Dependence of Escherichia coli hyperbaric oxygen toxicity on the lipid acyl chain composition

Abstract

This study examines certain membrane-related aspects of oxygen poisoning in E. coli K1060 (fabB fadE lacI) and its parent strain, K-12 Ymel. Cells were grown to exponential or stationary phase in a minimal medium and exposed to air plus 300 lb/in2 of O2 as a suspension in minimal salts. After an initial lag, both strains lost viability with apparent first-order kinetics. Hyperbaric oxygen was more toxic to cells harvested during the exponential phase of growth than to cells harvested from the stationary phase of growth for both strains K-12 Ymel and K1060. Control suspensions exposed to air plus 300 lb/in2 did not lose viability during a 96-h exposure. The sensitivity of the unsaturated fatty acid auxotroph, strain K1060, to hyperbaric oxygen increased as the degree of unsaturation of the fatty acid supplement increased. Cells grown with a cyclopropane fatty acid (9,10-methyleneoctadecanoate) were the most resistant; cells grown with a monounsaturated fatty acid (oleate) were intermediate; and those grown with polyunsaturated fatty acids (linoleate and linolenate) were most sensitive to hyperbaric oxygen. The parent strain, K-12 Ymel, lost viability in hyperbaric oxygen most similarly to strain K1060 supplemented with oleate. To determine the relative effect of hyperbaric oxygen on the survival of E. coli with saturated membranes, substrains of K1060 were selected for growth on 12-methyltetradecanoate or on 9 or 10-monobromostearate. Substrains grown with a saturated fatty acid supplement were equally or more sensitive to hyperbaric oxygen than when the same substrains were grown with a cyclopropane fatty acid supplement. The lipid acyl chain composition was determined in E. coli K1060 before and after exposure to hyperbaric oxygen or hyperbaric nitrogen. The proportion of nonsaturated acyl chain lipid of either the oleate- or the 9,10-methyleneoctadecanoate-supplemented K1060 remained unchanged after hyperbaric gas exposure. In strain K1060 supplemented with linoleate and grown to stationary phase, however, the relative unsaturated acyl chain content after hyperbaric exposure decreased in both gases. This finding prompted an investigation of the role of lipid oxidation in hyperbaric oxygen toxicity. Assays of potential lipid oxidation products were performed with linoleate-grown cells. The lipid hydroperoxide and peroxide content of the lipid extract increased by 6.9 times after 48 h of air plus 300 lb/in2 of O2; malondialdehyde and fluorescent complex lipid oxidation products showed much smaller or no changes. Lipid extracts from hyperbaric oxygen-exposed cells were not toxic to viable E. coli K1060, nor did they increase the rate of loss of viability in cells simultaneously exposed to hyperbaric oxygen. Linoleic acid hydroperoxide at 1.0 mM had no effect on the viability of E. coli K-12 Ymel and only marginally decreased the viability of E. coli K1060 supplemented with linoleate. We conclude that the kinetics of oxygen toxicity in E. coli are dependent upon the particular unsaturated fatty acyl chain composition of the membrane, that the membrane composition influences the rate-limiting step in hyperbaric oxygen toxicity, but that a direct structural effect of hyperbaric oxygen on the membrane lipid, such as oxidation of the unsaturated acyl chain moiety, is unlikely in these cells.