Complement activation by polysaccharide of lipopolysaccharide: an important virulence determinant of salmonellae

Salmonellae with differences only in the O-antigenic polysaccharide of their lipopolysaccharide were previously shown to differentially activate complement via the alternative pathway, causing them to be ingested at different rates by the mouse macrophage-like cell line J774. We now show that this mechanism could explain the different virulence of these strains in vivo. Mouse peritoneal macrophages (thioglycolate induced) ingest these salmonellae at rates that are inversely proportional to the known virulence of the organisms and virtually identical to the rates observed with J774. As with J774, complement is required for this differential uptake, since serum was required and heating (56 degrees C for 30 min) or zymosan treatment of the serum destroyed activity. The known receptor for nonreducing terminal mannose-, fucose-, N-acetylglucosamine, and glucose-containing glyco-proteins did not participate, since uptake was not inhibited by high concentrations of mannan. When clearance of bacteria from the bloodstream of mice was measured, the least virulent organism was cleared very much faster than the most virulent organism, in confirmation of earlier data. When complement in the mice was destroyed by pretreatment with cobra venom factor, the clearance of the least virulent strain was greatly reduced, whereas the very slow clearance of the most virulent strain was unaffected. These data strongly support the hypothesis that when bacteria have polysaccharide in lipopolysaccharide that activates complement efficiently, the bacteria will be phagocytosed, whereas if the polysaccharide activates complement poorly, the bacteria escape ingestion and may cause disease.