Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Popa, C. Articles by Netea, M. G. Search for Related Content PubMed PubMed Citation Articles by Popa, C. Articles by Netea, M. G.

 Previous Article  |  Next Article 

Infection and Immunity, October 2007, p. 4831-4837, Vol. 75, No. 10
0019-9567/07/$08.00+0     doi:10.1128/IAI.00237-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Bartonella quintana Lipopolysaccharide Is a Natural Antagonist of Toll-Like Receptor 4

Calin Popa,^1,2, Shahla Abdollahi-Roodsaz,^2, Leo A. B. Joosten,^1,2 Nozomi Takahashi,² Tom Sprong,^1,4 Giovanni Matera,³ Maria Carla Liberto,³ Alfredo Foca,³ Marcel van Deuren,^1,4 Bart Jan Kullberg,^1,4 Wim B. van den Berg,² Jos W. M. van der Meer,^1,4 and Mihai G. Netea^1,4*

Department of General Internal Medicine,¹ Rheumatology Research & Advanced Therapeutics, Department of Rheumatology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands,² Institute of Microbiology, Department of Medical Sciences, University of Catanzaro, Catanzaro, Italy,³ Nijmegen University Centre for Infectious Diseases, Nijmegen, The Netherlands⁴

Received 13 February 2007/ Returned for modification 24 March 2007/ Accepted 20 June 2007

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bartonella quintana is a gram-negative microorganism that causes trench fever and chronic bacteremia. B. quintana lipopolysaccharide (LPS) was unable to induce the production of proinflammatory cytokines in human monocytes. Interestingly, B. quintana LPS is a potent antagonist of Toll-like receptor 4 (TLR4), as it inhibited both mRNA transcription and the release of tumor necrosis factor alpha, interleukin 1ß (IL-1ß), and IL-6 by Escherichia coli LPS in human monocytes, at ratios ranging from 1,000:1 to 10:1 (B. quintana LPS to E. coli LPS). Likewise, B. quintana LPS blocked the interaction of E. coli LPS with TLR4 in transfected cell lines. The extent of the inhibitory effect of B. quintana LPS was demonstrated in microarray studies, which showed downregulation of practically all genes induced by LPS in monocytes. Because of the role of TLR4 in inflammation, B. quintana LPS may prove useful as a potent anti-TLR4 agent with therapeutic potential in both infections and autoimmune inflammation.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bartonella quintana is a gram-negative pathogen initially described during World War I as the agent of trench fever, a disease associated with recurrent fever and headaches. In the past few decades, B. quintana infection has been identified in homeless people (4). While most individuals with B. quintana infection recover, some 5 to 10% eventually develop chronic bacteremia (4) and subsequent complications, such as chronic endocarditis in the absence of preexisting heart valve lesions (20). New manifestations of B. quintana infections, such as bacillary angiomatosis, bacillary peliosis hepatitis, and chronic lymphadenopathy, have also been described (2). These manifestations have been attributed to proliferative and antiapoptotic effects of Bartonella spp. (6).

A characteristic of B. quintana bacteremia is the absence of symptoms of high fever and signs of septic shock, disseminated intravascular coagulation, or organ failure. Lipopolysaccharide (LPS, or endotoxin) is a main component of the outer membranes of gram-negative microorganisms, and the LPSs from gram-negative enteric bacteria (such as Escherichia coli and Salmonella enterica) are able to induce proinflammatory cytokines, chemokines, and adhesion molecules (23) and thereby to evoke the clinical signs of sepsis. The lipid A moiety of LPS interacts with a membrane receptor complex containing Toll-like receptor 4 (TLR4), MD-2, and CD14 (17, 19, 23). Because of strong proinflammatory effects of most species of LPS, the apparent absence of sepsis syndrome in patients with B. quintana bacteremia is a puzzling aspect of the infection. As an explanation, overproduction of the anti-inflammatory cytokine interleukin-10 (IL-10) and an attenuated inflammatory cytokine profile during B. quintana bacteremia have been proposed (5), but the molecular mechanisms have remained elusive.

Recently, the LPS of the related organism Bartonella henselae was purified and characterized as a penta-acylated deep-rough LPS with low endotoxic activity (17, 27). In the present study, we investigated the biologic activities of B. quintana LPS in terms of induction of proinflammatory cytokines and interaction with TLRs and other species of LPS.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Reagents and microorganisms. LPS (E. coli serotype O55:B5) was purchased from Sigma Chemical Co., and synthetic Pam3Cys was purchased from EMC Microcollections (Tübingen, Germany). The B. quintana Oklahoma strain was kindly provided by D. Raoult (Marseille, France) and grown on sheep blood agar at 37°C in a 5% CO₂ atmosphere. For stimulation experiments, 5-day cultures of B. quintana were heat killed for 60 min at 56°C. B. quintana LPS was extracted either by a single-step phenol-water extraction, as previously described (13), or by a two-step extraction method (9) which eliminates contamination with proteins. E. coli LPS from Sigma was also double purified (9). Both purified and nonpurified E. coli and B. quintana LPSs (100 µg of each) were run in a 10% polyacrylamide gel and subjected to silver staining to visualize contaminating proteins. Briefly, the gel was fixed in 40% methanol-10% acetic acid, sensitized using 0.2% sodium thiosulfate, and then stained with 0.2% silver nitrate for 20 min. Color was developed using a 3% sodium carbonate solution.

Signaling through human TLR2 and TLR4 in a transfected cell line. Chinese hamster ovary (CHO) fibroblasts stably transfected with human CD14 (3E10-CD14), a combination of CD14 and TLR4 (3E10-TLR4), and TLR2 (3E10-TLR2) were a kind gift from Robin Ingalls. These cell lines express inducible membrane CD25 under the control of a region of the human E-selectin (ELAM-1) promoter containing NF-B binding sites. Cells were maintained at 37°C in 5% CO₂ in Ham's F-12 medium (Gibco, Invitrogen, Breda, The Netherlands) supplemented with 10% fetal calf serum, 0.01% L-glutamine, 50 µg/ml gentamicin, 400 U/ml hygromycin, and 0.5 mg/ml of G418 (for 3E10-TLR2) or 0.05 mg/ml of puromycin (for 3E10-TLR4) as an additional selection antibiotic. TLR4 expression was confirmed by flow cytometry (Coulter Epics XL-MCL; Beckman Coulter, Mijdrecht, The Netherlands), using a phycoerythrin-labeled anti-TLR4 antibody (clone HTA125; Immunosource, Halle-Zoersel, Belgium).

For stimulation experiments, 500 µl of cells in culture medium at a density of 1 x 10⁵/ml was plated in 24-well culture plates. After an overnight incubation, cells were incubated with control medium, Pam3Cys (10 µg/ml), E. coli LPS (1 µg/ml), B. quintana LPS (10 µg/ml), or a combination of E. coli LPS and B. quintana LPS for 20 h at 37°C. Thereafter, cells were harvested using trypsin-EDTA (Cambrex, East Rutherford, NY) and prepared for flow cytometry (Coulter FACScan). CD25 expression of the CHO cells was measured using fluorescein isothiocyanate-labeled anti-CD25 (DAKO, Glostrup, Denmark) and was expressed as the x-fold increase above the mean.

Isolation of PBMC and stimulation of cytokine production. After informed consent was obtained, venous blood was drawn from the cubital veins of six healthy volunteers into three 10-ml lithium-heparin tubes (Monoject, 's Hertogenbosch, The Netherlands). The regional medical ethical committee approved the study protocol. The peripheral blood mononuclear cell (PBMC) fraction was obtained by density centrifugation of blood diluted 1:1 in pyrogen-free saline over Ficoll-Paque (Pharmacia Biotech AB, Uppsala, Sweden). Cells were washed twice in saline and suspended in culture medium (RPMI 1640 DM) supplemented with 10 µg/ml gentamicin, 10 mM L-glutamine, and 10 mM pyruvate. The cells were counted in a Coulter counter (Beckman Coulter, Mijdrecht, The Netherlands), and the number was adjusted to 5 x 10⁶ cells/ml.

PBMC (5 x 10⁵) in a 100-µl volume were added to round-bottomed 96-well plates (Greiner, Alphen a/d Rijn, The Netherlands) and incubated with either 100 µl of culture medium (negative control) or one of the following stimuli: B. quintana bacteria (1 x 10⁶ microorganisms/ml), B. quintana LPS (at concentrations of up to 10 µg/ml), and E. coli LPS (10 ng/ml). To assess the role of TLR4 in the induction of cytokines by B. quintana LPS, PBMC were preincubated for 30 min with 10 µg/ml control immunoglobulin G1 (IgG1) or a blocking anti-TLR4 antibody (eBioscience, AMDS Malden, The Netherlands). Similarly, the TLR4 antagonistic properties of B. quintana LPS were assessed by preincubating the cells with various concentrations of B. quintana LPS 30 min before stimulation with E. coli LPS.

Cytokine measurements. Human tumor necrosis factor alpha (TNF-), IL-1ß, IL-8, and IL-6 concentrations were measured by use of commercial enzyme-linked immunosorbent assay kits (Pelikine Compact, CLB, Amsterdam, The Netherlands) according to the manufacturer's instructions.

RNA extraction. Total RNA was extracted from 10 x 10⁶ cells by using 1 ml TRIzol reagent (Sigma, St. Louis, MO). Subsequently, 200 µl chloroform and 500 µl 2-propanol (Merck, Darmstadt, Germany) were used to separate the RNAs from DNAs and proteins. Finally, after a wash step with 75% ethanol (Merck, Darmstadt, Germany), the dry RNAs were dissolved in 30 µl of water.

PCR amplification. To obtain cDNA, we reverse transcribed 1 µg DNase-treated total RNA with oligo(dT) primers (0.01 µg/ml) in a reverse transcription-PCR mixture with a total volume of 20 µl. Subsequently, quantitative PCR was performed using an ABI PRISM 7000 sequence detection system (Applied Biosystems, Foster City, CA). PCRs to amplify glyceraldehyde-3-phosphate dehydrogenase, IL-1ß, TNF-, and IL-6 were performed with SYBR green PCR master mix (Applied Biosystems, Foster City, CA), 5 µl diluted (1/20) cDNA, and primers at a final concentration of 300 nmol/liter in a total volume of 25 µl. The following primers were developed using Primer Express (Applied Biosystems, Foster City, CA): CCTCTGATGGCACCACCAG (reverse primer) and TCTTCTCGAACCCCGAGTGA (forward primer) for TNF-, GGCAAGTCTCCTCATTGAATCC (reverse primer) and AGCCCTGAGAAAGGAGACATGTA (forward primer) for IL-6, and TGGGTAATTTTTGGGATCTACACTCT (reverse primer) and AATCTGTACCTGTCCTGCGTGTT (forward primer) for IL-1ß. Quantification of the PCR signals for each sample was performed by comparing the cycle threshold values, in duplicate, for the gene of interest with the cycle threshold values for the glyceraldehyde-3-phosphate dehydrogenase housekeeping gene. We validated all primers according to the protocol, and the standard curves were all within the dynamic range.

Microarray analysis. PBMC from two different healthy donors were preincubated or not with 500 ng/ml B. quintana LPS for 1 h and, thereafter, were left untreated for 8 h or treated with 1 ng/ml E. coli LPS. RNAs were extracted by the TRIzol method (Sigma, St. Louis, MO). Protocols from Affymetrix (Santa Clara, CA) (1a) were followed as described previously, with some modification, as briefly described below (21). Synthesis and labeling of double-stranded cDNA from 1 µg total RNA were performed through one-cycle amplification in vitro transcription/biotin labeling using a MessageAmp II aRNA amplification kit (Ambion, Austin, TX). After fragmentation, the hybridization mixture was spiked with Bio B, C, and D and cre and hybridized to human genome array U133 Plus2.0 (Affymetrix, Santa Clara, CA), containing over 47,000 probes. After 16 h, hybridization arrays were washed and incubated with streptavidin-coupled phycoerythrin (Molecular Probes, Eugene, OR). The GeneChips were then scanned using a GeneChipScanner laser scanner (Affymetrix, Santa Clara, CA) according to the manufacturer's instructions and were analyzed by using GCOS (Affymetrix, Santa Clara, CA). Normalization, model fitting, and filtering were performed by using dChip (Harvard) (12). Subsequently, pairwise and multiple comparisons were performed by dChip, applying different criteria (>1.2-fold change with P value of <0.05 and/or >2-fold change with P value of <0.05). High-level analyses such as cluster analysis, principal component analysis, and analysis of variance were performed by dChip on filtered genes and a gene list generated by multiple comparisons. Hierarchical cluster analysis was performed using algorithms that assemble all elements in a single tree based on computed similarity scores.

Statistical analysis. The differences between groups were analyzed by the unpaired Student t test and, where appropriate, the paired t test. The level of significance between groups was set to P values of <0.05. The data are given as means ± standard deviations (SD).

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Role of TLR2 and TLR4 in recognition of B. quintana. The first sets of experiments were designed to explore the induction of proinflammatory cytokines by whole B. quintana microorganisms and the role of TLR2 and TLR4 in this induction. Interestingly, heat-killed B. quintana did not stimulate TLR4-transfected CHO cells, while TLR2-transfected CHO cells reacted to B. quintana, indicating that the microorganism is able to signal through this receptor (Fig. 1A to D). To further assess the biological significance of these findings, we investigated the production of proinflammatory cytokines by human PBMC upon stimulation with heat-killed B. quintana. As shown in Fig. 1E, a blocking TLR4 antibody did not influence the capacity of B. quintana to stimulate the production of TNF-, IL-1ß, and IL-6 from human PBMC. This finding implies that the stimulation of cytokines by B. quintana is not induced by its LPS component or that B. quintana LPS stimulates cytokines through a TLR4-independent mechanism.

View larger version (25K): [in this window] [in a new window]

FIG. 1. TLR4 is not involved in the host response to B. quintana. CHO cells transfected with CD14, CD14 and TLR4, or CD14 and TLR2 were stimulated with 1 x 106 heat-killed B. quintana cells. Expression of CD25 on the cell membrane was measured by fluorescence-activated cell sorting analysis. B. quintana stimulated CD25 expression only in CHO-CD14/TLR2 cells (B and D), not in CHO-CD14 (A and D) or CHO-CD14/TLR4 (C and D) cells. (E) Human PBMC were stimulated for 24 h with 1 x 106 heat-killed B. quintana microorganisms/ml in the presence of either control IgG1 (open bars) or an anti-TLR4 antibody (aTLR4) (10 µg/ml; hatched bars). Unstimulated cells are presented as solid bars.

B. quintana LPS is a TLR4 antagonist. To investigate the two above-mentioned possibilities, we isolated B. quintana LPS in a two-step purification process which warrants the absence of protein contaminants (14). The purity of both E. coli and B. quintana LPSs after the two-step purification was determined using silver staining for proteins. Silver staining revealed the absence of any protein contaminants in both LPS samples after purification (Fig. 2A). As shown in Fig. 2B, purified B. quintana LPS did not stimulate the production of TNF-, IL-1ß, and IL-6 in human PBMC. IL-8 was also not induced by B. quintana LPS (not shown). Similar negative results were obtained when peritoneal murine macrophages were stimulated (not shown). These results demonstrate that B. quintana LPS is not able to induce cytokine production and suggest that it is devoid of direct biological activity.

View larger version (12K): [in this window] [in a new window]

FIG. 2. (A) Silver staining of E. coli and B. quintana LPSs before and after double purification (100 µg of each LPS). Two micrograms of bovine serum albumin (BSA) was also loaded into the gel as a control for staining. (B) Human PBMC were stimulated for 24 h with E. coli LPS (10 ng/ml; open bars) or B. quintana LPS (1 µg/ml; hatched bars). Unstimulated cells are presented as solid bars. Data are presented as means plus SD (n = 6).

It was shown previously that certain LPS structures with no direct agonistic activity can function as TLR4 antagonists. To assess whether B. quintana LPS also functions as a TLR4 antagonist, we preincubated human PBMC with various concentrations of B. quintana LPS before stimulating the cells with E. coli LPS. As shown in Fig. 3A, B. quintana LPS completely abolished E. coli LPS-induced TNF- production when ratios of at least 10:1 were used and inhibited the induction of cytokines approximately 50% when a 1:1 ratio of B. quintana LPS to E. coli LPS was employed. In addition, when a 10:1 ratio was used, B. quintana LPS was able to completely inhibit the expression of proinflammatory cytokine TNF-, IL-1ß, and IL-6 mRNAs from E. coli LPS-challenged human PBMC (Fig. 3B). The inhibitory effect of B. quintana LPS was exerted by blocking TLR4, as further demonstrated by the blockade of E. coli LPS-induced stimulation of CHO cells transfected with TLR4 (Fig. 3C). In addition, the effects of B. quintana LPS are targeted specifically to TLR4, since the stimulation of TLR2 was not influenced by the presence or absence of B. quintana LPS in the cultures (not shown).

View larger version (12K): [in this window] [in a new window]

FIG. 3. (A) Human PBMC were stimulated or not for 24 h with E. coli LPS at a concentration of 10 ng/ml. Before stimulation with E. coli LPS, the cells were preincubated with either RPMI or B. quintana LPS at concentrations ranging from 1 to 100 ng/ml. (B) In parallel, PBMC preincubated with B. quintana LPS at 100 ng/ml and thereafter stimulated with E. coli LPS at 10 ng/ml were used for mRNA measurements of proinflammatory cytokines. (C) The findings were also confirmed in CHO cells transfected with either human CD14 or a combination of human TLR4 and human CD14, which were stimulated with E. coli LPS (1 µg/ml), B. quintana LPS (10 µg/ml), or a combination of both. Expression of CD25 on the cell membrane was measured by fluorescence-activated cell sorting analysis. The TLR2 agonist Pam3Cys (5 µg/ml) served as a negative control stimulus. Data are presented as means ± SD (n = 6).

Microarray analysis. We performed oligonucleotide genome array (Affymetrix) analysis to evaluate the impact of B. quintana LPS on gene expression on its own and in combination with E. coli LPS. Cluster analysis revealed that samples treated with E. coli LPS clustered together, whereas samples treated with B. quintana LPS or E. coli LPS plus B. quintana LPS clustered together with the control (Fig. 4A). In spite of the strong influence of donor differences, the effect of B. quintana LPS was unambiguous, as the expression profile for B. quintana LPS alone or in combination with E. coli LPS was hardly different from that of the corresponding control. This indicated that B. quintana LPS exerted no effect on its own but totally inhibited the effect of E. coli LPS. The complete mode of inhibition indeed suggests that B. quintana LPS binds to TLR4 but does not transduce the signal, hereby acting as a competitive inhibitor of E. coli LPS. When the effects of different treatments on gene expression were compared, E. coli LPS regulated 679 genes (>2-fold change and P < 0.05; both up- and downregulation) compared to the control (Fig. 4B). In contrast, B. quintana LPS alone or in combination with E. coli LPS regulated no single gene compared to the control by these criteria.

View larger version (18K): [in this window] [in a new window]

FIG. 4. (A) PBMC were isolated from two individual donors (1 and 2). They were preincubated or not with B. quintana LPS (Inh) for 15 min and thereafter stimulated or not with E. coli LPS (LPS). The numbers designate the donor origins. Gene expression profiling was performed and analyzed as described in Materials and Methods. Hierarchical cluster analysis was performed on genes (vertical dendrogram) and samples (horizontal dendrogram) according to the algorithm described in Materials and Methods to visualize the gene expression patterns and relatedness of samples. Red, upregulation; blue, downregulation; white, average regulation. The magnified view in the box shows sample information in color blocks (first row, individual sample color code; second row, treatment groups, with dark green for control, light green for LPS, dark brown for inhibitor alone, and light brown for LPS plus inhibitor; third row, treatment groups, with dark and light gray for with and without E. coli LPS, respectively) and clustering of samples according to the gene expression profiles. (B) Samples were compared to the control, and the numbers of differentially expressed genes were determined using the criteria of a >2-fold change and a P value of <0.05. The number above each bar indicates the number of differentially expressed genes.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In the present study, we show for the first time that the gram-negative bacterium B. quintana does not trigger host defense mechanisms via TLR4-dependent pathways and, instead, that its LPS component is a powerful TLR4 antagonist. Septic shock is often preceded by gram-negative bacteremia, and the LPS component of these pathogens is known toplay a crucial role in this respect. Interestingly, B. quintana was previously described to induce chronic infections (4), whereas signs of an important acute response were absent in those infected with the pathogen (2, 20). In line with this, the results of our present study strongly suggest that TLR4 is not involved in the host defense against B. quintana, and this fact might also be responsible for the absence of hypotension and shock in B. quintana-infected persons. The limited role of LPS-TLR4 interaction in the induction of an immune response to Bartonella spp. is not singular to B. quintana, since previous reports indicated that other systems, such as the proteinaceous VirB type IV secretion system (17), outer membrane proteins (8), and BadA (22), are more responsible than LPS for the stimulation of cytokines by the related microorganism B. henselae. In line with this, recent evidence suggests that evading TLR4-dependent responses may in fact contribute to the virulence of various gram-negative bacteria (16). On the other hand, other LPS species, such as those isolated from Legionella pneumophila, Leptospira interrogans, Helicobacter pylori, and Porphyromonas gingivalis (11, 17, 18, 26), interact with TLR2 rather than with TLR4. In a recent study, we found that cytokine stimulation by B. quintana is mediated almost entirely by TLR2 (G. Matera et al., submitted for publication), further substantiating the TLR4 independence of B. quintana-induced cytokine production.

Previous studies reported that certain LPS structures could act as TLR4 antagonists. We investigated whether this could also be the case for B. quintana LPS. Our results indicated that interaction of B. quintana LPS with TLR4 does not induce gene transcription, as demonstrated by the lack of gene regulation in an Affymetrix assay of cells incubated with culture medium containing B. quintana LPS. This contrasts with the potent stimulation of a variety of gene families, including the NF-B proinflammatory cytokine genes, when human umbilical vein endothelial cells (7), but not HeLa229 cells (10), were stimulated with intact B. henselae microorganisms. This effect also differs from a recent report in which B. quintana LPS was indicated to induce the release of the chemokine IL-8 (15), while also being able to modulate the apoptosis of endothelial cells (13). One crucial difference with these initial reports is in the method of LPS purification. Indeed, when we stimulated PBMC with LPS only partially purified by the single-step method employed in the previous studies, we also observed TLR2-dependent stimulation of cytokines (not shown). However, single-step purification is prone to retain a high percentage of protein contaminants in the LPS preparation, which likely explains this discrepancy. B. quintana LPS is not the only LPS which is unable to stimulate cytokine production. Recently, another LPS-like molecule derived from Oscillatoria planktonthrix FP1, termed Cyp, was reported to have antagonistic effects on TLR4 (14). Similar properties have been described for the LPSs isolated from other microorganisms, such as Rhodobacter sphaeroides or Chromobacterium violaceum, which are similar to the lipid A precursor Ia (compound 406) (25).

The results of our study are the first to report the TLR4 antagonistic effects of an LPS from Bartonella spp. In line with a low endotoxic potency of an LPS from Bartonella spp., Zahringer and colleagues recently reported that LPS isolated from B. henselae is a penta-acylated deep-rough LPS with endotoxic activities at least 1,000-fold less potent than that of S. enterica LPS (27). B. quintana LPS has a similar deep-rough structure (24), which may explain part of the absent endotoxic activity. However, differences with B. henselae LPS must also be present, as B. quintana LPS is completely unable to stimulate cytokines at concentrations as high as 10 µg/ml. The nature of these differences remains to be established. It should be noted, however, that other LPS species with antagonistic properties, i.e., R. sphaeroides or C. violaceum LPS, have tetra-acylated lipid A structures (25). As we previously proposed, the three-dimensional structure of lipid A is responsible for the differential interaction with TLR4, with strictly cylindrical structures being antagonists, whereas more conical shapes function as strong TLR4 agonists (18). In addition, a hexa-acylated biphosphoryl LPS is likely to represent the structure that could best trigger a proinflammatory response upon binding to TLR4 (17). Other LPS structures with antagonistic properties for human TLR4, such as a mutant LPS from Neisseria meningitidis (T. Sprong, personal communication), have been shown to be agonists of murine TLR4. Initial studies suggest that this is not the case for B. quintana LPS (1).

The role of TLR4 in the pathogenesis of endotoxic shock has been demonstrated extensively, and recently the receptor was proposed to be a potential therapeutic target for several autoimmune diseases, including rheumatoid arthritis (3). In line with this, the results of our study underline the large potential of B. quintana LPS for therapeutic use in these diseases. Firstly, in experiments conducted in vitro, B. quintana LPS could block E. coli LPS-induced effects at ratios as small as 10:1. Secondly, B. quintana LPS specifically binds TLR4 and does not interfere with the stimulation of other receptor pathways. Finally, unlike studies using a selected number of genes, the genome array approach enables the evaluation of total transcript levels in an unbiased way. Therefore, these results clearly demonstrate that B. quintana LPS is a potent TLR4 antagonist and also suggest that its potential to interfere with other immunologic pathways, if any, is extremely limited. Eventually, this might be translated into a limitation in side effects in the case of therapeutic applications. This may give B. quintana LPS a significant advantage with respect to other biological drugs currently on the market.

In conclusion, we demonstrate in the present study that stimulation of cytokines by B. quintana is independent of both its LPS component and TLR4. Moreover, while being unable to stimulate cytokine production, B. quintana LPS is a potent TLR4 antagonist. Because TLR4 proinflammatory signals are involved in a variety of pathological inflammatory reactions, the use of the TLR4 antagonistic properties of B. quintana LPS may prove of potential therapeutic value.

 
We thank Johanna van der Ven-Jongekrijg for her contribution to the experiments involving flow cytometry.

M.G.N. was supported by a Vidi grant from The Netherlands Organization for Scientific Research (NWO).

 
* Corresponding author. Mailing address: Department of General Internal Medicine (463), Radboud University Nijmegen Medical Centre and Nijmegen University Centre for Infectious Diseases, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. Phone: 31-24-3618819. Fax: 31-24-3541734. E-mail: M.Netea{at}aig.umcn.nl

Published ahead of print on 7 July 2007.

Editor: J. L. Flynn

C.P. and S.A.-R. contributed equally to this study.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Abdollahi-Roodsaz, S., L. A. B. Joosten, M. F. Roelofs, T. R. D. J. Radstake, G. Matera, C. Popa, J. W. M. van der Meer, M. G. Netea, and W. B. van den Berg. Inhibition of TLR4 breaks the inflammatory loop in autoimmune destructive arthritis. Arthritis Rheum., in press.
1a
2. Affymetrix.2003. Expression analysis technical manual. Affymetrix, Santa Clara, CA.
2
3. Anderson, B. E., and M. A. Neuman. 1997. Bartonella spp. as emerging human pathogens. Clin. Microbiol. Rev. 10:203-219.[Abstract]
3
4. Brentano, F., D. Kyburz, O. Schorr, R. Gay, and S. Gay. 2005. The role of Toll-like receptor signalling in the pathogenesis of arthritis. Cell. Immunol. 233:90-96.[CrossRef][Medline]
4
5. Brouqui, P., B. Lascola, V. Roux, and D. Raoult. 1999. Chronic Bartonella quintana bacteremia in homeless patients. N. Engl. J. Med. 340:184-189.[Abstract/Free Full Text]
5
6. Capo, C., N. Amirayan-Chevillard, P. Brouqui, D. Raoult, and J. L. Mege. 2003. Bartonella quintana bacteremia and overproduction of interleukin-10: model of bacterial persistence in homeless people. Clin. Infect. Dis. 187:837-844.
6
7. Dehio, C. 2003. Recent progress in understanding Bartonella-induced vascular proliferation. Curr. Opin. Microbiol. 6:61-65.[CrossRef][Medline]
7
8. Dehio, M., M. Quebatte, S. Foser, and U. Certa. 2005. The transcriptional response of human endothelial cells to infection with Bartonella henselae is dominated by genes controlling innate immune responses, cell cycle, and vascular remodelling. Thromb. Haemost. 94:347-361.[Medline]
8
9. Fuhrmann, O., M. Arvand, A. Gohler, M. Schmid, M. Krull, S. Hippenstiel, J. Seybold, C. Dehio, and N. Suttorp. 2001. Bartonella henselae induces NF-kappaB-dependent upregulation of adhesion molecules in cultured human endothelial cells: possible role of outer membrane proteins as pathogenic factors. Infect. Immun. 69:5088-5097.[Abstract/Free Full Text]
9
10. Hirschfeld, M., C. J. Kirschning, R. Schwandner, H. Wesche, J. H. Weis, R. M. Wooten, and J. J. Weis. 1999. Inflammatory signaling by Borrelia burgdorferi lipoproteins is mediated by Toll-like receptor 2. J. Immunol. 163:2382-2386.[Abstract/Free Full Text]
10
11. Kempf, V. A., M. Lebiedziejewski, K. Alitalo, J. H. Walzlein, U. Ehehalt, J. Fiebig, S. Huber, B. Schutt, C. A. Sander, S. Muller, G. Grassl, A. S. Yazdi, B. Brehm, and I. B. Autenrieth. 2005. Activation of hypoxia-inducible factor-1 in bacillary angiomatosis: evidence for a role of hypoxia-inducible factor-1 in bacterial infections. Circulation 111:1054-1062.[Abstract/Free Full Text]
11
12. Lepper, P. M., M. Triantafilou, C. Schumann, E. M. Schneider, and K. Triantafilou. 2005. Lipopolysaccharides from Helicobacter pylori can act as antagonists for Toll-like receptor 4. Cell. Microbiol. 7:519-528.[CrossRef][Medline]
12
13. Li, C., and W. H. Wong. 2001. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc. Natl. Acad. Sci. USA 98:31-36.[Abstract/Free Full Text]
13
14. Liberto, M. C., G. Matera, A. G. Lamberti, G. S. Barreca, A. Quirino, and A. Foca. 2003. In vitro Bartonella quintana infection modulates the programmed cell death and inflammatory reaction of endothelial cells. Diagn. Microbiol. Infect. Dis. 45:107-115.[CrossRef][Medline]
14
15. Macagno, A., M. Molteni, A. Rinaldi, F. Bertoni, A. Lanzavecchia, C. Rossetti, and F. Sallusto. 2006. A cyanobacterial LPS antagonist prevents endotoxin shock and blocks sustained TLR4 stimulation required for cytokine expression. J. Exp. Med. 203:1481-1492.[Abstract/Free Full Text]
15
16. Matera, G., M. C. Liberto, A. Quirino, G. S. Barreca, A. G. Lamberti, M. Iannone, E. Mancuso, E. Palma, F. A. Cufari, D. Rotiroti, and A. Foca. 2003. Bartonella quintana lipopolysaccharide effects on leukocytes, CXC chemokines and apoptosis: a study on the human whole blood and a rat model. Int. Immunopharmacol. 3:853-864.[Medline]
16
17. Montminy, S. W., N. Khan, S. McGarth, M. J. Walkowicz, F. Sharp, J. E. Conlon, K. Fukase, S. Kusumoto, C. Sweet, K. Miyake, S. Akira, R. J. Cotter, J. D. Goguen, and E. Lien. 2006. Virulence factors of Yersinia pestis are overcome by strong lipopolysaccharide response. Nat. Immunol. 7:1066-1073.[CrossRef][Medline]
17
18. Munford, R. S., and A. W. Varley. 2006. Shield as signal: lipopolysaccharides and the evolution of immunity to gram-negative bacteria. PLoS Pathog. 2:e67.[CrossRef][Medline]
18
19. Netea, M. G., M. van Deuren, B. J. Kullberg, J. M. Cavaillon, and J. W. M. van der Meer. 2002. Does the shape of lipid A determine the interaction of lipopolysaccharides with Toll-like receptors? Trends Immunol. 23:135-139.[CrossRef][Medline]
19
20. Poltorak, A., X. He, I. Smirnova, M. Y. Liu, C. van Huffel, X. Hu, D. Birdwell, E. Alejos, M. Silva, C. Galanos, M. Freundenberg, P. Ricciardi-Castagnoli, B. Layton, and B. Beutler. 1998. Defective LPS signalling in C3H/HeJ and C57Bl/10ScCr mice: mutations in TLR4 gene. Science 282:2085-2088.[Abstract/Free Full Text]
20
21. Raoult, D., P. E. Fournier, M. Drancourt, T. J. Marrie, J. Etienne, J. Cosserat, P. Cacoub, Y. Poinsignon, P. Leclercq, and A. M. Sefton. 1996. Diagnosis of 22 new cases of Bartonella endocarditis. Ann. Intern. Med. 125:646-652.[Abstract/Free Full Text]
21
22. Rieneck, K., L. F. Bovin, K. Josefsen, K. Buschard, M. Svenson, and K. Bendtzen. 2000. Massive parallel gene expression profiling of RINm5F pancreatic islet beta-cells stimulated with interleukin-1beta. APMIS 108:855-872.[Medline]
22
23. Riess, T., S. G. Andersson, A. Lupas, M. Schaller, A. Schafer, P. Kyme, J. Martin, J. H. Walzlein, U. Ehehalt, H. Lindroos, M. Schirle, A. Nordheim, I. B. Autenrieth, and V. A. Kempf. 2004. Bartonella adhesin A mediates a proangiogenic host cell response. J. Exp. Med. 200:1267-1278.[Abstract/Free Full Text]
23
24. Rietschel, E. T., H. Brade, L. Brade, et al. 1987. Lipid A, the endotoxic center of bacterial lipopolysaccharides: relation of chemical structure to biological activity, p.25-53. In A. Sturk and S. J. van Deventer (ed.), Detection of bacterial endotoxins with the Limulus amebocyte lysate test. Alan R. Liss, Inc., New York, NY.
24
25. Schmid, M. C., R. Schulein, M. Dehio, G. Denecker, I. Carena, and C. Dehio. 2004. The VirB type IV secretion system of Bartonella henselae mediates invasion, proinflammatory activation and antiapoptotic protection of endothelial cells. Mol. Microbiol. 52:81-92.[CrossRef][Medline]
25
26. Schromm, A. B., K. Brandenburg, H. Loppnow, A. P. Moran, M. H. Koch, E. T. Rietschel, and U. Seydel. 2000. Biological activities of lipopolysaccharides are determined by the shape of their lipid A portion. Eur. J. Biochem. 267:2008-2013.[Medline]
26
27. 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.[Abstract/Free Full Text]
27
28. Zahringer, U., B. Lindner, Y. A. Knirel, W. M. van den Akker, R. Hiestand, H. Heine, and C. Dehio. 2004. Structure and biological activity of the short-chain lipopolysaccharide from Bartonella henselae ATCC 49882T. J. Biol. Chem. 279:21046-21054.[Abstract/Free Full Text]

---

Infection and Immunity, October 2007, p. 4831-4837, Vol. 75, No. 10
0019-9567/07/$08.00+0     doi:10.1128/IAI.00237-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Abdollahi-Roodsaz, S., Joosten, L. A.B., Koenders, M. I., van den Brand, B. T., van de Loo, F. A.J., van den Berg, W. B. (2009). Local Interleukin-1-Driven Joint Pathology Is Dependent on Toll-Like Receptor 4 Activation. Am. J. Pathol. 175: 2004-2013 [Abstract] [Full Text]  
• Chai, L. Y. A., Kullberg, B. J., Vonk, A. G., Warris, A., Cambi, A., Latge, J.-P., Joosten, L. A. B., van der Meer, J. W. M., Netea, M. G. (2009). Modulation of Toll-Like Receptor 2 (TLR2) and TLR4 Responses by Aspergillus fumigatus. Infect. Immun. 77: 2184-2192 [Abstract] [Full Text]  
• Netea, M. G., Nold-Petry, C. A., Nold, M. F., Joosten, L. A. B., Opitz, B., van der Meer, J. H. M., van de Veerdonk, F. L., Ferwerda, G., Heinhuis, B., Devesa, I., Funk, C. J., Mason, R. J., Kullberg, B. J., Rubartelli, A., van der Meer, J. W. M., Dinarello, C. A. (2009). Differential requirement for the activation of the inflammasome for processing and release of IL-1{beta} in monocytes and macrophages. Blood 113: 2324-2335 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Popa, C. Articles by Netea, M. G. Search for Related Content PubMed PubMed Citation Articles by Popa, C. Articles by Netea, M. G.