The lipoprotein of the outer membrane of Escherichia coli: a B-lymphocyte mitogen.

The lipoprotein of the outer membrane of Escherichia coli is a B-cell mitogen in mice. Polyclonal activation of B lymphocytes was measured by an increase in thymidine uptake, by the development of plaque-forming cells against densely coupled trinitrophenylated sheep red cells, and by selectively increased rates of synthesis and secretion of leucine-labeled IgM. Murein-free and muropeptides-containing lipoprotein are effective in B-cell activation, while free murein is inactive. Removal of ester-linked fatty acids from the amino-terminal end of the lipoprotein by alkaline hydrolysis abolishes the mitogenicity of the lipoprotein. B lymphocytes from high responder (C3H/Tif and BALB/c nu/nu) or from low responder (C3H/HeJ) mice to the mitogen lipopolysaccharide (LPS) both respond well to the lipoprotein. Anti-immunoglobulin antibodies inhibit the mitogenic stimulation of B cells by lipoprotein. A complex of structures including the Ig-receptor molecules, the LPS receptor, and the lipoprotein receptor appear involved in the regulation of mitogenic stimulation of B cells to proliferation and differentiation to IgM-secreting cells.

The mitogenic effect of lipopolysaccharide on bone marrow-derived mouse lymphocytes. Lipid A as the mitogenic part of the molecule.

Lipopolysaccharides with different structure, isolated from different mutant strains of Escherichia coli and Salmonella bacteria, and chemical degradation products of these lipopolysaccharides have been employed to investigate which part of the lipopolysaccharide molecule exerts mitogenic effects on bone marrow-derived mouse lymphocytes. Within the structure of lipopolysaccharide consisting of lipid A, a core polysaccharide, and the O-polysaccharide antigen, lipid A was found to be the mitogenic part. The mitogenic effect of lipid A, consisting of phosphorylated glucosamine disaccharide units with ester- and amide-linked fatty acids, was lost after alkali treatment, which removes ester-linked fatty acids. Insertion of the lipid A portion of lipopolysaccharides into the lipid bilayer of the plasma membranes of bone marrow-derived lymphocytes is discussed as the initial mitogenic action.

Lethal toxicity of lipopolysaccharide and tumor necrosis factor in normal and D-galactosamine-treated mice.

The toxic properties of human recombinant tumor necrosis factor (TNF) were investigated in mice made hypersensitive to endotoxin by treatment with D-galactosamine. C3H/TifF mice treated with D-galactosamine were rendered sensitive to the lethal effects of submicrogram amounts of TNF. In the absence of D-galactosamine, TNF caused approximately 80% lethality with 500 micrograms. The duration of sensitization to TNF lasted up to 8 h after D-galactosamine administration, that towards LPS, up to 4 h. As with LPS, with TNF sensitization could be inhibited by uridine administered up to 2 h after D-galactosamine/TNF, showing that the early biochemical alterations in the liver known to be necessary for sensitization to LPS are also necessary for sensitization to TNF. In contrast to LPS, the toxicity of TNF was expressed also in D-galactosamine-treated endotoxin-resistant C3H/HeJ mice. The susceptibility of these mice to TNF was identical to that of endotoxin sensitive mice. In the absence of D-galactosamine the toxicity of TNF in C3H/HeJ mice was comparable to that obtained in C3H/TifF mice, being lethal with amounts of the order of 500 micrograms. The present results support the hypothesis that TNF is a mediator of lethal toxicity of endotoxin.

Interferon beta, a cofactor in the interferon gamma production induced by gram-negative bacteria in mice.

The interferon (IFN) gamma production of splenocytes from closely related C57BL/10ScSn (Sn) and C57BL/10ScCr (Cr) mice was compared. Concanavalin A and CD3 monoclonal antibodies induced high levels of IFN-gamma in both Sn and Cr splenocytes. By contrast, treatment with gram-negative bacteria induced IFN-gamma only in Sn splenocytes; in Cr splenocytes, the IFN-gamma response was heavily impaired. The IFN-gamma induction by bacteria requires the cooperation of IFN-gamma-producing cells with macrophages. Depletion of macrophages from Sn splenocytes resulted in the loss of ability to produce IFN-gamma after bacterial stimulation. Reconstitution with new Sn macrophages restored the IFN-gamma responsiveness, whereas reconstitution with Cr macrophages failed to do so. Normal function of IFN-gamma-producing cells and a defective function of macrophages of Cr mice was demonstrated by evidence showing that whole or macrophage-depleted Cr splenocytes, when supplemented with Sn macrophages, acquire the ability to produce IFN-gamma in response to bacteria. A similar effect was achieved by supplementing Cr splenocytes with supernatants of bacteria-stimulated Sn macrophages or with recombinant murine IFN-beta or IFN-alpha. Preincubation of active macrophage supernatants with antibodies to IFN-beta suppressed the helper activity for Cr splenocytes. Moreover, the bacteria-induced production of IFN-gamma by Sn splenocytes could be inhibited by antibodies to murine IFN-beta. The results provide evidence that IFN-beta is an important cofactor of IFN-gamma induction, which is not induced in Cr mice by gram-negative bacteria.

B-cell activation by lipopolysaccharide. Distinct pathways for induction of mitosis and antibody production.

The role played by macrophages in two effects of lipopolysaccharide (LPS) on the immune system of the mouse-substitution for helper T cells and induction of B-cell mitosis-has been investigated. C3H/HeJ mice are unresponsive and do not produce (as other strains do) antibody to 2,4,6-trinitrophenol (TNP) conjugated with autologous mouse erythrocytes (MRBC-TNP) in the presence of LPS. We found that C3H/HeJ spleen cells produce antibody to MRBC-TNP when (a) LPS and macrophages from LPS-responsive C3HeB/FeJ mice or (b) tumor necrosis serum ([TNS] induced by LPS in responsive mice) are added. The mitotic response was not restored. The findings suggest that adjuvanticity and mitogenicity represent distinct pathways of B-cell activation by LPS, subject to different regulatory mechanisms.

Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene.

Mutations of the gene Lps selectively impede lipopolysaccharide (LPS) signal transduction in C3H/HeJ and C57BL/10ScCr mice, rendering them resistant to endotoxin yet highly susceptible to Gram-negative infection. The codominant Lpsd allele of C3H/HeJ mice was shown to correspond to a missense mutation in the third exon of the Toll-like receptor-4 gene (Tlr4), predicted to replace proline with histidine at position 712 of the polypeptide chain. C57BL/10ScCr mice are homozygous for a null mutation of Tlr4. Thus, the mammalian Tlr4 protein has been adapted primarily to subserve the recognition of LPS and presumably transduces the LPS signal across the plasma membrane. Destructive mutations of Tlr4 predispose to the development of Gram-negative sepsis, leaving most aspects of immune function intact.

Innate and adaptive immune responses in contact dermatitis: analogy with infections.

Allergic contact dermatitis (ACD) is an inflammatory skin disease of great and steadily increasing importance as an occupational health problem. The disease is induced by chemicals and metal ions which penetrate the skin and form complexes with host proteins. This process is accompanied by a strong, allergen-induced inflammatory reaction and leads to the migration of allergen-carrying dendritic cells (DC) from the skin to regional lymph nodes, where they promote generation of allergen-specific T cells. The latter are the ultimate effector cells of the disease. Re-exposure to the causative agent leads to the recruitment of the T effector cells, which then elicit the typical skin inflammatory reaction at the site of contact. Although DC and effector T cells play a protagonistic role in the sensitization and elicitation phase of ACD, respectively, other cell types including keratinocytes, NK cells, mast cells and B cells contribute to the pathogenesis of the disease. In this review the authors summarize recent findings that identify stress responses and innate immune pathways triggered by contact allergens and review recent data regarding the adaptive T cell response. The new data were collected mainly from studies on contact hypersensitivity (CHS), the corresponding experimental mouse model of human ACD. The elucidation of the molecular events involved in contact allergen-induced innate responses will help to design new treatment strategies and may allow to develop predictive in vitro assays for the identification of contact allergens.

Phase I trial of intravenously administered endotoxin (Salmonella abortus equi) in cancer patients.

We report a phase I study in cancer patients being treated with i.v. bolus injections of highly purified lipopolysaccharide (LPS) Salmonella abortus equi. Twenty-four patients with disseminated cancer received escalating doses of LPS at 2-week intervals. Dose escalation was performed in six dose levels treating 3-6 patients at each level. Dose levels 1 and 2 consisted of 0.15 and 0.3 ng/kg, respectively. Further dose escalation up to 5.0 ng/kg was enabled by pretreatment with ibuprofen, which attenuated the constitutional side effects of LPS. The maximum tolerated dose was 4.0 ng/kg with dose-limiting toxicity being World Health Organization grade III hepatic toxicity. Hematological changes included transient decreases in WBCs affecting granulocytes, monocytes, and lymphocytes in a marked different pattern. Endogenous cytokine release occurred in an LPS dose-dependent manner as measured by tumor necrosis factor-alpha, interleukin-6, and macrophage colony-stimulating factor serum levels. Moderate antitumor activity in colorectal cancer was observed in the case of 2 patients. Phase II trials of LPS are currently in progress.

Relationship between chemical composition and biological function of Pseudomonas aeruginosa lipopolysaccharide: effect on human neutrophil chemotaxis and oxidative burst.

There are conflicting data on the effect of bacterial lipopolysaccharides (LPS) on the function of human neutrophils. The present study was designed to examine the relationship between chemical composition and the modulatory effect of LPS on human neutrophil function. LPS was extracted from five strains of Pseudomonas aeruginosa isolated from cystic fibrosis patients by the hot phenol-water method. Chemical characterization included neutral sugars, amino components, and fatty acids. Neutrophils isolated from peripheral blood of healthy individuals were preincubated with different concentrations of LPS. After preincubation, the chemotaxis and chemiluminescence of neutrophils to various stimuli were determined. It was shown that LPS from different strains did not exert the same degree of regulatory effect on neutrophil functions. LPS from strain 174-O:9 exerted the most pronounced effect on neutrophil function seen as inhibition of neutrophil chemotaxis toward the chemotactic peptide f-Met-Leu-Phe and zymosan-activated serum (ZAS) and priming of the cells for less than or equal to 8-fold enhancement of chemiluminescence response to f-Met-Leu-Phe. Conversely, LPS from strain 1118-O:3 had no effect on neutrophil chemotaxis and a slight effect on chemiluminescence. The major differences in chemical composition of the LPS from these two strains are in the rhamnose and heptose content of the O side chain and in the alanine content of the core region. These data indicate that chemical composition of the LPS molecule may play an important role in biological activity of LPS.

Infection of C57BL/10ScCr and C57BL/10ScNCr mice with Leishmania major reveals a role for Toll-like receptor 4 in the control of parasite replication.

The innate immune system is essential for host defense; it senses the presence of potentially pathogenic-invading microorganisms, and the contribution of Toll-like receptors (TLRs) to this response is increasingly recognized. In the present study, we investigated the contribution of TLR4 to the course of cutaneous leishmaniasis in vivo. We used C57BL/10ScNCr (TLR4(0/0)) and C57BL/10ScCr [TLR4/interleukin-12 (IL-12)Rbeta2(0/0)] mice and compared the course of Leishmania major infection, parasite load, cell recruitment, and cytokine profile with those of wild-type C57BL/10ScSn mice. Our results confirm the importance of IL-12 receptor-mediated signaling in resistance to L. major infections. Importantly, we show that the lack of TLR4 results in an increased permissiveness for parasite growth during the innate and adaptive phase of the immune response and in delayed healing of the cutaneous lesions. The use of the tlr4 transgenic mouse strain TCr5 demonstrated unequivocally that TLR4 contributes to the efficient control of Leishmania growth in vivo.

Changes in dehydroepiandrosterone (DHEA) and DHEA-sulfate plasma levels during experimental endotoxinemia in healthy volunteers.

Dehydroepiandrosterone (DHEA) and DHEA-sulfate (DHEA-S) have immunomodulatory effects in vitro and in vivo. Additionally, their plasma levels are altered during chronic infection and inflammation. However, it remains unknown whether these steroids are involved in early host responses to infection in humans. We examined DHEA and DHEA-S levels during experimental endotoxinemia, a well established pathophysiological model of bacterial infections in humans. Purified Salmonella abortus equi endotoxin (0.2, 0.4, or 0.8 ng/kg body weight) was injected in a single-blind, placebo-controlled experiment to 17 healthy male volunteers. During the following 12 h, rectal temperature and the plasma levels of ACTH, cortisol, DHEA, DHEA-S, interleukin 6, and tumor necrosis factor alpha were determined. Confirming earlier studies, temperature and cytokine levels showed monophasic, dose-dependent increases in response to endotoxin. In contrast, endocrinological effects of endotoxin showed a complex, biphasic pattern: cortisol levels were not affected by 0. 2 ng/kg but significantly increased during the first 6 h following 0. 4 and 0.8 ng/kg endotoxin, whereas ACTH and DHEA levels were significantly enhanced during the first 6 h following 0.8 ng/kg only. ACTH, DHEA, and cortisol secretion was blunted 6-12 h following 0.8 ng/kg. DHEA-S levels were unaffected during the first 6 h following all dosages, but between 6-12 h after injection they were significantly increased following 0.2 ng/kg, unaffected by 0.4 ng/kg, and significantly decreased following 0.8 ng/kg endotoxin. The present results suggest that similarly to glucocorticoids, the adrenal androgens DHEA and DHEA-S play an important role during early host responses to bacterial infections in humans.

TNF-alpha hyper-responses to Gram-negative and Gram-positive bacteria in Propionibacterium acnes primed or Salmonella typhimurium infected mice.

IFN-gamma-dependent hypersensitivity to LPS is inducible in mice by infection or pre-treatment with killed bacteria. Hypersensitive mice exhibit enhanced inflammatory responses to LPS, including the overproduction of TNF-alpha. Using Lps(n) BALB/c and Lps(d) BALB/c/l mice, primed with Propionibacterium acnes or infected with Salmonella typhimurium, we show that concurrently to hypersensitivity to LPS, a hypersensitivity to other constituents of killed Gram-negative or Gram-positive bacteria and to staphylococcal enterotoxin B (SEB) develops. The TNF-alpha hyper-responses in sensitized mice induced by different Gram-positive bacteria, are generally weaker than those by Gram-negative bacteria and vary significantly, due to the absence of a common, LPS-equivalent component. Using IFN-gamma R(-/-) and the respective wild-type mice, we demonstrate that although sensitization to LPS and killed Listeria monocytogenes is exclusively IFN-gamma-dependent, an IFN-gamma-independent, moderate sensitization to certain TNF-alpha-inducing constituents in bacteria may develop in parallel.

MALP-2, a Mycoplasma lipopeptide with classical endotoxic properties: end of an era of LPS monopoly?

Although some activities of LPS are shared by other bacterial components, for half a century LPS has been regarded as unique in displaying many pathophysiological activities. Here we report on a synthetic lipopeptide, MALP-2 from Mycoplasma fermentans, which expresses potent endotoxin-like activity and whose lethal toxicity is comparable to that of LPS. With the exception of the Limulus lysate gelation test, in which MALP-2 was approximately 1000-fold less active than LPS, the synthetic lipopeptide induced all activities tested for, and in most cases to an extent comparable to that of LPS. Unlike LPS, the biological activities of MALP-2 were expressed both in LPS-responder and in LPS-non-responder mice (BALB/c/l, C57BL10/ScCr), indicating that MALP-2 signaling, unlike that of LPS, is not transduced via the Toll-like receptor (Tlr) 4 protein.MALP-2 expressed no toxicity in normal or sensitized Tlr2 knockout (Tlr2(-/-)) mice indicating that its toxic activity is induced via Tlr2 signaling. The phenomenology of the lethal shock induced by MALP-2 in normal or sensitized mice, i.e. the kinetics of its development and symptoms of illness exhibited by the treated animals, was very reminiscent of the lethal shock induced by LPS.

Differential clearance and induction of host responses by various administered or released lipopolysaccharides.

The clearance and activity of different types of lipopolysaccharide (LPS) released during infection with Gram-negative bacteria were investigated. When highly purified preparations differing in their specific endotoxin activity were administered intravenously to mice, the clearance of rough (R)-form LPS preparations from Salmonella minnesota and Escherichia coli was much faster than that of a smooth (S)-form LPS preparation from Salmonella abortus equi, but slower than that of lipo-oligosaccharides (LOS) preparations from Bordetella pertussis and Helicobacter pylori. After intraperitoneal infection with 10(7) and 10(8) CFU E. coli O111:B4, relatively high levels of LPS were detected dose-dependently in the plasma of infected mice and persisted for a long time. In addition, plasma sCD14 levels in infected mice were higher than in LPS-administered mice. These results indicate that continuously higher levels of plasma LPS followed by stronger host responses occur during infection and suggest that these differences between LPS-administered and infected mice should be taken into consideration when analyzing host responses induced by LPS.

Role of lipopolysaccharide susceptibility in the innate immune response to Salmonella typhimurium infection: LPS, a primary target for recognition of Gram-negative bacteria.

Lipopolysaccharide is an important recognition marker by virtue of which the innate immune system senses and reacts against Gram-negative bacteria invading the LPS susceptible host. This review deals with the factors affecting LPS susceptibility and with the role of the latter in the course and outcome of Salmonella typhimurium infection.

Phase II trial of intravenous endotoxin in patients with colorectal and non-small cell lung cancer.

We report the immunological and clinical results of a phase II trial with intravenously administered highly purified endotoxin (Salmonella abortus equi) in patients with advanced cancer. 15 patients with non-small cell lung cancer and 27 with colorectal cancer were entered into the study. 37 evaluable patients received at least four injections of endotoxin (4 ng/kg body weight) and 1600 mg ibuprofen orally in 2-week intervals. Transient renal (WHO grade 0-1) and hepatic (WHO grade 0-4) toxicities occurred in several patients. Constitutional side-effects such as fever, chills and hypotension could not be prevented completely by pretreatment with ibuprofen. 3 patients in the colorectal cancer group demonstrated objective responses (1 complete remission (CR), 2 partial remission (PR)). The complete remission has been maintained for more than 3 years, while the partial remissions were stable for 7 and 8 months, respectively. Only marginal antitumour effects were seen in the lung cancer group. Tolerance of the macrophage system to the stimulatory effect of endotoxin, as measured by human necrosis factor alpha (TNF-alpha) release into serum, built up after the first administration and remained at a steady-state level after each subsequent injection. In constrast, rising CD4:CD8 ratio and release of tumour necrosis factor beta (TNF-beta) indicated the continuing activation of the lymphocyte system by repetitive injections of endotoxin.

Immune carrier properties of acid-treated Salmonella minnesota R595 bacteria. The immune response to TNP-bacterial conjugates in rabbits and mice.

Salmonella minnesota R595 bacteria from which the core region of the lipopolysaccharide on the cell wall had previously been removed by mild acid treatment were trinitrophenylated. Differing amounts of these trinitrophenyl naked bacterial conjugates (TNP-NB), covering a range of epitope densities, were used for immunising mice and rabbits via the intraperitoneal or intravenous routes without adjuvants. It was found that such acid-treated, naked bacteria were effective carriers for the covalently linked hapten, TNP, with an optimum epitope density of 15 micrograms TNP/mg NB. Significant immune responses were obtained with dose levels as low as 50 ng TNP. The possible applications of acid-treated, naked bacteria as universal carriers having inherent adjuvant activity are discussed.

Bacterial lipopolysaccharides: structure, metabolism and mechanisms of action.

Endotoxins (lipopolysaccharides, LPS) are biologically active substances present in the outer membrane of gram-negative bacteria. They induce a spectrum of biological effects which may be harmful or beneficiary for the host. Lipid A is the biologically active part of the LPS molecule. This was demonstrated using soluble forms of lipid A and more recently confirmed further by employing synthetic lipid A. LPS administered into experimental animals circulates as LPS/HDL complex and is cleared from the blood mainly into the liver and spleen. In the liver LPS undergoes partial deacylation however without a loss of toxic activity. Its excretion is effected mainly via the bile into the gut. The lethal toxicity and tolerance inducing properties of LPS are mediated by macrophages through tumor necrosis factor alpha (TNF alpha), which is probably the most important endogenous mediator of the lethal effects of LPS. The lethal toxicity of LPS may be completely inhibited by anti-TNF alpha antibodies.

Mechanisms of endotoxin shock and endotoxin hypersensitivity.

Endotoxins (lipopolysaccharide, LPS) are biologically active substances present in Gram-negative bacteria. Injection of purified LPS into experimental animals leads to the development of many biological activities that can lead to shock with lethal outcome. The biological activities of LPS are not direct effects of the LPS molecule since LPS usually expresses no direct cytotoxic activity. The toxic and other biological properties of LPS are caused indirectly through the action of endogenous mediators that are formed following interaction of LPS with cellular targets, macrophages occupying a key position in the development of endotoxin shock. The interaction of LPS with macrophages may proceed directly leading to activation of these cells, with subsequent synthesis and secretion of a number of endogenous mediators which initiate the different biological activities of LPS. Tumor necrosis factor alpha (TNF-alpha), a macrophage derived cytokine, is a primary mediator of the lethal action of endotoxin. Sensitivity to LPS is genetically determined, varying considerably among different species. The sensitivity of normal animals (mice) to endotoxin may be enhanced considerably under different experimental conditions that include treatment with live (infection) or killed Gram-negative and -positive bacteria. Sensitization to endotoxin proceeds in all LPS-responder strains investigated and in the LPS-resistant mice of the strain C3H/HeJ. It does not proceed in a second LPS-resistant strain, C57BL/10ScCr. The absence of sensitization in the latter mice was found to be due to an impaired IFN-gamma production. IFN-gamma could be identified as the mediator of endotoxin hypersensitivity induced by bacteria.

Serum anti-lipid A antibodies in multiple myeloma and Waldenström's macroglobulinaemia.

Anti-lipid A antibodies were determined in sera from 38 patients with IgG multiple myeloma, 33 patients with IgA multiple myeloma and 38 patients with Waldenström's macroglobulinaemia using an enzyme-linked immunosorbent assay (ELISA). Compared to 34 healthy adults, significantly (p less than 0.025, p less than 0.005, p less than 0.0025) lower serum anti-lipid A antibody levels were measured for the respective patient groups. Low anti-lipid A antibody levels correlated with a higher infection rate with gram-negative bacteria in patients with monoclonal B-cell malignancies. The highest infection rate was seen in patients with simultaneous low anti-lipid A levels and secondary antibody deficiency.