Biotechnology

Two Novel Fungal Phenolic UDP Glycosyltransferases from Absidia coerulea and Rhizopus japonicus

Kebo Xie, Xiaoxiang Dou, Ridao Chen, Dawei Chen, Cheng Fang, Zhiyan Xiao, Jungui Dai
Daniel Cullen, Editor
DOI: 10.1128/AEM.03103-16

ABSTRACT

In the present study, two novel phenolic UDP glycosyltransferases (P-UGTs), UGT58A1 and UGT59A1, which can transfer sugar moieties from active donors to phenolic acceptors to generate corresponding glycosides, were identified in the fungal kingdom. UGT58A1 (from Absidia coerulea) and UGT59A1 (from Rhizopus japonicas) share a low degree of homology with known UGTs from animals, plants, bacteria, and viruses. These two P-UGTs are membrane-bound proteins with an N-terminal signal peptide and a transmembrane domain at the C terminus. Recombinant UGT58A1 and UGT59A1 are able to regioselectively and stereoselectively glycosylate a variety of phenolic aglycones to generate the corresponding glycosides. Phylogenetic analysis revealed the novelty of UGT58A1 and UGT59A1 in primary sequences in that they are distantly related to other UGTs and form a totally new evolutionary branch. Moreover, UGT58A1 and UGT59A1 represent the first members of the UGT58 and UGT59 families, respectively. Homology modeling and mutational analysis implied the sugar donor binding sites and key catalytic sites, which provided insights into the catalytic mechanism of UGT58A1. These results not only provide an efficient enzymatic tool for the synthesis of bioactive glycosides but also create a starting point for the identification of P-UGTs from fungi at the molecular level.

IMPORTANCE Thus far, there have been many reports on the glycosylation of phenolics by fungal cells. However, no P-UGTs have ever been identified in fungi. Our study identified fungal P-UGTs at the molecular level and confirmed the existence of the UGT58 and UGT59 families. The novel sequence information on UGT58A1 and UGT59A1 shed light on the exciting and new P-UGTs hiding in the fungal kingdom, which would lead to the characterization of novel P-UGTs from fungi. Molecular identification of fungal P-UGTs not only is theoretically significant for a better understanding of the evolution of UGT families but also can be applied as a powerful tool in the glycodiversification of bioactive natural products for drug discovery.

INTRODUCTION

Glycosylation plays an important role in the biosynthesis of natural products by enhancing their solubility and stability, facilitating their storage and accumulation in living cells, and regulating their chemical properties and bioactivities (1–4). Glycosylation is catalyzed by glycosyltransferases (GTs), which are highly divergent and polyphyletic. GTs can be categorized into 101 numbered families according to sequence similarity, signature motifs, and stereochemistry of the glycoside linkage formed (http://www.cazy.org/ ) (5). Of the 101 families, the family 1 GTs, which are often referred to as UDP glycosyltransferases (UGTs), are characterized by the utilization of UDP-activated sugar moieties as donors and contain a consensus UGT-defining sequence motif near the C terminus (6). Phenolic UGTs (P-UGTs), which are key enzymes in the biosynthesis of naturally occurring phenolic glycosides, are ubiquitous in nature and have been characterized in bacteria, plants, and animals (7). Bacterial P-UGTs have been exploited as enzymatic tools that can be used in glycodiversification and combinatorial applications to produce novel and bioactive glycosylated products (8–10). In plants, numerous soluble cytosolic UGTs have been studied and were determined to play a vital role in the structural modification of secondary metabolites (11–13). In mammals, membrane-bound UGTs are regarded as the dominant phase II drug-metabolizing enzymes, transferring activated sugars to a large number of xenobiotics as well as endobiotics (14, 15). Phylogenetic analysis of these UGTs revealed that members from different phyla clustered to generate animal-, plant-, bacterium-, and virus-specific clades, except for some sterol and lipid UGTs (16). However, the fungus-specific clade was still unknown due to the lack of related fungal UGTs.

In recent years, there have been many reports on the glycosylation of phenolics by whole cells of various fungi. For example, xanthohumol is glycosylated by Absidia coerulea AM93, Rhizopus nigricans UFP701, and Beauveria bassiana AM446 (17, 18); quercetin, kaempferol, and isorhamnetin can be converted to their glycosides by Cunninghamella elegans ATCC 9245 (19); and Trichoderma koningii KTCC 6042 glycosylates silybins A and B to silybin A 3-O-β-glucopyranoside, silybin A 7-O-β-glucopyranoside, silybin B 3-O-β-glucopyranoside, and silybin B 7-O-β-glucopyranoside, respectively (20, 21). However, the genes encoding P-UGTs in fungi have not yet been identified. Therefore, the genes, phylogeny, and catalytic traits of fungal P-UGTs remain enigmatic, which inspired us to search for these unknown enzymes.

In our previous work, whole cells of two fungi, A. coerulea CGMCC 3.2462 and R. japonicus ZW-4, showed the ability to glycosylate two phenolics, magnolol (compound 1) and honokiol (compound 2), to generate the corresponding bioactive glycosides with high yields (22). It is speculated that there are corresponding P-UGTs with high catalytic activity in these two fungi. Therefore, these two fungal strains were considered to be ideal materials for research on fungal P-UGTs. Here, we describe the identification of two novel fungal P-UGTs, UGT58A1 and UGT59A1, from A. coerulea and R. japonicus, respectively. UGT58A1 and UGT59A1 exhibit catalytic promiscuity, regioselectivity, and stereoselectivity. Mutational analysis suggested that UGT58A1 utilized an acceptor-His-Asp catalytic triad mechanism for catalysis. The phylogenetic tree of UGT families was rebuilt, and UGT58A1 and UGT59A1 clustered together to form a new and distinct evolutionary branch.

RESULTS

Biochemical identification of P-UGT activities in cultured mycelia of A. coerulea and R. japonicus.Crude enzyme extracts, including supernatants and microsomal fractions, from A. coerulea and R. japonicas were prepared to further biochemically confirm the activities and to investigate the properties of the P-UGTs. The enzyme assay showed that magnolol (compound 1) can be glycosylated to its corresponding glycoside when using UDP-glucose (UDP-Glc) as a donor with microsomal fractions of these two fungi. However, such activities were not detected in the supernatants (see Fig. S1 in the supplemental material). These results indicated that the P-UGTs responsible for the glycosylation of magnolol (compound 1) are membrane-bound proteins.

Cloning and identification of P-UGTs from A. coerulea and R. japonicus.Although UGTs possess the 44-amino-acid consensus signature motif in the C terminus, this conserved motif is species specific in nucleotide sequence. Thus, degenerate primers should be designed based on the fungal P-UGT genes. Since no fungal P-UGT genes have ever been characterized, we switched to Mucor circinelloides 1006PhL, a filamentous fungus classified in the family Mucoraceae (the same as A. coerulea and R. japonicus). The complete genome data for M. circinelloides were released on the Broad Institute website (http://www.broadinstitute.org/annotation/genome/rhizopus_oryzae/MultiHome.html ). According to the genome information, 81 of the total 12,227 protein-encoding sequences in M. circinelloides are annotated as GTs, most of which are involved in common cellular processes such as cell wall synthesis and protein modification. Of the 81 annotated GT genes, 10 integral hypothetical GT genes showed homology to family 1 GTs and possess the 44-amino-acid consensus UGT signature sequences, the presence of which strongly supports the identification of these proteins as members of the family 1 GTs (16). Figure 1A shows the alignment of the signature sequences of the 10 hypothetical UGTs, and Fig. 1B (red brackets) exhibits the typical 44-amino-acid consensus signature sequences of family 1 GTs. Family 1 GTs are usually responsible for the glycosylation of secondary metabolites and small molecules (23). Therefore, after aligning these 10 gene sequences and subsequently designing degenerated primers based on the conserved motif, we successfully amplified two full-length cDNAs (UGT58A1 and UGT59A1) from A. coerulea and R. japonicas, respectively. The encoded polypeptides of UGT58A1 (539 amino acids) and UGT59A1 (548 amino acids) have signal sequences in the N terminus and transmembrane-spanning domains in the C terminus, as predicted by the SignalP 4.1 server (http://www.cbs.dtu.dk/services/SignalP ) and the TMHMM 2.0 program (http://www.cbs.dtu.dk/services/TMHMM-2.0 ) (Fig. 1B). These results are in agreement with the hypothesis from the crude enzyme assays that the corresponding P-UGTs are membrane proteins.

FIG 1
FIG 1

Multiple alignment of representative UGTs. (A) Web logos of amino acids representing the signature conserved motifs of the 10 annotated UGTs from Mucor circinelloides 1006PhL. The degenerated primers AcUGT-3′ and RjUGT-3′ were designed based on these conserved motifs. (B) Multiple alignment of UGTs of different organisms, including UGT58A1 (A. coerulea, GenBank accession number KX610760 ), UGT59A1 (R. japonicus, accession number KX610761 ), UGT1A1 (Homo sapiens, accession number AAA63195 ), UGT2A1 (H. sapiens, accession number CAB41974 ), UGT2B7 (H. sapiens, accession number AAA36793 ), VvGT1 (Vitis vinifera, accession number AAB81683 ), UGT72B1 (Arabidopsis thaliana, accession number CAB80916 ), UGT85H2 (Medicago truncatula, accession number ABE87250 ), SsfS6 (Streptomyces sp. strain SF2575, accession number ADE34512 ), UrdGT2 (Streptomyces fradiae, accession number AAF00209 ), and LanGT2 (Streptomyces cyanogenus, accession number AAD13553 ). Conserved UGT motifs are highlighted by a red box. The transit peptides in the N termini are highlighted in a red background. The transmembrane sequences in the C domains are highlighted in a blue background.

Although the polypeptides of UGT58A1 and UGT59A1 possess 44 conserved amino acids in their C-terminal domains, they share low sequence identities (less than 30%) with the characterized UGTs in the NCBI database. UGT58A1 showed the highest identity (54%) to an uncharacterized hypothetical protein (GenBank accession number SAL98987 ) from Absidia glauca, and UGT59A1 also exhibited the highest identity (85%) to an uncharacterized hypothetical protein (accession number EIE90240 ) from Rhizopus delemar, which also revealed the scarcity of functional UGTs identified in fungi.

UGT58A1 and UGT59A1 are predicted to be membrane proteins, so the Pichia pastoris expression system was selected to functionally express UGT58A1 and UGT59A1. Interestingly, P-UGT activity was detected in the GT activity assays with the microsomal fractions of P. pastoris cells transformed with pPIC3.5K-UGT58A1/59A1 but not with an empty vector. Recombinant UGT58A1 and UGT59A1 were able to glycosylate magnolol (compound 1) with high conversion yields (95% and 78%, respectively) to generate magnolol-2-O-glucopyranoside (compound 1a) (Fig. 2), the same as the product obtained from whole-cell transformation in our previous work (22). Thus, UGT58A1 and UGT59A1 are characterized as fungal UGTs that possess phenolic glycosylation activity.

FIG 2
FIG 2

Functional characterization of recombinant UGT58A1 and UGT59A1. (A) Reactions catalyzed by UGT58A1 and UGT59A1. (B) HPLC analysis of incubation mixtures containing magnolol (compound 1) and UDP-Glc with the microsomal fractions of P. pastoris harboring the empty vector pPIC3.5K and the recombinant vectors pPIC3.5K-UGT58A1 and pPIC3.5K-UGT59A1, respectively, and the standard magnolol-2-O-β-glucopyranoside (compound 1a). mAU, milli-absorbance unit. (C) Typical negative-ion MS spectra for peaks of compound 1 and the glucosylated product compound 1a.

The systematic names of UGT58A1 and UGT59A1 were obtained based on evolutionary divergence by submitting the sequences to the UGT Naming Committee (http://www.flinders.edu.au/medicine/sites/clinical-pharmacology/ugt-homepage.cfm ) (5). According to the UGT Naming Committee, UGT58A1 and UGT59A1 are two members of the UGT58 and UGT59 families, respectively, and evidentially confirm the existence of these two new UGT families.

Biochemical properties of UGT58A1 and UGT59A1.Further investigations using magnolol (compound 1) as an acceptor and UDP-Glc as a sugar donor revealed that the highest activity occurred at ∼45°C and decreased rapidly at temperatures above 50°C (see Fig. S2A in the supplemental material). Analysis of enzymatic activity within the pH range of 4.0 to 11.0 showed that the optimal pH value was between 8.0 and 9.0 (Fig. S2B). Previous studies reported that divalent metal ions are not essential to GT B-type GTs but can influence their enzymatic activities (24–26). UGT58A1 activity was increased with Mg2+, Mn2+, Ca2+, Ba2+, and Fe2+ and decreased with Ni2+, Co2+, and Cu2+, while Zn2+ did not lead to the glycosylated product. UGT59A1 activity was enhanced with Mg2+, Mn2+, Ca2+, and Ba2+; unaffected by Fe2+; and downregulated with Co2+, Ni2+, and Cu2+, while Zn2+ did not lead to the glycosylated product (Fig. S2C). The apparent Km values of UGT58A1 and UGT59A1 for magnolol (compound 1) were 220.8 μM and 122.0 μM, respectively. For honokiol (compound 2), the apparent Km values of UGT58A1 and UGT59A1 were 65.9 μM and 594.1 μM, respectively.

Substrate specificity assays of UGT58A1 and UGT59A1.The substrate specificities of UGT58A1 and UGT59A1 were tested with 13 potential phenolic acceptor substrates with diverse structures, which included lignans (compounds 1 and 2), flavonoids (compounds 3 and 4), anthraquinone (compound 5), stilbene (compound 6), benzophenone (compound 7), curcuminoid (compound 8), xanthones (compounds 9 to 12), and coumarin (compound 13) (Fig. 3B). Steroids (compounds 14 to 17) (see Fig. S3 in the supplemental material) were also prepared to investigate the glycosylation capabilities of UGT58A1 and UGT59A1.

FIG 3
FIG 3

Substrate specificities of recombinant UGT58A1 and UGT59A1. (A) Conversion rates of enzymatic reactions with various substrates acting as acceptors and UDP-Glc as the sugar donor. (B) Chemical structures of the substrates used for analysis of the catalytic specificity and the corresponding glycosylated products.

From the first-pass analysis by high-performance liquid chromatography (HPLC)-UV/electrospray ionization mass spectrometry (ESI-MS), enzyme-catalyzed glucosylation of the 13 members, including 8 structurally different types, was observed (Fig. 3; see also Fig. S4 to S14 in the supplemental material). In addition to magnolol (compound 1), both UGT58A1 and UGT59A1 were able to catalyze the glucosylation of honokiol (compound 2), chrysin (compound 4), emodin (compound 5), curcumin (compound 8), and three xanthones (compounds 10 to 12). The conversion rates of the substrates described above by UGT58A1 were higher than those by UGT59A1. In addition, fisetin (compound 3), resveratrol (compound 6), 2,4-dihydroxy bezophenone (compound 7), 1,3,8-trihydroxylxanthone (compound 9), and 7-hydroxy-4-methylcoumarin (compound 13) were accepted only by UGT58A1. Above all, the catalytic activities of these two membrane-bound P-UGTs allow them to be applied as enzymatic tools in the synthesis of bioactive glycosides, especially UGT58A1, which is uncommon for a membrane-bound protein. The biological roles of UGT58A1 and UGT59A1 in fungi might be to protect themselves from the diverse toxic xenobiotics in the environment via detoxification by glycosylation, which forced the P-UGTs to evolve with catalytic promiscuity.

It is also important to note that several members (compounds 3 and 5 to 7) that contain multiple nucleophiles led to a single, chromatographically distinct monoglucosylated product, suggesting its regioselectivity, although the regioselectivity depends on the acceptors. For example, UGT58A1 glycosylated fisetin (compound 3) with a conversion rate of 94% to generate only one monoglucosylated product, while fisetin (compound 3) possesses four phenolic hydroxyl groups. Thus, to confirm the glycosylated position and to determine the anomeric stereoselectivity, product 3a was isolated from the preparative-scale reaction and analyzed by MS, 1H nuclear magnetic resonance (NMR), 13C NMR, heteronuclear single quantum coherence (HSQC), and heteronuclear multiple bond correlation (HMBC) (see Fig. S18 to S21 in the supplemental material). The glucosyl moiety of compound 3a was suggested by the 162-atomic-mass-unit-higher shift by MS and the characteristic signals for Glc-H1 (δH 5.17, 1H) and Glc-H2–6 (δH 3.20 to 3.70, 6H) by 1H NMR. The location of the glucosyl moiety was indicated by the correlation between Glc-H1 and C-3 (δC 136.0) in the HMBC spectrum. The large coupling constant (J = 7.8 Hz) of Glc-H1 revealed the β configuration of the glucosidic linkage. Thus, the glycosylated product of compound 3 was identified as fisetin-3-O-β-glucopyranoside (compound 3a), revealing that UGT58A1 also possesses regio- and stereospecificity. However, neither UGT58A1 nor UGT59A1 showed any glycosylation activity toward the representative steroids (compounds 14 to 17) that were accepted by sterol UGTs (Fig. S3). UGT58A1 also exhibited flexibility in tolerance to sugar donors. Of the three commercially available donors (UDP-Glc, UDP-glucuronic acid [UDP-GA], and UDP-galactose [UDP-Gal]), UGT58A1 was able to recognize UDP-Glc and UDP-glucuronic acid with magnolol (compound 1) as an acceptor (Fig. S15).

Phylogenetic analysis of UGT58A1, UGT59A1, and UGTs from other phyla.As the two newly identified P-UGTs from the fungal kingdom and the first members of UGT58 and UGT59 families, UGT58A1 and UGT59A1 will certainly have particularly significant effects on our understanding of the evolution of fungal UGTs. Thus, a neighbor-joining phylogenetic tree was constructed to analyze the evolutionary relationships of UGT58A1 and UGT59A1 with UGTs from different phyla (Fig. 4). At least one member of each UGT family is included in the tree. The two fungal P-UGTs UGT58A1 and UGT59A1 appeared to cluster together to form a new branch, which is clearly divergent from the UGTs of animals, plants, bacteria, and viruses.

FIG 4
FIG 4

Multiorganism neighbor-joining tree containing 103 full-length UGTs. At least one representative member of UGTs for each family was included. Different phyla are color-coded. Protein sequences were aligned by using ClustalW, and the neighbor-joining phylogenetic tree was drawn by using MEGA5. The optimal tree with the sum of a branch length of 46.35 is shown. Numbers below branches represent bootstrap values based on 1,000 replicates, and only values of ≥90% are shown. The branch lengths represent the relative genetic distances. The evolutionary distances were computed by using the Poisson correction method and are in units of the number of amino acid substitutions per site. The number of parsimony-informative sites is 642. Shown are UGTs from Mamestra brassicae multiple nucleopolyhedrovirus (MbMNPV), Buzura suppressaria nucleocapsid nucleopolyhedrovirus (BusuNPV), Lymantria dispar multiple nucleopolyhedrovirus (LdMNPV), Helicoverpa armigera nucleopolyhedrovirus (HaSNPV), Epinotia aporema granulovirus (EpapGV), Autographa californica multicapsid nucleopolyhedrovirus (AcMNPV), Choristoneura fumiferana multiple nucleopolyhedrovirus (CfMNPV), and Anticarsia gemmatalis multicapsid nucleopolyhedrovirus (AgMNPV). GenBank accession numbers and organisms for the relevant UGTs are listed in Table S3 in the supplemental material.

UGTs from animals, plants, bacteria, and viruses tend to form vastly expanded and species-specific clades. However, the fungus-specific clade was still unknown due to the lack of fungal UGTs, although some fungal sterol UGTs (UGT51/52) have been characterized. Generally, lipid and sterol UGTs are not distributed in their specific clades. As shown in Fig. 4, the plant lipid UGT family UGT81 shows more homology to the lipid UGT family UGT106 from bacteria. Similarly to the plant lipid UGTs, fungal sterol UGT family UGT51/52 also tends to be distributed in the bacterium-specific clade, as does another sterol UGT family, UGT80, from plants. Thus, these lipid and sterol UGTs do not belong to the species-specific clades. This is in accordance with sterols and lipids being biological molecules that have evolved before the radiation of the animal/plant/fungus/bacterium kingdoms. These sterol and lipid UGTs from different phyla with similar functions tend to cluster together, which provides a case of evolutionary convergence in these phyla (16). Accordingly, given the significant distinction of the new fungal branch and the catalytic activity of the members UGT58A1 and UGT59A1, this new branch is considered to be the fungus-specific clade that has not been identified previously and is different from the fungal sterol UGT clade.

Fungal phenolic UGTs (UGT58/59) and sterol UGTs (UGT51/52) are distributed in two different evolutionary branches, indicating that phenolic and sterol UGTs in fungi evolved independently. Consequently, the search for new phenolic UGTs from fungi should not focus solely on the degrees of sequence identity with known UGTs. In evolutionarily distant phyla, it is possible that as-yet-unidentified phenolic UGTs share low homology with known UGTs. Notably, with the guidance of sequence information on the UGT58/59 family, different fungal UGTs would be found and classified into this fungus-specific branch. In addition, the significant independence of this new branch (UGT58/59) further explained the failure of previous attempts to clone fungal P-UGTs via the known UGTs from other species.

Insights into the structure and glycosylation mechanism of UGT58A1.UGT crystal structures are highly similar, although the enzymes share relatively low sequence identities (25). Homology modeling of UGT58A1 was achieved via SWISS-MODEL (27). UGT58A1 showed the highest identity (20%) to the template UGT85H2 in the Protein Data Bank, a (iso)flavonoid UGT from Medicago truncatula. Structure-based research on UGT85H2 revealed that an acceptor-His21-Asp125 triad was formed to accomplish the glycosylation reaction, which is similar to the serine hydrolases (28, 29). In the serine hydrolase, the histidine group of the catalytic triad of Ser-His-Asp deprotonates serine for nucleophilic attack at the substrate; protonation of the histidine is subsequently stabilized through an interaction with the aspartate (28). Molecular docking of the acceptor magnolol (compound 1) into the three-dimensional (3D) structure of UGT58A1 built via homology modeling revealed key interactions between magnolol (compound 1) and the corresponding amino acid residues (His47 and Asp145). A similar acceptor-His-Asp catalytic triad was observed in UGT58A1 (Fig. 5).

FIG 5
FIG 5

UGT58A1 utilizes an acceptor-His47-Asp145 catalytic triad for catalysis. (A) Homologous model of UGT58A1 bound with magnolol (compound 1) and interactions between His47, Asp145, and magnolol (compound 1). The crystal structure of UGT85H2 (PDB accession number 2PQ6.1 ) was used as a template for homology modeling. (B) Sequence alignments of the key motifs involved in catalysis by UGT58A1, UGT59A1, and other representative P-UGTs, including UGT71G1 (M. truncatula, GenBank accession number AAW56092 ), UGT78G1 (M. truncatula, accession number A6XNC6 ), VvGT1 (V. vinifera, accession number AAB81683 ), UGT78K6 (Clitoria ternatea, accession number XP_003618665 ), and UGT85H2 (M. truncatula, accession number ABE87250 ). His47 and Asp145 (marked with red stars) were replaced with alanine.

In terms of the donor binding pocket, UGTs usually have a highly conserved motif in their C-terminal domains, which is involved in the binding of the sugar nucleotide with the UDP moiety (5, 6). Based on the alignment, UGT58A1 and UGT59A1 also possess this conserved motif in the donor binding pocket, consistent with other UGTs (Fig. 6).

FIG 6
FIG 6

Sequence alignment of the conserved motifs of UGT58A1, UGT59A1, and other representative UGTs in regions involved in donor binding. The conserved amino acids (Trp369, Pro371, Gln372, Lys376, His378, Phe384, Ser386, His387, Gln389, Asn391, Ser392, Glu395, Asp411, and Gln412) marked with red stars were replaced with alanine (for organisms and GenBank accession numbers, see Table S3 in the supplemental material).

Thus, the highly conserved sites were further investigated to illuminate the mechanism of sugar donor binding. Accordingly, to explore the catalytic mechanism of UGT58A1, potential active sites in the acceptor binding pocket (His47 and Asp145) and the donor binding pocket (Trp369, Pro371, Gln372, Lys376, His378, Phe384, Ser386, His387, Gln389, Asn391, Ser392, Glu395, Asp411, and Gln412) were mutated to Ala (Fig. 5 and 6). The protein expression of these 16 mutants was confirmed via Western blotting (Fig. 7A).

FIG 7
FIG 7

Mutational analysis of the key amino acids involved in the catalytic glycosylation reactions of UGT58A1. (A) Western blotting of wild-type (WT) UGT58A1 and the mutants in which the conserved amino acids were replaced with alanine. (B) Relative activities of wild-type UGT58A1 and the mutants. See Table S4 in the supplemental material for the primers used for mutational analysis. N.D., not determined.

According to the UGT activity assays, the H47A mutant completely lost its catalytic activity, and the D145A mutant showed <5% of the activity of wild-type UGT58A1 (Fig. 7). Similar mutations in plant and human UGTs have shown equivalent results (29). These results suggest that UGT58A1 utilizes a classic catalytic mechanism in which His47 acts as a general base and catalytic residue for enzymatic activity to abstract a proton from the acceptor substrate. A neighboring conserved Asp145 residue interacts with His47 through a hydrogen bond, and thus, an acceptor-His-Asp triad is generated. Asp145 could then stabilize His47 and neutralize its charge after deprotonating the acceptor magnolol (compound 1). The deprotonated magnolol (compound 1), a nucleophilic oxyanion, then attacks the C1 carbon center of the UDP-Glc with direct displacement of the UDP moiety and forms a β-glucosidic bond (30, 31).

In addition, all mutant proteins with a single-residue change in the conserved sites that are involved in the donor binding pocket dramatically lost their activity (Fig. 7). Similar results were also obtained by mutation of other UGTs (31). Generally, the donor binding pockets of UGTs are highly conversed in both sequences and 3D structures, as the donor specificity is relatively strict. These findings indicate that the mechanism of sugar donor recognition of UGT58A1 is also similar to that of known UGTs. The conversed key residues would form hydrogen bonds with the UDP part and interact with the sugar moiety of the sugar donor (28–31). Although a single residue plays a decisive role in recognizing the sugar, donor recognition was thought to be complex and to require multiple amino acids (30, 31). Therefore, UGT58A1, as a fungal UGT, is believed to adopt a GT B fold and utilize a typical acceptor-His-Asp catalytic triad in catalyzing glycosylation reactions.

DISCUSSION

Rationale for seeking fungal UGTs.Various fungi have been proven to have the ability to convert structurally different foreign phenolics into their corresponding glycosides with high yields (17–22). This glycosylation capability might be derived from the mechanism of self-protection from diverse xenobiotics. However, the genes encoding these P-UGTs in fungi remained unknown due to the low degree of homology with known UGTs and, more specifically, with nonfungal UGTs that glycosylate phenolic compounds.

As more whole-genome sequences of fungi become available, data mining of relevant multigene families within whole-genome sequences is possible. In addition to the filamentous fungus M. circinelloides 1006PhL described above, there are another 75 sets of fungal genome data available at FungiDB (http://fungidb.org/fungidb/showApplication.do ). However, determining how to mine the target GT genes from the mountain of data is our preliminary challenge. With sequence information on UGT58A1 and UGT59A1 in hand, we thus used these two sequences for BLASTP queries of FungiDB. Interestingly, only about 50 similar GTs were obtained from the database. Among the 50 GT sequences, 11 were found in Mucorcircinelloides, and 12 were found in Rhizopus delemar. With subsequent molecular cloning and in vitro expression, these cryptic fungal UGTs will be deciphered. The blasting results also indicated that only some fungi harbor the relevant UGT genes, which also means that glycosylation of exogenetic phenolics is not ubiquitous in fungi. It has been decades since the first fungus with the capability of glycosylating phenolics was found. Now, with the help of these two newly identified fungal P-UGTs, genome mining will provide an effective strategy to identify these new fungal UGTs.

Molecular evolution of fungal UGTs.Based on the newly identified fungal UGTs, the phylogenetic tree of the UGT family was rebuilt (Fig. 4). The plant UGTs include three distinct clades: a large plant-specific clade and two minor clades represented by the UGT80 and UGT81 families. For the members of minor clades, UGT80A1/80A2 (sterol UGTs) and UGT81A1 (lipid UGTs) show higher similarity to bacterial UGTs than to members of the plant-specific clade. Similarly to the plant UGTs, the fungal UGTs contain two clades, represented by UGT51/52 and UGT58/59. The clade of UGT51/52, containing two sterol UGTs (UGT51B1 and UGT52A1) from yeast involved in the synthesis of sterol glucosides, tends to be distributed in bacteria. However, the clade of UGT58/59 is significantly distinct from animal-, plant-, bacterium-, and virus-specific clades and possesses phenolic glycosylation activity, so it is reasonable to believe that the clade of UGT58/59 is the fungus-specific clade, which is significant in order to construct and understand the phylogenetic tree of the UGT family. After the findings of P-UGTs from animals, plants, and bacteria, fungal P-UGTs represent the lost piece of the P-UGT puzzle. According to the new phylogenetic tree, fungal sterol UGTs and P-UGTs evolved independently before the bifurcation of the animal/plant/fungus/bacterium kingdoms. However, our understanding of the fungal UGTs is currently based on just over 4 UGT families so far. Thus, different fungal UGTs need to be identified to clarify the molecular evolution of fungal UGTs.

In summary, we characterized two novel fungal P-UGTs, UGT58A1 and UGT59A1, from A. coerulea and R. japonicas, respectively, at the molecular level. These two fungal UGTs represent original members of the UGT58 and UGT59 families and feature catalytic regiospecificity, stereospecificity, and promiscuity. Under the guidance of homology modeling, molecular docking, and mutational analysis, UGT58A1 was speculated to utilize an acceptor-His47-Asp145 catalytic triad mechanism. Of particular note, rebuilding of the phylogenetic tree of the UGT family revealed that these novel P-UGTs evolved independently from other known UGTs and formed a new evolutionary branch. These findings provide insights for the identification of additional and evolutionarily varying P-UGTs from fungi. Furthermore, UGT58A1 and UGT59A1 might be applied as green and efficient biological catalysts in the glycosylation of diverse phenolics to produce bioactive glycosides for drug discovery in the future.

MATERIALS AND METHODS

Fungal materials and substrates. A. coerulea CGMCC 3.2462 and R. japonicus ZW-4 were kept and cultivated as previously described (22). Substrates 1 to 7 (Fig. 3B) were purchased from Nanjing Zelang Medical Technology Co. Ltd. (Nanjing, China). Compounds 8 and 9 (Fig. 3B) were purchased from Shanxi Sciphar Biotechnology Co. Ltd. (Shanxi, China). Xanthones 10 to 13 (Fig. 3B) were synthesized as described in the supplemental material. Steroids (compounds 14 to 16) (see Fig. S3 in the supplemental material) were purchased from Sigma-Aldrich Co. LLC (St. Louis, MO, USA), and compound 17 was isolated previously by our laboratory (32). UDP-Glc, UDP-GA, and UDP-Gal were purchased from Sigma-Aldrich Co. LLC (St. Louis, MO, USA).

Biochemical identification of P-UGT activities in cultured mycelia of A. coerulea and R. japonicus.To biochemically confirm the activities and investigate the properties of P-UGTs in A. coerulea and R. japonicas, crude enzymes, including the supernatant and microsomal fractions, were prepared. The fungal strains of A. coerulea and R. japonicas were cultivated for 4 days. The mycelium was harvested via filtration, washed with sterile water, and then ground and homogenized in liquid nitrogen. The homogenized mycelium was resuspended with extraction buffer (50 mM Tris-HCl [pH 9.0], 1 mM phenylmethylsulfonyl fluoride [PMSF]) to a final concentration of 1 mg/ml. The homogenized mixture was centrifuged at 10,000 × g for 30 min at 4°C. The resulting supernatant was then further ultracentrifuged at 160,000 × g for 90 min at 4°C to pellet the microsomal fraction. After two washes with extraction buffer, the microsomal fraction was resuspended in the same buffer. The total protein concentrations of the supernatant and the microsome were determined by using the Bradford method (33).

Assays were performed by using 50 mM Tris-HCl buffer (pH 9.0) containing 5 mM MgCl2, 400 μM UDP-Glc, 200 μM magnolol (compound 1), and 200 μg of protein from the microsomal fraction or the supernatant in a total volume of 200 μl. The reaction mixtures were incubated at 35°C for 12 h and quenched by the addition of 400 μl methanol. Protein was removed by centrifugation at 14,000 × g for 30 min, and the resulting supernatant was analyzed by HPLC-UV/ESI-MS. Three parallel assays were routinely performed.

Isolation of P-UGT cDNAs from A. coerulea and R. japonicus. A. coerulea and R. japonicus were cultivated for 4 days. Total RNAs were subsequently extracted with an EZNA plant RNA kit (Omega Bio-tek Inc., Doraville, GA, USA) and reverse transcribed by using a SMARTer rapid amplification of cDNA ends (RACE) kit (Clontech Inc., Mountain View, CA, USA). The reverse transcription (RT) products were subjected to RACE according to the manufacturer's protocol. The 3′ ends of the A. coerulea and R. japonicus UGT genes were obtained via 3′ RACE using the degenerated primers AcUGT-3′ (5′-TGGGCWCCTCARACMGCYRTYCT) and RjUGT-3′ (5′-CATCATAGGGTCGCCTAYRTBGCHTTTGG) as forward primers, respectively, and universal primer A mix (UPM) (see Table S1 in the supplemental material) as reverse primers. The International Union of Pure and Applied Chemistry (IUPAC) codes were used for these degenerate primers. The 3′ ends were then sequenced and analyzed. Subsequently, the corresponding 5′ ends of the UGT58A1 and UGT59A1 genes were obtained by 5′ RACE using UPM as forward primers and the gene-specific primers UGT58A1-5′-GSP (5′-CGAGTTTGACCCGCCATGAGATAAAAACAC) and UGT59A1-5′-GSP (5′-TTCTTCTTCATCCAGCGTCCATAGCCATTT) as reverse primers, respectively. Finally, full-length UGT58A1 and UGT59A1 genes were acquired by RT-PCR using the gene-specific primer pairs UGT58A1-Fw (5′-GGATCCATGGGGATTTTTCTTATGAAACTCC) and UGT58A1-Rv (5′-GCGGCCGCCTAGAGTATTTTTTGCTTGATGCGA) and UGT59A1-Fw (5′-GGATCCATGAAGCTTTCGACCCTATCCT) and UGT59A1-Rv (5′-GCGGCCGCTCAAGCAGACTTCACCTTCTTTGC), respectively (the restriction sites for subcloning are underlined). The PCR products were subsequently cloned into the pEASY-Blunt vector (TransGen Biotech, Beijing, China) for sequencing and then subcloned into the pPIC3.5K vector (Invitrogen, Carlsbad, CA, USA), which resulted in the expression constructs pPIC3.5K-UGT58A1 and pPIC3.5K-UGT59A1, respectively.

Heterologous expression in Pichia pastoris.The expression vectors pPIC3.5K-UGT58A1 and pPIC3.5K-UGT59A1 were introduced into P. pastoris strain GS115, resulting in recombinant yeast cells that carried pPIC3.5K-UGT58A1 and pPIC3.5K-UGT59A1, respectively. The empty vector pPIC3.5K was introduced into GS115 competent cells as a control. Both the recombinant yeast strains harboring the UGT58A1 and UGT59A1 genes and the mock transformant (control) were inoculated into 20 ml buffered minimal glycerol complex (BMGY) medium (10 g yeast extract, 20 g peptone, 100 mM potassium phosphate buffer [pH 6.0], and 10 ml glycerol per liter) and incubated at 30°C at 200 rpm for 24 h. The cells were then washed twice with sterile water via centrifugation, and the cell pellet was resuspended in 20 ml BMMY medium, which was the same as BMGY medium but contained 10 ml/liter methanol instead of 10 ml/liter glycerol. Cells were cultured at 30°C at 200 rpm for 4 days. Methanol was added to maintain 1% (vol/vol) to induce the expression of UGT58A1 and UGT59A1.

Cells were collected and resuspended in extraction buffer as described above, and the cell suspension was disrupted by using a high-pressure homogenizer (APV-2000) at 1.2 × 108 Pa. The resulting mixture was centrifuged at 10,000 × g for 30 min at 4°C. The supernatant was then further ultracentrifuged at 160,000 × g for 90 min at 4°C to pellet the microsome fraction. After two washes with 50 mM Tris-HCl buffer (pH 9.0), the membrane fraction was resuspended in the same buffer. The total protein concentration was determined by using the Bradford method (33).

Glycosyltransferase activity assays.For the quantitative determination of P-UGT activity, the standard reaction mixture (200 μl) contained 50 mM Tris-HCl (pH 9.0), 5 mM MgCl2, 400 μM sugar donors, 200 μM acceptors, and 200 μg of protein from the recombinant P. pastoris microsome. Reaction mixtures were incubated at 45°C for 12 h, and the reactions were terminated by the addition of 400 μl methanol to the mixture. Protein was removed by centrifugation at 14,000 × g for 30 min. The enzymatic reactions were analyzed by HPLC-UV/ESI-MS under the conditions described in Table S2 in the supplemental material.

Biochemical properties of the recombinant enzymes.To assess the optimal reaction temperature, the reaction mixtures, which contained 50 mM Tris-HCl (pH 7.4), 5 mM MgCl2, 400 μM UDP-Glc, 200 μM magnolol (compound 1), and 200 μg of protein from the recombinant yeast microsome, were analyzed at eight different temperatures ranging from 25°C to 60°C for 2 h. To investigate the optimal pH, the enzymatic reactions were performed in reaction buffer with pH values ranging from 4.0 to 6.0 (sodium acetate buffer), 6.0 to 7.0 (phosphate buffer), 7.0 to 9.0 (Tris-HCl buffer), and 9.0 to 11.0 (sodium carbonate buffer) at 45°C for 2 h. To test the requirement of GT activity for divalent cations, MgCl2, MnCl2, FeCl2, CaCl2, ZnCl2, CuCl2, CoCl2, NiCl2, and BaCl2 were individually used at a final concentration of 5 mM with UDP-Glc and magnolol (compound 1) in 50 mM Tris-HCl (pH 9.0) at 45°C for 2 h.

Assay mixtures for the determination of kinetic parameters of the acceptors magnolol (compound 1) and honokiol (compound 2) contained 50 mM Tris-HCl (pH 9.0), 1,000 μM UDP-Glc, 5 mM MgCl2, 100 μg of protein from the recombinant yeast microsome, and acceptors at final concentrations of 25, 30, 35, 40, 50, 60, 80, 125, and 250 μM. The reaction mixtures were incubated at 45°C for 20 min. The apparent Km values were calculated from Lineweaver-Burk plots by using Hyper32 software.

Enzymatic synthesis of the glycosylated products for structural elucidation.Assay mixtures for the isolation of enzymatic products contained 50 mM Tris-HCl (pH 9.0), 5 mM MgCl2, 1 mM UDP-Glc, 500 μM acceptors, and 50 mg of protein from the recombinant yeast microsome in a total volume of 25 ml. The reactions were performed at 30°C for up to 12 h, followed by the addition of 50 ml 5× ethyl acetate for extraction. The organic phase was evaporated to dryness under reduced pressure, and the residue was dissolved in methanol and purified by reverse-phase semipreparative HPLC. The obtained products (compounds 1a and 3a) were analyzed by MS and NMR (see Fig. S16 to S21 in the supplemental material). The glucosylated products of compounds 2, 5 to 8, and 13 were analyzed by HPLC-UV/MS and compared with the products obtained in our previous work (12, 22).

For magnolol-2-O-β-glucopyranoside (compound 1a), ESI-MS m/z 427.18 [M − H]−; 1H NMR (600 MHz, dimethyl sulfoxide [DMSO]-d6), δ 3.09 to 3.35 (4H, m, Glc-H), 3.25 to 3.34 (2H, overlapped, H-7, H-7′), 3.44 (1H, m, Glc-H), 3.69 (1H, d, J = 9.9 Hz, Glc-H), 4.94 (1H, d, J = 7.7 Hz, Glc-H1), 5.04 (4H, m, H-9, H-9′, Glc-OH, Glc-OH), 5.95 (2H, m, H-8, H-8′), 6.95 (1H, dd, J = 2.1, 8.2 Hz, H-4′), 6.82 (1H, d, J = 8.2 Hz, H-3′), 7.02 (1H, d, J = 2.1 Hz, H-6′), 7.03 (1H, d, J = 2.2 Hz, H-6), 7.10 (1H, dd, J = 2.2, 8.5 Hz, H-4), 7.12 (1H, d, J = 8.5 Hz, H-3); 13C NMR (150 MHz, DMSO-d6), δ 38.7 (C-7), 38.7 (C-7′), 60.8 (G-6), 69.7 (G-4), 73.4 (G-2), 76.4 (G-3), 77.1 (G-5), 100.4 (G-1), 114.4 (C-3), 115.2 (C-9′), 115.5 (C-9), 116.1 (C-3′), 125.5 (C-1′), 127.8 (C-1), 128.1 (C-4), 128.1 (C-4′), 130.1 (C-5′), 131.5 (C-6′), 131.8 (C-6), 132.3 (C-5), 138.0 (C-8), 138.3 (C-8′), 152.2 (C-2′), 152.7 (C-2).

For fisetin-3-O-β-glucopyranoside (compound 3a), ESI-MS (m/z) 448.72 [M + H]+; 1H NMR (600 MHz, methanol-d4), δ 3.20 (1H, m, Glc-H), 3.36 (1H, overlapped, Glc-H), 3.42 (1H, m, Glc-H), 3.48 (1H, m, Glc-H2), 3.58 (1H, dd, J = 11.9, 5.3 Hz, Glc-H6b), 3.70 (1H, dd, J = 11.9, 2.4 Hz, Glc-H6a), 5.17 (1H, d, J = 7.8 Hz, Glc-H1), 6.88 (1H, d, J = 8.4 Hz, H-5′), 6.90 (1H, d, J = 2.4 Hz, H-8), 6.92 (1H, dd, J = 8.4, 2.4 Hz, H-6), 7.61 (1H, dd, J = 8.4, 2.4 Hz, H-6′), 7.76 (1H, d, J = 2.4 Hz, H-2′), 7.99 (1H, dd, J = 8.4, 2.4 Hz, H-5); 13C NMR (150 MHz, methanol-d4), δ 61.1 (G-6), 69.7 (G-4), 74.3 (G-2), 76.8 (G-3), 76.9 (G-5), 101.7 (C-8), 103.3 (G-1), 114.5 (C-5′), 115.1 (C-6), 115.7 (C-10), 116.2 (C-2′), 121.7 (C-6′), 121.9 (C-1′), 126.4 (C-5), 136.0 (C-3), 144.4 (C-3′), 148.3 (C-4′), 157.3 (C-2), 157.4 (C-9), 163.9 (C-7), 174.7 (C-4).

Computer-assisted sequence analysis.Sequence identities were obtained via alignment of the amino acid sequences performed with the program BLAST 2 Sequences (http://sucest-fun.org/sucamet/blast/wblast2.html ). To construct the phylogenetic tree of the UGT family, the UGT proteins were aligned by using ClustalW (34). At least one member of each UGT family was included. A complete list of the UGTs chosen to represent the individual subfamilies is given in Table S3 in the supplemental material. From the alignments, a consensus phylogenetic tree was generated by the neighbor-joining method using MEGA5 (35). Bootstrap values are indicated as percentages (only those greater than 90% are presented) on the nodes. The bootstrap values were obtained from 1,000 bootstrap replicates. The bar corresponds to 0.2 estimated amino acid changes per site. The optimal tree with the sum of a branch length of 46.35 is shown. The number of parsimony-informative sites is 642.

Homology modeling and molecular docking.The 3D model of UGT58A1 was constructed by using the SWISS-MODEL server (http://swissmodel.expasy.org/ ) and the crystal structure of UGT85H2 (Medicago truncatula, PDB accession number 2PQ6.1 ) as a template. Docking of the glycosyl acceptor magnolol (compound 1) with UGT58A1 was performed by using the Flexible Docking module in Accelrys Discovery Studio 2.5 (36, 37).

Site-directed mutagenesis of UGT58A1 and Western blot analysis.To study the catalytic mechanism of UGT58A1, the predicted catalytic and binding sites according to homology modeling, molecular docking, and conservative regions were chosen for site-directed mutagenesis. The pPIC3.5K-UGT58A1 plasmid was used as the template for site-directed mutagenesis to introduce H47A, D145A, W369A, P371A, Q372A, L376A, H378A, F384A, S386A, H387A, G389A, N391A, S392A, E395A, D411A, and Q412A point mutations into UGT58A1 by using the Fast Mutagenesis system (TransGen Biotech, Beijing, China), with subsequent verification by DNA sequencing. Mutagenic primer pairs are presented in Table S4 in the supplemental material. The mutagenic nucleotides are shown in uppercase type. The recombinant plasmids with point mutations were introduced into the GS115 strain and heterologously expressed as described above. SDS-PAGE and Western blotting of the microsome membrane proteins were performed according to methods described previously to confirm the expression of the recombinant proteins (38). The anti-His tag monoclonal antibody (MAb) (purified IgG/mouse, lot number 006) used for Western blotting was purchased from MBL (Nagoya, Japan). The microsomes of mutants were subsequently used for further activity assays.

Accession number(s).The nucleotide sequences of UGT58A1 and UGT59A1 have been assigned GenBank accession numbers KX610760 and KX610761 , respectively.

ACKNOWLEDGMENTS

This study was financially supported by the CAMS Innovation Fund for Medical Sciences (CIFMS-2016-I2M-3-012) and the National Natural Science Foundation of China (grant numbers 21572277 and 81573317).

FOOTNOTES

    • Received 11 November 2016.
    • Accepted 27 January 2017.
    • Accepted manuscript posted online 3 February 2017.

  • Supplemental material for this article may be found at https://doi.org/10.1128/AEM.03103-16 .

REFERENCES

  1. 1.↵
    1. Modolo LV,
    2. Li L,
    3. Pan H,
    4. Blount JW,
    5. Dixon RA,
    6. Wang X
    . 2009. Crystal structures of glycosyltransferase UGT78G1 reveal the molecular basis for glycosylation and deglycosylation of (iso)flavonoids. J Mol Biol392:1292–1302. doi:10.1016/j.jmb.2009.08.017.
    OpenUrlCrossRefPubMedWeb of Science
  2. 2.↵
    1. Gachon CMM,
    2. Langlois-Meurinne M,
    3. Saindrenan P
    . 2005. Plant secondary metabolism glycosyltransferases: the emerging functional analysis. Trends Plant Sci10:542–549. doi:10.1016/j.tplants.2005.09.007.
    OpenUrlCrossRefPubMedWeb of Science
  3. 3.↵
    1. Jones P,
    2. Vogt T
    . 2001. Glycosyltransferases in secondary plant metabolism: tranquilizers and stimulant controllers. Planta213:164–174. doi:10.1007/s004250000492.
    OpenUrlCrossRefPubMedWeb of Science
  4. 4.↵
    1. Bowles D,
    2. Isayenkova J,
    3. Lim E,
    4. Poppenberger B
    . 2005. Glycosyltransferases: managers of small molecules. Curr Opin Plant Biol8:254–263. doi:10.1016/j.pbi.2005.03.007.
    OpenUrlCrossRefPubMedWeb of Science
  5. 5.↵
    1. Mackenzie PI,
    2. Owens IS,
    3. Burchell B,
    4. Bock KW,
    5. Bairoch A,
    6. Bélanger A,
    7. Fournel-Gigleux S,
    8. Green M,
    9. Hum DW,
    10. Iyanagi T,
    11. Lancet D,
    12. Louisot P,
    13. Magdalou J,
    14. Chowdhury JR,
    15. Ritter JK,
    16. Schachter H,
    17. Tephly TR,
    18. Tipton KF,
    19. Nebert DW
    . 1997. The UDP glycosyltransferase gene superfamily: recommended nomenclature update based on evolutionary divergence. Pharmacogenet Genomics7:255–269.
    OpenUrl
  6. 6.↵
    1. Yonekura-Sakakibara K,
    2. Hanada K
    . 2011. An evolutionary view of functional diversity in family 1 glycosyltransferases. Plant J66:182–193. doi:10.1111/j.1365-313X.2011.04493.x.
    OpenUrlCrossRefPubMedWeb of Science
  7. 7.↵
    1. Bowles D,
    2. Lim EK,
    3. Poppenberger B,
    4. Vaistij FE
    . 2006. Glycosyltransferases of lipophilic small molecules. Annu Rev Plant Biol57:567–597. doi:10.1146/annurev.arplant.57.032905.105429.
    OpenUrlCrossRefPubMedWeb of Science
  8. 8.↵
    1. Chang A,
    2. Singh S,
    3. Phillips GN Jr,
    4. Thorson JS
    . 2011. Glycosyltransferase structural biology and its role in the design of catalysts for glycosylation. Curr Opin Biotechnol22:800–808. doi:10.1016/j.copbio.2011.04.013.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Erb A,
    2. Weiß H,
    3. Härle J,
    4. Bechthold A
    . 2009. A bacterial glycosyltransferase gene toolbox: generation and applications. Phytochemistry70:1812–1821. doi:10.1016/j.phytochem.2009.05.019.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Schmid J,
    2. Heider D,
    3. Wendel NJ,
    4. Sperl N,
    5. Sieber V
    . 2016. Bacterial glycosyltransferases: challenges and opportunities of a highly diverse enzyme class toward tailoring natural products. Front Microbiol7:182. doi:10.3389/fmicb.2016.00182.
    OpenUrlCrossRef
  11. 11.↵
    1. Meech R,
    2. Miners JO,
    3. Lewis BC,
    4. Mackenzie PI
    . 2012. The glycosidation of xenobiotics and endogenous compounds: versatility and redundancy in the UDP glycosyltransferase superfamily. Pharmacol Ther134:200–218. doi:10.1016/j.pharmthera.2012.01.009.
    OpenUrlCrossRefPubMed
  12. 12.↵
    1. Xie K,
    2. Chen R,
    3. Li J,
    4. Wang R,
    5. Chen D,
    6. Dou X,
    7. Dai J
    . 2014. Exploring the catalytic promiscuity of a new glycosyltransferase from Carthamus tinctorius. Org Lett16:4874–4877. doi:10.1021/ol502380p.
    OpenUrlCrossRef
  13. 13.↵
    1. Chen D,
    2. Chen R,
    3. Wang R,
    4. Li J,
    5. Xie K,
    6. Bian C,
    7. Sun L,
    8. Zhang X,
    9. Liu J,
    10. Yang L,
    11. Ye F,
    12. Yu X,
    13. Dai J
    . 2015. Probing the catalytic promiscuity of a regio- and stereospecific C-glycosyltransferase from Mangifera indica. Angew Chem Int Ed Engl54:12678–12682. doi:10.1002/anie.201506505.
    OpenUrlCrossRef
  14. 14.↵
    1. Radominska-Pandya A,
    2. Bratton SM,
    3. Redinbo MR,
    4. Miley MJ
    . 2010. The crystal structure of human UDP-glucuronosyltransferase 2B7 C-terminal end is the first mammalian UGT target to be revealed: the significance for human UGTs from both the 1A and 2B families. Drug Metab Rev42:133–144. doi:10.3109/03602530903209049.
    OpenUrlCrossRefPubMed
  15. 15.↵
    1. Bock KW
    . 2016. The UDP-glycosyltransferase (UGT) superfamily expressed in humans, insects and plants: animal-plant arms-race and co-evolution. Biochem Pharmacol99:11–17. doi:10.1016/j.bcp.2015.10.001.
    OpenUrlCrossRef
  16. 16.↵
    1. Paquette S,
    2. Møller BL,
    3. Bak S
    . 2003. On the origin of family 1 plant glycosyltransferases. Phytochemistry62:399–413. doi:10.1016/S0031-9422(02)00558-7.
    OpenUrlCrossRefPubMedWeb of Science
  17. 17.↵
    1. Tronina T,
    2. Bartmanska A,
    3. Milczarek M,
    4. Wietrzyk J,
    5. Poplonski J,
    6. Roj E,
    7. Huszcza E
    . 2013. Antioxidant and antiproliferative activity of glycosides obtained by biotransformation of xanthohumol. Bioorg Med Chem Lett23:1957–1960. doi:10.1016/j.bmcl.2013.02.031.
    OpenUrlCrossRef
  18. 18.↵
    1. Huszcza E,
    2. Bartmanska A,
    3. Tronina T
    . 2008. Glycosylation of xanthohumol by fungi. Z Naturforsch C63:557–560.
    OpenUrl
  19. 19.↵
    1. Zi J,
    2. Valiente J,
    3. Zeng J,
    4. Zhan J
    . 2011. Metabolism of quercetin by Cunninghamella elegans ATCC 9245. J Biosci Bioeng112:360–362. doi:10.1016/j.jbiosc.2011.06.006.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Kim JH,
    2. Kim BG,
    3. Park Y,
    4. Ko JH,
    5. Lim CE,
    6. Lim J,
    7. Lim Y,
    8. Ahn JH
    . 2006. Characterization of flavonoid 7-O-glucosyltransferase from Arabidopsis thaliana. Biosci Biotechnol Biochem70:1471–1477. doi:10.1271/bbb.60006.
    OpenUrlCrossRefPubMedWeb of Science
  21. 21.↵
    1. Kim HJ,
    2. Park H,
    3. Lee I
    . 2006. Microbial transformation of silybin by Trichoderma koningii. Bioorg Med Chem Lett16:790–793. doi:10.1016/j.bmcl.2005.11.022.
    OpenUrlCrossRefPubMed
  22. 22.↵
    1. Yang L,
    2. Wang Z,
    3. Lei H,
    4. Chen R,
    5. Wang X,
    6. Peng Y,
    7. Dai J
    . 2014. Neuroprotective glucosides of magnolol and honokiol from microbial-specific glycosylation. Tetrahedron70:8244–8251. doi:10.1016/j.tet.2014.09.033.
    OpenUrlCrossRef
  23. 23.↵
    1. Coutinho PM,
    2. Deleury E,
    3. Davies GJ,
    4. Henrissat B
    . 2003. An evolving hierarchical family classification for glycosyltransferases. J Mol Biol328:307–317. doi:10.1016/S0022-2836(03)00307-3.
    OpenUrlCrossRefPubMedWeb of Science
  24. 24.↵
    1. Breton C,
    2. Šnajdrová L,
    3. Jeanneau C,
    4. Koča J,
    5. Imberty A
    . 2006. Structures and mechanisms of glycosyltransferases. Glycobiology16:29R–37R.
    OpenUrlCrossRefPubMedWeb of Science
  25. 25.↵
    1. Hu Y,
    2. Walker S
    . 2002. Remarkable structural similarities between diverse glycosyltransferases. Chem Biol9:1287–1296. doi:10.1016/S1074-5521(02)00295-8.
    OpenUrlCrossRefPubMedWeb of Science
  26. 26.↵
    1. Moréra S,
    2. Larivière L,
    3. Kurzeck J,
    4. Aschke-Sonnenborn U,
    5. Freemont PS,
    6. Janin J,
    7. Rüger W
    . 2001. High resolution crystal structures of T4 phage β-glucosyltransferase: induced fit and effect of substrate and metal binding. J Mol Biol311:569–577. doi:10.1006/jmbi.2001.4905.
    OpenUrlCrossRefPubMedWeb of Science
  27. 27.↵
    1. Biasini M,
    2. Bienert S,
    3. Waterhouse A,
    4. Arnold K,
    5. Studer G,
    6. Schmidt T,
    7. Kiefer F,
    8. Cassarino TG,
    9. Bertoni M,
    10. Bordoli L,
    11. Schwede T
    . 2014. SWISS-MODEL: modelling protein tertiary and quaternary structure using evolutionary information. Nucleic Acids Res42:W252–W258. doi:10.1093/nar/gku340.
    OpenUrlCrossRefPubMedWeb of Science
  28. 28.↵
    1. Li L,
    2. Modolo LV,
    3. Escamilla-Trevino LL,
    4. Achnine L,
    5. Dixon RA,
    6. Wang X
    . 2007. Crystal structure of Medicago truncatula UGT85H2—insights into the structural basis of a multifunctional (iso)flavonoid glycosyltransferase. J Mol Biol370:951–963. doi:10.1016/j.jmb.2007.05.036.
    OpenUrlCrossRefPubMedWeb of Science
  29. 29.↵
    1. Liang DM,
    2. Liu JH,
    3. Wu H,
    4. Wang BB,
    5. Zhu HJ,
    6. Qiao JJ
    . 2015. Glycosyltransferases: mechanisms and applications in natural product development. Chem Soc Rev44:8350–8374. doi:10.1039/C5CS00600G.
    OpenUrlCrossRefPubMed
  30. 30.↵
    1. Modolo LV,
    2. Escamilla-Treviño LL,
    3. Dixon RA,
    4. Wang X
    . 2009. Single amino acid mutations of Medicago glycosyltransferase UGT85H2 enhance activity and impart reversibility. FEBS Lett583:2131–2135. doi:10.1016/j.febslet.2009.05.046.
    OpenUrlCrossRefPubMedWeb of Science
  31. 31.↵
    1. Miley MJ,
    2. Zielinska AK,
    3. Keenan JE,
    4. Bratton SM,
    5. Radominska-Pandya A,
    6. Redinbo MR
    . 2007. Crystal structure of the cofactor-binding domain of the human phase II drug-metabolism enzyme UDP-glucuronosyltransferase 2B7. J Mol Biol369:498–511. doi:10.1016/j.jmb.2007.03.066.
    OpenUrlCrossRefPubMedWeb of Science
  32. 32.↵
    1. Liu J,
    2. Tang W,
    3. Chen R,
    4. Dai J
    . 2015. Microbial transformation of 14-anhydrodigoxigenin by Alternaria alternata. Chem Biodivers12:1871–1880. doi:10.1002/cbdv.201500024.
    OpenUrlCrossRef
  33. 33.↵
    1. Bradford MM
    . 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem72:248–254. doi:10.1016/0003-2697(76)90527-3.
    OpenUrlCrossRefPubMedWeb of Science
  34. 34.↵
    1. Larkin MA,
    2. Blackshields G,
    3. Brown NP,
    4. Chenna R,
    5. McGettigan PA,
    6. McWilliam H,
    7. Valentin F,
    8. Wallace IM,
    9. Wilm A,
    10. Lopez R,
    11. Thompson JD,
    12. Gibson TJ,
    13. Higgins DG
    . 2007. Clustal W and Clustal X version 2.0. Bioinformatics23:2947–2948. doi:10.1093/bioinformatics/btm404.
    OpenUrlCrossRefPubMedWeb of Science
  35. 35.↵
    1. Tamura K,
    2. Peterson D,
    3. Peterson N,
    4. Stecher G,
    5. Nei M,
    6. Kumar S
    . 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol28:2731–2739. doi:10.1093/molbev/msr121.
    OpenUrlCrossRefPubMedWeb of Science
  36. 36.↵
    1. Diller DJ,
    2. Merz KM Jr
    . 2001. High throughput docking for library design and library prioritization. Proteins43:113–124. doi:10.1002/1097-0134(20010501)43:2<113::AID-PROT1023>3.0.CO;2-T.
    OpenUrlCrossRefPubMedWeb of Science
  37. 37.↵
    1. Hodis E,
    2. Sussman JL
    . 2009. An encyclopedic effort to make 3D structures easier to understand. Trends Biochem Sci34:100–101. doi:10.1016/j.tibs.2009.01.001.
    OpenUrlCrossRefPubMed
  38. 38.↵
    1. Lv Q,
    2. Wang W,
    3. Xue J,
    4. Hua F,
    5. Mu R,
    6. Lin H,
    7. Yan J,
    8. Lv X,
    9. Chen X,
    10. Hu ZW
    . 2012. DEDD interacts with PI3KC3 to activate autophagy and attenuate epithelial-mesenchymal transition in human breast cancer. Cancer Res72:3238–3250. doi:10.1158/0008-5472.CAN-11-3832.
    OpenUrlAbstract/FREE Full Text
Back to top
Download PDF
Citation Tools
Print

Alerts
Email

KEYWORDS

Absidia
Glycosides
glycosyltransferases
Phenols
Rhizopus
Uridine Diphosphate
UDP glycosyltransferases
fungi
phenolics
glycosylation
phylogenetics

Related Articles

Cited By...