Nomenclature for sugar-binding subsites in glycosyl hydrolases.

The huge structural diversity of polysaccharides leads to their central roles in food storage and utilization, structure, cell–cell signalling, cell-wall expansion and turnover and viral invasion. Glycosyl hydrolases, enzymes hydrolysing the glycosidic bond in di-, oligoand poly-saccharides, are found in all living organisms. The first X-ray structural determination of an enzyme was of a glycosyl hydrolase: hen egg-white lysozyme (HEWL) [1]. Since then, over 57 sequence-based families of glycosyl hydrolases have been identified [2], and three-dimensional structures are known for representatives of over 22 of these [3]. This rapid growth of known three-dimensional structures of glycosyl hydrolases has been accompanied by a diverse and disparate array of nomenclature for the labelling of their sugar-binding subsites. In a number of depolymerizing enzymes, catalytic activity is affected by substrate-binding sites distant from the bond actually undergoing hydrolysis. Such a subsite system is not only encountered in glycosyl hydrolases, but also in proteinases and nucleases. The number of subsites, the energy of interaction of each subsite and the hydrolytic rate coefficients may be determined experimentally [4]. There is clearly a need for an appropriate and consistent nomenclature for the labelling of the subsites in glycosyl hydrolases. Sadly, the current literature is beset with problems regarding the naming of the enzyme subsites which bind saccharides. Whilst most enzymologists have chosen to use one system, the crystallographic community boasts almost as many nomenclatures as there are published papers. Comparisons between various complexes of the same enzyme are difficult, and between different enzymes, almost impossible. The fundamental basis for a consensus nomenclature must be that it allows comparison both between different enzymes with different numbers of subsites and between several complexes of the same enzyme. Two criteria are essential : it must indicate the position of the subsite relative to the point of cleavage and must not change the subsite labelling when new complexes, with extra sugar units at either the reducing or the non-reducing end, become known. We propose that the structural-biology community adopts the ®n to ­n subsite nomenclature widely used by molecular enzymologists. Subsites are labelled from ®n to ­n (where n is an integer). ®n represents the non-reducing end and ­n the reducing end, with cleavage taking place between the ®1 and ­1 subsites. Before detailing this nomenclature, we place the various labelling schemes for polysaccharide-degrading enzymes into a historical perspective and address their respective strengths and weaknesses.