Bound Substrate in the Structure of Cyanobacterial Branching Enzyme Supports a New Mechanistic Model

Branching enzyme (BE) catalyzes the formation of α-1,6-glucosidic linkages in amylopectin and glycogen. The reaction products are variable, depending on the organism sources, and the mechanistic basis for these different outcomes is unclear. Although most cyanobacteria have only one BE isoform belonging to glycoside hydrolase family 13, Cyanothece sp. ATCC 51142 has three isoforms (BE1, BE2, and BE3) with distinct enzymatic properties, suggesting that investigations of these enzymes might provide unique insights into this system. Here, we report the crystal structure of ligand-free wild-type BE1 (residues 5–759 of 1–773) at 1.85 Å resolution. The enzyme consists of four domains, including domain N, carbohydrate-binding module family 48 (CBM48), domain A containing the catalytic site, and domain C. The central domain A displays a (β/α)8-barrel fold, whereas the other domains adopt β-sandwich folds. Domain N was found in a new location at the back of the protein, forming hydrogen bonds and hydrophobic interactions with CBM48 and domain A. Site-directed mutational analysis identified a mutant (W610N) that bound maltoheptaose with sufficient affinity to enable structure determination at 2.30 Å resolution. In this structure, maltoheptaose was bound in the active site cleft, allowing us to assign subsites −7 to −1. Moreover, seven oligosaccharide-binding sites were identified on the protein surface, and we postulated that two of these in domain A served as the entrance and exit of the donor/acceptor glucan chains, respectively. Based on these structures, we propose a substrate binding model explaining the mechanism of glycosylation/deglycosylation reactions catalyzed by BE. Branching enzyme (BE) catalyzes the formation of α-1,6-glucosidic linkages in amylopectin and glycogen. The reaction products are variable, depending on the organism sources, and the mechanistic basis for these different outcomes is unclear. Although most cyanobacteria have only one BE isoform belonging to glycoside hydrolase family 13, Cyanothece sp. ATCC 51142 has three isoforms (BE1, BE2, and BE3) with distinct enzymatic properties, suggesting that investigations of these enzymes might provide unique insights into this system. Here, we report the crystal structure of ligand-free wild-type BE1 (residues 5–759 of 1–773) at 1.85 Å resolution. The enzyme consists of four domains, including domain N, carbohydrate-binding module family 48 (CBM48), domain A containing the catalytic site, and domain C. The central domain A displays a (β/α)8-barrel fold, whereas the other domains adopt β-sandwich folds. Domain N was found in a new location at the back of the protein, forming hydrogen bonds and hydrophobic interactions with CBM48 and domain A. Site-directed mutational analysis identified a mutant (W610N) that bound maltoheptaose with sufficient affinity to enable structure determination at 2.30 Å resolution. In this structure, maltoheptaose was bound in the active site cleft, allowing us to assign subsites −7 to −1. Moreover, seven oligosaccharide-binding sites were identified on the protein surface, and we postulated that two of these in domain A served as the entrance and exit of the donor/acceptor glucan chains, respectively. Based on these structures, we propose a substrate binding model explaining the mechanism of glycosylation/deglycosylation reactions catalyzed by BE. Branching enzyme (BE 4The abbreviations used are: BEbranching enzymeCBMcarbohydrate-binding modulecceBE1BE1 from Cyanothece sp. ATCC 51142CrISA1isoamylase 1 from C. reinhardtiiDPdegree of polymerizationEcBEBE from E. coliG3maltotrioseG5maltopentaoseG6maltohexaoseG7maltoheptaoseG8maltooctaoseG9maltononaoseG10maltodecaoseGHglycoside hydrolaseHsBEBE from H. sapiensLFligand-freeMtBEBE from M. tuberculosis H37RVOsBEIBEI from O. sativaRMSDroot mean square deviationSBSsurface binding siteWtBE1WT of cceBE1PDBProtein Data Bank.; EC.2.4.1.18) catalyzes the transglucosylation reaction of the α-glucan molecules glycogen and amylopectin by cleaving an α-1,4-linkage to form a new branch point via an α-1,6-linkage (1.Tetlow I.J. Emes M.J. A review of starch-branching enzymes and their role in amylopectin biosynthesis.IUBMB Life. 2014; 66: 546-558Crossref PubMed Scopus (109) Google Scholar, 2.Suzuki E. Suzuki R. Distribution of glucan-branching enzymes among prokaryotes.Cell Mol. Life Sci. 2016; 73: 2643-2660Crossref PubMed Scopus (31) Google Scholar). Glycogen and starch (primarily composed of amylopectin) are major carbohydrates that are accumulated for energy reserve in animals, plants, and microorganisms (3.Adeva-Andany M.M. González-Lucán M. Donapetry-García C. Fernández-Fernández C. Ameneiros-Rodríguez E. Glycogen metabolism in humans.BBA Clin. 2016; 5: 85-100Crossref PubMed Scopus (230) Google Scholar, 4.Pfister B. Zeeman S.C. Formation of starch in plant cells.Cell Mol. Life Sci. 2016; 73: 2781-2807Crossref PubMed Scopus (205) Google Scholar, 5.Wilson W.A. Roach P.J. Montero M. Baroja-Fernández E. Muñoz F.J. Eydallin G. Viale A.M. Pozueta-Romero J. Regulation of glycogen metabolism in yeast and bacteria.FEMS Microbiol Rev. 2010; 34: 952-985Crossref PubMed Scopus (259) Google Scholar, 6.Suzuki E. Suzuki R. Variation of storage polysaccharides in phototrophic microorganisms.J. Appl. Glycosci. 2013; 60: 21-27Crossref Google Scholar). Glycogen has a randomly branched structure and displays high water solubility. By contrast, amylopectin is a water-insoluble polymer due to an ordered pattern of branching structure called tandem cluster structure (7.Nakamura Y. Towards a better understanding of the metabolic system for amylopectin biosynthesis in plants: rice endosperm as a model tissue.Plant Cell Physiol. 2002; 43: 718-725Crossref PubMed Scopus (354) Google Scholar, 8.Ball S. Colleoni C. Cenci U. Raj J.N. Tirtiaux C. The evolution of glycogen and starch metabolism in eukaryotes gives molecular clues to understand the establishment of plastid endosymbiosis.J. Exp. Bot. 2011; 62: 1775-1801Crossref PubMed Scopus (168) Google Scholar). The discrete pattern of branching makes the amylopectin molecule display a much higher molecular weight than glycogen and exhibit distinct properties, such as crystallinity and gelatinization upon hydrothermal treatment (8.Ball S. Colleoni C. Cenci U. Raj J.N. Tirtiaux C. The evolution of glycogen and starch metabolism in eukaryotes gives molecular clues to understand the establishment of plastid endosymbiosis.J. Exp. Bot. 2011; 62: 1775-1801Crossref PubMed Scopus (168) Google Scholar, 9.Fujita N. Starch biosynthesis in rice endosperm.Agri-Biosci. Monogr. 2014; 4: 1-18Crossref Google Scholar). The action of BEs significantly influences the structure and property of these polysaccharides. branching enzyme carbohydrate-binding module BE1 from Cyanothece sp. ATCC 51142 isoamylase 1 from C. reinhardtii degree of polymerization BE from E. coli maltotriose maltopentaose maltohexaose maltoheptaose maltooctaose maltononaose maltodecaose glycoside hydrolase BE from H. sapiens ligand-free BE from M. tuberculosis H37RV BEI from O. sativa root mean square deviation surface binding site WT of cceBE1 Protein Data Bank. Distinct types of BE have been identified in various organisms. BEs found in many bacteria display sequence similarities to those in animals, fungi, and plants (10.Deschamps P. Colleoni C. Nakamura Y. Suzuki E. Putaux J.L. Buléon A. Haebel S. Ritte G. Steup M. Falcón L.I. Moreira D. Löffelhardt W. Raj J.N. Plancke C. d'Hulst C. et al.Metabolic symbiosis and the birth of the plant kingdom.Mol. Biol. Evol. 2008; 25: 536-548Crossref PubMed Scopus (120) Google Scholar). These BEs are also distantly similar to α-amylase and other amylolytic enzymes, and they are collectively classified into the glycoside hydrolase 13 (GH13) family in the CAZy database (a sequence-based database of carbohydrate-active enzymes) (11.Cantarel B.L. Coutinho P.M. Rancurel C. Bernard T. Lombard V. Henrissat B. The Carbohydrate-Active EnZymes database (CAZy): an expert resource for glycogenomics.Nucleic Acids Res. 2009; 37: D233-D238Crossref PubMed Scopus (4114) Google Scholar). As the number of GH13 family members became enormous, the family was subdivided into more than 40 subfamilies (12.Stam M.R. Danchin E.G. Rancurel C. Coutinho P.M. Henrissat B. Dividing the large glycoside hydrolase family 13 into subfamilies: towards improved functional annotations of α-amylase-related proteins.Protein Eng. Des. Sel. 2006; 19: 555-562Crossref PubMed Scopus (427) Google Scholar). The eukaryotic and bacterial BEs are classified into the GH13_8 and GH13_9 subfamilies, respectively (2.Suzuki E. Suzuki R. Distribution of glucan-branching enzymes among prokaryotes.Cell Mol. Life Sci. 2016; 73: 2643-2660Crossref PubMed Scopus (31) Google Scholar, 12.Stam M.R. Danchin E.G. Rancurel C. Coutinho P.M. Henrissat B. Dividing the large glycoside hydrolase family 13 into subfamilies: towards improved functional annotations of α-amylase-related proteins.Protein Eng. Des. Sel. 2006; 19: 555-562Crossref PubMed Scopus (427) Google Scholar). In addition, a different type of BE belonging to GH57 has been identified exclusively in prokaryotes (2.Suzuki E. Suzuki R. Distribution of glucan-branching enzymes among prokaryotes.Cell Mol. Life Sci. 2016; 73: 2643-2660Crossref PubMed Scopus (31) Google Scholar, 13.Murakami T. Kanai T. Takata H. Kuriki T. Imanaka T. A novel branching enzyme of the GH-57 family in the hyperthermophilic archaeon Thermococcus kodakaraensis KOD1.J. Bacteriol. 2006; 188: 5915-5924Crossref PubMed Scopus (76) Google Scholar). Cyanothece sp. ATCC 51142 is a unicellular cyanobacterium that performs oxygenic photosynthesis as well as nitrogen fixation (14.Reddy K.J. Haskell J.B. Sherman D.M. Sherman L.A. Unicellular, aerobic nitrogen-fixing cyanobacteria of the genus Cyanothece.J. Bacteriol. 1993; 175: 1284-1292Crossref PubMed Scopus (0) Google Scholar). Whereas most cyanobacteria produce glycogen, some species, including ATCC 51142 and Cyanobacterium NBRC 102756, produce amylopectin-like polysaccharide, designated as cyanobacterial starch (10.Deschamps P. Colleoni C. Nakamura Y. Suzuki E. Putaux J.L. Buléon A. Haebel S. Ritte G. Steup M. Falcón L.I. Moreira D. Löffelhardt W. Raj J.N. Plancke C. d'Hulst C. et al.Metabolic symbiosis and the birth of the plant kingdom.Mol. Biol. Evol. 2008; 25: 536-548Crossref PubMed Scopus (120) Google Scholar, 15.Nakamura Y. Takahashi J. Sakurai A. Inaba Y. Suzuki E. Nihei S. Fujiwara S. Tsuzuki M. Miyashita H. Ikemoto H. Kawachi M. Sekiguchi H. Kurano N. Some cyanobacteria synthesize semi-amylopectin type α-polyglucans instead of glycogen.Plant Cell Physiol. 2005; 46: 539-545Crossref PubMed Scopus (81) Google Scholar, 16.Suzuki E. Onoda M. Colleoni C. Ball S. Fujita N. Nakamura Y. Physicochemical variation of cyanobacterial starch, the insoluble α-glucans in cyanobacteria.Plant Cell Physiol. 2013; 54: 465-473Crossref PubMed Scopus (20) Google Scholar). A strong correlation has been observed between the type of α-glucan produced and the number of BE genes. Glycogen-producing cyanobacteria have one BE each from the GH13_9 and GH57 (sub)families. By contrast, ATCC 51142 and NBRC 102756 contain two extra GH13_9 subfamily isoforms (2.Suzuki E. Suzuki R. Distribution of glucan-branching enzymes among prokaryotes.Cell Mol. Life Sci. 2016; 73: 2643-2660Crossref PubMed Scopus (31) Google Scholar, 16.Suzuki E. Onoda M. Colleoni C. Ball S. Fujita N. Nakamura Y. Physicochemical variation of cyanobacterial starch, the insoluble α-glucans in cyanobacteria.Plant Cell Physiol. 2013; 54: 465-473Crossref PubMed Scopus (20) Google Scholar). The three GH13_9 BE isoforms from ATCC 51142 and NBRC 102756 have been characterized, and it was found that BE1 and BE2 transferred short glucan chains with a degree of polymerization (DP) of 6–7, whereas BE3 transferred short as well as long glucan chains (DP 30) (17.Suzuki R. Koide K. Hayashi M. Suzuki T. Sawada T. Ohdan T. Takahashi H. Nakamura Y. Fujita N. Suzuki E. Functional characterization of three (GH13) branching enzymes involved in cyanobacterial starch biosynthesis from Cyanobacterium sp. NBRC 102756.Biochim. Biophys. Acta. 2015; 1854: 476-484Crossref PubMed Scopus (18) Google Scholar, 18.Hayashi M. Suzuki R. Colleoni C. Ball S.G. Fujita N. Suzuki E. Crystallization and crystallographic analysis of branching enzymes from Cyanothece sp. ATCC 51142.Acta Crystallogr. F Struct. Biol. Commun. 2015; 71: 1109-1113Crossref PubMed Scopus (9) Google Scholar). Structural determination of the protein is an important factor for studying the catalytic mechanism of enzymes. Three-dimensional structures of BEs from Escherichia coli (EcBE) (19.Abad M.C. Binderup K. Rios-Steiner J. Arni R.K. Preiss J. Geiger J.H. The X-ray crystallographic structure of Escherichia coli branching enzyme.J. Biol. Chem. 2002; 277: 42164-42170Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar), Mycobacterium tuberculosis H37RV (MtBE) (20.Pal K. Kumar S. Sharma S. Garg S.K. Alam M.S. Xu H.E. Agrawal P. Swaminathan K. Crystal structure of full-length Mycobacterium tuberculosis H37Rv glycogen branching enzyme: insights of N-terminal β-sandwich in substrate specificity and enzymatic activity.J. Biol. Chem. 2010; 285: 20897-20903Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar), Oryza sativa (OsBEI) (21.Noguchi J. Chaen K. Vu N.T. Akasaka T. Shimada H. Nakashima T. Nishi A. Satoh H. Omori T. Kakuta Y. Kimura M. Crystal structure of the branching enzyme I (BEI) from Oryza sativa L with implications for catalysis and substrate binding.Glycobiology. 2011; 21: 1108-1116Crossref PubMed Scopus (37) Google Scholar), and Homo sapiens (HsBE) (22.Froese D.S. Michaeli A. McCorvie T.J. Krojer T. Sasi M. Melaev E. Goldblum A. Zatsepin M. Lossos A. Álvarez R. Escribá P.V. Minassian B.A. von Delft F. Kakhlon O. Yue W.W. Structural basis of glycogen branching enzyme deficiency and pharmacologic rescue by rational peptide design.Hum. Mol. Genet. 2015; 24: 5667-5676Crossref PubMed Scopus (39) Google Scholar) have been solved to date. These studies provided basic information on the overall and domain structures of GH13_9 (prokaryotic)- and GH13_8 (eukaryotic)-type BEs. The structures of OsBEI (23.Chaen K. Noguchi J. Omori T. Kakuta Y. Kimura M. Crystal structure of the rice branching enzyme I (BEI) in complex with maltopentaose.Biochem. Biophys. Res. Commun. 2012; 424: 508-511Crossref PubMed Scopus (32) Google Scholar), HsBE (22.Froese D.S. Michaeli A. McCorvie T.J. Krojer T. Sasi M. Melaev E. Goldblum A. Zatsepin M. Lossos A. Álvarez R. Escribá P.V. Minassian B.A. von Delft F. Kakhlon O. Yue W.W. Structural basis of glycogen branching enzyme deficiency and pharmacologic rescue by rational peptide design.Hum. Mol. Genet. 2015; 24: 5667-5676Crossref PubMed Scopus (39) Google Scholar), and EcBE (24.Feng L. Fawaz R. Hovde S. Gilbert L. Chiou J. Geiger J.H. Crystal Structures of Escherichia coli branching enzyme bound to linear oligosaccharides.Biochemistry. 2015; 54: 6207-6218Crossref PubMed Scopus (37) Google Scholar, 25.Feng L. Fawaz R. Hovde S. Sheng F. Nosrati M. Geiger J.H. Crystal structures of Escherichia coli branching enzyme in complex with cyclodextrins.Acta Crystallogr. D Struct. Biol. 2016; 72: 641-647Crossref PubMed Scopus (26) Google Scholar) in complex with linear maltooligosaccharides or cyclodextrins led to the identification of surface binding sites (SBSs) located on the enzyme surface at certain distances from the active site (26.Nielsen M.M. Bozonnet S. Seo E.S. Mótyán J.A. Andersen J.M. Dilokpimol A. Abou Hachem M. Gyémánt G. Naested H. Kandra L. Sigurskjold B.W. Svensson B. Two secondary carbohydrate binding sites on the surface of barley α-amylase 1 have distinct functions and display synergy in hydrolysis of starch granules.Biochemistry. 2009; 48: 7686-7697Crossref PubMed Scopus (61) Google Scholar, 27.Cuyvers S. Dornez E. Delcour J.A. Courtin C.M. Occurrence and functional significance of secondary carbohydrate binding sites in glycoside hydrolases.Crit. Rev. Biotechnol. 2012; 32: 93-107Crossref PubMed Scopus (73) Google Scholar, 28.Cockburn D. Nielsen M.M. Christiansen C. Andersen J.M. Rannes J.B. Blennow A. Svensson B. Surface binding sites in amylase have distinct roles in recognition of starch structure motifs and degradation.Int. J. Biol. Macromol. 2015; 75: 338-345Crossref PubMed Scopus (40) Google Scholar, 29.Wilkens C. Cockburn D. Andersen S. Petersen B.O. Ruzanski C. Field R.A. Hindsgaul O. Nakai H. McCleary B. Smith A.M. Hachem M.A. Svensson B. Analysis of surface binding sites (SBS) within GH62, GH13, and GH77.J. Appl. Glycosci. 2015; 62: 87-93Crossref Google Scholar). Possible roles of SBSs (substrate targeting, guiding substrate into the active site, and passing on reactions products) have been proposed (27.Cuyvers S. Dornez E. Delcour J.A. Courtin C.M. Occurrence and functional significance of secondary carbohydrate binding sites in glycoside hydrolases.Crit. Rev. Biotechnol. 2012; 32: 93-107Crossref PubMed Scopus (73) Google Scholar). However, bound oligosaccharides have never been identified in the active site cleft of BE structures. The interactions of glucan chains at both the active site cleft and SBSs may provide important clues for understanding the reaction mechanism and mode of substrate binding of BEs. In the present study, we determined the crystal structures of BE1 from ATCC 51142 (cceBE1), including ligand-free wild-type BE1 (ligand-free WtBE1 (LF-WtBE1)) and the W610N mutant (where Trp610 was replaced by Asn) in complex with G7 (W610N-G7). These structures revealed the unprecedented arrangement of the domains of cceBE1. Furthermore, the structure of W610N-G7 revealed for the first time the mode of oligosaccharide binding at the active site cleft. The significance of sugar binding at two SBSs was also identified. Based on the experimental evidence, we propose the substrate-binding model of cceBE1. The crystal structure of LF-WtBE1 was determined at 1.85 Å resolution (Table 1). All of the protein crystals analyzed in this study (as described below) contained one molecule each per asymmetric unit. All BE1 structures, including LF-WtBE1, contained 755 residues (residues 5–759 of 1–773) with no disordered region (Fig. 1A). The cceBE1 comprised domain N (residues 5–99), CBM48 (residues 100–206), the catalytic domain known as domain A (residues 207–651), and domain C (residues 652–759) (Fig. 1, A and B). Domain A had a (β/α)8-barrel fold common to all proteins belonging to GH13, whereas domain N, CBM48, and domain C adopted β-sandwich folds. The arrangement of domains in cceBE1 was unprecedented among the available BE structures in that domain N was located at the back (opposite to the active site cleft) of the protein (Fig. 1C). The relative position of the domains was stabilized by three hydrogen bonds and nine van der Waals contacts between domain N and CBM48 in addition to four hydrogen bonds and nine van der Waals contacts between domain N and domain A. Apart from this work, only one study has reported the crystal structure of BE (MtBE) determined with domain N (20.Pal K. Kumar S. Sharma S. Garg S.K. Alam M.S. Xu H.E. Agrawal P. Swaminathan K. Crystal structure of full-length Mycobacterium tuberculosis H37Rv glycogen branching enzyme: insights of N-terminal β-sandwich in substrate specificity and enzymatic activity.J. Biol. Chem. 2010; 285: 20897-20903Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). In MtBE (20.Pal K. Kumar S. Sharma S. Garg S.K. Alam M.S. Xu H.E. Agrawal P. Swaminathan K. Crystal structure of full-length Mycobacterium tuberculosis H37Rv glycogen branching enzyme: insights of N-terminal β-sandwich in substrate specificity and enzymatic activity.J. Biol. Chem. 2010; 285: 20897-20903Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar), domain N was located at one end of the elongated protein, making contact only with CBM48 via 10 hydrogen bonds and 26 van der Waals contacts (Fig. 1C). Excluding domain N, the overall structure of cceBE1 was similar to those of previously reported BEs, including EcBE (PDB code 1M7X; root mean square deviation (RMSD) for 569 Cα atoms is 1.08 Å) (19.Abad M.C. Binderup K. Rios-Steiner J. Arni R.K. Preiss J. Geiger J.H. The X-ray crystallographic structure of Escherichia coli branching enzyme.J. Biol. Chem. 2002; 277: 42164-42170Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar), MtBE (3K1D; RMSD for 596 Cα atoms is 0.98 Å) (20.Pal K. Kumar S. Sharma S. Garg S.K. Alam M.S. Xu H.E. Agrawal P. Swaminathan K. Crystal structure of full-length Mycobacterium tuberculosis H37Rv glycogen branching enzyme: insights of N-terminal β-sandwich in substrate specificity and enzymatic activity.J. Biol. Chem. 2010; 285: 20897-20903Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar), OsBEI (3AML; RMSD for 550 Cα atoms is 1.75 Å) (21.Noguchi J. Chaen K. Vu N.T. Akasaka T. Shimada H. Nakashima T. Nishi A. Satoh H. Omori T. Kakuta Y. Kimura M. Crystal structure of the branching enzyme I (BEI) from Oryza sativa L with implications for catalysis and substrate binding.Glycobiology. 2011; 21: 1108-1116Crossref PubMed Scopus (37) Google Scholar), and HsBE (4BZY; RMSD for 553 Cα atoms is 1.64 Å) (22.Froese D.S. Michaeli A. McCorvie T.J. Krojer T. Sasi M. Melaev E. Goldblum A. Zatsepin M. Lossos A. Álvarez R. Escribá P.V. Minassian B.A. von Delft F. Kakhlon O. Yue W.W. Structural basis of glycogen branching enzyme deficiency and pharmacologic rescue by rational peptide design.Hum. Mol. Genet. 2015; 24: 5667-5676Crossref PubMed Scopus (39) Google Scholar). When only domain N was considered, its structure was highly similar between cceBE1 and MtBE (Z score = 10.5; RMSD for 99 Cα atoms = 2.5 Å), as revealed by a structural homology search using the DALI server (30.Holm L. Rosenström P. Dali server: conservation mapping in 3D.Nucleic Acids Res. 2010; 38: W545-W549Crossref PubMed Scopus (3053) Google Scholar).TABLE 1Summary of data collection and refinement statistics for proteins investigated in this studyParametersValuesPDB code 5GQUPDB code 5GQVPDB code 5GQWPDB code 5GQXData collection statistics Data setLF-WtBE1WtBE1-G6LF-W610NW610N-G7 X-ray sourceKEK PF BL-5AKEK PF AR NW12AKEK PF AR NE3AKEK PF AR NW12A DetectorADSC Quantum 315rADSC Quantum 270PILATUS 2 mADSC Quantum 270 Wavelength (Å)1.000001.000001.000001.00000 Space groupP41212P41212P41212P41212 Unit-cell parameters (Å)a = 133.75, b = 133.75, c = 185.90a = 134.13, b = 134.13, c = 184.32a = 133.62, b = 133.62, c = 185.25a = 134.03, b = 134.03, c = 185.06 Resolution range (Å)50.00 to 1.85 (1.88 to 1.85)50.00 to 3.00 (3.05 to 3.00)50.00 to 1.80 (1.83 to 1.80)50.00 to 2.30 (2.34 to 2.30) Rmerge0.074 (0.274)0.128 (0.322)0.041 (0.198)0.075 (0.261) Completeness (%)100.0 (100.0)99.8 (100.0)100.0 (100.0)100.0 (100.0) Multiplicity14.6 (14.8)13.9 (14.4)13.2 (13.1)14.5 (14.6) Average I/σ (I)61.0 (10.7)40.17 (15.12)44.8 (9.0)52.5 (15.1) Unique reflections143,511 (7126)34,353 (1679)154,705 (7650)75,270 (3702) Total reflections2,096,822476,5492,036,6381,088,438Refinement statistics Resolution47.29 to 1.85 (1.90 to 1.85)47.42 to 3.00 (3.08 to 3.00)47.24 to 1.80 (1.85 to 1.80)47.38 to 2.30 (2.36 to 2.30) R-factor0.148 (0.163)0.166 (0.211)0.150 (0.164)0.151 (0.156) Rfree-factor0.170 (0.200)0.224 (0.340)0.171 (0.205)0.184 (0.227) RMSD from ideal value Bond lengths (Å)0.0280.0140.0260.021 Bond angles (degrees)2.3401.8902.2302.194 No. of water molecules852166863530 Average B-value25.540.722.126.8 Ramachandran plot Favored region (%)97.995.698.397.6 Allowed region (%)1.94.21.62.1 Outlier region (%)0.30.10.10.3 Open table in a new tab Site-directed mutational analyses were performed extensively in this study for amino acid residues around the active site and the known carbohydrate binding sites. The W610N mutant was created because Trp610 is positioned close to the active site, and it displays sequence variation between the related species (17.Suzuki R. Koide K. Hayashi M. Suzuki T. Sawada T. Ohdan T. Takahashi H. Nakamura Y. Fujita N. Suzuki E. Functional characterization of three (GH13) branching enzymes involved in cyanobacterial starch biosynthesis from Cyanobacterium sp. NBRC 102756.Biochim. Biophys. Acta. 2015; 1854: 476-484Crossref PubMed Scopus (18) Google Scholar). The mutant enzyme displayed alterations in catalytic specificity, which prompted us to determine its crystal structure as a ligand-free form or in complex with maltohexaose (G6) or maltoheptaose (G7). The crystal structure of the W610N mutant in complex with G7 (W610N-G7) was determined at 2.3 Å resolution (Table 1). The oligosaccharide molecule was observed at the active site cleft (Fig. 2, A and B) for the first time in the crystal structure of BE. An electron density map of the bound sugar molecule in the active site cleft was shown in Fig. 3A. The amino acid residues, namely Pro217, Glu284, Trp285, Phe323, Asp324, Trp327, Tyr329, Gln330, Asp373, Trp399, His554, and Asp555, involved in recognition of the G7 molecule bound at subsites −7 to −1 are shown in Fig. 4B. Catalytically important residues in the α-amylase family, including the nucleophile (Asp434), acid/base catalyst (Glu487), and second aspartate (Asp555), were located at subsite −1 (Figs. 1A and 5A). The hydrogen bonds and van der Waals contacts between the protein and G7 are summarized in Table 2. The G7 molecule adopted a twisted “S” conformation, as shown previously in the structure of barley α-amylase 1 (Amy1; another related enzyme belonging to GH13_6 subfamily) (Fig. 2, A and B) (31.Robert X. Haser R. Mori H. Svensson B. Aghajari N. Oligosaccharide binding to barley α-amylase 1.J. Biol. Chem. 2005; 280: 32968-32978Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar). A similar binding mode of G7 was also observed in debranching enzyme (isoamylase 1) from Chlamydomonas reinhardtii (CrISA1; belonging to GH13_11 subfamily) (32.Sim L. Beeren S.R. Findinier J. Dauvillée D. Ball S.G. Henriksen A. Palcic M.M. Crystal structure of the Chlamydomonas starch debranching enzyme isoamylase ISA1 reveals insights into the mechanism of branch trimming and complex assembly.J. Biol. Chem. 2014; 289: 22991-23003Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar). The G7 molecule reported in CrISA1 was superimposed onto the W610N-G7 structure (Fig. 5B). The binding modes of the G7 molecules in both structures were apparently identical to each other except that in W610N-G7, the pyranose ring of Glc801 (at the reducing end, as shown in Fig. 5B) was rotated by 91° so that the O1 atom of Glc801 obstructed the position of the Oδ1 atom of the catalytic Asp434. By contrast, G7 in CrISA1 was covalently attached to the catalytic Asp452 (CrISA1 numbering) to form a glycosylated intermediate (32.Sim L. Beeren S.R. Findinier J. Dauvillée D. Ball S.G. Henriksen A. Palcic M.M. Crystal structure of the Chlamydomonas starch debranching enzyme isoamylase ISA1 reveals insights into the mechanism of branch trimming and complex assembly.J. Biol. Chem. 2014; 289: 22991-23003Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar). In W610N-G7, the Oδ1 and Oδ2 atoms of Asp555 and the Nϵ2 atom of His554 hydrogen bonded to the O6, O5, and O6 atoms of Glc801, respectively (Table 2). The consecutive Asp and His residues are highly conserved in GH13 enzymes, and they typically form hydrogen bonds with the O2 and O3 atoms of the Glc unit at subsite −1 (33.Uitdehaag J.C. Mosi R. Kalk K.H. van der Veen B.A. Dijkhuizen L. Withers S.G. Dijkstra B.W. X-ray structures along the reaction pathway of cyclodextrin glycosyltransferase elucidate catalysis in the α-amylase family.Nat. Struct. Biol. 1999; 6: 432-436Crossref PubMed Scopus (366) Google Scholar). The orientation of Glc801 was therefore unusual, and it may not represent the physiological state of the Glc residue, because it was not covalently bound to the protein (Figs. 4B and 5B and Table 2).FIGURE 3Fo − Fc omit electron density maps of the bound maltooligosaccharides in the W610N mutant in complex with maltoheptaose calculated from the final refined structure. All maps were contoured at 2.0σ. A, map of the maltoheptaose unit bound at the active site cleft. B, map of the maltoheptaose unit bound at the A1 site. C, map of the maltotriose unit bound at the A2 site. The glucose units are shown as orange stick models and numbered according to coordinate files.View Large Image Figure ViewerDownload Hi-res image Download (PPT)FIGURE 4Sugar-binding structures of WtBE1 and the W610N mutant in the active site cleft, A1 site, and A2 site. A, G6 molecule bound to the active site cleft of WtBE1-G6. Residues participating in recognition of the G6 molecule are shown as green stick models. Catalytically important residues are indicated as magenta sticks. The bound G6 molecule is shown as an orange stick model. Reducing ends of the sugars are marked with φ. Subsites are labeled from −7 to −2. The estimated hydrogen bonds are shown as dashed lines. B, maltoheptaose (G7) molecule bound to the active site cleft of W610N-G7. Stick models are colored as described for A. The subsites are labeled from −7 to −1. C, stereo view of the superimposed structures of the A1 site in W610N-G7 and binding site III in EcBE in complex with G7 (PDB code 4LPC) (24.Feng L. Fawaz R. Hovde S. Gilbert L. Chiou J. Geiger J.H. Crystal Structures of Escherichia coli branching enzyme bound to linear oligosaccharides.Biochemistry. 2015; 54: 6207-6218Crossref PubMed Scopus (37) Google Scholar). W610N-G7 and EcBE are colored in green and gray, respectively. Residues