Biosynthesis of bacterial lipopolysaccharide. V. Lipid-linked intermediates in the biosynthesis of the O-antigen groups of Salmonella typhimurium.

Biosynthesis of the core proceeds by successive transfer of monosaccharide residues from nucleotide sugars to the incomplete lipopolysaccharide.1 The origin of the repeating oligosaccharide units of the O-antigenic side chains is less well understood. We previously described the incorporation of galactose, mannose, and rhamnose into 0-antigen repeating units in a mutant of S. typhimurium deficient in GDP-mannose.2 Similar results have also been described by Nikaido and Nikaido3 with a mutant of S. typhimurium lacking TDP-rhamnose and by Robbins4 with wild-type S. anatum. However, in these experiments the nature of the intermediate reactions was not established. Anderson. et al.5 have presented evidence for a lipid-linked disaccharide intermediate in the biosynthesis of cell-wall glycopeptide in Staphylococcus aureus and Micrococcus lysodeikticus, and our data suggest that analogous intermediates are involved in synthesis and polymerization of the 0-antigen of S. typhimurium. We have isolated lipid-linked derivatives of two oligosaccharides, rhamnosyl-galactosyl1-phosphate and mannosyl-rhamnosyl-galactosyl-l-phosphate, related to the 0antigen repeating units. Our evidence suggests that the trisaccharide-lipid is an intermediate in formation of polysaccharide chains containing the trisaccharide repeating units. These experiments have been carried out with a galactose-negative strain which contains an incomplete core polysaccharide. Under the conditions used, the enzymically synthesized polysaccharide is not transferred to lipopolysaccharide but apparently remains attached through a galactose-l-phosphate reducing terminus to the lipid acceptor. It is postulated that this lipid-linked polysaccharide is a precursor of the O-antigenic chains. Materials and Mlethods.-These were as previously described,2 6 including the galactose-negative strain of Salmonella tryphimurium deficient in UDP-galactose-4epimerase, the conditions of growth, and the preparation of the cell-envelope enzyme fraction. This fraction was routinely sedimented by centrifugation at 40,000