In silico analysis of family GH77 with focus on amylomaltases from borreliae and disproportionating enzymes DPE2 from plants and bacteria

https://doi.org/10.1016/j.bbapap.2015.05.009Get rights and content

Highlights

  • This in silico analysis delivers 416 4-α-glucanotransferases from the family GH77.

  • Thermus aquaticus amylomaltase represents the most typical prokaryotic group.

  • Borreliae produce evolutionary distinct amylomaltases with unique sequence features.

  • Eukaryotic disproportionating enzymes DPE2 have counterparts also among bacteria.

Abstract

The CAZy glycoside hydrolase (GH) family GH77 is a monospecific family containing 4-α-glucanotransferases that if from prokaryotes are known as amylomaltases and if from plants including algae are known as disproportionating enzymes (DPE). The family GH77 is a member of the α-amylase clan GH-H. The main difference discriminating a GH77 4-α-glucanotransferase from the main GH13 α-amylase family members is the lack of domain C succeeding the catalytic (β/α)8-barrel. Of more than 2400 GH77 members, bacterial amylomaltases clearly dominate with more than 2300 sequences; the rest being approximately equally represented by Archaea and Eucarya. The main goal of the present study was to deliver a detailed bioinformatics study of family GH77 (416 collected sequences) focused on amylomaltases from borreliae (containing unique sequence substitutions in functionally important positions) and plant DPE2 representatives (possessing an insert of ~ 140 residues between catalytic nucleophile and proton donor). The in silico analysis reveals that within the genus of Borrelia a gradual evolutionary transition from typical bacterial Thermus-like amylomaltases may exist to family-GH77 amylomaltase versions that currently possess progressively mutated the most important and otherwise invariantly conserved positions. With regard to plant DPE2, a large group of bacterial amylomaltases represented by the amylomaltase from Escherichia coli with a longer N-terminus was identified as a probable intermediary connection between Thermus-like and DPE2-like (existing also among bacteria) family GH77 members. The presented results concerning both groups, i.e. amylomaltases from borreliae and plant DPE2 representatives (with their bacterial counterpart), may thus indicate the direction for future experimental studies.

Introduction

The glycoside hydrolase (GH) family GH77 is a monospecific family containing 4-α-glucanotransferases (EC 2.4.1.25) defined in the well-established sequence-based classification system of all carbohydrate-active enzymes, the CAZy database [1]. This family forms the clan GH-H with the main α-amylase family GH13 and the family GH70 of glucan sucrases [2], [3], [4], [5], [6]. The enzyme 4-α-glucanotransferase is known also as amylomaltase in bacteria [7], [8], [9], [10], [11], [12], [13], [14], [15] and archaeons [16], [17] or as disproportionating enzyme (D-enzyme; DPE) in plants [18], [19], [20], [21]. Within the CAZy, the family GH77 counts almost 2300 sequenced members (1% being biochemically characterized) that originate predominantly from Bacteria (more than 2200 sequences) completed by only around 30 sequences from each Archaea and Eucarya (plants and green algae) [22].

4-α-Glucanotransferase employs the retaining reaction mechanism used in the entire clan GH-H [2] to preferably catalyze inter molecular transglycosylation, shuffling α-1,4-glucan chains by cleaving and reforming α-1,4-glycosidic bonds [3]. In fact, it catalyzes the transfer of a glucan-chain from one α-1,4-glucan to another α-1,4-glucan (or to 4-hydroxyl group of glucose) or within a single linear glucan molecule to produce a cyclic α-1,4-glucan [9], [16], [18]. Concerning the degree of polymerization, the size of the cyclic α-1,4-glucan starts from 17, which is usually much higher than α-, β- and γ-cyclodextrins produced by the family GH13 cyclodextrin glucanotransferase with 6, 7 and 8 glucose molecules, respectively [3]. Amylomaltase was also described for its ability to be used for the in vitro production of enzymatically synthesized glycogen [23]. Importantly, both the enzyme produced by cultivation of Bacillus subtilis expressing the amylomaltase gene from Thermus aquaticus as well as the unnatural enzymatically synthesized glycogen were found to be safe in food production and as a food ingredient for human consumption, respectively [24], [25]. Another interesting example was achieved with the plant DPE from adzuki beans in the enzymatical synthesis of acarviosyl maltooligosaccharides (derivatives of acarbose) that can be used as new inhibitors of glycoside hydrolases [26].

The family GH77 4-α-glucanotransferases adopt an α-amylase type (β/α)8-barrel (TIM-barrel) as a catalytic domain, like all the clan GH-H members [2] that is, however, disrupted by more insertions between the barrel β-strands [27]. Of the insertions that form the three subdomains called B1, B2 and B3, the subdomain B1 protruding out of the barrel in the place of the loop 3 connecting the strand β3 to helix α3, corresponds to domain B typical for the whole clan [2], [27]. While the subdomain B2 is unique to family GH77, the subdomain B3 could play the role of the family GH13 domain C [27]. Note that the absence of a typical antiparallel β-sandwich domain C succeeding the catalytic TIM-barrel represents the main difference discriminating a GH77 4-α-glucanotransferase from the members of the main α-amylase family GH13 [2], [6]. The catalytic machinery consists of a triad of residues — aspartic acid, glutamic acid and aspartic acid — positioned at the barrel strands β4 (catalytic nucleophile), β5 (proton donor) and β7 (transition-state stabilizer), respectively, e.g., Asp293, Glu340 and Asp395 observed in the tertiary structure of the amylomaltase from Thermus aquaticus [27]. In addition, the structures have been published also for the enzymes from Thermus thermophilus [28] and Thermus brockianus [29]. There are two more structures determined but still not published: the one of the amylomaltase from Aquifex aeolicus (Protein Data Bank (PDB) ID: 1TZ7) and the other one of potato D-enzyme (PDB ID: 1X1N); the crystallization of Corynebacterium glutamicum amylomaltase being also reported [30]. From the sequence point of view, all the family GH77 4-α-glucanotransferases can be characterized by 4–7 conserved sequence regions [11] applicable for the entire GH-H clan [31].

It is worth mentioning that the in silico analysis of hypothetical amylomaltase from Borrelia burgdorferi revealed remarkable mutations in functional positions from CSRs [32]. The most important substitution concerned the lysine in the CSR-II that replaced otherwise invariant and functional arginine positioned two residues in front of the strand-β4 catalytic nucleophile [31], [32]. Since the arginine along with the catalytic triad belonged to the only four residues conserved invariantly throughout the α-amylase family, i.e. the clan GH-H [31], the amylomaltase identified first in the B. burgdorferi genome [33] was cloned and characterized confirming that the GH77 protein really exhibits the amylomaltase activity [11]. Currently CAZy contains several putative amylomaltases from different borreliae [22]; while some of them exhibit the remarkable mutations seen in the enzyme from B. burgdorferi, there are others possessing a typical family GH77 amylomaltase sequence [11]. This phenomenon makes the amylomaltases from borreliae an attractive subject for evolutionarily oriented studies of the family GH77.

With regard to DPE, two forms named as DPE1 and DPE2 have been identified in plants and green algae [18], [19], [20], [34], [35], [36], [37]. Whereas the DPE1 represents a typical GH77 4-α-glucanotransferase with ~ 500 amino acid residues [18], [19], [20], [26], the DPE2 contains additional domains. It consists of a GH77 catalytic TIM-barrel domain (analogous to DPE1) that is preceded by two copies of starch-binding domain (SBD) of the carbohydrate-binding module (CBM) family 20 and interrupted by an insertion of ~ 140 amino acids between the catalytic nucleophile and proton donor [38]. Although there is obviously no relationship between the insert itself and the DPE2 catalytic action, the removing of the insert resulted in the enzyme inactivation [39].

The main goal of the present study was therefore to deliver a bioinformatics analysis of the family GH77 focused on amino acid sequences of amylomaltases from borreliae that in their primary structures contain unique sequence features, i.e. natural mutations in functionally important positions. A particular attention has also been paid to DPE2, containing an insert between catalytic nucleophile and proton donor and usually possessing two copies of CBM20 at their N-terminus.

Section snippets

Sequence collection

Sequences were collected based on the information in the: (i) CAZy database [1], [22] for the family GH77; (ii) previous bioinformatics analysis [40] focused on SBDs from families CBM20 and CBM48; (iii) results from the BLAST searches [41] using the insert sequence of the Arabidopsis thaliana DPE2 [38] and the complete sequence of Neospora caninum hypothetical 4-α-glucanotransferase [42].

With regard to CAZy database, all 29 and 24 4-α-glucanotransferases from Archaea and Eucarya (excluding

Sequence analysis

The present in silico study delivers a detailed comparison of 416 amino acid sequences of 4-α-glucanotransferases from the family GH77 (Table 1). The sequences cover all available biochemically characterized amylomaltases and both form of DPE as well as several hundreds of hypothetical family GH77 members (Table S1).

Based on the visual inspection of the alignment of their complete sequences (Fig. S1) as well as taking into account the information concerning their available tertiary structures 

Transparency document

Transparency document

Acknowledgments

This work was supported by the Slovak Science Grant Agency VEGA (the project No. 2/0150/14).

References (65)

  • T.R. Barends et al.

    Three-way stabilization of the covalent intermediate in amylomaltase, an α-amylase-like transglycosylase

    J. Biol. Chem.

    (2007)
  • J.M. Steichen et al.

    Domain characterization of a 4-α-glucanotransferase essential for maltose metabolism in photosynthetic leaves

    J. Biol. Chem.

    (2008)
  • C. Ruzanski et al.

    A bacterial glucanotransferase can replace the complex maltose metabolism required for starch to sucrose conversion in leaves at night

    J. Biol. Chem.

    (2013)
  • S. Janecek et al.

    Structural and evolutionary aspects of two families of non-catalytic domains present in starch and glycogen binding proteins from microbes, plants and animals

    Enzyme Microb. Technol.

    (2011)
  • S.F. Altschul et al.

    Basic local alignment search tool

    J. Mol. Biol.

    (1990)
  • K. Fujii et al.

    Function of second glucan binding site including tyrosines 54 and 101 in Thermus aquaticus amylomaltase

    J. Biosci. Bioeng.

    (2007)
  • W. Rachadech et al.

    Identification of essential tryptophan in amylomaltase from Corynebacterium glutamicum

    Int. J. Biol. Macromol.

    (2015)
  • B.L. Cantarel et al.

    The Carbohydrate-Active EnZymes database (CAZy): an expert resource for Glycogenomics

    Nucleic Acids Res.

    (2009)
  • T. Kuriki et al.

    The concept of the α-amylase family: a rational tool for interconverting glucanohydrolases/glucanotransferases, and their specificities

    J. Appl. Glycosci.

    (2006)
  • S. Janecek et al.

    α-Amylase—an enzyme specificity found in various families of glycoside hydrolases

    Cell. Mol. Life Sci.

    (2014)
  • A.P. Pugsley et al.

    Molecular characterization of malQ, the structural gene for the Escherichia coli enzyme amylomaltase

    Mol. Microbiol.

    (1988)
  • S.K. Goda et al.

    Molecular analysis of a Clostridium butyricum NCIMB 7423 gene encoding 4-α-glucanotransferase and characterization of the recombinant enzyme produced in Escherichia coli

    Microbiology

    (1997)
  • Y. Terada et al.

    Thermus aquaticus ATCC 33923 amylomaltase gene cloning and expression and enzyme characterization: production of cycloamylose

    Appl. Environ. Microbiol.

    (1999)
  • A. Godany et al.

    The unique glycoside hydrolase family 77 amylomaltase from Borrelia burgdorferi with only catalytic triad conserved

    FEMS Microbiol. Lett.

    (2008)
  • S. Hwang et al.

    Biochemical characterization of 4-α-glucanotransferase from Saccharophagus degradans 2–40 and its potential role in glycogen degradation

    FEMS Microbiol. Lett.

    (2013)
  • W. Srisimarat et al.

    A novel amylomaltase from Corynebacterium glutamicum and analysis of the large-ring cyclodextrin products

    J. Incl. Phenom. Macrocycl. Chem.

    (2011)
  • K. Sawasdee et al.

    Direct cloning of gene encoding a novel amylomaltase from soil bacterial DNA for large-ring cyclodextrin production

    Appl. Biochem. Microbiol.

    (2014)
  • T. Kaper et al.

    Amylomaltase of Pyrobaculum aerophilum IM2 produces thermoreversible starch gels

    Appl. Environ. Microbiol.

    (2005)
  • T. Kaper et al.

    Exploring and exploiting starch-modifying amylomaltases from thermophiles

    Biochem. Soc. Trans.

    (2004)
  • F. Wattebled et al.

    STA11, a Chlamydomonas reinhardtii locus required for normal starch granule biogenesis, encodes disproportionating enzyme. Further evidence for a function of α-1,4 glucanotransferases during starch granule biosynthesis in green algae

    Plant Physiol.

    (2003)
  • N.S. Bresolin et al.

    Characterisation of disproportionating enzyme from wheat endosperm

    Planta

    (2006)
  • A.M. Smith et al.

    Starch mobilization in leaves

    J. Exp. Bot.

    (2003)
  • Cited by (18)

    • Enzymatic potato starch modification and structure-function analysis of six diverse GH77 4-alpha-glucanotransferases

      2023, International Journal of Biological Macromolecules
      Citation Excerpt :

      Superimpositions and visualizations were done in PyMol (Schrödinger, LCC). 4αGTs of family GH77 represent four different domain architectures [29] denoted A through D (Fig. 1B). Architecture A consists solely of the catalytic domain, B has an extended N-terminal domain of structural similarity to carbohydrate binding modules (CBMs) [30], while the catalytic domain in architectures C and D is interrupted by a ~140 residue insert of unknown function [9], henceforth referred to as a domain-of-unknown-function (DUF).

    • Starch-binding domains as CBM families–history, occurrence, structure, function and evolution

      2019, Biotechnology Advances
      Citation Excerpt :

      By contrast, in plant 4-α-glucanotransferases of the family GH77 (a member of the α-amylase clan GH-H; MacGregor et al., 2001), which are known as the so-called disproportionating enzymes 2 (DPE2; Lloyd et al., 2004; Chia et al., 2004; Steichen et al., 2008), two copies of the CBM20 module are positioned N-terminally (Fig. 1). It is of note that bacterial counterparts of plant DPE2 exist but they usually lack the CBM20 modules (Kuchtova and Janecek, 2015). Among other CAZy GH families, the CBM20 is present in the α-amylase IgtZ from Bacillus circulans from family GH119 (Watanabe et al., 2006) that has been shown to be related to the second α-amylase family, GH57 (Janecek and Kuchtova, 2012).

    • Photometric assay of maltose and maltose-forming enzyme activity by using 4-alpha-glucanotransferase (DPE2) from higher plants

      2017, Analytical Biochemistry
      Citation Excerpt :

      The maltose exporter of the inner chloroplast envelope membrane (designated as MEX) has been identified in Arabidopsis thaliana [16]. The disproportionating (iso)enzyme 2 (DPE2; in Arabidopsis encoded by gene At2g40840; for other designations see Ref. [17]) is essential for the cytosolic maltose metabolism. DPE2-deficient Arabidopsis mutants are strongly compromised in growth [18–20].

    • Characterization of amylomaltase from Thermus filiformis and the increase in alkaline and thermo-stability by E27R substitution

      2015, Process Biochemistry
      Citation Excerpt :

      TfAM lied in between amylomaltases from Thermus group and of a hyperthermophilic bacterium (A. aeolicus) and archaea (P. aerophilum and T. litoralis), but TfAM is significantly different from amylomaltases from mesophilic bacteria with a long N-terminus (C. glutamicum and E. coli) and plant MeDPE1, with 14–17% and 36% sequence identities [20,25]. It is noticed that the amylomaltase of B. burgdorferi with several mutations at functional positions in conserved sequence regions was distinctly separated from the Thermus group, this finding was in agreement with previous report analyzing 32 Borrelia amylomaltases [52]. TfAM gene was cloned into the expression vector pET17b and the enzyme was overexpressed by T7 promoter system as described in Section 2.

    View all citing articles on Scopus
    View full text