Demonstration of peptidoglycan-binding sites on lymphocytes and macrophages by photoaffinity cross-linking

One dominant binding site (70 kDa 6.5 pI protein) for bacterial cell wall peptidoglycan (PGN), a macrophage activator and polyclonal B cell mitogen, was demonstrated on mouse B and T lymphocytes and macrophages by photoaffinity cross-linking and two-dimensional polyacrylamide gel electrophoresis. This binding site was not present on erythrocytes. The binding was specific for polymeric PGN and was competitively inhibited by unlabeled PGN with IC50 = 48 micrograms/ml (0.38 microM). The binding was partially inhibited by O-acetylated PGN monomers (IC50 = 469 micrograms/ml, 521 microM), dextran sulfate (IC50 = 1024 micrograms/ml, 124 microM), and (GlcNAc)3 (IC50 = 6.6 mg/ml, 10 mM), and was not inhibited by non-O-acetylated PGN monomers and dimers, muramyl dipeptide, PGN pentapeptide, GlcNAc, teichoic acid, protein A, and gelatin. The cell surface location of the 70-kDa PGN-binding protein was indicated by the ability of PGN to bind to this protein in intact metabolically inactive cells (at 4 degrees C and in the presence of 0.1% NaN3) and by the ability to extract the 70-kDa PGN-binding protein from viable B lymphocytes by noncytotoxic concentration of n-octyl-beta-D-glucopyranoside.     

Specific endotoxic lipopolysaccharide-binding proteins on murine splenocytes. II. Membrane localization and binding characteristics

We have characterized the binding of LPS to an 80-kDa LPS-binding protein detected by an LPS photoaffinity probe to be present on murine splenocytes. Specific binding of LPS to the 80-kDa protein is directly proportional to LPS concentration at low concentrations of LPS and is saturable at high concentrations of LPS. Binding is inhibited by both homologous and heterologous underivatized LPS as well as by polysaccharide-free lipid A, indicating a specificity for the biologically active component of LPS. Analysis of the kinetics of binding indicate a time-dependent increase over the first 15 min, but increases are not detected after this time. Binding of LPS to the 80-kDa LPS-binding protein is reduced but still readily detectable at 4 degrees C in the presence of azide. The presence of the 80-kDa LPS-binding protein in an isolated cytoplasmic membrane fraction of murine splenocytes as well as its release from intact splenocytes by octylglucoside suggest that this LPS-binding protein is membrane localized. The results are consistent with, but do not establish unequivocally, the identity of the 80-kDa LPS-binding protein as a specific membrane receptor for lipid A.     

Bacterial peptidoglycan binds to tubulin

A search for cellular binding proteins for peptidoglycan (PGN), a CD14- and TLR2-dependent macrophage activator from Gram-positive bacteria, using PGN-affinity chromatography and N-terminal micro-sequencing, revealed that tubulin was a major PGN-binding protein in mouse macrophages. Tubulin also co-eluted with PGN from anti-PGN vancomycin affinity column and bound to PGN coupled to agarose. Tubulin-PGN binding was preferential under the conditions that promote tubulin polymerization, required macromolecular PGN, was competitively inhibited by soluble PGN and tubulin, did not require microtubule-associated proteins, and had an affinity of 100-150 nM. By contrast, binding of tubulin to lipopolysaccharide (LPS) had 2-3 times lower affinity, faster kinetics of binding, and showed positive cooperativity. PGN enhanced tubulin polymerization in the presence of 4 M glycerol, but in the absence of glycerol, both PGN and LPS decreased microtubule polymerization. These results indicate that tubulin is a major PGN-binding protein and that PGN modulates tubulin polymerization.     

Identification of lipopolysaccharide-binding proteins in 70Z/3 cells by photoaffinity cross-linking

A radioiodinated, photoactivatable derivative of Salmonella minnesota Re595 lipopolysaccharide (LPS) was used to label LPS-binding proteins in 70Z/3 cells. The labeled proteins were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and visualized by autoradiography. 125I-Labeled-2-(p-azidosalycylamido)1,3'-dithiopropionamide S. minnesota Re595 LPS (125I-ASD-Re595) labeled a limited number of proteins. The most prominent of these had a apparent molecular mass of 18 kDa. Less prominent labeling of 25- and 28-kDa proteins was also seen. Labeling was saturated by 5 micrograms/ml 125I-ASD-Re595 and was inhibited by a 10-100-fold excess of unlabeled LPS or lipid A. Labeling was maximal within 30 min at 37 degrees C; much less labeling occurred at lower temperatures. The proteins labeled with 125I-ASD-Re595 appear to be on the surface of the cell, since they can be digested by trypsin and were found in the membrane fraction of the cell but not in the cytosol. Studies with competitive inhibitors suggested that the proteins bind to the lipid A region of the LPS molecule. Biologically inactive lipid A analogs were poor inhibitors of labeling, suggesting that the LPS-binding proteins could discriminate between active lipid A and inactive analogs. These studies suggest that the 18- and 25-kDa proteins bind specifically to the lipid A region of the LPS molecule and should be considered as candidates for a functional LPS receptor.     

Isolation of a lipopolysaccharide (LPS)-resistant mutant, with defective LPS binding, of cultured macrophage-like cells.

Lipopolysaccharide (LPS)-resistant mutants which did not respond to LPS were isolated from a macrophage-like mouse cell line, J774.1. Unlike the parental J774.1 cells, these mutants grew even in LPS added medium as well as in normal growth medium without any morphological changes. Assay of 125I-LPS binding to the cell monolayers revealed that one of these LPS-resistant mutants (LR-9) was strikingly defective in LPS-binding activity. Scatchard plot showed that LR-9 cells lacked the high affinity binding sites which were present in J774.1. The high affinity binding was inhibited by addition of excess unlabeled LPS, lipid A, lipid IVA (tetraacyl-beta(1'-6)-linked D-glucosamine disaccharide-1,4'-bisphosphate), and lipid X (2,3-diacylglucosamine 1-phosphate) and sensitive to proteinase K. LPS enhanced O2- generation and the release of arachidonic acid in J774.1 cells but not in LR-9 cells. Other stimulants such as zymosan and 12-O-tetradecanoylphorbol 13-acetate, however, induced the release of arachidonic acid in LR-9 cells as well as in J774.1 cells. LPS-photocross-linked assay allowed the identification of 65- and 55-kDa LPS-binding proteins in the membrane fraction of J774.1 cells. Both of the bands were not detectable in that of LR-9 cells and disappeared by competing with unlabeled LPS or lipid X. These results show that one or both of the two LPS-binding proteins might relate to the specific membrane receptor for LPS.     

Involvement of mitogen-activated protein kinase homologues in the regulation of lipopolysaccharide-mediated induction of cyclo-oxygenase-2 but not nitric oxide synthase in RAW 264.7 macrophages.

In RAW 264.7 macrophages lipopolysaccharide (LPS) stimulated the activation of p42 and p44 MAP kinases and their upstream activator mitogen-activated protein (MAP) kinase kinase (MAPKK), and induced the 69-kDa isoform of cyclo-oxygenase-2 (COX-2) and the 130-kDa isoform of nitric oxide synthase (iNOS). PD 098059, a specific inhibitor of the activation of MAPKK, prevented LPS-mediated activation of MAPKK (IC50 = 3.0 +/- 0.1 microM, n = 3) and p42/44 MAP kinases and substantially reduced the induction of COX-2 by approximately 40%-70%, but was without effect upon the induction of iNOS. In parallel, LPS also stimulated the activation of p38 MAP kinase and the MAPKAP kinase-2, a downstream target of p38 MAP kinase. SB 203580, a specific inhibitor of p38 MAP kinase prevented the activation of p38 MAP kinase (IC50 = 3.3 +/- 1.4 microM, n = 3) and MAPKAP kinase-2 by LPS and reduced the induction of COX-2 by approximately 50-90%, with no significant effect upon iNOS expression. These studies indicate the involvement of both the classical p42/44 MAP kinases and p38 MAP kinase in the regulation of COX-2 but not iNOS induction following exposure to LPS.     

Specific endotoxic lipopolysaccharide-binding proteins on murine splenocytes. I. Detection of lipopolysaccharide-binding sites on splenocytes and splenocyte subpopulations

Experiments have been carried out using a unique radio-iodinated, disulfide-reducible, photoactivatable LPS derivative (ASD-LPS) to detect specific LPS-binding proteins on murine splenocytes. Fractionation of LPS-photo-cross-linked, reduced, and solubilized splenocyte extracts on two-dimensional polyacrylamide gels has allowed the identification of an 80-kDa LPS-binding protein with approximate pI of 6.5. This LPS-binding protein is present on partially purified populations of splenic B lymphocytes, T lymphocytes, and macrophages. It is also the dominant LPS-binding protein on the murine 70Z/3 B cell line and the YAC-1 and EL4 T cell lines but is not detectable on the undifferentiated murine Sp2/0 myeloma cell line. Of potential importance is the fact that the 80-kDa protein appears to be indistinguishable when photolabeled extracts of splenocytes from the C3HeB/FeJ (lpsn) and LPS-nonresponder C3H/HeJ (lpsd) mice are compared.     

Purification and characterization of an endotoxin-binding protein with protease inhibitory activity from Limulus amebocytes

Using a lipopolysaccharide affinity column and ion exchange chromatography, a 12-kDa protein has been purified from Limulus amebocytes. In solid phase binding assays, the radiolabeled protein binds specifically to lipopolysaccharide (LPS) with a Kd value on the order of 10(-7) M. A cDNA coding for this protein has been isolated and sequenced. The amino acid sequence deduced from the cDNA indicates that this protein shares no sequence homology with LPS-binding proteins isolated from different species of vertebrates (Schumann, R. R., Leong, S. R., Flaggs, G. W., Gray, P. W., Wright, S. D., Mathison, J. C., Tobias, P. S., and Ulevitch, R. J. (1990) Science 249, 1429-1431) and invertebrates (Aketagawa, J., Miyata, T., Ohtsubo, S., Nakamura, T., Morita, T., Hayashida, H., Miyata, T., Iwanaga, S., Takao, T., and Shimonishi, Y. (1986) J. Biol. Chem. 261, 7357-7365). The binding to LPS can be displaced by the unlabeled 12-kDa protein, polymyxin B, lipid A, and to a lesser extent by D-glucosamine. In whole cell binding assays, the 12-kDa protein has also been shown to bind to Escherichia coli. Using both [14C]casein and a synthetic substrate, the protein has been shown to inhibit the proteolytic activity of trypsin, with an IC50 of approximately 10(-7) M. In the presence of LPS, the antitryptic acitivity of the Limulus endotoxin-binding protein-protease inhibitor remains unaffected. The protein is a major component of the cytoplasmic proteins (1%). Immunocytochemical analysis reveals that this protein exists in the secretory granules of the amebocytes where enzymes and substrates for the clotting cascade reside. Based on the unusual dual functional properties, the newly isolated protein was named a "Limulus endotoxin-binding protein-protease inhibitor" (LEBP-PI).     

Quantitative analysis of albumin uptake and transport in the rat microvessel endothelial monolayer

We determined the concentration dependence of albumin binding, uptake, and transport in confluent monolayers of cultured rat lung microvascular endothelial cells (RLMVEC). Transport of (125)I-albumin in RLMVEC monolayers occurred at a rate of 7.2 fmol. min(-1). 10(6) cells(-1). Albumin transport was inhibited by cell surface depletion of the 60-kDa albumin-binding glycoprotein gp60 and by disruption of caveolae using methyl-beta-cyclodextrin. By contrast, gp60 activation (by means of gp60 cross-linking using primary and secondary antibodies) increased (125)I-albumin uptake 2.3-fold. At 37 degrees C, (125)I-albumin uptake had a half time of 10 min and was competitively inhibited by unlabeled albumin (IC(50) = 1 microM). Using a two-site model, we estimated by Scatchard analysis the affinity (K(D)) and maximal capacity (B(max)) of albumin uptake to be 0.87 microM (K(D1)) and 0.47 pmol/10(6) cells (B(max1)) and 93.3 microM (K(D2)) and 20.2 pmol/10(6) cells (B(max2)). At 4 degrees C, we also observed two populations of specific binding sites, with high (K(D1) = 13.5 nM, 1% of the total) and low (K(D2) = 1.6 microM) affinity. On the basis of these data, we propose a model in which the two binding affinities represent the clustered and unclustered gp60 forms. The model predicts that fluid phase albumin in caveolae accounts for the bulk of albumin internalized and transported in the endothelial monolayer.     

Localization of the lipopolysaccharide-binding protein in phospholipid membranes by atomic force microscopy

Lipopolysaccharides (LPS; endotoxin) activate immunocompetent cells of the host via a transmembrane signaling process. In this study, we investigated the function of the LPS-binding protein (LBP) in this process. The cytoplasmic membrane of the cells was mimicked by lipid liposomes adsorbed on mica, and the lateral organization of LBP in these membranes and its interaction with LPS aggregates were characterized by atomic force microscopy. Using cantilever tips functionalized with anti-LBP antibodies, single LBP molecules were localized in the membrane at low concentrations. At higher concentrations, LBP formed clusters of several molecules and caused cross-linking of lipid bilayers. The addition of LPS to LBP-containing liposomes led to the formation of LPS domains in the membranes, which could be inhibited by anti-LBP antibodies. Thus, LBP mediates the fusion of lipid membranes and LPS aggregates.     

Preparation and characterization of truncated human lipopolysaccharide-binding protein in Escherichia coli

Lipopolysaccharide (LPS) is a component of the outer membrane of Gram-negative bacteria, and is the causative agent of endotoxin shock. LPS induces signal transduction in immune cells when it is recognized by the cell surface complex of toll-like receptor 4 (TLR4) and MD-2. The complex recognizes the lipid A structure in LPS, which is buried in the membrane of the outer envelope. To present the Lipid A structure to the TLR4/MD-2, processing of LPS by LPS-binding protein (LBP) and CD14 is required. In previous studies, we expressed recombinant proteins of human MD-2 and CD14 as fusion proteins with thioredoxin in Escherichia coli, and demonstrated their specific binding abilities to LPS. In this study, we prepared a recombinant fusion protein containing 212 amino terminal residues of human LBP (HLB212) by using the same expression system. The recombinant protein expressed in E. coli was purified as a complex form with host LPS. The binding was not affected by high concentrations of salt, but was prevented by low concentrations of various detergents. Both rough-type LPS lacking the O antigen and smooth-type LPS with the antigen bound to HLBP212. Therefore, oligosaccharide repeats appeared to be unnecessary for the binding. A nonpathogenic penta-acylated LPS also bound to HLBP212, but the binding was weaker than that of the wild type. The hydrophobic interaction between the LBP and acyl chains of lipid A appears to be important for the binding. The recombinant proteins of LPS-binding molecules would be useful for analyzing the defense mechanism against infections.     

Calmodulin-mediated adenylate cyclase from mammalian sperm

Calmodulin (CaM), the calcium binding protein that modulates the activity of a number of key regulatory enzymes, is present at high levels in sperm. To determine whether CaM regulates adenylate cyclase in mammalian sperm, the actions of EGTA and selected CaM antagonists on a solubilized adenylate cyclase from mature equine sperm were examined. The activity of equine sperm adenylate cyclase was inhibited by EGTA in a concentration-dependent manner with a half-maximal inhibitory concentration (IC50) of 2 mM. Equine sperm adenylate cyclase was also inhibited in a concentration-dependent manner by the CaM antagonists chlorpromazine and calmidazolium (IC50 = 400 and 50 microM, respectively). The inhibition of enzyme activity by these agents correlated with their known potency and specificity as anti-CaM agents. The activity of the enzyme in the presence of 200 microM calmidazolium was restored by the addition of authentic CaM (EC50 = 15 microM); full activity was restored by the addition of 50 microM CaM. La3+, an ion that dissociates CaM from tightly bound CaM-enzyme systems, inhibited equine sperm adenylate cyclase (IC50 = 1 mM). Incubation of equine sperm adenylate cyclase with La3+ dissociated endogenous CaM from the enzyme so that most of the enzyme bound to a CaM-Sepharose column equilibrated with Ca2+. Specific elution of CaM-binding proteins from the CaM-Sepharose column with EGTA yielded a CaM-depleted adenylate cyclase fraction that was stimulated 2-fold by the addition of exogenous CaM.     

Membrane interactions of amphiphilic polypeptides mastoparan, melittin, polymyxin B, and cardiotoxin. Differential inhibition of protein kinase C, Ca2+/calmodulin-dependent protein kinase II and synaptosomal membrane Na,K-ATPase, and Na+ pump and differentiation of HL60 cells.

Interactions of certain naturally occurring, amphiphilic polypeptides with membranes were investigated. Mastoparan (wasp venom toxin), melittin (bee venom toxin), cardiotoxin (cobra venom toxin), and polymyxin B (antibacterial antibiotic) inhibited protein kinase C stimulated by phosphatidylserine bilayer or arachidonate monomer and blocked binding of [3H] phorbol 12,13-dibutyrate to protein kinase C in the presence of phosphatidylserine bilayer, with IC50 values (concentrations causing 50% inhibition) of 1-8 microM. Mastoparan and polymyxin B were much less inhibitory (IC50, 10-20 microM), whereas melittin and cardiotoxin were similarly inhibitory (IC50, 1-4 microM), when protein kinase C was activated instead by synaptosomal membrane. Kinetic analysis indicate that mastoparan inhibited protein kinase C, assayed using phosphatidylserine or synaptosomal membrane as the phospholipid cofactor, competitively with the phospholipid cofactor, in a mixed manner with CaCl2 or diacylglycerol, noncompetitively with histone, and uncompetitively with ATP, with apparent Ki values of 1.6-18.7 microM. Inhibition of Na,K-ATPase in the membrane by these polypeptides had relative potencies different from those for their inhibition of protein kinase C activated by the same membrane preparation; mastoparan and melittin inhibited the two activities with comparable potencies, but polymyxin B and cardiotoxin were far less effective in inhibiting Na,K-ATPase. The same relative inhibitory potencies of the polypeptides (melittin greater than mastoparan greater than polymyxin B) for inhibition of Na,K-ATPase were also noted for their inhibition of Ca2+/calmodulin-dependent protein kinase II, 86Rb uptake (Na+ pump) by HL60 cells and the phorbol ester-induced differentiation of the leukemia cells. These findings were consistent with discrete interactions of the polypeptides with functionally distinct sites on the membrane, leading to differential inhibition of biological activities associated with the membrane. Actions of certain polypeptides appeared to be more specific compared to those of lipid second messengers such as lyso-phosphatidylcholine and sphingosine, and the antineoplastic ether lipid analogs such as 1-O-octadecyl-2-methyl-rac-glycero-3-ophosphocholine.     

Promoting protein, a silkworm hemolymph protein promoting in vitro replication of nucleopolyhedrovirus, binds to beta-glucans

A protein which bound to 125I-labeled peptidoglycan (PGN) was isolated from hemolymph of silkworm larvae. The N-terminal amino acid sequence and the molecular weight of the protein were in accord with those described for Promoting Protein (PP) from the silkworm. The binding of the protein to [125I]PGN was competitively inhibited by various beta-glucans. The binding kinetics of PGN and chitin to the protein were analyzed in a biosensor.     

Identification and characterization of calmodulin-binding proteins in mammalian sperm flagella

A calcium and calmodulin-regulated cyclic nucleotide phosphodiesterase has been shown to be an integral component of both rat and bovine sperm flagella. The calcium-activated enzyme was inhibited by both trifluoperazine (ID50 = 10 microM) and [ethylene-bis(oxyethylenenitrilo)]tetraacetic acid (EGTA), and the basal activity measured in the presence of EGTA was stimulated by limited proteolysis to that observed in the presence of calcium/calmodulin. 125I-Calmodulin binding to purified rat sperm flagella has been characterized and the flagellar-associated calmodulin-binding proteins identified by a combination of gel and nitrocellulose overlay procedures and by chemical cross-linking experiments using dimethyl suberimidate. 125I-Calmodulin bound to demembranated rat sperm flagella in a time- and concentration-dependent manner. At equilibrium, 30-40% of the bound 125I-calmodulin remains associated with the flagella after treatment with EGTA or trifluoperazine. The majority of the bound 125I-calmodulin, both the Ca2+-dependent and -independent, was displaced by excess calmodulin. A 67-kDa calmodulin-binding protein was identified by both the gel and nitrocellulose overlay procedures. In both cases, binding was dependent on Ca2+ and was totally inhibited by trifluoperazine, EGTA, and excess calmodulin. On nitrocellulose overlays, the concentration of calmodulin required to decrease binding of 125I-calmodulin by 50% was between 10(-10) and 10(-11) M. Limited proteolysis resulted in the total loss of all Ca2+-dependent binding to the 67-kDa polypeptide. Chemical cross-linking experiments identified a major calcium-dependent 125I-calmodulin:polypeptide complex in the 84-90-kDa molecular mass range and a minor complex of approximately 200 kDa. Immunoblot analysis showed that the major 67-kDa calmodulin-binding protein did not cross-react with polyclonal antibodies raised against either the calcium/calmodulin-regulated cyclic nucleotide phosphodiesterase or phosphoprotein phosphatase (calcineurin) from bovine brain.     

Inhibition by gossypol of phospholipid-sensitive Ca2+-dependent protein kinase from pig testis

Gossypol, a polyphenolic binaphthalene-dialdehyde extracted from cotton plants which possesses male antifertility action in mammals, is a potent inhibitor of phospholipid-sensitive Ca2+-dependent protein kinase from pig testis. Gossypol inhibited Ca2+-dependent activity of the enzyme without affecting its basal activity. The IC50 value (concentration causing 50% inhibition) was 31 microM when lysine-rich histone was used as substrate. Kinetic analysis indicated that the compound inhibited the enzyme non-competitively with respect to ATP (Ki = 31 microM) or lysine-rich histone (Ki = 30 microM), and competitively with respect to phosphatidylserine (Ki = 2.1 microM). With Ca2+, irrespective of the presence or absence of 1,3-diolein, the compound lowered Vmax and increased the apparent Ka for Ca2+. The compound also inhibited phosphorylation by the enzyme of high-mobility-group 1 protein (one of the endogenous substrates in the testis for the enzyme located in nucleosome), with an IC50 value of 88 microM. These results suggested that a phospholipid-sensitive Ca2+-dependent protein phosphorylation system in the testis is involved in the regulation of spermatogenesis.     

High-throughput characterization of lipopolysaccharide-binding proteins using mass spectrometry

Lipopolysaccharide (LPS)-binding proteins interact with LPS in human serum and mediate various immune responses. We describe a high-throughput LPS-binding protein profiling platform for discovering unknown LPS-binding proteins and potential inflammatory mediators. As a pull-down method, the LPS molecules were immobilized onto epoxy beads and then directly incubated with human serum to screen LPS-binding proteins. Through the "untargeted" mass spectrometric approach, potential LPS-binding proteins which elicit various immune responses in human serum were identified by a highly sensitive LTQ Orbitrap Hybrid Fourier Transform Mass Spectrometer (LTQ Orbitrap FT MS). Therefore, this mass spectrometry (MS)-based profiling method is straightforward for screening unknown LPS-binding proteins and provides physiologically relevant binding partners in human serum.     

The inhibitory effects of okadaic acid on platelet function.

Okadaic acid (OA), a potent inhibitor of protein phosphatases type 1 and type 2A, inhibited thrombin-induced platelet aggregation (IC50 = 0.8 microM), [14C]serotonin release and increase in intracellular Ca2+ ([Ca2+]i) in the same dose dependence. In the absence of thrombin OA increased the phosphorylation of 50-kDa protein and 20-kDa myosin light chain (MLC20). The 50-kDa protein phosphorylation was accomplished within a shorter time period and at a lower concentration than was the MLC20. OA decreased the thrombin-induced phosphorylation of 47-kDa protein and MLC20, although phosphorylation of MLC20 reincreased at higher concentrations of OA (5-10 microM). Since type 2A phosphatase is more sensitive to OA than type 1, these results suggest that type 2A phosphatases are involved in the regulation of Ca2+ signaling in thrombin-induced platelet activation.     

MD-2 binds to bacterial lipopolysaccharide

The exact roles and abilities of the individual components of the lipopolysaccharide (LPS) receptor complex of proteins remain unclear. MD-2 is a molecule found in association with toll-like receptor 4. We produced recombinant human MD-2 to explore its LPS binding ability and role in the LPS receptor complex. MD-2 binds to highly purified rough LPS derived from Salmonella minnesota and Escherichia coli in five different assays; one assay yielded an apparent KD of 65 nm. MD-2 binding to LPS did not require LPS-binding proteins LBP and CD14; in fact LBP competed with MD-2 for LPS. MD-2 enhanced the biological activity of LPS in toll-like receptor 4-transfected Chinese hamster ovary cells but inhibited LPS activation of U373 astrocytoma cells and of monocytes in human whole blood. These data indicate that MD-2 is a genuine LPS-binding protein and strongly suggest that MD-2 could play a role in regulation of cellular activation by LPS depending on its local availability.