) and conversion to sodium salts (35). Where indicated in the text, the P. gingivalis LPS preparations were further purified by the method of Manthey and Vogel, and analyzed by SDS-PAGE and enhanced colloidal gold staining (29). Colloidal gold staining of reextracted LPS preparations revealed no detectable protein bands compared to primary LPS preparations (data not shown). msbBLPS was isolated from the E. coli mutant strain BMS67C12 as described previously (45). Recombinant Mycobacterium tuberculosis HSP60.1 (a kind gift from Brian Henderson, University of London, London, United Kingdom) was isolated from E. coli by metal chelate affinity chromatography, and contaminating LPS was reported to contribute less than 0.12 ng/μg of purified protein, as described by the authors (26). Further tests performed in our laboratory revealed that polymyxin B (5 μg/ml) completely inhibited E. coli LPS (10 ng/ml) induction of E-selectin mRNA in HUVEC while only partially inhibiting HSP60.1 (10 μg/ml) induction of E-selectin mRNA (data not shown). Recombinant human IL-1β was purchased from R&D Systems (Minneapolis, Minn.). Staphylococcus aureus peptidoglycan was obtained from Fluka (St. Louis, Mo.).

Cell culture.

HUVECs were obtained from Clonetics (San Diego, Calif.). The cells were maintained in growth medium containing medium 199 (Life Technologies; Grand Island, NY) supplemented with 4 mM l-glutamine, 90 μg of heparin per ml, 1 mM sodium pyruvate, 30 μg of endothelial cell growth supplement (Bedford, Mass.) per ml, 100 units of penicillin per ml, 100 μg of streptomycin per ml, and 20% fetal bovine serum (HyClone Laboratories, Logan, Utah). HUVEC stimulation medium is medium 199 supplemented with 4 mM glutamine, 90 μg of heparin per ml, 1 mM sodium pyruvate, 1 mg of human serum albumin per ml, and 5 to 10% human serum (Gemini Bioproducts, Calabasas, Calif.). HMEC-1 (passage 11) were obtained from F. J. Candal (Centers for Disease Control and Prevention; Atlanta, Ga.). HMEC-1 cells were maintained in MCDB-131 medium (Life Technologies, Grand Island, N.Y.) supplemented with 10% fetal bovine serum, 10 ng of epidermal growth factor per ml, 100 units of penicillin per ml, and 100 μg of streptomycin per ml.

Plasmids and expression constructs.

The NF-κB firefly luciferase reporter construct pNF-κB-TA-Luc was obtained from Invitrogen (Carlsbad, Calif.). The β-actin-Renilla luciferase reporter construct and the modified pDisplay expression vector and the plasmids encoding human TLR2 (phTLR2), human mCD14 (phmCD14), and murine TLR4 (pmuTLR4) generously provided by C. Wilson (University of Washington, Seattle) (17, 18). By with PCR-based site-directed mutagenesis, the proline codon encoding amino acid number 712 of the wild-type murine TLR4 coding region was replaced with a histidine codon. This mutation is known to be responsible for inactivating TLR4 in C3H/HeJ mice (36) and confers dominant negative activity against the human TLR4 in HMEC-1 cells (13).

Detection of E-selectin protein expression on HUVEC.

HUVEC (passaged four times) were plated on 96-well plates coated with collagen and grown overnight in HUVEC growth medium. Plates were washed once with phosphate-buffered saline (PBS), and 0.1 ml of HUVEC stimulation medium containing 5% human serum and the appropriate LPS mixture as indicated in the text and figure legends was added to each well and incubated at 37°C for 4 h under 5% CO2. Following stimulation, the plates were washed twice in PBS and processed to detect E-selectin expression as described previously (9). E-selectin expression intensity was reported as absorbance at 450 nm. Data points were plotted with GraphPad Prism software (GraphPad Software Inc., San Diego, Calif.).

Reverse transcription-PCR assay of E-selectin mRNA expression.

HUVEC (passage 4) were plated on 1% gelatin in six-well plates and grown to confluency. Plates were washed once with PBS, and 1 ml of stimulation medium containing 5% human serum and the indicated activators was added to each well, and the stimulations proceeded for 4 h at 37°C and 5%CO2. In some experiments, cells were treated with P. gingivalis LPS (1 μg/ml) for 2 h and washed twice with stimulation medium containing no serum prior to stimulation with E. coli LPS as described above.

Total cellular RNA was harvested by standard phenol-guanidinium isothiocyanate extraction. Template RNAs were then subjected to reverse transcription-PCR with the Access RT-PCR system (Promega, Madison, Wis.); 0.5 μg total of HUVEC RNA was mixed with 0.25 μg of oligo(dT) (12-mer-18-mer mixture), and the mixture was then heated 65°C for 5 min and cooled to room temperature prior to being subjected to RT-PCR.

Reverse transcription was carried out at 48°C for 50 min. The number of cycles required for quantitative determination of target gene expression was determined empirically. Amplification with the E-selectin and β-actin primer sets was performed for 21 cycles at 94°C for 10 s, 55°C for 30 s, and 68°C for 2 min. The primers used to detect E-selectin mRNA expression were as follows: antisense, 5′-GCCAGTGTTCAGCCAGAACT-3′, and sense, 5′-GAATACAGAAGATGGATGT-3′. The primers used to detect β-actin mRNA expression were as follows: antisense, 5′-AGCCCTGGCTGCCTCCAC-3′, and sense, 5′-GTCGGTTGGAGCGAGCATC-3′. PCR products were resolved on 1.5% agarose gels, stained, and visualized, and images were stored with the Eagle-eye gel documentation system (Stratagene, La Jolla, Calif.). Images of PCR products were quantified with Image Quant software (Molecular Dynamics, Sunnyvale, Calif.). The densitometric values obtained for E-selectin PCR products were normalized to the values obtained for β-actin PCR products, and the resulting values were graphed with GraphPad Prism software (GraphPad Software Inc., San Diego, Calif.).

Transient transfection assays.

HMEC-1 cells were cotransfected with the plasmids pNF-κB-TA-Luc and pβ-actin-Renilla luciferase combined either with pDisplay, pmuTLR4P714H, phTLR2, or phmCD14, as described previously, in 24-well culture dishes (13). Transfected cells were stimulated with activators (in triplicate) as indicated in the text and figures. Following 5 h of stimulations in HUVEC stimulation medium containing 5 to 10% human serum, the cells were rinsed once with PBS and lysed with 100 μl of passive lysis buffer (Promega Corporation, Madison, Wis.). The luciferase activity of 12-μl samples was measured on a microplate luminometer with the dual luciferase assay kit (Promega Corporation; Madison, Wis.). Data points were plotted and statistical analyses were performed with GraphPad Prism software (GraphPad Software Inc., San Diego, Calif.). Results are presented as means ± standard deviations of triplicate determinations. Unpaired t tests were performed to determine statistical significance. Asterisks indicate statistically significant (P < 0.05) differences. All experiments were performed at least twice for confirmation.

RESULTS

Ability of crude P. gingivalis LPS and reextracted P. gingivalis LPS to antagonize E. coli LPS induction of E-selectin expression in HUVEC.

Earlier studies demonstrated that crude bacteria and LPS extracted from P. gingivalis ATCC 33277 are much less potent than LPS from bacteria such as E. coli in activating E-selectin expression on human umbilical vein endothelial cells (HUVEC) (9). These studies also demonstrated that P. gingivalis LPS could efficiently antagonize E. coli LPS activation of E-selectin expression between mass ratios of 10:1 and 100:1 (P. gingivalis LPS:E. coli LPS). Other research has raised concerns as to whether the biological activities attributed to P. gingivalis LPS preparations are due to small amounts of lipoproteins that may copurify in these preparations (20, 25, 33).

To address this concern, we subjected a standard preparation of P. gingivalis LPS to reextraction with phenol in order to remove trace proteins (20). As determined by gel electrophoresis followed by colloidal gold staining, protein was undetectable in P. gingivalis LPS preparations (1). Crude P. gingivalis LPS and reextracted P. gingivalis LPS (P. gingivalis LPSRE) were then compared to repurified E. coli LPS for relative ability to induce E-selectin expression in HUVEC (Fig. 1A). P. gingivalis LPS and P. gingivalis LPSRE were similarly poor stimulators of E-selectin expression compared to E. coli LPS over a dosage range of 0.1 ng/ml to 10 μg/ml. Antagonism experiments were performed to test the relative abilities of P. gingivalis LPS and P. gingivalis LPSRE to inhibit E. coli LPS induction of E-selectin expression in HUVEC. Figure 1B and Fig. 1C demonstrate that both P. gingivalis LPS (1 μg/ml) and P. gingivalis LPSRE (1 μg/ml) are effective antagonists to a range of E. coli LPS dosages as determined by induction of E-selectin expression.

FIG. 1.

FIG. 1.

Open in a new tab

P. gingivalis LPS antagonizes E. coli LPS-activation of E-selectin expression in HUVEC as measured by enzyme-linked immunosorbent assay. Cells were treated for 4 h with the indicated doses of LPSs, and E-selectin expression was determined as described in Materials and Methods. Data are represented as the mean absorbance from duplicate samples. (A) Ability of E. coli LPS (EcLPS), P. gingivalis LPS (PgLPS), and P. gingivalis LPSRE to activate E-selectin expression in HUVEC. (B) Ability of P. gingivalis LPS to antagonize E. coli LPS-dependent activation of E-selectin expression in HUVEC. (C) Ability of P. gingivalis LPSRE to antagonize E. coli LPS-dependent activation of E-selectin expression in HUVEC. Results shown are representative of three independent experiments.

P. gingivalis LPS antagonizes E. coli LPS induction but not IL-1β induction of E-selectin mRNA in HUVEC.

P. gingivalis LPS antagonism of E. coli LPS-dependent E-selectin protein expression in HUVEC is due to the suppression of mRNA expression as determined by previous experiments (9). In that report, the specificity of the antagonism was demonstrated by the ability of P. gingivalis LPS to inhibit E. coli LPS-dependent but not tumor necrosis factor alpha-dependent induction of E-selectin mRNA. Since E. coli LPS-dependent signaling pathways and tumor necrosis factor alpha signaling pathways are likely to be divergent both intracellularly and extracellularly (11, 53), the site of antagonism is unclear.

It has recently been reported that TLR4 is the major LPS receptor in HUVEC and HMEC-1 cells, based on RNA expression profiles, anti-TLR4 antibody-blocking experiments, and experiments with dominant negative versions of murine TLR4 (13, 19). The IL-1 receptor (IL-1R) and TLR4 converge intracellularly at the plasma membrane via interaction with myeloid differentiation factor 88 (MyD88) (31, 53). To further probe the specificity and site of P. gingivalis LPS-dependent antagonism, we tested the ability of P. gingivalis LPS to antagonize either E. coli LPS or IL-1β-dependent activation of E-selectin mRNA expression in HUVEC (Fig. 2). P. gingivalis LPS (1 μg/ml) potently inhibited the ability of E. coli LPS (10 ng/ml) to induce E-selectin mRNA expression without modulating the ability of IL-1β (10 ng/ml) to activate E-selectin expression. These data demonstrate the specificity of P. gingivalis LPS-dependent antagonism for the TLR4 pathway and are consistent with the site of P. gingivalis LPS antagonism occurring upstream from the MyD88 adaptor protein.

FIG. 2.

FIG. 2.

Open in a new tab

P. gingivalis LPS (PgLPS) antagonizes E. coli LPS (EcLPS) induction but not Il-1β induction of E-selectin mRNA expression in HUVEC. Cells were plated in six-well culture dishes and treated with the indicated activators for 4 h. Total RNA was harvested, and RT-PCR analysis was performed to detect E-selectin and β-actin mRNA expression as described in Materials and Methods. The resulting RT-PCR products (lower panels) were imaged and subjected to densitometric analysis. E-selectin mRNA expression was normalized to β-actin mRNA expression, and the resulting values were expressed as E-selectin mRNA induction relative to the unstimulated control (upper panel). Results shown are representative of three independent experiments.

P. gingivalis LPS and msbB LPS antagonize M. tuberculosis HSP60.1-dependent as well as E. coli LPS-dependent induction of E-selectin mRNA in HUVEC.

If P. gingivalis LPS antagonizes E. coli LPS at the TLR4 complex in HUVEC, then a non-LPS activator predicted to signal through TLR4 might also be subject to antagonism by P. gingivalis LPS. In addition, if TLR4 is the site of antagonism in HUVEC, then an alternative LPS which is a known antagonist for E. coli LPS might also antagonize the non-LPS TLR4 agonist. The non-LPS substances HSP60 from humans and Chlamydia have been shown to be agonists for TLR4 by other investigators (5, 34). We used M. tuberculosis HSP60.1 for this purpose (26). We also observed that HEK 293 cells transfected with plasmids bearing recombinant human TLR4 and human MD-2 respond to M. tuberculosis HSP60.1. In addition, polymyxin B (5 μg/ml) completely inhibited E. coli LPS (10 ng/ml), but only partially inhibited M. tuberculosis HSP60.1(10 μg/ml) activity, indicating that contaminating endotoxin may have been responsible for only part of the total observed activity (data not shown).

For an alternative LPS antagonist, we used the E. coli LPS variant msbB LPS, which lacks the 14:0 myristic acid on lipid A and can inhibit E. coli LPS stimulation of E-selectin protein expression in HUVEC at doses similar to that of P. gingivalis LPS (45). Results from antagonism experiments with M. tuberculosis HSP60.1 and msbB LPS are shown in Fig. 3 and demonstrate that both P. gingivalis LPS (1 μg/ml) and msbB LPS (1 μg/ml) effectively antagonized either E. coli LPS (10 ng/ml)-dependent or M. tuberculosis HSP60.1 (10 μg/ml)-dependent activation of E-selectin mRNA expression in HUVEC. These data are in good agreement with earlier findings in which P. gingivalis LPS was observed to antagonize the protein TLR4 agonist recombinant FimA in human THP-1 monocytes (16). These results also suggest that HSP60.1 (10 μg/ml) interacts more weakly with the TLR4 signaling complex than E. coli LPS (10 ng/ml), since P. gingivalis LPS and msbB LPS inhibited HSP60.1-dependent cell activation at an approximately 1:1 molar ratio versus the 100:1 molar ratio required for inhibition of E. coli LPS-dependent cell activation. Similarly, it was reported that Rhodobacter sphaeroides lipid A (1 μg/ml) antagonized Chlamydia pneumoniae HSP60 (10 μg/ml) and E. coli LPS (20 ng/ml) activation of NF-κB in HMEC-1 cells at similar efficiencies (5).

FIG. 3.

FIG. 3.

Open in a new tab

P. gingivalis LPS (PgLPS) and msbBLPS antagonize both E. coli LPS (EcLPS) induction and HSP60.1 induction of E-selectin mRNA expression in HUVEC. Cells were plated in six-well culture dishes and treated with the indicated stimulants and doses for 4 h. Total RNA was harvested, and RT-PCR analysis was performed to detect E-selectin and β-actin mRNA expression as described in Materials and Methods. The resulting RT-PCR products (lower panels) were imaged and subjected to densitometric analysis. E-selectin mRNA expression was normalized to β-actin mRNA expression, and the resulting values were expressed as E-selectin mRNA induction relative to the unstimulated control (upper panel). Results shown are representative of three independent experiments.

P. gingivalis LPS antagonizes E. coli LPS-dependent activation of NF-κB in HMEC-1 cells.

We speculated that the ability of P. gingivalis LPS to inhibit E. coli LPS-dependent TLR4 activation involves the suppression of NF-κB activation in endothelial cells. NF-κB is known to play a pivotal role in the ability of LPS and cytokines to stimulate E-selectin expression in HUVEC (4, 40). To test this possibility, HMEC-1 cells were transiently transfected with an NF-κB-luciferase reporter construct, and antagonism experiments were performed. Figure 4 shows that E. coli LPS (10 ng/ml)-dependent NF-κB activation was effectively inhibited by P. gingivalis LPS (1 μg/ml), while P. gingivalis LPS (1 μg/ml) was unable to antagonize IL-1β (10 ng/ml)-mediated NF-κB activation in HMEC-1 cells. Similar results were obtained with both P. gingivalis LPS and P. gingivalis LPSRE (data not shown). This correlates well with P. gingivalis LPS antagonism of E. coli LPS-dependent E-selectin mRNA induction in HUVEC (compare Fig. 2 and Fig. 4). These results indicate that P. gingivalis LPS antagonism of E. coli LPS is due in part to the inhibition of NF-κB activation via blockage of TLR4-mediated signaling.

FIG. 4.

FIG. 4.

Open in a new tab

P. gingivalis LPS (PgLPS) antagonizes E. coli LPS (EcLPS)-dependent activation of NF-κB in HMEC-1. Cells were plated in 24-well culture dishes and cotransfected with pNF-κB-TA-Luc and pβ-actin Renilla luciferase as described in Materials and Methods. The following day, transfected cells were exposed to the indicated stimulants and doses for 5 h and washed with PBS, and the resulting cell lysates were analyzed for firefly luciferase activity and Renilla luciferase activity. Firefly luciferase values were normalized to Renilla luciferase values, and the resulting values are represented as NF-κB activation. Results are presented as means ± standard deviations of triplicate determinations and are representative of three independent experiments. Two independent preparations of P. gingivalis LPS that were used for these experiments yielded similar results.

P. gingivalis LPS and msbB LPS antagonize M. tuberculosis HSP60.1-dependent as well as E. coli LPS-dependent activation of NF-κB in HMEC-1 cells.

Further experiments were performed to explore the hypothesis that P. gingivalis LPS antagonizes E. coli LPS-dependent activation of HMEC-1 cells via modulation of TLR4 signaling. Figure 5 shows that P. gingivalis LPS (1 μg/ml) effectively antagonized M. tuberculosis HSP60.1 (10 μg/ml) as well as E. coli LPS (10 ng/ml). In addition, msbB LPS (1 μg/ml) antagonized both E. coli LPS and M. tuberculosis HSP60.1 effectively. Again, a striking parallel in the pattern of antagonism observed in HUVEC (E-selectin mRNA) and HMEC-1 (NF-κB activation) cells was noted (compare Fig. 3 and Fig. 5). The ability of P. gingivalis LPS and the E. coli LPS variant msbB LPS to similarly antagonize distinct TLR4 agonists (E. coli LPS and M. tuberculosis HSP60.1) provides further support for the hypothesis that the mechanism of P. gingivalis LPS antagonism involves proximal or direct interactions with the TLR4 signaling system.

FIG. 5.

FIG. 5.

Open in a new tab

P. gingivalis LPS (PgLPS) and msbB LPS antagonize both E. coli LPS (EcLPS)-dependent and M. tuberculosis HSP60.1-dependent activation of NF-κB in HMEC-1 cells. Cells were plated in 24-well culture dishes and cotransfected with pNF-κB-TA-Luc and pβ-actin Renilla luciferase as described in Materials and Methods. The following day, transfected cells were exposed to the indicated stimulants and doses for 5 h and washed with PBS, and the resulting cell lysates were analyzed for firefly luciferase activity and Renilla luciferase activity. Firefly luciferase values were normalized to Renilla luciferase values, and the resulting values are represented as NF-κB activation. Results are presented as means ± standard deviations of triplicate determinations and are representative of two independent experiments. Two independent preparations of P. gingivalis LPS that were used for these experiments yielded similar results.

P. gingivalis LPS activates transcription factor NF-κB via the TLR4 receptor complex in HMEC-1 cells.

In the absence of positive functional activity of P. gingivalis LPS at TLR4 in HMEC-1 cells (i.e., P. gingivalis LPS agonism), it remains uncertain whether or not P. gingivalis LPS antagonism of E. coli LPS depends on direct, proximal, or distal interactions relative to the TLR4 receptor complex. Expression studies in vitro or in vivo indicate that crude P. gingivalis LPS does not strongly stimulate innate immune responses in endothelial cells (7, 9, 38). To assess the ability of P. gingivalis LPS to elicit a detectable activation of HMEC-1 cells, cells were transfected with the NF-κB luciferase reporter construct and then exposed to various concentrations of P. gingivalis LPS in the absence of E. coli LPS.

As shown in Fig. 6A, P. gingivalis LPS (100 ng/ml and 1 μg/ml) was able to weakly but significantly (P < 0.05) activate NF-κB in HMEC-1 cells. The TLR2 agonist S. aureus peptidoglycan (10 μg/ml) (42) did not activate HMEC-1 cells unless the cells were transiently cotransfected with a plasmid construct bearing recombinant human TLR2 (Fig. 6B) (13). These data provide strong evidence that the weak activation elicited by P. gingivalis LPS is due to interaction with the TLR4 complex in human endothelial cells (33). In support of this, the weak activation of NF-κB elicited by P. gingivalis LPS was eliminated by ectopic expression of the dominant negative murine TLR4 (Fig. 6C) (13). The dominant negative murine TLR4 effect was also specific for E. coli LPS, a TLR4 agonist, but not for IL-1β, an IL-1R agonist (13) (Fig. 6D). In contrast to the results shown above, other cell types such as murine macrophages and human monocytes are believed to be activated by P. gingivalis LPS through interaction with TLR2 (20, 30). Based upon the data shown in Fig. 6A to 6D, we conclude that P. gingivalis LPS can function as a weak TLR4 agonist in HMEC-1 cells in the absence of E. coli LPS.

FIG. 6.

FIG. 6.

Open in a new tab

P. gingivalis LPS -dependent activation of NF-κB is mediated by TLR4 in HMEC-1 cells. Cells were plated in 24-well culture dishes and cotransfected with pNF-κB-TA-Luc, pβ-actin Renilla luciferase, and either (A) the irrelevant plasmid control, pDisplay, (B) phTLR2, or (C and D) pmuTLR4P712H. The following day, transfected cells were exposed to the indicated doses of P. gingivalis LPS (PgLPS), E. coli LPS (EcLPS), or IL-1β for 5 h and washed with PBS, and the resulting cell lysates were analyzed for firefly luciferase activity and Renilla luciferase activity. Firefly luciferase values were normalized to Renilla luciferase values, and the resulting values are represented as NF-κB activation. Results are presented as means ± standard deviations of triplicate determinations. Asterisks indicate statistically significant (P < 0.05) differences between P. gingivalis LPS- and peptidoglycan-treated cells and the unstimulated control. Results shown are representative of three independent experiments.

Ectopic expression of recombinant mCD14 does not alter the ability of P. gingivalis LPS to antagonize E. coli LPS-dependent NF-κB activation of HMEC-1 cells.

P. gingivalis LPS has been reported to act as an agonist or an antagonist for cell activation depending upon the cell type that is tested (7, 9, 30, 52). The more pronounced ability of P. gingivalis LPS to activate monocytes and gingival fibroblasts has been ascribed to the presence of significant levels of the TLR4 coreceptor mCD14 (43, 50). The lack of substantial mCD14 expression on the surface of HUVEC and HMEC-1 cells may be sufficient to account for the relative inability of these cells to respond efficiently to P. gingivalis LPS (22, 37). This might also explain the ability of P. gingivalis LPS to efficiently antagonize E. coli LPS in endothelial cells.

In order to examine this possibility, HMEC-1 cells were transfected with phmCD14 and then stimulated with P. gingivalis LPS (10 ng/ml to 1 μg/ml) (Fig. 7). These data show that transient expression of recombinant human mCD14 rendered HMEC-1 cells slightly more responsive to P. gingivalis LPS (10 ng/ml) activation of NF-κB compared to control transfections (Fig. 7A). However, the maximal activation of NF-κB was not significantly influenced. Likewise, the ability of P. gingivalis LPS to antagonize E. coli LPS was not appreciably reduced when human mCD14 was expressed in HMEC-1 cells, although E. coli LPS responsiveness was increased compared to that in cells that were transfected with an irrelevant plasmid control, pDisplay (Fig. 7B).

FIG. 7.

FIG. 7.

Open in a new tab

Ectopic expression of mCD14 does not reduce the ability of P. gingivalis LPS (PgLPS) to antagonize E. coli LPS (EcLPS) in HMEC-1 cells. Cells were plated in 24-well culture dishes and cotransfected with pNF-κB-TA-Luc, pβ-actin Renilla luciferase, and either pDisplay or phmCD14 as indicated in panels A and B. The following day, transfected cells were exposed to the indicated doses for 5 h and washed with PBS, and the resulting cell lysates were analyzed for firefly luciferase activity and Renilla luciferase activity. Firefly luciferase values were normalized to Renilla luciferase values, and the resulting values are represented as NF-κB activation. Each treatment was performed in triplicate and expressed as the standard deviation of the mean. (A) P. gingivalis LPS-dependent agonism in HMEC-1 cells transiently expressing human mCD14. (B) P. gingivalis LPS-dependent antagonism of E. coli LPS in HMEC-1 cells transiently expressing human mCD14. Results shown are representative of two independent experiments.

Pretreatment of HUVEC with P. gingivalis LPS does not inhibit the ability of E. coli LPS to stimulate E-selectin mRNA expression.

In the antagonism experiments presented in this study, P. gingivalis LPS and E. coli LPS mixtures are administered simultaneously. However, it cannot be ruled out that P. gingivalis LPS indirectly modulates the ability of endothelial cells to respond to E. coli LPS by negatively regulating signaling components specific to the TLR4 pathway. To test this possibility HUVEC were either mock treated or treated with P. gingivalis LPS (1 μg/ml) for 2 h and washed briefly prior to stimulation with E. coli LPS (10 ng/ml) alone or in combination with P. gingivalis LPS (1 μg/ml) (Fig. 8). As shown in Fig. 8, P. gingivalis LPS pretreatment did not significantly alter the ability of E. coli LPS to induce E-selectin mRNA expression compared to nontreated cells. In addition, the ability of simultaneously administered P. gingivalis LPS to antagonize E. coli LPS-dependent activation of E-selectin mRNA expression was retained. The results of this experiment indicate that exposure of cells to P. gingivalis LPS is not sufficient for sustained inhibition of E. coli LPS-dependent cell activation and that P. gingivalis LPS-dependent antagonism of E. coli LPS in HUVEC requires simultaneous application of these LPSs.

FIG. 8.

FIG. 8.

Open in a new tab

Pretreatment of HUVEC with P. gingivalis LPS (PgLPS) does not influence the ability of E. coli LPS (EcLPS) to induce E-selectin mRNA expression. HUVEC were plated in six-well culture dishes and treated for 2 h with either growth medium alone or growth medium containing P. gingivalis LPS (1 μg/ml). Subsequently, the cells were treated with the indicated LPSs for 4 h. Total RNA was harvested, and RT-PCR analysis was performed to detect E-selectin and β-actin mRNA expression as described in Materials and Methods. The resulting RT-PCR products (lower panels) were imaged and subjected to densitometric analysis. E-selectin mRNA expression was normalized to β-actin mRNA expression, and the resulting values are expressed as E-selectin mRNA induction relative to the unstimulated control (upper panel). Results shown are representative of three independent experiments.

DISCUSSION

It is not known if P. gingivalis LPS antagonizes E. coli LPS-dependent activation of human endothelial cells through direct or indirect interactions with the TLR4 signaling pathway. Since E. coli LPS has been shown to activate human endothelial cells (HUVEC and HMEC-1) through TLR4, P. gingivalis LPS could act directly at the TLR4 receptor complex to compete with E. coli LPS for binding and function (7, 13, 19, 52). Alternatively, P. gingivalis LPS engagement of the TLR4 receptor or another surface receptor might antagonize E. coli LPS-dependent cell activation through an indirect mechanism such as negative regulation of TLR4 signaling components. For example, the TLR4 and IL-1R signaling pathways converge early at the level of the MyD88 adaptor protein and share multiple intracellular signaling molecules that lead to NF-κB activation (3, 31, 51, 53). Therefore, it is plausible that an indirect P. gingivalis LPS-dependent inhibition of TLR4 signaling components would also inhibit IL-1R-mediated signaling. However, we failed to observe any such influence on IL-1β-dependent cell activation in the presence of P. gingivalis LPS (Fig. 2 to 4), indicating that P. gingivalis LPS-dependent antagonism likely occurs upstream of the MyD88 adaptor protein and that it is specific for the TLR4 signaling pathway.

The experimental approach used to investigate antagonism in the present study involves administering a mixture of agonist and antagonist simultaneously to the endothelial cell in the presence of human serum for a relatively short-term exposure (4 h). Consequently, it is not likely that an indirect mechanism such as LPS-induced tolerance is involved in P. gingivalis LPS-dependent antagonism of E. coli LPS in endothelial cells since an extended preexposure of cells to LPS is required (typically 24 h) to elicit desensitization to LPS responsiveness (12, 16, 27, 30). In addition, the establishment of LPS-induced tolerance is likely to require long-term transcriptional changes that occur in negative regulatory TLR signaling components, including IRAK-M, SOCS-1, and MyD88s (6, 15, 23, 24). Nevertheless, it is conceivable that a specific, rapid, and sustained negative regulatory event desensitizes the TLR4 pathway during cell exposure to P. gingivalis LPS, resulting in an indirect inhibition of E. coli LPS-dependent cell activation. Indeed, evidence is emerging for the existence of an early-phase negative regulatory circuit in the TLR4 signaling pathway in monocytes (15). However, we observed that E. coli LPS was able to strongly activate HUVEC regardless of P. gingivalis LPS preexposure and that P. gingivalis LPS must be added concurrently with E. coli LPS to elicit antagonism (Fig. 8). These data demonstrate that P. gingivalis LPS-dependent antagonism of E. coli LPS in human endothelial cells does not involve indirect inhibitory effects such as sustained desensitization of the TLR4 signaling pathway.

The data discussed above do not provide support for an indirect mechanism of antagonism and are consistent with a direct mechanism such as competitive inhibition. However, to date it has been difficult to demonstrate that the P. gingivalis LPS interacts with the TLR4 receptor in human endothelial cells because it has only been reported to act as an antagonist in these cells (7, 9). Interestingly, P. gingivalis LPS may function as an agonist in murine macrophages or human monocytes, presumably through utilization of either TLR4 or TLR2 (7, 20, 33). Human endothelial cells express TLR4 but do not express TLR2 (13, 19) (Fig. 6A and 6B), and they do not express E-selectin in response to P. gingivalis LPS (9). Similarly, CHO cells expressing mCD14 and human TLR4 fail to respond to P. gingivalis LPS, as determined by NF-κB-dependent CD25 expression (16, 52) or p38 mitogen-activated protein kinase activation (7), but P. gingivalis LPS can clearly antagonize E. coli LPS in this system (7, 16, 52).

Unexpectedly, we observed that P. gingivalis LPS was capable of acting as a weak agonist in HMEC-1 cells, as determined by highly sensitive NF-κB-luciferase reporter assays (Fig. 6). We conclude that P. gingivalis LPS-dependent activation of NF-κB in HMEC-1 cells occurs via interaction with TLR4, as determined by the ability of an ectopically expressed dominant negative murine TLR4 to suppress P. gingivalis LPS-dependent NF-κB activation (Fig. 6A and 6C). Thus, the present study provides the first evidence that P. gingivalis LPS acts as both a weak agonist and a strong antagonist at the same receptor (TLR4) in human endothelial cells. These data suggest that direct P. gingivalis LPS interaction with the TLR4 complex is able to modulate the ability of E. coli LPS to activate human endothelial cells. This finding is also consistent with the hypothesis that P. gingivalis LPS antagonizes E. coli LPS by competitive binding of LPS-engaged TLR4 complexes. Interestingly, an analogous mechanism has recently been reported to occur for TLR2. In this case, lung collectin surfactant protein A directly blocks zymosan interaction with TLR2 by competitive binding. This results in functional antagonism of zymosan-dependent macrophage activation (39).

In conclusion, the main findings of this study demonstrate for the first time that P. gingivalis LPS can act as either a weak agonist alone or a strong antagonist in combination with E. coli LPS at the TLR4 signaling complex in human endothelial cells (HUVEC and HMEC-1). P. gingivalis LPS does not appear to antagonize E. coli LPS via an indirect and nonspecific mechanism, since IL-1β-dependent cell activation was not influenced in antagonism experiments. In addition, P. gingivalis LPS pretreatment of HUVEC did not influence the ability of E. coli LPS to activate HUVEC, and antagonism was observed only when E. coli LPS and P. gingivalis LPS were administered simultaneously. Taken together, the data suggest that P. gingivalis LPS antagonism is the outcome of the ability of high concentrations of P. gingivalis LPS to directly compete for E. coli LPS binding sites at the human TLR4 complex in endothelial cells and that in the absence of E. coli LPS, P. gingivalis LPS binding manifests simply as weak agonism. The development of effective E. coli LPS and P. gingivalis LPS binding assays for TLR4 would help to test this possibility. Future studies with recombinant TLR4-MD-2 receptor complexes should also assist in further elucidating the mechanism of P. gingivalis LPS-dependent antagonism of E. coli LPS.

Acknowledgments

We thank A. M. Hajjar and C. B. Wilson for providing the recombinant human TLR2 and human mCD14 plasmid constructs used in this study.

This work was supported by National Institutes of Health grant R01 DE12768.

Editor: W. A. Petri, Jr.

REFERENCES

  • 1.Bainbridge, B. W., S. R. Coats, and R. P. Darveau. 2003. Porphyromonas gingivalis lipopolysaccharide displays functionally diverse interactions with the innate host defense system. Ann. Periodontol. 7:29-37. [DOI] [PubMed] [Google Scholar]
  • 2.Beutler, B. 2000. Tlr4: central component of the sole mammalian LPS sensor. Curr. Opin. Immunol. 12:20-26. [DOI] [PubMed] [Google Scholar]
  • 3.Bowie, A., E. Kiss-Toth, J. A. Symons, G. L. Smith, S. K. Dower, and L. A. O'Neill. 2000. A46R and A52R from vaccinia virus are antagonists of host IL-1 and toll-like receptor signaling. Proc. Natl. Acad. Sci. USA 97:10162-10167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Boyle, E. M., Jr., J. C. Kovacich, T. G. Canty, Jr., E. N. Morgan, E. Chi, E. D. Verrier, and T. H. Pohlman. 1998. Inhibition of nuclear factor-kappa B nuclear localization reduces human E-selectin expression and the systemic inflammatory response. Circulation 98:282-288. [PubMed] [Google Scholar]
  • 5.Bulut, Y., E. Faure, L. Thomas, H. Karahashi, K. S. Michelsen, O. Equils, S. G. Morrison, R. P. Morrison, and M. Arditi. 2002. Chlamydial heat shock protein 60 activates macrophages and endothelial cells through Toll-like receptor 4 and MD2 in a MyD88-dependent pathway. J. Immunol. 168:1435-1440. [DOI] [PubMed] [Google Scholar]
  • 6.Burns, K., S. Janssens, B. Brissoni, N. Olivos, R. Beyaert, and J. Tschopp. 2003. Inhibition of interleukin 1 receptor/Toll-like receptor signaling through the alternatively spliced, short form of MyD88 is due to its failure to recruit IRAK-4. J. Exp. Med. 197:263-268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Darveau, R. P., S. Arbabi, I. Garcia, B. Bainbridge, and R. V. Maier. 2002. Porphyromonas gingivalis lipopolysaccharide is both agonist and antagonist for p38 mitogen-activated protein kinase activation. Infect. Immun. 70:1867-1873. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Darveau, R. P., C. M. Belton, R. A. Reife, and R. J. Lamont. 1998. Local chemokine paralysis, a novel pathogenic mechanism for Porphyromonas gingivalis. Infect. Immun. 66:1660-1665. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Darveau, R. P., M. D. Cunningham, T. Bailey, C. Seachord, K. Ratcliffe, B. Bainbridge, M. Dietsch, R. C. Page, and A. Aruffo. 1995. Ability of bacteria associated with chronic inflammatory disease to stimulate E-selectin expression and promote neutrophil adhesion. Infect. Immun. 63:1311-1317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Darveau, R. P., and R. E. Hancock. 1983. Procedure for isolation of bacterial lipopolysaccharides from both smooth and rough Pseudomonas aeruginosa and Salmonella typhimurium strains. J. Bacteriol. 155:831-838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Delude, R. L., A. Yoshimura, R. R. Ingalls, and D. T. Golenbock. 1998. Construction of a lipopolysaccharide reporter cell line and its use in identifying mutants defective in endotoxin, but not tumor necrosis factor alpha-alpha, signal transduction. J. Immunol. 161:3001-3009. [PubMed] [Google Scholar]
  • 12.Dobrovolskaia, M. A., A. E. Medvedev, K. E. Thomas, N. Cuesta, V. Toshchakov, T. Ren, M. J. Cody, S. M. Michalek, N. R. Rice, and S. N. Vogel. 2003. Induction of in vitro reprogramming by Toll-like receptor (TLR)2 and TLR4 agonists in murine macrophages: effects of TLR “homotolerance” versus “heterotolerance” on NF-kappa B signaling pathway components. J. Immunol. 170:508-519. [DOI] [PubMed] [Google Scholar]
  • 13.Faure, E., O. Equils, P. A. Sieling, L. Thomas, F. X. Zhang, C. J. Kirschning, N. Polentarutti, M. Muzio, and M. Arditi. 2000. Bacterial lipopolysaccharide activates NF-kappaB through toll-like receptor 4 (TLR-4) in cultured human dermal endothelial cells. Differential expression of TLR-4 and TLR-2 in endothelial cells. J. Biol. Chem. 275:11058-11063. [DOI] [PubMed] [Google Scholar]
  • 14.Folch, J., M. Lees, and G. H. S. Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:497-509. [PubMed] [Google Scholar]
  • 15.Fukao, T., and S. Koyasu. 2003. PI3K and negative regulation of TLR signaling. Trends Immunol. 24:358-363. [DOI] [PubMed] [Google Scholar]
  • 16.Hajishengallis, G., M. Martin, R. E. Schifferle, and R. J. Genco. 2002. Counteracting interactions between lipopolysaccharide molecules with differential activation of toll-like receptors. Infect. Immun. 70:6658-6664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hajjar, A. M., R. K. Ernst, J. H. Tsai, C. B. Wilson, and S. I. Miller. 2002. Human Toll-like receptor 4 recognizes host-specific LPS modifications. Nat. Immunol. 3:354-359. [DOI] [PubMed] [Google Scholar]
  • 18.Hajjar, A. M., D. S. O'Mahony, A. Ozinsky, D. M. Underhill, A. Aderem, S. J. Klebanoff, and C. B. Wilson. 2001. Cutting edge: functional interactions between toll-like receptor (TLR) 2 and TLR1 or TLR6 in response to phenol-soluble modulin. J. Immunol. 166:15-19. [DOI] [PubMed] [Google Scholar]
  • 19.Hijiya, N., K. Miyake, S. Akashi, K. Matsuura, Y. Higuchi, and S. Yamamoto. 2002. Possible involvement of toll-like receptor 4 in endothelial cell activation of larger vessels in response to lipopolysaccharide. Pathobiology 70:18-25. [DOI] [PubMed] [Google Scholar]
  • 20.Hirschfeld, M., J. J. Weis, V. Toshchakov, C. A. Salkowski, M. J. Cody, D. C. Ward, N. Qureshi, S. M. Michalek, and S. N. Vogel. 2001. Signaling by toll-like receptor 2 and 4 agonists results in differential gene expression in murine macrophages. Infect. Immun. 69:1477-1482. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Hoshino, K., O. Takeuchi, T. Kawai, H. Sanjo, T. Ogawa, Y. Takeda, K. Takeda, and S. Akira. 1999. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J. Immunol. 162:3749-3752. [PubMed] [Google Scholar]
  • 22.Jersmann, H. P., C. S. Hii, G. L. Hodge, and A. Ferrante. 2001. Synthesis and surface expression of CD14 by human endothelial cells. Infect. Immun. 69:479-485. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Kinjyo, I., T. Hanada, K. Inagaki-Ohara, H. Mori, D. Aki, M. Ohishi, H. Yoshida, M. Kubo, and A. Yoshimura. 2002. SOCS1/JAB is a negative regulator of LPS-induced macrophage activation. Immunity 17:583-591. [DOI] [PubMed] [Google Scholar]
  • 24.Kobayashi, K., L. D. Hernandez, J. E. Galan, C. A. Janeway, Jr., R. Medzhitov, and R. A. Flavell. 2002. IRAK-M is a negative regulator of Toll-like receptor signaling. Cell 110:191-202. [DOI] [PubMed] [Google Scholar]
  • 25.Lee, H. K., J. Lee, and P. S. Tobias. 2002. Two lipoproteins extracted from Escherichia coli K-12 LCD25 lipopolysaccharide are the major components responsible for Toll-like receptor 2-mediated signaling. J. Immunol. 168:4012-4017. [DOI] [PubMed] [Google Scholar]
  • 26.Lewthwaite, J. C., A. R. Coates, P. Tormay, M. Singh, P. Mascagni, S. Poole, M. Roberts, L. Sharp, and B. Henderson. 2001. Mycobacterium tuberculosis chaperonin 60.1 is a more potent cytokine stimulator than chaperonin 60.2 (Hsp 65) and contains a CD14-binding domain. Infect. Immun. 69:7349-7355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Lush, C. W., G. Cepinskas, and P. R. Kvietys. 2000. LPS tolerance in human endothelial cells: reduced PMN adhesion, E-selectin expression, and NF-kappaB mobilization. Am. J. Physiol. Heart Circ. Physiol. 278:H853-861. [DOI] [PubMed] [Google Scholar]
  • 28.Madianos, P. N., P. N. Papapanou, and J. Sandros. 1997. Porphyromonas gingivalis infection of oral epithelium inhibits neutrophil transepithelial migration. Infect. Immun. 65:3983-3990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Manthey, C. L., and S. N. Vogel. 1994. Elimination of trace endotoxin protein from rough chemotype LPS. J. Endotoxin Res. 1:84-91. [Google Scholar]
  • 30.Martin, M., J. Katz, S. N. Vogel, and S. M. Michalek. 2001. Differential induction of endotoxin tolerance by lipopolysaccharides derived from Porphyromonas gingivalis and Escherichia coli. J. Immunol. 167:5278-5285. [DOI] [PubMed] [Google Scholar]
  • 31.Medzhitov, R., P. Preston-Hurlburt, E. Kopp, A. Stadlen, C. Chen, S. Ghosh, and C. A. Janeway, Jr. 1998. MyD88 is an adaptor protein in the hToll/IL-1 receptor family signaling pathways. Mol. Cell 2:253-258. [DOI] [PubMed] [Google Scholar]
  • 32.Nassar, H., H. H. Chou, M. Khlgatian, F. C. Gibson 3rd, T. E. Van Dyke, and C. A. Genco. 2002. Role for fimbriae and lysine-specific cysteine proteinase gingipain K in expression of interleukin-8 and monocyte chemoattractant protein in Porphyromonas gingivalis-infected endothelial cells. Infect. Immun. 70:268-276. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Ogawa, T., Y. Asai, M. Hashimoto, O. Takeuchi, T. Kurita, Y. Yoshikai, K. Miyake, and S. Akira. 2002. Cell activation by Porphyromonas gingivalis lipid A molecule through Toll-like receptor 4- and myeloid differentiation factor 88-dependent signaling pathway. Int. Immunol. 14:1325-1332. [DOI] [PubMed] [Google Scholar]
  • 34.Ohashi, K., V. Burkart, S. Flohe, and H. Kolb. 2000. Cutting edge: heat shock protein 60 is a putative endogenous ligand of the toll-like receptor-4 complex. J. Immunol. 164:558-561. [DOI] [PubMed] [Google Scholar]
  • 35.Peterson, A. A., A. Haug, and E. J. McGroarty. 1986. Physical properties of short- and long-O-antigen-containing fractions of lipopolysaccharide from Escherichia coli O111:B4. J. Bacteriol. 165:116-122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Poltorak, A., X. He, I. Smirnova, M. Y. Liu, C. Van Huffel, X. Du, D. Birdwell, E. Alejos, M. Silva, C. Galanos, M. Freudenberg, P. Ricciardi-Castagnoli, B. Layton, and B. Beutler. 1998. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282:2085-2088. [DOI] [PubMed] [Google Scholar]
  • 37.Pugin, J., C. C. Schurer-Maly, D. Leturcq, A. Moriarty, R. J. Ulevitch, and P. S. Tobias. 1993. Lipopolysaccharide activation of human endothelial and epithelial cells is mediated by lipopolysaccharide-binding protein and soluble CD14. Proc. Natl. Acad. Sci. USA 90:2744-2748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Reife, R. A., R. A. Shapiro, B. A. Bamber, K. K. Berry, G. E. Mick, and R. P. Darveau. 1995. Porphyromonas gingivalis lipopolysaccharide is poorly recognized by molecular components of innate host defense in a mouse model of early inflammation. Infect. Immun. 63:4686-4694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Sato, M., H. Sano, D. Iwaki, K. Kudo, M. Konishi, H. Takahashi, T. Takahashi, H. Imaizumi, Y. Asai, and Y. Kuroki. 2003. Direct binding of toll-like receptor 2 to zymosan, and zymosan-induced NF-kappaB activation and tumor necrosis factor alpha-alpha secretion are down-regulated by lung collectin surfactant protein A. J. Immunol. 171:417-425. [DOI] [PubMed] [Google Scholar]
  • 40.Schindler, U., and V. R. Baichwal. 1994. Three NF-κB binding sites in the human E-selectin gene required for maximal tumor necrosis factor alpha-induced expression. Mol. Cell. Biol. 14:5820-5831. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Schromm, A. B., E. Lien, P. Henneke, J. C. Chow, A. Yoshimura, H. Heine, E. Latz, B. G. Monks, D. A. Schwartz, K. Miyake, and D. T. Golenbock. 2001. Molecular genetic analysis of an endotoxin nonresponder mutant cell line: a point mutation in a conserved region of MD-2 abolishes endotoxin-induced signaling. J. Exp. Med. 194:79-88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Schwandner, R., R. Dziarski, H. Wesche, M. Rothe, and C. J. Kirschning. 1999. Peptidoglycan- and lipoteichoic acid-induced cell activation is mediated by toll-like receptor 2. J. Biol. Chem. 274:17406-17409. [DOI] [PubMed] [Google Scholar]
  • 43.Shapira, L., S. Takashiba, S. Amar, and T. E. Van Dyke. 1994. Porphyromonas gingivalis lipopolysaccharide stimulation of human monocytes: dependence on serum and CD14 receptor. Oral Microbiol. Immunol. 9:112-117. [DOI] [PubMed] [Google Scholar]
  • 44.Shimazu, R., S. Akashi, H. Ogata, Y. Nagai, K. Fukudome, K. Miyake, and M. Kimoto. 1999. MD-2, a molecule that confers lipopolysaccharide responsiveness on Toll-like receptor 4. J. Exp. Med. 189:1777-1782. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Somerville, J. E., Jr., L. Cassiano, B. Bainbridge, M. D. Cunningham, and R. P. Darveau. 1996. A novel Escherichia coli lipid A mutant that produces an antiinflammatory lipopolysaccharide. J. Clin. Investig. 97:359-365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Tada, H., S. Sugawara, E. Nemoto, N. Takahashi, T. Imamura, J. Potempa, J. Travis, H. Shimauchi, and H. Takada. 2002. Proteolysis of CD14 on human gingival fibroblasts by arginine-specific cysteine proteinases from Porphyromonas gingivalis leading to down-regulation of lipopolysaccharide-induced interleukin-8 production. Infect. Immun. 70:3304-3307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Tobias, P. S., J. Gegner, R. Tapping, S. Orr, J. Mathison, J. D. Lee, V. Kravchenko, J. Han, and R. J. Ulevitch. 1997. Lipopolysaccharide dependent cellular activation. J. Periodontal Res. 32:99-103. [DOI] [PubMed] [Google Scholar]
  • 48.Tonetti, M. S., M. A. Imboden, L. Gerber, N. P. Lang, J. Laissue, and C. Mueller. 1994. Localized expression of mRNA for phagocyte-specific chemotactic cytokines in human periodontal infections. Infect. Immun. 62:4005-4014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Wang, P. L., and K. Ohura. 2002. Porphyromonas gingivalis lipopolysaccharide signaling in gingival fibroblasts-cd14 and toll-like receptors. Crit. Rev. Oral Biol. Med. 13:132-142. [DOI] [PubMed] [Google Scholar]
  • 50.Wang, P. L., K. Sato, M. Oido, T. Fujii, Y. Kowashi, M. Shinohara, K. Ohura, H. Tani, and Y. Kuboki. 1998. Involvement of CD14 on human gingival fibroblasts in Porphyromonas gingivalis lipopolysaccharide-mediated interleukin-6 secretion. Arch. Oral Biol. 43:687-694. [DOI] [PubMed] [Google Scholar]
  • 51.Xu, Y., X. Tao, B. Shen, T. Horng, R. Medzhitov, J. L. Manley, and L. Tong. 2000. Structural basis for signal transduction by the Toll/interleukin-1 receptor domains. Nature 408:111-115. [DOI] [PubMed] [Google Scholar]
  • 52.Yoshimura, A., T. Kaneko, Y. Kato, D. T. Golenbock, and Y. Hara. 2002. Lipopolysaccharides from periodontopathic bacteria Porphyromonas gingivalis and Capnocytophaga ochracea are antagonists for human toll-like receptor 4. Infect. Immun. 70:218-225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Zhang, F. X., C. J. Kirschning, R. Mancinelli, X. P. Xu, Y. Jin, E. Faure, A. Mantovani, M. Rothe, M. Muzio, and M. Arditi. 1999. Bacterial lipopolysaccharide activates nuclear factor-kappaB through interleukin-1 signaling mediators in cultured human dermal endothelial cells and mononuclear phagocytes. J. Biol. Chem. 274:7611-7614. [DOI] [PubMed] [Google Scholar]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES