SUMMARYThe extracellular production of ' free endotoxin ' from Escherichia coli serotype 078K80 is discussed. Isolation of purified toxic and non-toxic fractions from crude extracellular material and the physico-chemical properties of these fractions are described. Purified free endotoxin possessed similar properties to endotoxins extracted from the bacteria by conventional procedures ; a purified non-toxic extracellular fraction corresponded to the so-called ' native hapten '. When dissolved in buffers containing sodium dodecylsulphate, the non-toxic fraction was not disaggregated, but the endotoxin produced sub-units of a smaller particle size than the non-toxic material. From these experiments it was concluded that the non-toxic material did not constitute simple sub-units of endotoxin.
I N T R O D U C T I O NThe term ' endotoxin ' describes certain lipopolysaccharides of high molecular weight found in Gram-negative bacterial cell walls. Although Shear and his associates (e.g. Shear & Turner, 1943) have isolated tumour-necrotizing substances (endotoxins) from supernatant fluids of cultures of Gram-negative bacteria, most other workers have preferred to extract endotoxins from the bacteria themselves. An exception has been Work and her collaborators (Bishop & Work, 1965, Taylor, Knox & Work, 1966' who found endotoxic lipopolysaccharides in culture supernatant fluids of lysinerequiring mutants of Escherichia coli grown under lysine-limited conditions.The release of large quantities of free endotoxin into the culture supernatant fluids of Gram-negative bacteria, under favourable growth conditions, has recently been described (Crutchley, Marsh & Cameron, 1967a). The present paper will deal with the fractionation of supernatant fluids of cultures of Escherichia coli serotype 078K80 and the subsequent physical and chemical characterization of the fractions ; the accompanying paper (Crutchley et al. 1967 b) will discuss the biological properties of these fractions.Origin of toxicity. charide, was essential for toxicity; but a more recent investigation of endotoxins from rough Salmonella strains led them to suggest more strongly that the lipid moiety ('lipid A') was 'the factor decisive for at least some endotoxic effects' (Liideritz & Westphal, 1966). They also believe that, with respect to endotoxic activity, combined polysaccharide may function only as a solubilizing agent. However, the American school of Landy, Ribi and their associates (Ribi, Haskins, Landy & Milner, 1961a;Fukushi et al. 1964) have generally found very little lipid in their highly toxic fractions from several of the Enterobacteriaceae ; all their lipid extracts, including lipid A, possessed less than 1 % of the biological activity of the parent endotoxins. The balance of their published evidence suggests that a carbohydrate moiety is implicated in the toxicity, although certain other factors may be involved (Ribi, Haskins, Landy & Milner, 1961b).There have also been conflicting reports about the minimal molecular size of the complex required fo...
