Glucoamylase (1,4-α-D-glucan glucohydrolase, EC 126.96.36.199; GA) is an exohydrolase that catalyzes the release of β-D-glucose by hydrolyzing α-1,4- and α-1,6-glucosidic linkages at the non-reducing ends of raw or soluble starches and related oligosaccharides [1-3]. GA has been used in industrial processes such as the production of glucose syrup and other food-processing applications [2,4]. Although many fungal species are capable of producing GA under different growth conditions , the industrial development of GA has focused only on GA from Aspergillus niger (AnGA; identical to Aspergillus awamori GA) and Rhizopus oryzae (RoGA) because of their stability and high activity [3,6,7]. The overall domain structure of AnGA consists of an N-terminal catalytic region and a C-terminal starch-binding domain (SBD). In contrast, the organization of that of the RoGA consists of an N-terminal SBD and a C-terminal catalytic region. The biochemical properties of AnGA have been well characterized [2,8-16], whereas less is known about RoGA [17,18]. This work focuses on functional analysis of RoGA.
RoGA is synthesized as a precursor containing a typical hydrophobic secretory signal sequence of 25 amino acids. The mature form of RoGA is a single-chain protein composed of three domains: an SBD (residues 26–131), a Thr/Ser-rich linker region (residues 132–167), and a catalytic domain (residues 168–604) . The schematic representation of RoGA is shown in Figure 1. TheN-terminal SBD belongs to the carbohydrate-binding module (CBM) family 21 and shows a relatively low level of similarity to SBDs in other starch-degrading enzymes [20-23]. The C-terminal catalytic domain of RoGA plays an active role in hydrolyzing starch and has a high degree of sequence similarity to those of other fungal GAs . The linker region between these two functional domains is rich in hydroxyl-amino acid residues, but information about its function is quite limited [16,24,25]. In the CAZy classification based on amino acid sequences of the catalytic domain, GAs are classified into glycoside hydrolase (GH) family 15 , and on two recently published articles [27,28], the N-terminally positioned SBD (CBM21) and the C-terminally positioned SBD (CBM20) are classified to be grouped into a common CBM clan. However, the sequences of the linker regions are highly variable. Comparison of the primary structures of different fungal GAs reveals that the linker sequences vary greatly in length and composition (Table 1).
In some fungal hydrolases, the substrate-binding and catalytic domains are separated by a linker segment rich in proline and hydroxy amino acid residues, some of which have been shown to be involved in various functions including enzymatic activity, stability, protein secretion, and ligand binding [29,30]. In AnGA, the heavily O-glycosylated linker domain is essential for secretion and is responsible for enzyme stability as well as activity toward raw starch [11,16,24]. Aside from the knowledge that the linker region in RoGA acts as an interdomain spacer, very little is known about this specific sequence stretch. The linker region of RoGA contains high percentages of Thr (44%) and Ser (16%) residues. Along with numerous putative O-glycosylation sites, the RoGA linker also contains one potential N-glycosylation site adjacent to the catalytic domain (Asn167-Ser-Thr, Figure 1).
Here the baker's yeast, Saccharomyces cerevisiae, was used as a host system to study the structure and function of RoGA, with special attention to the interdomain linker region. The effects of the linker segment with specific length, composition, and glycosylation on the properties of the protein expression, ligand or substrate binding, enzyme activity and stability have been thoroughly investigated.