Teichoic Acids Are Not Required for Streptococcus pneumoniae and ...

2 downloads 2 Views 123KB Size Report
Jan 22, 2003 - stimulating discussion, Michael Potter of the National Cancer Institute for the gift of the TEPC-15 myeloma and ascites, and Marlies Knaup.

INFECTION AND IMMUNITY, July 2003, p. 3707–3713 0019-9567/03/$08.00⫹0 DOI: 10.1128/IAI.71.7.3707–3713.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Vol. 71, No. 7

Teichoic Acids Are Not Required for Streptococcus pneumoniae and Staphylococcus aureus Cell Walls To Trigger the Release of Tumor Necrosis Factor by Peripheral Blood Monocytes P. A. Majcherczyk,1 E. Rubli,1 D. Heumann,2 M. P. Glauser,2 and P. Moreillon1* Institute of Fundamental Microbiology, University of Lausanne, 1015 Lausanne,1 and Division of Infectious Diseases, Department of Internal Medicine, Centre Hospitalier Universitaire Vaudois, 1011 Lausanne,2 Switzerland Received 22 January 2003/Returned for modification 27 February 2003/Accepted 9 April 2003

In gram-negative bacteria, the outer membrane lipopolysaccharide is the main component triggering cytokine release from peripheral blood mononuclear cells (PBMCs). In gram-positive bacteria, purified walls also induce cytokine release, but stimulation requires 100 times more material. Gram-positive walls are complex megamolecules reassembling distinct structures. Only some of them might be inflammatory, whereas others are not. Teichoic acids (TA) are an important portion (>50%) of gram-positive walls. TA directly interact with C3b of complement and the cellular receptor for platelet-activating factor. However, their contribution to wall-induced cytokine-release by PBMCs has not been studied in much detail. In contrast, their membranebound lipoteichoic acids (LTA) counterparts were shown to trigger inflammation and synergize with peptidoglycan (PGN) for releasing nitric oxide (NO). This raised the question as to whether TA are also inflammatory. We determined the release of tumor necrosis factor (TNF) by PBMCs exposed to a variety of TA-rich and TAfree wall fragments from Streptocccus pneumoniae and Staphylococcus aureus. TA-rich walls from both organisms induced measurable TNF release at concentrations of 1 ␮g/ml. Removal of wall-attached TA did not alter this activity. Moreover, purified pneumococcal and staphylococcal TA did not trigger TNF release at concentrations as high as >100 ␮g/ml. In contrast, purified LTA triggered TNF release at 1 ␮g/ml. PGN-stem peptide oligomers lacking TA or amino-sugars were highly active and triggered TNF release at concentrations as low as 0.01 ␮g/ml (P. A. Majcherczyk, H. Langen, et al., J. Biol. Chem. 274:12537-12543,1999). Thus, although TA is an important part of gram-positive walls, it did not participate to the TNF-releasing activity of PGN. 20, 27, 30, 46, 49, 52). Indeed, it contains very-much-conserved motifs that might reveal a common inflammatory pathway for gram-positive bacteria. Previous studies revealed both similarities and differences between cytokine release mediated by gram-negative LPS and gram-positive PGN. Like LPS, soluble PGN prepared from penicillin-treated S. aureus can bind to the CD14 receptor of target cells (10, 37, 49, 50), and even its minimal immunomodulatory subunit muramyl-dipeptide can do so (50). However, CD14 contains discrete binding sites that are different for LPS and PGN (10, 18). Moreover, although soluble CD14 binds both of these bacterial molecules, activation of CD14deficient endothelial and epithelial cells depends on soluble CD14 for LPS but not for PGN (23). Finally, unlike LPS, PGN does not bind to LBP (28) and does not activate intracellular signal transcription via TLR4 (39). It uses its TLR2 homologue instead (39). It also generates a unique pattern of gene expression when human monocytes are tested by the technique of cDNA arrays (48). Thus, LPS and PGN do not entirely share similar host recognition mechanisms. Attempting to identify essential inflammatory components of the S. pneumoniae PGN, we recently observed that oligomeric stem peptides—but not simple monopeptides or dipeptides—devoid of their aminosugars were highly inflammatory and sufficient to trigger the release of tumor necrosis factor (TNF) from PBMCs (27). This suggested that some minimal structural constraints of the PGN network were sufficient to activate inflammation, with the rest of the wall being more inert with regard to this reaction. However, this view might be

Inflammation and septic shock are features of both gramnegative and gram-positive bacterial infection (41). In gramnegative bacteria, the major molecule triggering cytokine-release from peripheral blood mononuclear cells (PBMCs) and other responsive cells is the outer membrane lipopolysaccharide (LPS) (5, 29, 35). LPS-mediated cytokine-release results from a cascade of events, starting with its binding to the serum acute-phase LPS-binding protein (LBP). The LBP-LPS complex then attaches to the glycerophosphate-inositol (GPI)linked CD14 receptor of target cells, which eventually become activated via the transmembrane coreceptor TLR4 (31, 43, 52). In contrast to gram-negative organisms, gram-positive bacteria do not contain LPS. However, they produce a variety of extrinsic and intrinsic molecules that may trigger inflammation. Staphylococcus aureus and Streptococcus pyogenes may secrete superantigens, which nonspecifically activate T lymphocytes (33, 35), and Staphylococcus epidermidis were recently shown to produce toxin-like inflammatory polypeptides (32). Components of the cell envelope, such as peptidoglycan (PGN) and lipoteichoic acids (LTA), may trigger cytokine release from a variety of host cells (4, 20, 24, 25, 30, 42, 45). Even unprocessed bacterial proteins containing formylmethionine amino-terminal (14) and specific DNA (40) motifs can promote granulocyte migration and cytokine production. Among these, the gram-positive PGN has attracted much attention (9, 10, 17, * Corresponding author. Mailing address: Institute of Fundamental Microbiology, University of Lausanne, 1015 Lausanne, Switzerland. Phone: 41-21-692-5601. Fax: 41-21-692-5605. E-mail: Philippe [email protected] 3707



too reductive. Far from being a static protection against osmotic explosion, the PGN is a dynamic scaffold and the site of attachment of multiple surface molecules, which act as the interface between the environment and the bacterial cells (44). In the majority of cases the cell wall preparations tested for their inflammatory activity in the laboratory do not reflect this complexity. During wall purification, the loosely attached polysaccharides (including the capsule) and the numerous cell wall proteins are removed (16) and thus never analyzed for their inflammatory power. Only PGN and its covalently attached teichoic acids (TA) withstand such harsh treatment. Although TA represent up to 60% of the purified wall’s dry weight, little is known of their participation in inflammation (20, 30, 36). Moreover, TA vary between different organisms and thus might be important for the host response. They are mainly constituted of either polyribitol-phosphate or polyglycerol-phosphate polymers and may be decorated with various side chains, including phosphorylcholine in S. pneumoniae and D-alanine in S. aureus and other organisms (13, 22). The TA structurally resemble their membrane-bound LTA analogs, which were shown to synergistically interact with PGN subcomponents for nitric oxide release in vitro and in animal experiments (25, 26). Therefore, the question arises as to whether TA might also be important for inflammation. The following experiments tested the TNF-releasing activity of both S. pneumoniae and S. aureus TA used alone or in the presence of purified cell walls. Various preparations of cell wall, TA, and LTA were used. MATERIALS AND METHODS Bacterial strains and growth conditions. S. aureus strain COL (34) was grown in tryptic soy broth and the noncapsulated lysis-prone S. pneumoniae strain R6 (2) was grown in the chemically defined liquid medium Cden (A. Tomasz, unpublished data). Growth of the cultures was monitored by determining their optical density at 620 nm with a spectrophotometer (Pharmacia Biotech, Uppsala, Sweden). The cells were stored at ⫺70°C in medium containing 20% glycerol. In certain experiments a radioactive label was introduced into the cell walls by either of the following methods: (i) to label the stem peptides, L-[4,5-3H]lysine (Amersham Pharmacia Biotech, Inc., Uppsala, Sweden) was added to the Cden medium at final concentrations of 500 mCi/liter and 0.9 mg/liter; (ii) to label the glycan chains, N-acetyl-D-[1-3H]glucosamine was added at final concentrations of 500 mCi/liter and 2.46 mg/liter; (iii) to label pneumococcal TA, [methyl-3H]choline chloride (Amersham) was added at final concentrations of 500 mCi/liter and 2.5 mg/liter. The cultures were allowed to grow for four to five generation times in the presence of the label before being harvested, and their cell walls purified as described below. Escherichia coli O111 LPS (Sigma Chemical Co., St. Louis, Mo.) was used as a stimulant for comparison to gram-positive wall constituents. Preparation of purified walls, TA, and LTA. Cell walls were prepared as previously described (27). To remove the teichoic acids, purified walls were suspended in 49% hydrofluoric (HF) acid at a concentration of 1 mg/ml and stirred gently at 4°C for 48 h. The PGN was recovered as previously described (8, 19), precipitated twice in acetone, resuspended in endotoxin-free water, and dried by rotary evaporation. For pneumococcal walls, the loss of TA was monitored by determining the decrease in the radioactivity of [3H]choline-labeled TA in the material. For staphylococci, the reduction was measured both by determining the loss in mass and in phosphate concentration as measured by a colorometric test (1). TA were prepared from S. pneumoniae ([3H]choline labeled) and S. aureus (unlabeled) cell walls by trichloroacetic acid extraction, followed by precipitation with ethanol as previously described (18, 30). S. aureus LTA was a generous gift from Siegfried Morath (University of Konstanz, Konstanz, Germany). Solubilization of pneumococcal walls and purification of stem peptides and TA-rich and TA-free wall fragments. Purified walls were solubilized with the natural pneumococcal autolysin N-acetylmuramyl-L-alanine amidase (referred to as amidase in the following) as described previously (27). After solubilization,

INFECT. IMMUN. peptide-rich and glycan-rich fractions were separated by differential precipitation in a mixture of water–acetonitrile–propan-2-ol (at the ratio of 50/25/25) containing 0.1% trifluoroacetic acid (15). Typically, 600 ␮g of amidase-solubilized cell walls were suspended in this solvent and kept on ice for 30 min. The glycan-rich fraction is poorly soluble in this solvent and was collected by centrifugation at 15,000 ⫻ g for 15 min (4°C). The supernatant contained the stem peptides. Solvents were removed by rotary evaporation, and the peptides were dissolved in water. The amount of material present in each fraction was quantified by liquid scintillation counting. Digested walls were precipitated twice in acetone to remove possible contamination by endotoxin (27). The peptide-rich fraction was used to purify stem peptides. Stem peptides were separated by high-pressure liquid chromatography (HPLC) as previously described (27). The system consisted of the L-7200 autosampler; the L-7100 gradient pump, with low-pressure mixing; and the L-7400 UV detector (Hitachi Instruments, Ichige, Hitachinaka, Japan). Column temperature was maintained at 25°C by using the L-7360 column oven. The results were analyzed by using the D-7000 HPLC system manager program (Hitachi). Separation was performed by injection of a 100-␮l sample, containing 100 ␮g of wall digest, into a C18 reversed-phase column (SuperPac Sephasil C18; 5 mm, 4-by-250-mm column; Pharmacia Biotech, Uppsala, Sweden), protected with a guard cartridge (C18, 5 mm, 4 ⫻ 10 mm). The mixture was separated by using a linear gradient of 0 to 15% acetonitrile in 0.1% trifluoroacetic acid over 100 min at a flow rate of 0.5 ml/min. Detection was at 210 nm. One-minute fractions (500 ␮l) were collected, and the radioactivity in 100 ␮l was measured by liquid scintillation counting the Ultima Gold scintillant (Packard Instrument B. V. Chemical Operations, Groningen, The Netherlands). The glycan-rich fraction was used to purify TA-rich and TA-free fragments. TA-rich and TA-free fragments were separated by immunoaffinity chromatography as described previously (12). Briefly, the glycan-rich fraction of amidasedigested walls was dissolved in 0.1 M sodium acetate (pH 6.7) and loaded onto an immunoaffinity column containing the TEPC-15 antibody. This antibody selectively retains fragments containing choline-decorated TA. The column was washed until no further radioactivity eluted from the column, and the washings were collected. Bound material was eluted with 10 mM phosphorylcholine chloride in ammonium acetate. In order to assess the possibility of column overload, the unbound fractions were pooled and reloaded onto it. Collected fractions were dried by rotary evaporation, precipitated twice in acetone to remove any possible endotoxin contaminants, dissolved in water, and used to stimulate monocytes as described below. Preparation, stimulation of human PBMCs, and measurement of TNF-␣ release. Human PBMCs were extracted from heparinized blood of healthy volunteers by Ficoll-Hypaque (Seromed, Munich, Germany) density gradient centrifugation as described previously (20). The cells were suspended in RPMI 1640 medium (Gibco Laboratories, Basel, Switzerland) and distributed into the wells of a flat-bottom 96-well tissue culture plate at a concentration of 0.5 ⫻ 106 cells per well. Each well contained a final volume of 200 ␮l, which comprised the RPMI medium (140 ␮l), plasma from the donor (20 ␮l), and sample (20 ␮l). LPS from E. coli O111 (Sigma) in the concentration range of 0.01 to 100 ng/ml was used as a positive control. PBMCs incubated with plasma or with medium alone were used as negative controls. The plates were incubated at 37°C in an atmosphere of 5% CO2. After 8 h of incubation, samples (20 ␮l) of the supernatants were taken, diluted 20-fold in RPMI medium and stored at ⫺80°C for measurement of TNF-␣ concentrations. A number of control experiments with this material in the presence or the absence of polymyxin B indicated that it was not contaminated with LPS (data not presented). WEHI clone 13 murine fibroblast cells (104 cells/well) were used for quantitation of TNF-␣ as described previously (3). Recombinant mouse TNF-␣ were used as a standard. The sensitivity of the assay was 25 pg/ml.

RESULTS TNF release from PBMCs exposed to purified S. pneumoniae and S. aureus cell walls. TNF release by TA-rich purified cell walls was concentration dependent as previously described (27). The stimulatory activity was expressed as the minimum concentration of material required to induce a ⱖ10-fold increase in TNF levels above background (minimum stimulatory concentration, 10-fold [MSC10⫻]). Both S. pneumoniae and S. aureus walls were equivalently active and required 0.1 to 1 ␮g of material/ml to stimulate the monocytes. Although inflam-

VOL. 71, 2003

FIG. 1. Steps used in the dissection of the pneumococcal wall in order to investigate the role of TA in the induction of TNF release from PBMCs. Radiolabeled pneumococcal walls were prepared from S. pneumoniae grown in the presence of either [3H]lysine, [3H]GlcNAc, or [3H]choline, resulting in the labeling of peptides (vertical lines), glycan chains (horizontal lines), or TA (black circles), respectively. TA were removed from the insoluble walls by HF acid treatment (panel A, step 2). In order to isolate the native TA-rich and TA-free constituents of the cell wall, the walls were digested with amidase (panel B, step 2), which hydrolyzed the bonds between the stem peptides and the glycan chains wherever an adjacent TA was present. The glycan- and peptiderich fractions were separated by differential precipitation (step 3). The glycan-rich fraction was then separated into its TA-containing TA-free constituents by affinity chromatography by using antibodies against the phophorylcholine moiety of TA (step 4). The proinflammatory activity of these constituents was then assessed in parallel. The peptide-rich fraction was separated by HPLC (step 4), and the activity of their eluted material was assayed (see the text for details).

matory, this concentration was up to 1,000 times greater, in a weight/weight ratio, than equally inflammatory gram-negative LPS preparations (data not presented). To determine the roles of TA in these preparations, the two types of walls were separately processed for TA removal, and the resulting TA-rich and TA-free materials were compared for their inflammatory activity. Effect of TA removal from S. pneumoniae purified walls. Figure 1 illustrates the various experimental steps and products used in this experiment. In a preliminary study (Fig. 1A), TA were removed from the walls by HF acid (8, 19). To assess



the efficacy of this procedure, the walls were first biosynthetically labeled with [3H]-choline—a tracer that specifically incorporates in the pneumococcal TA—and then tested for the loss of radioactivity due to acid treatment. The specific activity (in counts per minute [cpm]) before HF acid treatment was of 103 cpm/␮g of wall. After HF acid treatment, this value dropped to ⬍10 cpm/␮g, which was equivalent to background counts. This ⬎99% loss of TA corresponded to a parallel loss of 60% of the original dry weight of the preparation, which was in agreement with a previous study (13). Thus, TA represent a substantial proportion of the purified pneumococcal cell wall. Both the TA-rich and TA-free preparations were evaluated for their MSC10⫻, as described above (27). Table 1 indicates that, in spite of HF acid treatment, both types of walls had very similar inflammatory power. In a second study (steps 1 to 4 in Fig. 1B), the purified walls were first solubilized with the major pneumococcal autolysin amidase and the glycan-enriched fraction, containing the TA, was isolated by differential precipitation (15). The fraction contained ca. 18% of the total peptides. These peptides had either remained attached to the sugar backbone after amidase digestion or had coprecipitated during the purification process (11, 15). The glycan-rich fraction was further partitioned in TA-rich and TA-free fragments by affinity chromatography on a TEPC-15 column retaining choline-decorated residues (step 4 in Fig. 1B) (16). During TEPC-15 affinity chromatography, the amounts of materials eluting from the column was monitored by using glycans labeled with 3H-labeled N-acetylglucosamine. Eightynine percent of the radioactivity was retained by the column (choline-rich), whereas 11% did not attach to it (choline-free). The unbound material did not attach to the column upon a second passage, confirming that the experimental conditions were not saturating. The two fractions were tested for their TNF-releasing activ-

TABLE 1. Relative stimulatory potency of LPS and various pneumococcal wall components exposed to human PBMCs Putative structure


E. coli 0111 LPS

MSC10⫻a (␮g/ml)


Pneumococcal purified walls Insoluble TA-rich walls


Insoluble TA-free walls


TA-free glycans


TA-rich glycans


Stem peptide monomers and dimersb Stem peptide trimers and polymers a


ⱖ0.1 ⬍0.01

That is, the minimum concentration of material required to induce a ⱖ10fold increase in TNF levels above the background level. Values are the means of at least three experiments. b HPLC-separated peptides 1 to 10. c HPLC-separated peptides 9 to 21.



ity. Although they contained glycans variously tailored regarding their TA and residual peptides contents (step 4 in Fig. 1B), these two fractions had similar specific TNF-stimulating activities (Table 1). Finally, since we previously described that the pneumococcal peptide-enriched fraction on its own contained highly proinflammatory components (27), we further tested whether certain of these components could contain contaminating TA and/ or glycan constituents. Walls biosynthetically labeled with either [3H]lysine, [3H]-N-acetylglucosamine, or [3H]choline were digested with amidase, and these peptide-enriched fractions processed for HPLC separation as described previously (steps 3 to 4⬘ in Fig. 1B) (27). This so-called peptide-enriched fraction still contained ca. 56% of the total radioactivity due to glycans ([3H]-N-acetylglucosamine) and TA ([3H]choline), respectively. This is in accordance with the fact that amidase hydrolyzes ca. 50% of the muramyl-peptide amide bonds in the wall (16). After HPLC separation, the various fractions were compared both for their proinflammatory activity on PBMCs and for their content in the specific radiotracers of peptides, glycans, and TA (Fig. 2). The results are depicted in Fig. 2. Figure 2A confirms that the peptide-enriched fraction could be resolved by HPLC in a series of components carrying either no proinflammatory activity (UV peaks 1 to 8) or high proinflammatory activity (UV peaks 9 to 21) (27). Figure 2B indicates that the majority of UV peaks corresponded to [3H]lysine-containing fractions and thus contained wall peptides. Figures 2C and D indicate that, although the UV peaks contained radioactively labeled peptides, they did not contain labeled glycans or TA. Most (96%) of both [3H]-N-acetylglucosamine (Fig. 2C) and [3H]choline (Fig. 2D) radioactivity was retained on the column and did not appear on the peptide chromatogram. It required stringent conditions to be recovered. In contrast, most (⬎80%) of the [3H]lysine radioactivity was recovered in the HPLC chromatogram presented in Fig. 2B. Thus, the active peptide peaks, which carried ⱖ100 times more TNF-releasing power than whole walls or purified PGN, were not contaminated by components of the wall glycan backbone. Taken together, these results indicate that TA did not significantly contribute to the proinflammatory activity of purified pneumococcal cells walls, whether they were presented to the PBMCs in gross insoluble form or in more subtle soluble fragments. Effect of TA removal from S. aureus purified walls. To test whether the observation described above was specific to the peculiar S. pneumoniae TA or whether it could also apply to other pathogens, we repeated the basic experiment illustrated in Fig. 1A with S. aureus walls from the methicillin-resistant S. aureus strain COL. As for pneumococci, 48 h of treatment of purified S. aureus walls with HF resulted in a loss of 55% of their original dry weight, and there was a ⬎99% decrease in the phosphate content. Longer incubation with HF did not solubilize more material. However, in spite of TA removal, the proinflammatory activities of the pre-HF and post-HF preparations were similar (Table 2). Thus, as for S. pneumoniae, TA removal from insoluble S. aureus walls did not affect their proinflammatory activity. TNF release activity of purified TA, purified LTA, and purified walls combined or not with TA and LTA. The inflam-


matory activity of purified TA and LTA was determined alone or when combined with purified walls. Table 2 indicates that purified TA were at least 1,000 times less active than purified walls in a weight/weight ratio. This lack of activity was observed with both pneumococcal and staphylococcal TA. In contrast, purified LTA, which originated from S. aureus, had an inflammatory activity that was equivalent to that of purified walls. Table 2 also presents the TNF-triggering activity of purified calls combined or not combined with exogenous TA or LTA. In the present experiments, no additive or synergistic effect was observed. DISCUSSION The present experiments tested the implication of TA in the ability of pneumococcal and staphylococcal purified cell walls to trigger cytokine release from PBMCs. The question is relevant because TA are part of the outermost layer of the grampositive envelope and thus are likely to interact preferentially with the constituents of the host. As far as the innate immunity is concerned, TA from pneumococci were shown to bind the C3b opsonic component of the complement (21), as well as to the cellular receptor for platelet-activating factor (7). However, only a few studies attempted to determine their contribution to cell wall-induced release of inflammatory cytokines (20, 30, 36). In contrast, LTA—the membrane-bound counterpart of TA—demonstrated proinflammatory activity in several studies (4, 24, 25, 25, 47). The intrinsic inflammatory power of these molecules was relatively low. It was comparable to that of PGN and 100 to 1,000 times lower than that of the gram-negative LPS control (25). On the other hand, LTA combined with PGN subcomponents as simple as muramyl-dipeptides demonstrated a synergism in stimulating the release of NO by the macrophage cell line, J774.2 cells (25). This suggests that some cooperation between these two intrinsically poorly active compounds might be important for recognition by the innate immune system. Previous studies indicated that the structural integrity of LTA was critical for this activity (4, 24). LTA is a glycolipid consisting of one hydrophilic TA chain linked to four fatty acids that serve as a membrane anchor (13). Cytokine release was substantially decreased by the loss of two of the fatty acids and completely abolished by the loss of all four of them (4, 24). In the view of this information, it was interesting to test the impact of PGN-attached TA on the inflammatory power of purified cell walls. As LTA, the TA-PGN complex contains a hydrophilic TA moiety. Unlike LTA, on the other hand, this moiety is linked to a PGN scaffold instead of a membrane anchor. Thus, the question arises as to whether this PGN scaffold can replace the membrane-anchor moiety in inflammatory reactions. We addressed this question in two types of organisms carrying structurally different TA. Several methods were used to separate TA-rich and TA-free components of the purified walls. The results did not detect any impact of TA on the wall proinflammatory activity. This was true whether the material used was insoluble or whether it was solubilized and separated in TA-rich and TA-free glycan-enriched strands, as demonstrated with pneumococcal walls. Moreover, TA alone showed

VOL. 71, 2003



FIG. 2. Separation of radiolabeled pneumococcal cell wall stem peptides by HPLC. The solid line shows the chromatogram obtained with UV detection. One-minute fractions were collected. The crossed line in panel A shows the TNF released after stimulation of PBMCs by the collected fractions. Major peaks were numbered according to Garcia-Bustos et al. (15a). Some peaks had no stimulatory activity (1 to 5, inactive region), whereas others had a high stimulating activity (6 to 9, active region). The crossed lines in panels B, C, and D show the chromatograms obtained with liquid scintillation detection. (B) Separation of stem peptides labeled with [3H]lysine gave UV and radioactivity chromatograms that were superimposable, confirming the presence of this amino acid in the resolved components. (C and D) However, the chromatogram obtained with UV detection did not follow the chromatogram obtained with radioactivity detection for cell walls labeled either with [3H]GlcNAc or [3H]choline, confirming the absence of substantial amounts of the sugars (C) and TA (D), respectively, from the resolved components. TNF and radioactive labeling datum points are aligned at start of the 1-min fractions. When medium or diluents alone are added, no TNF was released.

very low inflammatory activity compared to purified LTA, which triggered cytokine release, as reported elsewhere (6, 30). In the present experiments neither TA nor LTA synergized with PGN. This could be due to differences in the readout

systems and/or the tested materials. The cells and cytokines used in the present experiments were different from those used in previous work (25). The bacterial material was also different. It was composed of whole mature cell walls in the present study




TABLE 2. Effect of TA and LTA on the relative stimulatory potency of pneumococcal and staphylococcal cell walls exposed to human PBMCsa


Pneumococcal purified walls Insoluble TA-rich walls Insoluble TA-free walls TA Staphylococcal purified walls Insoluble TA-rich walls Insoluble TA-free walls TA LTA

MSC10⫻ (␮g/ml)

MSC10⫻ (␮g/ml) of insoluble walls combined with LTA or TA LTA (0.1 ␮g/ml)b

TA (10 ␮g/ml)c

0.1 0.1

0.1 0.1



0.1 0.1

0.1 0.1







0.1 0.1 ⱖ100 0.1 0.1


Details are as as described in Table 1. LTA purified from S. aureus (kindly provided by S. Morath). TA purified from S. pneumoniae and S. aureus, respectively. d ND, not determined. b c

compared to the nontranspeptidated soluble PGN released during penicillin treatment in earlier experiments (25). Such structural differences might account for the somewhat different observations. On the other hand, oligomeric stem peptides recently described in pneumococcal PGN (27) were highly inflammatory and clearly devoid of TA or polysaccharides. Thus, the high proinflammatory power of PGN constituents does not require TA as an adjuvant for inflammation. Although these results reasonably rule out a significant inflammatory role of TA in the present experimental conditions, other studies suggested that TA might be important for inflammation (30, 36). First, Mattson et al. (30) reported that S. epidermidis TA could trigger the release of TNF, interleukin-1 (IL-1), and IL-6 by human monocytes. However, the cytokinereleasing activity of TA was 10 times lower than that of PGN, which itself had a low inflammatory activity. Second, Riesenfeld-Orn et al. (36) reported very high IL-1releasing activity by TA-rich, but not TA-free soluble fragments of purified pneumococcal walls. A priori, this finding is in contradiction with the present results. However, beyond the difference in the main cytokine tested (IL-1 in the study by Riesenfeld-Orn et al. versus TNF in the present work), the PGN cutout between the two experiments was quite different. Riesenfeld-Orn et al. solubilized the wall with a glycosidase (M1 muramidase) and isolated relatively large TA-rich fragments (ca. 20,000 kDa) (16). Such fragments were likely to contain representatives of the highly active oligomeric stem peptides recently described in pneumococci (27). Moreover, these fragments were compared to TA-free simple disaccharide-peptide monomers and dimers, which do not carry intrinsic inflammatory activity (10, 27, 42). Thus, the present study did not compare strictly similar structures containing or not containing TA. In the present experiments, TA-rich and TAfree walls were made up of the same basic PGN scaffold that was stripped or not from its decorating TA by HF acid treatment. Further different stimulatory tests were used in these

studies. Riesenfeld-Orn et al. (36) used PBMCs in the absence of plasma, whereas the present study was conducted in the presence of plasma. Plasma has been shown to enhance the stimulatory activity of cell walls when exposed to PBMCs (20). Our results indicate that TA are unlikely to interfere with PGN-induced release of cytokines, at least within the limits of this experimental system. The observation is compatible both with the poor intrinsic inflammatory activity of purified TA (30) and the lack of synergism between PGN and deacylated LTA (24). On the other hand, TA are implicated in the hostparasite relationship by numerous other means. Complement and platelet-activating factor binding is one of them. Facilitating autolysis is another. In pneumococci, TA are decorated with choline residues that are critical for the activity of the major autolysin amidase (13). Amidase-induced bacterial lysis results in the release of numerous bacterial products, some of which are known as pathogenic determinants, such as pneumolysin (38), and others of which are highly inflammatory, such as PGN subcomponents and intracellular determinants (27). A possible scenario is that live bacteria keep hiding from host defenses by burying their proinflammatory components inside the intact cell. Autolysis would be dangerous to them because the release of proinflammatory molecules would encourage recognition by immunocompetent cells. This idea is compatible with the observation that pneumococci undergo phase variation in the production of TA between colonization and invasive infection (51). In invasive infection, pneumococci switch to low TA production, a phenotype that confers increased resistance to autolysis. Thus, although not directly involved in cytokine release, TA play other major roles in the host-parasite relationship, the subtlety of which is only beginning to be understood. ACKNOWLEDGMENTS P.A.M. and E.R. contributed equally to this study. This work was supported by the Fondation Leenards and grant 3200-063253.00 of the Swiss National Fund for Scientific Research. We are grateful to Alex Tomasz of the Rockefeller University for stimulating discussion, Michael Potter of the National Cancer Institute for the gift of the TEPC-15 myeloma and ascites, and Marlies Knaup for outstanding technical assistance. REFERENCES 1. Ames, B. N., and D. T. Dubin. 1960. The role of polyamines in the neutralization of bacteriophages. J. Biol. Chem. 235:769–775. 2. Avery, O. T., C. M. MacLeod, and M. McCarty. 1944. Studies on the chemical nature of the substance inducing transformation of pneumococcal types: induction of transformation by a desoxyribonucleic acid fraction isolated from pneumococcus type III. J. Exp. Med. 79:137–158. 3. Baumgartner, J. D., D. Heumann, J. Gerain, P. Weinbreck, G. E. Grau, and M. P. Glauser. 1990. Association between protective efficacy of anti-lipopolysaccharide (LPS) antibodies and suppression of LPS-induced tumor necrosis factor alpha and interleukin 6: comparison of O side chain-specific antibodies with core LPS antibodies. J. Exp. Med. 171:889–896. 4. Bhakdi, S., T. Klonisch, P. Nuber, and W. Fischer. 1991. Stimulation of monokine production by lipoteichoic acids. Infect. Immun. 59:4614–4620. 5. Bone, R. C. 1993. How gram-positive organisms cause sepsis. J. Crit. Care 8:51–59. 6. Cleveland, M. G., J. D. Gorham, T. L. Murphy, E. Tuomanen, and K. M. Murphy. 1996. Lipoteichoic acid preparations of gram-positive bacteria induce interleukin-12 through a CD14-dependent pathway. Infect. Immun. 64:1906–1912. 7. Cundell, D. R., N. P. Gerard, C. Gerard, I. Idanpaan-Heikkila, and E. I. Tuomanen. 1995. Streptococcus pneumoniae anchor to activated human cells by the receptor for platelet-activating factor. Nature 377:435–438. 8. de Jonge, B. L., Y. S. Chang, D. Gage, and A. Tomasz. 1992. Peptidoglycan

VOL. 71, 2003 composition of a highly methicillin-resistant Staphylococcus aureus strain: the role of penicillin-binding protein 2A. J. Biol. Chem. 267:11248–11254. 9. Dziarski, R. 1982. Studies on the mechanism of peptidoglycan- and lipopolysaccharide-induced polyclonal activation. Infect. Immun. 35:507–514. 10. Dziarski, R., R. I. Tapping, and P. S. Tobias. 1998. Binding of bacterial peptidoglycan to CD14. J. Biol. Chem. 273:8680–8690. 11. Fischer, H., and A. Tomasz. 1984. Production and release of peptidoglycan and wall teichoic acid polymers in pneumococci treated with beta-lactam antibiotics. J. Bacteriol. 157:507–513. 12. Fischer, H., and A. Tomasz. 1985. Peptidoglycan cross-linking and teichoic acid attachment in Streptococcus pneumoniae. J. Bacteriol. 163:46–54. 13. Fischer, W. 1997. Pneumococcal lipoteichoic and teichoic acid. Microb. Drug Resist. 3:309–325. 14. Furie, M. B., B. L. Naprstek, and S. C. Silverstein. 1987. Migration of neutrophils across monolayers of cultured microvascular endothelial cells: an in vitro model of leucocyte extravasation. J. Cell Sci. 88:161–175. 15. Garcia-Bustos, J. F., B. T. Chait, and A. Tomasz. 1988. Altered peptidoglycan structure in a pneumococcal transformant resistant to penicillin. J. Bacteriol. 170:2143–2147. 15a.Garcia-Bustos, J. F., B. T. Chait, and A. Tomasz. 1987. Structure of the peptide network of pneumoccal peptidoglycan. J. Biol. Chem. 262:15400– 15405. 16. Garcia-Bustos, J. F., and A. Tomasz. 1987. Teichoic acid-containing muropeptides from Streptococcus pneumoniae as substrates for the pneumococcal autolysin. J. Bacteriol. 169:447–453. 17. Gupta, D., Y. P. Jin, and R. Dziarski. 1995. Peptidoglycan induces transcription and secretion of TNF-alpha and activation of lyn, extracellular signalregulated kinase, and rsk signal transduction proteins in mouse macrophages. J. Immunol. 155:2620–2630. 18. Gupta, D., T. N. Kirkland, S. Viriyakosol, and R. Dziarski. 1996. CD14 is a cell-activating receptor for bacterial peptidoglycan. J. Biol. Chem. 271: 23310–23316. 19. Heckels, J. E., and M. Virji. 1988. Separation and purification of surface components, p. 67–135. In I. Hancock and I. Poxton (ed.), Bacterial cell surface techniques. John Wiley & Sons, Ltd., London, United Kingdom. 20. Heumann, D., C. Barras, A. Severin, M. P. Glauser, and A. Tomasz. 1994. Gram-positive cell walls stimulate synthesis of tumor necrosis factor alpha and interleukin-6 by human monocytes. Infect. Immun. 62:2715–2721. 21. Hostetter, M. K. 1999. Opsonic and nonopsonic interactions of C3 with Streptococcus pneumoniae. Microb. Drug Resist. 5:85–89. 22. Jenni, R., and B. Berger-Bachi. 1998. Teichoic acid content in different lineages of Staphylococcus aureus NCTC 8325. Arch. Microbiol. 170:171– 178. 23. Jin, Y., D. Gupta, and 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. 177:1629–1638. 24. Keller, R., W. Fischer, R. Keist, and S. Bassetti. 1992. Macrophage response to bacteria: induction of marked secretory and cellular activities by lipoteichoic acids. Infect. Immun. 60:3664–3672. 25. Kengatharan, K. M., S. De Kimpe, C. Robson, S. J. Foster, and 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. 188:305–315. 26. Kengatharan, K. M., S. J. De Kimpe, and C. Thiemermann. 1996. Role of nitric oxide in the circulatory failure and organ injury in a rodent model of gram-positive shock. Br. J. Pharmacol. 119:1411–1421. 27. Majcherczyk, P. A., H. Langen, D. Heumann, M. Fountoulakis, M. P. Glauser, and P. Moreillon. 1999. Digestion of Streptococcus pneumoniae cell walls with its major peptidoglycan hydrolase releases branched stem peptides carrying proinflammatory activity. J. Biol. Chem. 274:12537–12543. 28. Mathison, J. C., P. S. Tobias, E. Wolfson, and R. J. Ulevitch. 1992. Plasma lipopolysaccharide (LPS)-binding protein: a key component in macrophage recognition of gram-negative LPS. J. Immunol. 149:200–206. 29. Mattsson, E., H. Van Dijk, K. Van Kessel, J. Verhoef, A. Fleer, and J. Rollof. 1996. Intracellular pathways involved in tumor necrosis factor-alpha release by human monocytes on stimulation with lipopolysaccharide or staphylococcal peptidoglycan are partly similar. J. Infect. Dis. 173:212–218. 30. Mattsson, E., L. Verhage, J. Rollof, A. Fleer, J. Verhoef, and H. Van Dijk. 1993. Peptidoglycan and teichoic acid from Staphylococcus epidermidis stimulate human monocytes to release tumour necrosis factor-alpha, interleukin-1 beta, and interleukin-6. FEMS Immunol. Med. Microbiol. 7:281–287. 31. Medzhitov, R., P. Preston-Hurlburt, and C. A. Janeway, Jr. 1997. A human

Editor: J. N. Weiser


32. 33.

34. 35. 36.



39. 40. 41. 42.

43. 44. 45.

46. 47.

48. 49.


51. 52.


homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 388:394–397. Mehlin, C., C. M. Headley, and S. J. Klebanoff. 1999. An inflammatory polypeptide complex from Staphylococcus epidermidis: isolation and characterization. J. Exp. Med. 189:907–918. Miethke, T., C. Wahl, K. Heeg, B. Echtenacher, P. H. Krammer, and H. Wagner. 1992. T cell-mediated lethal shock triggered in mice by the superantigen staphylococcal enterotoxin B: critical role of tumor necrosis factor. J. Exp. Med. 175:91–98. Murakami, K., and A. Tomasz. 1989. Involvement of multiple genetic determinants in high-level methicillin resistance in Staphylococcus aureus. J. Bacteriol. 171:874–879. Parsonnet, J. 1989. Mediators in the pathogenesis of toxic shock syndrome: overview. Rev. Infect. Dis. 11(Suppl. 1):263–269. Riesenfeld-Orn, I., S. Wolpe, J. F. Garcia-Bustos, M. K. Hoffmann, and E. Tuomanen. 1989. Production of interleukin-1 but not tumor necrosis factor by human monocytes stimulated with pneumococcal cell surface components. Infect. Immun. 57:1890–1893. Rietschel, E. T., J. Schletter, B. Weidemann, V. El-Samalouti, T. Mattern, U. Zahringer, U. Seydel, H. Brade, H. D. Flad, S. Kusumoto, D. Gupta, R. Dziarski, and A. J. Ulmer. 1998. Lipopolysaccharide and peptidoglycan: CD14-dependent bacterial inducers of inflammation. Microb. Drug Resist. 4:37–44. Rossjohn, J., R. J. Gilbert, D. Crane, P. J. Morgan, T. J. Mitchell, A. J. Rowe, P. W. Andrew, J. C. Paton, R. K. Tweten, and M. W. Parker. 1998. The molecular mechanism of pneumolysin, a virulence factor from Streptococcus pneumoniae. J. Mol. Biol. 284:449–461. Schwandner, R., R. Dziarski, H. Wesche, M. Rothe, and C. J. Kirschning. 1999. Peptidoglycan- and lipoteichoic acid-induced cell activation is mediated by Toll-like receptor 2. J. Biol. Chem. 274:17406–17409. Sparwasser, T., T. Miethke, G. Lipford, K. Borschert, H. Hacker, K. Heeg, and H. Wagner. 1997. Bacterial DNA causes septic shock. Nature 386:336– 337. Sriskandan, S., and J. Cohen. 1999. Gram-positive sepsis: mechanisms and differences from gram-negative sepsis. Infect. Dis. Clin. N. Am. 13:397–412. Timmerman, C. P., E. Mattsson, L. Martinez-Martinez, L. De Graaf, J. A. Van Strijp, H. A. Verbrugh, J. Verhoef, and A. Fleer. 1993. Induction of release of tumor necrosis factor from human monocytes by staphylococci and staphylococcal peptidoglycans. Infect. Immun. 61:4167–4172. Tobias, P. S., K. Soldau, J. A. Gegner, D. Mintz, and R. J. Ulevitch. 1995. Lipopolysaccharide binding protein-mediated complexation of lipopolysaccharide with soluble CD14. J. Biol. Chem. 270:10482–10488. Ton-That, H., K. F. Faull, and O. Schneewind. 1997. Anchor structure of staphylococcal surface proteins: a branched peptide that links the carboxyl terminus of proteins to the cell wall. J. Biol. Chem. 272:22285–22292. van Langevelde, P., J. T. van Dissel, E. Ravensbergen, B. J. Appelmelk, I. A. Schrijver, and P. H. Groeneveld. 1998. Antibiotic-induced release of lipoteichoic acid and peptidoglycan from Staphylococcus aureus: quantitative measurements and biological reactivities. Antimicrob. Agents Chemother. 42:3073–3078. Verhoef, J., and E. Mattsson. 1995. The role of cytokines in gram-positive bacterial shock. Trends Microbiol. 3:136–140. Wang, J. E., P. F. Jorgensen, M. Almlof, C. Thiemermann, S. J. Foster, A. O. Aasen, and R. Solberg. 2000. Peptidoglycan and lipoteichoic acid from Staphylococcus aureus induce tumor necrosis factor alpha, interleukin 6 (IL-6), and IL-10 production in both T cells and monocytes in a human whole blood model. Infect. Immun. 68:3965–3970. Wang, Z. M., C. Liu, and R. Dziarski. 2000. Chemokines are the main proinflammatory mediators in human monocytes activated by Staphylococcus aureus, peptidoglycan, and endotoxin. J. Biol. Chem. 275:20260–20267. Weidemann, B., H. Brade, E. T. Rietschel, R. Dziarski, V. Bazil, S. Kusumoto, H. D. Flad, and A. J. Ulmer. 1994. Soluble peptidoglycan-induced monokine production can be blocked by anti-CD14 monoclonal antibodies and by lipid A partial structures. Infect. Immun. 62:4709–4715. Weidemann, B., J. Schletter, R. Dziarski, S. Kusumoto, F. Stelter, E. T. Rietschel, H. D. Flad, and A. J. Ulmer. 1997. Specific binding of soluble peptidoglycan and muramyldipeptide to CD14 on human monocytes. Infect. Immun. 65:858–864. Weiser, J. N. 1998. Phase variation in colony opacity by Streptococcus pneumoniae. Microb. Drug Resist. 4:129–135. Yoshimura, A., E. Lien, R. R. Ingalls, E. Tuomanen, R. Dziarski, and D. Golenbock. 1999. Cutting edge: recognition of gram-positive bacterial cell wall components by the innate immune system occurs via Toll-like receptor 2. J. Immunol. 163:1–5.