The action of Trypanosoma cruzi trans-sialidase on glycolipids and glycoproteins.

Addition of sialic acid residues in the human pathogen Trypanosoma cruzi glycoconjugates is mediated by a trans-sialidase and not by a CMP-sialic acid:glycoconjugate sialyltransferase. Incubation of trans-sialidase with N-[galactose-14C]acetyllactosamine and O-linked oligosaccharides, N-linked glycopeptides (both obtained from fetuin) or sialyllactose showed that the last three compounds were donors of sialic acid residues to the first one. Moreover, N- and O-linked oligosaccharides in asialofetuin and asialomucin, respectively, served as acceptors of sialic acid units. Gangliosides GM3, GD1a and GT1b but not GM2, GM1a nor GD1b donated sialic acid units to N-acetyllactos amine when incubated with trans-sialidase. This showed that only sialic acid units bound to terminal galactosyl residues were transferred. GM1a was converted to GD1a, and GD1b to GT1b when incubated with the appropriate donor. The fact that asialo-GM1a was converted to a ganglioside migrating as GD1a on thin-layer chromatography suggested that sialic acid units may be transferred to internal galactosyl residues, although once linked to those residues they can not be further transferred to other glycoconjugates. Sialic acid residues linked alpha 2,3- but not alpha 2,6- or alpha 2,8- were transferred by the trans-sialidase. Methyl beta-galactoside but not methyl alpha-galactoside served as acceptor of sialic acid units, thus suggesting that terminal alpha-linked galactosyl units in T. cruzi and mammalian glycoproteins are not sialylated by the enzyme. As the trans-sialidase employed in these experiments has been shown to be located on the external surface of the parasite and to be shed to the medium, the relatively broad specificity shown by the enzyme with respect to protein- and lipid-linked oligosaccharides strongly suggests that infection by T. cruzi might alter the sialic acid distribution in glycoproteins and glycolipids of the mammalian host.