This minireview covers comparative sequence analysis of proteins using hydrophobic cluster analysis. One of the major points to come from this paper is the discovery of two domains for glycosyl transferases. Non-processive glycosyl transferases form proteins involved in succinoglycan synthesis or dolichol-phosphatase mannose synthesis in yeasts or Rhizobium etc., have a single domain with two polar Asp residues found to be highly conserved. On the other hand, processive glycosyl transferases such as cellulose synthase from Acetobacter, hyaluronan synthase from Streptococcus, alginate synthesis from Pseudomonas, nod factor synthesis form Azorhizobium, chitin synthase from Saccharomyces and even DG42 from Xenopus, all have the same domain as the non-processive group, but in addition, have domain B which has an additional Asp residue and a highly conserved QXXRW motif. From these data analyses come a new hypothesis- the double addition model for polymerization of glucan chains in cellulose biosynthesis. Rather than a single UDP-glucose binding to the enzyme site, two UDP-glucoses bind to enzyme and add two glucoses to the growing chain. This new model obviates the necessity of rotating the glucose molecules which is the final conformation in ß-1,4 linked glucans. The common mechanisms for glycosylation reactions among prokaryotic and eukaryotic organisms seems to explain the mechanisms for processive and non-processive polymerization mechanisms. This review has provided the basis for identification of cellulose synthases from vascular plants.
For Online Viewing (1 Mb)
For Printing (5 Mb)
Up to the 1995 Publications Page
Up to Publication Listing Page