The interactions of polycationic antibiotics with lipopolysaccharide (LPS) isolated from parental and polymyxin-resistant strains of Salmonella typhimurium and Escherichia coli were measured by using a cationic spin probe. Electron spin resonance spectra indicated that increasing concentrations of cations competitively displaced probe from LPS aggregates. Polymyxin B and other cations displaced less probe from LPS of porymyxin-resistant strains than from LPS of the parental strains, whereas the same amount or more probe was displaced from isolates of the mutants by the structurally similar antibiotic, EM 49 (octapeptin). In general, the differential affinities of these antibiotics for LPS correlated with their antibiotic activity in vivo, suggesting that resistance results from a decrease in antibiotic permeability across the outer membrane due to alterations in the LPS which affect antibiotic binding. The alterations in the structure of LPS from the polymyxin-resistant mutants of E. coli were characterized using 31P nuclear magnetic resonance spectroscopy. The results suggested that esterification of the core-lipid A phosphates is responsible for increased resistance to polymyxin B and that this alteration is different from that previously proposed for the S. typhimurium strains.In both cases, however, resistance was the result of modifications that result in a less acidic lipid A.

Lipopolysaccharide from smooth strains of Salmonella typhimurium, Salmonela minnesota, and Escherichia coli 0111:B4, 055:B5, and 0127:B8 was fractionated by gel filtration chromatography. All lipopolysaccharide samples separated into three major populations. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the fractions from S. typhimurium and S. minnesota indicated that the three peaks were made up of molecules with average 0-antigen lengths of (i) 70 or more repeat units, (ii) 30 and 20 repeats units in the samples from S. typhimurium and S. minnesota, respectively, and (iii) 1 repeat unit. In contrast to the Salmonella samples, peak 1 from the E. coli samples was not detected on polyacrylamide gels and lacked detectable phosphate. This high-molecular-weight material had a sugar composition similar to that of 0-antigen and was tentatively identified as capsular polysaccharide. Peaks 2 and 3 of the E. coli samples were analogous to those of the Salmonella isolates, containing lipopolysaccharide molecules with averages of 18 and 1 0-antigen repeat units, respectively. These lipopolysaccharide molecules did not completely dissociate during electrophoresis, and multimers were detected as distinct, anomalous, slow-migrating bands. Increasing the concentration of sodium dodecyl sulfate in the gels resulted in the dissociation of these multimers.

Polycations, such as aminoglycoside and peptide antibiotics, and naturally occurring polyamines were found to bind to the lipopolysaccharide of Pseudomonas aeruginosa and alter its packing arrangement. Binding of cations was measured by the displacement of a cationic spin probe from lipopolysaccharide into the aqueous environment upon addition of competitive cations. The level of probe displacement was dependent on the concentration and charge of the competing cation, with the more highly charged cations being more effective at displacing probe. The relative affinity of several antibiotics for lipopolysaccharide correlated with their ability to increase outer membrane permeability, while the relative affinity of several polyamines correlated with their ability to stabilize the outer membrane. Probe mobility within the lipopolysaccharide,head group was shown to be decreased by cationic antibiotics and unaltered or increased by polyamines. We propose that antibiotic permeability and disruption of outer membrane integrity by polycationic antibiotics results from binding of the antibiotic to anionic groups on lipopolysaccharide with a consequent change in the conformation of lipopolysaccharide aggregate structure.

Aggregates of short-and long-chain 0-antigen-containing fractions of lipopolysaccharide were analyzed by electron spin resonance probing to reveal differences in their physical properties. The fluidities of the lipid regions of the two fractions were quite similar, although the long-chain lipopolysaccharide aggregates appeared to be more hydrated as reflected by the polarity determined with a lipid probe. In contrast, the head-group region of the long-chain fraction was dramatically more mobile than that of the short-chain sample. The binding of polycations (e.g., polymyxin B, spermine) to lipopolysaccharide aggregates was measured by the partitioning of a cationic spin probe. Less probe was displaced from the long-chain fraction and unseparated lipopolysaccharide than from the short-chain fraction by the addition of cations, suggesting that the long 0-antigen masks anionic sites on lipopolysaccharide. These results indicate that the aggregate shape and reactivity of lipopolysaccharide are affected by 0-antigen length. Thus, the biological activity of lipopolysaccharide may be modulated directly by the presence of 0-antigen and indirectly by the effects of 0-antigen on the lipopolysaccharide aggregate structure.Gram-negative bacteria possess an outer membrane which serves as a permeability barrier to toxic compounds. A key component in this barrier function is the lipopolysaccharide (LPS) which is a major part of the outer monolayer of the membrane. LPS forms a rigid, highly charged surface that prevents the diffusion of hydrophobic compounds across the bilayer (24). In addition, LPS can form blebs off the bacterium and is found in sera of patients suffering from septicemia. This LPS can cause adverse host responses, including fever, shock, and even death (22).In its interactions within a host or on the surface of the bacteria, LPS does not act in monomeric form but as large aggregates. The size and shape of these aggregates depend on temperature (3), ion content (4, 11, 17), and pH (R. T. Coughlin, A. A. Peterson, A. Haug, H. T. Pownall, and E. J. McGroarty, Biochim. Biophys. Acta, in press), as well as the composition of the LPS (17). Variations in biological activities between preparations may then be caused by variations in the physical properties of different aggregate structures. The LPS isolated from Rhodospirillum tenue is moderately toxic, whereas the lipid A prepared from it is 70 to 140 times more toxic than the intact LPS (21). Differences in toxicity may thus result from covering the lipid A by the core 0-antigen of R. tenue LPS or from differences in the physical state of the aggregates modulating whether the inner regions of LPS are exposed and able to bind to or react with target structures.LPS, as isolated from bacteria, is a heterogeneous collection of molecules which can vary as to substitution of the core and lipid A (20) and in the length of the 0-antigen polysaccharide chain (12,16,23,25,27). The presence of covalently bound phosphate and acidic sugars in the core and lipid A make LPS a hig...

The dependence of colicin channel activity on membrane potential and peptide concentration was studied in large unilamellar vesicles using colicin E1, its COOH-terminal thermolytic peptide and other channel-forming colicins. Channel activity was assayed by release of vesicle-entrapped chloride, and could be detected at a peptide: lipid molar ratio as low as 10(-7). The channel activity was dependent on the magnitude of a transnegative potassium diffusion potential, with larger potentials yielding faster rates of solute efflux. For membrane potentials greater than -60 mV (K+in/K+out greater than or equal to 10), addition of valinomycin resulted in a 10-fold increase in the rate of Cl- efflux. A delay in Cl-efflux observed when the peptide was added to vesicles in the presence of a membrane potential implied a potential-independent binding-insertion mechanism. The initial rate of Cl- efflux was about 1% of the single-channel conductance, implying that only a small fraction of channels were initially open, due to the delay or latency of channel formation known to occur in planar bilayers. The amount of Cl- released as a function of added peptide increased monotonically to a concentration of 0.7 ng peptide/ml, corresponding to release of 75% of the entrapped chloride. It was estimated from this high activity and consideration of vesicle number that 50-100% of the peptide molecules were active. The dependence of the initial rate of Cl- efflux on peptide concentration was linear to approximately the same concentration, implying that the active channel consists of a monomeric unit.