The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Blease, K.
Right arrow Articles by Burke-Gaffney, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Blease, K.
Right arrow Articles by Burke-Gaffney, A.
The Journal of Immunology, 1999, 163: 6139-6147.
Copyright © 1999 by The American Association of Immunologists

Lipoteichoic Acid inhibits Lipopolysaccharide-Induced Adhesion Molecule Expression and IL-8 Release in Human Lung Microvascular Endothelial Cells1

Kate Blease2,*, Yan Chen{dagger}, Paul G. Hellewell{ddagger} and Anne Burke-Gaffney3,*,{dagger}

* Applied Pharmacology and {dagger} Unit of Critical Care, National Heart and Lung Institute Division, Imperial College School of Medicine, London, United Kingdom; and {ddagger} Section of Vascular Biology, University of Sheffield, Clinical Sciences Centre, Northern General Hospital, Sheffield, United Kingdom


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Cell adhesion molecule expression (CAM) and IL-8 release in lung microvascular endothelium facilitate neutrophil accumulation in the lung. This study investigated the effects of lipoteichoic acid (LTA), a cell wall component of Gram-positive bacteria, alone and with LPS or TNF-{alpha}, on CAM expression and IL-8 release in human lung microvascular endothelial cells (HLMVEC). The concentration-dependent effects of Staphylococcus aureus (S. aureus) LTA (0.3–30 µg/ml) on ICAM-1 and E-selectin expression and IL-8 release were bell shaped. Streptococcus pyogenes (S. pyogenes) LTA had no effect on CAM expression, but caused a concentration-dependent increase in IL-8 release. S. aureus and S. pyogenes LTA (30 µg/ml) abolished LPS-induced CAM expression, and S. aureus LTA reduced LPS-induced IL-8 release. In contrast, the effects of S. aureus LTA with TNF-{alpha} on CAM expression and IL-8 release were additive. Inhibitory effects of LTA were not due to decreased HLMVEC viability, as assessed by ethidium homodimer-1 uptake. Changes in neutrophil adhesion to HLMVEC paralleled changes in CAM expression. Using RT-PCR to assess mRNA levels, S. aureus LTA (3 µg/ml) caused a protein synthesis-dependent reduction (75%) in LPS-induced IL-8 mRNA and decreased the IL-8 mRNA half-life from >6 h with LPS to ~2 h. These results suggest that mechanisms exist to prevent excessive endothelial cell activation in the presence of high concentrations of bacterial products. However, inhibition of HLMVEC CAM expression and IL-8 release ultimately may contribute to decreased neutrophil accumulation, persistence of bacteria in the lung, and increased severity of infection.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Severe bacterial infection is the major cause of sepsis, a syndrome characterized by a widespread inflammatory response that triggers organ damage and ultimately may result in multiple organ failure (1, 2). The involvement of Gram-negative bacteria in sepsis is well established (3), but there is an increasing prevalence of Gram-positive bacteria, alone or in mixed infections, as a cause of sepsis; however, less is known about the mechanisms involved (4, 5, 6). Gram-positive bacteria are now thought to be responsible for one-third to one-half of all cases of sepsis, and with the resurgence of hospital-acquired Gram-positive infections, it is likely that these will be predominant in the future (5, 6). In recent years, there has also been a change in the main origin of sepsis from the abdomen to the lung (6, 7). Regardless of the organ in which sepsis originates, however, the lung is usually the first to fail (8). This may in part result from a rapid accumulation of neutrophils in the narrow lumen of lung capillaries (9). Toxic products released from sequestered neutrophils damage the endothelial cells lining the capillary walls and lead to an increase in permeability, edema formation, and ultimately pulmonary failure (10).

Mechanisms that facilitate neutrophil recruitment to the lung in response to an inflammatory stimulus may contribute to endothelial cell damage and pulmonary failure. Two mechanisms that play a key role in the recruitment of neutrophils to sites of inflammation are the induction of cell adhesion molecules (CAM)4 and the release of chemokines in the vascular endothelium (11, 12, 13, 14). Neutrophils bind E-selectin or ICAM-1 expressed on endothelial cells via sialyl-Lewis X and related carbohydrate structures or the ß2 integrins, CD11a/CD18 (LFA-1) and CD11b/CD18 (Mac-1), respectively, on neutrophils (11, 12, 13, 14). The CXC chemokine, IL-8, acts in concert with CAM to provide a directional cue for neutrophil recruitment and to activate neutrophil function (12). CAM expression and IL-8 release are increased in the lung microvascular endothelium of patients with sepsis, and raised levels of IL-8 and soluble CAM are also detected in the plasma and bronchoaveolar lavage fluid of these patients (15, 16, 17, 18, 19).

LPS, a major cell wall component of Gram-negative organisms (20), triggers the inflammatory responses associated with sepsis, including increased CAM expression and release of IL-8 in endothelial cells (21, 22). Little is known, however, about the endothelial-activating properties of cell wall components of Gram-positive bacteria or how these effects contribute to the pathogenesis of sepsis. The cell walls of Gram-positive bacteria are made up of a layer of peptidoglycan (PepG) embedded with teichoic acids, including lipoteichoic acid (LTA) (4). The effects of LTA and PepG on cells other than endothelial cells are quite well characterized and include induction of IL-8 release and CAM expression in monocytes (23, 24). LTA also induces the release of hepatocyte growth factor from human gingival fibroblasts (25), and PepG, alone and in combination with LTA, triggers cytokine production in monocytes and up-regulates inducible nitric oxide synthase expression in macrophages (26, 27, 28).

The only endothelial-activating property of LTA described to date is CAM induction in human umbilical vein endothelial cells (HUVEC) by Staphylococcus aureus LTA (29), although infection of HUVEC with live S. aureus induces CAM expression and IL-8 release (30, 31). Also, the effects of LTA in combination with LPS or cytokines on endothelial cell function in vitro have not been investigated. In vivo, LPS-treated rats failed to recruit neutrophils to the lung following aerosolized challenge with LTA or Gram-positive bacteria, resulting in the proliferation of bacteria in the lung, which suggests an inhibitory interaction between LPS and Gram-positive bacteria/bacterial components (32, 33). The aims of the present study were to investigate the effects of LTA from S. aureus alone and in combination with LPS or TNF-{alpha} on ICAM-1 and E-selectin expression, neutrophil adhesion, and IL-8 release in human lung microvascular endothelial cells (HLMVEC). In selected studies, the effects of LTA from Streptococcus pyogenes (S. pyogenes) were compared with those of S. aureus LTA. These Gram-positive bacteria were chosen because they are often associated with lung infections and may also trigger sepsis (34, 35).


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture reagents

HLMVEC, prepared by Clonetics (San Diego, CA), were obtained from three donors, as cryopreserved third-passage cultures from TCS Biologicals (Buckingham, U.K.) and used at passages 5–10 . Microvascular endothelial growth medium (EGM-MV) was also obtained from TCS Biologicals.

Cytokines and other reagents

Human recombinant (hr) TNF-{alpha} was obtained from Boehringer Mannheim U.K. (Lewes, U.K.; sp. act., >1 x 108). Hr IL-8 (72 aa) was a gift from Dr. J. White, SmithKline Beecham (King of Prussia, PA). LPS from Escherichia coli (055:B5) and LTA from S. aureus (DSM 20233) and S. pyogenes (IID 698) were purchased from Sigma (Poole, U.K.). The preparations of LTA contained <1 ng of LPS/mg LTA (0.03 ng/ml LPS per 30 µg/ml LTA, maximum concentration used in this study) according to the manufacturer’s quality control. Peptidoglycan from S. aureus was a gift from Dr. C. Thiemermann (William Harvey Research Institute, London, U.K.). Percoll was obtained from Pharmacia Biotech (St. Albans, U.K.), sterile normal saline (0.9%) was obtained from FL (Manufacturing), Fresenius Health Care Group (Basingstoke, U.K.), and Dulbecco’s PBS (with Ca2+/Mg2+) was purchased from Life Technologies (Paisley, Scotland). Calcein-AM and ethidium homodimer-1 (EtH-1) were obtained from Cambridge Bioscience (Cambridge, U.K.). Iodogen radioiodination reagent was purchased from Pierce and Warriner, (Chester, U.K.), and Na125I was obtained from Amersham (Aylesbury, U.K.). All other reagents were purchased from Sigma or BDH Chemicals (Poole, U.K.).

Antibodies

Affinity-isolated goat anti-mouse peroxidase conjugate {gamma}- and light chain specific was purchased from TCS Biologicals, and donkey anti-goat IgG mAb was obtained from Nordic Immunological Laboratories (Tilburg, The Netherlands). The following were obtained as gifts: mouse anti-human ICAM-1 (RR1/1) IgG1 mAb (36) from Dr. R. Rothlein (Boehringer Inglheim Pharmaceuticals, Ridgefield, CT), mouse anti-human IgG1 mAb against E-selectin (BBIG-E1) (37) from Dr. R. Pigott (British Biotech, Oxford, U.K.), and goat anti-human IL-8 antiserum from Dr. H. Showell (Pfizer Central Research, Groton, CT).

Molecular biology reagents

Primers for PCR amplification were synthesized by Life Technologies (Paisley, Scotland). These were as follows: IL-8 sense primer, 5'-TAC TCC AAA CCT TTC CAC CC-3'; IL-8 antisense primer, 5'-AAC TTC TCC ACA ACC CTC TG-3'; ß-actin sense primer, 5'-GGC ACC ACA CCT TCT ACA ATG-3'; and ß-actin antisense primer, 5'-CAG GAA GGA AGG TTG GAA GAG-3'. The following were also obtained from Life Technologies: Moloney murine leukemia virus reverse transcriptase supplied with 5x First Strand buffer and DTT; Taq DNA polymerase supplied with 10x PCR buffer and MgCl; and oligo(dT)12–18 primer. RNasin and dNTP (containing 10 mM dATP, dCTP, dGTP, and dTTP) were obtained from Promega (Southampton, U.K.), and actinomycin D was purchased from Calbiochem Novabiochem (Nottingham, U.K.). All other reagents were obtained from Sigma or BDH Chemicals.

Cell culture

HLMVEC purchased for this study were isolated from peripheral lung tissue (avoiding pleura and large blood vessels) and have been shown to retain a number of properties of endothelial cells, including the production of human factor-VIII related Ag, uptake of acetylated low density lipoprotein, and expression of CD31 (22, 38). HLMVEC were maintained in EGM-MV medium, a modification of MCDB 131, supplemented with 10 ng/ml human recombinant epidermal growth factor, 1 µg/ml hydrocortisone, 5% heat-inactivated FCS, 50 µg/ml gentamicin, 50 ng/ml amphotericin B, bovine brain extract containing 12 µg/ml protein, and 10 µg/ml heparin. The magnitude of basal and stimulated CAM expression or IL-8 release was not significantly altered as HLMVEC were passaged or by hydrocortisone in the culture medium. Confluent cells were subcultured and seeded onto 96-well plates at 3200 cells/well, as described previously, unless otherwise stated (22). Confluent monolayers (4 days in culture) were incubated with stimuli diluted in serum-supplemented complete culture medium for the times indicated in Results.

ELISA for ICAM-1 and E-selectin expression

The ICAM-1 and E-selectin were detected by a specific ELISA method using mouse anti-human ICAM-1 (RR1/1) or E-selectin (1D2) primary mAbs, a peroxidase-linked goat anti-mouse secondary Ab, and the chromophore 2,2'-azinobis-3-ethylbenzthiazoline-6-sulfonic acid as described previously (22, 37). Optical density was determined at 405 nm, and adhesion molecule expression is given as OD405. Also, the OD405 for the theoretical additive value of CAM expression was calculated by addition of LTA-stimulated CAM expression to that induced by LPS or TNF-{alpha} and is the value predicted whether the effects of these agents acting in concert is additive. LPS and TNF-{alpha} have been shown previously to have an additive effect on HLMVEC function (39, 40).

Purification of human peripheral blood neutrophils

Neutrophils (>98% pure) were purified from peripheral blood of normal adult donors by discontinuous plasma-Percoll density gradient centrifugation using a modification of the Haslett method (41) as we have described previously (42). Blood was sampled according to a Royal Brompton Hospital ethics committee-approved protocol. Purified neutrophils were suspended at 5 x 106 cells/ml in Krebs-Ringer phosphate dextrose (KRPD) buffer (4.8 mM KCl, 3.1 mM NaH2PO4, 12.5 mM Na2HPO4, and 5% glucose) and labeled (30 min, 37°C) with a fluorescent dye, calcein-AM (10 µM dissolved in 1% DMSO in KRPD). Cells were washed twice in KRPD and resuspended at 1.25 x 106 cells/ml in KRPD containing 2.5% FCS, 0.93 mM CaCl2, and 1.2 mM MgSO4.

Measurement of neutrophil adhesion to HLMVEC

HLMVEC monolayers were washed three times with PBS (containing Ca2+/Mg2+) to remove stimuli before carrying out the adhesion assay. One hundred microliters of KRPD with Ca2+/Mg2+ were added per well, followed by 100 µl of calcein-AM labeled neutrophils, and the plate was incubated at 37°C for 30 min (22, 43). Fluorescence was measured using a Biolite F1 plate reader (Labtech, Ringmer, East Sussex, U.K.) at an excitation wavelength of 485 ± 25 nm and an emission wavelength of 530 ± 25 nm before and after washing the plate to remove nonadherent cells by two gentle washes with PBS containing 1% horse serum. Percent adhesion was expressed as the fluorescence after washing the plate minus the background fluorescence, divided by the fluorescence before washing the plate minus the background x 100. Theoretical additive values for adhesion were also calculated (see ELISA section).

Measurement of endothelial cell viability

Endothelial cell damage was assessed using EtH-1, a high affinity red fluorescent DNA dye that enters cells through damaged cell membranes, as we have described previously (44). Following stimulation, conditioned medium was removed, and monolayers were incubated with 2.5 µM EtH-1 (dissolved in PBS containing 0.025% DMSO) for 1 h at room temperature. A measure of 100% cell damage was obtained from monolayers treated with saponin detergent (0.1%, 10 min, room temperature). Fluorescence was measured with a Biolite F1 multiwell fluorescent plate reader (excitation at 530 ± 25 nm and emission at 645 ± 40 nm). Background fluorescence was subtracted, and results were expressed as arbitrary units of fluorescence.

RIA detection of IL-8

IL-8 levels in conditioned medium collected from HLMVEC monolayers (pooled triplicate wells) were determined by a RIA method described previously by Au and colleagues (45). Human rIL-8 was iodinated with Na125I using Iodogen radioiodination reagent as described previously (46), and the sp. act. of [125I]IL-8 was 65 ± 4 µCi/µg protein (n = 4). Radiolabeled IL-8 was diluted in 10 mM PBS containing 0.1% NaN3, 0.2% gelatin, and 0.5% protamine sulfate (PBS.GP). Polyclonal goat-anti-human IL-8 anti-serum was diluted 1/2000 in PBS.GP, and donkey anti-goat IgG mAb was diluted 1/20 in PBS.GP. Spun sample buffer, prepared from 0.9% normal saline incubated (1 h, 4°C) with polyethylene glycol (11%), protamine sulfate (0.5%), EDTA (10 mM), and NaN3 (0.1%) and spun at 5420 x g for 20 min, was used to dilute standards.

One hundred microliters of duplicate samples (undiluted) or standards (9.5–10,000 pM) were mixed with 100 µl of spun sample buffer, 50 µl of [125I]human IL-8 (30 fmol/tube), and 50 µl of antiserum (24 h, room temperature). Each sample was then incubated with 25 µl of donkey anti-goat IgG mAb for 16 h at room temperature. Radioactivity associated with the Ab-bound fraction was determined by gamma counting. Nonspecific binding of [125I]IL-8 was <6% of the total binding. IL-8 concentrations were calculated from standards, and results are expressed as picomolar concentrations, percentage of LPS- or TNF-{alpha}-induced IL-8 release, and theoretical additive values, where appropriate.

RNA isolation and semiquantitative RT-PCR

The HLMVEC monolayers, grown in 48-well plates (7, 500 cells/well), were activated, and total cellular RNA was extracted using Tri-Reagent according to the manufacturer’s instructions (Sigma, Poole, UK.). Total RNA was quantified by measuring absorbance at 260 nm (Gene Quant Pharmacia, St. Albans, U.K.), and the integrity of the RNA was analyzed by agarose gel electrophoresis and ethidium bromide staining.

Total RNA (600 ng), was converted to ssRNA by incubation at 65°C for 10 min in the presence of 0.5 µg of oligo(dT)12–18 primer, and the reaction was terminated by placing the samples on ice. RT was conducted in First Strand buffer (39 mM Tris-HCl (pH 8.3), 58 mM KCl, 2.3 mM MgCl, and 11 mM DTT) with 1.1 mM dNTP, 20 U of RNasin, and 200 U of Moloney murine leukemia virus reverse transcriptase. The cDNA was formed following incubation at 37°C for 1 h and at 65°C for 10 min. PCR was performed using 3 µl of cDNA in a final volume of 25 µl of PCR buffer (20 mM Tris-HCl (pH 8.3) with 50 mM KCl). The reaction mixture also contained 0.2 mM dNTPs, 1.5 mM MgCl2, 15 pmol of sense and antisense primers for IL-8 or ß-actin (see above for details), and 1.25 U of Taq DNA polymerase. Samples were amplified in a DNA thermal cycler (Crocodile III, Appligene Oncor, Durham, U.K.) for 24 cycles (IL-8) or 28 cycles (ß-actin) of denaturation at 95°C for 30 s, annealing at 64°C for 1 min, and extension at 72°C for 1 min. The appropriate number of cycles for each primer was selected from the linear portion of the cycle curve. Amplified cDNA was analyzed by 1% agarose gel electrophoresis and ethidium bromide staining. Band densities were measured using a scanning laser densitometer attached to a computer with image analysis software (Adobe Photoshop, Abode Systems, San Jose, CA), which was calibrated before use and gave linear readings over 4 OD units. Results are expressed as a ratio of IL-8/ß-actin band density, and this represents semiquantitative analysis. The identities of the major products were confirmed by sequencing a 158-bp fragment of IL-8 and a 551-bp fragment of ß-actin (Advanced Biotechnology Centre, Charing Cross Hospital, London, U.K.).

Messenger RNA stability analysis

The rate of IL-8 mRNA decay in HLMVEC was determined by mRNA decay analysis. HLMVEC monolayers were activated with LPS and/or LTA for 2 h. RNA synthesis was blocked by addition of actinomycin D (5 µg/ml), and total RNA was isolated at 0, 2, 4, and 6 h after addition. Semiquantitative RT-PCR was then performed for IL-8 and ß-actin mRNA, as described in the previous section, and the decay of mRNA was determined from band density ratios.

Statistics

Results are expressed as the mean ± SEM of n experiments. Statistical analysis was conducted unless otherwise stated using one-way ANOVA followed by Dunnett’s multiple comparison test, which compares all values with a control. Instat GraphPad software was used to perform statistical analysis. Results were deemed significant if p < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Cell adhesion molecule expression and IL-8 release in HLMVEC treated with LTA or PepG

Constitutive ICAM-1, but not E-selectin, expression was detected on resting HLMVEC monolayers (Fig. 1Go, A and B). We have reported previously that maximum E-selectin and ICAM-1 expressions in HLMVEC are detected at 6 and 24 h, respectively, following LPS or TNF-{alpha} stimulation, and these times were therefore used in the present study. The concentration-dependent effect of S. aureus LTA (0.3–30 µg/ml) on ICAM-1 (24 h) or E-selectin expression (6 h) was bell shaped (Fig. 1Go, A and B). Significant increases in ICAM-1 were detected with S. aureus LTA at 1 (p < 0.05), 3 or 10 µg/ml (p < 0.01), but not 0.3 or 30 µg/ml (Fig. 1GoA), and significant induction (p < 0.01) of E-selectin was also detected at these concentrations (Fig. 1GoB). In contrast, S. pyogenes LTA or S. aureus PepG did not induce significantly ICAM-1 or E-selectin expression (0.3–30 µg/ml; Fig. 1Go, A and B). Furthermore, PepG (0.3–30 µg/ml) did not alter S. aureus LTA-induced ICAM-1 expression (data not shown).



View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 1. Effects of LTA and PepG from S. aureus and LTA from S. pyogenes on CAM expression and IL-8 release in HLMVEC monolayers. The HLMVEC were treated for 24 h with culture medium, 0.3–30 µg/ml S. aureus LTA (•), S. pyogenes LTA ({circ}), and S. aureus PepG ({square}) for 6 h (B) or 24 h (A and C). ICAM-1 (A) and E-selectin (B) expressions were determined by ELISA, and IL-8 release (C) was determined by RIA. Results are shown as the mean ± SEM of five experiments. To ensure clarity, statistics and SEM are given in Results.

 
IL-8 release from resting HLMVEC monolayers was 39 ± 8 pM. Increasing concentrations of S. aureus LTA (0.3–30 µg/ml) had a bell-shaped effect on IL-8 release at 24 h (Fig. 1GoC). Significant increases were measured with LTA at 1 (p < 0.05), 3, or 10 (p < 0.01), but not 0.3 or 30 µg/ml (Fig. 1GoC). S. aureus PepG (0.3–30 µg/ml) did not alter basal (Fig. 1GoC) or S. aureus LTA-induced IL-8 release (data not shown). In contrast to the bell-shaped effect seen with S. aureus LTA, the effect of increasing concentrations of S. pyogenes LTA on IL-8 release was sigmoid shaped (Fig. 1GoC). Significant increases (p < 0.01) were detected with 10 and 30 µg/ml S. pyogenes LTA (Fig. 1GoC).

LTA inhibits LPS-induced ICAM and E-selectin expression

Incubation of HLMVEC monolayers with LPS (0.01 or 0.1 µg/ml) significantly induced ICAM-1 and E-selectin expression (Figs. 2Go and 3), and these responses were deemed submaximal and maximal, respectively, because 1 µg/ml LPS did not further increase CAM expression (data not shown). LPS-induced ICAM-1 or E-selectin expression was not increased further in the presence of S. aureus LTA, despite significant induction with 1, 3, and 10 µg/ml LTA alone (Figs. 1Go, A and B; 2; and 3). Expression of ICAM-1 and E-selectin in HLMVEC treated with LPS in combination with LTA was also less than the predicted additive effect of these stimuli (Figs. 2Go and 3Go). We have described these predicted effects as theoretical additive values, and they are within the range of maximal CAM induction detected previously in HLMVEC (22). LTA at 30 µg/ml abolished LPS-induced ICAM-1 and E-selectin expression (Figs. 2Go and 3Go). A lower concentration of LTA (10 µg/ml) also significantly inhibited (p < 0.01) E-selectin induced by 0.1 µg/ml LPS, but had no effect on expression induced by 0.01 µg/ml LPS (Fig. 3Go). Finally, although S. pyogenes LTA (0.3–30 µg/ml) had no effect alone on either ICAM-1 or E-selectin expression, 10 µg/ml significantly inhibited (p < 0.01) E-selectin induced by 0.1 µg/ml LPS, and 30 µg/ml abolished LPS-induced expression of these CAM (Fig. 4Go).



View larger version (27K):
[in this window]
[in a new window]
  		FIGURE 2. The S. aureus LTA inhibits LPS-induced ICAM-1 expression in HLMVEC monolayers. HLMVEC were treated for 24 h with culture medium ({square}; M), LPS ({blacksquare}) at 0.01 µg/ml (A) or 0.1 µg/ml (B), and increasing concentrations of S. aureus LTA (0.3–30 µg/ml) alone () or in combination with LPS (•), and ICAM-1 was determined by ELISA. Results show the mean ± SEM of four experiments. The theoretical additive values ({circ}) were determined by addition of ICAM-1 expression induced by LPS with LTA-induced expression. *, p < 0.05; **, p < 0.01 (significant induction of ICAM-1 expression compared with treatment with culture medium). #, p < 0.05; ##, p < 0.01 (significant inhibition of ICAM-1 expression compared with LPS-induced expression). To ensure clarity, statistical significance for the effects of LTA alone is given in the appropriate Results section.

 


View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 3. The S. aureus LTA inhibits LPS-induced E-selectin expression in HLMVEC monolayers. HLMVEC were treated for 6 h with culture medium ({square}; M), LPS ({blacksquare}) at 0.01 µg/ml (A) or 0.1 µg/ml (B), and increasing concentrations of S. aureus LTA (0.3–30 µg/ml) alone () or in combination with LPS (•), and E-selectin expression was determined by ELISA. Results show the mean ± SEM of four experiments, except where SEM lies within a data point. The theoretical additive values ({circ}) were determined by addition of E-selectin expression induced by LPS with LTA-induced expression. *, p < 0.01; **, p < 0.001 (significant induction of E-selectin expression compared with treatment with culture medium). #, p < 0.01; ##, p < 0.001 (significant inhibition of E-selectin expression compared with LPS-induced expression). To ensure clarity, statistical significance for the effects of LTA alone is given in the appropriate Results section.

 


View larger version (21K):
[in this window]
[in a new window]
  		FIGURE 4. The S. pyogenes LTA inhibits LPS-induced ICAM-1 and E-selectin expression in HLMVEC monolayers. HLMVEC were treated for 24 h (A) or 6 h (B) with culture medium ({square}; M), LPS ({blacksquare}; 0.1 µg/ml), and S. aureus LTA (0.3–30 µg/ml) alone () or with LPS (•). ICAM-1 (A) and E-selectin (B) expressions were determined by ELISA. Results show the mean ± SEM of three experiments. *, p < 0.01 (significant induction of ICAM-1 or E-selectin expression compared with treatment with culture medium). #, p < 0.01 (significant inhibition of LPS-induced ICAM-1 or E-selectin expression).

 
Additive effect of LTA on TNF-{alpha}-induced ICAM and E-selectin expression

Tumor necrosis factor-{alpha} (0.1 ng/ml) significantly induced ICAM-1 (24 h) and E-selectin expression (6 h) in HLMVEC monolayers (Fig. 5Go). The ICAM-1 and E-selectin expression in HLMVEC treated with TNF-{alpha} and S. aureus LTA in combination were similar to the theoretical additive values for these stimuli (Fig. 5Go). In particular, E-selectin induced by TNF-{alpha} in combination with 1, 3, or 10 µg/ml LTA was significantly greater (p < 0.05) than that induced by TNF-{alpha} alone (Fig. 5GoB). In contrast to the inhibitory effect on LPS-induced ICAM-1 or E-selectin expression of LTA at 30 µg/ml, TNF-{alpha}-induced expression was not significantly altered by S. aureus LTA at this concentration (Fig. 5Go).



View larger version (28K):
[in this window]
[in a new window]
  		FIGURE 5. The S. aureus LTA in combination with TNF-{alpha} increases ICAM-1 and E-selectin expression in HLMVEC monolayers. HLMVEC were treated for 24 h (A) or 6 h (B) with culture medium ({square}; M), TNF-{alpha} ({blacksquare}; 0.1 ng/ml), and S. aureus LTA (0.3–30 µg/ml) alone () or with TNF-{alpha} (•). ICAM-1 (A) and E-selectin (B) expressions were determined by ELISA. Results show the mean ± SEM of four experiments. The theoretical additive values ({circ}) were determined by addition of CAM expression induced by TNF-{alpha} with LTA-induced expression. *, p < 0.05; **, p < 0.001 (significant induction of CAM expression compared with treatment with culture medium). #, p < 0.05 (significant increase in expression compared with TNF-{alpha}-induced CAM expression).

 
Neutrophil adhesion to HLMVEC pretreated with LPS, TNF-{alpha}, and LTA

In these experiments we investigated whether the effects of S. aureus LTA on LPS- and TNF-{alpha}-induced CAM expression in HLMVEC were paralleled by similar effects on neutrophil adhesion. Stimulation of HLMVEC monolayers (6 h) with LPS (0.1 µg/ml) and with S. aureus LTA at concentrations of 1, 3, and 10 µg/ml, but not 0.3 or 30 µg/ml, caused a significant increase (p < 0.01, 0.05, 0.01, and 0.05, respectively) in neutrophil adhesion (Fig. 6GoA). However, pretreatment of HLMVEC with LPS and LTA together gave a percent adhesion of neutrophils that was less than that predicted for an additive effect. Also, pretreatment of HLMVEC with LPS in combination with 30 µg/ml LTA reduced neutrophil adhesion to basal levels (Fig. 6GoA). Pretreatment of HLMVEC with TNF-{alpha} (0.1 ng/ml, 6 h) also significantly increased (p < 0.05) neutrophil adhesion from basal levels (Fig. 6GoB). Neutrophil adhesion to HLMVEC pretreated with TNF-{alpha} combined with 1, 3, or 10 µg/ml of LTA was similar to the theoretical additive values and was not inhibited in the presence of 30 µg/ml LTA (Fig. 6GoB).



View larger version (28K):
[in this window]
[in a new window]
  		FIGURE 6. Neutrophil adhesion to HLMVEC monolayers pretreated with LPS, TNF-{alpha}, and S. aureus LTA. HLMVEC were pretreated for 6 h with culture medium ({square}; M), LPS ({blacksquare}; 0.1 µg/ml; A), TNF-{alpha} ({blacksquare}; 0.1 ng/ml; B), and LTA (0.3–30 µg/ml) alone () or in combination with LPS or TNF-{alpha} (•). Results show the mean ± SEM of three separate experiments. The theoretical additive values ({circ}) were determined by addition of neutrophil adhesion to monolayers treated with LPS or TNF-{alpha} and LTA. *, p < 0.05; **, p < 0.01 (significant increases in neutrophil adhesion compared with monolayers treated with medium alone). #, p < 0.01 (significant inhibition of neutrophil adhesion compared with monolayers treated with LPS alone). To ensure clarity, statistical significance for the effects of LTA alone is given in Results.

 
Treatment of HLMVEC monolayers with LTA, LPS, or TNF-{alpha} did not alter cell viability

Uptake of EtH-1 was used to determine whether decreases in CAM expression and neutrophil adhesion were due to a reduction in HLMVEC viability. Fluorescence was increased 3.6-fold after treatment of HLMVEC with saponin (0.1%, 10 min; Table IGo) compared with that in control cells. Treatment of HLMVEC monolayers for 24 h with S. aureus LTA alone or in combination with LPS or TNF-{alpha} did not significantly increase fluorescence (Table IGo). Similar data were obtained with S. pyogenes LTA (data not shown). These results suggest that the inhibitory effects of LTA are unlikely to be due to decreased HLMVEC viability.


View this table:
[in this window]
[in a new window]
  		Table I. Uptake of ethidium homodimer-1 by HLMVEC monolayers after 24-h treatment

 
Effect of LTA in combination with LPS or TNF-{alpha} on IL-8 release in HLMVEC

In the experiments described in this section, the effects of activating HLMVEC for 24 h with S. aureus LTA in combination with LPS (0.1 µg/ml) or TNF-{alpha} (0.1 ng/ml) and also with S. pyogenes LTA in combination with LPS on IL-8 release were assessed. Induction was normalized so that LPS- or TNF-{alpha}-induced IL-8 was 100%, and the effects of LTA alone and in combination were expressed as a percentage of the effect of LPS or TNF-{alpha} alone, where appropriate. The theoretical additive values of IL-8 release in these experiments are within the range of maximal release from HLMVEC (data not shown).

LPS-induced IL-8 release was not increased by S. aureus LTA (Fig. 7GoA), was less than the theoretical additive values, and was significantly inhibited (p < 0.01) with 30 µg/ml LTA (Fig. 7GoA). Similar effects were seen with S. pyogenes LTA, although an inhibitory effect with 30 µg/ml LTA was not detected (Fig. 7GoB). Finally, IL-8 release in HLMVEC treated with TNF-{alpha} and S. aureus LTA in combination was similar to the theoretical additive value for these stimuli and was not inhibited in the presence of 30 µg/ml LTA (Fig. 7GoC).



View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 7. Effects of S. aureus LTA in the presence of LPS or TNF-{alpha} and of S. pyogenes LTA with LPS on IL-8 release in HLMVEC monolayers. HLMVEC were treated for 24 h with LPS (0.1 µg/ml) and S. aureus LTA (0.3–30 µg/ml; A), LPS and S. pyogenes LTA (0.3–30 µg/ml; B), and S. aureus LTA and TNF-{alpha} (0.1 ng/ml). IL-8 release was normalized so that LPS-induced (A and B) or TNF-{alpha}-induced (C) IL-8 was 100%, and release in the presence of the other stimuli was expressed as a percentage of these. Results show the mean ± SEM of four experiments. #, p < 0.01 (significant difference compared with LPS- or TNF-{alpha}-induced IL-8 release from HLMVEC monolayers; analysis was performed on nontransformed data). Basal release of IL-8 (95 ± 18 pM (A and C) and 35 ± 5 pM (B)) was increased to 708 ± 56 pM (p < 0.01; A) or 1831 ± 303 pM (p < 0.001; B) with LPS and with TNF-{alpha} to 425 ± 75 pM (p < 0.05; C). To ensure clarity, statistical significance for the effects of LTA alone is given in Results.

 
LTA inhibits LPS-induced, but not TNF-{alpha}-induced, IL-8 mRNA expression

In these experiments we investigated whether the effects of S. aureus LTA with LPS or TNF-{alpha} on IL-8 release in HLMVEC were paralleled by similar effects on IL-8 mRNA induction. Total RNA was isolated from HLMVEC following 2-h stimulation, and semiquantitative RT-PCR analysis was used to assess the level of IL-8 and ß-actin mRNA. Induction was expressed as a ratio of IL-8/ß-actin band densities. Figs. 8Go and 9 show blots for ß-actin (A) or IL-8 (B) RT-PCR products from representative experiments and bar graphs (C) of mean IL-8/ß-actin ratios for three experiments. The ratio for HLMVEC monolayers treated with LPS (0.1 µg/ml) or LTA (3 µg/ml) was significantly increased (p < 0.001 or p < 0.01) compared with that in untreated monolayers (Fig. 8GoC). However, LTA in combination with LPS significantly decreased (p < 0.05) the IL-8/ß-actin ratio compared with that in LPS-treated monolayers (Fig. 8GoC). TNF-{alpha} also increased IL-8 mRNA, but this was not significantly altered by LTA (Fig. 9GoC).



View larger version (37K):
[in this window]
[in a new window]
  		FIGURE 8. Effects of S. aureus LTA on LPS-induced IL-8 mRNA expression in HLMVEC. Total RNA was isolated from HLMVEC after treatment for 2 h with culture medium, LPS (0.1 µg/ml), LTA (3 µg/ml), or LPS/LTA in the absence ({square}) or the presence ({blacksquare}) of cycloheximide (CHX; 5 µg/ml), and semiquantitative RT-PCR was used to analyze IL-8 or ß-actin mRNA. Induction of mRNA is expressed as a ratio of the band densities for IL-8/ß-actin RT-PCR products. Blots for ß-actin (A) or IL-8 (B) mRNA for a representative experiment and a bar graph (C) of the mean ± SEM of IL-8/ß-actin ratios from three separate experiments are shown. *, p < 0.01; **, p < 0.001 (significant increases in IL-8 mRNA expression compared with control cells treated with medium alone and without CHX). #, p < 0.05 (significant inhibition of IL-8 mRNA expression compared with LPS-induced expression without CHX). +, p < 0.01 (significant superinduction of IL-8 mRNA expression by CHX compared with respective controls without CHX). Statistical analysis was conducted using one-way ANOVA on log-transformed data, followed by the Tukey-Kramer multiple-comparison test.

 


View larger version (36K):
[in this window]
[in a new window]
  		FIGURE 9. Effects of S. aureus LTA on TNF-{alpha}-induced IL-8 mRNA expression in HLMVEC. Total RNA was isolated from HLMVEC after treatment for 2 h with culture medium, TNF-{alpha} (0.1 µg/ml), LTA (3 µg/ml), or TNF-{alpha}/LTA in the absence ({square}) or the presence ({blacksquare}) of cycloheximide (CHX; 5 µg/ml), and semiquantitative RT-PCR was used to analyze IL-8 or ß-actin mRNA. Induction of mRNA is expressed as a ratio of the band densities for IL-8/ß-actin RT-PCR products. Blots for ß-actin (A) or IL-8 (B) mRNA for a representative experiment and a bar graph (C) of the mean ± SEM of IL-8/ß-actin ratios from three separate experiments are shown. *, p < 0.001 (significant increases in IL-8 mRNA expression compared with control cells treated with medium alone and without CHX). +, p < 0.01 (significant superinduction of IL-8 mRNA expression by CHX compared with control without CHX). Statistical analysis was conducted using one-way ANOVA on log-transformed data, followed by the Tukey-Kramer multiple comparison test.

 
Inhibition of protein synthesis increases LPS/LTA-induced IL-8 mRNA expression

To assess whether levels of IL-8 mRNA expression were regulated by de novo protein synthesis, HLMVEC monolayers were treated for 2 h with cycloheximide (5 µg/ml), alone or in combination with LPS, LTA, TNF-{alpha}, LPS/LTA, and TNF-{alpha}/LTA. Cycloheximide is a protein synthesis inhibitor that can cause the superinduction of many genes by preventing the degradation of otherwise labile mRNA. Cycloheximide significantly increased (p < 0.01) the IL-8/ß-actin ratio in control cells, but not in cells treated with LPS, TNF-{alpha}, or LTA alone (Figs. 8Go and 9Go). The IL-8/ß-actin ratio for HLMVEC treated with LPS/LTA was, however, significantly increased (p < 0.01) in the presence of cycloheximide by ~6-fold (Fig. 8Go), whereas the ratio for TNF-{alpha}/LTA-treated cells was not increased significantly (Fig. 9Go).

LTA destabilizes LPS-induced IL-8 mRNA

The stability of IL-8 mRNA in HLMVEC following treatment for 2 h with LPS, S. aureus LTA, or LPS with LTA was examined after addition of actinomycin D (5 µg/ml) to stop transcription (Fig. 10Go). IL-8 and ß-actin mRNA were measured 0 (time zero), 2, 4, and 6 h after actinomycin D administration. Induction was expressed as a ratio of IL-8/ß-actin band density, normalized so that the ratio for induction at time zero for each stimuli was 100. The IL-8/ß-actin ratio in LPS-treated HLMVEC at 2, 4, and 6 h was not significantly different from that at time zero (Fig. 10Go), and therefore, the half-life of IL-8 mRNA under these conditions was >6 h. Although the ratios for IL-8/ß-actin in LTA-treated cells decreased compared with the effects of LPS alone at corresponding times, they were not significantly different (Fig. 10Go). In contrast, the IL-8/ß-actin band density ratios in LTA/LPS-treated HLMVEC were significantly (p < 0.01) reduced, compared with LPS, at 4 and 6 h to 40% of the ratio at time zero (Fig. 10Go). The half-life of IL-8 mRNA in LTA/LPS-treated cells was therefore ~2 h. Destabilization of IL-8 mRNA in LTA/LPS-treated cells was reversed by cycloheximide (5 µg/ml; data not shown).



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 10. Effects of S. aureus LTA on LPS-induced mRNA stability. HLMVEC monolayers were stimulated with LPS (0.1 µg/ml), LTA (3 µg/ml), or LPS/LTA; after 2 h, actinomycin D (5 µg/ml; an inhibitor of mRNA translation) was added to the monolayers, and total RNA was isolated at 0, 2, 4, or 6 h after actinomycin D addition. This allows assessment of mRNA stability over time. Semiquantitative RT-PCR was used to measure mRNA for IL-8 or the housekeeping gene ß-actin. The mRNA induction is expressed as a ratio of IL-8/ß-actin mRNA band density, normalized so that the ratio at 2 h, before actinomycin D addition, for LPS, LTA, and LTA plus LPS was 100. Results show the mean ± SEM of IL-8/ß-actin ratios for six experiments. *, p < 0.01 (significant decreases in IL-8/ß-actin ratio compared with the ratio for LPS-treated HLMVEC monolayers at the corresponding time).

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The complex interaction between an infectious agent and the host may result in induction or suppression of the inflammatory response, and the balance is critical in determining the outcome of infection. This study shows that S. aureus LTA had a bell-shaped effect on CAM expression and release of IL-8 in HLMVEC, inhibited LPS-induced expression of these properties, but had an additive effect with TNF-{alpha}. In contrast, S. pyogenes LTA did not induce CAM expression, but caused a concentration-dependent increase in IL-8 release in HLMVEC and inhibited LPS-induced CAM expression, but not IL-8 release. The stimulatory effects of LTA are unlikely to be the result of LPS contamination of LTA, because the amounts of LPS associated with the concentrations used in this study did not activate HLMVEC (22). Also, as LTA did not decrease HLMVEC viability, the inhibitory effects are unlikely to result from a detergent-like action associated with micelle formation at high concentrations of LTA (47). To our knowledge, this is the first report to show that LTA inhibits LPS-induced endothelial cell function and to highlight important differences between the effects of LTA from S. aureus and S. pyogenes in HLMVEC.

The bell-shaped effects of S. aureus LTA on CAM expression and neutrophil adhesion in HLMVEC contrast with the concentration-dependent increase in HUVEC CAM expression (29). In vivo, a bell-shaped effect of platelet-activating factor on the release of inflammatory mediators has been speculated to protect the endothelium from overstimulation during severe sepsis (48). Down-regulation of HLMVEC function in our study may represent a similar protective mechanism. The maximal effects of S. aureus LTA on ICAM-1 or E-selectin expression in HLMVEC were 2- and 3-fold lower, respectively, than those in HUVEC at corresponding times (29) and may also represent the control of endothelial activation locally in the lung microvasculature.

The importance of being able to down-regulate inflammation locally is illustrated in a mouse model of S. aureus-induced arthritis in which leukocyte recruitment, tissue injury, and mortality were compared in genetically manipulated mice lacking ICAM-1 and wild-type controls (49). The results showed that ICAM-1-deficient mice had fewer leukocytes at the site of infection and reduced leukocyte-mediated tissue damage, but that mortality was greater. This suggests that while induction of ICAM-1 is essential for protection against systemic disease, it may exacerbate disease manifestation locally. Thus, local down-regulation in response to bacteria may limit leukocyte-mediated tissue/organ damage without negating the systemic inflammatory response. This protection may be of particular importance in the lung, because the capillary microvascular endothelium is more likely than other vascular beds, with the exception of the gut, to be exposed to high concentrations of bacterial products.

In contrast to the effects seen with S. aureus LTA on CAM expression, S. pyogenes LTA did not induce ICAM-1 or E-selectin in HLMVEC under the conditions used in this study. It is of note that clinically a striking feature in some patients with S. pyogenes infection is an absence of inflammatory cells at infection sites and also a relatively high mortality (50). In contrast, S. aureus infection is associated with brisk neutrophil infiltrates at the site of infection resulting from the orchestrated expression of CAM on endothelial cells (35). The observation that CAM expression in endothelial cells stimulated with LTA parallels neutrophil accumulation in patients infected with S. aureus or S. pyogenes establishes an important link between in vitro and in vivo results.

The stimulatory effects of S. aureus and S. pyogenes LTA on IL-8 release in HLMVEC were similar, apart from a lack of effect with S. aureus LTA at 30 µg/ml. In contrast to the largely stimulatory effects of LTA, PepG had no effect on IL-8 release or CAM expression in HLMVEC or on similar properties in HUVEC (27) despite the fact that PepG makes up 50% of the cell wall of staphylococcal bacteria (35). The lack of effect of PepG may result from the absence of membrane-associated CD14 on endothelial cells, a receptor to which PepG and also LPS and LTA bind, or the inability of PepG to bind soluble CD14 in serum (27).

In the second part of this study we investigated the effects of LTA in combination with LPS or TNF-{alpha} on HLMVEC CAM expression and IL-8 release. When several bacteria are present during infection, the interactions between the products of these organisms and/or the cytokines they induce may be instrumental in determining the outcome of infection. An understanding of these interactions may be essential for improving therapeutic intervention for bacterial infection. In this study we showed that while LTA had an additive effect with TNF-{alpha} on HLMVEC function, the effects of LTA with LPS were inhibitory. This provides evidence that local down-regulation of an inflammatory response in the lung, as suggested above, may also have the advantage of being stimulus specific.

Having established that LTA inhibited LPS-induced HLMVEC function, we addressed the mechanism(s) that may be responsible for the inhibitory effects. Our results provide insight into the mechanisms that may be involved and also help to eliminate others. A mechanism that is unlikely to account for the inhibitory effect is competitive antagonism for a receptor such as CD14, as shown for LTA inhibition of LPS-induced responses in monocytes and fibroblasts (51). For competitive antagonism, a greater inhibitory effect would normally be expected in the presence of a lower concentration of LPS, whereas our results show the effect was greater in the presence of a higher concentration. Evidence that CD14-dependent antagonism in particular is unlikely to account for the inhibitory effects is provided by our observation that LPS-induced CAM expression in HLMVEC becomes independent of CD14 with increasing concentrations of LPS, being <20% dependent at 0.1 µg/ml (our unpublished observation). A similar independence of CD14, with increasing concentrations of LPS, has also been reported in macrophages (52).

One explanation for the inhibitory effects of LTA is that LTA induces the synthesis of inhibitory proteins that limit HLMVEC activation by bacteria stimuli. At high concentrations of LTA (30 µg/ml), sufficient inhibitory proteins may be induced to prevent its own action and thus account for the bell-shaped effect. However, LPS may enhance LTA-induced synthesis of such inhibitory proteins, and this may in part account for the greater effect with a higher (0.1 µg/ml), rather than a lower (0.01 µg/ml), concentration of LPS. Further evidence that the synthesis of inhibitory protein(s) plays an important role in the inhibitory effects is provided by the observation that low levels of IL-8 mRNA in LPS/LTA-treated HLMVEC were increased by cycloheximide, a protein synthesis inhibitor. Cycloheximide also significantly increased levels of IL-8 mRNA in resting HLMVEC, which suggests that the inhibitory protein(s) regulating IL-8 production is synthesized constitutively, but that expression is increased with LTA or LPS in combination with LTA. Similar inhibitory proteins are also thought to regulate IL-8 production in monocytes and account for the inability of smooth muscle cells to express E-selectin (23, 53). Constitutive expression of inhibitory proteins that limit basal expression of inflammatory genes and are also inducible in response to appropriate inflammatory stimuli may provide an important control mechanism for inflammation.

We further investigated the inhibitory effect of LTA on LPS-induced IL-8 release and established that LTA destabilized LPS-induced IL-8 mRNA and reduced the half-life from >6 h to ~2 h. The mRNAs of certain cellular proteins, including IL-8, share a common AUUUA pentamer in their 3' untranslated region that confers instability (54). A number of groups have searched for proteins that could potentially modulate the turnover of AUUUA mRNA, and an AU-binding factor, Auf, has been identified that binds to the AUUUA region (54). It is speculated that the binding of this factor leaves the mRNA unprotected and susceptible to degradation (54). At present it is not known, however, whether LTA alone or in combination with LPS induces the expression of Auf or similar AU-binding factors.

Finally, the extent of neutrophil accumulation at sites of infection in patients with mixed Gram-negative and Gram-positive bacterial infection has not, to our knowledge, been specifically addressed. In animal models of mixed infection, aerosolized challenge of LTA or whole Gram-positive bacteria in LPS-treated rats prevented neutrophil accumulation in the lung. However, the contribution to suppressed neutrophil accumulation of CAM expression or chemokine release in lung endothelial cells was not specifically addressed in these studies (32, 33). To establish a direct link, these parameters would need to be determined in tissue taken from an animal model following measurement of appropriate in vivo parameters and is the subject of further studies. Clinical predictions based on findings with in vitro or in vivo models should be made with caution, but, on balance, these results together with ours suggest that inhibition of neutrophil accumulation in the lung may occur in mixed bacterial infection.

In conclusion, our results suggest that LTA from Gram-positive bacteria can limit the extent of HLMVEC activation induced either by itself or by LPS from Gram-negative bacteria, and the effect is dependent on synthesis of inhibitory protein. HLMVEC do not completely lose the ability to express CAM and release IL-8 in response to LTA because their effect with TNF-{alpha} remains additive. Whether a selective decrease in endothelial cell function has evolved to provide local protection from bacterial insult in the lung, as we have suggested, or a mechanism exists by which bacteria evade detection can only be speculated. However, these results increase our knowledge of the effects and mechanisms of action of bacteria on endothelial cell inflammatory properties and may have important implications for the treatment of bacterial infections in the future.


   Footnotes
 
1 This work was supported by the British Heart Foundation and the National Asthma Campaign (U.K.). Back

2 Current address: Department of Pathology, University of Michigan, 1303 Catherine Road, Ann Arbor, MI 48109. Back

3 Address correspondence and reprint requests to Dr. Anne Burke-Gaffney, Unit of Critical Care, National Heart and Lung Institute Division, Imperial College School of Medicine, Royal Brompton Campus, Sydney Street, London, United Kingdom SW3 6NP. E-mail address: Back

4 Abbreviations used in this paper: CAM, cell adhesion molecule; LTA, lipoteichoic acid; PepG, peptidoglycan; HUVEC, human umbilical vein endothelial cells; HLMVEC, human lung microvascular endothelial cells; EtH-1, ethidium homodimer-1; EGM-MV, microvascular endothelial growth medium; KRPD, Krebs-Ringer phosphate buffer. Back

Received for publication March 12, 1999. Accepted for publication September 13, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Bone, R. C.. 1991. The pathogenesis of sepsis. Ann. Intern. Med. 115:457.
  2. Bone, R. C.. 1994. Sepsis and its complications: the clinical problems. Crit. Care Med. 22:S8.[Medline]
  3. Bone, R. C.. 1991. Gram-negative sepsis: background, clinical features and intervention. Chest 100:802.[Abstract/Free Full Text]
  4. Bone, R. C.. 1993. How gram-positive organisms cause sepsis. J. Crit. Care Med. 8:51.
  5. Bone, R. C.. 1994. Gram-positive organisms and sepsis. Arch. Intern. Med. 154:26.[Abstract]
  6. Friedman, G., E. Silva, J. L. Vincent. 1998. Has the mortality of septic shock changed with time?. Crit. Care Med. 26:2078.[Medline]
  7. Wheeler, A. P., G. R. Bernard. 1999. Treating patients with severe sepsis. N. Engl. J. Med. 340:207.[Free Full Text]
  8. Welbourne, C. R., Y. Young. 1992. Endotoxin, septic shock and acute lung injury: neutrophils, macrophages and inflammatory mediators. Br. J. Surg. 79:998.[Medline]
  9. Worthern, G. S., M. G. Tonnesen, D. C. Lien, P. M. Hensen. 1987. Interaction of leukocytes with the pulmonary endothelium. U. S. Ryan, ed. Pulmonary Endothelium in Health and Disease 123. Marcel Dekker, New York and Basel.
  10. Worthern, G. S., G. P. Downey. 1996. Mechanisms of neutrophil-mediated lung injury. T. W. Evans, and C. Haslett, eds. ARDS Acute Respiratory Diseases in Adults 99. Chapman and Hall, London.
  11. Williams, T. J., P. G. Hellewell. 1992. Adhesion molecules involved in the microvascular inflammatory response. Am. Rev. Respir. Dis. 146:S45.[Medline]
  12. Albelda, S. M., C. W. Smith, P. A. Ward. 1994. Adhesion molecules and inflammatory injury. FASEB J. 8:504.[Abstract]
  13. Carlos, T. M., J. M. Harlan. 1994. Leukocyte-endothelial adhesion molecules. Blood 84:2068.[Abstract/Free Full Text]
  14. Malik, A. B., S. K. Lo. 1996. Vascular endothelial adhesion molecules and tissue inflammation. Pharmacol. Rev. 48:213.[Medline]
  15. Cowley, H. C., D. Heney, A. J. Gearing, I. Hemmingway, N. R. Webster. 1994. Increased circulating adhesion molecule concentration in patients with the systemic inflammatory response syndrome: a prospective cohort study. Crit. Care Med. 22:651.[Medline]
  16. Fujishima, S., J. Sasaki, Y. Shinozawa, K. Takuma, H. Kimura, M. Suzuki, M. Kanazawa, S. Hori, N. Aikawa. 1996. Serum MIP-1{alpha} and IL-8 in septic patients. Inten. Care Med. 22:1169.
  17. Grau, G. E., N. Mili, J. N. Lou, D. R. Morel, B. Ricoy, R. Lucas, P. M. Suter. 1996. Phenotypic and functional analysis of pulmonary microvascular endothelial cells from patients with acute respiratory distress syndrome. Lab. Invest. 74:761.[Medline]
  18. Moss, M., M. K. Gillespie, L. Ackerson, F. A. Moore, E. E. Moore, P. E. Parsons. 1996. Endothelial cell activity varies in patients at risk for the adult respiratory distress syndrome. Crit. Care Med. 24:1782.[Medline]
  19. Kiehl, M. G., H. Ostermann, M. Thomas, C. Muller, U. Cassens, J. Kienast. 1998. Inflammatory mediators in bronchoaveolar lavage fluid and plasma in leukocytopenic patients with septic shock-induced acute respiratory distress syndrome. Crit. Care Med. 26:1194.[Medline]
  20. Rietschel, E. T., H. Brade. 1992. Bacterial endotoxins. Sci. Am. 267:54.[Medline]
  21. Brown, Z., M. E. Gerritsen, W. W. Carley, R. M. Strieter, S. L. Kunkel, J. Westwick. 1994. Chemokine gene expression and secretion by cytokine-activated human microvascular endothelial cells: differential regulation of monocyte chemoattractant protein-1 and interleukin-8 in response to IFN{gamma}. Am. J. Pathol. 145:913.[Abstract]
  22. Blease, K., J. Seybold, I. M. Adcock, P. G. Hellewell, A. Burke-Gaffney. 1998. Interleukin-4 and lipopolysaccharide synergise to induce vascular cell adhesion molecule-1 expression in human lung microvascular endothelial cells. Am. J. Respir. Cell Mol. Biol. 18:620.[Abstract/Free Full Text]
  23. Standiford, T. J., D. A. Arenberg, J. M. Danforth, S. L. Kunkel, G. M. VanOtteren, R. M. Strieter. 1994. Lipoteichoic acid induces secretion of interleukin-8 from human blood monocytes: a cellular and molecular analysis. Infect. Immun. 62:119.[Abstract/Free Full Text]
  24. Carratelli, C. R., I. Nuzzo, C. Bentivoglio, M. Galdiero. 1996. CD11a/CD18 and CD11b/CD18 modulation by lipoteichoic acid, N-acetyl-muramyl-{alpha}-alanyl-D-isoglutamine, muramic acid and protein A from Staphylococcus aureus. FEMS Immunol. Med. Microbiol. 16:309.
  25. Sugiyama, A., R. Arakaki, T. Ohnishi, N. Arakaki, Y. Daikuhara, H. Takada. 1996. Lipoteichoic acid and interleukin 1 stimulate synergistically production of hepatocyte growth factor (scatter factor) in human gingival fibroblasts in culture. Infect. Immun. 64:1426.[Abstract]
  26. DeKimpe, S. J., K. M. Kengatharan, C. Thiemermann, J. R. Vane. 1995. The cell wall components peptidoglycan and lipoteichoic acid from Staphylococcus aureus act in synergy to cause shock and multiple organ failure. Proc. Natl. Acad. Sci. USA 92:10359.[Abstract/Free Full Text]
  27. Jin, Y., D. Gupta, R. Dziarski. 1998. Endothelial and epithelial cells do not respond to complexes of peptidoglycan with soluble CD14 but are activated indirectly by peptidoglycan-induced tumor necrosis factor-alpha and interleukin-1 from monocytes. J. Infect. Dis. 177:1629.[Medline]
  28. Kengatharan, K. M., S. DeKimpe, C. Robson, S. J. Foster, C. Thiemermann. 1998. Mechanism of Gram-positive shock: identification of peptidoglycan and lipoteichoic acid moieties essential in the induction of nitric oxide synthase, shock, and multiple organ failure. J. Exp. Med. 188:305.[Abstract/Free Full Text]
  29. Kawamura, N., N. Imanishi, H. Koike, H. Nakahara, L. Phillips, S. Morooka. 1995. Lipoteichoic acid-induced neutrophil adhesion via E-selectin to human umbilical vein endothelial cells. Biochem. Biophys. Res. Commun. 217:1208.[Medline]
  30. Beekhuizen, H., J. S. van de Gevel, B. Olsson, I. J. van Benten, R. van Furth. 1997. Infection of human vascular endothelial cells with Staphylococcus aureus induces hyperadhesiveness for human monocytes and granulocytes. J. Immunol. 158:774.[Abstract]
  31. Yao, L., F. D. Lowy, J. W. Berman. 1996. Interleukin-8 gene expression in Staphylococcus aureus-infected endothelial cells. Infect. Immun. 64:3407.[Abstract]
  32. Nelson, S., C. Chidiac, G. Bagby, W. R. Summer. 1990. Endotoxin-induced suppression of lung host defenses. J. Med. 21:85.[Medline]
  33. Wagner, J. G., K. E. Driscoll, R. A. Roth. 1999. Inhibition of pulmonary neutrophil trafficking during endotoxemia is dependent on the stimulus for migration. Am. J. Respir. Cell Mol. Biol. 20:769.[Abstract/Free Full Text]
  34. Spencer, R.. 1995. Invasive streptococci. Eur. J. Clin. Microbiol. Infect. Dis. 14:S26.
  35. Lowy, F. D.. 1998. Staphylococcus aureus infections. N. Engl. J. Med. 339:520.[Free Full Text]
  36. Rothlein, R., M. Czajkowski, M. M. O’Neill, S. D. Marlin, E. Mainolfi, V. J. Merluzzi. 1988. Induction of intercellular adhesion molecule 1 on primary and continuous cell lines by pro-inflammatory cytokines. J. Immunol. 141:1665.[Abstract]
  37. Pigott, R., L. A. Needham, R. M. Edwards, C. Walker, C. Power. 1991. Structural and functional studies of the endothelial activation antigen endothelial leukocyte adhesion molecule-1 using a panel of monoclonal antibodies. J. Immunol. 147:130.[Abstract]
  38. Shen, J., R.G. Ham, S. Karmiol. 1995. Expression of adhesion molecules in cultured human pulmonary microvascular endothelial cells. Microvasc. Res. 50:360.[Medline]
  39. Visner, G. A., S. E. Chesrown, J. Monnier, U. S. Ryan, H. S. Nicks. 1992. Regulation of manganese superoxide dismutase: IL-1 and TNF induction in pulmonary artery and microvascular endothelial cells. Biochem. Biophys. Res. Commun. 188:453.[Medline]
  40. Pan, M., M. Wasa, U. Ryan, W. Souba. 1996. Lipopolysaccharide and tumor necrosis factor stimulate lung microvascular arginine uptake, a response attenuated by dexamethasone. J. Parenter. Enteral. Nutr. 20:50.[Abstract]
  41. Haslett, C., L. A. Guthrie, M. M. Kopaniak, R. B. Johnston, P. M. Henson. 1985. Modulation of multiple neutrophil function by preparative methods or trace concentrations of bacterial lipopolysaccharide. Am. J. Pathol. 119:101.[Abstract]
  42. Burke-Gaffney, A., P. G. Hellewell. 1996. Regulation of ICAM-1 by dexamethasone in a human vascular endothelial cell line EAhy926. Am. J. Physiol. 270:C552.[Abstract/Free Full Text]
  43. Burke-Gaffney, A., P. G. Hellewell. 1996. Eotaxin stimulates adhesion to human lung microvascular endothelial cells. Biochem. Biophys. Res. Commun. 227:35.[Medline]
  44. Burke-Gaffney, A., P. G. Hellewell. 1997. Endogenous nitric oxide limits cytokine-induced damage of murine lung epithelial cells. Am. J. Physiol. 272:707.
  45. Au, B.-T., T. J. Williams, P. D. Collins. 1994. Zymosan-induced interleukin-8 release from human neutrophils involves activation via the CD11b/CD18 receptor and endogenous platelet-activating factor as an autocrine modulator. J. Immunol. 152:5411.[Abstract]
  46. Collins, P. D., P. J. Jose, T. J. Williams. 1991. The sequential generation of neutrophil chemoattractant proteins in acute inflammation in the rabbit in vivo: relationship between C5a and a protein with the characteristics of IL-8. J. Immunol. 146:677.[Abstract]
  47. Courtney, H. S., W. A. Simpson, E. H. Beachey. 1986. Relationship of critical micelle concentrations of bacterial lipoteichoic acids to biological activities. Infect. Immun. 51:414.[Abstract/Free Full Text]
  48. Koltai, M., D. Hosford, P. G. Braquet. 1993. Platelet-activating factor in septic shock. New Horiz. 1:87.[Medline]
  49. Verdrengh, M., T. A. Springer, J. C. Gutierrez-Ramos, A. Tarkowski. 1996. Role of intercellular adhesion molecule 1 in pathogenesis of staphylococcal arthritis and in host defense against staphylococcal bacteremia. Infect. Immun. 64:2807.
  50. III Cockerill, F. R., R. L. Thompson, J. M. Musser, P. M. Schlievert, J. Talbot, K. E. Holley, W. S. Harmsen, D. M. Ilstrup, P. C. Kohner, M. H. Kim, et al 1998. Molecular, serological, and clinical features of 16 consecutive cases of invasive streptococcal disease. Clin. Infect. Dis. 26:1448.[Medline]
  51. Sugawara, S., R. Arakaki, H. Rikiishi, H. Takada. 1999. Lipoteichoic acid acts as an antagonist and an agonist of lipopolysaccharide on human gingival fibroblasts and monocytes in a CD14-dependent manner. Infect. Immun. 67:1623.[Abstract/Free Full Text]
  52. Gangloff, S., N. Hijiya, A. Haziot, S. M. Goyert. 1999. Lipopolysaccharide structure influences the macrophage response via CD14-independent and CD14-dependent pathways. Clin. Infect. Dis. 28:491.[Medline]
  53. Chen, X.-L. L., P. E. Tummala, L. Olliff, R. M. Medford. 1997. E-selectin gene expression in vascular smooth muscle cells: evidence for a tissue-specific repressor protein. Circ. Res. 80:305.[Abstract/Free Full Text]
  54. Carter, B. Z., J. S. Malter. 1991. Regulation of mRNA stability and its relevance to disease. Lab Invest. 65:610.[Medline]



This article has been cited by other articles:


Home page