Structural analysis of glucoamylase encoded by the STA1 gene of Saccharomyces cerevisiae (var. diastaticus)

Abstract

The sequence of the STA1-encoded glucoamylase of amylolytic Saccharomyces cerevisiae (var. diastaticus) strains shows two well-defined regions: an amino-terminal part rich in serine and threonine residues and a carboxy-terminal part very similar to the catalytic domain of other fungal glucoamylases. A version of the enzyme in which most of the amino-terminal region was deleted still has glucoamylase activity, indicating that the remaining carboxy-terminal part forms a functional catalytic domain. Homology-based models of the two parts of the protein have been obtained. As expected, the shortened form of the enzyme is very similar to the catalytic domain of related glucoamylases of known structure. However, the amino-terminal part yielded a structure revealing an unexpected similarity to bacterial invasins, suggesting functional connections between several yeast proteins homologous to STA1-encoded glucoamylase and invasins. A characteristic of Saccharomyces glucoamylase in its native form is its extreme degree of glycosylation. Despite its high molecular mass (about 300 kDa), and in contrast with what occurs with other extracellular glycoproteins produced by yeast, the enzyme does not remain attached to the cell wall, being fully and efficiently secreted into the medium, even when it is produced in large amounts by overexpression of its gene.

Overexpression of the glucoamylase-encoding STA1 gene of Saccharomyces cerevisiae var. diastaticus in laboratory and industrial strains of Saccharomyces

“…It belongs to family 15 glycosyl hydrolases along with homologous enzymes from bacteria and fungi but has the unique feature of having a Ser/Thr-rich domain in the N-terminal part of its protein sequence. The efficiency of this enzyme is limited by two factors: the absence of a substrate (starch)-binding domain present in most fungal glucoamylases and poor activity against the a-1,6 glycosidic linkages present at the branching point of the starch molecules (Adam et al 2004;Latorre-García et al 2005).…”

Yeast glucoamylases: molecular-genetic and structural characterization

“…Modelling of the catalytic domain, on the basis of S. fibuligera glucoamylase Glu, confirmed the (α/α) 6 -barrel structure. The model of the N-terminal region resembles the structure of invasin fromYersinia pseudotuberculosis (Adam et al 2004). The model of the glucoamylase Stp1 with the SBD from A. niger fused to its C-terminal part has also been proposed (LatorreGarcia et al 2005).…”

“…Glucoamylase from S.cerevisiae var. diastaticus, in its native form has an extreme degree of glycosylation causing high molecular mass about 300 kDa (Adam et al 2004). Nevertheless, the enzyme is efficiently secreted into medium where it is found in multiple isoforms, attributed to different extent of proteolysis and N-linked glycosylation reactions (Yamashita et al 1986).…”

“…The glucoamylase S1 (StaIp) deduced from the structural gene STA1 (Yamashita et al 1985c;Adam et al 2004) is 767 amino acid-residue long protein consisting of a signal peptide (1-21), an N-terminal part of a threonine-rich tract with direct repeat sequences of 35 amino acids (22-347) and a C-terminal part comprising subunits H (348-691) and Y (692-767). The truncated protein in which the Nterminal part was deleted still possesses the glucoamylase activity indicating that the C-terminal part forms a functional catalytic domain.…”

“…diastaticus glucoamylase Sta1p, encoded by the STA1 gene, has also been modelled (Adam et al 2004). The enzyme contains two domains that are structurally and functionally different: a catalytic domain corresponding to the C-terminal part of the enzyme with glucoamylase activity (and homology to glucoamylases) and a Ser/Thr-rich domain, with homology to proteins encoded by FLO genes and structural similarity to bacterial invasins (without any starch-binding function observed).…”