Monocyte cytokine secretion induced by chemically-defined derivatives of Klebsiella pneumoniae.

The capacity of a K. pneumoniae membrane proteoglycan (Kp-MPG) and four of its chemically defined derivatives to activate human monocytes was studied by measuring immunoreactive IL-1 beta, IL-6 and tumour necrosis factor-alpha (TNF-alpha) in culture supernatants. Monocyte culture supernatants were also tested for their comitogenic activity on concanavalin A-stimulated thymocytes and for their cytotoxic activity on the mouse fibroblastic L929 cell line. The four Kp-MPG derivatives were: (i) an acylpoly(1-3)galactoside (APG); (ii) an APG preparation submitted to acid hydrolysis which removed all fatty acids but left intact the galactose chain of APG (GC-APG); (iii) a preparation obtained by mild alkaline hydrolysis, containing additional ester-linked C14 and C16 fatty acids bound to the APG molecule (EFA-APG); and (iv) a polymer of the latter compound (APG pol). Kp-MPG induced the synthesis of IL-1 beta, IL-6 and TNF-alpha with dose-responses and kinetics similar to those of Salmonella minnesota lipopolysaccharide (Sm-Re-LPS). APG pol and EFA-APG induced the secretion of the three cytokines with lower potency than Kp-MPG or Sm-Re-LPS. APG did not trigger any detectable cytokine production and GC-APG induced only borderline and inconsistent responses. Our data demonstrate the critical role of ester-linked C14 and C16 fatty acids in the triggering of monocyte response to Kp-MPG derivatives.     

Activation of human monocyte chemiluminescence response by acylpoly(1,3)galactosides derived from Klebsiella pneumoniae.

The stimulating activity of several preparations isolated from a membrane proteoglycan of a nonencapsulated smooth strain of Klebsiella pneumoniae (Kp-MPG) on the oxidative burst of human blood monocytes was assessed by luminol-enhanced chemiluminescence (CL). Five Kp derivatives were studied: a 34-kd acylpoly(1,3)galactoside (APG), obtained by drastic alkaline hydrolysis and purified by chromatography; an APG preparation subjected to acid hydrolysis that removed the core part and all fatty acids, leaving intact the galactose chain of APG (GC-APG); an APG preparation subjected to mild oxidation (ox APG); a preparation obtained by mild alkaline hydrolysis of Kp-MPG, containing additional ester-linked C14 and C16 fatty acids bound to the APG molecule (EFA-APG); and a polymer of the latter compound, APG pol. EFA-APG directly stimulated monocyte CL, whereas Kp-MPG, APG pol, and the whole bacterial cells had little or no activity. APG itself and ox APG induced a weaker response than EFA-APG. Polymyxin B sulfate completely inhibited the CL response to bacterial lipopolysaccharide (LPS) but not to EFA-APG. The stimulating action of EFA-APG on blood monocytes was dependent on the extracellular levels of both calcium and magnesium. Preincubation of monocytes with monoclonal antibody anti-Mac-1 directed against CD11b, the alpha chain of complement receptor type 3 (CR3; CD11b/CD18), strongly inhibited CL activation by EFA-APG and to a lesser extent CL activation by unopsonized zymosan and rough LPS. Altogether, these findings provide indirect evidence for the contribution of the CD11b/CD18 integrin in the functional interaction of EFA-APG with monocyte membranes. They demonstrate the role of fatty acids in the triggering of monocyte oxidative burst, while the polysaccharide chain itself does not contribute to induction of the CL response in this model. In keeping with the effects of EFA-APG and APG, we show that the monocyte CL response was triggered by bacterial LPS from the rough strain of Salmonella minnesota Re 595 and its lipid A, but not by LPS from smooth strains, again suggesting a critical role for the lipid moiety.     

Binding to leukocytes and induction of TNFα production involve different sites within the lipid-A region of LPS-like molecules

The induction of TNFα synthesis in whole blood culture assay and isolated peripheral blood mononuclear cells was investigated, using LPS from Klebsiella pneumoniae and two water-soluble 34 kDa derivatives designed as acylpolygalactoside (APG) and EFA-APG, an APG molecule bearing two additional ester-linked fatty acids. Both APG and EFA-APG bind to monocytes by specific ligand receptor interaction but only EFA-APG could induce TNFα synthesis. It is concluded that ester-linked fatty acids are not involved in LPS binding to the cell surface, but play a critical role in the triggering of cellular responses.     

CD14 and CD11b mediate serum-independent binding to human monocytes of an acylpolygalactoside isolated from Klebsiella pneumoniae.

A water-soluble acylpolygalactosyl (APG) of 34 kDa was obtained from the Klebsiella pneumoniae membrane by alkaline hydrolysis and delipidation. APG comprises a poly(1,3)galactose chain, a core, and a lipid moiety made of a glucosamine disaccharide with two N-linked beta OH-myristates. The monocyte binding sites for APG were investigated by flow cytometry. Biotin-labelled APG (Biot-APG) bound to monocytes at 4 degrees C in the absence of serum, calcium, and magnesium. The binding was dose dependent, saturable, and displaced by unlabelled APG. Neither the polysaccharide chain present in APG-related molecules nor the PPi group or additional ester-linked myristates and palmitates were required for APG binding. The role of CD11b and CD14 was demonstrated by competitive inhibition with monoclonal antibodies and by the uptake of APG by these solubilized proteins. APG was rapidly internalized into monocytes at 37 degrees C while CD14 and CD11b/CD18 molecules were partially down-modulated. Lipopolysaccharides (LPS) from the same K. pneumoniae strain and from Escherichia coli and Salmonella minnesota partially competed for Biot-APG binding in the absence but not in the presence of serum. When altered by alkaline hydrolysis, those LPS became strong competitors for APG binding. It was concluded that alkaline hydrolysis of the K. pneumoniae membrane yielded molecules structurally related to LPS which bind to LPS membrane receptors in the absence of serum.     

Binding of a Bacterial Acylpoly(1,3)Galactoside to Human Blood Leucocytes

The binding of a 34-kDa (mol. wt.) acylpoly(1,3)galactoside (APG) extracted from a membrane proteoglycan of Klebsiella pneumoniae to human blood leucocytes was investigated. APG is made of a long poly(1,3)galactose chain, a core-like region and a lipid moiety which comprises two glucosamine residues bound to a phosphate group and two beta OH myristic acids. Fluoresceinated APG was shown to bind preferentially to monocytes and to a lesser extent to polymorphonuclear neutrophils, as determined by flow cytometry. Binding of fluoresceinated APG was inhibited by unlabelled APG; it was concentration dependent, but not saturable, with rapid kinetics. It occurred at +4 degrees C but was markedly increased at 37 degrees C. It involved trypsin-sensitive molecules on the membrane of monocytes. Neither the parent proteoglycan nor lipopolysaccharide from K. pneumoniae or Salmonella minnesota competed for APG binding. A minor non-specific binding to lymphocytes, occurring predominantly on B cells, was observed. Unlike that of lipopolysaccharide, the APG binding was not blocked by polymyxin B sulphate. Interaction between the galactose chain of APG and the galactose receptor does not account for the binding of APG to monocytes because the galactose receptor (Mac-2) is expressed at high density on activated macrophages but not on monocytes. Despite its strong binding to human blood monocytes, APG displayed a much weaker activity than K. pneumoniae membrane proteoglycan with respect to induction of monocyte cytokine synthesis. When administered as a Technetium 99 conjugate, APG was shown to label inflammatory foci in experimental animals, and its property as a marker of macrophages is currently being evaluated in clinical trials.     

Correlation Between Modification of Membrane Phospholipids and Some Biological Activity of Lymphocytes, Neutrophils and Macrophages

Our study considered the possibility of modifying the functional response of human neutrophils, of mouse lymphocytes and macrophages treated with phospholipids having different polar groups, different isomerisms with saturated and unsaturated fatty acids from C12 to C20 carbon atoms. the results are as follows.

a) Most of the phospholipids containing fatty acids from C12 to C20 cause inhibition of the blastogenic capacity of the polyclonal activators tested.

b) the phospholipids tested cause a decrease in adherence of polymorphonuclear leukocytes with the exception of the phosphatidyl-choline containing saturated and unsaturated fatty acids.

c) A decrease in polymorphonuclear leukocytes migrational capacity almost always occurs.

d) the cells treated with L-phosphatidyl-ethanolamine having fatty acids from C14 to C17 show an increase in chemiluminescence; those treated with phosphatidyl-choline and L-phosphatidyl-glycerol show a decrease of the chemiluminescence; L-phosphatidic acid and L-phosphatidyl-ethanolamine having Microbial fatty acids (FAs) at c16 cause a decrease in the formation of phagolisosomes in the macrophages tested.     

Comparison of the effects of acid and base hydrolyses on hydroxy and cyclopropane fatty acids in bacteria

The cellular fatty acid compositions of Legionella oakridgensis, Brucella suis, Pseudomonas aeruginosa, and Francisella tularensis were compared after base hydrolysis (saponification), acid hydrolysis, and acid methanolysis procedures were used to release the fatty acids. The branched-chain, unsaturated, saturated, and ester-linked hydroxy acids were released as effectively with saponification at 100 degrees C for 30 min as with acid hydrolysis or acid methanolysis at 85 degrees C for 16 h. Although the amide-linked hydroxy acids were released more effectively by acid hydrolysis or acid methanolysis, these methods degraded the cyclopropane fatty acids, producing a number of new peaks or artifacts in the chromatograms. Cyclopropane fatty acids were not degraded by saponification, and at least 50% of the hydroxy acids were released when the cells were saponified with 15% NaOH in 50% aqueous methanol. Thus, the results show that saponification for 30 min at 100 degrees C with 15% NaOH, followed by methylation is an excellent method for routine fatty acid analysis of bacteria and for screening cultures whose identity and fatty acid composition are unknown.     

Lipopolysaccharide complexed with soluble CD14 binds to normal human monocytes

Using flow cytometry we have compared the binding of Neisseria meningitidis lipopolysaccharide labeled with fluorescein isothiocyanate (FITC-LPS) to normal human monocytes in whole blood with the binding to Chinese hamster ovary (CHO) cells transfected with human CD14 gene (hCD14-CHO cells). Binding of FITC-LPS to cells was dose dependent, saturable and enhanced in the presence of increasing concentrations of serum. Blockade of membrane CD14 with saturating concentrations of anti-CD14 monoclonal antibody (mAb) My4 inhibited 50% of the binding of FITC-LPS to monocytes and 100% to hCD14-CHO cells. Similarly, removal of membrane CD14 by phosphatidylinositol phospholipase C (Pl-PLC) treatment of the cells partially decreased the binding of FITC-LPS to monocytes but totally inhibited the binding to hCD14-CHO-transfected cells. These results suggest that binding of FITC-LPS to monocytes is not only mediated by membrane CD14. Using two-color flow cytometry, we observed that FITC-LPS binds to My4-saturated monocytes in association with soluble (s)CD14 present in serum as revealed by staining with rhodaminelabeled My4 mAb. The binding of FITC-LPS/sCD14 complexes to monocytes treated with saturating amounts of unlabeled My4 prior to addition of the complexes was completely inhibited by anti-CD14 mAb 10G33. When cells were first saturated with a mixture of My4 and 10G33 mAb, washed and further incubated with FITC-LPS/sCD14, inhibition of the binding of LPS was similar to that observed with cells saturated with My4 alone, showing that the binding of FITC-LPS is not mediated by the 10G33 epitope present on mCD14. These results suggest that either the 10G33 epitope on sCD14 is involved in the binding of LPS/sCD14 complexes to the cells, or that 10G33 mAb inhibits the binding of FITC-LPS to sCD14. Taken together, these data indicate that sCD14 which is present in normal serum, in addition to membrane CD14, enables LPS to bind monocytes through an as yet unidentified molecule and that sCD14 does not simply serve as a shuttle for transfer of LPS to membrane CD14.     

Evidence that the Receptor for Soluble CD14:LPS Complexes may not be the Putative Signal-Transducing Molecule Associated with Membrane-Bound CD14

Membrane-bound CD14 acts as a receptor for lipopolysaccharide (LPS) on monocytes/macrophages and neutrophils. Studies have suggested that the activation of monocytes/macrophages by the binding of LPS to membrane-bound CD14 may require the association of a signal-transducing molecule with membrane-bound CD14. The observation that non-CD14 expressing cells, such as endothelial cells, can nevertheless be activated by a complex of LPS and a soluble form of CD14 (sCD14) suggests that the receptor for this complex may be identical to the signal transducing molecule associated with membrane-bound CD14. The studies described show that two CD14-specific MoAb are able to block the LPS-induced activation of endothelial cells but do not affect the response of monocytes to LPS. This suggests that the interaction of the sCD14:LPS complex with endothelial cells is distinct from the interaction of membrane-bound CD14 with its putative signal-transducing molecule.     

The involvement of CD14 in stimulation of cytokine production by uronic acid polymers

In this study the molecular mechanisms behind the stimulatory activities of the uronic acid polymers poly mannuronic acid (poly M), high M alginate and oxidized cellulose (C60XY) were investigated and compared with lipopolysaccharide (LPS). The cytokine-inducing abilities of the uronic acid polymers and LPS were examined on CD14-positive human monocytes and CD14-negative U373 astrocytoma cells. It was found that LPS induced monocytes and U373 cells to produce tumor necrosis factor (TNF) and interleukin(IL)-6, respectively, by different mechanisms. The poly uronic acids induced monocytes to produce TNF, but with 100-1000 times less potency compared to LPS. On U373 cells, LPS at concentrations ? 32 ng/ml resulted in a dose-related IL-6 production, whereas the poly uronic acids had negligible effects even at 1 mg/ml. The binding data demonstrate that only the CD14-positive monocytes in the peripheral blood mononuclear cells population bound poly M. Furthermore, poly M was found to bind to CD14 in the presence of serum. Antibodies against CD14 also inhibited the TNF-inducing activity of the three uronic acid polymers tested. In conclusion, these results demonstrate that uronic acid polymers induce TNF production through mechanisms which involve CD 14.     

Production of anti-CD14 monoclonal antibodies using CD14 expressing COS cells as immunizing antigen

CD14 is a leukocyte surface molecule expressed on monocytes but not on lymphocytes. Recently, CD14 molecule was demonstrated to function as a receptor for endotoxin. CD14 specific monoclonal antibody (MAb), therefore, can be used to identify monocytes and study the host defense mechanism to bacterial endotoxin. To produce MAb against CD14 protein, in this study cDNA encoding CD14 protein and COS cell expression systems were used to prepare CD14 expressing COS cells. The CD14 transfectants were then used as antigen for mouse immunization. The spleen cells of the immunized mouse were then fused with myeloma cells by conventional hybridoma technique. By using this strategy, 5 hybridroma clones secreting antibody specific for CD14 molecule were generated within one fusion. The generated CD14 MAbs were strongly positive with monocytes, weakly positive with neutrophils but negative with lymphocytes. In addition, the generated CD14 MAb blocked the binding of lipopolysaccharide (LPS) to the CD14 molecules. These CD14 MAbs could be used to enumerate peripheral blood monocytes as well as using referent CD14 MAb. We, therefore, introduce an alternative method for preparation of antigen for production of monoclonal antibody. This type of antigen is a very effective antigen for the production of monoclonal antibodies against cell surface molecules.     

Heparin inhibits lipopolysaccharide (LPS) binding to leukocytes and LPS-induced cytokine production

The glycosaminoglycan heparin is known to exhibit anti-inflammatory properties unrelated to its anticoagulant activity. However, in a generalized inflammatory response with implanted or extracorporeal devices, the beneficial effect of heparin coating and/or systemic administration is still unclear as well as the precise mechanisms of action. In the present study, we have first studied the effect of heparin on lipopolysaccharide (LPS)-induced cytokine production by human blood monocytes. Our results indicated that the production of interleukin-1alpha, tumor necrosis factor-alpha, and interleukin-8 was significantly decreased when heparin was simultaneously incubated with Escherichia coli LPS. Because the modulation of heparin on monocyte activation could be mediated by its binding via CD14, the main LPS receptor on monocytes, we then studied the binding of LPS and heparin to leukocytes from human blood and to Chinese hamster ovary cells transfected with the human CD14 gene. The data by flow cytometry showed the binding of biotinylated heparin to leukocytes. Moreover, the experiments performed on leukocytes and on CD14-positive Chinese hamster ovary cells indicated that heparin inhibited LPS binding. From our results, we conclude that: 1. heparin is an effective inhibitor of LPS-induced monocyte activation, and 2. heparin inhibits the binding of LPS to cells via a CD14-independent pathway. This study suggests a potentially important therapeutic application for heparin or heparin analogs to prevent inflammation with biomaterials.     

The M-DC8-positive leukocytes are a subpopulation of the CD14+ CD16+ monocytes

Recently, a population of M-DC8-positive leukocytes has been described as a new subpopulation of human dendritic cells (DC). In view of the expression of the CD16 antigen on these cells, as well as the finding that DC can arise from blood monocytes, we hypothesized that the expression of M-DC8 is mainly associated with the CD14+ CD16+ phenotype of blood monocytes. Immunofluorescence analysis of whole blood showed that the percentage of M-DC8+ cells is about three times lower than the percentage of CD14+ CD16+ monocytes among all leukocytes (0.32% versus 1.10%). Further, in addition to the expression of CD16, these M-DC8+ cells were also positive for CD14 at low density. Multicolor flow cytometric analysis of whole blood revealed that more than one third of the CD14+ CD16+ monocyte population expressed the M-DC8 antigen (42.3%), and almost all M-DC8+ cells were CD14-CD16-double-positive (87.5%). Finally, the M-DC8 antigen was also expressed on alveolar macrophages from healthy individuals, i.e., cells that are phenotypically and functionally related to the CD14+ CD16+ monocytes. Taken together, the data presented here clearly demonstrate that the M-DC8+ leukocytes are a subpopulation of the CD14+ CD16+ monocytes and may represent DC     

The Fc receptor, FcRI, and other activation molecules on human mononuclear phagocytes after treatment with interferon-gamma.

Human monocytes and the myeloid cell lines U937 and HL60 have been tested with monoclonal antibodies (MoAbs) reactive with 22 different cell surface molecules before and after treatment with interferon-gamma (IFN-gamma). An increase in the expression of the high affinity Fc receptor, FcRI, and the receptor for interleukin 2, IL-2R, were the most consistent alterations which were observed. In addition, expression of the gp55 molecule recognized by CD14 MoAbs was decreased on monocytes. Of the MHC Class II molecules, there was little expression by the myeloid cell lines and no enhancement after IFN-gamma treatment. In contrast monocytes expressed all three MHC Class II subloci with DR much greater than DQ and DP. However there was much variation in IFN-gamma-mediated increase in expression of the individual subregions. In monocytes, the alteration in expression of FcRI, IL-2R, gp55 and MHC Class II molecules took place in a co-ordinate fashion and reached a plateau only after 48 h. In U937 cells, activation proceeded more rapidly and was at maximum levels between 12-16 h. This increase in FcRI appears to be a hallmark of IFN-gamma activation for mononuclear phagocytes (Mph) as the other alterations are either not found on all types of Mph (gp55, MHC Class II) or are induced by other cytokines on Mph and on other cells (IL-2R, MHC Class II). Conversely, other cytokines do not induce FcRI on Mph. These results also suggest that the cell membrane phenotypic changes induced in Mph by IFN-gamma may not be extensive and that FcRI must play a specific role in the IFN-gamma-activated Mph.     

Flow cytometric analysis of different adhesion molecules expression on circulating CD14- and CD64- human dendritic cell precursors

Blood dendritic cell precursors (DCps) are identified as mononuclear leukocytes expressing HLA-DR but lacking the characteristic antigens associated with T cells (CD3), NK cells (CD16 and CD56) and B cells (CD 19). Dendritic cell precursors are distinguished from monocytes by their lack of expression of CD64 rather than of CD14. This study investigated whether CD14- DCps differed from CD64-DCps, which were predominantly CD14+, in their expression of five well-characterised adhesion molecules. There were significantly fewer cells expressing CD11b, CD18 and CD29 in the CD64-DCp population compared with CD14- DCps, and this CD64- DCp subpopulation also had a lower expression of CD11b and CD18. Our results suggest that the two DC precursor subpopulations may differ from one another in their binding characteristics to blood vessel walls and to other leukocytes.     

The Staphyloccous aureus Eap protein activates expression of proinflammatory cytokines

The extracellular adhesion protein (Eap) secreted by the major human pathogen Staphylococcus aureus is known to have several effects on human immunity. We have recently added to knowledge of these roles by demonstrating that Eap enhances interactions between major histocompatibility complex molecules and human leukocytes. Several studies have indicated that Eap can induce cytokine production by human peripheral blood mononuclear cells (PBMCs). To date, there has been no rigorous attempt to identify the breadth of cytokines produced by Eap stimulation or to identify the cell subsets that respond. Here, we demonstrate that Eap induces the secretion of the proinflammatory cytokines interleukin 6 (IL-6) and tumor necrosis factor alpha (TNF-alpha) by CD14(+) leukocytes (monocytes and macrophages) within direct ex vivo PBMC populations (note that granulocytes are also CD14(+) but are largely depleted from PBMC preparations). Anti-intercellular adhesion molecule 1 (CD54) antibodies inhibited this induction and implicated a role for this known Eap binding protein in cellular activation. IL-6 and TNF-alpha secretion by murine cells exposed to Eap was also observed. The activation of CD14(+) cells by Eap suggests that it could play a significant role in both septic shock and fever, two of the major pathological features of S. aureus infections.     

Serum-independent binding of lipopolysaccharide to human monocytes is trypsin sensitive and does not involve CD14.

The nature of the binding sites for lipopolysaccharide (LPS) on human monocytes was investigated using fluorescein isothiocyanate (FITC)-labelled LPS from Salmonella minnesota R595 (ReLPS). In the absence of serum, ReLPS bound to monocytes and this interaction was trypsin sensitive. A concentration of 0.1 mg/ml resulted in a 90% loss of LPS binding, while low concentrations increased this binding. Trypsin-treated monocytes recovered FITC-ReLPS binding after 20 hr culture, which was abrogated in the presence of cycloheximide and actinomycin D. This showed that de novo protein and mRNA synthesis were essential. A number of different proteins have been implicated in cellular binding of LPS to monocytes. In this paper we show that CD14 is not involved in direct binding of FITC-ReLPS to monocytes, since anti-CD14 monoclonal antibody (mAb) (3C10) and removal of most of cell-surface CD14 by phosphatidylinositol-specific phospholipase C did not prevent FITC-ReLPS binding. Furthermore, LPS also bound to CD14-deficient cells from a patient with paroxysmal nocturnal haemoglobinuria (PNH). FITC-ReLPS binding was not mediated by the CD11/CD18 complex since mAb to the alpha and beta chains of the CD11/CD18 complex did not alter the binding of FITC-ReLPS to cells. These observations indicate that ReLPS may interact with monocyte membrane protein(s) in the absence of serum. This binding site(s) for LPS might be different from those previously described by others.     

Membrane molecules which trigger the production of interleukin-1 and tumor necrosis factor-alpha by lipopolysaccharide-stimulated human monocytes.

We have investigated the role of the membrane molecules CD11/CD18 and CD14 which may mediate the binding of lipopolysaccharide (LPS) to human monocytes, in the induction of the production and release of interleukin (IL)-1 and tumor necrosis factor-alpha (TNF-alpha) by LPS-stimulated cells. Blockade of CD11a, CD11b and CD18 with saturating concentrations of specific mAb did not inhibit the release of cytokines from LPS-stimulated monocytes. In contrast, inhibition of the release of IL-1 beta and TNF-alpha occurred in monocytes cultures that had been pretreated with either of two monoclonal antibodies (mAb) recognizing different epitopes on the CD14 molecule. The binding of LPS to CD14 has been previously shown to require serum factors. In the present study, we found that serum had an enhancing effect on the release of IL-1 and TNF-alpha from LPS-stimulated cultures of normal human monocytes. The inhibitory effect of anti-CD14 mAb was, however, observed in cultures performed in the presence or in the absence of serum, suggesting that triggering of IL-1/TNF-alpha release by CD14 is independent of LPS-binding proteins or other serum proteins. IL-1 beta and TNF-alpha were also released from LPS-stimulated cultures of monocytes from patients with paroxysmal nocturnal hemoglobinuria lacking expression of CD14. Thus, CD14 but not CD11/CD18 can trigger serum-dependent and independent cytokine release from endotoxin-stimulated normal human monocytes; CD14 is not, however, the only LPS receptor that is involved in the secretory response of endotoxin-stimulated cells.