| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print December 12, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Institute for Medical Microbiology and Hygiene, University of Lübeck, Germany; and
Department of Biochemical Pharmacology, University of Konstanz, Germany
1Correspondence: Institute for Medical Microbiology and Hygiene, University of Lübeck, Ratzeburger Allee 160, D-23538 Lübeck, Germany. E-mail: Tamas.Laskay{at}hygiene.ukl.mu-luebeck.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, and granulocyte-colony stimulating factor by PMN. The effects of LTA on PMN were found to be associated with nuclear factor-
B activation. Of particular interest was that LTA inhibited the spontaneous apoptosis and therefore, increased the lifespan of PMN. Experiments using blocking antibodies revealed that CD14 and Toll-like receptor 2 (TLR2) but not TLR4 play a major role in LTA-mediated effects on PMN. These data clearly show that LTA, a component of gram-positive bacteria, directly activates neutrophil granulocytes, the primary effector cells in the first line of defense against infectious challenge.
Key Words: PMN • LTA • gram-positive bacteria • cellular activation • inflammation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although the activity of pure LTA on many cell types has been studied [9 10 11 ], it is still not clear whether LTA plays a role in the activation of neutrophil granulocytes. In the present study, the reaction of highly purified neutrophil granulocytes to endotoxin-free and bioactive LTA was investigated. Our results indicated that LTA directly activates granulocytes, resulting in morphological changes, shedding of CD62L, degranulation, cytokine release, and priming for formyl-Met-Leu-Phe (fMLP)-mediated oxidative burst. It is most interesting that through delaying the spontaneous apoptosis, LTA treatment led to an extended lifespan of PMN. Blocking experiments revealed that CD14 and Toll-like receptor 2 (TLR2) participate in the LTA-induced effect on neutrophil granulocytes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Isolation of neutrophil granulocytes
Peripheral blood was collected by venipuncture from healthy adult volunteers using lithium-heparin. Granulocytes were isolated as described previously [12
, 13
]. Briefly, blood was layered on a two-layer Histopaque gradient consisting of Histopaque 1077 (upper layer) and Histopaque 1119 (bottom layer; Sigma, Taufkirchen, Germany) and centrifuged for 5 min at 300 g followed by 20 min at 800 g. The plasma and the Histopaque 1077 layer consisting mainly of lymphocytes and monocytes were discarded. The granulocyte-rich layer in the Histopaque 1119 layer was collected, leaving the erythrocyte pellet at the bottom of the tube. Granulocytes were washed twice in phosphate-buffered saline (PBS), resuspended in complete medium (RPMI-1640 medium supplemented with 50 µM 2-mercaptoethanol, 2 mM L-glutamine, 10 mM HEPES, 100 U/ml penicillin, 100 µg/ml streptomycin, all from Biochrom, Berlin, Germany) and 10% low endotoxin fetal calf serum (FCS; Gibco, Karlsruhe, Germany; LPS content, 0.523 ng/ml), and further fractionated on a discontinuous Percoll (Amersham Biosciences, Uppsala, Sweden) gradient consisting of layers with densities of 1.105 g/ml (85%), 1.100 g/ml (80%), 1.093 g/ml (75%), 1.087 g/ml (70%), and 1.081 g/ml (65%). After centrifugation for 20 min at 800 g, the interface between the 80% and 85% Percoll layers was collected, washed twice in PBS, and resuspended in complete medium. All procedures were conducted at RT. Cell viability was >99%, as determined by trypan blue exclusion. The cell preparations contained >99% granulocytes, of which >95% were neutrophils, and 1–4% were eosinophils, as determined by morphological examination of Giemsa-stained cytocentrifuge (Shandon, Pittsburgh, PA) preparates.
As monocytes are known to respond to LTA with cytokine release [9 ], the number of contaminating monocytes in the granulocyte preparations was determined using an immunofluorescent staining. Freshly isolated neutrophil granulocytes (5x105 cells) were resuspended in fluorescein-activated cell sorter (FACS) buffer (PBS containing 1% human serum, 1% bovine serum albumin, and 0.01% sodium azide) in a V-bottom 96-well plate. After washing with FACS buffer, PMN were stained with a phycoerythrin (PE)-conjugated monoclonal antibody (mAb) to CD14 (clone TÜK4, Dako, Hamburg, Germany). Following two washing steps, the cells were fixed with paraformaldehyde (1% in PBS). Cytocentrifuge preparates were embedded in IMAGEN® mounting fluid (Dako), and the cells were observed under a Zeiss Axioskop2 fluorescent microscope. Monocytes that have a high expression of CD14 could be detected as brightly stained red cells. After counting cells in 50 high power fields (10,000 cells), an average of two monocytes was detected, which corresponds to a monocyte contamination of 0.02%.
Cell culture
PMN were incubated in complete medium (see above) at a concentration of 5 x 106/ml at 37°C in a humidified atmosphere containing 5% CO2 in tissue-culture plates with 96 flat-bottom wells (Greiner, Frickenhausen, Germany). All components of the cell-culture medium were of low endotoxin content. The complete cell-culture medium contained between 50 and 100 pg/ml endotoxin as determined by the Limulus amoebocyte lysate assay (Charles River Laboratories, Sulzfeld, Germany). LTA was used in a concentration range of 0.3–30 µg/ml. As the LTA preparation had an endotoxin content as low as 30 pg/mg LTA, LTA contributed solely 0.9 pg LPS/ml culture medium, even in the highest concentration (30 µg/ml) used. This endotoxin level is far below the endotoxin content of the culture medium and therefore, cannot account for the biological effects of the LTA preparation. LPS from Escherichia coli 0111:B4 (Sigma) was used at a concentration of 100 ng/ml. The morphology of PMN in the cell culture was visualized under an invert microscope (Zeiss Axiovert 25) and photographed with a Ricoh HR-10m camera (Ricoh, Tokyo, Japan).
Blocking experiments
Blocking experiments were performed to investigate the role of CD14, TLR2, and TLR4 in the LTA-mediated effects on PMN. Freshly isolated PMN were pretreated with 10 µg/ml anti-CD14 mAb [immunoglobulin G (IgG)1; clone 18D11, Diatec, Hamburg, Germany], 10 µg/ml anti-TLR2 mAb (IgG2a, clone TL2.1, Imgenex, San Diego, CA), and 10 µg/ml anti-TLR4 mAb (IgG2a, clone HAT125, Imgenex) for 30 min at RT and subsequently exposed to LTA. As isotype controls, 10 µg/ml IgG1 (Diatec) or 10 µg/ml IgG2a (eBioscience, San Diego, CA) was used.
Flow cytometry analysis
The cell-surface expression of CD66b as a specific marker for granulocytes [14
] and of CD62L and CD11b as markers for PMN activation was analyzed [15
]. PMN (5x105) were resuspended in FACS buffer (see above) in a V-bottom 96-well plate. After washing with FACS buffer, PMN were stained with fluorescein isothiocyanate (FITC)-conjugated mAb to human CD66b (clone 80H3, IgG1, Immunotech, Hamburg, Germany) and PE-conjugated mAb to CD62L (clone Dreg-56, IgG1, BD Biosciences, Heidelberg, Germany) or PE-conjugated mAb to CD11b (clone 2LPM19c, IgG1, Dako) in FACS buffer for 20 min on ice. Following two wash steps, the cells were fixed with paraformaldehyde (1% in PBS) and analyzed with a FACSCalibur® flow cytometer using CellQuest® software (Becton Dickinson, San Diego, CA). FITC- and PE-conjugated mouse IgG1 isotype-control antibodies were purchased from Dako.
Analysis of intracellular interleukin (IL)-8 by flow cytometry
PMN were cultured at a concentration of 5 x 106/ml in complete medium for 3 h before Brefeldin A (GolgiPlug®, BD Biosciences) was added, and the cultures were further incubated for 12 h. Subsequently, the cells were stained with FITC-conjugated anti-CD66b mAb (Immunotech) followed by permeabilization and intracellular staining of IL-8 using the Cytofix/Cytoperm Plus kit (BD Biosciences) and PE-labeled anti-IL-8 mAb (clone G 265-8, IgG2b, BD Biosciences), respectively. PE-labeled mouse IgG2b (BD Biosciences) was used as an isotype control. The cells were analyzed with a FACSCalibur® flow cytometer using CellQuest® software (Becton Dickinson).
Determination of cytokines in culture supernatants
Cell-free supernatants of PMN cultures were collected and stored at –20°C until cytokine determination. IL-8 was measured using an enzyme-linked immunosorbent assay (ELISA; CytoSets, Biosource, Camarillo, CA), according to the manufacturer’s instructions. The detection limit was 30 pg/ml. Granuloctye-colony stimulating factor (G-CSF) was determined by using the Quantikine HS immunoassay (R&D Systems, Wiesbaden-Nordenstadt, Germany) with a detection limit of 1 pg/ml. Tumor necrosis factor (TNF-) was analyzed by using the Opteia human TNF- set (BD Biosciences) with a detection limit of 3 pg/ml.
Oxidative burst assay
The intracellular production of reactive oxygen radicals was assayed by using the substrate dihydrorhodamine 123 (DHR; Molecular Probes, Leiden, Netherlands), which is fluorescent upon interaction with reactive oxygen species (ROS; H2O2 and HO·) [16
].
To investigate the priming effect of LTA on fMLP-induced oxidative burst, 1 x 106/200 µl PMN were pretreated with various concentrations of LTA for 90 min or without LTA in RPMI + 1% FCS. Then, DHR was added for 5 min followed by stimulation with 1 µM fMLP (Sigma) for 5 min. The reaction was stopped on ice, and the fluorescent intensity of the cells was analyzed immediately by flow cytometry.
Analysis of PMN apoptosis
Morphological assessment of PMN apoptosis
In apoptotic neutrophils, morphological changes are striking and include separation of nuclear lobes and darkly stained pyknotic nuclei [17
, 18
]. Nuclear morphology was assessed on Giemsa-stained cytocentrifuge slides. Cell morphology was examined under oil immersion light microscopy. A minimum of 200 cells/slide was examined and graded as apoptotic/nonapoptotic.
Annexin-V binding
Annexin-V exhibits calcium-dependent binding to phosphatidylserine (PS), expressed in the outer membrane leaflet of apoptotic cells [19
]. Labeling of apoptotic PMN with Annexin-V–FITC (Roche Molecular Biologicals, Mannheim, Germany) was performed as recommended by the manufacturer. Labeled cells were analyzed by flow cytometry using a FACSCalibur® with CellQuest® software (Becton Dickinson).
Deoxyuridine triphosphate (dUTP) nick end-labeling (TUNEL) assay of chromatin fragmentation
The TUNEL assay with fluorescein-dUTP (in situ cell death detection kit, Roche Molecular Biologicals) was used to detect DNA strand breaks in apoptotic cells [20
]. Briefly, freshly isolated PMN (106 cells per probe) were incubated for 24 h in the presence or absence of LTA. The cells were fixed with 4% paraformaldeyde in PBS and incubated for 1 h on a shaker at RT followed by a washing step. The cells were then permeabilized with 100 µl 0.1% Triton X-100 in 0.1% sodium citrate for 2 min on ice. After washing twice, the cells were resuspended in 50 µl TUNEL reaction mixture or in LABEL solution as negative control for 60 min at 37°C in a humidified atmosphere in the dark. The cells were washed again and resuspended in 400 µl PBS and analyzed by flow cytometry (see above).
Western blot analysis of phosphorylated inhibitor of B (P-IB)
PMN were cultured in the absence or presence of 10 or 30 µg/ml LTA for 30, 90, or 150 min. After adding ice-cold RPMI-1640 medium, the cells were pelleted at 800 g for 5 min at 4°C. Cell pellets were washed and dissolved in 2x sample buffer [50 mM Tris, pH 6.80, 4% sodium dodecyl sulfate (wt/vol), 10% 2-mercaptoethanol (vol/vol), 20% glycerol (vol/vol)], containing a protease and phosphatase-inhibitor cocktail [1 mM EDTA (Sigma), 0.5 mM EGTA (Sigma), 1 mM phenylmethylsulfonyl fluoride (Sigma), and 10 µg/ml each of aprotinin (Sigma), leupeptin (Calbiochem, Schwalbach, Germany), pepstatin-A (Sigma), 0.5 mM dithiothreitol (Sigma), 2 mM levamisol (Sigma), 0.5 mM benzamidin (Sigma), 1 mM sodium orthovanadate (Sigma), 10 mM ß-glycerophosphate (Sigma), and complete miniprotease inhibitor (Roche Molecular Biologicals; one tablet dissolved in 1 ml distilled water and used in a final dilution of 1:5]. Samples were then heated for 7 min at 95°C, and equivalents of 1 x 106 cells/lane were loaded in 15% denaturing polyacrylamide gels.
Proteins were blotted onto a nitrocellulose membrane at 21 V constant voltage for 60 min in a Transblot® semidry transfer cell (BioRad, Hercules, CA). The membrane was blocked in PBS containing 0.05% Tween-20 and 3% low-fat skimmed milk for 60 min at RT and was subsequently incubated with a 1:500 dilution of mouse antiphospho-IB (IgG1) antibody (clone IMG-156, Biocarta, Hamburg, Germany) overnight at 4°C. After extensive washing, the membranes were incubated with a horseradish peroxidase-coupled polyclonal goat anti-mouse-IgG secondary antibody (Dako) for 90 min at RT, and the bands were visualized using the chemiluminescent Western blotting detection system (SuperSignal® West Dura extended duration substrate, Pierce, Bonn, Germany). To control equal protein load, the membranes were stripped and incubated with a 1:5000 dilution of anti-ß-actin antibody (clone AC-74, Sigma) for 90 min at RT followed by chemiluminescence detection as described above.
Statistical analysis
Data are expressed as mean ± SD. Statistical significance was assessed by the paired or unpaired Student’s t-test. Throughout the figures, P values of <0.05 are indicated by one asterisk; P values of <0.01 by double asterisks.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
LTA alone does not induce oxidative burst in PMN but primes for fMLP-induced production of reactive oxygen intermediates (ROI)
The ability to induce the production of ROI is characteristic for many neutrophil-activating agents [26
]. Hence,we analyzed the ability of LTA to induce an oxidative burst in human PMN. Intracellular production of ROI was assayed by using DHR, which is fluorescent on interaction with ROS. PMN were incubated in medium alone or with LTA, which did not induce intracellular production of ROI (not shown).
A characteristic feature of low-level PMN activation is the priming, i.e., the amplification of the neutrophil’s functional response to other stimuli [28 ]. LPS, for example, has a priming effect on the fMLP-mediated oxidative burst of PMN [29 ]. fMLP alone induced a relatively weak oxidative response, where only 34% of PMN contained ROI. Preincubation with 10 µg/ml LTA markedly increased the number of positive cells (Fig. 4 ). These data show that although LTA alone does not induce an oxidative response in PMN, treatment with LTA results in priming of PMN to fMLP-induced oxidative burst.
|
, and G-CSF [30
]. Although IL-8 is produced by a variety of cell types, neutrophil granulocytes are the major source of this proinflammatory cytokine [30
]. The release of IL-8 by neutrophils has been proven as a very sensitive method to detect low-level activation of neutrophils, where the stimulation is not strong enough to activate the oxidative burst [31
]. PMN were incubated for 24 h with various concentrations of LTA. PMN in medium alone released low levels of IL-8. In the presence of LTA, a dose-dependent induction of IL-8 release was observed (Fig. 5 ). The amounts of IL-8 released after exposure to 10 µg/ml LTA were comparable with those measured in PMN cultures after coincubation with 100 ng/ml LPS (not shown).
|
|
and G-CSF as well (Fig. 5)
.
CD14 and TLR2 play a major role in the LTA-mediated activation of PMN
Having established a direct, stimulatory effect of LTA on PMN, we addressed the question of LTA-binding receptors on PMN. In a previous study, CD14 was found to be involved in the LTA-induced effects on monocytes [9
, 10
]. As PMN express CD14, we tested whether blockage of the CD14 receptor affects the response of PMN to LTA. PMN were pretreated for 30 min with the anti-CD14 mAb and subsequently exposed to 10 µg/ml LTA. CD14 blockage prevented the LTA-induced shedding of CD62L (Fig. 7A
) and markedly inhibited the LTA-induced IL-8 release (Fig. 7B)
by human PMN. These data indicate that CD14 plays a major role in the LTA-mediated activation of PMN.
|
Coincubation with LTA inhibits the apoptosis of PMN in a CD14- and TLR2-dependent manner
Neutrophils undergo constitutive apoptosis when aged in vitro. Aging granulocytes exhibit classical features of apoptosis such as cell shrinkage, cytoplasmic condensation, and condensation of nuclear heterochromatin [17
, 18
]. Accordingly, neutrophil apoptosis can be assessed by various parameters including changes in cellular morphology. Using these criteria, the percentage of apoptotic cells was determined in highly purified granulocytes cultured in vitro in the absence or presence of LTA. The ratio of apoptotic cells was strongly reduced when neutrophils were coincubated with LTA (Fig. 8A
). LTA protected neutrophils from apoptosis without evidence of significant necrotic death as assessed by trypan blue exclusion (not shown).
|
An early marker of apoptosis is the appearance of PS on the outer membrane, a process that is called membrane flip-flop [19 ]. PS can be detected by staining with Annexin-V. As shown in Figure 7B , freshly isolated neutrophil granulocytes are PS-negative. Sixty-two percent of the cells became Annexin-V-positive after 24 h of incubation in medium alone (Fig. 8C) . Coincubation with LTA (10 µg/ml) resulted in a marked decrease in the ratio of Annexin-V-positive cells (Fig. 8C) .
Taken together, our data, obtained with morphological analysis, Annexin-V staining, and TUNEL assay, show that LTA protects neutrophils from spontaneous apoptosis. In an independent set of experiments, we could demonstrate that the antiapoptotic effect of LTA lasted for several days. After incubation with 10 µg/ml LTA for 2 days, less than half of the cells were apoptotic, whereas nearly all PMN in the control culture had apoptotic nuclear morphology (Fig. 8A) . Even after 3 days, when all PMN in the control cultures had an apoptotic nuclear morphology, 34% of the LTA-treated cells were still nonapoptotic (Fig. 8A) .
Stimulation of PMN with granulocyte macrophage (GM)-CSF and interferon-
(IFN-) has been reported to delay the PMN apoptosis [34
]. It is more important that after treatment with these cytokines for 3 days, PMN were reported to express major histocompatibility complex class II (MHC-II) molecules [35
], which could be confirmed in our laboratory. We tested whether LTA-mediated activation leads to MHC-II expression on PMN. However, LTA in a concentration of 10 µg/ml, neither alone nor in combination with IFN-, induced MHC-II expression by PMN (data not shown).
As CD14 and TLR2 were found to be involved in the LTA-induced IL-8 release by PMN, blocking experiments were performed to investigate whether the same receptors participate in the antiapoptotic effect of LTA. PMN were pretreated for 30 min with anti-CD14, anti-TLR2, and anti-TLR4 mAb and were subsequently exposed to 10 µg/ml LTA. The rate of Annexin-V-positive apoptotic PMN was determined after 24 h. The antiapoptotic effect of LTA was diminished after blocking CD14 and TLR2 but not TLR4 (Fig. 7C) . These data show that CD14 and TLR2 are involved in the LTA-mediated, antiapoptotic effect on human PMN.
LTA treatment leads to nuclear factor (NF)-B activation in neutrophil granulocytes
The above data clearly demonstrated activation of PMN by LTA. Although the NF-B pathway is a major way of PMN activation [36
], NF-B-independent mechanisms of PMN activation have been also described [37
]. Therefore, we tested whether NF-B was involved in the LTA-induced effects on PMN functions. As a marker for NF-B activation, the amount of P-IB was assessed by Western blotting. Dose-dependent IB phosphorylation was observed in LTA-treated PMN within the time period of 30–150 min (Fig. 9
). Therefore, it is evident that the NF-B pathway is involved in the activation of PMN through LTA.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B.
The highly purified, biologically active LTA preparation used in our present study was previously shown to stimulate the cytokine production of monocytes [9
]. These cytokines include the proinflammatory cytokines TNF-, IL-1ß, IL-6, and IL-8, which are known to possess the capacity to stimulate PMN. The goal of our study was to investigate the direct effects of LTA on PMN. Therefore, indirect effects of LTA mediated by monocytes had to be ruled out; e.g., special care had to be taken to avoid monocyte contamination. Our technique for PMN isolation resulted in a highly pure granulocyte population containing solely two or less monocytes per 10,000 granulocytes. The marked biological effect of LTA on this highly pure PMN population argues strongly against a possible indirect effect mediated by soluble factors released by contaminating leukocytes other then PMN.
In our in vitro studies, LTA concentrations of 0.3–10 µg/ml had detectable, stimulatory activity on PMN functions. This concentration is relatively high as compared with the 200 ng/ml, which activates the release of TNF- in peripheral blood mononuclear cell cultures [10
]. However, similar differences were reported between monocytes and granulocytes concerning their responsiveness to LPS. Whereas 10–100 ng/ml LPS can stimulate monocytes [38
], a concentration as high as 1 µg/ml is the generally used LPS concentration for the direct activation of PMN [39
, 40
]. This is thought to be associated with the low-level expression of the LPS-binding receptors on PMN. For example, only 3300 CD14 molecules were found on a single PMN, whereas over 100,000 were on a monocyte [41
]. As CD14 is also a receptor for LTA [10
], this difference may explain the relatively low response of PMN to LTA.
LTA concentrations can reach high levels at infectious sites with high bacterial load. LTA concentrations up to 10 µg/ml were detected in cultures of S. aureusin vitro [42 ]. Moreover, a previous study showed that the local concentration of LTA and TA can reach a level as high as 26 µg/ml in infected tissues in vivo [43 ]. This is presumably associated with the fact that 107 gram-positive bacteria contain as much as 1 µg LTA, whereas 107 gram-negative contain solely 20 ng LPS [44 ].
In infected tissues, however, LTA represents only one of several stimuli that act on PMN. Additional bacterial products such as fMLP, CpG DNA, and flagellin can all be recognized by pattern-recognition receptors. These various factors often act in a synergistic way, as demonstrated for muramyldipeptide with LTA [45 ]. Moreover, in infected tissues, various cell types, which can act in a synergistic way when acting together or in concert with microbial components such as LPS or LTA, release cytokines [46 ]. Such interactions, however, were not analyzed in our present work. We used a highly purified population (>99.9%) of PMN to investigate the direct effect of LTA on PMN and to rule out that the observed activating effect was a result of indirect activation mediated by LTA-induced products of other cells such as monocytes.
Neutrophil granulocytes contain a large number of granules (hence, granulocytes), which based on their content, can be divided into azurophil, specific, gelatinase granules as well as secretory vesicles. The membrane of specific granules serves as a reservoir of receptors and membrane-bound proteins involved in cell adhesion, signal transduction, and activation of microbicidal pathways [47 ]. The fusion of granules with the cell membrane during degranulation changes the surface expression of these membrane receptors [26 ], and this principle has been used in this study to measure and quantify neutrophil degranulation using flow cytometry. We demonstrated the enhanced cell-surface expression of CD11b and CD66b on PMN upon exposure to LTA. Using this approach, we could clearly show that treatment with LTA results in PMN granule exocytosis, which is a clear sign of PMN activation.
After being recruited to the site of infection, PMN recognize and phagocytose gram-negative [48 ] and -positive bacteria [49 ], resulting in the activation of these antimicrobial effector cells. PMN activation includes a broad range of phenotypic and functional changes, which occur in several stages [26 , 50 ]. A low level of activation leads to events such as the modulation of receptor expression on the cell surface and release of cytokines. Such low-level activation results also in the "priming" of PMN [25 ]; i.e., the cells become highly responsive to subsequent stimuli. Functions that require higher levels of activating stimuli include the oxidative burst. Some activating agents cannot evoke all functions [31 ]. LPS, for example, as a single stimulus does not have the capacity to induce oxidative burst, albeit it can induce activation-associated changes in the surface-receptor expression, prime neutrophils for enhanced induction of ROS by another activating stimulus, and mediate cytokine release [25 ]. LTA treatment alone did not activate a full-blown oxidative burst in PMN but resulted in the priming of PMN for fMLP-induced production of ROI.
We presented data that LTA treatment leads to release of cytokines from PMN. After pathogen challenge, IL-8 is one of the most important chemotactic factors that mediate local neutrophil recruitment [30
]. The autocrine production of IL-8 by activated PMN is regarded as an amplifying loop to attract more neutrophils to the site of infection [31
]. Intracellular staining in PMN revealed that treatment with LTA results not only in the release of preformed IL-8 but stimulates the new synthesis of this neutrophil-derived chemokine as well. IL-8 is a prototype of proinflammatory cytokines under NF-B regulation [51
]. Measuring IB phosphorylation as a read-out system, rapid activation of NF-B was observed in PMN upon exposure to LTA. In addition to IL-8, LTA treatment also induced the release of G-CSF and TNF-, which is a further, additional sign of the activated state of PMN.
PMN are inherently short-lived cells with a half-life of only 6–10 h, after which they undergo spontaneous apoptosis [17
]. Although apoptosis is an intrinsic cell process, it is modulated by signals from the local micromilieu. The lifespan of mature neutrophils can be extended in vitro by incubation with proinflammatory cytokines including GM-CSF and G-CSF, IL-8, and IL-1ß and bacterial products such as LPS and fMLP [34
, 52
]. This implies that PMN die rapidly via apoptosis if not engaged in function. After activation, however, inflammatory PMN have an extended lifespan that enables these important cells of the innate immune system to execute their antimicrobial effector functions. In this study, we demonstrated for the first time that LTA, a component of gram-positive bacteria, can directly delay the spontaneous apoptosis of PMN, resulting in an extended lifespan of these cells. In case of cytokine-induced delay of PMN apoptosis, NF-B activation was shown to be involved [53
]. As LTA activates NF-B in PMN very rapidly, this way is also likely to play a role in the LTA-induced modulation of programmed cell death in PMN.
IL-8 and G-CSF, which are released after exposure to LTA, could mediate the observed delay of spontaneous PMN apoptosis. However, the levels of these cytokines in the PMN cultures were much lower as compared with concentrations that were reported to inhibit PMN apoptosis [54 , 55 ]. Therefore, the delay in spontaneous PMN apoptosis is likely a direct effect of LTA rather than being mediated by LTA-induced cytokines.
Using blocking experiments with specific mAb, here, we showed for the first time the CD14- and TLR2-dependent, direct effect of LTA on PMN. Blocking CD14 in primary human monocytes was previously shown to diminish LTA-mediated effects [56 ]. In contrast to monocytes, CD14 is expressed at a low level on PMN [41 ]. The strongly attenuated effect of LTA on PMN after CD14 blockade shows, however, that this molecule plays a major role in the LTA-mediated effects on PMN.
Here, we showed that in primary human neutrophils, in addition to blocking CD14, treatment with anti-TLR2 antibody also reduced a LTA-induced cellular response. Transfection studies with human embryonic kidney and Chinese hamster ovary cells revealed that CD14 and TLR2 participate in the response to LTA [56 , 57 ]. However, few previous reports dealt with the effect of different TLR-2 ligands on PMN. In these works, TLR2 agonists such as peptidoglycan, zymosan, and tripalmitoyl-cyteinyl-seryl-(lysyl)3-lysine (Pam3CSK4) were studied [46 , 58 ], but the effect of LTA on PMN was not investigated. In one study, the stimulation of PMN with the TLR2 agonists zymosan, peptidoglycan, and nonmannose-capped lipoarabinomannan resulted in a release of low amounts of IL-8 by PMN but did not trigger the generation of ROI [46 ]. Hayashi et al. [58 ] showed a stimulatory effect of TLR ligands zymosan and Pam3CSK4 on PMN. However, the low purity of granulocytes in the mentioned study makes it difficult to say whether the observed effect was a result of the direct interaction with PMN. Studies with TLR2 ligands such as Pam3CSK4, however, may be of limited value regarding the role of LTA, as LTA and Pam3CSK4 were recently shown to have distinct binding domains on the TLR2 molecule [59 ]. The importance of the response of PMN to LTA is highlighted by the fact that in addition to peptidoglycan, LTA is the main immunostimulatory cell-wall component of gram-positive bacteria [60 ], which are responsible for an increasing number of systemic infections [61 ].
Ligand specificity for a number of TLR2 activators requires heterodimerization with additional TLR molecules, TLR1 and TLR6 [62 , 63 ]. Studies suggest that TLR2/TLR1 heterodimers mediate responses to Pam3CSK4. It is still to be clarified whether LTA binds to TLR2/TLR1 or TLR2/TLR6 heterodimers. TLR2/TLR1 and TLR2/TLR6 heterodimers are capable of inducing differential cellular responses [64 , 65 ]. In line with these data, different TLR2 agonists were shown to induce distinct effects on human mast cells [66 ].
Using blocking experiments with specific mAb, here, we showed for the first time that CD14 and TLR2 are involved in the LTA-mediated, antiapoptotic effect on human PMN. Interaction with distinct binding domains on TLR2 may explain the apparent discrepancy of our present finding with a previous report, in which TLR2 activation via Pam3CSK4 did not prevent neutrophil apoptosis [67 ]. Moreover, no direct comparison of the LTA effect with that of other TLR2 ligands can be made, as for example, the biological activity of Pam3-Cys-Ser-Ser-Asn-Ala was shown to be independent of CD14 [68 ].
Taken together, LTA has a direct stimulatory activity on the functions of human PMN, which when exposed to LTA, show all typical features of activation, i.e., elongated cell shape, granule degranulation, shedding of CD62L, release of IL-8, TNF-, and G-CSF, a primed state for fMLP-induced oxidative burst, and an extended lifespan. The LTA-induced stimulation of PMN involves CD14, TLR2, and the activation of NF-B. These data suggest that LTA is a potent component of gram-positive bacteria in terms of recognition by and stimulation of PMN.
|
ACKNOWLEDGEMENTS |
|---|
Received August 1, 2003; revised October 28, 2003; accepted November 12, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
release Eur. J. Immunol. 32,541-551[CrossRef][Medline]
RIII and acquire Annexin V binding sites during apoptosis in vitro Blood 85,532-540m ß2) in human neutrophils J. Clin. Invest. 92,1467-1476
, and IL-3 J. Immunol. 151,1482-1490[Abstract]
B pathway by inflammatory stimuli in human neutrophils Blood 89,3421-3433B and AP-1 activation, and IL-8 gene expression in human leukocytes Proc. Natl. Acad. Sci. USA 99,13266-13271B translocation is required for TRAF-1 synthesis Shock 14,290-294[Medline]
-mediated apoptosis of human neutrophils Kidney Int. 53,84-91[CrossRef][Medline]
