1. The first approach is based on the ability of glycosides with hydrophobic aglycon (long alkyl, phenyl, "Lemieux-type" spacer) to bind reversibly to hydrophobic adsorbents (for example, Sep Pack C18). Being hydrophobic enough, biotinylated monomeric probes and also the products of their interaction with radioactively labelled nucleoside-phosphate sugars can be adsorbed onto C18 column from water or buffer solutions. After the sorbent washing fron the non-reacted glycosyl donor, the radioactively labelled glycosylation product can then be eluted from the sorbent using methanol: the subsequent measurement of incorporated radioactivity is proportional to enzyme activity.

2. The second approach seems to be the most practical since it is both radioactivity-free and allows many assays simultaneously. These are particularly important features for large scale protocols such as those used in the search for glycosyltransferase antagonists or the rigorous demands of routine pathology screening. The principle of the third approach involves an enzymatic reaction on the surface of polystyrene plates coated by Sug-PAA. Unlabelled nucleoside-phosphate sugar attaches to the bound glycoconjugate and the reaction product is quantified with the help of antibodies or lectin (see Fig). It has been shown that an assay of this type may be of comparable sensitivity to radioactivity based measurements.

3. The third approach is analogous to second one exept that gives the possibility to perform the transferase-substrate reaction in solution which may be advantageous in some cases. Enzyme incubation with biotinylated polymeric probe in the presence of "cold" nucleoside-phosphate sugar in 96-well nonadsorbing plastic gives rise to a glycosylation product. This, together with the non-reacted probe, can be quantitatively adsorbed onto a solid phase by the transfer of the reaction mixture to another 96-well streptavidin-coated plate. The product of enzymatic reaction can be determined with the help of antibodies or lectin (see Fig.).