Human Blood Lipopolysaccharide-Induced Responses in Serum ...

5 downloads 116219 Views 123KB Size Report
Receive free email-alerts when new articles cite this article. Sign up at: Print ISSN: .... (TRLAFPKECLNQF), were prepared by automated simultaneous multiple.
A Synthetic Lipopolysaccharide-Binding Peptide Based on Amino Acids 27–39 of Serum Amyloid P Component Inhibits Lipopolysaccharide-Induced Responses in Human Blood Carla J. C. de Haas,1 Marijke E. van der Tol, Kok P. M. Van Kessel, Jan Verhoef, and Jos A. G. Van Strijp LPS-binding proteins in plasma play an important role in modifying LPS toxicity. Significant properties have already been attributed to the LPS-binding protein (LBP). It accelerates LPS toxicity as well as incorporation into high-density lipoproteins, leading to neutralization of LPS in serum. A search for other LPS-binding components in serum, using LPS-coated magnetic beads, revealed a new LPS-binding protein. N-terminal microsequencing identified this protein as serum amyloid P component (SAP). Purified SAP bound to smooth and rough types of LPS via the lipid A part. SAP inhibited the binding of FITC-labeled ReLPS (LPS from Salmonella minnesota strain R595) to human monocytes and the ReLPS-induced priming of the oxidative burst of human neutrophils only in the presence of low concentrations of LBP. In search for the LPS binding site of SAP, we found that pep27–39, a 13-mer peptide consisting of amino acids 27–39 of SAP, competitively inhibited the binding of LPS to SAP. In addition, pep27–39 significantly inhibited ReLPS-induced responses in phagocytes in the presence of serum, as well as in human whole blood. Carboxamidomethylated pep27–39 showed an even more pronounced reduction of the ReLPS-induced priming of phagocytes in human blood. Performing gel filtration of FITC-labeled ReLPS incubated with soluble CD14, we showed that SAP could not prevent binding of LPS to soluble CD14, in contrast to pep27–39. The ability of pep27–39 to antagonize specifically the effects of LPS in the complex environment of human blood suggests that pep27–39 may be a novel therapeutic agent in the treatment of Gram-negative sepsis. The Journal of Immunology, 1998, 161: 3607–3615.

L

ipopolysaccharide is the major component of the outer membrane of Gram-negative bacteria. The specific interaction of LPS with soluble and cell surface localized components is a prerequisite for the orchestration of the immune response to a Gramnegative infection. Cell activation requires CD14, which is not only expressed on monocytes, macrophages, and neutrophils, but is also present in a soluble form in serum as soluble CD14 (sCD14).2 CD14 is the only known receptor for LPS that is able to transduce a signal, albeit probably indirectly (1, 2). CD14-negative cells, such as endothelial, epithelial, and smooth muscle cells, can be activated by LPS via the interaction of LPS with sCD14 (3). LPS-binding protein (LBP) is a lipid-transfer protein that promotes the movement of LPS from micelles to (s)CD14 (4). The interaction of LPS with (s)CD14 is markedly enhanced by LBP and initiates a cascade of cellular responses, which are necessary to fight Gram-negative infection, but under certain circumstances also lead to septic shock (5). LBP can also catalyze the movement of LPS to high-density lipoproteins (HDL), which neutralizes the capacity of LPS to stimulate cells (6, 7). There are more LPS-binding proteins known to play a role in LPSmediated effects, several of which have been recognized among the antimicrobial arsenal secreted by neutrophils (8). Bactericidal/permeEijkman Winkler Institute, Department of Inflammation, Utrecht, The Netherlands Received for publication January 14, 1998. Accepted for publication June 3, 1998. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Address correspondence and reprint requests to Dr. Carla J. C. de Haas, Eijkman Winkler Institute. Department of Inflammation, AZU, G04.614, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands. E-mail address: [email protected] 2 Abbreviations used in this paper: sCD14, soluble CD14; BPI, bactericidal/permeability-increasing protein; CAP, cationic protein; HDL, high-density lipoprotein; HSA, human serum albumin; LBP, LPS-binding protein; mCD14, membrane-bound CD14; ReLPS, LPS from Salmonella minnesota strain R595; SAP, serum amyloid P.

Copyright © 1998 by The American Association of Immunologists

ability-increasing protein (BPI), a cationic protein, not only exerts strong antimicrobial effects against Gram-negative bacteria, but is also potent in binding and detoxifying LPS (9, 10). Neutrophil granule components, such as cationic protein 18 (CAP18), lysozyme, lactoferrin, and azurocidin/CAP37, have also been described as LPS-binding proteins with a neutralizing capacity (8, 11, 12). Serum amyloid P component (SAP) is a decameric serum glycoprotein composed of identical 25.5-kDa subunits noncovalently associated in two pentameric rings interacting face to face. It has been associated with all forms of amyloid deposits, for example with those in Alzheimer’s and Parkinson’s disease, Down’s syndrome, and Creutzfeldt-Jacob syndrome (13). It is described to protect amyloid deposits from proteolytic degradation in vivo (14). Recently, the participation of SAP in the pathogenesis of amyloidosis was demonstrated using mice with targeted deletion of the SAP gene (15). However, SAP has also been reported to inhibit Alzheimer b-peptide fibril formation in an in vitro model (16). Furthermore, it is present in the normal glomerular basement membrane covalently associated with collagen and is associated with elastic fibers in skin and blood vessels (17). SAP belongs to the family of pentraxins, lectin-like serum proteins, which have been stably conserved throughout vertebrate evolution. This protein has a 51% amino acid homology with C-reactive protein, the classical acute-phase protein found in humans. SAP is an acute-phase reactant in mice, while it is constitutively present in human serum at 40 mg/ml, with a maximum twofold increase during sepsis (14). SAP shows calcium-dependent binding to DNA (18), chromatin (19), and glycosaminoglycans such as heparin, heparan, and dermatan sulfate (20), and has been described to play a role in the complement cascade since it can bind to several complement components, such as C4b-binding protein, C1q, and C3bi (21, 22), and to immune complexes, probably via the F(ab9)2 fragment of IgG 0022-1767/98/$02.00

3608

SAP PEPTIDE INHIBITS LPS RESPONSES IN HUMAN BLOOD

(23, 24). Although its exact physiologic function is still unknown, it is believed to play a role in the binding and clearance of host- or pathogen-derived cellular debris at sites of inflammation (25). At present, it is still not possible to reconstruct the actual modulating role of plasma in phagocyte activation by LPS and the pathophysiology of Gram-negative sepsis. Therefore, we sought other components that fulfill an important role in these events.

creasing amounts (1–100 mg/ml) of lipid A, the rough types (Re, Rc, Ra) of LPS from S. typhimurium, LPS from S. minnesota R595, LPS from E. coli O111:B4, or from its Rc-mutant J5; washed three times with HBSS containing 0.05% Tween-20; and incubated with mouse anti-human SAP mAb clone 5 (Sigma) for 30 min at room temperature with a subsequent washing procedure. Detection of SAP binding to the beads was performed by incubating the beads for 30 min with FITC-labeled goat anti-mouse Ig (Becton Dickinson, Mountain View, CA) and analysis on a FACScan (Becton Dickinson).

Materials and Methods

Isolation of SAP from serum

Reagents LPS from Salmonella minnesota strain R595, Escherichia coli O111:B4, its Rc mutant J5, monophosphoryl lipid A, and LPS from Salmonella typhimurium and S. typhimurium TV119 (Ra), SL684 (Rc), and SL1181 (Re) were obtained from Sigma (St. Louis, MO). Human rLBP was a generous gift from H. Lichenstein (Amgen, Boulder, CO).

Serum and plasma Blood was drawn from healthy human volunteers. Human serum was obtained after pooling the sera of three or more donors and stored until use at 270°C. Plasma was obtained from blood in heparinized or EDTA tubes. Fresh human serum was used for SAP isolation.

Peptide synthesis A 13-mer peptide, pep27–39 (EKPLQNFTLCFRA), corresponding to amino acids 27–39 of SAP, and a scrambled peptide, pep27–39scr (TRLAFPKECLNQF), were prepared by automated simultaneous multiple peptide synthesis. The simultaneous multiple peptide synthesis setup was developed using a standard autosampler, as described previously (26). Briefly, standard Fmoc chemistry with in situ PyBop/N-methylmorpholine (NMM) (Novabiochem, Laufelfingen, Switzerland) activation of the amino acids in a fivefold molar excess with respect to 2 mmol/peptide PALPEG-PS resin (Perseptive Biosystems, Framingham, MA) was employed. Peptides were obtained as C-terminal amides after cleavage with 90 to 95% trifluoroacetic acid/scavenger mixtures. Peptides were dissolved in 50 mM HAc at a concentration of 5 mM and further diluted in 0.25 M Tris-HCl, pH 7.5, to a concentration of 0.6 mM. Before use in biologic assays, the peptides were further diluted in HBSS, containing 0.2% human serum albumin (HSA; Central Laboratory Blood transfusion, Amsterdam, The Netherlands). In some experiments, pep27–39 was carboxamidomethylated to prevent formation of dimers, as follows: Pep27–39 in 50 mM HAc was first reduced using Tris(2-carboxyethyl)phosphine hydrochloride (TCEPHCl; Pierce, Rockford, IL; 2 mg/1.5 mmol peptide) for 1.5 h at room temperature under constant agitation. Then the pH was adjusted to pH 7 using 4 M NaOH, and 100 ml 0.5 M iodoacetamide (Merck-Schuchardt, Hohenbrunn bei Munchen, Germany) was added. Carboxamidomethylation was allowed to take place overnight under constant agitation. Carboxamidomethylated pep27–39 was then dialyzed against 50 mM HAc in a 100 D cutoff dialysis membrane. Ellman’s reagent (5, 59-dithio-bis-(2nitrobenzoic acid); Sigma) (27) was used to determine the presence of free sulfhydryl groups before and after carboxamidomethylation. Optimal reduction was verified and subsequent efficient carboxamidomethylation was confirmed by the total absence of free sulfhydryl groups after the procedure.

Detection of LPS-binding proteins in serum and plasma LPS from S. minnesota strain R595 (ReLPS) was used to coat magnetic beads (Tosyl-activated Dyna M-450 beads; Dynal A. S., Oslo, Norway), as described by Weersink et al. (28). Control beads were treated similarly without the addition of ReLPS. For the detection of LPS-binding proteins, ReLPS-coated magnetic beads (5 3 106) were incubated with 300 ml 10% human serum, heparin plasma, or EDTA plasma for 30 min at 37°C on a rotator. Except for EDTA plasma, which was diluted in PBS, samples were diluted in HBSS. After incubation, the beads were washed three times with HBSS containing 0.05% Tween-20. Proteins were eluted from the beads by heating at 100°C for 2 min in 20 ml sample buffer (2% SDS, 2.5% DTT, 20% glycerol, 0.001% bromophenol blue, in 0.05 M Tris-HCl, pH 6.9) and detected by SDS-PAGE on 12.5%, 0.75 mm minigels. Gels were stained with Coomassie brilliant blue. Proteins were blotted onto Problott membrane, stained with Coomassie brilliant blue, according to the manufacturer’s descriptions (Applied Biosystems, Foster City, CA), and sequenced on an Applied Biosystems protein sequencer model 476A. In other experiments, magnetic beads were coated with 1 mg/ml smooth type LPS from S. typhimurium. The beads (5 3 106) were then incubated with 1 mg/ml SAP for 30 min at 37°C, under constant agitation, in the presence of in-

Isolation of SAP from serum was performed as described by Skinner and Cohen (29) with modifications. Briefly, fresh human serum was centrifuged at 17,000 3 g for 5 h at 4°C to remove the top lipid layer. The delipidated serum was then applied to a Sepharose 4B (Pharmacia, Uppsala, Sweden) column, equilibrated with a calcium buffer (140 mM NaCl, 0.01 M Tris-HCL, 2 mM CaCl2, pH 7.8). SAP was eluted with an EDTA buffer (140 mM NaCl, 0.01 M Tris-HCl, 10 mM EDTA, pH 8) and applied to a gel-filtration column (Superdex 200; Pharmacia) equilibrated in the same EDTA buffer. Fractions containing SAP were concentrated in an Amicon filter system (10-kDa cutoff) and dialyzed against PBS or saline. Purity of the SAP isolate was checked with SDS-PAGE, and subsequent Coomassie brilliant blue staining. The SAP concentration was determined by ELISA. Therefore, microtiter plates (96-well polyvinyl; Costar, Cambridge, MA) were coated overnight at 4°C with anti-human SAP mAb 5.4A in PBS (1 mg/ml; Monosan; Sanbio, Uden, The Netherlands) and blocked for 1 h at 37°C with PBS/0.05% Tween/4% BSA. Samples and a SAP standard (Calbiochem-Novabiochem, La Jolla, CA) were diluted in PBS/ Tween/1% BSA and incubated for 1 h at 37°C, followed by a 1-h incubation with a second biotinylated anti-human SAP mAb 5.4A (1 mg/ml; Monosan; Sanbio). Then peroxidase-labeled streptavidin (Southern Biotechnology, Birmingham, AL) was added, and after 1 h, the substrate composed of TMB (tetramethylbenzidine; Sigma) and H2O2 in 0.1 M acetate buffer was allowed to be converted for 10 min. To stop the enzymatic reaction, 2 N H2SO4 was added and the OD was determined using a microtiter plate reader (Bio-Rad Laboratories, Hercules, CA) operating at 450 nm. In between incubations, the plate was washed five times with H2O/ 0.05% Tween. LPS contamination of SAP preparations was about 10 to 20 ng/mg SAP, as determined by the Limulus amebocyte lysate assay (Chromogenix AB, Mo¨lndal, Sweden).

Cell isolation Human neutrophils and PBMC were isolated from heparinized blood drawn from healthy volunteers, as described by Troelstra et al. (30).

Binding of fluorescein-labeled ReLPS to human monocytes FITC-labeled ReLPS (FITC-LPS) was prepared as described by Troelstra et al. (31), with a molar labeling efficiency of 1:1. For FITC-LPS binding studies, 2.5 ng/ml FITC-LPS was preincubated with increasing amounts of SAP (0 –30 mg/ml) or pep27–39 (0 –10 mM) for 0 to 30 min at 37°C in HBBS containing 0.2% HSA. Then LBP (10 ng/ml) and PBMC (6 3 106/ml) in the same buffer were added to a final volume of 50 ml, gently shaken for 30 min at 37°C, and put on ice. Binding of FITC-LPS to monocytes was analyzed on a FACScan, using forward and sideward scatter parameters to gate on monocytes. The results were expressed as the mean fluorescence of 10,000 cells. The percentage of inhibition of binding was calculated using the following formula: 1-(A-bgrA/B-bgrB) 3 100%, where A is the mean fluorescence of cells incubated with FITC-LPS 1 LBP 1 SAP or pep27–39; B, the mean fluorescence of cells incubated with FITC-LPS 1 LBP; bgrA, the background fluorescence of cells incubated with FITC-LPS 1 SAP or pep27–39; and bgrB, the background fluorescence of cells incubated with FITC-LPS alone.

LPS-induced priming of human neutrophils This procedure has been described in detail elsewhere (30). Briefly, neutrophils (5 3 106/ml) were added to a mixture of 1 ng/ml LPS alone or 1 ng/ml LPS with increasing amounts of SAP (0 –30 mg/ml) or peptides (0 –30 mM) in the presence of 1 to 100 ng/ml LBP in HBBS/1.8% HSA. In some experiments, increasing amounts of serum (0.1–3%) were used. In experiments using peptides, LPS and peptides were preincubated for 30 min before addition to the cells. Cells were incubated with the mixtures for 30 min at 37°C under constant agitation. Next, chemoluminescence response was measured in a luminometer (Autolumat LB 953; Berthold GmbH, Wildbad, Germany) after automated injection of FMLP (1 mM final concentration) and HBSS containing 180 mM luminol (Sigma). The chemoluminescence response was measured automatically over a period of

The Journal of Immunology

3609

10 min. Data were analyzed with the AXIS software package (ExOxEmis, San Antonio, TX). Curves were obtained for all samples presenting the chemoluminescence response in cpm versus time. Absolute counts were obtained by calculating the area under the curve of the chemoluminescence for 10 min. In experiments using human blood, 80 ml of human blood was incubated with 20 ml of LPS/peptide mixture for 30 min at 37°C. Then 900 ml PBS/0.05% glucose was added, and 100 ml of this mixture was used to measure the chemoluminescence response, as described. In some experiments, PMA (25 ng/ml) was used to activate the neutrophils or blood for a chemoluminescence response. In other experiments, TNF-a (1 nM) was used to prime neutrophils or blood for an enhanced FMLP response in the presence of peptides.

J5-LPS ELISA J5-LPS was coated to 96-well flat-bottom plates (Greiner, Nu¨rtingen, Germany) at a concentration of 1 mg/ml in PBS for 1 h at 37°C, with a subsequent overnight incubation at 4°C. The plate was washed five times with H2O/0.05% Tween, and blocked for 1 h at 37°C with PBS/4% BSA/0.05% Tween. Then 0.3 mg/ml SAP was incubated with increasing concentrations of pep27–39 (0 –30 mM) in HBSS/0.2% BSA/0.05% Tween for 1 h at 37°C. Subsequently, the binding of SAP was detected, as described for the SAP ELISA in Materials and Methods (isolation of SAP from serum), with the only exception of using HBSS/0.2% BSA/0.05% Tween as a dilution buffer for the second biotinylated anti-human SAP 5.4A mAb and the peroxidase-labeled streptavidin.

FIGURE 1. SDS-PAGE of proteins extracted from serum and plasma using ReLPS-coated beads. ReLPS-coated beads (A) and control beads (B) were incubated with 10% serum (lane 1), EDTA plasma (lane 2), and heparin plasma (lane 3) for 30 min. Beads were washed and heated for 3 min at 100°C in sample buffer. Eluted proteins were detected on SDSPAGE with Coomassie brilliant blue staining. M represents the molecular mass (kDa) of marker proteins.

FITC-LPS gel filtration To study the effect of SAP and pep27–39 on the LPS binding to recombinant sCD14 (rsCD14; kindly provided by Dr. Henri S. Lichenstein, Amgen), gel filtration of FITC-LPS in combination with on-line fluorescence detection was used. In principle, this system resembles the gel-shift assay described earlier by Hailman et al. (32) using 3H-labeled LPS. FITC-LPS alone forms self-quenching aggregates. This LPS aggregate will migrate as a molecule of about 500 kDa with a very low fluorescence signal. Addition of sCD14 will monomerize FITC-LPS, resulting in a rise in fluorescence of the rsCD14/FITC-LPS complex and a comigration of FITC-LPS with rsCD14. FITC-LPS (0.5 mg/ml) was incubated with 5 mg/ml rsCD14 and 100 ng/ml LBP with or without addition of 100 mg/ml SAP or 10 to 30 mM of pep27–39. Pep27–39 was preincubated with FITC-LPS for 30 min at 37°C before addition of rsCD14 and LBP. After 30-min incubation at 37°C, 100 ml of the mixture was loaded onto a Superdex TM 200 HR 10/30 column (Pharmacia) and run at a flow of 0.5 ml/min for 35 ml. The effect of SAP and pep27–39 on the binding of LPS to rsCD14 was also studied in the absence of LBP. In these experiments, the incubation time of rsCD14 and FITC-LPS was also 30 min. To determine the retention time of FITCLPS, a sample of 50 mg/ml FITC-LPS alone was run. Fluorescence was recorded using a Perkin-Elmer (Norwalk, CT) LS30 luminometer with excitation wavelength of 475 nm and emission set at 514 nm.

These data suggest that SAP has a specific binding capacity for LPS. As shown in Figure 3, experiments performed with beads coated with smooth type LPS from S. typhimurium demonstrated that purified SAP can also bind to a smooth type of LPS. Competition experiments with other rough and smooth types of LPS from E. coli and other Salmonella strains showed that SAP not only exhibits a specific binding to ReLPS from S. minnesota, but it also specifically binds to all other tested forms of LPS, including monophosphoryl lipid A. SAP inhibits binding of FITC-LPS to human monocytes The finding that SAP specifically binds to LPS prompted us to investigate the effect of SAP on the LPS-induced effects on phagocytes. For this purpose, we studied the effect of purified SAP on LPS binding to monocytes. FITC-LPS was preincubated with various amounts of SAP, whereafter the binding of FITC-LPS to monocytes was studied by flow cytometry. As shown in Figure 4,

Results Detection of LPS-binding proteins in human serum and plasma Magnetic beads coated with LPS from S. minnesota strain R595 (ReLPS beads) were used to capture LPS-binding proteins from serum and plasma samples. Figure 1A shows the binding of three proteins, with Mr values of approximately 70, 45, and 30 kDa, in the presence of 10% serum and heparin plasma, while hardly any proteins bound to the ReLPS beads in the presence of 10% EDTA plasma. Control nonReLPS-coated beads showed binding of the 70-kDa protein only (Fig. 1B). N-terminal amino acid sequencing of the blotted 45- and 70-kDa proteins showed 100% homology with the a- and the b-chain of complement component C3bi, respectively. The 30-kDa protein yielded a sequence of 17 amino acids that was 100% homologous with that of the N-terminal sequence of human SAP. To check whether binding of these proteins to ReLPS was specific, we incubated ReLPS beads in 10% serum in the presence of increasing amounts of free ReLPS. Figure 2 shows that in the presence of 10 or 100 mg/ml free ReLPS, binding of the 30-kDa protein to the beads was inhibited, suggesting competition between free LPS and LPS coated on the beads for binding to SAP. Additional experiments showed that purified SAP, isolated from human serum, also bound to ReLPS beads. This binding could be inhibited by the addition of free ReLPS (data not shown).

FIGURE 2. SDS-PAGE showing binding of SAP to ReLPS-coated beads and to free ReLPS. ReLPS-coated beads were incubated with 10% serum for 30 min, then washed and heated for 3 min at 100°C in sample buffer. Eluted proteins were detected on SDS-PAGE with Coomassie brilliant blue staining. Lanes 1 to 4 show proteins binding to ReLPS-coated beads in the absence (lane 1) and presence of 100 mg/ml (lane 2), 10 mg/ml (lane 3), and 1 mg/ml (lane 4) free ReLPS. M represents the molecular mass (kDa) of marker proteins.

3610

SAP PEPTIDE INHIBITS LPS RESPONSES IN HUMAN BLOOD

FIGURE 3. Flow-cytometric analysis of SAP binding to various LPS types and monophosphoryl lipid A. Magnetic beads were coated with smooth type LPS from S. typhimurium. Binding of 1 mg/ml SAP to the beads was tested in the presence of increasing amounts (1–100 mg/ml) of monophosphoryl lipid A, smooth and rough types of LPS from S. typhimurium, LPS from S. minnesota R595, and LPS from E. coli O111:B4 and its Rc mutant J5. SAP binding was determined by incubating the beads with an anti-human SAP mAb and then FITC-labeled goat anti-mouse Ig. Beads were analyzed on a FACScan. Binding of SAP to the beads without the addition of one of the aforementioned free types of LPS represents 100% SAP binding. Data represent the mean of two separate experiments.

preincubation of FITC-LPS with SAP dose dependently inhibited the binding of FITC-LPS to monocytes up to 90%. This effect was already evident at a concentration of 1 mg/ml SAP. To study kinetics of SAP binding to LPS, FITC-LPS was preincubated for 0 to 30 min with 10 mg/ml SAP, and the competence of FITC-LPS to bind to monocytes was analyzed. Even without preincubation, SAP inhibited binding of FITC-LPS to monocytes to the same extent as was achieved after 30 min of preincubation (data not shown). To exclude the possibility that SAP prevents LPS binding to the monocytes, via binding to the monocytes itself, a control experiment was performed. Therefore, PBMC were preincubated with SAP for 30 min, and washed three times to remove unbound SAP. Subsequent incubation of the PMBC with FITC-LPS, and analysis on the FACS, showed that preincubation of PBMC with SAP did not inhibit binding of FITC-LPS to monocytes. SAP inhibits LPS-induced priming of human neutrophils To evaluate the effects of SAP on LPS toxicity in a functional assay, we investigated whether SAP could inhibit LPS-induced priming of neutrophils. Neutrophils were primed with 1 ng/ml LPS for an enhanced FMLP response in the presence of 1 ng/ml LBP for 30 min. Addition of SAP revealed a dose-dependent inhibition on LPS-induced priming of neutrophils with a 70 to 80% inhibition at 30 mg/ml SAP (Fig. 5). However, the inhibitory effect of SAP on LPS-induced priming of neutrophils was profoundly reduced when higher concentrations of LBP were used (Fig. 6). In the presence of serum concentrations over 0.1%, the inhibitory effect of SAP on LPS-induced priming was strongly reduced as well (data not shown). SAP peptide, pep27–39, inhibits binding of SAP to LPS In literature, a peptide, comprising the amino acids 27–39 of the SAP sequence, was described to interfere with the binding of SAP

FIGURE 4. Flow-cytometric analysis demonstrating inhibition of FITC-LPS binding to monocytes by SAP. Human monocytes were incubated with FITC-LPS (2.5 ng/ml) in the presence or absence of increasing amounts of SAP for 30 min. FITC-LPS binding was assessed by FACScan analysis. FITC-LPS binding to monocytes is shown in the presence of 10 ng/ml LBP without SAP (A), and with 1 (B), 3 (C), 10 (D), and 30 (E) mg/ml SAP. The background binding of FITC-LPS in the absence of LBP is shown in all figures (thin lines). The data are representative of five separate experiments. Percentage of inhibition of FITC-LPS binding to monocytes is shown in the presence of increasing amounts of SAP 6 SEM (n 5 5) (F).

The Journal of Immunology

3611

FIGURE 6. LBP decreases the SAP inhibition of the LPS-induced priming of the oxidative burst in human neutrophils. Neutrophils were primed for 30 min with 1 ng/ml ReLPS and 30 mg/ml SAP in the presence of increasing amounts of LBP. Thereafter, FMLP-induced chemoluminescence was measured for 10 min. Data represent the percentage of inhibition of the 10-min integral (AUC) of four separate experiments 6 SEM.

FIGURE 5. Chemoluminescence assay illustrating inhibition of LPSinduced priming of the oxidative burst in human neutrophils by SAP. Human neutrophils were incubated with 1 ng/ml ReLPS in the presence or absence of increasing amounts of SAP for 30 min, after which the FMLPinduced chemoluminescence was measured. Effects of SAP on LPS priming of neutrophils are shown with LPS together with increasing amounts of SAP (0 –30 mg/ml), all in the presence of 1 ng/ml LBP. Background luminescence was measured with 1 ng/ml LPS without addition of LBP. The data are representative of four separate experiments (A). Percentage of inhibition of the 10-min integral (AUC) of the LPS-induced priming of neutrophils is shown in the presence of increasing amounts of SAP 6 SEM (n 5 4) (B).

interfere with the binding of FITC-LPS to monocytes (Fig. 8). A control experiment testing the possibility that binding of pep27–39 to the monocytes would influence subsequent binding of FITCLPS, as we performed earlier for SAP, showed that prebound pep27–39 did not influence the binding of FITC-LPS to the monocytes. Testing the effect of pep27–39 on the LPS-induced priming of neutrophils in the presence of increasing concentrations of LBP, it was found that 3 mM of pep27–39 profoundly reduced the LPSinduced priming of neutrophils, even at high concentrations of LBP (Fig. 9). Also in the presence of serum, 30 mM of pep27–39 was able to almost completely inhibit the LPS-induced priming of neutrophils (1% serum; percentage of inhibition, 90.2 6 11.7;

to some of its ligands. Therefore, we tested whether this peptide (pep27–39) could interfere with the binding of SAP to LPS as well. A microtiterplate was coated with J5-LPS and incubated with SAP in the presence of increasing amounts of pep27–39. Figure 7 shows that the binding of SAP to J5-LPS was completely inhibited by 30 mM of pep27–39. A control peptide, comprising the same amino acids in a scrambled order (pep27–39scr), did not interfere with SAP binding to LPS. Pep27–39 neutralizes LPS in human serum and blood Pep27–39 was anticipated to inhibit binding of FITC-LPS to monocytes. In a flow-cytometric assay, as little as 0.1 mM of pep27–39 inhibited the binding of FITC-LPS to monocytes by 50%, while a complete inhibition was reached at a concentration of 3 mM of pep27–39. The scrambled peptide, pep27–39scr, did not

FIGURE 7. Pep27–39 inhibits binding of SAP to J5-LPS. SAP binding to J5-LPS, coated to a 96-well plate, was monitored in the presence of increasing amounts of pep27–39 or scrambled pep27–39 (pep27–39scr) via detection with a biotinylated anti-human SAP mAb, followed by a streptavidin-horseradish peroxidase conjugate. Data represent three separate experiments 6 SEM.

3612

FIGURE 8. Pep27–39 inhibits binding of FITC-LPS to human monocytes. Human monocytes were incubated with 2.5 ng/ml FITC-LPS and 10 ng/ml LBP for 30 min in the presence of increasing amounts of pep27–39 or scrambled pep27–39 (pep27–39scr). Binding of FITC-LPS was analyzed on a FACScan. Data represent the percentage of inhibition of FITCLPS binding of five separate experiments 6 SEM.

n 5 3). As pep27–39 contains a Cys residue, it will spontaneously form dimers. To investigate the effect of dimerization of pep27–39 on its LPS-inhibitory effects, the free sulfhydryl group of the Cys residue was blocked by carboxamidomethylation. In human whole blood, 30 mM of pep27–39 decreased the LPS-induced priming of neutrophils to about 50% compared with the chemoluminescence response in the presence of 1 ng/ml LPS alone. Carboxamidomethylated pep27–39 was even more efficient in inhibiting the LPS-induced priming of human blood (Fig. 10). The peptide with the scrambled sequence, pep27–39scr, showed no effect. In addition, carboxamidomethylated pep27–39 was more potent than pep27–39 in the inhibition of FITC-LPS binding to monocytes (IC50 of 0.03 mM compared with 0.1 mM). Control experiments

FIGURE 9. Pep27–39 inhibits the LPS-induced priming of neutrophils independent of the LBP concentration. Human neutrophils were primed for 30 min with 1 ng/ml ReLPS and 3 mM pep27–39 in the presence of increasing amounts of LBP. Subsequently, FMLP-induced chemoluminescence was measured for 10 min. Data represent the percentage of inhibition of the 10-min integral (AUC) of five separate experiments 6 SEM.

SAP PEPTIDE INHIBITS LPS RESPONSES IN HUMAN BLOOD

FIGURE 10. Pep27–39 and carboxamidomethylated pep27–39 inhibit LPS priming of human blood. Human blood was primed for 30 min with 1 ng/ml ReLPS and increasing amounts of peptides. An FMLP-induced chemoluminescence in 10-fold diluted blood was measured for 10 min. Data represent the percentage of inhibition of the 10-min integral (AUC) of four separate experiments 6 SEM.

using TNF-a as the primer for the FMLP-induced oxidative burst of human blood did not show any inhibitory effects of pep27–39 or carboxamidomethylated pep27–39. The activation of human blood by PMA was not affected by either of the peptides (data not shown). This indicates that (carboxamidomethylated) pep27–39 specifically antagonizes the LPS-induced priming of neutrophils and that this inhibitory effect was not caused by nonspecific cytotoxicity. SAP-derived peptide, pep27–39, but not SAP, inhibits binding of LPS to rsCD14 To study the effect of SAP and pep27–39 on the LPS binding to sCD14, we used a FITC-LPS gel-filtration technique in which we determined the capacity of rsCD14 to bind FITC-LPS, by monitoring the change in retention time of FITC-LPS on a gel-filtration column, in the presence of SAP or pep27–39. In Figure 11, we show that FITC-LPS runs at about 7 ml, just after the void volume of this column, representing 500 kDa (Fig. 11A). A 30-min preincubation of FITC-LPS with rsCD14 and LBP resulted in a shift of fluorescence from this quenched fluorescence signal at 7 to 14 ml (60 kDa), the place at which rsCD14 elutes from the column (Fig. 11B). Addition of SAP did not decrease the 60-kDa signal (data not shown). Even without LBP, SAP could not inhibit binding of FITC-LPS to rsCD14 (Fig. 11F). Lower concentrations of rsCD14 did not result in a retention time shift, so that the effect of SAP could not be tested at these concentrations of rsCD14. However, preincubation of FITC-LPS with 30 mM of pep27–39 very potently inhibited the binding of FITC-LPS to rsCD14, as is shown by the disappearance of the 60-kDa peak (Fig. 11D). The concentrations of FITC-LPS in Figure 11, B to G, are 10-fold lower then in Figure 11A. Therefore, the 500-kDa peak, representing the quenched form of FITC-LPS aggregates, is not visible in the lower panels. Figure 11, E and G, show that pep27–39 also prevents binding of LPS to rsCD14 in the absence of LBP.

Discussion In literature, SAP is described to bind many ligands, although no clear biologic function has been ascribed to it as yet (19, 20, 22,

The Journal of Immunology

3613

FIGURE 11. SAP cannot prevent binding of LPS to sCD14, in contrast to pep27–39. FITC-LPS (0.5 mg/ml), incubated with 5 mg/ml sCD14, in the presence (B, C, D) or absence (E, F, G) of 100 ng/ml LBP, was run on a gel-filtration column in combination with on-line fluorescence detection. Line A shows the retention time of aggregated FITC-LPS (50 mg/ml) in the absence of sCD14. FITC-LPS was preincubated with 10 mM (C) and 30 mM (D and G) pep27–39 or with 100 mg/ml SAP (F). The y-axis represents relative fluorescence units. The different trails are superimposed with the same fluorescence scale for each line.

33, 34). In our study, using magnetic beads coated with ReLPS, we identified SAP as a new LPS-binding protein present in human plasma. Competition experiments showed that SAP specifically binds to rough as well as smooth types of LPS; however, the affinity of SAP for smooth types of LPS seems less as compared with rough types of LPS (Fig. 4). This can be partly explained by differences in m.w. of the LPS, but the affinity of SAP to smooth types of LPS could also be lower because of sterical hindrance caused by the oligosaccharide chain of smooth LPS. Monophosphoryl lipid A was a very potent inhibitor for the binding of SAP to LPS, indicating that SAP binds to the lipid A part of LPS. We showed that SAP profoundly inhibited LPS responses in human granulocytes in the presence of low concentrations of LBP. This interference of LPS binding to CD14 by SAP was not the result of direct SAP binding to CD14, thereby inhibiting LPS/ CD14 interactions, as preincubation of monocytes with SAP did not affect subsequent binding of LPS to the cells (data not shown). SAP was not able to neutralize LPS in serum or human blood. However, we demonstrated that SAP binds to LPS in the presence of serum, suggesting a role for SAP binding to LPS in vivo. Not much is known about the fate of LPS in vivo. To date, LPS has been described to bind either (s)CD14 or HDL, after entering the circulation, via interaction with LBP (4, 6). In the circulation, LPSbinding proteins as BPI and CAP18 cannot play a role in the immediate binding of LPS since they are constituents of neutrophil granules and are supposed to play a role in LPS neutralization only at specific sites of inflammation (8, 10, 35). Because the binding of LPS to HDL is a slow process (6, 36) and we have shown that

binding of SAP to LPS occurs rapidly, we propose that, on entering the circulation, LPS is immediately captured by SAP. Although SAP is not able to neutralize LPS in vivo directly, it could serve as a carrier protein to transport LPS to the liver for rapid detoxification, and thus indirectly contribute to LPS clearance. Experiments in SAP-knockout mice are needed to further investigate the exact role of SAP binding to LPS in endotoxemia. It has been shown that SAP can interact with phagocytes (37, 38). It can prime neutrophils (37) and enhance macrophage listericidal activity (39, 40). Furthermore, it has been described that substrate-bound SAP can activate C3b and C3bi receptors of monocytes (38). Since SAP can bind to phagocytes, bacteria (41, 42), and complement components (21–24), it might serve a role as an opsonin, potentiating phagocytosis of C3- or SAP-coated pathogens. The direct interaction with LPS on bacteria clearly fits in this model. In search for the LPS-binding region of SAP, we found a SAP peptide, pep27–39, which could compete for the binding of SAP to LPS. This 13-mer synthetic SAP peptide, comprising the amino acids 27–39 of SAP, was described to interfere with the interaction of SAP with heparin and C4b-binding protein (21, 33, 43). In addition, a 12mer synthetic peptide that corresponds to amino acids 27–38 was reported to support cell attachment (44). We showed that pep27–39 was able to inhibit LPS responses in human phagocytes even in the presence of human blood. Carboxaminomethylation of pep27–39, which prevents formation of dimers via blockage of the free sulfydryl groups of Cys residues, resulted in a peptide that was about 4 times more active. In other studies concerning cell attachment, it was shown

3614

SAP PEPTIDE INHIBITS LPS RESPONSES IN HUMAN BLOOD

FIGURE 12. Comparison of the LPS-binding motifs of several LPS-binding proteins with the proposed LPS-binding region of SAP. Basic residues are marked in boldface. Hydrophobic residues are italicized.

that 83% of the initial activity of the SAP peptide, pep27–38, was confined to a hexapeptide, pep33–38 (44). When heparin binding was studied, pep33–38 was even found to have 10-fold higher activity than pep27–38 (43). However, in the present study, investigating the interactions between SAP and LPS, the hexapeptide, pep33–38, was about 40-fold less active than pep27–39 in binding to LPS (data not shown). The LPS-binding motifs of several LPS-binding proteins have already been described. LPS-binding motifs for Limulus anti-LPS factor (LALF), BPI, and LBP show high sequence similarity with an alternating series of positively charged and hydrophobic residues with a proposed ability to produce an amphipatic loop that binds to the lipid A part of LPS (45, 46). In addition, CAP18 shows an LPS-binding region that contains a high number of basic and hydrophobic residues (47). We show a new LPS-binding motif within SAP, which does not contain the usual stretches of positively charged residues that are found in other LPS-binding proteins (Fig. 12). Much to our surprise, a peptide of only 13 amino acids mimicked the effect of the whole protein in binding to and neutralization of LPS. Even more surprising, this peptide inhibited LPSinduced responses in human blood, whereas SAP did not. We propose that at least one reason for this phenomenon is a competition between SAP and LBP for binding to LPS, as we showed that increasing LBP concentrations abolished the inhibiting effects of SAP. Pep27–39, in contrary, still was able to inhibit LPS responses in the presence of high concentrations of LBP. This could be due to the fact that the pep27–39 sequence within the SAP molecule appears not to be readily available, as it is partly situated on a b-strand under a short a-helix (14). Once liberated from the rest of the protein, pep27–39, better than SAP, competes with LBP for binding to LPS. Another explanation might be the fact that SAP is a rather large protein. Therefore, pep27–39, only 13 amino acids in size, will be more capable than SAP in binding all LPS molecules in an LPS aggregate, and thereby more efficiently shield them from the action of LBP. Until now, we described the effects of SAP on LPS binding to membrane-bound CD14 (mCD14). Using the ability of FITC-LPS to increase its fluorescence signal and shift its retention time via binding to rsCD14, we showed that SAP was not able to prevent FITC-LPS binding to rsCD14, even in the absence of LBP, while it could inhibit binding of LPS to mCD14. sCD14, just like LBP, acts as a lipid transfer molecule. In contrast to mCD14, sCD14 does not need LBP to bind LPS, although LBP accelerates binding of LPS to sCD14 (32). We could also demonstrate this accelerated LPS binding to sCD14 by LBP in our gel-filtration assay, as the fluorescence peak in the presence of LBP was about fivefold higher as compared with the fluorescence peak when no LBP was added. As SAP was not able to interfere in the binding of LPS to rsCD14, we hypothesize that rsCD14 competes with SAP for binding to LPS, just as LBP does. The finding that pep27–39 is still capable of interfering with the binding of LPS to rsCD14 strengthens this hypothesis. We can thus conclude that SAP competes with both LBP and sCD14 for binding to LPS, but that SAP-derived

peptide pep27–39 is capable of preventing LPS binding to mCD14 as well as sCD14. We discovered SAP as a novel LPS-binding protein in human plasma. As SAP did not neutralize LPS responses in human blood, its role in the pathophysiology of Gram-negative infections has yet to be elucidated. However, a 13-mer peptide, pep27–39, derived from SAP was found to bind to LPS. Its carboxamidomethylated form was even more potent in binding to LPS. The ability of pep27–39 to antagonize specifically the effects of LPS in the complex environment of human blood suggests that pep27–39 may be a novel therapeutic agent in the defense against Gram-negative sepsis. We are currently investigating the capacity of other SAP-derived peptides to bind and neutralize LPS.

Acknowledgments We thank Fridolin van der Lecq and Dr. Ton Aarsman from the Sequence Center Utrecht (Institute of Biomembranes, Utrecht University, The Netherlands) for their help in obtaining protein sequences, and Dr. Ruurd van der Zee (Institute of Infectious Diseases and Immunology, Utrecht University) for the synthesis of the peptides.

References 1. Haziot, A., S. Chen, E. Ferrero, M. G. Low, R. Silber, and S. M. Goyert. 1988. The monocyte differentiation antigen, CD14, is anchored to the cell membrane by a phosphatidylinositol linkage. J. Immunol. 141:547. 2. Wright, S. D., R. A. Ramos, A. Hermanowski Vosatka, P. Rockwell, and P. A. Detmers. 1991. Activation of the adhesive capacity of CR3 on neutrophils by endotoxin: dependence on lipopolysaccharide binding protein and CD14. J. Exp. Med. 173:1281. 3. Pugin, J., C. C. Schurer Maly, D. Leturcq, A. Moriarty, R. J. Ulevitch, and P. S. Tobias. 1993. Lipopolysaccharide activation of human endothelial and epithelial cells is mediated by lipopolysaccharide-binding protein and soluble CD14. Proc. Natl. Acad. Sci. USA 90:2744. 4. Yu, B., and S. D. Wright. 1996. Catalytic properties of lipopolysaccharide (LPS) binding protein: transfer of LPS to soluble CD14. J. Biol. Chem. 271:4100. 5. Bone, R. C. 1991. The pathogenesis of sepsis. Ann. Intern. Med. 115:457. 6. Wurfel, M. M., S. T. Kunitake, H. Lichenstein, J. P. Kane, and S. D. Wright. 1994. Lipopolysaccharide (LPS)-binding protein is carried on lipoproteins and acts as a cofactor in the neutralization of LPS. J. Exp. Med. 180:1025. 7. Parker, T. S., D. M. Levine, J. C. Chang, J. Laxer, C. C. Coffin, and A. L. Rubin. 1995. Reconstituted high-density lipoprotein neutralizes Gram-negative bacterial lipopolysaccharides in human whole blood. Infect. Immun. 63:253. 8. Cowland, J. B., A. H. Johnsen, and N. Borregaard. 1995. hCAP-18, a cathelin/ pro-bactenecin-like protein of human neutrophil specific granules. FEBS Lett. 368:173. 9. Weiss, J., P. Elsbach, C. Shu, J. Castillo, L. Grinna, A. Horwitz, and G. Theofan. 1992. Human bactericidal/permeability-increasing protein and a recombinant NH2-terminal fragment cause killing of serum-resistant Gram-negative bacteria in whole blood and inhibit tumor necrosis factor release induced by the bacteria. J. Clin. Invest. 90:1122. 10. Marra, M. N., C. G. Wilde, M. S. Collins, J. L. Snable, M. B. Thornton, and R. W. Scott. 1992. The role of bactericidal/permeability-increasing protein as a natural inhibitor of bacterial endotoxin. J. Immunol. 148:532. 11. Ohno, N., and D. C. Morrison. 1989. Lipopolysaccharide interaction with lysozyme: binding of lipopolysaccharide to lysozyme and inhibition of lysozyme enzymatic activity. J. Biol. Chem. 264:4434. 12. Wang, D., K. M. Pabst, Y. Aida, and M. J. Pabst. 1995. Lipopolysaccharideinactivating activity of neutrophils is due to lactoferrin. J. Leukocyte Biol. 57:865. 13. Kalaria, R. N., P. G. Galloway, and G. Perry. 1991. Widespread serum amyloid P immunoreactivity in cortical amyloid deposits and the neurofibrillary pathology of Alzheimer’s disease and other degenerative disorders. Neuropathol. Appl. Neurobiol. 17:189.

The Journal of Immunology 14. Emsley, J., H. E. White, B. P. O’Hara, G. Oliva, N. Srinivasan, I. J. Tickle, T. L. Blundell, M. B. Pepys, and S. P. Wood. 1994. Structure of pentameric human serum amyloid P component. Nature 367:338. 15. Botto, M., P. N. Hawkins, M. C. M. Bickerstaff, J. Herbert, A. E. Bygrave, A. McBride, W. L. Hutchinson, G. A. Tennent, M. J. Walport, and M. B. Pepys. 1997. Amyloid deposition is delayed in mice with targeted deletion of the serum amyloid P component gene. Nat. Med. 3:855. 16. Janciauskiene, S., P. Garcia de Frutos, E. Carlemalm, B. Dahlback, and S. Eriksson. 1995. Inhibition of Alzheimer b-peptide fibril formation by serum amyloid P component. J. Biol. Chem. 270:26041. 17. Steel, D. M., and A. S. Whitehead. 1994. The major acute phase reactants: Creactive protein, serum amyloid P component and serum amyloid A protein. Immunol. Today 15:81. 18. Serban, D., and C. Rordorf Adam. 1987. Binding characteristics of human serum amyloid P component. Scand. J. Immunol. 25:275. 19. Hicks, P. S., L. Saunero Nava, T. W. Du Clos, and C. Mold. 1992. Serum amyloid P component binds to histones and activates the classical complement pathway. J. Immunol. 149:3689. 20. Heegaard, N. H., H. D. Mortensen, and P. Roepstorff. 1995. Demonstration of a heparin-binding site in serum amyloid P component using affinity capillary electrophoresis as an adjunct technique. J. Chromatogr. A. 717:83. 21. Garcia de Frutos, P., Y. Hardig, and B. Dahlback. 1995. Serum amyloid P component binding to C4b-binding protein. J. Biol. Chem. 270:26950. 22. Ying, S. C., A. T. Gewurz, H. Jiang, and H. Gewurz. 1993. Human serum amyloid P component oligomers bind and activate the classical complement pathway via residues 14 –26 and 76 –92 of the A chain collagen-like region of C1q. J. Immunol. 150:169. 23. Brown, M. R., and B. E. Anderson. 1993. Receptor-ligand interactions between serum amyloid P component and model soluble immune complexes. J. Immunol. 151:2087. 24. Bristow, C. L., and R. J. Boackle. 1986. Evidence for the binding of human serum amyloid P component to Clq and Fab. Mol. Immunol. 23:1045. 25. Gewurz, H., X. H. Zhang, and T. F. Lint. 1995. Structure and function of the pentraxins. Curr. Opin. Immunol. 7:54. 26. Van der Zee, R., S. M. Anderton, C. A. F. Buskens, E. Alonso de Velasco, and W. Van Eden. 1995. Heat shock protein T cell epitopes as immunogenic carriers in subunit vaccines. In Peptides 1994, Proceedings of the Twenty-Third European Peptide Symposium. H. L. S. Maya, ed. ESCOM, Leiden, pp. 841– 842. 27. Ellman, G. L. 1959. Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82:70. 28. Weersink, A. J., K. P. van Kessel, M. E. van den Tol, J. A. van Strijp, R. Torensma, J. Verhoef, P. Elsbach, and J. Weiss. 1993. Human granulocytes express a 55-kDa lipopolysaccharide-binding protein on the cell surface that is identical to the bactericidal/permeability-increasing protein. J. Immunol. 150: 253. 29. Skinner, M., and A. S. Cohen. 1988. Amyloid P component. Methods Enzymol. 163:523. 30. Troelstra, A., B. N. G. Giepmans, K. P. M. Van Kessel, H. S. Lichenstein, J. Verhoef, and J. A. G. Van Strijp. 1997. Dual effects of soluble CD14 on LPS priming of neutrophils. J. Leukocyte Biol. 61:173. 31. Troelstra, A., P. Antal-Szalmas, L. A. M. de Graaf-Miltenburg, A. J. L. Weersink, J. Verhoef, K. P. M. Van Kessel, and J. A. G. Van Strijp. 1997. Saturable CD14dependent binding of fluorescein-labeled lipopolysaccharide to human monocytes. Infect. Immun. 65:2272.

3615 32. Hailman, E., H. S. Lichenstein, M. M. Wurfel, D. S. Miller, D. A. Johnson, M. Kelley, L. A. Busse, M. M. Zukowski, and S. D. Wright. 1994. Lipopolysaccharide (LPS)-binding protein accelerates the binding of LPS to CD14. J. Exp. Med. 179:269. 33. Loveless, R. W., O. S. Floyd, J. G. Raynes, C. T. Yuen, and T. Feizi. 1992. Human serum amyloid P is a multispecific adhesive protein whose ligands include 6-phosphorylated mannose and the 3-sulphated saccharides galactose, Nacetylgalactosamine and glucuronic acid. EMBO J. 11:813. 34. Schwalbe, R. A., B. Dahlback, J. E. Coe, and G. L. Nelsestuen. 1992. Pentraxin family of proteins interact specifically with phosphorylcholine and/or phosphorylethanolamine. Biochemistry 31:4907. 35. Weiss, J., P. Elsbach, I. Olsson, and H. Odeberg. 1978. Purification and characterization of a potent bactericidal and membrane active protein from the granules of human polymorphonuclear leukocytes. J. Biol. Chem. 253:2664. 36. Flegel, W. A., M. W. Baumstark, C. Weinstock, A. Berg, and H. Northoff. 1993. Prevention of endotoxin-induced monokine release by human low- and highdensity lipoproteins and by apolipoprotein A-I. Infect. Immun. 61:5140. 37. Landsmann, P., O. Rosen, M. Pontet, M. Pras, D. Levartowsky, E. G. Shephard, and M. Fridkin. 1994. Binding of human serum amyloid P component (hSAP) to human neutrophils. Eur. J. Biochem. 223:805. 38. Wright, S. D., L. S. Craigmyle, and S. C. Silverstein. 1983. Fibronectin and serum amyloid P component stimulate C3b- and C3bi-mediated phagocytosis in cultured human monocytes. J. Exp. Med. 158:1338. 39. Siripont, J., J. M. Tebo, and R. F. Mortensen. 1988. Receptor-mediated binding of the acute-phase reactant mouse serum amyloid. Cell. Immunol. 117:239. 40. Singh, P. P., F. Gervais, E. Skamene, and R. F. Mortensen. 1986. Serum amyloid P-component-induced enhancement of macrophage listericidal activity. Infect. Immun. 52:688. 41. Pepys, M. B., M. L. Baltz, F. C. de Beer, R. F. Dyck, S. Holford, and S. M. Breathnach. 1982. Biology of serum amyloid P component. Ann. NY Acad. Sci. 389:286. 42. Hind, C. R., P. M. Collins, M. L. Baltz, and M. B. Pepys. 1985. Human serum amyloid P component, a circulating lectin with specificity for the cyclic 4,6pyruvate acetal of galactose: interactions with various bacteria. Biochem. J. 225: 107. 43. Heegaard, N. H. H., P. M. H. Heegaard, P. Roepstorff, and F. A. Robey. 1996. Ligand-binding sites in human serum amyloid P component. Eur. J. Biochem. 239:850. 44. Dhawan, S., R. L. Fields, and F. A. Robey. 1990. A novel peptide from amyloid P component supports cell attachment. Biochem. Biophys. Res. Commun. 171: 1284. 45. Hubacek, J. A., C. Bu¨chler, C. Aslanidis, and G. Schmitz. 1997. The genomic organization of the genes for human lipopolysaccharide binding protein (LBP) and bactericidal permeability increasing protein (BPI) is highly conserved. Biochem. Biophys. Res. Commun. 236:427. 46. Hoess, A., S. Watson, G. R. Siber, and R. Liddington. 1993. Crystal structure of an endotoxin-neutralizing protein from the horseshoe crab, Limulus anti-LPS factor, at 1.5 A resolution. EMBO J. 12:3351. 47. Larrick, J. W., M. Hirata, H. Zheng, J. Zhong, D. Bolin, J. M. Cavaillon, H. S. Warren, and S. C. Wright. 1994. A novel granulocyte-derived peptide with lipopolysaccharide-neutralizing activity. J. Immunol. 152:231.