Protein-specific glycosylation: signal patches and cis-controlling peptidic elements

The term 'protein-specific glycosylation' refers to important functional implications of a subset of glycosylation types that are under direct control of recognition determinants on the protein. Examples of the latter are found in the formation of the mannose-6-phosphate receptor ligand on lysosomal hydrolases, and in polysialylation of NCAM, which are regulated via conformational signal patches on the protein. Distinct from these examples, the beta4-GalNAc modification of N-linked glycans on a selected panel of proteins, such as carbonic anhydrase or glycodelin, was demonstrated recently to require specific protein (sequence) determinants proximal to the glycosylation site that function as cis-regulatory elements. Another example of such a cis-regulatory element was described for the control of mammalian O-mannosylation. In this case, the structural features of substrate sites within the mucin domain of alpha-dystroglycan are necessary, but not sufficient for determining the transfer of mannose to Ser/Thr. Evidence has been provided that an upstream-located peptide is also essential. Such cis-controlling elements provide a higher level of protein specificity, because a putative glycosylation site cannot result from a single point mutation. Here, we highlight recent work on protein-specific glycosylation with particular emphasis on the above-cited examples and we will try to link protein-specific glycosylation to function.