Abstract
We compared the abilities of free and liposome-incorporated Salmonella minnesota wild-type lipopolysaccharide (LPS) and lipid A to activate peritoneal macrophages and induce lethal toxicity in mice. Incorporation of lipid A into multilamellar vesicles resulted in a 100-fold-decreased potency to prime macrophages for phorbol myristate acetate-triggered release of H2O2. In addition, liposome incorporation reduced the lethality of LPS and lipid A at least 10-fold in dactinomycin-sensitized mice. Similar results were obtained with multilamellar liposomes delivered intravenously and when small unilamellar vesicles were employed. The observed difference in toxicity was not dependent on dactinomycin treatment, since a similar decrease was obtained with large doses of liposomal LPS in unsensitized mice. Control liposomes, prepared without LPS and lipid A, did not reduce the activities of the free compounds. The administration of a sublethal amount of liposomal LPS induced within 20 days, but not during the first week, tolerance to a subsequently injected lethal dose of free endotoxin. The latter observation suggests that early-phase tolerance is not the mechanism responsible for the reduced toxicity of liposomal LPS. These data show that liposomal LPS and lipid A have reduced endotoxic activity in vivo and are consistent with our hypothesis that a direct interaction of lipid A with appropriate plasma membrane components is necessary to efficiently trigger biologic responses. This interaction, however, is prevented by the stable insertion of LPS into the liposomal membrane.