Enzymatic Biosynthesis of Novel Resveratrol Glucoside and GlycosideDerivatives

Ramesh Prasad Pandey,a Prakash Parajuli,a Ju Yong Shin,a Jisun Lee,b Seul Lee,b Young-Soo Hong,c Yong Il Park,b Joong Su Kim,d

Jae Kyung Sohnga

Institute of Biomolecule Reconstruction, Department of Pharmaceutical Engineering, Sun Moon University, Tangjeonmyun, Asan-si, Chungnam, South Koreaa;Department of Biotechnology, Catholic University of Korea, Bucheon, Gyeonggi-do, South Koreab; Chemical Biology Research Center, Korea Research Institute ofBioscience and Biotechnology, Ochang-eup, Chungbuk, South Koreac; Bioindustry Process Center, Jeonbuk Branch Institute, Korea Research Institute of Bioscience andBiotechnology, Jeonbuk, Jeong-Ub, South Koread

A UDP glucosyltransferase from Bacillus licheniformis was overexpressed, purified, and incubated with nucleotide diphosphate(NDP) D- and L-sugars to produce glucose, galactose, 2-deoxyglucose, viosamine, rhamnose, and fucose sugar-conjugated res-veratrol glycosides. Significantly higher (90%) bioconversion of resveratrol was achieved with �-D-glucose as the sugar donor toproduce four different glucosides of resveratrol: resveratrol 3-O-�-D-glucoside, resveratrol 4=-O-�-D-glucoside, resveratrol 3,5-O-�-D-diglucoside, and resveratrol 3,5,4=-O-�-D-triglucoside. The conversion rates and numbers of products formed were foundto vary with the other NDP sugar donors. Resveratrol 3-O-�-D-2-deoxyglucoside and resveratrol 3,5-O-�-D-di-2-deoxyglucosidewere found to be produced using TDP-2-deoxyglucose as a donor; however, the monoglycosides resveratrol 4=-O-�-D-galacto-side, resveratrol 4=-O-�-D-viosaminoside, resveratrol 3-O-�-L-rhamnoside, and resveratrol 3-O-�-L-fucoside were producedfrom the respective sugar donors. Altogether, 10 diverse glycoside derivatives of the medically important resveratrol were gener-ated, demonstrating the capacity of YjiC to produce structurally diverse resveratrol glycosides.

Resveratrol is a phytoalexin that is naturally produced by vari-ous plants and is present in grapes, peanuts, blueberries, and

hops (1–4). Resveratrol is often produced as a defense againstmicrobial infections by Botrytis cinerea (5), radical damage, UVirradiation (6), and other stressors. The trans and cis configura-tions of resveratrol exist in glucosides, dimers (pallidol) (7), tri-mers (grandiphenol C) (8), �-viniferin (9), and polymers (Fig. 1).Moreover, the biological activities of resveratrol are mostly asso-ciated with its trans configuration (10–12). Other postmodifiedresveratrol derivatives with potent biological activities, such asantioxidant, cardioprotective, neuroprotective, anticancer, anti-estrogenic, and antiaging properties, have also been identified indifferent plant sources, including pterostilbene, trans-polydatin(trans-piceid), piceatannol (synonym astringinin), and astringin(Fig. 1) (12–14).

Due to the popularity of the “French paradox,” a wide array ofstudies on resveratrol have been performed. As a result, the bio-logical activities of resveratrol against various diseases, includingtype II diabetes (15–17), obesity, atherosclerosis (18, 19), Alzhei-mer’s disease (20), cardiovascular diseases (including hyperten-sion [21, 22] and ischemic injury [23–25]), and various cancers(26), have been demonstrated. Similar to other antioxidants, res-veratrol neutralizes self-generated free radical species, includingreactive oxygen species and reactive nitrogen species, and preventsradical damage to organs. Thus, most of the biological activities ofresveratrol are attributed to its intrinsic radical-scavenging activ-ity (27). Resveratrol exhibits anti-inflammatory properties, asdemonstrated by its ability to inhibit the production of nitric ox-ide (NO) and proinflammatory cytokines, including tumor ne-crosis factor alpha and interleukin-6 (IL-6), from macrophages(28–30). Resveratrol also has numerous functional targets insidecells, including receptors (either membrane bound or intracellu-lar), signaling molecules (sirutins [31] and the 5=-AMP-activatedkinase signaling pathway [32]), transcription and DNA repair fac-

tors, and oxidative enzymes (33, 34). Furthermore, based on thedatabase available at the U.S. National Institutes of Health website(http://www.clinicaltrials.gov/), a series of clinical investigationsof resveratrol are in progress, as well. However, additional studiesare needed to address the safety, pharmacokinetics, pharmacody-namics, and clinical efficacy of resveratrol, as these parameters areessential for the advancement of resveratrol for clinical applica-tion (35).

Most biologically active natural products (NPs) and recentlyused therapeutics, including calicheamicin (36), daunomycin(37), streptomycin (38), vancomycin (39), digitotoxin (40), andamphotericin (41), are decorated with diverse types of sugar moi-eties to enhance their physical, chemical, and biological proper-ties. These added sugar moieties can also be important for therecognition of different target molecules in a cell (42). Moreover,the adsorption, distribution, metabolism, and excretion proper-ties of drugs are also influenced by the type of sugar moiety at-tached to them (43). Hence, glycosylation has been developed asan efficient tool for generating biologically potent and novel NPglycoside compounds with enhanced drug efficacy.

Various NPs have been successfully glycodiversified, either byenzymatic or by chemical synthesis approaches. Digitoxin (44),novobiocin (45), vancomycin (46), colchicine (47), and 4-methyl-

Received 22 June 2014 Accepted 27 August 2014

Published ahead of print 19 September 2014

Editor: H. Nojiri

Address correspondence to Jae Kyung Sohng, sohng@sunmoon.ac.kr.

Supplemental material for this article may be found at http://dx.doi.org/10.1128/AEM.02076-14.

Copyright © 2014, American Society for Microbiology. All Rights Reserved.

doi:10.1128/AEM.02076-14

December 2014 Volume 80 Number 23 Applied and Environmental Microbiology p. 7235–7243 aem.asm.org 7235

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

umbelliferone (48) have been glycorandomized, thereby produc-ing a number of unusual sugar-bearing NPs. Among these, somehave shown enhanced biological properties compared to their re-spective parent molecules (44). In addition, glycosyltransferases(GTs) (e.g., oleandomycin GT [OleD] and its variants [49, 50], kana-mycin GT-KanE, and vancomycin GT-GtfE [51]) have been engi-neered to increase their flexibility in accepting various donor andacceptor substrates. For the present study, a GT, YjiC (GenBank ac-cession no. AAU40842.1) from Bacillus licheniformis DSM 13, witha flexible donor and acceptor substrate profile was selected in or-der to synthesize glycodiversified resveratrol derivatives. YjiC hasbeen shown to accept geldanamycin analogues (52), epothilone A(53), various flavonoids (54, 55), chalcone (56), and isoflavonoids(57) as substrates for glucosylation. In this study, resveratrol wasused as an acceptor substrate and various rare nucleotide diphos-phate sugars were used as sugar donor substrates to produce gly-codiversified resveratrol derivatives that have not been previouslydescribed.

MATERIALS AND METHODSChemicals and reagents. Resveratrol, UDP-D-glucose, UDP-D-galactose,UDP-D-glucuronic acid, and UDP-D-N-acetylglucosamine were pur-chased from Sigma-Aldrich (St. Louis, MO, USA). TDP-L-rhamnose andGDP-L-fucose were purchased from GeneChem (Daejeon, South Korea).TDP-D-2-deoxyglucose and TDP-D-viosamine were gifts from Dae HeeKim (GeneChem, Daejeon, South Korea). All other chemicals and re-agents were of the highest chemical grade.

Gene cloning and protein expression and purification. The details ofthe methodology for the cloning, expression, and purification of YjiC aredescribed in our previous reports (53–57).

General laboratory-scale glycosyltransferase enzymatic reaction.The glycosylation reactions using YjiC were carried out in a total reactionmixture volume of 100 �l containing 100 mmol Tris-Cl (pH 8.0), 4 mmolnucleotide diphosphate (NDP) sugar, 2 mmol resveratrol, 10 mmolMgCl2 · 6H2O, and 50 �g/ml (final concentration) enzyme. The reactionmixtures were incubated at 37°C for 3 h. Assay mixtures lacking enzymeserved as controls. The reaction was quenched by adding 400 �l chilledmethanol and mixed by vortexing for several minutes. Then, the aliquotswere centrifuged at 12,000 rpm to remove denatured proteins, and the

FIG 1 Structures of resveratrol and its polymers, along with examples of methoxylated and glycosylated resveratrol derivatives. The latter derivatives were eitherisolated from natural sources or enzymatically synthesized.

Pandey et al.

7236 aem.asm.org Applied and Environmental Microbiology

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

reaction product was directly monitored by high-performance liquidchromatography (HPLC) coupled with a photodiode array (PDA) afterproper dilution.

Preparative-scale reaction of resveratrol with UDP-D-glucose. Thereaction mixture contained purified YjiC enzyme (500-�g/ml final con-centration in the reaction mixture), 20 mmol (final concentration) ofUDP-D-glucose (�122 mg) as a donor substrate, 5 mmol resveratrol (11.4mg dissolved in dimethyl sulfoxide [DMSO]) as an acceptor substrate,and 100 mmol Tris-Cl (pH 8.0) buffer containing 10 mmol MgCl2 · 6H2O.The reaction was carried out in a 10-ml final volume for 12 h at 37°C andstopped by adding a triple volume of chilled methanol. Then, the reactionmixture was mixed by vortexing and centrifuged at 12,000 rpm for 30 minat 4°C to remove denatured protein. The supernatant was concentrated byevaporation and lyophilization. Finally, the compounds were dissolved in2 ml HPLC grade methanol and subjected to preparative HPLC purifica-tion (Shimadzu, Tokyo, Japan). The purified compounds were dried, ly-ophilized, and dissolved in methanol-D4 and further analyzed by nuclearmagnetic resonance (NMR).

Deglycosylation reaction assay. The deglycosylation assay was car-ried out using purified resveratrol 3-O-�-D-glucoside as the substrate. Thereaction mixture contained 100 mmol Tris-Cl (pH 8.0), 10 mmol UDP, 10mmol MgCl2, and a 50-�g/ml final concentration of enzyme. The reactionmixture was incubated at 37°C for 24 h. Ten microliters of reaction samplewas taken and analyzed by HPLC-PDA after proper dilution.

Analytical methods. For the thin-layer chromatography (TLC) anal-ysis of reaction mixtures, samples were loaded on a normal-phase silicaTLC plate and developed in an ethyl acetate-methanol-water-toluene (10:2:1.5:0.2 ml) mixture in a closed TLC chamber. After a complete run of thesamples, the plate was air dried and visualized under UV light (UV) at 254nm. The reverse-phase HPLC-PDA analysis was performed with a C18

column (YMC-Pack ODS-AQ; 4.6 mm internal diameter [i.d.], 150 mmlong, 5 �m particle size) connected to a PDA (308 nm) using binaryconditions of H2O (0.1% trifluoroacetic acid buffer) and 100% acetoni-trile (ACN) at a flow rate of 1 ml/min for 25 min. The ACN concentrationswere as follows: 20% (0 to 5 min), 50% (5 to 10 min), 70% (10 to 15 min),90% (15 to 20 min), and 10% (20 to 25 min). The purification of com-pounds was carried out by preparative HPLC with a C18 column (YMC-Pack ODS-AQ; 150 mm long, 20 mm i.d., 10 �m particle size) connectedto a UV detector (308 nm) using a 36-min binary program with ACN at20% (0 to 5 min), 40% (5 to 10 min), 40% (10 to 15 min), 90% (15 to 25min), 90% (25 to 30 min), and 10% (30 to 35 min) at a flow rate of 10ml/min. The high-resolution quantitative time of flight electrospray ion-ization mass spectrometry (HRQTOF–ESI-MS) was carried out in posi-tive ion mode on an Acquity mass spectrometer (with UPLC; Waters,Milford, MA, USA) coupled with a Synapt G2-S system (Waters). Thecompounds were further characterized with a 900-MHz Avance II 900Bruker (Germany) BioSpin NMR spectrometer equipped with a TCICryoProbe (5 mm). All the samples were prepared in methanol-D4

(Sigma-Aldrich). One-dimensional NMR (1H-NMR and 13C-NMR) andtwo-dimensional NMR (correlation spectroscopy [COSY], nuclear Over-hauser effect spectroscopy [NOESY], rotating-frame NOE spectroscopy[ROESY], heteronuclear single-quantum correlation [HSQC], and het-eronuclear multiple-bond correlation [HMBC]) were performed to elu-cidate the structures of the resveratrol glucosides. For 1H-NMR experi-ments, 16 transient spectra were acquired with a spectral width of9,014.423 Hz (10 ppm). For 13C-NMR experiments, 512 transient spectrawere acquired with a spectral width of 56,818.184 Hz (250 ppm). Thepulse sequences for each NMR were as follows: 1H-NMR, zg30 or zgpr(water suppression); 13C-NMR, zgpg30; COSY, cosygpqf; ROSEY,roseyesgp; HSQC, hsqcetgpsi; and HMBC, hmbcgplpndqf. All the rawdata were processed by using TopSpin 3.1 software (Bruker) and furtheranalyzed by using MestReNova 8.0 software (Mestrelab Research S. L.,Spain).

Molecular modeling and docking of resveratrol in YjiC. DiscoveryStudio v. 3.1 (2011) (DS 3.1) was used for homology-modeling and dock-

ing studies of YjiC protein (Accelrys, San Diego, CA, USA). PSI-BLAST(comparison matrix, BLOSUM 62; E threshold, 10) was carried out usingthe ExPASy (Expert Protein Analysis System; 2010) server (http://ca.expasy.org) to search the homologs of YjiC. A template, OleD (ProteinData Bank [PDB] ID 2IYA; crystal structure resolved at 1.7 Å) from Strep-tomyces antibioticus, was selected to build the model on the basis of highsimilarity and identity with YjiC. An approach similar to one previouslydescribed was followed to construct and validate the YjiC model (58). Theamino acid sequence of YjiC was aligned with that of OleD and used forthe model construction using the “build homology model” command andMODELLER (59), applying the Accelrys CHARMm force field through-out the process of modeling. The refined model was validated with ProSA2003 and Ramachandran Plot7 (58). Root mean square deviation(RMSD) analysis of the YjiC model was done using SUPERPOSE (60).Resveratrol was docked to the validated model of YjiC using the “Ligand-Fit” procedure implemented in Discovery Studio (61). Twenty-five dif-ferent conformations were generated for the selection of the best bindingmode of resveratrol in the YjiC protein (Carlo trails). The conformationsof each ligand were fitted in the binding cavity of YjiC with a grid-basedenergy function. The docked structures (diverse conformations) weresaved, and a scoring function was applied to each docked structure (bind-ing affinity prediction). All the conformations were scored using the Dis-covery Studio 3.1 scoring function, which includes Ligscore 1, Ligscore 2,Piecewise Linear Potential 1 (PLP1), PLP2, Jain, and Potential of MeanForce (PMF). Out of 25 conformations generated, the most suitable dock-ing mode for resveratrol with a high score from the consensus scoringfunction was finally selected.

RESULTSEnzymatic synthesis of resveratrol glucosides. The HPLC-PDAanalysis of the reaction mixture containing resveratrol, UDP-D-glucose, and the YjiC enzyme showed four different glucosidepeaks at 308 nm (Fig. 2). By analyzing the same reaction mixturewith an HPLC-PDA combined with HRQTOF–ESI-MS, the exactmass for each peak was determined. Peaks PI and PII had retentiontimes (tRs) of 13.32 min and 12.9 min, respectively. Both peakshad an exact mass [(PI or PII) � H�] of �391.1384 and repre-sented monoglucosides of resveratrol. The third peak (PIII) hada tR of 11.4 min with an exact mass, [PIII � Na�], of �575.1755and represented a resveratrol with two glucoses conjugated. Sim-ilarly, the fourth peak (PIV) had a tR of 10.1 min and an exact mass,[PIV � Na�], of �737.2266, representing a triglucoside of resvera-trol (see Fig. S1 and Tables S1 and S2 in the supplemental mate-rial). To further characterize the reaction products obtained, apreparative-scale reaction was carried out in a 10-ml reactionvolume. The products were purified as described in Materialsand Methods and were subjected to various NMR analyses, in-cluding 1H-NMR, 13C-NMR, and two-dimensional NMR analy-ses (COSY, NOESY, ROESY, HSQC, and HMBC) (see Fig. S2to S4 in the supplemental material). The PI product was iden-tified as (E)-1-(3-�-D-glucopyranosyloxy-5-hydroxyphenyl)-2-(4-hydroxyphenyl) ethene (common name, resveratrol 3-O-�-D-gluco-side) (1.81 mg; 465 �mol; �9.3%), which has previously beenidentified from different plant sources, as well as by enzymatic reac-tions (62, 63). The PII product was identified as (E)-1-(3,5-dihy-droxyphenyl)-2-(4-�-D-glucopyranosyl-oxyphenyl) ethene (com-mon name, resveratrol 4=-O-�-D-glucoside) (3.63 mg; 931.5�mol; �18.6%), as described previously (62, 63). According tothe mass analysis performed, PIII is hypothesized to represent adiglucoside of resveratrol identified as (E)-1-(3,5-bis-�-D-gluco-pyranosyloxyphenyl)-2-(4-hydroxyphenyl) ethane (common name,resveratrol 3,5-di-O-�-D-glucoside) (14.58 mg; 2639 �mol;

Glycosylation of Flavonoids

December 2014 Volume 80 Number 23 aem.asm.org 7237

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

�52.8%) (62–64). The triglucoside product, PIV, was identifiedas (E)-1-(3,5-bis-�-D-glucopyranosyloxyphenyl)-2-(4-�-D-glu-copyranosyl-oxyphenyl) ethene (common name, resveratrol3,5,4=-tri-O-�-D-glucoside) (3.28 mg; 460 �mol; �9.2%), and allavailable reactive hydroxyl groups of resveratrol were glucosylated(Fig. 2). The purity of each product was greater than 95%, and theoverall conversion rate of resveratrol to its glucoside forms was89.92%. In addition, YjiC exhibited nonregiospecific enzyme ac-tivity by conjugating glucose at all the available reactive hydroxylpositions of resveratrol. In our previous report, YjiC exhibitedhigh flexibility when phloretin was used as an acceptor molecule,producing five different glucosylated derivatives, including twomonoglucosides, two diglucosides, and a triglucoside. Recently,Zhou and colleagues reported the production of two monogluco-sides and two diglucosides of resveratrol using OleD and its vari-ants (ASP, AIP, TDP16, and 3–1-H12) (63). However, the authorsdid not report the detection of a triglucoside resveratrol.

Biosynthesis of diverse resveratrol glycosides. To character-ize the donor substrate specificity of YjiC for different NDP sugarsusing resveratrol as an acceptor substrate, NDP-D-sugars andNDP-L-sugars were used in glycosyltransferase assays. For theNDP-D-sugars, four different types of sugar donors were used.Specifically, C-2 position-modified sugars (TDP-D-2-deoxyglu-cose and UDP-N-acetylglucosamine), a C-4 position-modifiedsugar (UDP-D-galactose), a C-6 position-modified sugar (UDP-D-glucuronic acid), and a C-4 and C-6 position-modified sugar(TDP-D-viosamine) were used. Two different NDP-L-sugars,TDP-L-rhamnose and GDP-L-fucose, were also used to assay theactivity of YjiC with resveratrol under identical reaction condi-tions. As a preliminary analysis, TLC on normal-phase silica plates

(F254; Merck, Darmstadt, Germany) was performed. Extra spotswith a lower tR than that of unmodified resveratrol were distinctlyobserved in most of the reaction mixtures (see Fig. S5 in the sup-plemental material). All the reaction mixtures were further ana-lyzed by HPLC-PDA (Fig. 3), which detected the conjugation ofvarious sugars with resveratrol at 308 nm. HPLC-PDA andHRQTOF–ESI-MS analysis in positive mode finally confirmed theconjugation of various sugars with resveratrol (Fig. 3; see Fig. S6and Tables S1 and S2 in the supplemental material).

Because of the limited availability and difficulties in enzymaticsynthesis of various NDP sugars, all products were characterized byHPLC-PDA followed by HRQTOF–ESI-MS, with tRs for glycoran-domized products compared with those of the aforementioned glu-cosylated products of resveratrol that were structurally characterizedusing various NMR analyses (Fig. 3; see Table S1 in the supplementalmaterial). Both resveratrol 3-O-�-D-2-deoxyglucoside (tR, 13.8 min;[M � Na]� �397.1267) and resveratrol 3,5-O-�-D-di-2-deoxyglu-coside (tR, 13.1 min; [M � Na]� �543.1867) had longer tRs than therespective monoglucoside and diglucoside forms of resveratrol due todeoxy sugar conjugation. In contrast, resveratrol 4=-O-�-D-galacto-side (tR, 12.6 min; [M � Na]� �413.1229) and resveratrol 4=-O-�-D-viosaminoside (tR, 12.6 min; [M � Na]� �396.1412) exhibited tRssimilar to those of the corresponding resveratrol 4=-O-�-D-glucoside.However, no products were detected when UDP-D-glucuronicacid and UDP-N-acetylglucosamine were used as sugar donors.When the NDP-L-sugars, TDP-L-rhamnose and GDP-L-fucose, wereused for independent glycosyltransferase reactions, the monoglyco-sylated products, resveratrol 3-O-�-L-rhamnoside (tR, 13.8 min;[M � H]� �375.1462]) and resveratrol 3-O-�-L-fucoside (tR, 13.8min; [M � H]� �375.1446) were detected (see Fig. S6 and Table S2

FIG 2 HPLC-PDA analysis. The chromatogram presents the reaction products of resveratrol and UDP-�-D-glucose catalyzed by the YjiC enzyme. The insetprovides the full chromatogram, and the shaded region is magnified. Above each peak, the structure of the identified product is presented. Two resveratrolmonoglucosides (resveratrol 3-O-�-D-glucoside [PI] and resveratrol 4=-O-�-D-glucoside [PII]), a diglucoside (resveratrol 3,5-O-�-D-diglucoside [PIII]), and atriglucoside (resveratrol 3,5,4=-O-�-D-triglucoside [PIV]) were produced in the reaction mixture.

Pandey et al.

7238 aem.asm.org Applied and Environmental Microbiology

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

in the supplemental material). These products had tRs similar to thatof resveratrol 3-O-�-D-glucoside (see Table S1 in the supplementalmaterial).

The substrate conversion percentage, as well as the numberof products formed, varied with the NDP sugars used in thereaction mixtures. The conversion percentages of resveratrol to

resveratrol glycosides were determined at different time inter-vals in all reaction mixtures (Fig. 4). The highest conversion ofresveratrol to resveratrol glucosides was obtained with UDP-D-glucose (�90%), followed by TDP-D-viosamine (�49%),UDP-D-galactose (�45%), and TDP-2 deoxy-D-glucose (�45%).In comparison, the reactions that used TDP-L-rhamnose and

FIG 3 HPLC-PDA analyses of reaction mixtures of resveratrol and YjiC with different NDP sugar donor substrates. (A) Resveratrol standard; (B) TDP-2-deoxy-D-glucose; (C) UDP-D-galactose; (D) TDP-D-viosamine; (E) TDP-L-rhamnose; (F) GDP-L-fucose; (G) UDP-D-glucuronic acid; (H) UDP-N-acetylglucosamine.The open circles represent resveratrol standard, the filled circles are products, and the four-pointed stars represent unidentified products. The structures of theresveratrol glycoside derivatives identified are presented, along with HPLC chromatograms. The structures were identified by HPLC-PDA coupled withHRQTOF–ESI-MS analyses and by comparing the retention times with those of resveratrol glucoside analogues identified by NMR.

Glycosylation of Flavonoids

December 2014 Volume 80 Number 23 aem.asm.org 7239

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

GDP-L-fucose exhibited the lowest conversion (21.6% and 15.4%,respectively) of resveratrol. The reaction mixtures that used UDP-D-glucuronic acid and UDP-D-N-acetylglucosamine did not ex-hibit any conversion of resveratrol to its glycoside derivatives. Theconversion percentage of resveratrol to the respective resveratrolglycosides increased in all the reactions upon longer incubation(Fig. 4). However, an unexpected result was that the concentra-tion of resveratrol 3-O-�-D-glucoside was higher after a 1-h incu-bation period and lower concentrations were detected after 2 hand 3 h of incubation (data not shown). This decrease in theamount of product initially formed could be due to the instabilityof the product, or the product may be readily converted to res-veratrol and UDP-D-glucose in a reverse-glycosylation reactionmediated by YjiC as previously reported for phloretin glycosyla-tion (56). In addition, the product resveratrol 3-O-�-D-glucosidecould be converted to a resveratrol 3,5-O-�-D-diglucoside or res-veratrol 3,5,4=-O-�-D-triglucoside, as in the case of phloretin, byYjiC (56).

Deglucosylation of resveratrol glucoside. To further evaluatethe instability of resveratrol 3-O-�-D-glucoside in the reactionmixture, a deglucosylation reaction was performed to transfer aglucose molecule from resveratrol 3-O-�-D-glucoside to UDP toform UDP-D-glucose and aglycon resveratrol. HPLC-PDA wasused to analyze the reaction mixture at different time intervals.The concentration of resveratrol 3-O-�-D-glucoside was found todecrease by 20% (to 80% of the initial concentration) within 1 h,and it decreased to 20% of the initial concentration in 12 h at 37°C.Concomitantly, the concentration of resveratrol increased (Fig.5). After 12 h and 24 h, reaction mixture analyses revealed thesimultaneous formation of the products resveratrol 4=-O-�-D-glucoside and resveratrol 3,5-O-�-D-diglucoside, along with agly-

con resveratrol. UDP-D-glucose and resveratrol were also detectedafter deglucosylation of resveratrol 3-O-�-D-glucoside occurredand were utilized by YjiC to produce the newly formed products.However, resveratrol 3,5,4=-O-�-D-triglucoside was not detected,and this may be due to a lack of excess UDP-D-glucose in thereaction mixture. Taken together, these results indicate that YjiCcan deglucosylate resveratrol glucoside to produce UDP-D-glu-cose and resveratrol, and these products can eventually be utilizedby the same enzyme to produce other products (Fig. 5).

Docking of resveratrol in the YjiC model. OleD from S. anti-bioticus (PDB ID 2IYA) (65) was selected as a template to build theYjiC model using Accelrys Discovery Studio 3.1 software (AccelrysInc., San Diego, CA) based on the high similarity and identity ofthe amino acid sequences (see Fig. S7 in the supplemental mate-rial). The three-dimensional (3D) structure of the YjiC modelresembles the Rossmann-like domains belonging to the GT-B foldof GT1 family proteins (Fig. 6A). Following active-site optimiza-tion, molecular dynamics were used to dock resveratrol to the YjiCmodel. The study of all docked conformations of resveratrol in theYjiC model showed that resveratrol has two binding modes (headfirst versus tail first) (Fig. 6). In head-first mode, 3-OH and 5-OHgroups were in close proximity to UDP in a deep cleft formedbetween N-terminal (acceptor binding pocket) and C-terminal(donor binding pocket) domains. In the tail-first mode, the4=-OH group was near the UDP (Fig. 6B and C). Thus, there was apossibility of production of multiple glycosylated products of res-veratrol. This in silico docking result is consistent with our in vitroglucosylation reaction elucidating the production of four differentresveratrol glucoside derivatives while using UDP-D-glucose as aglucose donor substrate (Fig. 2), as well as the conjugation of othersugars at different positions of resveratrol by YjiC (Fig. 3). Similarbinding modes of resveratrol were also observed in docking stud-ies of resveratrol in OleD and its variants, producing two mono-glucosides and two diglucosides of resveratrol (63).

DISCUSSION

Besides a number of biological activities of resveratrol and its de-rivatives beneficial to health (12–14), recent evidence has alsodemonstrated an important role of resveratrol in the activation ofSIRT1, a protein that has been shown to lengthen the life spans

FIG 4 Conversion percentages of resveratrol to resveratrol glycosides in gly-cosylation reactions catalyzed by YjiC involving resveratrol as an acceptorsubstrate and different NDP sugars as donor substrates at different time inter-vals. The error bars indicate the standard deviation values for three indepen-dent experiments.

FIG 5 Deglucosylation of purified resveratrol 3-O-�-D-glucoside by YjiC inthe presence of UDP. When incubation times were increased from 12 h to 24 h,resveratrol glucoside derivatives (resveratrol 4=-O-�-D-glucoside and resvera-trol 3,5-O-�-D-diglucoside) were produced in the same reaction mixture,while resveratrol standard also accumulated. The error bars indicate the stan-dard deviation values for three independent experiments.

Pandey et al.

7240 aem.asm.org Applied and Environmental Microbiology

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

of yeast, worms, flies, fish, and mice (31, 66, 67). Upregulation ofSIRT1 has also been shown to reduce cellular oxidative stress inthe diabetic milieu (68). In this study, we attempted to expand thenumber of resveratrol derivatives by means of an enzymatic gly-cosylation approach that could help to generate novel therapeu-tics. Since the YjiC glycosyltransferase has been identified as ac-cepting diverse kinds of NDP glucoses, as well as NDP sugars, asdonor substrates, the use of the same enzyme has eased the pro-duction of diverse resveratrol glucosides and glycosides by in vitroglycosylation. We have recently reported the wide promiscuity ofthe enzyme, which raised the possibility of production of diverseflavonol glycosides. As resveratrol is one of the most studied plant-derived lead compounds and is under clinical trials for develop-ment as a drug for diverse diseases, generation of novel resveratrolglycosides could facilitate the development of drugs to treat oxi-dative-stress-induced diseases and to provide antiaging functions.

Resveratrol occurs in two forms, cis-resveratrol and trans-res-veratrol. The latter form of resveratrol changes to the cis formupon exposure to UV light. Importantly, the trans form of resvera-trol is biologically potent and more abundantly present in naturethan the cis form (69–71). The glycosylation of resveratrol couldmake it more stable, since most studies have proven that glycosy-lation enhances the stability of the compound (72–74). Further-more, glycosylation increases its solubility in water, which ulti-mately enhances the bioavailability of the compound for diversemedical and cosmetic applications. Importantly, the conjugationof resveratrol with diverse sugars could modify its biological po-tency, as well.

The sugar moieties of most NPs have been found to be involvedin cell-cell recognition, stable binding of the compound at thetarget site, and other biochemical processes (42). Most of theplant-derived glycosides contain sugars, like D-glucose, D-galac-tose, D-xylose, D-arabinose, L-rhamnose, and L-fucose. However,most of the microbe-originated therapeutics were found to beconjugated with deoxy sugars and amino sugars. Thus, we useddiverse NDP sugars, including deoxy sugars (TDP-D-2-deoxyglu-cose, TDP-L-rhamnose, and GDP-L-fucose) and amino sugars(TDP-D-viosamine and UDP-D-N-acetylglucosamine) for theproduction of nonnatural sugars bearing resveratrol glycosides,along with glucosides and galactosides. We successfully produced10 diverse resveratrol glycosides—resveratrol 3-O-�-D-glucoside,resveratrol 4=-O-�-D-glucoside, resveratrol 3,5-O-�-D-digluco-side, resveratrol 3,5,4=-O-�-D-triglucoside, resveratrol 3-O-�-D-2-

deoxyglucoside, resveratrol 3,5-O-�-D-di-2-deoxyglucoside, res-veratrol 4=-O-�-D-galactoside, resveratrol 4=-O-�-D-viosaminoside,resveratrol 3-O-�-L-rhamnoside, and resveratrol 3-O-�-L-fuco-side—by in vitro glycosylation reactions. Most of the resveratrol de-rivatives produced are novel compounds, except for two monoglu-cosides and a diglucoside, which were isolated from different plantsources. The diverse resveratrol glycosides could exhibit different bio-activities in comparison to the parent lead compound.

In general, the secondary metabolites produced by plants andactinomycetes are unstable, toxic, hydrophobic, or volatile and arepostmodified to glycosylated forms by GTs (75). Moreover, theseglycosylated compounds have been found to exhibit increased sta-bility and solubility in water while also being nontoxic. Currently,the relevance of metabolic stability following deglycosylation ofNP-glycosides remains unknown. Thus, the formation of resvera-trol 4=-O-�-D-glucoside and resveratrol 3,5-O-�-D-diglucosidefollowed by deglucosylation of resveratrol 3-O-�-D-glucoside is asyet unknown, but this could be due to the higher thermodynamicstability of resveratrol 4=-O-�-D-glucoside and resveratrol 3,5-O-�-D-diglucoside than of resveratrol 3-O-�-D-glucoside. TheHPLC-PDA analysis of the resveratrol glucosylation reaction(Fig. 2) also detected a greater abundance of resveratrol 4=-O-�-D-glucoside and resveratrol 3,5-O-�-D-diglucoside than ofresveratrol 3-O-�-D-glucoside. However, it is also possible thatresveratrol 3-O-�-D-glucoside is directly converted to themore stable resveratrol 3,5-O-�-D-diglucoside following se-quential glycosylation of resveratrol 3-O-�-D-glucoside at theC-5 hydroxyl position.

The molecular modeling and docking studies showed that res-veratrol has two binding modes (head first versus tail first) withYjiC, similar to OleD. The dual binding modes of resveratrol havesupported the possibility of production of multiple glucosylatedproducts. Thus, glycosylation can occur at the 3-, 5-, and 4=-hy-droxyl positions of resveratrol. Our in vitro results, which in-cluded the detection of diglucoside and triglucoside forms of res-veratrol, are consistent with this model. Detailed studies ofprotein structure, as well as mutagenesis studies, are needed tofurther elucidate the mechanistic details of adopting various sug-ars in the donor substrate binding pocket, as well as harboring thesubstrate in dual conformations by YjiC, which could be a keyfollow-up study to explore the substrate promiscuity of the en-zyme. Though the biological activities of the newly synthesizedresveratrol glucosides and various glycoside analogues in this

FIG 6 Molecular modeling of YjiC and docking of resveratrol in the YjiC model. (A) Superimposed ribbon diagram of the YjiC model (light brown) with thecrystal structure of OleD (yellow). The ribbon diagram of the 3D structure of YjiC resembles the Rossmann-like domains belonging to the GT-B fold of GT1family proteins. The C-terminal domains of the enzyme, like those of other family 1 glycosyltransferases, lodge nucleotide sugar donors, whereas the N-terminaldomain accommodates acceptors. (B) Docking of resveratrol in head-first mode. (C) Docking of resveratrol in tail-first mode.

Glycosylation of Flavonoids

December 2014 Volume 80 Number 23 aem.asm.org 7241

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

study are still unknown, they could be beneficial for the generationof novel therapeutics.

ACKNOWLEDGMENTS

This study was supported by a grant from the Next-Generation BioGreen21 Program (J.K.S. [PJ0094832] and Y.I.P. [PJ0094835]), by the RuralDevelopment Administration, and by the Republic of Korea.

REFERENCES1. Roldán A, Palacios V, Caro I, Pérez L. 2003. Resveratrol content of

Palomino fino grapes: influence of vintage and fungal infection. J. Agric.Food Chem. 51:1464 –1468. http://dx.doi.org/10.1021/jf020774u.

2. Sanders TH, McMichael JRW, Hendrix KW. 2000. Occurrence of res-veratrol in edible peanuts. J. Agric. Food Chem. 48:1243–1246. http://dx.doi.org/10.1021/jf990737b.

3. Lyons MM, Yu C, Toma RB, Cho SY, Reiboldt W, Lee J, Breemen RB.2003. Resveratrol in raw and baked blueberries and bilberries. J. Agric.Food Chem. 51:5867–5870. http://dx.doi.org/10.1021/jf034150f.

4. Jerkovic V, Collin S. 2007. Occurrence of resveratrol and piceid in Amer-ican and European hop cones. J. Agric. Food Chem. 55:8754 – 8758. http://dx.doi.org/10.1021/jf071792k.

5. Langcake P, Pryce RJ. 1976. The production of resveratrol by Vitis vinif-era and other members of the Vitaceae as a response to infection or injury.Physiol. Plant Pathol. 9:77– 86. http://dx.doi.org/10.1016/0048-4059(76)90077-1.

6. Roggero JP, Garciaparrilla C. 1995. Effects of ultraviolet irradiation onresveratrol and changes in resveratrol and various of its derivatives in theskins of ripening grapes. Sci. Aliments 15:411– 422.

7. He S, Jiang L, Wu B, Pan Y, Sun C. 2009. Pallidol, a resveratrol dimer fromred wine, is a selective singlet oxygen quencher. Biochem. Biophys. Res. Com-mun. 379:283–287. http://dx.doi.org/10.1016/j.bbrc.2008.12.039.

8. Ito T, Abe N, Oyama M, Tanaka T, Murata J, Darnaedi D, Linuma M.2009. Two novel resveratrol trimers from Dipterocarpus grandiflorus. Helv.Chim. Acta 92:1203–1216. http://dx.doi.org/10.1002/hlca.200800429.

9. Soleas GJ, Diamandis EP, Goldberg DM. 1997. Resveratrol: a moleculewhose time has come? And gone? Clin. Biochem. 30:91–113. http://dx.doi.org/10.1016/S0009-9120(96)00155-5.

10. Orallo F. 2006. Biological effects of cis- versus trans-resveratrol, p 577–600. In Aggarwal BB, Shishodia S (ed), Resveratrol in health and disease.CRC Press, Boca Raton, FL.

11. Orallo F. 2006. Comparative studies of the antioxidant effects of cis- andtrans-resveratrol. Curr. Med. Chem. 13:87–98. http://dx.doi.org/10.2174/092986706775197962.

12. Anisimova NY, Kiselevsky MV, Sosnov AV, Sadovnikov SV, StankovIN, Gakh AA. 2011. Trans-, cis-, and dihydro-resveratrol: a comparativestudy. Chem. Cent. J. 5:88. http://dx.doi.org/10.1186/1752-153X-5-88.

13. King RE, Bomser JA, Min DB. 2006. Bioactivity of resveratrol. Compr.Rev. Food Sci. F 5:493–506. http://dx.doi.org/10.1111/j.1541-4337.2006.00001.x.

14. Stuart JA, Robb EL. 2013. Bioactive polyphenols from wine grapes.Springer, New York, NY.

15. Szkudelski T. 2006. Resveratrol inhibits insulin secretion from rat pan-creatic islets. Eur. J. Pharmacol. 552:176 –181. http://dx.doi.org/10.1016/j.ejphar.2006.09.046.

16. Sun C, Zhang F, Ge X, Yan T, Chen X, Shi X, Zhai Q. 2007. SIRT1improves insulin sensitivity under insulin-resistant conditions by repress-ing PTP1B. Cell Metab. 6:307–319. http://dx.doi.org/10.1016/j.cmet.2007.08.014.

17. Baur JA, Pearson KJ, Price NL, Jamieson HA, Lerin C, Kalra A, PrabhuVV, Allard JS, Lopez-Lluch G, Lewis K, Pistell PJ, Poosala S, Becker KG,Boss O, Gwinn D, Wang M, Ramaswamy S, Fishbein KW, Spencer RG,Lakatta EG, Le Couteur D, Shaw RJ, Navas P, Puigserver P, Ingram DK,de Cabo R, Sinclair DA. 2006. Resveratrol improves health and survivalof mice on a high-calorie diet. Nature 444:337–342. http://dx.doi.org/10.1038/nature05354.

18. Wang MJ, Huang HM, Hsieh SJ, Jeng KC, Kuo JS. 2001. Resveratrolinhibits interleukin-6 production in cortical mixed glial cells under hy-poxia/hypoglycemia followed by reoxygenation. J. Neuroimmunol. 112:28 –34. http://dx.doi.org/10.1016/S0165-5728(00)00374-X.

19. Frémont L, Belguendouz L, Delpal S. 1999. Antioxidant activity ofresveratrol and alcohol-free wine polyphenols related to LDL oxidation

and polyunsaturated fatty acids. Life Sci. 64:2511–2521. http://dx.doi.org/10.1016/S0024-3205(99)00209-X.

20. Pezzuto JM. 2008. Grapes and human health: a perspective. J. Agric. FoodChem. 56:6777– 6784. http://dx.doi.org/10.1021/jf800898p.

21. Mizutani K, Ikeda K, Kawai Y, Yamori Y. 2000. Resveratrol attenuatesovariectomy-induced hypertension and bone loss in stroke-prone spon-taneously hypertensive rats. J. Nutr. Sci. Vitaminol. 46:78 – 83. http://dx.doi.org/10.3177/jnsv.46.78.

22. Bhatt SR, Lokhandwala MF, Banday AA. 2011. Resveratrol preventsendothelial nitric oxide synthase uncoupling and attenuates developmentof hypertension in spontaneously hypertensive rats. Eur. J. Pharmacol.667:258 –264. http://dx.doi.org/10.1016/j.ejphar.2011.05.026.

23. Petrovski G, Gurusamy N, Das DK. 2011. Resveratrol in cardiovascularhealth and disease. Ann. N. Y. Acad. Sci. 1215:22–33. http://dx.doi.org/10.1111/j.1749-6632.2010.05843.x.

24. Sato M, Ray PS, Maulik G, Maulik N, Engelman RM, Bertelli AA,Bertelli A, Das DK. 2000. Myocardial protection with red wine extract. J.Cardiovasc. Pharmacol. 35:263–268. http://dx.doi.org/10.1097/00005344-200002000-00013.

25. Gorbunov N, Petrovski G, Gurusamy N, Ray D, Kim Do H, Das DK.2012. Regeneration of infarcted myocardium with resveratrol-modifiedcardiac stem cells. J. Cell. Mol. Med. 16:174 –184. http://dx.doi.org/10.1111/j.1582-4934.2011.01281.x.

26. Jang M, Cai L, Udeani GO, Slowing KV, Thomas CF, Beecher CW,Fong HH, Farnsworth NR, Kinghorn AD, Mehta RG, Moon RC,Pezzuto JM. 1997. Cancer chemopreventive activity of resveratrol, a nat-ural product derived from grapes. Science 275:218 –220. http://dx.doi.org/10.1126/science.275.5297.218.

27. Pangeni R, Sahni JK, Ali J, Sharma S, Baboota S. 2014. Resveratrol:review on therapeutic potential and recent advances in drug delivery.Expert Opin. Drug Deliv. 11:1285–1298. http://dx.doi.org/10.1517/17425247.2014.919253.

28. Tsai SH, Lin-Shiau SY, Lin JK. 1999. Suppression of nitric oxide synthaseand the down-regulation of the activation of NF�B in macrophages byresveratrol. Br. J. Pharmacol. 126:673– 680. http://dx.doi.org/10.1038/sj.bjp.0702357.

29. Wadsworth TL, Koop DR. 1999. Effects of the wine polyphenolics quer-cetin and resveratrol on pro-inflammatory cytokine expression in RAW264.7 macrophages. Biochem. Pharmacol. 57:941–949. http://dx.doi.org/10.1016/S0006-2952(99)00002-7.

30. Feng YH, Zhou WL, Wu QL, Li XY, Zhao WM, Zou JP. 2002. Low doseof resveratrol enhanced immune response of mice. Acta Pharmacol. Sin.23:893– 897.

31. Knutson MD, Leeuwenburgh C. 2008. Resveratrol and novel potentactivators of SIRT1: effects on aging and age-related diseases. Nutr. Rev.66:591–596. http://dx.doi.org/10.1111/j.1753-4887.2008.00109.x.

32. Park SJ, Ahmad F, Philp A, Baar K, Williams T, Luo H, Ke H, RehmannH, Taussig R, Brown AL, Kim MK, Beaven MA, Burgin AB, Mangan-iello V, Chung JH. 2012. Resveratrol ameliorates aging-related metabolicphenotypes by inhibiting cAMP phosphodiesterases. Cell 148:421– 433.http://dx.doi.org/10.1016/j.cell.2012.01.017.

33. Robb EL, Page MM, Wiens BE, Stuart JA. 2008. Molecular mechanismsof oxidative stress resistance induced by resveratrol: specific and progres-sive induction of MnSOD. Biochem. Biophys. Res. Commun. 367:406 –412. http://dx.doi.org/10.1016/j.bbrc.2007.12.138.

34. Gatz SA, Wiesmüller L. 2008. Take a break—resveratrol in action on DNA.Carcinogenesis 29:321–332. http://dx.doi.org/10.1093/carcin/bgm276.

35. Patel KR, Scott E, Brown VA, Gescher AJ, Steward WP, Brown K. 2011.Clinical trials of resveratrol. Ann. N. Y. Acad. Sci. 1215:161–169. http://dx.doi.org/10.1111/j.1749-6632.2010.05853.x.

36. Zein N, Sinha AM, McGahren WJ, Ellestad GA. 1988. Calicheamicingamma 1I: an antitumor antibiotic that cleaves double-stranded DNA sitespecifically. Science 240:1198 –1201. http://dx.doi.org/10.1126/science.3240341.

37. Tan C, Tasaka H, Yu KP, Murphy ML, Karnofsky DA. 1967. Dauno-mycin, an antitumor antibiotic, in the treatment of neoplastic disease.Clinical evaluation with special reference to childhood leukemia. Cancer20:333–353.

38. Schatz A, Bugie E, Waksman SA. 1944. Streptomycin, a substance ex-hibiting antibiotic activity against Gram-positive and Gram-negative bac-teria. Exp. Biol. Med. 55:66 – 69. http://dx.doi.org/10.3181/00379727-55-14461.

Pandey et al.

7242 aem.asm.org Applied and Environmental Microbiology

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

39. Griffith RS. 1981. Session I: history, bacteriology, and animal toxicity:introduction to vancomycin. Clin. Infect. Dis. 3:S200 –S204.

40. Radford DJ, Gillies AD, Hinds JA, Duffy P. 1986. Naturally occurringcardiac glycosides. Med. J. Aust. 144:540 –544.

41. Abu-Salah KM. 1996. Amphotericin B: an update. Br. J. Biomed. Sci.53:122–133.

42. Weymouth-Wilson AC. 1997. The role of carbohydrates in biologicallyactive natural products. Nat. Prod. Rep. 14:99 –110. http://dx.doi.org/10.1039/np9971400099.

43. Thorson JS, Barton WA, Hoffmeister D, Albermann C, Nikolov DB.2004. Structure-based enzyme engineering and its impact on in vitroglycorandomization. Chembiochem 5:16 –25. http://dx.doi.org/10.1002/cbic.200300620.

44. Langenhan JM, Peters NR, Guzei IA, Hoffmann FM, Thorson JS. 2005.Enhancing the anticancer properties of cardiac glycosides by neoglycoran-domization. Proc. Natl. Acad. Sci. U. S. A. 102:12305–12310. http://dx.doi.org/10.1073/pnas.0503270102.

45. Albermann C, Soriano A, Jiang J, Vollmer H, Biggins JB, Barton WA,Lesniak J, Nikolov DB, Thorson JS. 2003. Substrate specificity of NovM:implications for novobiocin biosynthesis and glycorandomization. Org.Lett. 5:933–936. http://dx.doi.org/10.1021/ol0341086.

46. Fu X, Albermann C, Zhang C, Thorson JS. 2005. Diversifying vanco-mycin via chemoenzymatic strategies. Org. Lett. 7:1513–1515. http://dx.doi.org/10.1021/ol0501626.

47. Ahmed A, Peters NR, Fitzgerald MK, Watson JA, Jr, Hoffmann FM,Thorson JS. 2006. Colchicine glycorandomization influences cytotoxicityand mechanism of action. J. Am. Chem. Soc. 128:14224 –14225. http://dx.doi.org/10.1021/ja064686s.

48. Williams GJ, Yang J, Zhang C, Thorson JS. 2011. Recombinant E. coliprototype strains for in vivo glycorandomization. ACS Chem. Biol. 6:95–100. http://dx.doi.org/10.1021/cb100267k.

49. Gantt RW, Peltier-Pain P, Cournoyer WJ, Thorson JS. 2011. Usingsimple donors to drive the equilibria of glycosyltransferase-catalyzed re-actions. Nat. Chem. Biol. 7:685– 691. http://dx.doi.org/10.1038/nchembio.638.

50. Gantt RW, Peltier-Pain P, Singh S, Zhou M, Thorson JS. 2013. Broad-ening the scope of glycosyltransferase-catalyzed sugar nucleotide synthe-sis. Proc. Natl. Acad. Sci. U. S. A. 110:7648 –7653. http://dx.doi.org/10.1073/pnas.1220220110.

51. Park SH, Park HY, Sohng JK, Lee HC, Liou K, Yoon YJ, Kim BG. 2009.Expanding substrate specificity of GT-B fold glycosyltransferase via do-main swapping and high-throughput screening. Biotechnol. Bioeng. 102:988 –994. http://dx.doi.org/10.1002/bit.22150.

52. Wu CZ, Jang JH, Woo M, Ahn JS, Kim JS, Hong YS. 2012. Enzymaticglycosylation of nonbenzoquinone geldanamycin analogs via BacillusUDP-glycosyltransferase. Appl. Environ. Microbiol. 78:7680 –7686. http://dx.doi.org/10.1128/AEM.02004-12.

53. Parajuli P, Pandey RP, Koirala N, Yoon YJ, Kim BG, Sohng JK. 2014.Enzymatic synthesis of epothilone A glycosides. AMB Express 4:31. http://dx.doi.org/10.1186/s13568-014-0031-1.

54. Pandey RP, Gurung RB, Parajuli P, Koirala N, Tuoi Le T, Sohng JK.2014. Assessing acceptor substrate promiscuity of YjiC-mediated glycosy-lation toward flavonoids. Carbohydr. Res. 393:26 –31. http://dx.doi.org/10.1016/j.carres.2014.03.011.

55. Pandey RP, Parajuli P, Koirala N, Park JW, Sohng JK. 2013. Probing3-hydroxyflavone for in vitro glycorandomization of flavonols by YjiC.Appl. Environ. Microbiol. 79:6833– 6838. http://dx.doi.org/10.1128/AEM.02057-13.

56. Pandey RP, Li TF, Kim EH, Yamaguchi T, Park YI, Kim JS, Sohng JK.2013. Enzymatic synthesis of novel phloretin glucosides. Appl. Environ.Microbiol. 79:3516 –3521. http://dx.doi.org/10.1128/AEM.00409-13.

57. Pandey RP, Parajuli P, Koirala N, Lee JH, Park YI, Sohng JK. 2014.Glucosylation of isoflavonoids in engineered Escherichia coli. Mol. Cells37:172–177. http://dx.doi.org/10.14348/molcells.2014.2348.

58. Bhattarai S, Liou K, Oh TJ. 2012. Homology modeling and docking

studies of Streptomyces peucetius CYP147F1 as limonene hydroxylase. J.Microbiol. Biotechnol. 22:917–922. http://dx.doi.org/10.4014/jmb.1112.12053.

59. Sali A, Potterton L, Yuan F, van Vlijmen H, Karplus M. 1995. Evalua-tion of comparative protein modeling by MODELLER. Proteins 23:318 –326. http://dx.doi.org/10.1002/prot.340230306.

60. Maiti R, Van Domselaar GH, Zhang H, Wishart DS. 2004. SuperPose:a simple server for sophisticated structural superposition. Nucleic AcidsRes. 32:W590 –W594. http://dx.doi.org/10.1093/nar/gkh477.

61. Venkatachalam CM, Jiang X, Oldfield T, Waldman M. 2003. LigandFit:a novel method for the shape-directed rapid docking of ligands to proteinactive sites. J. Mol. Graph. Model. 21:289 –307. http://dx.doi.org/10.1016/S1093-3263(02)00164-X.

62. Zhang Z, Yu B, Schmidt RR. 2006. Synthesis of mono- and di-O-�-D-glucopyranoside conjugates of (E)-resveratrol. Synthesis 8:1301–1306.

63. Zhou M, Hamza A, Zhan CG, Thorson JS. 2013. Assessing the regiose-lectivity of OleD-catalyzed glycosylation with a diverse set of acceptors. J.Nat. Prod. 76:279 –286. http://dx.doi.org/10.1021/np300890h.

64. Larronde F, Richard T, Delaunay JC, Decendit A, Monti JP, Krisa S,Mérillon JM. 2005. New stilbenoid glucosides isolated from Vitis viniferacell suspension cultures (cv. Cabernet Sauvignon). Planta Med. 71:888 –890. http://dx.doi.org/10.1055/s-2005-871294.

65. Bolam DN, Roberts S, Proctor MR, Turkenburg JP, Dodson EJ, Mar-tinez-Fleites C, Yang M, Davis BG, Davies GJ, Gilbert HJ. 2007. Thecrystal structure of two macrolide glycosyltransferases provides a blue-print for host cell antibiotic immunity. Proc. Natl. Acad. Sci. U. S. A.104:5336 –5341. http://dx.doi.org/10.1073/pnas.0607897104.

66. Quideau S, Deffieux D, Pouységu L. 2012. Resveratrol still has somethingto say about aging! Angew. Chem. Int. Ed. Engl. 51:6824 – 6826. http://dx.doi.org/10.1002/anie.201203059.

67. Hubbard BP, Gomes AP, Dai H, Li J, Case AW, Considine T, Riera TV,Lee JE, E SY, Lamming DW, Pentelute BL, Schuman ER, Stevens LA,Ling AJ, Armour SM, Michan S, Zhao H, Jiang Y, Sweitzer SM, BlumCA, Disch JS, Ng PY, Howitz KT, Rolo AP, Hamuro Y, Moss J, PerniRB, Ellis JL, Vlasuk GP, Sinclair DA. 2013. Evidence for a commonmechanism of SIRT1 regulation by allosteric activators. Science 339:1216 –1219. http://dx.doi.org/10.1126/science.1231097.

68. Yun JM, Chien A, Jialal I, Devaraj S. 2012. Resveratrol up-regulatesSIRT1 and inhibits cellular oxidative stress in the diabetic milieu: mecha-nistic insights. J. Nutr. Biochem. 23:699 –705. http://dx.doi.org/10.1016/j.jnutbio.2011.03.012.

69. Athar M, Back JH, Tang X, Kim KH, Kopelovich L, Bickers DR, KimAL. 2007. Resveratrol: a review of preclinical studies for human cancerprevention. Toxicol. Appl. Pharmacol. 224:274 –283. http://dx.doi.org/10.1016/j.taap.2006.12.025.

70. Wenzel E, Soldo T, Erbersdobler H, Somoza V. 2005. Bioactivity andmetabolism of trans-resveratrol orally administered to Wistar rats. Mol.Nutr. Food Res. 49:482– 494. http://dx.doi.org/10.1002/mnfr.200500003.

71. Burns J, Yokota T, Ashihara H, Lean ME, Crozier A. 2002. Plant foodsand herbal sources of resveratrol. J. Agric. Food Chem. 50:3337–3340.http://dx.doi.org/10.1021/jf0112973.

72. Jones P, Vogt T. 2001. Glycosyltransferases in secondary plant metabo-lism: tranquilizers and stimulant controllers. Planta 213:164 –174. http://dx.doi.org/10.1007/s004250000492.

73. Gachon CM, Langlois-Meurinne M, Saindrenan P. 2005. Plant second-ary metabolism glycosyltransferases: the emerging functional analysis.Trends Plant Sci. 10:542–549. http://dx.doi.org/10.1016/j.tplants.2005.09.007.

74. Bowles D, Isayenkova J, Lim E, Poppenberger B. 2005. Glycosyltrans-ferases: managers of small molecules. Curr. Opin. Plant Biol. 8:254 –263.http://dx.doi.org/10.1016/j.pbi.2005.03.007.

75. Thorson JS, Vogt T. 2005. Glycosylated natural products, p 685–710. InWong CH (ed), Carbohydrate-based drug discovery. Wiley-VCH, Wein-heim, Germany.

Glycosylation of Flavonoids

December 2014 Volume 80 Number 23 aem.asm.org 7243

on April 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from