Bacteremia versus endotoxemia in experimental mouse leukopenia--role of antibiotic chemotherapy.

Imipenem and ceftazidime have different specificities for penicillin-binding proteins and cause a differential release of lipopolysaccharide (LPS) from gram-negative bacteria in vitro. In studies in mice made leukopenic by the administration of cyclophosphamide, the innate relative resistance to the lethal effects of LPS was not significantly changed, but these animals became highly sensitized to bacterial infection. When leukopenic mice were challenged with graded doses of Escherichia coli O111:B4, an LD50 was achieved at a dose of approximately 10(6) cfu. Administration of either antibiotic resulted in a shift in the LD50 of approximately 500-fold, in contrast to D-galactosamine-treated LPS-sensitized mice, in which a < 10-fold increase in the LD50 was observed with antibiotic therapy. Further, if mice were made LPS-sensitive with D-galactosamine, no differences between leukopenic and normal mice were noted with antibiotic therapy.

Evidence for antibiotic-mediated endotoxin release as a contributing factor to lethality in experimental gram-negative sepsis.

Endotoxic lipopolysaccharide (LPS) is a major constituent of the outer membrane of the Gram-negative microbe. Following its release from the bacterium, LPS serves as a potent proinflammatory stimulus by interacting with humoral and cellular mediator systems to stimulate production of an array of inflammatory molecules. Cell-wall active antibiotics are known to promote endotoxin release. To assess the contribution of antibiotic-induced endotoxin release in the pathogenesis of Gram-negative sepsis, we have developed several experimental models in which mice have been pretreated with various agents to make them sensitive to Gram-negative (E. coli, pseudomonas) infection and/or the lethal effects of endotoxin. For the former, both cyclophosphamide (which renders mice neutropenic) and the reversible hepatotoxin D-galactosamine (D-gal) have been used. D-gal also sensitized mice to the lethal effects of LPS. Infected mice treated with cell-wall active antibiotics are protected approximately five- to 10-fold (as assessed by increases in LD50) if they are sensitive to LPS lethality (D-gal treatment) but 500-fold if they are resistant to LPS lethality. Importantly, different antibiotics that have been documented to cause different amounts of endotoxin release in vitro also differ in their protective efficacy in vivo. Thus, imipenem, which causes relatively low endotoxin release, is significantly more protective (8-fold) than ceftazidime or meropenem (3-fold, P < 0.005) under conditions of equivalent MICs. Lethality data correlate well with circulating levels of interleukin-6 (Il-6) in vivo and with induction of Il-6 in ex vivo studies in which anticoagulated mouse blood is incubated with bacteria and antibiotics. Finally, antiendotoxin agents manifest additional levels of protection in vivo under conditions in which antibiotics alone are not protective. Collectively, these results strongly implicate antibiotic-induced endotoxin release as a significant contributing factor in experimental Gram-negative sepsis.

Differences in therapeutic efficacy among cell wall-active antibiotics in a mouse model of gram-negative sepsis.

The in vivo efficacy of three cell wall-active antibiotics, imipenem, meropenem, and ceftazidime, was compared in mice rendered hypersusceptible to the pathophysiologic effects of lipopolysaccharide by treatment with D-galactosamine. When CF-1 mice were administered Escherichia coli, D-galactosamine, and saline intraperitoneally, an LD50 was achieved at an inoculum of approximately 2 x 10(4) cfu. Administration of antibiotic at 20 mg/kg resulted in significant but widely variable protective efficacy from E. coli lethality among the three antibiotics. At this dose, an approximately 3-fold increase in LD50 was observed with either meropenem or ceftazidime, whereas administration of imipenem resulted in an approximately 8-fold increase in LD50 (P = .0053). When the dose of antibiotic was decreased to 2 mg/kg, neither meropenem nor ceftazidime could provide measurable protection, whereas imipenem was almost fully protective (P < .002). These differences in protective efficacy were also noted with experimental Pseudomonas aeruginosa but not Staphylococcus aureus infection.

Therapeutic efficacy of a polymyxin B-dextran 70 conjugate in experimental model of endotoxemia.

Numerous studies have suggested that lipopolysaccharide (LPS), a major component of the cell wall of gram-negative bacteria, is responsible for the initiation of gram-negative septic shock. Previously, others have designed therapeutic regimens to target the biologically active lipid A region of LPS by either neutralization of the biological properties of LPS or enhancement of clearance of this molecule. One such compound capable of neutralizing lipid A is the antibiotic polymyxin B. However, the clinical utility of polymyxin B is limited by its toxicity. We therefore covalently conjugated this antibiotic to the high-molecular-weight polysaccharide dextran 70, resulting in reduced toxicity of polymyxin B but retention of its endotoxin-neutralizing ability. The studies described in this report were designed to test the in vivo efficacy of this compound in an experimental animal model of gram-negative septic shock. Mice were administered graded doses of Escherichia coli or Pseudomonas aeruginosa along with D-galactosamine and the antibiotic imipenem. We had previously determined that antibiotic chemotherapy provides significant protection against E. coli-mediated lethality with smaller doses of bacteria; however, the antibiotic does not provide protection against larger doses of bacteria, but it is effective at killing the bacterial inoculum in vivo. Administration of the polymyxin B-dextran 70 conjugate provided significant protection when given with an antibiotic but was not effective by itself. A requirement for a pretreatment period prior to E. coli challenge was shown to depend upon the bacterial challenge dose. In other studies using this D-galactosamine sensitization model, we demonstrated that the lipid A-specific conjugate had no effect on lethality caused by staphylococcus aureus or tumor necrosis factor alpha. The results of these studies indicate that this compound is effective in preventing lethal gram-negative septic shock in mice and may be useful as a potential therapeutic agent in humans as well.

Differential antibiotic-induced release of endotoxin from gram-negative bacteria.

Treatment of log phase cultures of Escherichia coli with cell wall active antibiotics results in increased exposure of immunologically reactive lipid A epitopes of lipopolysaccharide (LPS) and release of soluble LPS into culture supernatants. Comparison of the efficacy of two cell wall active antibiotics, ceftazidime, a penicillin-binding protein 3 selective antibiotic, and imipenem, a penicillin-binding protein 2 selective antibiotic, for their relative efficacy in mediating LPS release indicated quantitative but not qualitative differences, with the former antibiotic manifesting a significantly broader range of concentrations at which LPS release could be demonstrated. Comparison of the relative efficacy of these two antibiotics in a mouse bacteraemia model in which animals were made hypersensitive to the lethal effects of endotoxin by treatment with D-galactosamine indicated that the latter antibiotic may provide a greater level of protection. These studies suggest that the release of endotoxin mediated by antibiotic treatment may contribute to the pathogenesis of disease in infectious due to gram-negative organisms.

An interleukin-6-induced acute-phase response does not confer protection against lipopolysaccharide lethality.

Lipopolysaccharide (LPS), a component of gram-negative bacterial outer cell walls, can stimulate lymphoreticular cells to produce cytokines such as tumor necrosis factor alpha (TNF-alpha), interleukin-1 (IL-1), and IL-6. One of these proinflammatory cytokines, IL-6, induces hepatic synthesis of a class of proteins termed acute-phase proteins. D-Galactosamine inhibits acute-phase protein synthesis and concurrently sensitizes mice to a lethal dose of LPS approximately 10,000-fold. From these observations, we hypothesized that the acute-phase response may serve as a defense mechanism for protection of the host against the deleterious effects of LPS. To test this hypothesis, murine recombinant IL-6 (mrIL-6) was used to induce an acute-phase response prior to a lethal LPS challenge in both D-galactosamine-treated and normal mice. Induction of the acute-phase response by mrIL-6 was quantitated by measuring the concentrations of fibrinogen and complement component C3, two well-characterized acute-phase proteins, in the circulation. The effect of acute-phase and normal serum on TNF-alpha release by peritoneal macrophages stimulated with LPS in vitro was also examined. The results of these studies confirmed the induction of the acute-phase response by mrIL-6, as reflected in an approximate doubling in circulating levels of fibrinogen and C3. However, when either D-galactosamine-sensitized or normal mice were challenged with a lethal dose of LPS at various times after mrIL-6 administration, the acute-phase response induced by mrIL-6 did not alter either cumulative lethality or the kinetics of lethality. Additionally, compared with normal serum, acute-phase serum did not affect TNF-alpha release by peritoneal macrophages following LPS-mediated stimulation in vitro. Collectively, these studies would not support a dominant role for an IL-6-mediated acute-phase response as contributing to the resistance of normal mice compared with D-galactosamine-sensitized mice in LPS-induced lethal toxicity.