The American Laryngological, Rhinological & Otological Society, Inc.

Volume 117(5), May 2007, pp 841-847

Pseudomonas aeruginosa Lipopolysaccharide Induces Osteoclastogenesis Through a Toll-Like Receptor 4 Mediated Pathway in Vitro and in Vivo
[Independent Papers] Zhuang, Lei MD; Jung, Jae Y. MD, PhD; Wang, Eric W. MD; Houlihan, Patrick MS; Ramos, Lisette MD; Pashia, Mary MS; Chole, Richard A. MD, PhD From the Department of Otolaryngology (l.z., j.y.j., p.h., l.r., m.p., r.a.c.), and the Fay and Carl Simons Center for the Study of the Biology of Hearing and Deafness of the Central Institute for the Deaf (r.a.c.), Washington University School of Medicine, Saint Louis, MO, U.S.A. Editor's Note: This Manuscript was accepted for publication January 5, 2007. This work was supported by NIH grant DC00263-18 (PI: RA Chole), training grant T32 DC00022 (PI: JG Neely), and grant P30 DC004665 (PI: D Simmons) from the National Institute on Deafness and Other Communicative Disorders, National Institute of Health, Bethesda, MD, and the Fay and Carl Simons Center for the Study of the Biology of Hearing and Deafness of the Central Institute for the Deaf, Washington University School of Medicine, St Louis, MO, U.S.A. Send correspondence to: Richard A. Chole, MD, PhD, Department of Otolaryngology/Head and Neck Surgery, Washington University School of Medicine, Campus Box 8115, 660 S Euclid Ave, Saint Louis, MO 63110, U.S.A. E-mail: rchole@wustl.edu.

Abstract
Objectives: Bacterial infections near bone result in localized inflammatory osteolysis, a significant complication of chronic ear infections. While many bacterial products may be involved, lipopolysaccharide (LPS) has been implicated as a major mediator of inflammation and osteolysis. However, the mechanisms by which LPS promotes bone resorption have not been clearly established. There is no consensus on whether LPS acts directly or indirectly on osteoclast precursors (bone marrow monocytes [BMM]) to induce bone resorption. In light of the role of Pseudomonas aeruginosa, in chronic ear infections, we investigated the effects of P. aeruginosa LPS on osteoclastogenesis in vivo and in vitro.

Methods: Wild-type C57BL/6J and toll-like receptor 4 knock-out (TLR4-/-) mice received subcutaneous calvarial injections of 250 g of P. aeruginosa LPS or phosphate buffered saline (PBS) only (n = 5 per group). Osteoclastic bone resorption was assessed histologically. The effect of P. aeruginosa LPS on bone resorption was assessed in vitro using combinations of BMMs and osteoblasts with and without functional toll-like receptor 4 (TLR4). Results: In vivo, P. aeruginosa LPS induced robust osteolysis, and this effect was completely abrogated in mice lacking expression of TLR4. In vitro, P. aeruginosa LPS failed to induce development of osteoclasts directly in BMMs. However, P. aeruginosa LPS did stimulate osteoclastogenesis in BMM-osteoblast cocultures. Conclusions: P. aeruginosa LPS acts indirectly through osteoblasts to induce bone resorption. Optimal osteoclastogenesis in vitro required functional TLR4 expression in both BMMs and osteoblasts.

INTRODUCTION
When present near bone, bacterial infections are associated with an inflammatory process that results in localized osteolysis. Lipopolysaccharide (LPS), a major component of the outer membrane of Gram-negative bacteria, is believed to be an important mediator of osteolysis. In chronic otitis media with and without formation of cholesteatoma, bacterial infection and chronic inflammation are associated with localized erosion of adjacent bony structures including the ossicles, cochlea, and mastoid, which can result in both conductive and sensorineural hearing loss (HL). LPS has been detected in human ear infections and is correlated with bone destruction.1 Pseudomonas aeruginosa is the most common Gramnegative species isolated from chronically infected ears.2 It has been shown that P. aeruginosa LPS is immunogenic, mitogenic, and toxic in C3H/HEJ mice, which are unresponsive to Escherichia coli LPS,3 suggesting that LPS from different bacteria have varying effects in different assays. However, to our knowledge, there are no reports examining the effect of P. aeruginosa LPS on osteoclastogenesis. Regardless of the initial pathology, bone destruction occurs through a common pathwaythe recruitment and activation of osteoclasts. Osteoclasts are multinucleated giant cells originating from hematopoietic precursors (bone marrow monocytes [BMMs]). Development of osteoclasts occurs in the presence of

monocyte/macrophage colony-stimulating factor (mCSF) and through the stimulation of the receptor activator of nuclear factor kappa B (NF[kappa]B) ligand (RANKL) by stromal cells or osteoblasts. These trophic factors are both necessary and sufficient for the development of osteoclasts in vitro. LPS may modulate this pathway and has been shown, in vitro, to increase RANKL expression by fibroblasts 4 and osteoblasts.5 LPS is a potent stimulator of the innate immune system through its putative receptor, toll-like receptor 4 (TLR4).6 TLR4 is a member of the toll-like receptor family of pattern recognition receptors that binds common structural motifs from a wide variety of pathogens including bacteria, fungi, protozoa, and viruses.6 These receptors are expressed on the surface of different types of cells involved in osteoclast regulation including primary murine osteoblasts and ST2 stromal cells, and primary murine BMMs and osteoclasts. 7 A functional role for TLR4 in osteoclastogenesis is suggested by a report demonstrating reduced bone resorption in TLR4-deficient mice in response to mixed anaerobic endodontic infections in vivo. 8 Previous studies have supported a role for LPS in bone resorption. Hou et al. showed significant bone resorption following oral infection with mixed anaerobic bacteria (Prevotella intermedia, Fusobacterium nucleatum, Streptococcus intermedius, and Peptostreptococcus micros).8 More directly, bone resorption in vivo was found with injections of purified LPS from E. coli in alveolar bone.9 Although bacterial infection and purified LPS induced robust bone resorption in vivo in many different model systems, there is a lack of consensus on the effects of LPS on osteoclast development in vitro. To investigate the effects of LPS on osteoclastogenesis, we tested P. aeruginosa LPS in a murine in vivo bone resorption model as well as two in vitro culture systems to further delineate LPS activity on BMMs and osteoblasts. We demonstrated that P. aeruginosa LPS robustly induced bone resorption in vivo and that this effect was abrogated in mice lacking TLR4. In vitro, P. aeruginosa LPS was unable to stimulate osteoclast development in isolated cultures of BMMs but resulted in a dose-dependent increase in osteoclastogenesis in osteoblast-BMM cocultures. In these experiments, we carefully controlled for the developmental stage of the osteoclast precursors by suppressing any endogenous RANKL stimulation by pretreatment with osteoprotegrin (OPG). OPG is a soluble decoy receptor that competes with RANKL for RANK on the osteoclast. Our findings suggest that, in vitro, LPS is principally acting indirectly through TLR4 on osteoblasts and not directly on osteoclast precursors (BMM).

MATERIALS AND METHODS

Wild-type (C57BL/6J) and toll-like receptor 4 knock-out (TLR4-/-) (C57BL/10ScNJ) mice were obtained from Jackson Laboratories (Bar Harbor, ME). C57BL/10ScNJ mice are homozygous for a spontaneous mutation in the TLR4 locus that eliminates expression of the mRNA and TLR protein.10 Age-matched, adult (57 weeks old) male mice were used for these studies in accordance with the Public Health Service (PHS) Policy on Humane Care and Use of Laboratory Animals, the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The protocol was approved by the Institutional Animal Care and Use Committee of the Washington School of Medicine in St. Louis. The glutathione-S-transferase tagged receptor activator of NF[kappa]B construct (GST-RANKL) and CMG14-12 cells were generously contributed by Drs Steven Teitelbaum and F. Patrick Ross (Department of Pathology, Washington University School of Medicine, St Louis, MO). All other reagents were obtained from Sigma (St. Louis, MO) unless otherwise specified. RANKL was purified, as described previously.11 The purified RANKL protein was tested for endotoxin contamination by limulus amoebocyte lysate assay (Cambrex Bio Services, Walkersville, MD); 0.000014 ng/L endotoxin was identified per 1 mg of RANKL. Supernatants collected from cultured CMG14-12 cells producing high concentrations of mouse mCSF were used to obtain purified BMMs using a protocol adapted from Dr Sunao Takeshita 12 (Department of Pathology, Washington University School of Medicine, St Louis, MO). Growing colonies of mCSF-producing cells were expanded, and culture supernatants were collected. The activity of mCSF activity was measured by 3-(4, 5-dimethyl-2-thiazol)-2, 5-dihenyltetrazolium bromide (MTT) assay using mNFS-60 cells. CMG14-12 produced approximately 1 to 3 g/mL of mCSF in culture supernatant, as measured by a colony formation assay using mouse bone marrow cells.12

In Vivo Studies
Animals were anesthetized, and subcutaneous injections were administered with sterile 25-gauge needle over the calvaria. Wildtype (C57BL/6J) and TLR4-/- (C57BL/10ScNJ) groups received 250 g of P. aeruginosa LPS (Sigma, St Louis, MO) in 250 L of sterile PBS. As a control, age-matched, wild-type, and TLR4-/- mice were injected with 250 L of PBS only (n = 5 per group). After 5 days, animals were euthanized, and their calvaria were processed for routine histology. Calvarial sections were visualized on an Olympus BH2 microscope (Sigma Chemical Co., St Louis, MO) and photographed on a Sony digital photograph camera DKC-5000 (Sony Corp., Tokyo, Japan). Image analysis was performed using ImageJ (Rasband WS, ImageJ, U.S. National Institutes of Health, Bethesda,

Maryland, USA, http://rsb.info . nih.gov/ij/, 19972005). Bone resorption was assessed by counting the number of resorption pits and by measuring the length of the resorption pits seen on the dorsal surface of the calvaria in accordance with standardized histomorphometry methods.13

Bone Marrow Monocytes (BMMs)

Mice were euthanized by cervical dislocation under halothane anesthesia. Femurs and tibias of mice were aseptically removed and dissected free of adhering tissue. The bone marrow cavity was flushed with a minimum essential medium alpha ([alpha]-MEM) (Invitrogen, Carlsbad, CA) using a sterile 26-gauge needle. The marrow cells were collected, washed with [alpha]-MEM, and treated with 0.747% NH4Cl-0.017% Tris-Cl (pH of 7.2) PBS solution to lyse the erythrocytes. After washing, cells were cultured in [alpha]-MEM containing 10% fetal bovine serum (FBS) (Invitrogen), 100 g/mL penicillin G, 100 g/mL streptomycin, 100 ng/mL OPG (R&D Systems, Minneapolis, MN), and 10% CMG14-12 culture supernatant at 5 106 cells per 10-cm suspension culture dish (Corning, Acton, MA). After 3 days of incubation at 37C in a humidified atmosphere containing 5% CO2, cultures were washed twice with PBS containing 0.02% EDTA to remove nonadherent cells. Cells were harvested by pipetting with 0.02% EDTA in PBS, and seeded at 5 106 cells in a new 10-cm tissue culture dish. After three cycles of harvesting and seeding, cells were counted and plated for experiments.

Osteoblasts
Osteoblastic cells were isolated from calvaria of 1- to 3-day-old neonatal mice. Ten to 20 calvaria were collected and digested sequentially five times with a solution of 0.1% collagenase (Sigma, St. Louis, MO) and 0.2% Dispase (Roche, Indianapolis, IN). Isolated cells were combined, suspended with [alpha]-MEM containing 10% FBS, 100 g/mL penicillin G, and 100 g/mL streptomycin and cultured for 4 days until 75% confluent. Cells were washed with PBS and harvested with 0.2% trypsin/0.02% EDTA/PBS.

Cell Cultures
Isolated bone marrow monocyte (BMM) cultures were established by plating 5 105 BMM cells/mL of [alpha]-MEM containing 10% FBS, 10 ng/mL m-CSF, 100 g/mL penicillin G, and 100 g/mL streptomycin. Cocultures were established by combining osteoblasts with BMM cells at a density of 5 104 cells/mL. Cells were maintained in [alpha]MEM containing 10% FBS, 10 ng/mL mCSF, 100 g/mL penicillin G, 100 g/mL streptomycin, 10 nM 1,25(OH)2D3, and 10 nM

dexamethasone (Sigma Chemical Co., St Louis, MO). Cells were incubated at 37C in a humidified atmosphere containing 5% CO 2 and supplemented with fresh media and cytokines every other day for 9 days.

Tartrate-Resistant Acid Phosphatase Assay

The presence of osteoclast cells was evaluated by cytochemical staining of tartrate-resistant acid phosphatase (TRAP) using a commercial kit (Sigma Chemical Co., St. Louis, Mo). A cell that was TRAP positive and contained three or more nuclei was considered an osteoclastic cell. Cells were visualized on an Olympus BH2 microscope (Sigma Chemical Co., St Louis, MO). Photographs were taken on a Sony digital photograph camera DKC-5000. A high-throughput assay evaluating osteoclast formation (TRAP-positive cells) was determined by the addition of a colorimetric substrate, 10 mmol/L p-nitrophenyl phosphate (Jas Diagnostics, Miami, FL), in the presence of sodium tartrate (Sigma, St. Louis, MO) at a pH of 5 to 6. The reaction product was quantified by measuring optical absorbance at 405 nm on a Synergy HT plate reader (BioTek, Winooski, VT).

Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)

Total RNA was isolated from mouse calvaria and in vitro cultures using a TRIZOL reagent (GIBCO, Rockville, MD), following the manufacturer's protocols. RNA concentration was determined by absorption at 260 nm on a Synergy HT microplate reader (BioTek, Winooski, VT). Isolated total RNA (1.0 g) was digested with DNAse I (Roche Molecular Biochemicals, Indianapolis, IN) for 15 minutes at room temperature, then heat inactivated at 65C for 10 minutes. Total RNA was then used to synthesize cDNA using the kit 1st strand cDNA Synthesis Kit for RT-PCR (Roche Molecular Biochemicals, Indianapolis, IN). All primers were synthesized by Integrated DNA Technologies (Coralville, IA); primers are listed in Table I. RT-PCR products were fractionated by electrophoresis on a 5% acrylamide gel stained with ethidium bromide and visualized on a GelDoc 2000 (BioRad, Hercules, CA).

Fig. 3. P. aeruginosa LPS TABLE I. Reverse Transcriptase-Polymerase dose response in Chain Reaction (RT-PCR) Primers. cocultures. (A) Cocultures consisting of wild-type BMMs

(WTBMM) and osteoblasts (WTOB) demonstrated a dosedependent increase in osteoclast development when exposed to LPS for 9 days, although there was some suppression at the highest concentration of LPS (P < .05). Both WTBMM cocultured with TLR4-/osteoblasts (TLR4-/OB), as well as TLR4-/BMM co-cultured with TLR4-/- OB failed to form osteoclasts at varying concentrations of LPS. TLR4-/- BMMs cocultured with WTOB demonstrated an attenuated dosedependent increase in osteoclast formation when exposed to LPS (P < .05). Fig. 2. P. aeruginosa LPS dose response in BMMs. (A) Pure populations of BMMs were treated with 10 ng/mL mCSF plus varying concentrations (0.001 g to 10 g/mL) of LPS. After 9 days, no significant formation of osteoclasts was seen. (B) Addition of a low concentration of RANKL (5 ng/mL) in conjunction with varying concentrations of LPS also failed to induce any significant development of osteoclasts. (C) BMMs were treated with 5 ng/mL RANKL plus 10

ng/mL mCSF. Cells received 0.1 g/mL LPS following 1 to 5 days of exposure to RANKL. LPS failed to induce osteoclast formation. (D) RANKL dose dependently stimulated osteoclastogenesis. When these cells were simultaneously treated with moderate levels of LPS (1 g/mL), osteoclast development was suppressed (P <= . 01). Photographs of TRAP-stained BMMs (E): (A) control condition, (B) with 50 ng/mL RANKL, (C) with 1 g/mL LPS plus 50 ng/mL RANKL, and (D) with 1 g/mL LPS. (F). Increased expression of macrophage markers CD11b, CD14, and F4/80 in BMMs treated with LPS, compared with BMMs treated with 50 ng/mL RANKL. Expression of surface marker RANK was detected in the RANKLtreated BMMs. Fig. 1. P. aeruginosa LPS induced bone resorption in wildtype mice, but not in TLR4-/- mice. A single subcutaneous injection of P. aeruginosa LPS (250 g) was applied over the mouse calvaria versus sterile PBS. LPSinjected wild-type mice (wild-type LPS) exhibited a significant

increase in bone resorption compared with PBS-injected wildtype and TLR4-/- mice and LPS-injected TLR4-/- mice (TLR4-/LPS), both in terms of (A) resorption pit number and (B) percent resorption surface (P <= .01). (C) Calvarial sections of wild-type mice injected with PBS (A) or LPS (B); and TLR4-/- mice injected with PBS (C) or LPS (D). Microscopic magnification, 200.
[Help with image viewing] [Email Jumpstart To Image]

The BMMs used in these experiments only differentiated into osteoclasts in the absence of P. aeruginosa LPS and in the presence of high doses of RANKL (10 and 50 ng/mL) (Fig. 2D, white bars). Osteoclastogenesis is dramatically suppressed with the addition of 1 g/mL P. aeruginosa LPS (Fig. 2D, black bars). However, at the highest concentrations of RANKL (50 ng/mL) with 1 g/mL P. aeruginosa LPS, some cells stained weakly TRAP positive. These cells were mostly mononuclear with rare multinucleated cells (Fig. 2E, panel C). BMMs treated with P.

aeruginosa LPS demonstrate enhanced expression of the macrophage markers CD11b, CD14, and F4/80 and decreased expression of RANK compared with osteoclasts (Fig. 2F), suggesting that BMMs treated with P. aeruginosa LPS commit to an alternate cell lineage.

P. aeruginosa LPS Potentiates Osteoclast Development in Osteoblast-BMM Cocultures

While P. aeruginosa LPS stimulated bone resorption in vivo, it failed to induce osteoclastogenesis when applied to BMM monoculture, suggesting that the in vivo effect of LPS is mediated, at least in part by another cell type. Since osteoblasts have a critical role in the development of osteoclasts in vivo, we examined the effect of LPS in osteoblast-bone marrow monocyte cocultures (Ob-BMM cocultures). Addition of P. aeruginosa LPS alone to osteoblast-BMM cocultures did not result in significant levels of

osteoclastogenesis (data not shown). However, the simultaneous addition of low levels of RANKL (5 ng/mL) and P. aeruginosa LPS did stimulate osteoclastogenesis. The increase was dose dependent except at the highest concentration of P. aeruginosa LPS, which appears to suppress osteoclastogenesis (Fig. 3A).

[Help with image viewing] [Email Jumpstart To Image]

P. aeruginosa LPS Does Not Induce Osteoclast Development in BMM Cultures

P. aeruginosa LPS alone, at concentrations from 0.001 to 10 g/mL, failed to induce osteoclast formation in BMMs in vitro, and no TRAP-positive cells were detected (Fig. 2A). It has previously been shown 12 that low levels of RANKL, insufficient to induce osteoclastogenesis by itself, can potentiate the ability of tumor necrosis factor-alpha (TNF-alpha) to induce

osteoclastogenesis. Following the same logic, we supplemented various doses of P. aeruginosa LPS with a dose of RANKL previously determined to be nonpermissive for osteoclastogenesis (5 ng/mL for these experiments). The combinations of P. aeruginosa LPS and RANKL did not induce osteoclastogenesis in purified murine BMMs (Fig. 2B). Pretreatment of BMMs with a nonpermissive dose of RANKL for 2 to 5 days before exposure to P. aeruginosa LPS also failed to potentiate osteoclastogenesis (Fig. 2C).

[Help with image viewing] [Email Jumpstart To Image]

The Phenotype of TLR4 Mutations in Cocultures is Cell Type Dependent

Our results demonstrated that TLR4-/- mice have attenuated bone resorption in response to in vivo exposure of P. aeruginosa LPS (Fig. 1). To determine whether this was due to the loss of TLR4 function in the osteoblast or osteoclast cell lineages, we conducted a series of coculture experiments pairing cells derived from TLR4 wild-type C57BL/6J mice with those derived from TLR4-/- C57BL/10ScNJ mice. Our results demonstrated a striking suppression of osteoclastogenesis when the osteoblasts lacked TLR4 (Fig. 3). This suppression was observed with both wild type and TLR4-/- bone marrow monocytes,

suggesting that TLR4 in the osteoblastic lineage is the primary determinant of whether P. aeruginosa LPS induces osteoclastogenesis in vitro. This conclusion is supported by the finding that cocultures containing wild-type osteoblasts and TLR4-/- BMM respond to P. aeruginosa LPS by a dose-dependent increase in osteoclast formation (Fig. 3). However, it should be noted that this increase in osteoclast formation is attenuated compared with the response observed in P. aeruginosa LPS treated cocultures combining wild-type cells from both lineages (Fig. 3). These results suggest that TLR4 also plays some role in the BMMs themselves but that this function is secondary to its role in osteoblasts.

DISCUSSION
Our results demonstrate that LPS purified from P. aeruginosa is a potent activator of bone resorption in vivo and that this effect is mediated through TLR4 expression primarily on the cells of osteoblastic lineage. Although there are many cell types that are potentially involved in the recruitment and activation of osteoclasts in vivo, we focused on the two main cell types involved: BMMs (osteoclast precursors) and osteoblasts. Our results suggest that P. aeruginosa LPS does not directly act on osteoclast precursors to generate osteoclasts. This conclusion is in contrast to the results from Suda et al.14 and Jiang et al.,15 which demonstrated direct effects of LPS on osteoclast precursors. It seems likely that the differences in these studies are methodological in nature. In the protocol used by Suda et al., the preosteoclasts may have been exposed to significant amounts of endogenous RANKL since BMMs were cultured for 5 to 6 days in the presence of dihydroxyvitamin D3, which supports the survival of osteoblast-like cells. Additionally, more than 70% of the cells were already TRAP positive at the time LPS was added, suggesting that these precursors were already committed to an osteoclast lineage. In our experiments, we minimized endogenous RANKL exposure to osteoclast precursors with OPG treatment. The difference between the E. coli and P. aeruginosa LPS may also exist and contribute to the methodological differences between these studies. In the study by Jiang et al., disaggregated splenocytes rather than BMMs were used to obtain osteoclast precursors, and whole-killed bacteria, rather than purified LPS, were used to stimulate osteoclastogenesis. In light of these methodological differences and varying outcomes, it may be that the direct effect of LPS on BMMs is dependent on their state of differentiation. Although low doses of LPS do not induce osteoclastogenesis in monoculture, higher doses have a clear inhibitory effect on BMMs.

Along with inhibition of osteoclastogenesis, we also observed upregulation of macrophage markers CD11b, CD14, and F4/80 and a downregulation of RANK, suggesting that these cells develop along an alternate pathway. P. aeruginosa LPS also demonstrated a clear inhibition of osteoclast development when added to BMM monocultures simultaneously treated with RANKL and mCSF. Based on these studies, we conclude that P. aeruginosa LPS directly inhibits BMMs, suppressing osteoclastogenesis, but we cannot rule out the possibility that LPS may have a direct effect on the survival of preformed osteoclasts (stimulated with RANKL and mCSF before the addition of LPS) as previously shown.7 The coculture experiments suggest that P. aeruginosa LPS acts primarily on the osteoblast to induce osteoclastogenesis in vivo but that maximal induction requires TLR4 function in both isolated BMM and osteoblasts. This is consistent with previous reports suggesting that LPS may upregulate RANKL on osteoblasts and stromal cells. 5

CONCLUSION
Our studies found that P. aeruginosa LPS act indirectly through osteoblasts to induce bone resorption. Optimal osteoclastogenesis in vitro required functional TLR4 expression in both BMMs and osteoblasts. Additionally, these findings highlight the potential overlap between RANKL and TLR signaling, which share common intracellular signaling molecules such as TRAF6 and NF[kappa]B. Continuing studies will further investigate the relationship between LPS, TLR4, and osteoclastogenesis. Elucidating the relationship between inflammatory and osteoclastogenic factors will contribute fundamentally to the understanding of bacterial pathogenesis and bone destruction. These are important prerequisites to future studies that might yield new therapeutic intervention for the management and possible amelioration of cholesteatoma.

Acknowledgments
The authors thank Drs Steve Teitelbaum, F. Patrick Ross, and Sunao Takeshita (Department of Pathology, Washington University School of Medicine, St Louis, MO) for the RANKL construct, the CMG14-12 cells, and for helpful discussion; and Dr Steve Scholnick for his assistance and critical review of the paper.

BIBLIOGRAPHY
1. Peek FA, Huisman MA, Berckmans RJ, Sturk A, Van Loon J, Grote JJ. Lipopolysaccharide concentration and bone resorption in cholesteatoma. Otol Neurotol 2003;24:709713. Ovid Full Text Bibliographic Links [Context
Link]

2. Attallah MS. Microbiology of chronic suppurative otitis media with cholesteatoma. Saudi Med J 2000;21:924927. Bibliographic Links [Context
Link]

3. Pier GB, Markham RB, Eardley D. Correlation of the biologic responses of C3H/HEJ mice to endotoxin with the chemical and structural properties of the lipopolysaccharides from Pseudomonas aeruginosa and Escherichia coli. J Immunol 1981;127:184191. Bibliographic Links [Context Link] 4. Wada N, Maeda H, Yoshimine Y, Akamine A. Lipopolysaccharide stimulates expression of osteoprotegerin and receptor activator of NF-kappa B ligand in periodontal ligament fibroblasts through the induction of interleukin-1 beta and tumor necrosis factor-alpha. Bone 2004;35:629635.
Bibliographic Links [Context Link]

5. Suda K, Udagawa N, Sato N, et al. Suppression of osteoprotegerin expression by prostaglandin E2 is crucially involved in lipopolysaccharideinduced osteoclast formation. J Immunol 2004;172:25042510.
Bibliographic Links [Context Link]

6. Pasare C, Medzhitov R. Toll-like receptors: linking innate and adaptive immunity. Microbes Infect 2004;6:13821387. Bibliographic Links [Context
Link]

7. Itoh K, Udagawa N, Kobayashi K, et al. Lipopolysaccharide promotes the survival of osteoclasts via toll-like receptor 4, but cytokine production of osteoclasts in response to lipopolysaccharide is different from that of macrophages. J Immunol 2003;170:36883695. Bibliographic Links [Context
Link]

8. Hou L, Sasaki H, Stashenko P. Toll-like receptor 4-deficient mice have reduced bone destruction following mixed anaerobic infection. Infect Immun 2000;68:46814687. Bibliographic Links [Context Link] 9. Umezu A, Kaneko N, Toyama Y, Watanabe Y, Itoh H. Appearance of osteoclasts by injections of lipopolysaccharides in rat periodontal tissue. J Periodontal Res 1989;24:378383. Bibliographic Links [Context Link] 10. Vogel SN, Hansen CT, Rosenstreich DL. Characterization of a congenitally LPS-resistant, athymic mouse strain. J Immunol 1979;122:619622. Bibliographic Links [Context Link] 11. Jung JY, Lin AC, Ramos LM, Faddis BT, Chole RA. Nitric oxide synthase I mediates osteoclast activity in vitro and in vivo. J Cell Biochem 2003;89:613621. Bibliographic Links [Context Link] 12. Lam J, Takeshita S, Barker JE, Kanagawa O, Ross FP, Teitelbaum SL. TNF-alpha induces osteoclastogenesis by direct stimulation of macrophages exposed to permissive levels of RANK ligand. J Clin Invest 2000;106:1481 1488. Ovid Full Text Bibliographic Links [Context Link]

13. Arnett TR. Methods in Bone Biology. London: Chapman and Hall; 1998.
[Context Link]

14. Suda K, Woo JT, Takami M, Sexton PM, Nagai K. Lipopolysaccharide supports survival and fusion of preosteoclasts independent of TNF-alpha, IL1, and RANKL. J Cell Physiol 2002;190:101108. Bibliographic Links
[Context Link]

15. Jiang Y, Mehta CK, Hsu TY, Alsulaimani FF. Bacteria induce osteoclastogenesis via an osteoblast-independent pathway. Infect Immun 2002;70:31433148. Bibliographic Links [Context Link]

Key Words: Toll-like receptor 4 (TLR4); lipopolysaccharide (LPS); bone marrow monocyte (BMM); osteoclast; osteoblast

Statistical Analysis
Statistical analysis was performed using SigmaStat (SPSS Inc., Chicago, IL). One-way analysis of variance (ANOVA) was used for all analyses with the power of the performed tests at [alpha] = 0.05.

RESULTS

P. aeruginosa LPS Induces Bone Resorption in Wild-Type Mice but Not the TLR4-/- Mice in Vivo
P. aeruginosa LPS-injected wild-type mice exhibited a greater than 2fold increase in bone resorption compared with the PBS-injected mice as measured by resorption pit number and percent resorbed surface (Fig. 1). The enhanced bone resorption seen in the calvaria of the P. aeruginosa LPS-injected wild-type mice was significantly attenuated in the TLR4-/- mice. These results demonstrate that P. aeruginosa LPS is a potent inducer of bone resorption in vivo and that this effect is mediated through TLR4.

[Help with image viewing] [Email Jumpstart To Image]