Detection of lipopolysaccharide-binding proteins on membranes of murine lymphocyte and macrophage-like cell lines

Rhodobacter sphaeroides diphosphoryl lipid A inhibits interleukin-6 production in CD14-negative murine marrow stromal ST2 cells stimulated with lipopolysaccharide or paclitaxel (taxol)

Paclitaxel (taxol), a microtubule stabilizer with anticancer activity, mimics the actions of lipopolysaccharide (LPS) on murine macrophages in vitro. Recent studies have shown that the Rhodobacter sphaeroides diphosphoryl lipid A (RsDPLA) inhibits both LPS- and paclitaxel-induced activation of murine macrophages, and have suggested that LPS, RsDPLA, and paclitaxel share the same receptor site on murine macrophages. To analyze this receptor site, the present study focused on the interactions between LPS, RsDPLA and paclitaxel in the activation of ST2 cells derived from murine bone marrow stroma. The ST2 cells did not express CD14 mRNA. The cells produced IL-6 molecules and expressed IL-6 mRNA in response to LPS, but did not produce TNF and nitric oxide. Paclitaxel induced IL-6 mRNA expression in ST2 cells. RsDPLA inhibited both LPS- and paclitaxel-induced IL-6 mRNA expression in a dose-dependent manner. These results suggest that LPS, RsDPLA, and paclitaxel are recognized by the same receptor complex on ST2 cells, and that the receptor functions without membrane CD14.

Genetic Analysis of Innate Immunity

The inflammatory response to microbes--and host perception of microbes in general--is largely initiated by a single class of receptors, named for their similarity to the prototypic Toll receptor of Drosophila. The mammalian Toll-like receptors (TLRs) are ultimately responsible for most phenomena associated with infection. This includes both "good" effects of infection (e.g., the induction of lasting specific immunity to an infectious agent) and "bad" effects of infection (systemic inflammation and shock). Although they are essential for host defense, no other endogenous proteins can match their lethal potential. The TLR complexes transduce the toxicity of lipopolysaccharide (LPS), cysteinyl lipopeptides, and many other molecules of microbial origin. The identification of the TLRs as the key conduit to host awareness of microbial infection was a victory for reductionism, proving that the complexity of infectious inflammation as a phenomenon belies the simplicity of its origins. It was achieved by a classical genetic approach, proceeding from phenotype to gene. Further analysis of the signaling pathways activated by the TLRs has depended on both classical and reverse genetic methods. Additional work will ultimately disclose the extent to which sterile inflammatory diseases are mediated by aberrations in these pathways.

Interactions of bacterial lipopolysaccharide and peptidoglycan with a 70 kDa and an 80 kDa protein on the cell surface of CD14+ and CD14- cells

Bacterial cell wall components, lipopolysaccharide (LPS), lipoteichoic acid (LTA), and peptidoglycan (PGN) are known to stimulate cells of the immune, inflammatory and vascular systems contributing to septic shock. CD14 has been identified as the main LPS receptor, a process that is accelerated by the serum protein LPS-binding protein (LBP). CD14 has also been found to bind LTA and PGN from the cell wall of gram positive bacteria. Recently, toll-like receptor proteins TLR-2 and TLR-4 have been shown to be required for LPS and LTA-induced intracellular signalling. Although CD14 functions as either a glycosylphosphatidylinositol (GPI)-anchored molecule that does not transverse the cell membrane or as a soluble serum protein, the mechanisms by which the CD14-LPS/LTA complex interacts with the TLRs remains to be elucidated. We have looked directly for cell surface protein(s) that bind LPS or LTA in a CD14-dependent manner. Using biochemical approaches we have identified two proteins of molecular weight 70 kDa (LAP-1) and 80 kDa (LAP-2) that can be precipitated from both CD14(+) and CD14(-) cells with LPS- or LTA-specific antibodies. Binding of LPS and LTA to LAP-1 and -2 required serum. While soluble CD14 (sCD14) was sufficient to allow precipitation of these two proteins from CD14(-) cells, serum could not be replaced by purified sCD14 and/or LBP when mCD14-expressing cells were used.

Genes, receptors, signals and responses to lipopolysaccharide endotoxin

C3H/HeJ inbred mice have been very useful for identifying genetic elements responsible for endotoxin mediated responses. Depending on the type of assays employed, Tlr-2, Tlr-4 and Lps/Ran have been shown to be important in lipopolysaccharide (LPS)-mediated responses. The concept of a single LPS gene being responsible for the genetic defect found in C3H/HeJ mice should therefore be re-examined more closely. Given the most recent discoveries, it is probable that more than one signal transduction pathway is involved. One is a CD14-dependent pathway, the other a CD14-independent pathway. Identification of the genetic elements involved in these pathways will be beneficial in designing therapeutic strategies for treating patients with endotoxic or septic shock.

Detection of lipopolysaccharide (LPS)-binding membrane proteins by immuno-coprecipitation with LPS and anti-LPS antibodies

In this study we describe a general method for the detection and characterization of endotoxin-(lipopolysaccharide, LPS)-binding membrane proteins. In the past, experimental procedures to detect LPS-binding sites on cells were generally performed with chemically modified LPS derivates. Since any modification of a ligand may lead to a modification of its binding characteristics, the results of those studies are controversial. In our assay, cell membrane preparations are treated with free lipid A, the endotoxic center of LPS, in the presence of normal human serum. After binding of lipid A, membrane proteins are solubilized by mild detergent treatment without disruption of the lipid A-protein complexes. Addition of anti-(lipid A) mAbs and subsequent adding of protein A agarose lead to the precipitation of complexes of lipid A and its binding proteins. By SDS/PAGE and western blot, these precipitates can be screened for the presence of LPS/lipid A-binding proteins. We describe the use of this method for the immuno-coprecipitation of lipid A (or LPS) with an 80-kDa LPS-binding membrane protein (LMP80), which we have previously identified on several human cells. In addition, CD14, the well-known functional LPS receptor on monocytes and macrophages, can be detected. By means of this immuno-coprecipitation approach we could demonstrate binding of either purified LPS preparations or synthetic lipid A to these LPS/lipid A-binding membrane proteins at physiological pH under conditions in which the proteins are in their natural membranous environment.

Hyporesponsiveness of inflamed human gingival fibroblasts from patients with chronic periodontal diseases against cell surface components of Porphyromonas gingivalis

Inflamed human gingival fibroblasts (HGF) of patients with chronic periodontal diseases have less active interleukin-8 (IL-8) production compared with normal HGF of volunteers with healthy gingival tissues, after stimulation with Porphyromonas gingivalis surface components such as fimbriae, lipopolysaccharide (LPS) and its lipid A, but not LPS or lipid A from other bacterial species. A decrease in number of specific binding sites for P. gingivalis fimbrial molecules in inflamed HGF is also observed by Scatchard plot analysis. A short exposure (6 h) to P. gingivalis LPS resulted in significant potentiation of the LPS-dependent IL-8 production in normal HGF, whereas a long exposure (48 h) to the LPS significantly reduced IL-8 production. Tyrosine phosphorylation of proteins of 127 kDa and 186 kDa in inflamed HGF stimulated with P. gingivalis fimbriae or its LPS was observed by immunoblotting, and these two phosphoproteins were termed tolerance-induced protein, TIP. Protein bands of 45 kDa which bound to radioiodinated P. gingivalis fimbriae in the presence and absence of fetal bovine serum (FBS), and major 73-kDa and minor 30-kDa and 45-kDa bands which bound to radioiodinated P. gingivalis LPS in the presence of FBS in normal and inflamed HGF were observed by using photocrosslinking. These findings suggest that the hyporesponsiveness of HGF induced by a prolonged exposure to P. gingivalis may emerge because of HGF damage or result from host defense in chronic periodontal lesions.

Splenic B-cell activation in lipopolysaccharide-non-responsive C3H/HeJ mice by lipopolysaccharide of Porphyromonas gingivalis

Porphyromonas gingivalis 381 lipopolysaccharide (LPS) definitely exhibited mitogenic activity in purified B-cells, separated from spleens of LPS-responsive C3H/HeN mice and LPS-non-responsive C3H/HeJ mice by using a magnetic cell sorting system. The mitogenic activity induced by P. gingivalis LPS was incompletely inhibited by polymyxin B. P. gingivalis LPS also induced a higher production of interleukin-6 (IL-6) in splenic B-cells of C3H/HeN mice as compared with Escherichia coli LPS. Furthermore, P. gingivalis LPS, but not E. coli LPS, induced definite IL-6 production in C3H/HeJ mice. P. gingivalis LPS increased tyrosine, serine/threonine phosphorylation of proteins with various major induced bands in splenic B-cells of both C3H/HeN and C3H/HeJ mice. Additionally, radioiodinated P. gingivalis LPS, similarly to E. coli LPS, bound to a 73-kDa protein on C3H/HeJ as well as C3H/HeN B-cells. Thus P. gingivalis LPS may activate B-cells of C3H/HeJ as well as C3H/HeN mice via the LPS-specific binding protein on the cells.

Binding of lipopolysaccharide (LPS) to an 80-kilodalton membrane protein of human cells is mediated by soluble CD14 and LPS-binding protein

Activation of cells by bacterial lipopolysaccharide (LPS) plays a key role in the pathogenesis of gram-negative septic shock. The 55-kDa glycoprotein CD14 is known to bind LPS and initiate cell activation. However, there must be additional LPS receptors because CD14 is linked by a glycosylphosphatidyl inositol anchor to the cell membrane and therefore unable to perform transmembrane signalling. Searching for potential LPS receptors, we investigated the binding of LPS to membrane proteins of the human monocytic cell line Mono-Mac-6. Membrane proteins were electrophoretically separated under reducing conditions, transferred to nitrocellulose, and exposed to LPS, which was visualized with anti-LPS antibody. Smooth- and rough-type LPS, as well as free lipid A, bound to a variety of proteins in the absence of serum. However, in the presence of serum, additional or preferential binding to a protein of approximately 80-kDa was observed. Experiments with differently acylated lipid A structures showed that the synthetic tetraacyl compound 406 was still able to bind, whereas no binding was detected with the bisacyl compound 606. The 80-kDa membrane protein was also detected on human peripheral blood monocytes and endothelial cells. The serum factors mediating the binding of lipid A to the 80-kDa membrane protein were identified as soluble CD14 and LPS-binding protein. From these results, we conclude that this 80-kDa protein is a candidate for the hypothetical molecule for LPS and/or LPS-CD14 recognition and signal transduction.

Endotoxin activates human vascular smooth muscle cells despite lack of expression of CD14 mRNA or endogenous membrane CD14

During infection or inflammation, cells of the blood vessel wall, such as endothelial cells (EC) and smooth muscle cells (SMC), contribute to the regulation of the immune response by production of cytokines or expression of adhesion molecules. Little is known about the mechanism(s) involved in the stimulation of vascular cells by endotoxin (lipopolysaccharide [LPS]). As reported previously, LPS antagonists reduce LPS-induced cytokine production or adhesion in vitro specifically, suggesting a specific LPS recognition mechanism. We thus investigated the role of CD14 for stimulation of vascular SMC by LPS. Complement-fixing antibodies directed against CD14 (LeuM3, RoMo I, or Mo2) lysed monocytes but failed to mediate lysis of EC or SMC, indicating the lack of endogenous membrane CD14 in vascular cells. In addition, we did not detect expression of CD14 protein on EC and SMC in cell sorting analysis or cell immunoassay experiments. These observations are in line with our finding that a CD14 probe did not hybridize with mRNA or EC or SMC in Northern (RNA) blot experiments, although it hybridized well with monocyte-derived mRNA. We obtained the same results with the much more sensitive reverse transcription-PCR. Since the vascular SMC did not express endogenous CD14, we investigated the role of human serum-derived soluble CD14 (sCD14) for activation of SMC by LPS. In medium containing human serum, anti-CD14 antibodies inhibited activation of SMC by LPS. In contrast, the same antibodies did not inhibit activation of cells cultured in medium containing fetal calf serum. SMC cultured in sCD14-depleted medium responded 1,000-fold less to LPS than cells cultured in presence of sCD14. Reconstitution of sCD14-depleted serum or supplementation of serum-free medium with recombinant CD14 restored the capacity of the cells to respond to LPS. These results show that specific activation of vascular SMC by LPS does not involve binding to endogenous membrane CD14, but that the activation of vascular SMC by LPS is mediated to a great extent by serum-derived sCD14.

CD14 is not involved in Rhodobacter sphaeroides diphosphoryl lipid A inhibition of tumor necrosis factor alpha and nitric oxide induction by taxol in murine macrophages

Taxol, a microtubule stabilizer with anticancer activity, mimics the actions of lipopolysaccharide (LPS) on murine macrophages in vitro. Recently, it was shown that taxol-induced macrophage activation was inhibited by the LPS antagonist Rhodobacter sphaeroides diphosphoryl lipid A (RsDPLA). To investigate the mechanisms of taxol-induced macrophage activation, the present study focused on the interaction of LPS, RsDPLA, and taxol in the activation of and binding to macrophages. Taxol alone induced murine C3H/He macrophages to secrete tumor necrosis factor alpha (TNF) and to produce nitric oxide (NO) with kinetics similar to that of LPS. Macrophages from LPS-hyporesponsive C3H/HeJ mice, in contrast, did not yield any detectable TNF and NO production in response to LPS or taxol. RsDPLA inhibited taxol-induced TNF and NO production from C3H/He macrophages in a dose-dependent manner. The inhibition by RsDPLA was specific for LPS and taxol in that RsDPLA did not inhibit heat-killed Listeria monocytogenes- or zymosan-induced TNF production. Polymyxin B blocked the inhibitory effect of RsDPLA on taxol-induced TNF production. The inhibitory activity of RsDPLA appeared to be reversible since macrophages still responded to taxol in inducing TNF production after the RsDPLA was washed out with phosphate-buffered saline prior to the addition of taxol. Taxol-induced TNF production was not inhibited by colchicine, vinblastine, or 10-deacetylbaccatine III. A mutant cell line, J7.DEF3, defective in expression of a CD14 antigen, responded equally well to taxol by producing TNF as did the parent J774.1 cells. This suggested that the activation of macrophages by taxol does not require CD14. Taxol-induced TNF production by the mutant cells was also inhibited by RsDPLA. 125I-labeled LPS and 3H-labeled taxol was reported to bind to J774.1 cells predominantly via CD14 and microtubules, respectively. The binding of 125I-labeled LPS to J7.DEF3 cells was about 30 to 40% of that to J774.1 cells. The binding of 125I-LPS to J774.1 cells was inhibited by unlabeled LPS and RsDPLA but not by taxol. On the other hand, 3H-labeled taxol bound to both J774.1 cells and J7.DEF3 cells in similar time- and dose-dependent manners. The binding of [3H]taxol to these cells was inhibited by taxol but not by LPS or RsDPLA. Although the binding studies failed to examine cross competition for binding to macrophages, a possible explanation of these results is that LPS, RsDPLA, and taxol share the same molecule(s) on murine macrophages for their functional receptor(s), which is neither CD14 nor tubulin.

The significance of the hydrophilic backbone and the hydrophobic fatty acid regions of lipid A for macrophage binding and cytokine induction

Natural partial structures of lipopolysaccharide (LPS) as well as synthetic analogues and derivatives of lipid A were compared with respect to inhibit the binding of 125I-labelled Re-chemotype LPS to mouse macrophage-like J774.1 cells and to induce cytokine-release in J774.1 cells. LPS, synthetic Escherichia coli-type lipid A (compound 506) and tetraacyl precursor Ia (compound 406) inhibited the binding of 125I-LPS to macrophage-like J774.1 cells and induced the release of tumor necrosis factor alpha (TNF alpha) and interleukin 6 (IL-6). Deacylated R-chemotype LPS preparations were completely inactive in inhibiting binding and in inducing cytokine-release. Among tetraacyl compounds, the inhibition-capacity of LPS-binding was in decreasing order: PE-4 (alpha-phosphonooxyethyl analogue of 406) > 406 >> 404 (4'-monophosphoryl partial structure of 406) > 405 (1-monophosphoryl partial structure of 406). In the case of hexaacyl preparations, compounds 506, PE-1 (alpha-phosphonooxyethyl analogue of 506) and PE-2 (differing from PE-1 in having 14:0 at positions 2 and 3 of the reducing GlcN) inhibited LPS-binding and induced cytokine release equally well, whereas preparation PE-3 (differing from PE-2 in containing a beta-phosphonooxyethyl group) showed a substantially lower capacity in binding-inhibition and cytokine-induction. The conclusion is that chemical changes in the hydrophilic lipid A backbone reduce the capacity of lipid A to bind to cells, whereas the number of fatty acids determines the capacity of lipid A to activate cells. These results indicate that the bisphosphorylated hexosamine backbone of lipid A is essential for specific binding of LPS to macrophages and that the acylation pattern plays a critical role for LPS-promoted cell activation, i.e. cytokine induction.

Biochemical and electron microscopy analysis of the endotoxin binding to microtubules in vitro

The mechanisms involved in cellular activation and damage by bacterial endotoxins are not completely defined. In particular, there is little information about possible intracellular targets of endotoxins. Recently, the participation of a microtubule associated protein in endotoxin actions on macrophages has been suggested. In the present work, we have studied the effect of E. coli lipopolysaccharide on the polymerization of microtubular protein in vitro. Electrophoretic analysis of the polymerization mixtures showed that the endotoxin inhibited the polymerization when present at high concentrations. At lower concentrations, LPS selectively displaced the microtubule associated protein MAP-2 from the polymerized microtubules. Electron microscopy showed that LPS binds to microtubules of tubulin + MAPs and to microtubules of purified tubulin (without MAPs) polymerized with taxol. Gel filtration experiments confirmed the binding of LPS to tubulin, and by ligand blot assays an interaction LPS-MAP-2 was detected. The ability of LPS to interact with microtubular proteins suggests a possible participation of microtubules on the cellular effects of endotoxins.

Endotoxin receptors on mammalian cells

Recent experiments from our laboratory, as well as those of several other investigators, have focused upon the identification and characterization of specific membrane-localized receptors for bacterial endotoxic lipopolysaccharides on mammalian cells. In this article, we have summarize the results of these studies with a primary emphasis upon experiments from our own laboratory. It would appear that some differences between laboratories in the identification of LPS receptor may be the result of different experimental techniques employed. In spite of these differences, however, there is an emerging picture which defines several LPS binding proteins as potentially dominant candidates for the functional LPS receptor on the cell membrane, including a protein of 70-80 kDa and perhaps a second protein of 30-40 kDa as well. Other membrane proteins with important functions in the cellular response to LPS include CD14, the CD11/18 family of adhesins and the 95 kDa scavenger receptor. Identification of specific cell membrane targets for LPS has important implications for immunotherapy of septic shock and perhaps cancer as well.

Intracellular protein phosphorylation in murine peritoneal macrophages in response to bacterial lipopolysaccharide (LPS): effects of kinase-inhibitors and LPS-induced tolerance

Intracellular protein phosphorylation is thought to be the initial step in cell activation. Bacterial lipopolysaccharide (LPS) induces a special set of the protein phosphorylation in the murine peritoneal macrophages, including p65 (molecular mass of 65 kDa) which is a substrate of serine kinase and the most dominant phosphorylated cytosolic protein. This article deals with the relation between the LPS-induced protein phosphorylation in the murine peritoneal macrophages and their productions of IL-1 beta and TNF-alpha. LPS-induced p65 phosphorylation seems to be dependent on protein kinase C (PKC) and calmodulin (CaM), because it diminishes in the presence of inhibitors to PKC or CaM. Tyrosine kinase inhibitors do not affect the p65 phosphorylation. The PKC inhibitors also affect the mRNA expressions and the productions of active molecules of IL-1 beta and TNF-alpha. Though the CaM inhibitor inhibits the mRNA expression and the active molecule production of IL-1 beta, it does not affect those of TNF-alpha. These results suggest that LPS-induced p65 phosphorylation is closely related to PKC and CaM, and that IL-1 beta production depends on PKC and CaM, while the TNF-alpha production is not dependent on CaM. These findings indicate the existence of multiple pathways and different regulatory mechanisms for transduction of LPS signal in the macrophages. Furthermore, LPS-induced phosphorylation is not observed in endotoxin tolerant macrophages after re-stimulation with LPS, suggesting that the LPS-stimulus signal is blocked at a site in the signal transduction-pathway before the point of phosphorylation of proteins in the tolerant macrophages.

Agonists and antagonists for lipopolysaccharide-induced cytokines

Agonistic and antagonistic properties of LPS and partial structures in the induction of cytokines are reviewed. Studies on structure-activity relationships of LPS and lipid A with human mononuclear cells reveal that S- and notably R-form LPS are very potent cytokine inducers. Synthetic E. coli lipid A is somewhat less active, whereas synthetic S. minnesota-type lipid A is significantly less active. Pentaacylated forms of lipid A are less potent than hexaacylated forms, and tetraacylated synthetic precursor Ia and bisacylated disaccharides and monosaccharides are completely inactive, indicating that a structure-dependent hierarchy of LPS and lipid A partial structures determines the monokine-inducing capacity in human mononuclear cells. Precursor Ia is a potent LPS antagonist. The mechanism of its inhibitory activity is shown to be due to competitive binding to cellular binding sites (receptors). Proinflammatory and antiinflammatory cytokines, receptor antagonists, and soluble cytokine receptors influence the cytokine-inducing activity of LPS, suggesting a complex regulatory network.

Lipopolysaccharide nonresponder cells: the C3H/HeJ defect

Since its initial discovery as endotoxin resistant, the C3H/HeJ mouse has been extensively studied and used as a comparative model to help reveal the mechanism under genetic control which governs host responses to endotoxin. Most of the research has focused on the B lymphocyte and macrophage of this strain which fail to be activated by LPS. Recently, specific LPS binding proteins have been isolated on lymphocytes and other cells; however a receptor which transduces an activation signal has not been isolated as yet from responder cells which is missing or altered on C3H/HeJ nonresponder cells. Investigations into the signal transduction pathways used by C3H/HeJ B cells when they are activated by a protein mitogen have been found to be similar to those used by LPS responder cells when activated by LPS. Protein kinase C and tyrosine kinase, which phosphorylate signal proteins in cells have been found to be operative in C3H/HeJ and C3H/OuJ B cells. In both cases, DNA synthesis is shut off by either PKC or PTK blockade; however, PTK inhibition will also block activation of PKC stimulated DNA synthesis, indicating tyrosine kinase initiated phosphorylation may regulate the PKC signal pathway. Further analysis of the proteins that are phosphorylated in LPS responder and LPS nonresponder B cells is needed before conclusions can be drawn as to whether the defect in C3H/HeJ cells resides in the signal pathway leading to gene activation and proliferation. Nevertheless, the notion of a missing or defective signal receptor still remains as a working hypothesis to explain C3H/HeJ cell hyporesponsiveness to LPS. Isolation of the Lpsn gene and its product will provide the evidence needed for a clearer understanding of how LPS reacts with cells.

Modulation of endotoxin-induced monokine release in human monocytes by lipid A partial structures that inhibit binding of 125I-lipopolysaccharide

We have previously shown that the synthetic tetraacyl precursor Ia (compound 406, LA-14-PP, or lipid IVa) was not able to induce the production of tumor necrosis factor, interleukin-1, and interleukin-6 in human monocytes but strongly antagonized lipopolysaccharide (LPS)-induced formation of these monokines. This inhibition was detectable at the level of mRNA production. To achieve a better understanding of molecular basis of this inhibition, we investigated whether lipid A precursor Ia (LA-14-PP), Escherichia coli-type lipid A (LA-15-PP), Chromobacterium violaceum-type lipid A (LA-22-PP), and synthetic lipid A partial structures and analogs (LA-23-PP, LA-24-PP, and PE-4) were able to influence the binding of 125I-LPS to human monocytes and compared this inhibitory activity with the agonistic and antagonistic action in the induction of monokines in human monocytes. 125I-LPS (20 ng per well) was added to human monocytes in the presence or absence of unlabeled rough Re mutant-derived LPS (Re-LPS) or lipid A compounds, and specific LPS binding was determined after 7 h. This binding was found to be dependent on CD14 as shown by the use of an anti-CD14 monoclonal antibody. Compound LA-14-PP was found to inhibit the binding of 125I-LPS to the cells in a similar dose-response to that of unlabeled LPS. This shows that the inhibitory capacity on LPS binding does not correlate with the monokine-inducing capacity because Re-LPS is active in inducing tumor necrosis factor, interleukin-1, and interleukin-6, while LA-14-PP is not. The strong capacity of LA-14-PP to inhibit 125I-LPS binding, however, correlates with the strong inhibitory capacity of this compound on LPS-induced monokine production. Compounds LA-15-PP, LA-23-PP, and LA-24-PP were active in the inhibition of 125I-LPS binding but were 5- to 10-fold weaker than Re-LPS and LA-14-PP. Of all lipid A structures tested, compound LA-22-PP expressed the weakest inhibitory capacity on LPS binding. These compounds showed again that the activity of binding inhibition does not correlate with the monokine-inducing capacity. We assume that the inhibitory effects of lipid A partial structures on LPS-induced monokine production have their origin in a competitive inhibition between LPS and the lipid A partial structures for the same binding site on the cell membrane.