Profound Differences in Leukocyte-Endothelial Cell Responses to Versus Lipoteichoic Acid

This information is current as Bryan G. Yipp, Graciela Andonegui, Christopher J. Howlett, of September 25, 2021. Stephen M. Robbins, Thomas Hartung, May Ho and Paul Kubes J Immunol 2002; 168:4650-4658; ; doi: 10.4049/jimmunol.168.9.4650

http://www.jimmunol.org/content/168/9/4650 Downloaded from

References This article cites 49 articles, 24 of which you can access for free at: http://www.jimmunol.org/content/168/9/4650.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 25, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Profound Differences in Leukocyte-Endothelial Cell Responses to Lipopolysaccharide Versus Lipoteichoic Acid1

Bryan G. Yipp,* Graciela Andonegui,† Christopher J. Howlett,‡ Stephen M. Robbins,‡ Thomas Hartung,§ May Ho,* and Paul Kubes2†

We have investigated the effects of LPS from Escherichia coli, lipoteichoic acid (LTA), and (PepG) from Staphy- lococcus aureus, and live S. aureus on leukocyte-endothelial interactions in vivo using intravital microscopy to visualize muscle microvasculature. Systemic vs local administration of LPS induced very different responses. Local administration of LPS into muscle induced significant leukocyte rolling, adhesion, and emigration in postcapillary venules at the site of injection. LPS given systemically dramatically dropped circulating leukocyte counts and increased neutrophils in the lung. However, the drop in circulating leukocytes was not associated with leukocyte sequestration to the site of injection (peritoneum) nor to peripheral microvessels in muscles. Unlike LPS, various preparations of LTA had no systemic and very minor local effect on leukocyte- Downloaded from endothelial interactions, even at high doses and for prolonged duration. LPS, but not LTA, potently activated human endothelium to recruit leukocytes under flow conditions in vitro. Endothelial adhesion molecule expression was also increased extensively with LPS, but not LTA. Interestingly, systemic administration of live S. aureus induced leukocyte-endothelial cell responses similar to LPS. PepG was able to induce leukocyte-endothelial interactions in muscle and peritoneum, but had no effect systemically (no increase in neutrophils in lungs and no decrease in circulating neutrophil counts). These results demonstrate that: 1) LPS has potent, but divergent local and systemic effects on leukocyte-endothelial interactions; 2) S. aureus can induce a systemic response http://www.jimmunol.org/ similar to LPS, but this response is unlikely to be due to LTA, but more likely to be mediated in part by PepG. The Journal of Immunology, 2002, 168: 4650–4658.

key feature of inflammation is the sequestration of leu- tol-linked protein, and an associate signal transducer Toll-like re- kocytes from the circulation to the endothelium, result- ceptor 4 (TLR4)3 (10, 11). Therefore, TLR4 is a critical receptor A ing in leukocyte emigration into the surrounding tissue. involved in the LPS detection system (12) that confers responsive- Leukocyte recruitment occurs in a sequential multistep fashion (1, ness to circulating and tissue leukocytes as well as endothelium ␣

2). Endothelium activated, for example, with LPS or TNF- ex- (13). By contrast, lipoteichoic acid (LTA), the Gram-positive cell by guest on September 25, 2021 presses the adhesion molecules that mediate tethering, rolling, and membrane equivalent to LPS, does not appear to activate TLR4, adhesion. Therefore, a localized response to LPS or infection will but rather functions through a second TLR, namely TLR2, to ac- induce brisk recruitment of leukocytes without undue damage to tivate immune responses. Nevertheless, in macrophages or cell line the surrounding tissue or organs. In contrast, an inappropriate in- equivalents, the in vitro work has suggested very similar mecha- flammatory response such as systemic will cause leukocytes nisms of action for the two . Both LPS and LTA act by to localize to the lung, where they become physically trapped and binding a TLR (14Ð18), activating various tyrosine kinases, and are not able to recirculate (3Ð6). Indeed, a key feature in human translocating the transcription factor NF-␬B (19), resulting in the septic shock from Gram-negative or Gram-positive is in- production and release of numerous proinflammatory mediators, appropriate leukocyte recruitment predominantly in lung with a including TNF-␣, IL-6, IL-1␤, IL-12, and NO (20Ð24). However, tendency for impaired lung function (7). two very recent in vitro studies reported some significant differ- Gram-negative septic shock is believed to result from the effects ences in mediator release in response to TLR2 vs TLR4 activation of LPS (8, 9). Recent evidence has established a model for LPS (25, 26). Although in vivo studies have suggested that both LPS recognition that involves both CD14, a glycosylphosphatidylinosi- and LTA can induce the same end result, namely shock (8, 27, 28), at least for leukocyte-endothelial cell interactions LPS and LTA may elicit very different responses. For example, TLR4, the re- Immunology Research Group, *Departments of and Infectious Disease, †Physiology and Biophysics, and ‡Biochemistry and Molecular Biology, Faculty of ceptor for LPS, was abundant on endothelium, whereas TLR2, the Medicine, University of Calgary, Alberta, Canada; and ¤Biochemical Pharmacology, receptor for LTA, could not be detected (13). Since the endothe- University of Konstanz, Konstanz, Germany lium affects leukocyte trafficking by expressing adhesion mole- Received for publication October 22, 2001. Accepted for publication February 26, 2002. cules in lungs as well as peripheral microvasculature, LPS vs LTA responses may be quite different. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance We have investigated leukocyte recruitment in response to local with 18 U.S.C. Section 1734 solely to indicate this fact. and systemic LPS or LTA by directly visualizing leukocyte be- 1 This work was supported by a group grant from the Canadian Institutes of Health havior in vivo, further characterizing the direct effects of each Research and the Alberta Heritage Foundation for Medical Research (AHFMR). P.K. and S.M.R. are Canada Research chairs and P.K. is an AHFMR scientist. M.H. and S.M.R. are AHFMR senior scholars. G.A. is an AHFMR postdoctoral fellow. C.J.H. and B.G.Y. are supported by AHFMR studentships. 2 Address correspondence and reprint requests to Dr. Paul Kubes, Health Sciences 3 Abbreviations used in this paper: TLR, Toll-like receptor; HDMEC, human dermal Center, 3330 Hospital Drive NW, Calgary, Alberta, Canada, T2N 4N1. E-mail ad- microvascular endothelial cell; LTA, lipoteichoic acid; MAPK, mitogen-activated dress: [email protected] protein kinase; MPO, myeloperoxidase; PepG, peptidoglycan.

Copyright © 2002 by The American Association of Immunologists 0022-1767/02/$02.00 The Journal of Immunology 4651 on various microvascular beds. Our findings demon- phil) infiltration into the lung (3, 30). The samples were stored at Ϫ20¡C strate that LTA and LPS induce profoundly different responses in for no more than 1 wk before the MPO assay was performed. MPO activity terms of leukocyte recruitment. Whereas LPS elicited profound was determined using an assay described previously (31), but with the volumes of each reagent modified for use in 96-well microtiter plates. leukocyte recruitment, LTA had no noticeable effects in vivo or in Change in absorbance at 450 nm over a 90-s period was determined using a human endothelial-leukocyte assay in vitro. Our results provide a kinetic microplate reader (Molecular Devices, Sunnyvale, CA). direct evidence that the in vitro macrophage activity of LTA is not predictive of its physiological effects upon leukocyte-endothelial Circulating and peritoneal neutrophil counts cell interactions. At the end of each experiment, whole blood was collected via cardiac puncture. Total leukocyte counts were performed, using a Bright-line he- Materials and Methods mocytometer (Hausser Scientific, Horsham, PA). Mice were sacrificed, and Reagents the peritoneal cavity was washed with 10 ml HBSS. The cells were sedi- mented by centrifugation at 260 ϫ g for 5 min. The pellet was resuspended All reagents for tissue culture, unless specified, were purchased from Life in DMEM containing 5% FCS. The total number of cells in the lavage fluid Technologies (Gaithersburg, MD). LTA from Staphylococcus aureus (Sig- was counted using a hemocytometer. ma-Aldrich, Oakville, Ontario, Canada) was reconstituted in sterile PBS, filtered, and stored at Ϫ20¡C. Highly purified LTA, which was Ͼ99% pure Quantitation of expression of P-selectin and contained Ͻ6 pg LPS/mg LTA, as assessed by Limulus amebocyte assay, was isolated using a novel isolation procedure (24). Peptidoglycan Expression of the adhesion molecules P-selectin was quantified using a (PepG) from S. aureus (Fluka, Oakville, Ontario, Canada) was reconsti- modified dual-radiolabeled Ab technique (32, 33). The Abs RB40.34 tuted in sterile PBS and sonicated for 1 h before use. Smooth LPS from (against P-selectin) and A110-1 (a rat IgG1, ␭ isotype standard) were la- 125 131 E. coli was provided by S. Goyert (North Shore University Hospital/New beled with either I (RB40.34) or I (A110-1) using the iodogen Downloaded from York University School of Medicine, Manhasset, NY.). rTNF-␣ was pur- method, as previously described (32, 33). A110-1 was used to detect non- chased from R&D Systems (Minneapolis, MN). specific binding in the murine system. To study P-selectin, animals were injected i.v. with a mixture of 10 ␮g In vivo experiments 125I-anti-labeled P-selectin (RB40.34), and a variable dose of 131I-labeled A110-1. The Abs were allowed to circulate for 5 min, then the animals Male BALB/c mice (The Jackson Laboratory, Bar Harbor, ME) weighing were heparinized. A blood sample was obtained from a carotid artery cath- between 20 and 25 g were used at 6Ð10 wk of age and were housed in a eter, then the mice were exsanguinated by blood withdrawal through the double barrier pathogen-free facility. One postcapillary venule was visu- carotid artery catheter and simultaneous i.v. infusion with bicarbonate- http://www.jimmunol.org/ alized per mouse; therefore, “n” refers to the total number of mice. For ␮ buffered saline. The lung, heart, pancreas, mesentery, small intestine, co- consistency, we chose unbranched vessels that were 25Ð35 m in diameter, lon, skin, and muscle were harvested and weighed. Both 131I and 125I consistent with postcapillary venules. activities were measured in plasma and tissue samples. Intravital microscopy P-selectin expression was calculated per gram of tissue, by subtracting the accumulated activity of the nonbinding Ab (131I-labeled A110-1) from The mouse cremaster preparation, which was approved by the animal care the accumulated activity of the binding Ab (125I-labeled RB40.34). Data committee, University of Calgary, was used to study the behavior of leu- for P-selectin were represented as the percentage of the injected dose of Ab kocytes in the microcirculation (29). Mice were anesthetized by i.p. injec- per gram of tissue. It has been demonstrated previously that this approach tion of a mixture of xylazine hydrochloride (10 mg/kg; MTC Pharmaceu- provides reliable quantitative values of adhesion molecule expression, and

ticals, Cambridge, Ontario, Canada) and ketamine hydrochloride (200 that radiolabeled binding Ab can be displaced specifically with sufficient by guest on September 25, 2021 mg/kg; Rogar/STB, London, Ontario, Canada). The jugular vein was can- amounts of unlabeled Ab. The technique is sufficiently sensitive that very nulated and used to administer additional anesthetic. The cremaster muscle small, basal levels of P-selectin can be detected in wild-type mice relative was dissected free of tissues and exteriorized onto an optically clear view- to P-selectin-deficient mice, in which values are zero (32). ing pedestal. The muscle was cut longitudinally with a cautery and held flat against the pedestal by attaching silk sutures to the corners of the tissue. Experimental protocol The muscle was then superfused with bicarbonate-buffered saline. An intravital microscope (Axioskop; Carl Zeiss Canada, Don Mills, On- To determine the in vivo effect of LTA, PepG, and LPS, we examined the tario, Canada) with a ϫ25 objective lens (Wetzlar L25/0.35; E. Leitz, Mu- local and systemic responses in BALB/c mice. Our initial experiments nich, Germany) and a ϫ10 eyepiece was used to examine the cremasteric were performed using LTA, PepG, or LPS concentrations that were similar microcirculation. A video camera (Panasonic 5100 HS, Osaka, Japan) was to the concentrations of LPS used in previous studies. used to project the images onto a monitor, and the images were recorded for playback analysis using a videocassette recorder. Single unbranched Local LTA or LPS administration cremasteric venules (25Ð40 ␮m in diameter) were selected, and to mini- The local response to LTA was examined by injecting varying concentra- mize variability, the same section of cremasteric venule was observed ␮ throughout the experiment. The number of rolling and adherent leukocytes tions of LTA in 200 l sterile saline s.c. beneath the scrotal skin using a was determined off-line during video playback analysis. Rolling leukocytes 30-gauge needle. The animals were prepared for intravital microscopy at were defined as those cells moving at a velocity less than that of erythro- 3.5 or 23.5 h after LTA administration, and the microvasculature was ob- cytes within a given vessel. Leukocyte rolling velocity was determined by served for 60 min. At 4.5 or 24.5 h, the animals were sacrificed, the lungs measuring the time required for a leukocyte to roll along a 100-␮m length were harvested for MPO assay, and peripheral blood was drawn for cir- culating leukocyte counts. Preliminary experiments with LPS demon- of venule. Rolling velocity was determined for 20 leukocytes at each time ␮ interval. Leukocytes were considered adherent to the venular endothelium strated a dramatic effect of 0.05 g/kg at4honthelocal microvasculature; if they remained stationary for 30 s or longer. Leukocyte emigration was therefore, our initial experiments with LTA were performed under the same defined as the number of extravascular leukocytes per microscopic field of conditions. Increased concentrations and prolonged times were subse- view (ϫ25 objective lens), and was determined by averaging the data de- quently examined. rived from four to five fields adjacent to postcapillary venules. Venular diameter (Dv) was measured on-line using a video caliper (Microcircula- Systemic LTA, PepG, LPS, or live S. aureus administration tion Research Institute, Texas A&M University, College Station, TX). The systemic response was examined by injecting 500 ␮l of varying con- Centerline RBC velocity (VRBC) was also measured on-line using an op- centrations of LTA, PepG, or LPS i.p. for 4 and/or 24 h. The microvas- tical Doppler velocimeter (Microcirculation Research Institute), and mean culature was directly observed at 3.5 or 23.5 h for 60 min. At the end of the RBC velocity (Vmean) was determined as VRBC/1.6. Venular wall shear rate ␥ ␥ ϭ intravital observations, the lungs from each animal were immediately har- ( ) was calculated based on the Newtonian definition: 8(Vmean/Dv). vested and frozen at Ϫ70¡C, peripheral blood was drawn, and a peritoneal Determination of tissue myeloperoxidase activity lavage was performed. The initial concentration of 500 ␮g/kg for4hwas chosen to make a direct comparison between LTA and LPS. Subsequent At the end of each experiment, samples of the lung were weighed, frozen experiments with LTA determined the effects of increasing concentrations on dry ice, and processed for determination of myeloperoxidase (MPO) and longer response times. S. aureus ATCC strain 25923 was grown on activity. MPO is an enzyme found in cells of myeloid origin, and has been blood agar plates (Difco, Detroit, MI) for 18 h, harvested, and washed used extensively as a biochemical marker of granulocyte (mainly neutro- twice with saline. Mice were injected i.p. with 1010 CFU/g body weight. 4652 LPS, BUT NOT LTA, POTENTLY STIMULATES INFLAMMATION IN VIVO

In vitro experiments As in vivo studies revealed very minor leukocyte-endothelium response to LTA, we performed in vitro studies: 1) in macrophage cell lines to deter- mine whether our LTA had biological activity, and 2) in a human endo- thelial-leukocyte system to ensure that the lack of biological activity of LTA was not restricted to mice. Activation of macrophage cell lines THP-1 (American Type Culture Collection, Manassas, VA) and RAW 264.7 cells were maintained in RPMI 1640 supplemented with 10% heat- inactivated FBS, 1% penicillin/streptomycin, 1 mM sodium pyruvate at

37¡Cina5%CO2 atmosphere. To induce the differentiation of the THP-1 cells to the more mature macrophage-like phenotype, cells were suspended to a cell density of 5 ϫ 105 cells/ml in fresh medium containing 5 ϫ 10Ϫ5 mM 1,25-dihydroxyvitamin D3 and incubated for 72 h (34). Both cell lines were stimulated with either LPS (100 ng/ml) or LTA (100 ng/ml) for the indicated times. Western blots were performed on cell-free lysates (35) with the phospho-specific mitogen-activated protein kinase (MAPK) Ab (Promega, Madison, WI), phospho-specific p38 Ab (New England Biolabs, Beverly, MA), or the total p38 Ab (Santa Cruz Biotechnology, Santa Cruz, CA) as a loading control, according to manufacturer’s specification. Downloaded from Endothelial cell isolation Microvascular endothelium was isolated using a previously established protocol (36). Discarded neonatal human foreskin was cut into 2- to 3-mm2 pieces and digested with 0.5 mg/ml collagenase type A1 (Boehringer Mannheim Biochemicals, Indianapolis, IN) in supplemented M199 for 16 h at 4¡C. Each segment of skin is placed keratinized side down, and mi- crovessels were obtained by compressing the tissue with a spatula. The http://www.jimmunol.org/ collected endothelial cells were cultured in EBM-2 with supplements (Clo- netics, San Diego, CA) on 35-mm gelatin-coated tissue culture dishes. Human dermal microvascular endothelial cell (HDMEC) was grown to confluence for each flow chamber experiment, and cells were used before the seventh passage. Adhesion molecule expression was previously shown to be stable up to and including the seventh passage. This protocol was approved by the Ethics Committee of the University of Calgary. Leukocyte recruitment under flow conditions FIGURE 1. Effect of local LTA and LPS on leukocyte kinetics in crem- Leukocyte recruitment was determined using a parallel plate flow chamber asteric postcapillary venules. Mice received LTA (0.05, 0.5, or 5 ␮g/kg) or by guest on September 25, 2021 (36). All experiments were performed with heparinized whole blood ob- LPS (0.05 ␮g/kg) intrascrotally, and were observed using intravital mi- tained from healthy human donors. The blood was diluted (10%) in sterile croscopy at 4 or 24 h postinjection. A, Leukocyte rolling flux; B, leukocyte HBSS at 37¡C. The human blood was drawn through the flow chamber and over the HDMEC at a rate of 2 dynes/cm2 using an infusion pump. Ex- adhesion; C, leukocyte emigration. Data are expressed as the arithmetic p Ͻ 0.001 compared with ,ءءء .periments were recorded and analyzed off-line. Leukocyte rolling flux was mean Ϯ SEM from at least three mice determined by counting the number of leukocytes that roll past a given control groups. plane that is perpendicular to the direction of the flow. Leukocytes that stayed stationary for at least 10 s were counted as adherent cells. For each condition, at least six fields were visualized. dothelial cell interactions was observed. Local LPS administration Statistical analysis at these concentrations causes stasis of the local microcirculation (P. Kubes, unpublished observation). No vascular changes, includ- Data are shown as mean Ϯ SEM. The mean for intravital experiments was ing decreased blood flow or a change in vessel diameter, were determined by combining the data from three time points (3.5, 4, and 4.5 h postinjection) for each vessel. All data were analyzed using one-way noted with LTA (data not shown). As the effect of local LTA ANOVA, and Bonferroni correction was applied where multiple compar- administration may not follow the same time course as local LPS isons were necessary. administration, we examined the response 24 h after the initial administration of LTA. LTA did not have any effect on leukocyte Results rolling (Fig. 1A), adhesion (Fig. 1B), or emigration (Fig. 1C)at Local LPS, but not LTA, causes profound leukocyte-endothelial 24 h. cell interactions A recent report (24) demonstrated that a highly active form of Under normal conditions, baseline leukocyte rolling flux within a LTA could be purified using butanol instead of phenol. The local postcapillary venule was approximately 75 cells/min (Fig. 1A). effect of this highly active form of LTA was investigated. Highly None of these cells adhered (Fig. 1B) or emigrated out of the purified LTA did not have any local effect on leukocyte rolling, microvessels (Fig. 1C). Four hours after local administration of adhesion, or emigration (Table I). LPS, the rolling flux ranged between 100 and 200 cells/min (Fig. 1A). Leukocyte adhesion was apparent, with an average of 12.5 Systemic LPS, but not LTA, causes profound changes in the cells/100 ␮m (Fig. 1B). Additionally, an average of 20 cells em- microcirculation igrated into the surrounding tissue (Fig. 1C). In contrast, LTA did Although the local administration of LPS had no apparent systemic not have any effect on leukocyte rolling, adhesion, or emigration effects (e.g., no change in circulating leukocyte counts), systemic (Fig. 1, AÐC). It was conceivable that the effective dose of LTA administration of LPS induced a striking decline in peripheral cir- may be higher than LPS, so we decided to increase the dose by 10- culating leukocytes (Fig. 2A). Concurrent with a reduction in cir- and 100-fold (0.5 and 5 ␮g/kg) (Fig. 1). Even under these ultra- culating leukocytes was a dramatic increase of neutrophil accu- pathological concentrations of LTA, no increase in leukocyte-en- mulation in the lung (Fig. 2B). Interestingly, addition of LPS i.p. The Journal of Immunology 4653

Table I. Effect of local and systemic administration of highly purified LTA on leukocyte kineticsa

Leukocyte Leukocyte Leukocyte Emigration Circulating Peritoneal Rolling Flux Adhesion (cells/field of Leukocytes Lung MPO Leukocytes (cells/min) (cells/100 ␮m) view) (ϫ106/ml) (U/mg tissue) (ϫ106/ml)

Local control 75.0 Ϯ 9.0 0.5 Ϯ 0.5 0 Ϯ 0 Local LTA (0.05 ␮g/kg) 59.1 Ϯ 18.27 0.33 Ϯ 0.33 0.5 Ϯ 0.5 Systemic control 70.37 Ϯ 7.12 0.125 Ϯ 0.125 0.375 Ϯ 0.183 7.2 Ϯ 0.6 5.66 Ϯ 0.95 1.47 Ϯ 0.16 Systemic LTA (500 ␮g/kg) 73.83 Ϯ 14.79 2.3 Ϯ 0.66 1.66 Ϯ 0.33 5.96 Ϯ 0.36 5.79 Ϯ 0.83 1.35 Ϯ 0.22

a LTA was administered i.p. at 0.05 or 500 ␮g/kg for 4 h. Leukocyte kinetics were directly determined by observing the microvasculature of the cremaster muscle under an intravital microscope. Harvesting lung tissue and assaying for MPO determined the level of neutrophil sequestration to the lung. Leukocyte counts were determined under light microscopy from blood drawn by cardiac puncture or peritoneal lavage. No significant differences were found between any of the groups. Each parameter represents n ϭ 3. at concentrations that had systemic effects did not recruit any leu- in neutrophil sequestration to the lung within the same 4-h time kocytes to the site of injection, as there was no change in peritoneal frame (Fig. 2B). Moreover, administration of LTA for 24 h did not leukocyte numbers (Fig. 2C). In direct contrast, LTA did not cause have any effect on the circulating leukocyte counts, lung MPO a drop in the circulating leukocyte counts (Fig. 2A) nor an increase values, or peritoneal leukocyte counts (Fig. 2, AÐC). We examined

systemic LTA at 10 times (5,000 ␮g/kg) and 100 times (50,000 Downloaded from ␮g/kg) the LPS dose. High doses of LTA did not have any effect on the number of peripheral circulating leukocytes, neutrophil se- questration to the lung, or peritoneal sequestration (Fig. 2, AÐC). We administered the highly purified LTA systemically, but did not observe any effect on leukocyte rolling, adhesion, or emigration

(Table I). http://www.jimmunol.org/

Systemic LPS prevents leukocyte-endothelial cell interactions in the periphery The decrease in circulating leukocyte counts may have been re- sponsible for the lack of recruitment into the peritoneum. To fur- ther explore leukocyte-endothelial cell interactions in the periph- ery following systemic LPS administration, we exteriorized the cremaster muscle. Remarkably, the number of white blood cells rolling through the vessels dropped from 50Ð100 cells/min in the by guest on September 25, 2021 control group to less than 5 cells/min in the LPS group (Fig. 3A). Due to the 95% reduction in rolling cells, very few leukocytes adhered (Fig. 3B) and emigrated (Fig. 3C) out of the vessels com- pared with local LPS administration and consistent with the ob- servation in the peritoneum. Fig. 3 highlights that neither LTA (500 ␮g/kg) nor concentrations of LTA 10- or 100-fold higher lowered the number of rolling leukocytes (Fig. 3A), but did have a minor effect on leukocyte adhesion (Fig. 3B), which did not result in emigration (Fig. 3C). Twenty-four-hour administration of LTA did not have any effects on leukocyte-endothelial cell interactions (Fig. 3). Therefore, whereas systemic LPS caused profound changes in leukocyte trafficking, systemic LTA had very minor effect. Fig. 4 summarizes quantitative P-selectin expression in various murine vascular beds. All tissues had minimal basal P-selectin expression and responded very significantly to LPS. The pulmo- nary, stomach, and cremaster muscle microvasculature revealed minimal P-selectin expression (Fig. 4, AÐC) following LTA ad- ministration. Similar lack of effect was noted in heart, other parts of the gastrointestinal tract (small and large intestine), skin, and spleen (data not shown). By contrast, the mesentery (Fig. 4D) and pancreas (data not shown) demonstrated some significant increase FIGURE 2. The effect of systemic LTA or LPS on different microcir- in P-selectin in response to LTA, but the values were still much culations. Mice received LTA (500, 5,000, or 50,000 ␮g/kg) or LPS (500 lower than with LPS. ␮g/kg) i.p. for 4 or 24 h, at which time they were sacrificed and different procedures were performed. A, A cardiac puncture was performed to de- LTA does not induce leukocyte-endothelial cell interactions in a termine the number of circulating leukocytes; B, lung tissue was harvested human system to assess the level of MPO; and C, a peritoneal lavage was done to assess leukocyte numbers. Data are expressed as the arithmetic mean Ϯ SEM To ensure that these observations were not related to species re- ِ ﺳ 4654 LPS, BUT NOT LTA, POTENTLY STIMULATES INFLAMMATION IN VIVO

FIGURE 4. P-selectin expression in lung (A), stomach (B), cremaster muscle (C), and mesentery (D) under basal control conditions or in re- -p Ͻ 0.001 com ,ءءء .sponse to4hofLPSorLTAsystemic stimulation pared with control groups; % I.D., the percentage of injected dose of Ab. Downloaded from

6B). In addition, 100 ng/ml of both LPS and LTA induced the phosphorylation of MAPK (p42/p44, also known as extracellular signal-related kinase 1/2) (Fig. 6C). Finally, message for various chemokines was detected with both lipoproteins (data not shown). To eliminate the possibility that LTA only activates human http://www.jimmunol.org/ THP-1 cells, we also tested the murine macrophage cell line RAW 264.7. These cells were stimulated with the same concentrations of LTA or LPS, and again induction of MAPK was determined using Western blot analysis. Both LPS and LTA induced the activation of MAPK (p42/p44) as well as another member of the MAPK family, p38 (Fig. 6D). Interestingly, in both cell lines, there was a FIGURE 3. Effect of systemic LTA or LPS on leukocyte kinetics. Mice noticeable difference between LPS and LTA stimulation in terms were treated with LTA (500, 5,000, or 50,000 ␮g/kg) or LPS at 500 ␮g/kg of amplitude and timing. LPS induced a more rapid and robust and observed using intravital microscopy at 4 or 24 h postinjection. A, by guest on September 25, 2021 Leukocyte rolling flux; B, leukocyte adhesion; C, leukocyte emigration. Data are expressed as the arithmetic mean Ϯ SEM from at least three mice p Ͻ 0.001 ,ءءء .p Ͻ 0.05 compared with control groups ,ء .in each group compared with control groups. human endothelium). Perfusion of leukocytes across unstimulated endothelium caused no rolling or adhesion. Addition of LPS for 4 h caused a profound increase in rolling and adhesion that was comparable with values observed with our most potent inducer (TNF-␣) of leukocyte-endothelium interactions in human systems (Fig. 5). In direct contrast, LTA at similar concentrations to LPS caused no leukocyte rolling or adhesion (Fig. 5). Twenty-four-hour incubations with LTA had no effect on leukocyte rolling and ad- hesion (data not shown). The concentration of LTA was examined at 0.1, 1, 10, and 100 ␮g/ml. None of the LTA concentrations induced leukocyte rolling and adhesion. The highly purified form of LTA failed to induce leukocyte-endothelial cell interactions (data not shown). Each concentration was tested at least four times. These data are consistent with the observations that LPS and TNF-␣ induced E-selectin and VCAM-1 on human endothelium, but LTA failed to induce any response (data not shown).

LPS and LTA both elicit biological responses in murine and human macrophages FIGURE 5. The direct effect of LTA and LPS on HDMEC-mediated To ensure that our LTA was active, we used a monocyte cell line leukocyte recruitment. HDMEC was stimulated with 100 ␮g/ml LTA for THP-1 that is known to respond to LTA. Unstimulated THP-1 cells 4 h. Diluted fresh whole blood was then perfused over the endothelium, did not adhere to plastic, whereas when LPS (100 ng/ml) was and leukocyte-endothelial cell interactions were observed. TNF-␣ and LPS added to the cell suspension, essentially all THP-1 cells firmly were used at 20 ng/ml and 0.1 ␮g/ml, respectively. Data are expressed as adhered. LTA (100 ng/ml) induced an identical proadhesive phe- the arithmetic mean Ϯ SEM of six separate experiments for LTA and three notype (Fig. 6A), but was quantitatively lower than with LPS (Fig. separate experiments for LPS and TNF-␣. The Journal of Immunology 4655 Downloaded from http://www.jimmunol.org/

FIGURE 7. The effects of systemic live S. aureus or PepG on different 10 FIGURE 6. Macrophage cell line activation as a result of LPS or LTA microcirculations. Mice were treated with 10 CFU/g body weight or 500 ␮ treatment. THP-1 cells were treated with either 100 ng/ml LPS or LTA for g/kg PepG for 4 h, at which time they were sacrificed and different pro- the times indicated. A, THP-1 cells were photographed at times 0 and 60 cedures were performed. A, A cardiac puncture was performed to deter- min after LPS or LTA treatment, and B, the adherent cells were quantified. mine the number of circulating leukocytes; B, lung tissue was harvested to by guest on September 25, 2021 C, LPS- or LTA-stimulated cell-free lysates were probed with a phospho- assess the level of MPO; and C, a peritoneal lavage was done to assess specific MAPK Ab. D, Adherent RAW 264.7 cells were stimulated with leukocyte numbers. The mouse microcirculation was observed using in- either 100 ng/ml LPS or LTA, and cell-free lysates were probed with phos- travital microscopy at 4 h. D, Leukocyte rolling flux; E, leukocyte adhe- pho-specific MAPK Ab, phospho-specific p38 Ab, or an Ab that recog- sion; and F, leukocyte emigration. Data are expressed as the arithmetic Ͻ ء Ϯ nized total p38 levels (total p38) as a loading control. mean SEM from three mice for each group. , p 0.05 compared with p Ͻ 0.001 ,ءءء .p Ͻ 0.01 compared with control groups ,ءء .control groups compared with control groups. kinase activation, which was easily detected within 15 min of stim- ulation compared with the weaker activation by LTA, which was first detected at 30 min. This difference in signaling between LPS leukocyte responses to this substance. Surprisingly, our data re- and LTA was also consistent when highly purified LTA was used vealed that some, but not all aspects of Gram-positive or LPS (data not shown). responses were seen with i.p. PepG administration. Although PepG had no effect on circulating leukocyte counts (Fig. 7A) or the num- Gram-positive bacteria induce leukocyte-endothelial cell ber of leukocytes that sequestered into the lung (Fig. 7B), there interactions were some effects on leukocyte behavior in the periphery. A subtle To ensure that Gram-positive bacteria could induce leukocyte-en- increase in peritoneal leukocyte recruitment was noted with PepG dothelial cell interactions, animals received an i.p. injection of live (Fig. 7C). Whereas a very profound drop in rolling cells was seen S. aureus (1010 CFU/g). In 4 h, the bacteria caused a 50% decrease in the peripheral muscle microcirculation with LPS or Gram-pos- in circulating leukocyte counts (Fig. 7A). The MPO assay showed itive bacteria, the PepG did not decrease rolling leukocytes, and in that a significant number of neutrophils had sequestered in the lung fact an increase was noted (Fig. 7D). Moreover, a very significant (Fig. 7B). Interestingly, the numbers of neutrophils sequestered in rise in adhesion (Fig. 7E) and some increase in emigration were the lung were similar between S. aureus- and LPS-treated animals. also noted (Fig. 7F). Addition of LTA to PepG in no way enhanced The smaller drop in circulating leukocyte counts allowed for some the PepG response (data not shown). infiltration into the peritoneum (Fig. 7C). Intravital microscopy revealed that S. aureus induced the characteristic drop in the num- Discussion ber of rolling leukocytes that was seen with LPS, but not LTA Sepsis and septic shock are characterized as widespread inflam- (Fig. 7D). Much like LPS, which essentially eliminated all rolling, matory responses that ultimately lead to failure of multiple organs. only a few leukocytes rolled and adhered with S. aureus (Fig. 7, D Regardless of the organ of septic origin, the lung is generally the and E) with some emigration into the surrounding tissue (Fig. 7F). first to fail, at least in part because of rapid accumulation of neu- Since PepG, an alternative Gram-positive molecule, has previ- trophils in the narrow pulmonary capillaries (7). It is well appre- ously been shown to have some biologic activity, we examined ciated that Gram-negative bacteria can cause sepsis and that LPS, 4656 LPS, BUT NOT LTA, POTENTLY STIMULATES INFLAMMATION IN VIVO a major component of these organisms, is involved. In- alternative explanation is that some LTA preparations have been deed, local exposure of the microcirculation to LPS can result in a demonstrated to contain sufficient LPS to account for cellular ac- decrease in microcirculatory flow and an increase in leukocyte- tivation (44). Although the aforementioned studies that observed endothelial cell interactions (37). The impact of systemic LPS ad- endothelial responses to LTA also included studies that attempted ministration on leukocyte-endothelial cell interaction in peripheral to discount potential LPS-contaminating effects, LPS as a potential microcirculatory beds is less well characterized, although it is well synergistic or priming factor for LTA remains a serious concern. appreciated that as an early event many of the leukocytes infiltrate Consistent with this concern is the fact that the receptor essential the pulmonary vasculature (7). With the increased prevalence of for LTA effects, namely TLR2, was not detected on endothelium Gram-positive sepsis, many laboratories have begun to investigate (13). Presently, the work as a whole suggests that LTA does not the molecular mechanisms leading to activation of the immune directly stimulate endothelium sufficiently in vivo or under shear system by components of these organisms. LTA has received a conditions in vitro to induce leukocyte-endothelial cell interac- tremendous amount of attention as the key cell wall component of tions. Our only caveat at this stage is the small increase in P- Gram-positive bacteria that functions as an immune activator with selectin in mesentery and pancreas, which raises some minor tissue characteristics similar to LPS. The impact of local or systemic specificities. LTA on leukocyte-endothelium interactions in vivo remains Our data also provide some interesting new observations with unknown. respect to leukocyte-endothelial cell responses with systemic LPS. In this study, we directly visualized the activation of the innate A number of investigators have reported that leukocyte recruit- immune system in the microvasculature in response to local or ment is reduced to the primary site of infection in systemic septic systemic LTA or LPS administration in an attempt to establish the response (45). Lack of adhesiveness or responsiveness perhaps due Downloaded from importance of these two molecules as inducers of leukocyte-endo- to shedding of critical adhesion molecules or internalization of thelium interactions in vivo. Our in vivo data reveal very profound chemokine receptors has been proposed as an explanation for the differences between LPS and LTA with respect to leukocyte-en- lack of leukocyte recruitment (46Ð48). Our data suggest another dothelial cell interactions. Whereas local LPS induced a cascade of response to systemic LPS worth noting. Addition of LPS to the events, including leukocyte rolling, firm adhesion, and migration peritoneal cavity at concentrations that had systemic effects caused of leukocytes into the surrounding tissue, LTA had a very limited a very profound drop in circulating leukocytes and dramatic se- http://www.jimmunol.org/ impact on any of these parameters. Neither extremely high con- questration of leukocytes in the lung. Interestingly, this leukocy- centrations of LTA, different preparations of LTA, nor longer topenia in wild-type animals translated into fewer leukocyte-en- times of exposure to LTA could induce leukocyte-endothelial cell dothelial cell interactions in peripheral vasculature and an inability interactions. These data are not consistent with the many studies to recruit leukocytes (adhesion and emigration) into peripheral mi- that have demonstrated that LTA, devoid of other contaminants, crocirculations, including the peritoneum and skeletal muscle. This could stimulate various macrophage cell lines to produce and re- delay in leukocyte recruitment into nonpulmonary tissues in septic lease cytokines and chemokines in vitro (21Ð24, 38). Clearly, our humans has been appreciated for many years and may account for data would suggest that if indeed these cells were activated in vivo, the multiorgan dysfunction associated with septic shock (49). Al- the amounts of cytokine and chemokine were simply not sufficient though LPS has been shown to increase endothelial adhesion mol- by guest on September 25, 2021 to induce leukocyte-endothelial cell interactions. ecule expression in all tissues (32), the preferential recruitment of An alternative explanation is that the Gram-positive and Gram- leukocytes into lungs may not be related to molecular adhesive negative sepsis has very different pathological mechanisms, with events, but rather due to physical trapping of activated (rigid) leu- only the latter recruiting neutrophils to the sites of infection. How- kocytes within narrow architecture of pulmonary capillaries (3, 6, ever, this does not appear to be the case. The inappropriate in- 30), which precedes the adhesion molecule expression in other flammatory immune responses observed with Gram-positive and tissues. In this regard, our PepG data are consistent with the im- Gram-negative bacteria are strikingly similar and are in fact indis- portance of the lung in systemic endotoxemia. PepG did not in- tinguishable clinically (39). Comparing experiments with live S. crease leukocyte trapping in the lung; hence, no reduction in cir- aureus (Gram positive) or E. coli (Gram negative) reveal endpoint culating counts and no decrease in leukocyte rolling in the similarities, including similar cardiovascular dysfunction (40), periphery were observed, allowing for substantial leukocyte adhe- rapid neutrophil infiltration into lungs, and increased cytokine and sion and recruitment. adhesion molecule expression (41). In fact, in our study, admin- In some respects, it is not entirely surprising that LTA is not the istration of S. aureus induced a similar pattern of leukocyte re- activator of leukocyte-endothelial cell interactions. First, LTA rep- sponses, as did LPS, but not LTA, suggesting perhaps that the resents a very small fraction of the outer membrane of Gram- latter molecule does not account for the leukocyte responses ob- positive bacteria, whereas LPS makes up a substantial amount of served in Gram-positive sepsis. Gram-negative outer membrane. Second, Gram-positive bacteria At odds with our results are reports that LTA has been able to are often coated with capsules so that from a teleological stand- induce E-selectin, VCAM-1, and ICAM-1 expression on human point, evolving recognition receptors against a reasonably rare endothelium from both umbilical vein (42) and lung (43), leading outer membrane molecule that is often not seen by the innate de- to leukocyte adhesion. Although it is tempting to speculate that our tection system may not be likely. This of course begs the question data differ due to human vs mouse sensitivity to LTA, when we as to which Gram-positive molecules do activate leukocyte-endo- stimulated human endothelium with LTA, no notable increase in thelial cell interactions? Certainly, there are many potential can- leukocyte rolling and adhesion was observed. The other difference didates. PepG have been postulated as important inducers of the between in vivo and previously published in vitro systems is that immune system. Indeed, our data do suggest that PepG may be the in vitro work was done under static conditions. This is not responsible for the adhesion of leukocytes in peripheral microvas- entirely reflective of the dynamic shear conditions that limit leu- culatures with Gram-positive bacteria. However, the trapping of kocyte-endothelium interactions. Indeed, our in vitro work was leukocytes in the lung and the drop in leukocyte numbers (leading also performed under shear through flow chambers, but when we to fewer rolling leukocytes) clearly require some molecule other arrested flow and allowed the cells to settle onto the endothelium, than PepG or LTA or PepG together with LTA (P. Kubes, M. Ho, leukocyte adhesion was still not observed (data not shown). An and B. G. Yipp, unpublished data). The Journal of Immunology 4657

Many groups have demonstrated dramatic LTA responses via 19. Opitz, B., N. W. Schroder, I. Spreitzer, K. S. Michelsen, C. J. Kirschning, TLR2 in numerous macrophage cell lines. Although the tendency W. Hallatschek, U. Zahringer, T. Hartung, U. B. Gobel, and R. R. Schumann. 2001. Toll-like receptor-2 mediates Treponema glycolipid and lipoteichoic acid- is to assume that the detection systems on macrophage cell lines induced NF-␬B translocation. J. Biol. Chem. 276:22041. and endothelium/neutrophils are the same, our data suggest that 20. Keller, R., W. Fischer, R. Keist, and S. Bassetti. 1992. Macrophage response to only macrophages are activated via LTA and this is not sufficient bacteria: induction of marked secretory and cellular activities by lipoteichoic to induce an inflammatory response similar to that seen with LPS acids. Infect. Immun. 60:3664. 21. Wang, J. E., P. F. Jorgensen, M. Almlof, C. Thiemermann, S. J. Foster, or S. aureus in vivo. Indeed, a recent study has reported ample A. O. Aasen, and R. Solberg. 2000. Peptidoglycan and lipoteichoic acid from TLR4, the LPS receptor, but minimal TLR2, the LTA receptor, on Staphylococcus aureus induce tumor necrosis factor ␣, interleukin 6 (IL-6), and endothelium, suggesting that endothelium may use other mecha- IL-10 production in both T cells and monocytes in a human whole blood model. nisms to detect Gram-positive bacteria. Infect. Immun. 68:3965. 22. Hattor, Y., K. Kasai, K. Akimoto, and C. Thiemermann. 1997. Induction of NO synthesis by lipoteichoic acid from Staphylococcus aureus in J774 macrophages: involvement of a CD14-dependent pathway. Biochim. Biophys. Acta 233:375. Acknowledgments 23. Cleveland, M. G., J. D. Gorham, T. L. Murphy, E. Tuomanen, and K. M. Murphy. 1996. Lipoteichoic acid preparations of Gram-positive bacteria induce interleu- We thank Dr. Tom Louie (University of Calgary) for providing the S. kin-12 through a CD14-dependent pathway. Infect. Immun. 64:1906. aureus; Wally Krulicki for technical assistance; and Dr. Carolyn Lane and 24. Morath, S., A. Geyer, and T. Hartung. 2001. Structure-function relationship of her colleagues at Valley View Family Practice Clinic (Calgary, Alberta, cytokine induction by lipoteichoic acid from Staphylococcus aureus. J. Exp. Med. Canada) for providing the skin specimens. 193:393. 25. Hirschfeld, M., J. J. Weis, V. Toshchakov, C. A. Salkowski, M. J. Cody,

D. C. Ward, N. Qureshi, S. M. Michalek, and S. N. Vogel. 2001. Signaling by Downloaded from Toll-like receptor 2 and 4 agonists results in differential gene expression in mu- References rine macrophages. Infect. Immun. 69:1477. 1. Lawrence, M. B., and T. A. Springer. 1991. Leukocytes roll on a selectin at 26. Jones, B. W., T. K. Means, K. A. Heldwein, M. A. Keen, P. J. Hill, J. T. Belisle, physiologic flow rates: distinction from and prerequisite for adhesion through and M. J. Fenton. 2001. Different Toll-like receptor agonists induce distinct mac- integrins. Cell 65:859. rophage responses. J. Leukocyte Biol. 69:1036. 2. Butcher, E. C. 1991. Leukocyte-endothelial cell recognition: three (or more) steps 27. De Kimpe, S. J., M. Kengatharan, C. Thiemermann, and J. R. Vane. 1995. The to specificity and diversity. Cell 67:1033. cell wall components peptidoglycan and lipoteichoic acid from Staphylococcus 3. Meyrick, B., and K. L. Brigham. 1983. Acute effects of Escherichia coli endo- aureus act in synergy to cause shock and multiple organ failure. Proc. Natl. Acad. http://www.jimmunol.org/ toxin on the pulmonary microcirculation of anesthetized sheep structure:function Sci. USA 92:10359. relationships. Lab. Invest. 48:458. 28. Kengatharan, K. M., S. De Kimpe, C. Robson, S. J. Foster, and C. Thiemermann. 4. Williams, J. H., S. K. Patel, D. Hatakeyama, R. Arian, K. Guo, T. J. Hickey, 1998. Mechanism of Gram-positive shock: identification of peptidoglycan and S. Y. Liao, and T. R. Ulich. 1993. Activated pulmonary vascular neutrophils as lipoteichoic acid moieties essential in the induction of nitric oxide synthase, early mediators of endotoxin-induced lung inflammation. Am. J. Respir. Cell Mol. shock, and multiple organ failure. J. Exp. Med. 188:305. Biol. 8:134. 29. Hickey, M. J., K. A. Sharkey, E. G. Sihota, P. H. Reinhardt, J. D. Macmicking, 5. Worthen, G. S., B. Schwab, III, E. L. Elson, and G. P. Downey. 1989. Mechanics C. Nathan, and P. Kubes. 1997. Inducible nitric oxide synthase-deficient mice of stimulated neutrophils: cell stiffening induces retention in capillaries. Science 245:183. have enhanced leukocyte-endothelium interactions in endotoxemia. FASEB J. 11:955. 6. Erzurum, S. C., G. P. Downey, D. E. Doherty, B. Schwab, III, E. L. Elson, and 30. Haslett, C., G. S. Worthen, P. C. Giclas, D. C. Morrison, J. E. Henson, and G. S. Worthen. 1992. Mechanisms of lipopolysaccharide-induced neutrophil re- by guest on September 25, 2021 tention: relative contributions of adhesive and cellular mechanical properties. P. M. Henson. 1987. The pulmonary vascular sequestration of neutrophils in J. Immunol. 149:154. endotoxemia is initiated by an effect of endotoxin on the neutrophil in the rabbit. Am. Rev. Respir. Dis. 136:9. 7. Welbourn, C. R., and Y. Young. 1992. Endotoxin, septic shock and acute lung injury: neutrophils, macrophages and inflammatory mediators. Br. J. Surg. 79: 31. Krawisz, J. E., P. Sharon, and W. F. Stenson. 1984. Quantitative assay for acute 998. intestinal inflammation based on myeloperoxidase activity: assessment of inflam- 8. Danner, R. L., R. J. Elin, J. M. Hosseini, R. A. Wesley, J. M. Reilly, and mation in rat and hamster models. Gastroenterology 87:1344. J. E. Parillo. 1991. Endotoxemia in human septic shock. Chest 99:169. 32. Eppihimer, M. J., B. Wolitzky, D. C. Anderson, M. A. Labow, and 9. Van Deventer, S. J., H. R. Buller, J. W. ten Cate, A. Sturk, and W. Pauw. 1988. D. N. Granger. 1996. Heterogeneity of expression of E- and P-selectins in vivo. Endotoxaemia: an early predictor of septicaemia in febrile patients. Lancet 1:605. Circ. Res. 79:560. 10. Haziot, A., E. Ferrero, F. Kontgen, N. Hijiya, S. Yamamoto, J. Silver, 33. Henninger, D. D., J. Panes, M. Eppihimer, J. Russell, M. Gerritsen, C. L. Stewart, and S. M. Goyert. 1996. Resistance to endotoxin shock and re- D. C. Anderson, and D. N. Granger. 1997. Cytokine-induced VCAM-1 and duced dissemination of Gram-negative bacteria in CD14-deficient mice. Immu- ICAM-1 expression in different organs of the mouse. J. Immunol. 158:1825. nity 4:407. 34. Schwende, H., E. Fitzke, P. Ambs, and P. Dieter. 1996. Differences in the state 11. Takeuchi, O., K. Hoshino, T. Kawai, H. Sanjo, H. Takada, T. Ogawa, K. Takeda, of differentiation of THP-1 cells induced by phorbol ester and 1,25-dihydroxy- and S. Akira. 1999. Differential roles of TLR2 and TLR4 in recognition of Gram- vitamin D3. J. Leukocyte Biol. 59:555. negative and Gram-positive bacterial cell wall components. Immunity 11:443. 35. Howlett, C. J., S. A. Bisson, M. E. Resek, A. W. Tigley, and S. M. Robbins. 1999. 12. Beutler, B. 2000. Tlr4: central component of the sole mammalian LPS sensor. The proto-oncogene p120Cbl is a downstream substrate of the Hck protein-ty- Curr. Opin. Immunol. 12:20. rosine kinase. Biochim. Biophys. Acta 257:129. 13. Faure, E., O. Equils, P. A. Sieling, L. Thomas, F. X. Zhang, C. J. Kirschning, 36. Yipp, B. G., S. Anand, T. Schollaardt, K. D. Patel, S. Looareesuwan, and M. Ho. N. Polentarutti, M. Muzio, and M. Arditi. 2000. Bacterial lipopolysaccharide 2000. Synergism of multiple adhesion molecules in mediating cytoadherence of ␬ activates NF- B through Toll-like receptor 4 (TLR-4) in cultured human dermal Plasmodium falciparum-infected erythrocytes to microvascular endothelial cells endothelial cells: differential expression of TLR-4 and TLR-2 in endothelial cells. under flow. Blood 96:2292. J. Biol. Chem. 275:11058. 37. Johnston, B., U. M. Walter, A. C. Issekutz, T. B. Issekutz, D. C. Anderson, and 14. Schwandner, R., R. Dziarski, H. Wesche, M. Rothe, and C. J. Kirschning. 1999. P. Kubes. 1997. Differential roles of selectins and the ␣ -integrin in acute, sub- Peptidoglycan- and lipoteichoic acid-induced cell activation is mediated by Toll- 4 acute, and chronic leukocyte recruitment in vivo. J. Immunol. 159:4514. like receptor 2. J. Biol. Chem. 274:17406. 38. Keller, R., W. Fischer, R. Keist, and S. Bassetti. 1992. Macrophage response to 15. Lehner, M. D., S. Morath, K. S. Michelsen, R. R. Schumann, and T. Hartung. bacteria: induction of marked secretory and cellular activities by lipoteichoic 2001. Induction of cross-tolerance by lipopolysaccharide and highly purified li- poteichoic acid via different Toll-like receptors independent of paracrine medi- acids. Infect. Immun. 60:3664. ators. J. Immunol. 166:5161. 39. Balk, R. A., and R. C. Bone. 1989. The septic syndrome: definition and clinical 16. Du, X., A. Poltorak, M. Silva, and B. Beutler. 1999. Analysis of Tlr4-mediated implications. Crit. Care Clin. 5:1. LPS signal transduction in macrophages by mutational modification of the re- 40. Natanson, C., R. L. Danner, R. J. Elin, J. M. Hosseini, K. W. Peart, S. M. Banks, ceptor. Blood Cells Mol. Dis. 25:328. T. J. MacVittie, R. I. Walker, and J. E. Parrillo. 1989. Role of endotoxemia in 17. Yang, R. B., M. R. Mark, A. Gray, A. Huang, M. H. Xie, M. Zhang, A. Goddard, cardiovascular dysfunction and mortality: Escherichia coli and Staphylococcus W. I. Wood, A. L. Gurney, and P. J. Godowski. 1998. Toll-like receptor-2 me- aureus challenges in a canine model of human septic shock. J. Clin. Invest. diates lipopolysaccharide-induced cellular signalling. Nature 395:284. 83:243. 18. Yang, R. B., M. R. Mark, A. L. Gurney, and P. J. Godowski. 1999. Signaling 41. Wakabayashi, G., J. A. Gelfand, W. K. Jung, R. J. Connolly, J. F. Burke, and events induced by lipopolysaccharide-activated Toll-like receptor 2. J. Immunol. C. A. Dinarello. 1991. Staphylococcus epidermidis induces complement activa- 163:639. tion, tumor necrosis factor and interleukin-1, a shock-like state and tissue injury 4658 LPS, BUT NOT LTA, POTENTLY STIMULATES INFLAMMATION IN VIVO

in rabbits without endotoxemia: comparison to Escherichia coli. J. Clin. Invest. of specific receptor-based response and inhibition by lipid A-based antagonists. 87:1925. J. Immunol. 147:3072. 42. Kawamura, N., N. Imanishi, H. Koike, H. Nakahara, L. Phillips, and 47. Khandaker, M. H., L. Xu, R. Rahimpour, G. Mitchell, M. E. DeVries, S. Morooka. 1995. Lipoteichoic acid-induced neutrophil adhesion via E-se- J. G. Pickering, S. K. Singhal, R. D. Feldman, and D. J. Kelvin. 1998. CXCR1 lectin to human umbilical vein endothelial cells (HUVECs). Biochim. Bio- and CXCR2 are rapidly down-modulated by bacterial endotoxin through a unique phys. Acta 217:1208. agonist-independent, tyrosine kinase-dependent mechanism. J. Immunol. 43. Blease, K., Y. Chen, P. G. Hellewell, and A. Burke-Gaffney. 1999. Lipotei- 161:1930. choic acid inhibits lipopolysaccharide-induced adhesion molecule expression and IL-8 release in human lung microvascular endothelial cells. J. Immunol. 48. Lloyd, A. R., A. Biragyn, J. A. Johnston, D. D. Taub, L. Xu, D. Michiel, 163:6139. H. Sprenger, J. J. Oppenheim, and D. J. Kelvin. 1995. Granulocyte-colony 44. Gao, J. J., Q. Xue, E. G. Zuvanich, K. R. Haghi, and D. C. Morrison. 2001. stimulating factor and lipopolysaccharide regulate the expression of inter- Commercial preparations of lipoteichoic acid contain endotoxin that contributes leukin 8 receptors on polymorphonuclear leukocytes. J. Biol. Chem. 270: to activation of mouse macrophages in vitro. Infect. Immun. 69:751. 28188. 45. Wagner, J. G., and R. A. Roth. 1999. Neutrophil migration during endotoxemia. 49. Ahmed, N. A., S. McGill, J. Yee, F. Hu, R. P. Michel, and N. V. Christou. 1999. J. Leukocyte Biol. 66:10. Mechanisms for the diminished neutrophil exudation to secondary inflammatory 46. Lynn, W. A., C. R. Raetz, N. Qureshi, and D. T. Golenbock. 1991. Lipopolysac- sites in infected patients with a systemic inflammatory response (sepsis). Crit. charide-induced stimulation of CD11b/CD18 expression on neutrophils: evidence Care Med. 27:2459. Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021