American Journal of Clinical Nutrition, Vol 32, 267-267, Copyright © 1979 by The American Society for Nutrition

Sensitivity to and site of oxygen poisoning in Escherichia coli

¹ From the University of Missouri, Columbia, Missouri 65201

Oxygen at concentrations greater than atmospheric is toxic for most life forms from bacteria to man. Although a variety of cellular processes in Escherichia coli is reversibly inhibited by hyperbaric oxygen, inhibition of growth is among the first to occur. There is evidence for inhibition in the following biosynthetic pathways: 1) branched-chain amino acids, 2) aromatic amino acids, 3) NAD-niacin, 4) thiamine, 5) RNA, and 6) reverse glycolysis. The inhibition of amino acid biosynthesis would theoretically bring on the "stringent response" with production of pp-guanine-pp which probably accounts for much of the reported nonspecific shut-down of metabolism. Two enzymes were possible sites of inhibition in NAD-niacin biosynthesis: phosphoribosylpyrophosphate (PRPP) synthetase and quinolinate transferase. Quinolinate transferase was suspect, because niacin and intermediates beyond quinolinate, but not quinolinate, protected cells from hyperoxia. PRPP is also required for this reaction. Therefore, assays of the activities of both enzymes were done in control cultures grown in air and in cells following exposure to growth-inhibiting concentrations of oxygen (4.2 atm). Basal salts plus glucose medium and this medium supplemented with the amino acids that give protection from oxygen toxicity were used. The effect of addition of the amino acids was studied since significantly more growth and oxygen uptake occur in this medium upon exposure to hyperoxia. Under these conditions, more toxic oxygen radicals would be metabolically generated and the potential for damage should be greater. Quinolinate synthetase activity was reduced by 25% in cells exposed to 4.2 atm of oxygen for 1 hr without amino acids (41.9 ± versus 31.5 ± 1.54 unit/mg of protein). Greater inactivation (92%) occurred when the cells were incubated in medium containing amino acids (41.9 ± 2.11 versus 3.3 ± 0.31). PRPP synthetase was more oxygen-sensitive and was more than 90% inactivated by exposure of cells to only 10 min of oxygen at 4.2 atm with or without amino acids present (27.5 ± 2.99 versus 0.95 ± 0.17, and 22.8 ± 1.03 versus 1.5 ± 0.19 nmoles PRPP per milligram of protein per hour, respectively). When nonmetabolizing cells (in the absence of carbon, energy, and nitrogen sources) were exposed to hyperoxia, PRPP synthetase was not inactivated. This further substantiates low or nonproduction of toxic oxygen radicals in the absence of significant metabolism. Upon reincubation of cells in air for only 10 min PRPP synthetase activity was restored to 82% of normal. However, with chloramphenicol present to prevent protein synthesis during recovery with air as the gas phase, the enzyme activity was restored to only 15% of normal. Dithiothreitol (a reducing agent) will not reactivate the oxygen-poisoned enzyme. The inhibitions of these enzymes are significant causes of oxygen toxicity in E. coil and perhaps other bacteria, and have intriguing possibilities with respect to the mechanisms of oxygen toxicity in higher organisms, including man.