点击显示 收起
Department of Pathology, Boston University Medical Center, Departments of Anesthesia and Pediatrics, Harvard Medical School, and Departments of Anesthesia and Critical Care, Pediatrics
Medicine
Infectious Disease Unit, Massachusetts General Hospital, Boston
Division of Pulmonary and Critical Care Medicine, University of Maryland School of Medicine, Baltimore
Laboratoire de Microbiologie et Génétique Moléculaire, Université Claude Bernard Lyon, France
Sepsis is initiated by interactions between microbial products and host inflammatory cells. Toll-like receptors (TLRs) are central innate immune mediators of sepsis that recognize different components of microorganisms. Peptidoglycan-associated lipoprotein (PAL) is a ubiquitous gram-negative bacterial outer-membrane protein that is shed by bacteria into the circulation of septic animals. We explored the inflammatory effects of purified PAL and of a naturally occurring form of PAL that is shed into serum. PAL is released into human serum by Escherichia coli bacteria in a form that induces cytokine production by macrophages and is tightly associated with lipopolysaccharide (LPS). PAL activates inflammation through TLR2. PAL and LPS synergistically activate macrophages. These data suggest that PAL may play an important role in the pathogenesis of sepsis and imply that physiologically relevant PAL and LPS are shed into serum and act in concert to initiate inflammation in sepsis.
Sepsis-induced inflammation is believed to result from interactions between host cells and microbial toxins. Toll-like receptors (TLRs) are critically involved in early, innate immune responses to infection [1]. TLRs recognize various classes of pathogen-associated molecular patterns (PAMPs) [220]. It is generally accepted that TLR4 mediates responses to most lipopolysaccharides (LPSs) [36] and that TLR2 mediates responses to microbial lipoproteins [1316]. Although different TLRs share common pathways, there are differences in patterns of activation and signaling pathways between various TLR agonists, which suggests that interplay between TLRs may be important in defining inflammatory responses in sepsis [10, 2124].
To eliminate confounding effects of contaminants, investigators have used chemically prepared or synthesized compounds to study TLR-mediated responses. However, it seems likely that the composition of microbial components that are released into the circulation in sepsis differs substantially from these highly pure agonists.
LPS is a critical bacterial mediator of sepsis [25, 26]. Bacterial peptidoglycan-associated lipoprotein (PAL) is an outer-membrane protein (OMP) that is highly conserved among different genera of enteric gram-negative bacteria [2729]. PAL is involved in maintaining cell-wall integrity and is structurally similar to other microbial lipoproteins [2932]. PAL is composed of 152 amino-acid residues and glyceride and acyl groups that are attached to the N-terminal cysteine [27, 28, 32]. We previously reported that Escherichia coli bacteria release complexes of LPS and several OMPsincluding PAL, murine lipoprotein (MLP), and OmpAinto the blood of septic rats [3335]. We also found that PAL is shed into the circulation in a peritonitis model, that purified PAL induces inflammation in LPS-hyporesponsive (C3H/HeJ) mice, and that E. coli bacteria that express abnormal PAL are less virulent than their wild-type (wt) progenitors in a peritonitis sepsis model [36]. In the present study, we explored the inflammatory effects of a "physiological" form of PAL that is shed into serum by bacteria, confirmed that PAL activity is TLR2-specific in vitro and in vivo, and tested the hypothesis that PAL and LPS are synergistic.
MATERIALS AND METHODS
Bacteria.
E. coli O18:K1:H7 (E. coli O18) was a gift from A. Cross (University of Maryland, Baltimore). E. coli K12 CH202(pRC2) was a gift from U. Henning (Max-Planck-Institut, Tübingen, Germany) [28]. Other E. coli K12 strains were prepared by J.-C.L. and included JC1129 (PAL wt), JC2721 (1129 pal892 [nonsense mutation resulting in PAL-null phenotype] nadA::Tn10), JC8056 (PAL wt), and JC7752 (PAL/TolB deletion derivative of JC8056) transformed with PJC2514, a plasmid that expresses TolB (which renders the strain a functional PAL deletion) [30, 37, 38]
LPS.
E. coli J5 LPS was purchased (List Biological Laboratories) and was determined to be free of protein by use of gold stains and immunoblots for OmpA, PAL, and MLP [39].
Cell lines.
Cell lines were incubated at 37°C in humidified 5% CO2. Human embryonic kidney (HEK) 293 cells were a gift from D. Golenbock (University of Massachusetts, Worcester). RAW 264.7 cells were purchased (American Type Culture Collection). HEK-293 and RAW 264.7 cells were maintained in Dulbecco's modified Eagle medium (DMEM) with 10% fetal calf serum and penicillin/streptomycin (PCN/Strep).
Plasmids.
Human TLR2 and TLR4 expression plasmids were a gift from C. Kirschning and M. Rothe (Tularik, South San Francisco, CA) [40]. A plasmid expressing -galactosidase (-gal) via the constitutively active heat-shock protein promoter (hsp-lacZ) was used as an internal standard. A wt E-selectin promoter reporter (ELAM-luciferase) has been described elsewhere [41]. A plasmid encoding a dominant-negative form of MyD88 (DN MyD88) was a gift from D. Golenbock [42].
Mice.
The Institutional Animal Care and Use Committee at Massachusetts General Hospital (MGH) and Boston Medical Center approved the animal studies. The mice used were C3H/HeJ and C3H/HeOuJ (Jackson Laboratories), C3H/HeN and C57BL/6 (Charles River Laboratories), and TLR2 knockout (KO) mice. C3H/HeJ mice contain a mutation in the TLR4 gene that makes them hyporesponsive to LPS [3]. TLR2 KO mice were generated and bred onto the C57BL/6 background strain by S. Akira (Osaka University, Osaka, Japan) [7].
Antibodies.
Mouse monoclonal antiPAL IgG and rabbit polyclonal antiPAL IgG were prepared as described elsewhere [33, 34, 36]. AntiPAL IgGs do not bind to serum components or to other bacterial products by immunoblotting.
PAL purification.
PAL was prepared from E. coli K12 CH202(pRC2), as described elsewhere [36]; it contained a single protein, as assessed by gold staining. All PAL preparations contained <5 pg LPS/1 g protein, according to the Limulus-amebocyte lysate (LAL) assay (Associates of Cape Cod) [43].
Preparation of mouse macrophages.
Thioglycollate-elicited peritoneal macrophages (PMs) were prepared and plated as described elsewhere [36, 44]. Bone marrowderived macrophages (BMDMs) were prepared as described elsewhere [45]. Briefly, stem cells were harvested from the femurs of 12 mice and placed in 5 × 30-cm Teflon bags (BioFOLIE; Sartorius) that were prepared with the hydrophobic side facing inward. BMDMs were incubated for 7 days and were harvested, plated (at 4 × 105 cells/cm2), and allowed to adhere for 2448 h.
Preparation of IgG-conjugated sepharose beads for capture experiments.
Cyanogen bromideactivated sepharose 4B beads (Amersham Pharmacia Biotech) were coupled with mouse antiPAL IgG and control mouse IgG. The IgG-conjugated beads were blocked overnight (in 5% milk at 4°C) and then were washed with PBS. Beads were then suspended in PBS that contained 1.8 mol/L NaCl and 0.1% Tween-20 (10 L beads/400 L PBS).
Capture of serum-released PAL.
E. coli O18 and wt and PAL-deletion E. coli K12 bacteria were cultured to OD550 0.8 (4 × 108 bacteria/mL) and washed in sterile saline. The bacterial pellets were suspended in an equal volume of normal human serum and incubated (for 4 h at 37°C). The serum was then centrifuged at 6000 g and passed through a 0.45-m filter to remove bacteria. The resultant bacterial serum filtrate (BSF) was added to anti-PAL and control immunoglobulin beads (100 L BSF/400 L washed and resuspended beads). Samples were incubated overnight at 4°C with end-over-end rotation. Unbound material was separated from the beads by centrifugation (at 250 g for 5 min), and the beads were washed with PBS. The final samples (hereafter called "anti-PAL/BSF" and "control immunoglobulin/BSF") were used in macrophage assays and were analyzed for PAL by immunoblots and for LPS by use of the LAL assay. Control samples were prepared from serum that was treated in a fashion identical to that described above but without bacteria.
Immunoblots.
Immunoblots were used to detect PAL in anti-PAL/BSF and control immunoglobulin/BSF samples, as described elsewhere [33, 46]. Densitometry was performed to quantify PAL in lanes of the immunoblots by use of a ChemiImager 5500 imaging system (Alpha Innotech) and a standard curve of purified PAL.
Macrophage assays using anti-PAL/BSF and control immunoglobulin/BSF samples.
Anti-PAL/BSF and control immunoglobulin/BSF samples were suspended in medium and incubated overnight with C3H/HeN or C3H/HeJ BMDMs. Interleukin (IL)6 levels were measured in culture supernatants by ELISA (R&D Systems). Purified LPS was used as a control to verify LPS unresponsiveness of the C3H/HeJ macrophages. IL-6 responses to chemically purified PAL were comparable in BMDMs from C3H/HeJ and C3H/HeN mice.
Human whole-blood and peripheral-blood macrophage assays.
The institutional review board at MGH approved these studies. Citrated whole blood was collected by venipuncture from healthy human volunteers. Whole blood was diluted 4-fold with RPMI 1640 medium and L-glutamine (Cellgro; Mediatech) and then incubated for 20 h with purified PAL. The cells were removed by centrifugation (at 1000 g for 5 min), and IL-6 in the supernatants was quantified by ELISA. Monocytes and macrophages were purified from peripheral blood mononuclear cells (PBMCs) that were prepared from citrated whole blood as described elsewhere [44]. PBMCs were washed with PBS, suspended in PBS with 2 mmol/L EDTA and 10% autologous plasma, and passed through a filter to remove clumps. Macrophages were prepared by magnetic depletion of other cells by use of an antilineage antibody mixture and passage through a magnetic column, according to the manufacturer's instructions (VarioMacs; Miltenyi Biotec). Cells in the flow-through were washed and suspended in medium (DMEM with glutamine, 10% autologous plasma, and PCN/Strep) and were plated (5 × 105 cells/cm2) and allowed to adhere for 12 h. Monocytes and macrophages were incubated for 20 h with PAL, and IL-6 levels were measured in culture supernatants by ELISA.
Stimulation of mouse macrophages with PAL and LPS.
Studies were performed with use of macrophages from C3H/HeJ, C3H/HeN, C3H/HeOuJ, C57BL/6, and TLR2 KO mice and RAW 264.7 cells. Macrophages were treated with PAL and/or LPS for 20 h, and cytokine levels were measured in the culture supernatants by ELISA. Nitrite production was quantified in the absence and presence of interferon (IFN) (10 U/mL) by use of the Greiss reaction [44]. Components were added to the wells in the following order: medium, LPS, and PAL. In some experiments, polymyxin B (PmxB; 5 g/mL) was also added to wells.
Preparation of heat-killed bacteria.
wt and PAL-mutant E. coli K12 strains were cultured to OD550 0.8 and were washed and resuspended in sterile saline. Quantitative cultures were performed on tryptic soy agar plates. Bacteria were killed by incubation at 100°C for 40 min. Heat-killed bacteria were diluted in medium to the concentration required to achieve an MOI of 5 bacteria/HEK-293 cell. Quantitative cultures were performed to verify that the heat-treated bacteria were dead.
Transient transfections.
Transfections were performed when HEK-293 cells reached 70%80% confluence by use of FuGENE 6 Transfection Reagent (Roche Diagnostics). A total of 2 g of plasmid DNA was suspended in 100 L of OptiMEM reduced serum medium (Life Technologies). Each condition included 100 ng of ELAM-luciferase reporter plasmid and 50 ng of hsp-lacZ as an internal standard. Additional human constructs added to different wells included (1) TLR2 (10 ng), (2) TLR2 (10 ng) and DN MyD88 (100 ng), and (3) TLR4 (10 ng). A constant amount of plasmid DNA was maintained by the inclusion of the empty vector pcDNA3.1. Transfected cells were treated for 6 h with PAL, LPS, or heat-killed bacteria, and cells were harvested. Luciferase activity was measured on 20-L cell lysates by use of the Luciferase Assay System (Promega). -gal activity was also measured from each transfection condition. Equal volumes of cell lysate and 2-fold concentrated -gal assay buffer (200 mmol/L sodium phosphate, dibasic dehydrate [pH 7.3], 2 mmol/L MgCl2, 100 mmol/L 2-mercaptoethanol, and 1.33 mg/mL -nitrophenyl--D-galactopyranoside) were incubated together (for 30 min at 37°C), and the optical density at 420 nm was measured. The luciferase activity of each condition was divided by its corresponding -gal activity, to accommodate differences in replicate transfection efficiency. The corrected luciferase activity was then divided by the luciferase activity in HEK-293 cells that were transfected by ELAM-luciferase and hsp-lacZ alone.
Splenocyte proliferation.
Splenocytes were prepared from C57BL/6 and TLR2 KO mice and used as described elsewhere [36, 44]. Briefly, splenocytes were treated with PAL and then were pulsed with 3H-thymidine. The incorporation of 3H-thymidine was assessed by liquid scintillation counting.
Systemic cytokine responses of wt and TLR2 KO mice.
Mice received an intravenous injection of purified PAL or LPS. At 90 min, mice were killed, and blood was collected by cardiac puncture. Cytokine levels were quantified in plasma by ELISA.
Statistical analysis.
Error bars represent SDs. Responses to purified PAL were analyzed by 1-way ANOVA, followed by Dunnett's multiple comparisons post-hoc test to determine threshold activating concentrations of PAL. Bonferroni's multiple comparison test was used to compare responses of PAL-treated C3H/HeOuJ macrophages in the presence and absence of PmxB, wt and TLR2 KO cellular and systemic responses, transfected HEK-293 cell responses to PAL and heat-killed bacteria, and macrophage responses to anti-PAL/BSF and control immunoglobulin/BSF. Unless otherwise indicated, representative data from at least 3 experiments are presented. P < .05 was considered to be statistically significant.
RESULTS
LPS copurification with PAL.
LPS was quantified in affinity-purified samples by use of the LAL assay. For samples prepared with E. coli O18 bacteria, LPS levels per 10 L of beads were 93 pg (experiment 1) and 99 pg (experiment 2) for anti-PAL/BSF samples and were 57 pg (experiment 1) and 50 pg (experiment 2) for control immunoglobulin/BSF samples. The anti-PAL and control immunoglobulin beads were free of LPS, according to the LAL assay. Because anti-PAL IgG does not specifically bind LPS, the 1.62-fold higher levels of LPS in antiPAL/BSF samples suggest that serum-released PAL and LPS are tightly associated.
Activation of macrophages by serum-released PAL.
Cytokines were quantified in supernatants of BMDMs that were treated with anti-PAL/BSF and control immunoglobulin/BSF. Initial experiments were performed by use of E. coli O18 bacteria. Anti-PAL/BSF induced IL-6 production by both C3H/HeJ and C3H/HeN macrophages, whereas IL-6 was not significantly induced in macrophages treated with control immunoglobulin/BSF (P < .001) (figure 1B).
We performed additional experiments to verify that PAL is responsible for the inflammatory effects of the anti-PAL/BSF samples. Macrophages were treated with anti-PAL/BSF and control immunoglobulin/BSF samples that were prepared by use of wt and PAL-deletion bacteria. Cytokine production was reduced (C3H/HeN) or eliminated (C3H/HeJ) in anti-PAL/BSF samples that were prepared with the PAL deletion, compared with the wt bacteria (P < .001) (figure 1C, shown for IL-6).
Induction of cytokine production in human cells by PAL.
PAL was incubated with whole blood and adherent peripheral blood monocytes and macrophages from 4 and 5 donors, respectively. PAL caused an increase in IL-6 production by whole blood (P < .0001) (figure 2A) and by peripheral bloodadherent monocytes and macrophages (P < .0001) (figure 2B). There was considerable variability between responses of cells from different donors. Whole-blood responses were statistically significant for all donors at PAL concentrations 400 ng/mL. Monocyte and macrophage responses were significant for all donors at PAL concentrations 10 ng/mL.
Activation of mouse macrophages by purified PAL.
To extend our prior studies that indicated that PAL activates PMs from LPS-hyporesponsive mice, we studied responses of macrophages from LPS-responsive mice and from different macrophage compartments. PAL increased cytokine production by BMDMs and PMs from all strains of mice tested at threshold PAL concentrations of 1100 ng/mL, depending on the cell lineage and the cytokine (P < .0001) (figure 3AC). PmxB did not inhibit the effects of PAL on IL-6 production by C3H/HeOuJ macrophages, which supports the LAL data that our purified PAL did not contain biologically relevant quantities of LPS (P > .05; data not shown). PAL also activated production of nitrite by PMs and BMDMs at threshold PAL concentrations of 110 ng/mL (P < .001) (figure 3D). IFN- was not required for the PAL-induced production of nitrite by C57BL/6 PM (P < .0001), but it was required for the PAL-induced production of nitrite by C57BL/6 BMDMs and C3H/HeN PMs (figure 3D). The source of the difference in this requirement of IFN- between the cell types is not clear. Other investigators have also reported heterogeneity in responses of leukocytes from different compartments [47]. It is possible that signaling pathways vary between cells in different compartments or that the PMs were primed by the thioglycollate that was used to elicit the PMs. PAL caused an increase in tumor necrosis factor (TNF), IL-6, and nitrite production by the macrophage-like cell line RAW 264.7 at threshold concentrations of 110 ng/mL (P < .01; data not shown). We have previously reported that PAL is released into the blood of mice in a cecal ligation and puncture (CLP) sepsis model and estimated that PAL levels were 130 ng/mL in some CLP mice [36], which is considerably higher than the threshold activating concentrations of PAL in the present study.
TLR2- and MyD88-mediated inflammatory effects of PAL.
HEK-293 cells were transiently transfected with reporter plasmids (ELAM-luciferase and hsp-lacZ) plus expression plasmids encoding human TLR2, human TLR2 plus DN MyD88, or human TLR4, and then were treated with PAL or LPS. MyD88 is an intracellular adapter molecule that is required for signaling by some TLRs. The DN MyD88 construct was used to confirm that PAL activation is fully dependent on TLRs. PAL activated ELAM-luciferase in TLR2-transfected cells but not in cells transfected with only the reporter construct or with the TLR4 construct (P < .001) (figure 4). The DN MyD88 construct fully inhibited activation of TLR2-transfected cells by PAL (P < .001).
Mediation of PAL-induced inflammation by TLR2 in mice.
We compared responses of BMDMs and splenocytes from C57BL/6 and TLR2 KO mice. PAL induced IL-6, TNF-, macrophage inflammatory protein (MIP)1, and nitrite production by macrophages from wt mice but not by macrophages from TLR2 KO mice (P < .001) (figure 5AB, shown for IL-6 only). PAL also increased the proliferation of splenocytes from wt but not from TLR2 KO mice (P < .001) (figure 5C). Plasma TNF-, IL-6, and MIP-1 levels were measured in PAL-treated mice. PAL activated systemic cytokine responses in wt but not in TLR2 KO mice (P < .001) (figure 5D, shown for IL-6 and nitrite only).
Effects of wt versus PAL-mutant bacteria on TLR2-transfected cells.
To determine whether PAL contributes to the TLR2-mediated activation of whole bacteria, TLR2-transfected HEK-293 cells were incubated with heat-killed strains of wt E. coli K12 bacteria and derivatives that contained PAL-nonsense or PAL-deletion mutations. There was no reduction in luciferase activation in TLR2-transfected cells in the PAL-mutant strains, compared with the wt strain (figure 6).
Synergistic activation of macrophages by PAL and LPS.
Studies were performed by use of PMs (C3H/HeN and C57BL/6), BMDMs (C57BL/6), and RAW 264.7 cells. There was marked synergy between PAL and low levels of LPS in inducing cytokine and nitrite production for all cell lineages tested (figure 7, shown for C57BL/6 PMs and BMDMs only). There was variability in the doses of PAL and LPS required for synergy between different cell types and readouts. Synergy was most evident at LPS and PAL concentrations that were near the threshold of activity for the individual component.
DISCUSSION
The present study shows that PAL is a naturally occurring ubiquitous TLR2 agonist that is shed into serum in a biologically active form. We also found that released PAL is tightly associated with LPS and that PAL and LPS are markedly synergistic, which supports the notion that multiple microbial components circulate and modulate the effects of one another in sepsis.
Our studies provide insight into the composition and biological activity of naturally occurring bacterial products that circulate in sepsis. Previous studies of TLR-mediated processes have used synthesized or chemically purified microbial products. We believe that the shed form of PAL that we used is closer in structure and additional content to forms that circulate and contribute to the pathogenesis of human sepsis. The higher IL-6 responses of C3H/HeN versus C3H/HeJ macrophages to anti-PAL/BSF suggest that the PAL and LPS in these naturally occurring products are synergistic.
Specific TLR agonists have been shown to modulate inflammatory effects of other TLR agonists. However, the ability of different microbial TLR agonists to influence the inflammatory effects of one another varies markedly between different studies and for different readouts. For instance, consistent with our data, results of other studies have shown synergy between other TLR2 agonists and LPS if they are administered concurrently [4850]. There are also reports of cross-tolerance between LPS and TLR2 agonists and between LPS and the TLR9 agonist bacterial DNA [48, 5153]. Finally, another study showed that the pretreatment of macrophages with LPS inhibited TNF release and augmented IL-1 release in response to zymosan, a TLR2 agonist, whereas zymosan pretreatment augmented release of both TNF and IL-1 in response to LPS [54]. Thus, it appears that the mechanism(s) by which different microbial products influence responses to one another is highly complex. Further studies will be required to define the pathways that are involved in PAL-LPS synergy.
Over the past several decades, thousands of studies have used various LPS preparations to gain insight into the processes that occur in sepsis. Investigators have noted that LPS preparations often contain proinflammatory protein contaminants [5557]. PAL frequently contaminates purified rough and smooth LPSs [39], a finding that, together with the present results, raises the possibility that PAL may have contributed to responses in some of these studies aimed at LPS.
In an earlier study [36], we found that E. coli K12 bacteria expressing abnormal PAL because of a nonsense mutation were less virulent than were their wt progenitors in a peritonitis model. The lack of reduced activity of the heat-killed PAL-mutant versus wt bacteria on TLR2-transfected HEK-293 cells in the present study is likely explained by the presence of other gram-negative bacterial TLR2 agonists, such as OmpA and MLP [14, 58], and suggests that the reduced virulence observed in our earlier study was not due to decreased PAL in whole bacteria but may have been due to an altered release of PAL. Our data showing that biologically active PAL is shed into serum in tight association with LPS and that PAL and LPS are synergistic suggest that circulating released PAL directly induces inflammation alone and/or synergistically with other bacterial products.
We hypothesize that, during human sepsis, microorganisms shed fragments that contain multiple PAMPs that activate different pathways and modulate inflammatory responses. Our data suggest that PAL and LPS are part of these released fragments. These results imply that, during sepsis, even very low levels of microbial toxins that are hardly capable of eliciting immune cell responses on their own may contribute significantly to inflammation, through synergy with other microbial components.
References
1. Akira S, Takeda K, Kaisho T. Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immunol 2001; 2:67580. First citation in article
2. Rock FL, Hardiman G, Timans JC, Kastelein RA, Bazan JF. A family of human receptors structurally related to Drosophila Toll. Proc Natl Acad Sci USA 1998; 95:58893. First citation in article
3. Poltorak A, He X, Smirnova I, et al. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in TLR4 gene. Science 1998; 282:20858. First citation in article
4. Qureshi ST, Lariviere L, Leveque G, et al. Endotoxin-tolerant mice have mutations in Toll-like receptor 4 (TLR4). J Exp Med 1999; 189:61525. First citation in article
5. Rhee SH, Hwang D. Murine Toll-like receptor 4 confers lipopolysaccharide responsiveness as determined by activation of NF B and expression of the inducible cyclooxygenase. J Biol Chem 2000; 275:3403540. First citation in article
6. Tapping RI, Akashi S, Miyake K, Godowski PJ, Tobias PS. Toll-like receptor 4, but not Toll-like receptor 2, is a signaling receptor for Escherichia and Salmonella lipopolysaccharides. J Immunol 2000; 165:57807. First citation in article
7. Takeuchi O, Hoshino K, Kawai T, et al. Differential roles of TLR2 and TLR4 in recognition of gram-negative and gram-positive bacterial cell wall components. Immunity 1999; 11:44351. First citation in article
8. Schwandner R, Dziarski R, Wesche H, Rothe M, Kirschning CJ. Peptidoglycan- and lipoteichoic acidinduced cell activation is mediated by Toll-like receptor 2. J Biol Chem 1999; 274:174069. First citation in article
9. Means TK, Lien E, Yoshimura A, Wang S, Golenbock DT, Fenton MJ. The CD14 ligands lipoarabinomannan and lipopolysaccharide differ in their requirement for Toll-like receptors. J Immunol 1999; 163:674855. First citation in article
10. Hirschfeld M, Weis JJ, Toshchakov V, et al. Signaling by Toll-like receptor 2 and 4 agonists results in differential gene expression in murine macrophages. Infect Immun 2001; 69:147782. First citation in article
11. Werts C, Tapping RI, Mathison JC, et al. Leptospiral lipopolysaccharide activates cells through a TLR2-dependent mechanism. Nat Immunol 2001; 2:34652. First citation in article
12. Hayashi F, Smith KD, Ozinsky A, et al. The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 2001; 410:1099103. First citation in article
13. Brightbill HD, Libraty DH, Krutzik SR, et al. Host defense mechanisms triggered by microbial lipoproteins through Toll-like receptors. Science 1999; 285:7325. First citation in article
14. Aliprantis AO, Yang RB, Mark MR, et al. Cell activation and apoptosis by bacterial lipoproteins through Toll-like receptor2. Science 1999; 285:7369. First citation in article
15. Hirschfeld M, Kirschning CJ, Schwandner R, et al. Cutting edge: inflammatory signaling by Borrelia burgdorferi lipoproteins is mediated by Toll-like receptor 2. J Immunol 1999; 163:23826. First citation in article
16. Lee HK, Lee J, Tobias PS. Two lipoproteins extracted from Escherichia coli K-12 LCD25 lipopolysaccharide are the major components responsible for Toll-like receptor 2mediated signaling. J Immunol 2002; 168:40127. First citation in article
17. Hemmi H, Takeuchi O, Kawai T, et al. A Toll-like receptor recognizes bacterial DNA. Nature 2000; 408:7405. First citation in article
18. Alexopoulou L, Holt AC, Medzhitov R, Flavell RA. Recognition of double-stranded RNA and activation of NF-B by Toll-like receptor 3. Nature 2001; 413:7328. First citation in article
19. Ozinsky A, Underhill DM, Fontenot JD, et al. 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 2000; 97:1376671. First citation in article
20. Zhang D, Zhang G, Hayden MS, et al. A Toll-like receptor that prevents infection by uropathogenic bacteria. Science 2004; 303:15226. First citation in article
21. Jones BW, Means TK, Heldwein KA, et al. Different Toll-like receptor agonists induce distinct macrophage responses. J Leukoc Biol 2001; 69:103644. First citation in article
22. Vogel SN, Fenton M. Toll-like receptor 4 signalling: new perspectives on a complex signal-transduction problem. Biochem Soc Trans 2003; 31:6648. First citation in article
23. Yamamoto M, Sato S, Hemmi H, et al. Role of adaptor TRIF in the MyD88-independent Toll-like receptor signaling pathway. Science 2003; 301:6403. First citation in article
24. Wietek C, Miggin SM, Jefferies CA, O'Neill LA. Interferon regulatory factor-3mediated activation of the interferon-sensitive response element by Toll-like receptor (TLR) 4 but not TLR3 requires the p65 subunit of NF-. J Biol Chem 2003; 278:5092331. First citation in article
25. Beutler B, Milsark IW, Cerami AC. Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effect of endotoxin. Science 1985; 229:86971. First citation in article
26. Galanos C, Freudenberg MA. Mechanisms of endotoxin shock and endotoxin hypersensitivity. Immunobiology 1993; 187:34656. First citation in article
27. Mizuno T. Structure of peptidolglycan-associated lipoprotein (PAL) of the Proteus mirabilis outer membrane. I. Isolation and characterization of fatty acidcontaining peptides from PAL. J Biochem 1981; 89:10518. First citation in article
28. Chen R, Henning U. Nucleotide sequence of the gene for the peptidoglycan-associated lipoprotein of Escherichia coli K12. Eur J Biochem 1987; 163:737. First citation in article
29. Rodriguez-Herva J, Ramos-Gonzalez M, Ramos J. The Pseudomonas putida peptidoglycan-associated outer membrane lipoprotein is involved in maintenance of the integrity of the cell envelope. J Bacteriol 1996; 178:1699706. First citation in article
30. Lazzaroni J-C, Portalier R. The excC gene of Escherichia coli K-12 required for cell envelope integrity encodes the peptidoglycan-associated lipoprotein (PAL). Mol Microbiol 1992; 6:73542. First citation in article
31. Lazzaroni J, Lefebvre N, Portalier R. Cloning of the excC and excD genes involved in the release of periplasmic proteins by Escherichia coli K12. Mol Gen Genet 1989; 218:4604. First citation in article
32. Mizushima S. Post-translational modification and processing of outer membrane prolipoproteins in Escherichia coli. Mol Cell Biochem 1984; 60:515. First citation in article
33. Hellman J, Loiselle PM, Zanzot EM, et al. Release of gram-negative outer membrane proteins into human serum and septic rat blood and their interactions with immunoglobulin in antiserum to Escherichia coli J5. J Infect Dis 2000; 181:103443. First citation in article
34. Hellman J, Loiselle PM, Tehan MM, et al. Outer membrane protein A, peptidoglycan-associated lipoprotein, and murein lipoprotein are released by Escherichia coli bacteria into serum. Infect Immun 2000; 68:256672. First citation in article
35. Hellman J, Warren HS. Outer membrane protein A (OmpA), peptidoglycan-associated lipoprotein (PAL), and murein lipoprotein (MLP) are released in experimental gram-negative sepsis. J Endotoxin Res 2001; 7:6972. First citation in article
36. Hellman J, Roberts JDJ, Tehan MM, Allaire JE, Warren HS. Bacterial peptidoglycan-associated lipoprotein is released into the bloodstream in gram-negative sepsis and causes inflammation and death in mice. J Biol Chem 2002; 277:1427480. First citation in article
37. Clavel T, Germon P, Vianney A, Portalier R, Lazzaroni JC. TolB protein of Escherichia coli K-12 interacts with the outer membrane peptidoglycan-associated proteins Pal, Lpp, and OmpA. Mol Microbiol 1998; 29:35967. First citation in article
38. Dubuisson JF, Vianney A, Lazzaroni JC. Mutational analysis of the TolA C-terminal domain of Escherichia coli and genetic evidence for an interaction between TolA and TolB. J Bacteriol 2002; 184:46205. First citation in article
39. Hellman J, Tehan MM, Warren HS. Murein lipoprotein (MLP), Peptidoglycan-associated lipoprotein (PAL), and outer membrane protein A (OmpA) are present in purified rough and smooth LPS. J Infect Dis 2003; 188:2869. First citation in article
40. Kirschning CJ, Wesche H, Merrill AT, Rothe M. Human toll-like receptor 2 confers responsiveness to bacterial lipopolysaccharide. J Exp Med 1998; 188:20917. First citation in article
41. Means TK, Wang S, Lien E, Yoshimura A, Golenbock DT, Fenton MJ. Human toll-like receptors mediate cellular activation by Mycobacterium tuberculosis. J Immunol 1999; 163:39207. First citation in article
42. Latz E, Visintin A, Lien E, et al. Lipopolysaccharide rapidly traffics to and from the Golgi apparatus with the Toll-like receptor 4MD-2CD14 complex in a process that is distinct from the initiation of signal transduction. J Biol Chem 2002; 277:4783443. First citation in article
43. Novitsky TJ, Roslansky PF, Siber GR, Warren HS. Turbidometric method for quantifying serum inhibition of Limulus amoebocyte lysate. J Clin Microbiol 1985; 21:2116. First citation in article
44. Bierer B, Coligan JE, Margulies DH, Shevach EM, Strober W, eds. Current protocols in immunology. John Wiley and Sons, 2004. First citation in article
45. Schilling D, Thomas K, Nixdorff K, Vogel SN, Fenton MJ. Toll-like receptor 4 and TollIL-1 receptor domaincontaining adapter protein (TIRAP)/myeloid differentiation protein 88 adapterlike (Mal) contribute to maximal IL-6 expression in macrophages. J Immunol 2002; 169:587480. First citation in article
46. Hellman J, Zanzot EM, Loiselle PM, et al. Antiserum against Escherichia coli J5 contains antibodies reactive with outer membrane proteins of heterologous gram-negative bacteria. J Infect Dis 1997; 176:12608. First citation in article
47. Fitting C, Dhawan S, Cavaillon JM. Compartmentalization of tolerance to endotoxin. J Infect Dis 2004; 189:1295303. First citation in article
48. Sato S, Nomura F, Kawai T, et al. Synergy and cross-tolerance between Toll-like receptor (TLR) 2 and TLR4-mediated signaling pathways. J Immunol 2000; 165:7096101. First citation in article
49. Zhang H, Peterson JW, Niesel DW, Klimpel GR. Bacterial lipoprotein and lipopolysaccharide act synergistically to induce lethal shock and proinflammatory cytokine production. J Immunol 1997; 159:486878. First citation in article
50. Wang JE, Jorgensen PF, Ellingsen EA, et al. Peptidoglycan primes for LPS-induced release of proinflammatory cytokines in whole human blood. Shock 2001; 16:17882. First citation in article
51. Lehner MD, Morath S, Michelsen KS, Schumann RR, Hartung T. Induction of cross-tolerance by lipopolysaccharide and highly purified lipoteichoic acid via different Toll-like receptors independent of paracrine mediators. J Immunol 2001; 166:51617. First citation in article
52. Schwartz DA, Wohlford-Lenane CL, Quinn TJ, Krieg AM. Bacterial DNA or oligonucleotides containing unmethylated CpG motifs can minimize lipopolysaccharide-induced inflammation in the lower respiratory tract through an IL-12dependent pathway. J Immunol 1999; 163:22431. First citation in article
53. Wang JH, Doyle M, Manning BJ, Di Wu Q, Blankson S, Redmond HP. Induction of bacterial lipoprotein tolerance is associated with suppression of Toll-like receptor 2 expression. J Biol Chem 2002; 277:3606875. First citation in article
54. West MA, Bennet T, Clair L. Reprogrammed macrophage tumor necrosis factor and interleukin-1 release with inflammatory pretreatment: differential regulation by endotoxin and zymosan. J Trauma 1995; 39:40410. First citation in article
55. Manthey CL, Perera P-Y, Henricson BE, Hamilton TA, Qureshi N, Vogel SN. Endotoxin-induced early gene expression in C3H/HeJ (Lpsd) macrophages. J Immunol 1994; 153:265363. First citation in article
56. Doe WF, Yang ST, Morrison DC, Betz SJ, Henson PM. Macrophage stimulation by bacterial lipopolysaccharides. II. Evidence for differentiation signals delivered by lipid A and by a protein rich fraction of lipopolysaccharides. J Exp Med 1978; 148:55768. First citation in article
57. Hirschfeld M, Ma Y, Weis JH, Vogel SN, Weis JJ. Cutting edge: repurification of lipopolysaccharide eliminates signaling through both human and murine Toll-like receptor 2. J Immunol 2000; 165:61822. First citation in article
58. Jeannin P, Magistrelli G, Goetsch L, et al. Outer membrane protein A (OmpA): a new pathogen-associated molecular pattern that interacts with antigen presenting cells-impact on vaccine strategies. Vaccine 2002; 20(Suppl 4):A237. First citation in article