Homology model of starch synthases

Abstract

Starch synthases are enzymes that catalyze the synthesis of starch in plants using ADP-glucose as the donor to form α-1,4-glucosidic bonds. Despite the importance of these enzymes, no information about their three-dimensional structure is available. To understand more about the architecture of the active site and its relationship with other distantly related enzymes, we built homology models of the catalytic domain of soluble starch synthases. To obtain a model of the closed active form we combined templates using the program Modeller. The crystal structure coordinates of the open inactive form of glycogen synthase from Agrobacterium tumefaciens and the closed form of maltodextrin phosphorylase from Escherichia coli were used as templates for the modeling. Glycogen synthase has the highest identity (~30%) and provides the best possible local information, whereas maltodextrin phosphorylase provides the correct relationship between the N- and C-terminal domains for a closed structure. To build a model of the open inactive form only glycogen synthase was used as a template. We found that there is a striking similarity between the active site of starch synthases and bacterial glycogen synthases, which supports the idea of a close evolutionary origin. Our model was validated by the observation that critical negatively charged residues found in the literature were located in the active site. This model will provide an optimum guide for aligning glycogen synthases and starch synthases to study their phylogenetic origins. This work was supported by start-up funds from Loyola University Chicago.