molecules

Article The Functional Characterization of a Site-Specific 40-O-methyltransferase Synthesized by the Liverwort Species Plagiochasma appendiculatum

Hui Liu, Rui-Xue Xu, Shuai Gao and Ai-Xia Cheng *

Key Laboratory of Chemical Biology of Natural Products, Ministry of Education, School of Pharmaceutical Sciences, Shandong University, Jinan 250012, China; [email protected] (H.L.); [email protected] (R.-X.X.); [email protected] (S.G.) * Correspondence: [email protected]; Tel.: +86-531-8838-2012; Fax: +86-531-8838-2019

Academic Editors: Qing-Wen Zhang and Chuangchuang Li Received: 5 April 2017; Accepted: 4 May 2017; Published: 7 May 2017

Abstract: Apigenin, a widely distributed flavone, exhibits excellent antioxidant, anti-inflammatory, and antitumor properties. In addition, the methylation of apigenin is generally considered to result in better absorption and greatly increased bioavailability. Here, four putative Class II methyltransferase genes were identified from the transcriptome sequences generated from the liverwort species Plagiochasma appendiculatum. Each was heterologously expressed as a His-fusion protein in Escherichia coli and their methylation activity against apigenin was tested. One of the four Class II OMT enzymes named 40-O-methyltransferase (Pa40OMT) was shown to react effectively with apigenin, catalyzing its conversion to . Besides the favorite substrate apigenin, the recombinant PaF40OMT was shown to catalyze , naringenin, , , genistein, , and genkwanin to the corresponding 40-methylation products. In vivo feeding experiments indicated that PaF40OMT could convert apigenin to acacetin efficiently in E. coli and approximately 88.8 µM (25.2 mg/L) of product was synthesized when 100 µM of apigenin was supplemented. This is the first time that a Class II plant O-methyltransferase has been characterized in liverworts.

Keywords: Plagiochasma appendiculatum; Class II O-methyltransferase; apigenin; acacetin; feeding

1. Introduction Flavonoids are a family composed of various polyphenolic compounds synthesized by plant species. Among this family, apigenin (4,5,7-trihydroxyflavone) is a widely distributed flavone and exhibits antioxidant, anti-inflammatory, and antitumor properties [1–4]. It was found that methylation of the free hydroxy groups in the flavonoids dramatically increases their stability and enhances the membrane transport, resulting in better absorption and greatly increased bioavailability [5,6]. Acacetin, 40-methylated form of apigenin, has been proven to have broad-spectrum biological activities and has gradually been used in the treatment of multifarious diseases. A previous study revealed that acacetin is a promising atrium-selective agent for the treatment of anti-atrial fibrillation [7] and has been well documented to show novel pharmacological properties such as antinociceptive, anti-inflammatory, and anti-cancer activities [8–10]. Enzyme-catalyzed synthesis of acacetin can be achieved by the para-O-methylation of apigenin using S-adenosyl-L-methionine as a methyl donor catalyzed by O-methyltransferases (OMTs). O-methyltransferases are known to be involved in the reaction transforming methyl to hydroxy groups of phenylpropanoids and flavonoids. Based on their size, amino acid sequence, and cation dependency, they have been classified into two major groups, referred to as Classes I and II. Class IOMTs (caffeoyl-coenzyme A O-methyltransferases; CCoAOMT)

Molecules 2017, 22, 759; doi:10.3390/molecules22050759 www.mdpi.com/journal/molecules Molecules 2017, 22, 759 2 of 11 Molecules 2017, 22, 759 2 of 11 KDa methylate the lignin precursors caffeoyl-coenzyme A (CCoA) and 5-hydroxyferuloyl CoA in the presence of an Mg2+ cation premise. Substrates preferred by CCoAOMT possess two vicinal hydroxylwith a molecular groups weightand their of 26–29 methylation KDa methylate affects thethe ligninmeta- precursorshydroxyl group. caffeoyl-coenzyme COMTs (caffeic A (CCoA) acid 2+ Oand-methyltransferases) 5-hydroxyferuloyl are CoA cation-independent in the presence enzymes of an Mg capablecation of methylating premise. Substratesnot only flavonoids, preferred but by - alsoCCoAOMT both 3-hydroxyl-and possess two vicinal 5-hydroxyl-containing hydroxyl groups phenylpropanoid-derived and their methylation affects lignin the precursors meta hydroxyl [11]. group.Class O IICOMTs OMTs’ (caffeic molecular acid weights-methyltransferases) range from are38 cation-independentto 43 KDa, and their enzymes activity capable does of not methylating require the not presenceonly flavonoids, of a metal but alsocation. both Beside 3-hydroxyl-ands the typical 5-hydroxyl-containing COMT activity of catalyzing phenylpropanoid-derived the methylation of lignin the ligninprecursors precursors [11]. Class caffeic II OMTs’ acid and molecular 5-hydroxy weights coniferic range acid, from some 38 to 43Class KDa, II andOMTs their also activity play a does role not in therequire methylation the presence of flavonoids. of a metal cation. For instance, Besides thethe typicalArabidopsis COMT thaliana activityenzyme of catalyzing AtCOMT1 the methylation is able to methylateof the lignin the precursors flavonol quercetin, caffeic acid which and 5-hydroxy bears an orthodihydroxy coniferic acid, some group, Class while II OMTs the Populus also play deltoides a role enzymein the methylation POMT-7 transfers of flavonoids. a methyl For instance,group to the flavonoidsArabidopsis (apigenin, thalianaenzyme kaempferol, AtCOMT1 luteolin, is able and to quercetin)methylate to the the flavonol C-7 OH quercetin, group [12]. which bears an orthodihydroxy group, while the Populus deltoides enzymeIn recent POMT-7 years, transfers metabolic a methyl engineering group to containing flavonoids OMT (apigenin, genes kaempferol, has been one luteolin, of theand hottest quercetin) fields into theterms C-7 of OH use group for the [12 ].production of valuable medical compounds, especially for polyhydroxyl molecules.In recent O-methylation years, metabolic of a engineering specific OH containing group th OMTrough genes chemical has been synthesis one of theconfronts hottest many fields drawbacks:in terms of harsh use for catalytic the production conditions, of valuablea long reac medicaltion time, compounds, protection especially of unwanted for polyhydroxyl groups, and expensivemolecules. methylationO-methylation reagents. of a For specific the production OH group of through acacetin, chemical direct C-4 synthesis′ methylation confronts of apigenin many isdrawbacks: impossible harsh by chemical catalytic means, conditions, due ato long the reactionhigher acidity time, protectionof C-7 hydroxy of unwanted group. groups,It has been and 0 synthesizedexpensive methylation from largely reagents. available For naringin the production via dehydrogenation of acacetin, direct (to C-4 )methylation and hydrolysis of apigenin of is glycosidicimpossible side by chemical chain [13,14]. means, However, due to the microbial higher acidity biotransformation of C-7 hydroxy is group. an alternative It has been strategy synthesized with greatfrom largelypotential available to produce naringin novel via bioactive dehydrogenation compounds (to in rhoifolin) a one-step and reaction hydrolysis and of a glycosidic much milder side reactionchain [13 condition.,14]. However, Microbes microbial have biotransformation a highly tractable is genetic an alternative system strategyand favorable with great fermentation potential conditionsto produce [15]. novel Several bioactive studies compounds have shown in a that one-step high reactionproductivities and a of much O-methylation milder reaction for flavonoids condition. areMicrobes viable. have SaOMT-2, a highly isolated tractable from genetic Streptomyces system and avermitilis favorable fermentationMA-4680, converted conditions naringenin [15]. Several to sakuranetinstudies have in shown E. coli, thatwhich high has productivities an antifungal ofactivityO-methylation against Magnaporthe for flavonoids grisea are [16]. viable. SaOMT-2, isolatedLiverworts from Streptomyces produce methylated avermitilis MA-4680,flavonoids, converted lignans, lignin-like naringenin compound to sakuranetins, and in bis-bibenzyls.E. coli, which Here,has an a antifungaldescription activity is given against of theMagnaporthe isolation of griseafour Plagiochasma[16]. appendiculatum Class II OMTs, one of whichLiverworts has been produce functionally methylated characterized flavonoids, as lignans,a flavone lignin-like 4′-O-methyltransferase compounds, and (Pa4 bis-bibenzyls.′OMT) with apigeninHere, a description as the favorite is given substrate of the isolation(Figure of1). fourThePlagiochasma site-specific appendiculatumbiotransformationClass of II apigenin OMTs, one to 0 0 acacetinof which in has engineered been functionally E. coli expressing characterized PaF4′ asOMT a flavone was explored 4 -O-methyltransferase and approximately (Pa4 25.2OMT) mg/L with of acacetinapigenin was as the synthesized. favorite substrate This is (Figure the first1). Thetime site-specific that a Class biotransformation II plant O-methyltransfe of apigeninrase, to acacetin which 0 catalyzedin engineered the E.O-methylation coli expressing of PaF4apigeninOMT to was form explored acacetin and in approximately E. coli, has been 25.2 mg/Lcharacterized of acacetin in liverworts.was synthesized. This is the first time that a Class II plant O-methyltransferase, which catalyzed the O-methylation of apigenin to form acacetin in E. coli, has been characterized in liverworts.

0 0 Figure 1. TheThe methylation of apigenin catalyzed by 4 ′--OO-methyltransferase-methyltransferase (Pa4 (Pa4′OMT).OMT).

2. Results Results

2.1. The The PaCOMT PaCOMT Sequences Sequences Four distinct distinct putative putative caffeic caffeic acid acid OMT OMT unigenes unigenes were were identified identified in in the the P.P. appendiculatum 0 transcriptome,transcriptome, and and designated designated as as PaCOMT1-PaCOMT4PaCOMT1-PaCOMT4. The. The 3′ end 3 end of PaCOMT2 of PaCOMT2 was was obtained obtained by RACE-PCR.by RACE-PCR. Amplif Amplificationication from from P.P. appendiculatum appendiculatum ofof the four fullfull lengthlength sequences sequences revealed revealed ORFs ORFs of ofthe the following following lengths: lengths: 1176 1176 bp, 1194 bp, bp,1194 1137 bp, bp, 1137 and bp, 1137 and bp, 1137 corresponding bp, corresponding to predicted to translationpredicted

Molecules 2017, 22, 759 3 of 11

Molecules 2017, 22, 759 3 of 11 product sizes of 43.4 KDa, 44.1 KDa, 41.5 KDa, and 41.5 KDa, respectively. The PaCOMT1 sequence includedtranslation two start product codons sizes separated of 43.4 by KDa, 27 amino 44.1 acids.KDa, The41.5 shorterKDa, and product 41.5 (transcribedKDa, respectively. from the The more 30 placedPaCOMT1 ATG) sequence was designated includedPaCOMT1-Tr two start codons, translating separated to a by 40.7 27 KDaamino product. acids. The shorter peptide product sequence identity(transcribed between from PaCOMT1 the more and 3′ placed PaCOMT2 ATG) was was 68.3%, designated while PaCOMT1-Tr that between, translating PaCOMT3 to and a 40.7 PaCOMT4 KDa wasproduct. 87.6%. AThe multiple peptide sequence sequence alignmentidentity between involving PaCOMT1 PaCOMT1 and andPaCOMT2 PaCOMT3 was 68.3%, with some while related that between PaCOMT3 and PaCOMT4 was 87.6%. A multiple sequence alignment involving PaCOMT1 proteins revealed a rather low level of identity: respectively, 31.8% and 34.3% with Medicago sativa and PaCOMT3 with some related proteins revealed a rather low level of identity: respectively, COMT (MsCOMT), and 23.5% and 26.2% with Medicago sativa IOMT (MsIOMT) (Figure2). The three 31.8% and 34.3% with Medicago sativa COMT (MsCOMT), and 23.5% and 26.2% with Medicago sativa catalytic residues (His-269, Glu-297, and Glu-329) present in MsIOMT [17] were conserved in the IOMT (MsIOMT) (Figure 2). The three catalytic residues (His-269, Glu-297, and Glu-329) present in PaCOMTs. With respect to the SAM-binding residues, most of them were retained in PaCOMT3, but MsIOMT [17] were conserved in the PaCOMTs. With respect to the SAM-binding residues, most of onlythem a few were in PaCOMT1. retained in TherePaCOMT3, was little but evidenceonly a few for in the PaCOMT1. conservation There of was substrate-binding little evidence for residues the in theconservation two PaCOMTs. of substrate-binding A phylogenetic residues analysis in ofthe the two PaCOMTs PaCOMTs. suggested A phylogen threeetic clades,analysis reflecting of the substratePaCOMTs specificity suggested (Figure three3). clades, The Class reflecting I OMT subs sequencestrate specificity clustered (Figure together, 3). The whereas Class theI OMT Class IIs weresequences distributed clustered in togeth a separateer, whereas group. the The Class four IIsPaCOMT were distributedgenes clustered in a separate with group. the flavone The four OMT genesPaCOMT rather genes than withclustered the caffeicwith the acid flavone OMT OMT genes genes and ra theyther werethan with located the caffeic in the acid root OMT of this genes clade. Theand implication they were was located that, unlike in the most root COMT of this enzymes, clade. The the implication PaCOMTs likelywas that, had unlike a substrate most preference COMT for flavonoids.enzymes, the PaCOMTs likely had a substrate preference for flavonoids.

PaF4'OMT MAVSTNGKTSEGVSDVSRTKNITNGNGMDMSVSANVASIGSVSATPLSPEDKRDHVVAIN 60 PaCOMT3 ...... MVPQDQRPLNPSVPLVADKRMDPSVPASQSDGAALALLD 39 MsIOMT ...... MASSLNGRKPSEIFKAQALLYK 22 MsCOMT ...... MGSTGETQITPTHISDEE..ANLFAMQ 25

PaF4'OMT FLLGWVQTWALKAVVTMRIPDIISKAPMD..ALTAREIISQIQCKNPSNAVPLLERLMRF 118 PaCOMT3 LSLIMGTPMALRAAITLGVPEILARHSEPLTAEEILLEISGCKDKHDKLAAENLRRILAI 99 MsIOMT HIFAFIDSMSLKWAVEMNIPNIIHNHGKP.....ISLSNLVSILQVPSSKIGNVRRLMRY 77 MsCOMT LASASVLPMILKSALELDLLEIIAKAGP..GAQISPIEIASQLPTTNPDAPVMLDRMLRL 83 * PaF4'OMT MVCKGFFSES....VNGENEIAYELNGVSKMYVT.DHEMSVVEWVNVAFHRSGIPALTRM 173 PaCOMT3 LVQNNVFSD..ESGEESGRSPLYGLTPLSKLLVD.EPECKLMCSFEVHQQAKLTKAWLHL 156 MsIOMT LAHNGFFEII...... TKEEESYALTVASELLVR.GSDLCLAPMVECVLDPTLSGSYHEL 130 MsCOMT LACYIILTCSVRTQQDGKVQRLYGLATVAKYLVKNEDGVSISALNLMNQDKVLMESWYHL 143 * PaF4'OMT HEAITED.VCPWDQAHGMPIYAHNPKFPELGDFFHKAMIVHSNFLTKILAEEYEGFKDFK 232 PaCOMT3 HEAVLDSSARPFEVANGMTIWQAFSEDKELAAHFHSAFQSGVHSHAKIFLQGYDGFEGVQ 216 MsIOMT KKWIYEEDLTLFGVTLGSGFWDFLDKNPEYNRSFNDAMASDSKLINLALRDCDFVFDGLE 190 MsCOMT KDAVLDG.GIPFNKAYGMTAFEYHGTDPRFNKVFNKGMSDHSTITMKKILETYTGFEGLK 202 * * PaF4'OMT VLVDVGGSLGNNLALILRFHPHLRGINFDMPSVIAGAPSFPGVEHVGGNFYESVPSGDAI 292 PaCOMT3 TLVDVGGSYGTLLGIIVQAKPHIRGINFDRPHIIASAPELPGVEHVGGDFFESVPKGDTI 276 MsIOMT SIVDVGGGNGTTGKIICETFPKLKCIVFDRPQVVENLSGSNNLTYVGGDMFTSIPNADAV 250 MsCOMT SLVDVGGGTGAVINTIVSKYPTIKGINFDLPHVIEDAPSYPGVEHVGGDMFVSIPKADAV 262

PaF4'OMT MLMAILHNNSDENCVKILRSVHAALPEDGK...LIVAEYIMTN..EADHNGERIRLLDML 347 PaCOMT3 LLKSILHNWDDEQCIKILHNCRKSLPESGK...VIIIEEVLLPPSNTSCKSRVAHKMNVS 333 MsIOMT LLKYILHNWTDKDCTRILKKCKEAVTNDGKKGKVIIIDMVINEKKDENQVTQIKLLMDVN 310 MsCOMT FMKWICHDWSDEHCLKFLKNCYEALPDNGK...VIVAECILPVAPDSSLATKGVVHIDVI 319 * * PaF4'OMT MLVQFRGGEERTFEHHKTLYEAAGFRHYKLIKTSGPMDIIEVRK. 391 PaCOMT3 MLATCDG.KERYAEDWSNLISSAGFQRHRMITLPNSFFDIIEVPM 377 MsIOMT MACLNGK..ERNEEEWKKLFIEAGFQDYKISPLTGFLSLIEIYP. 352 MsCOMT MLAHNPGGKERTQKEFEDLAKGAGFQGFKVHCNAFNTYIMEFLKK 364 * *

Figure 2. Sequence alignment of PaF40OMT, PaCOMT3, Medicago sativa IOMT (MsIOMT), and Figure 2. Sequence alignment of PaF4′OMT, PaCOMT3, Medicago sativa IOMT (MsIOMT), and Medicago sativa COMT (MsCOMT). Residues involved in SAM binding ( ), substrate binding (*), Medicago sativa COMT (MsCOMT). Residues involved in SAM binding (●), substrate binding (*), and catalysis ( ) are shown. Markers are located at the bottom of the sequences. and catalysisF (★) are shown. Markers are located at the bottom of the sequences.

Molecules 2017, 22, 759 4 of 11 Molecules 2017, 22, 759 4 of 11

97 Glycyrrhiza echinata D7OMT Medicago sativa IOMT 84 Medicago truncatula IOMT4 FOMT 81 Glycyrrhiza echinata HI4OMT 97 Medicago truncatula IOMT7 43 Hordeum vulgare F7OMT 98 Plagiochasma appendiculatum F4'OMT Plagiochasma appendiculatum COMT2 67 Plagiochasma appendiculatum COMT3 98 Plagiochasma appendiculatum COMT4 89 Sorghum bicolor COMT Zea mays COMT Arabidopsis thaliana COMT Prunus dulcis COMT 55 COMT 48 Populus tomentosa COMT 43 Medicago sativa COMT 67 Stylosanthes humilis COMT Mesembryanthemum crystallinum PFOMT Medicago sativa CCoAOMT 100 Plagiochasma appendiculatum OMT1 49 Populus tomentosa CCoAOMT CCoAOMT 28 Eucalyptus gunnii CCoAOMT

20 Broussonetia papyrifera CCoAOMT 21 Nicotiana tabacum CCoAOMT

0.5

Figure 3. Phylogeny of the PaCOMTs and related OMTs. The scale indicates evolutionary distance. Figure 3. Phylogeny of the PaCOMTs and related OMTs. The scale indicates evolutionary distance. The numbers shown are bootstrap values, based on 1000 replicates. The bar represents evolutionary The numbers shown are bootstrap values, based on 1000 replicates. The bar represents evolutionary O O distance.distance. COMT: COMT: caffeic caffeic acid acid O-methyltransferase;-methyltransferase; IOMT:IOMT: isoflavonoidisoflavonoid O-methyltransferase;-methyltransferase; DOMT: DOMT: 0 0 daidzeindaidzeinO -methyltransferase;O-methyltransferase; HI4HI4OMT:′OMT: 2,7,42,7,4-hydroxyflavonone′-hydroxyflavonone OO-methyltransferase;-methyltransferase; FOMT: FOMT: flavonoidflavonoidO-methyltransferase. O-methyltransferase.

2.2.2.2. Purification Purification and and In In Vitro Vitro Enzyme Enzyme Assays Assays of of Recombinant Recombinant PaCOMT1-4PaCOMT1-4 WhenWhen each each of theof PaCOMTsthe PaCOMTs was expressedwas expressed as a polyhistidine-tagged as a polyhistidine-tagged fusion infusionE. coli, inSDS-PAGE E. coli, analysisSDS-PAGE confirmed analysis that confirmed a high level that ofa recombinanthigh level ofprotein recombinant expression protein had expression been achieved. had been The molecularachieved. mass The ofmolecular each of mass the recombinant of each of the proteins recombinant was aroundproteins 60was KDa around (Figure 60 KDa S1). (Figure As putative S1). COMTs,As putative caffeic COMTs, acid was caffeic tested acid by enzymewas tested assays, by en whilezyme PaCOMTsassays, while failed PaCOMTs to convert failed it to ferulicconvert acid. it Amongto ferulic the other acid. possibleAmong substrates,the other possible PaCOMT1 substrates, displayed PaCOMT1 activity towardsdisplayed the activity flavonoids towards apigenin, the luteolin,flavonoids scutellarein, apigenin, genkwanin, luteolin, scutellarein, naringenin, quercetin,genkwanin, kaempferol, naringenin, and quercetin, eriodictyol. kaempferol, The identities and of alleriodictyol. the catalytic The productsidentities fromof all the thein catalytic vitro assays products were from confirmed the in vitro by comparing assays were the confirmed retention timeby comparing the retention time and mass spectrum with their respective authentic standards. and mass spectrum with their respective authentic standards. According to a standard calibration According to a standard calibration curve, the quantity of product could be calculated. Their curve, the quantity of product could be calculated. Their reaction activities were summarized (Table1, reaction activities were summarized (Table 1, Figure 4). Its optimal substrate was apigenin (100% Figure4). Its optimal substrate was apigenin (100% relative activity, 86.8 nmol/mg/min), followed by relative activity, 86.8 nmol/mg/min), followed by luteolin (37.8%, 32.9 nmol/mg/min) (Figure 4). The luteolin (37.8%, 32.9 nmol/mg/min) (Figure4). The products from apigenin and luteolin shared an products from apigenin and luteolin shared an identical retention time as, respectively, acacetin identicaland . retention Mass time spectrometry as, respectively, identified acacetin that and the diosmetin.catalytic product Mass of spectrometry apigenin had identified a parent ion that the catalytic product of apigenin had a parent ion peak [M − H]− at an m/z of 283, and among its peak [M − H]− at an m/z of 283, and among its fragmentation products was an [M – H − CH3]− ion − fragmentationwith an m/z productsof 268, consistent was an with [M – acacetin. H − CH The3] ionluteolin with product an m/z sharedof 268, the consistent same parent with ion acacetin. [M− − TheH] luteolin−at an m product/z of 299 sharedand a fragment the same ion parent at an ionm/z [Mof −284H] as diosmetinat an m/z (Fofigure 299 and5). The a fragment PaCOMT1-Tr ion at an proteinm/z of displayed 284 as diosmetin a much lower (Figure level5). Theof substrate PaCOMT1-Tr selectivity protein and activity displayed (Table a much 1), suggesting lower level that of substratethe longer selectivity PaCOMT1 and activitysequence (Table dominates1), suggesting in vivo. thatNone the of longer PaCOMT PaCOMT1 2-4 displayed sequence any dominates sign of in vivo. None of PaCOMT 2-4 displayed any sign of activity against any of the putative substrates.

MoleculesMolecules2017 2017, 22,, 22 759, 759 55 of of 1111

activity against any of the putative substrates. The conclusion was that, since PaCOMT1 was able to 0 Thecatalyze conclusion flavonoids was that, to sincetheir PaCOMT14′-O-methylether, was able it towas catalyze admissible flavonoids to rename to their it 4as-O PaF4-methylether,′OMT. Its 0 ◦ 0 it wasoptimal admissible temperature to rename was it 37°C. as PaF4 PaF4OMT.′OMT Its showed optimal temperaturea high catalytic was activity 37 C. PaF4 overOMT the showedpH range a high7.5–9.0, catalytic although activity the over substrates the pH became range 7.5–9.0,instable althoughat pH >8 the(Figure substrates S2). The became absence instable of Mg at2+ had pH >8no (Figureeffect S2).on Theits activity, absence in of accordance Mg2+ had nowith effect the onbehavior its activity, of Class in accordance II OMTs. withUnder the optimal behavior conditions of Class ◦ II OMTs.(37°C, pH Under 7.5), optimalthe enzyme’s conditions Km was (37 31.0C, µM pH 7.5),and its the K enzyme’scat was 0.12K ms−1waswhen 31.0 providedµM and with its apigenin,Kcat was − − 0.12and, s 1respectively,when provided 52.1 withµM apigenin,and 0.08 s and,−1when respectively, provided 52.1withµ luteolinM and 0.08(Table s 12),when indicating provided that with the luteolinenzyme (Table has2 a), stronger indicating binding that the affinity enzyme and has thus a stronger a higher binding catalytic affinity efficiency and thustowards a higher apigenin catalytic than efficiencythat towards towards luteolin apigenin in vitro than. that towards luteolin in vitro.

HO Flavone

O O Flavanone OH

HO Flavonol 100 Isoflavone 80

60

40

20

0 Relative activity of PaF4'OMT of (%) activity Relative

Tested substrates 0 FigureFigure 4. The4. The substrate substrate preference preference of of PaF4 PaF4OMT.′OMT. The The activity activity towards towards apigenin apigenin was was setset atat 100,100, andand otherother substrates substrates were were normalized normalized accordingly. accordingly.

TableTable 1. 1.Specific Specific activity activity of of purified purified PaF4 PaF40OMT′OMT andand PaCOMT1-TrPaCOMT1-Tr recombinantrecombinant proteinsproteins withwith selectedselected substrates. substrates.

SubstratesSubstrates PaF4PaF40OMT′OMT PaCOMT1-Tr PaCOMT1-Tr Apigenin 86.84 ± 1.76a 11.30 ± 1.07 Apigenin 86.84 ± 1.76a 11.30 ± 1.07 LuteolinLuteolin 32.89 32.89± 2.63 ± 2.63 5.26 5.26 ± ±0.590.59 ScutellareinScutellarein 14.26 14.26± 1.64 ± 1.64 NDNDb b GenkwaninGenkwanin 7.77 7.77± 0.75 ± 0.75 ND ND BaicaleinBaicalein NDND ND ND AcacetinAcacetin NDND ND ND Eriodictyol 1.61 ± 0.12 ND NaringeninEriodictyol 1.61 <1 ± 0.12 ND ND QuercetinNaringenin 7.60 ± 0.79<1 ND ND KaempferolQuercetin 8.17 7.60± 0.50 ± 0.79 ND ND GenisteinKaempferol 2.21 8.17± 0.53 ± 0.50 ND ND CaffeicGenistein acid 2.21 ND ± 0.53 ND ND Caffeoyl aldehyde ND ND CaffeoylCaffeic alcohol acid ND ND ND ND 5-HydroxyconiferylCaffeoyl aldehyde aldehyde ND ND ND ND 5-HydroxyconiferylCaffeoyl alcohol alcohol ND ND ND ND a Activities5-Hydroxyconiferyl presented in nmol aldehyde mg−1 min −1 ± STDEV. ND b No product ND detected. 5-Hydroxyconiferyl alcohol ND ND a Activities presented in nmol mg−1 min−1 ± STDEV. b No product detected.

Molecules 2017, 22, 759 6 of 11 Molecules 2017, 22, 759 6 of 11

200 A D 100

100 50

0 0 B E 200 100

100 50

0 0 C F 200 Absorbance at 346 nm (mAU) nm 346 at Absorbance 100 100

50 0 0 10 12 14 16 18 11 12 13 14 15 16 17 18 Retention Time (min) Retention Time (min)

Max. 4.7e7 cps. Max. 3.3e7 cps. 268.3 284.3 100 a 100 c 80 80

60 60

40 40 283.4 20 20 299.5 256.3 240.3 0 0 100 140 180 220 260 300 340 380 100 140 180 220 260 300 340 380

Max. 1.6e7 cps. Rel. Int (%) Max. 5.3e7 cps. 284.3 100 268.3 b 100 d

80 80

60 60

40 283.3 40

20 20 256.3 299.4 0 0 100 140 180 220 260 300 340 380 100 140 180 220 260 300 340 380

m/z, Da m/z, Da

FigureFigure 5. 5. InIn vitro vitro enzymeenzyme activity ofof recombinantrecombinant proteins proteins produced produced in E.in coliE. ,coli using, using apigenin apigenin or luteolin or luteolinas the substrate.as the substrate. (A, B) Apigenin (A, B) Apigenin supplied assupplied the substrate; as the HPLC substrat analysise; HPLC of reaction analysis products of reaction of cells productsharboring of (A)cells an harboring empty vector (A) an and empty (B) PaF4vector0OMT. and (B) (D ,PaF4E) Luteolin′OMT. (D supplied, E) Luteolin as the supplied substrate; as the HPLC substrate;analysis ofHPLC reaction analysis products of re ofaction cells harboringproducts (D)of cells an empty harboring vector (D) and an (E) empty PaF4 0OMT.vector ( Cand) Acacetin (E) PaF4standard.′OMT. (F(C) Diosmetin) Acacetin standard.standard. The (F) lower Diosmetin panel showsstandard. the massThe spectrometrylower panel spectrumshows the of (massa, c) the spectrometryproduct of PaF4 spectrum0OMT of supplied (a, c) the with product (a) apigenin of PaF4′ andOMT (c supplied) luteolin with as substrate. (a) apigenin (b) Acacetinand (c) luteolin standard. as(d substrate.) Diosmetin (b) standard.Acacetin standard. (d) Diosmetin standard.

TableTable 2. 2. KineticKinetic parameters parameters of ofrecombinant recombinant PaF4 PaF4′OMT0OMT using using apigenin apigenin and and luteolin luteolin as substrates. as substrates.

−1 −1 −1 −1 −1 Substrates Km (μM) Vmax (nmol mg−1 min−1) Kcat (s−1) Kenz (M − 1s −)1 Ki(μM) Substrates Km (µM) Vmax (nmol mg min ) Kcat (s ) Kenz (M s ) Ki(µM) Apigenin 31.0 ± 5.90 162.5 ± 7.29 0.117 ± 0.005 3795.9 − Apigenin 31.0 ± 5.90 162.5 ± 7.29 0.117 ± 0.005 3795.9 − LuteolinLuteolin 52.1 52.1 ± ±10.5410.54 110.8 110.8 ±± 12.6312.63 0.080 0.080 ± 0.009 1537.8 1537.8 292.8 292.8 ± 69.03± 69.03

Molecules 2017, 22, 759 7 of 11 Molecules 2017, 22, 759 7 of 11

2.3. Synthesis2.3. Synthesis of 40 -Methylatedof 4′-Methylated Form Form of Apigenin of Apigenin in E. in coli E. coli The E.The coli E.strain coli strain harboring harboring the pET32a-PaF4the pET32a-PaF40OMT′OMT construct construct was was fed fed with with apigenin apigenin in orderin order to to produceproduce acacetin. acacetin. HPLC HPLC analysis analysis of the reactionof the reaction product confirmedproduct confirmed that a single that product a single was product produced, was as inproduced, the in vitro asreactions. in the in vitro The conversionreactions. The after conversion 12 h was 15.0after mg/L12 h was (52.9 15.0µM), mg/L after (52.9 24 hµM), 17.5 after mg/L 24 h 17.5 mg/L (61.6 µM), after 36 h 21.3 mg/L (74.9 µM), and after 48 h 25.2 mg/L (88.8 µM) (Figure 6). (61.6 µM), after 36 h 21.3 mg/L (74.9 µM), and after 48 h 25.2 mg/L (88.8 µM) (Figure6). An 88.8% An 88.8% conversion of apigenin to acacetin was achieved after the 48 h incubation, a level which is conversion of apigenin to acacetin was achieved after the 48 h incubation, a level which is promising promising for the bio-based synthesis of 4′-methylated form of apigenin. for the bio-based synthesis of 40-methylated form of apigenin.

100 M)

μ 80

60 pET32a PaF4'OMT 40

20

0 Concentration of acacetin ofConcentration ( 0 12 24 36 48 Time of incubition (h)

FigureFigure 6. The 6. catalysisThe catalysis of apigenin of apigenin by PaF4 by PaF40OMT′OMT in E. in coli E.cells coli cells provided provided with with 100 µ 100M apigeninµM apigenin after after a 48 h incubation.a 48 h incubation. The control The control is represented is represented by cells by harboring cells harboring an empty an empty pET32a pET32a vector. vector.

3. Discussion3. Discussion In theIn presentthe present investigation, investigation, four four putative putative COMT-encoding COMT-encoding genes genes have have been been isolated isolated from from the the liverwortliverwort species speciesP. appendiculatum P. appendiculatumand and expressed expressed heterologously heterologously in E.in coli.E. coli.Three Three of of these these were were unableunable to catalyze to catalyze a range a range of phenylpropanoid of phenylpropanoid and and flavonoid flavonoid substrates, substrates, and and PaF4 PaF40OMT′OMT was was tested tested to reactto react most most effectively effectively with with apigenin. apigenin. Although Although the the PaF4 PaF40OMT′OMT sequence sequence was was more more like like that that of of a a COMTCOMT than athan flavonoid a flavonoid OMT, theOMT, present the experimentalpresent experi substratemental substrate specificity specificity tests and kinetictests and analyses kinetic (Tablesanalyses1 and2 (Tables) indicated 1 and that 2) indicated it was a flavonoid, that it was B-ring-specific a flavonoid, B-ring-specific OMT. A similar OMT. outcome A similar resulted outcome fromresulted the characterization from the characterization of wheat OMT2 of wheat (TaOMT2), OMT2 which,(TaOMT2), inspite which, of its inspite homology of its tohomology the COMTs, to the displaysCOMTs, a pronounced displays preferencea pronounced for apreference flavone substrate for a flavone [18]. The substrate TaOMT2 [18]. catalyzed The TaOMT2 three sequential catalyzed O-methylationsthree sequential of a flavonoid O-methylations substrate, whileof a PaF4flavon0OMToid showedsubstrate, strict while site-specificity PaF4′OMT to theshowed flavonoid strict 40-OHsite-specificity group. PaF4 0toOMT the flavonoid exhibited a4′-OH high group. level of PaF4 site-specificity′OMT exhibited and aa narrowhigh level substrate of site-specificity recognition. and Its differentiala narrow substrate response torecognition. the flavanoids Its differential naringenin resp andonse apigenin to the showedflavanoids that naringenin its catalytic and efficiency apigenin was promotedshowed that by theits presencecatalytic ofefficiency a double bondwas promot betweened Positions by the presence 2 and 3. Meanwhile, of a double the bond presence between of a 30-OHPositions group 2 impaired and 3. Meanwhile, catalysis, since the presence the rate of of luteolin a 3′-OH conversion group impaired was only catalysis, 38% that since of apigenin. the rate of A methylluteolin group conversion at Position was only 7 was 38% not that beneficial, of apigenin. as demonstrated A methyl group by theat Position contrasting 7 was responses not beneficial, to genkwaninas demonstrated and apigenin. by the Scutellarein contrasting (which responses carries to a 4genkwanin0-OH group and in itsapigenin. B ring) wasScutellarein converted (which to carries a 4′-OH group in its B ring) was converted to by PaF4′OMT, while was hispidulin by PaF40OMT, while baicalein was not accepted as substrate; these results further confirmed not accepted as substrate; these results further confirmed that the presence of a 4′-OH group is that the presence of a 40-OH group is essential for PaF40OMT activity. The isoflavone genistein showed essential for PaF4′OMT activity. The isoflavone genistein showed the same 4′-OH group as apigenin the same 40-OH group as apigenin except for the attachment position of the B ring. This slight difference except for the attachment position of the B ring. This slight difference was sufficient to greatly was sufficient to greatly compromise enzyme activity. When provided with methyl apigenin already compromise enzyme activity. When provided with methyl apigenin already methylated on its A methylated on its A ring (Position 7), the ability of PaF40OMT to transfer a methyl group to the 40-OH ring (Position 7), the ability of PaF4′OMT to transfer a methyl group to the 4′-OH group was group was inhibited. inhibited. Given the great potential application of acacetin in the field of medicine, initial researchers focused Given the great potential application of acacetin in the field of medicine, initial researchers its production on plants [19–21]; however, these extractions were restricted by the relative low content focused its production on plants [19–21]; however, these extractions were restricted by the relative of the metabolite in the plant. The leading outcome of the present research is the identification of a low content of the metabolite in the plant. The leading outcome of the present research is the

Molecules 2017, 22, 759 8 of 11 candidate gene encoding an enzyme able to synthesize acacetin from apigenin. The chemical synthesis of methylated flavonoids requires that non-target hydroxyl groups be protected, and this is assured by site-specific enzyme-based protocol. While enzymatic alkylation remains an important target of biocatalysis, it is impeded by a lack of effective SAM recycling methods. In contrast, E. coli has a highly tractable genetic system, gradually being one of the most widely used microorganisms for the production of natural products and their derivatives [22]. In the present investigation, a biotransformation method to produce acacetin from apigenin was introduced. The transformation system was based on engineered E. coli harboring PaF40OMT as a host and exogenous addition of substrate. An 88.8% conversion of apigenin to acacetin in E. coli after 48 h was implied, which is promising for a method of producing acacetin without exogenous SAM. The E. coli system is designed to be a simple factory allowing production of acacetin in much less time and with much fewer economic materials, so this plant methyltransferase appears to act as an efficient tool to represent viable biocatalysis.

4. Materials and Methods

4.1. Plant Materials and Reagents P. appendiculatum plants were maintained in a greenhouse under a 12 h photoperiod and a temperature of ~25◦C. Flavonoids and their methylated products, caffeic acid, SAM, and other reagents were purchased from Chengdu Must Bio-technology (Chengdu, China), Alfa Aesar (Heysham, UK), and Sigma-Aldrich (St. Louis, MO, USA), respectively. The synthesis of caffeoyl aldehyde, caffeoyl alcohol, 5-hydroxy coniferyl aldehyde, and 5-hydroxy coniferyl alcohol was performed following procedures published elsewhere [23,24]. CCoA was synthesized from caffeic acid using liverwort 4-coumarate CoA ligase [25].

4.2. Sequence Analysis and cDNA Cloning Total RNA extracted from rinsed, snap-frozen two-month-old P. appendiculatum thalli, and it was used as template for cDNA synthesis, based on the use of a PrimeScript™ RT Master Mix (Takara, Otsu, Japan), following the manufacturer’s protocol [26]. Four OMT sequences (PaCOMT1-PaCOMT4) were extracted from the resulting transcriptome sequence set and aligned with known Class II OMT sequences using DNAMAN v7 software (Lynnon LLC, San Ramon, CA, USA). A phylogenetic analysis was performed based on the maximum likelihood method implemented in MEGA v4.0 software [27]. The 962 bp PaCOMT2 sequence, lying at the 50 end of the coding region, was used to generate three gene-specific primers denoted PaCOMT2-GSP-1/-2/-3 (Table S1) in order to amplify its 30 end via RACE-PCR. A SMART RACE cDNA Amplification Kit (Clontech, Mountain View, CA, USA) was used for this purpose. The complete cDNA sequence was inferred from that of the partial cDNA fragments. The full-length sequences of PaCOMT1 through 4 were amplified from P. appendiculatum using various gene-specific primer pairs (Table S2), A-T cloned into the pMD19-T vector (Takara, Otsu, Japan), then transformed into E. coli DH5α.

4.3. Heterologous Expression and Purification The PaCOMT1 cDNA sequence proved to be about 80 bp longer than that of a typical COMT. It harbored two start codons, resulting in two possible transcripts, denoted PaCOMT1-Tr and PaCOMT1. The corresponding open reading frames (ORFs), along with those from PaCOMT2 through 4, were amplified from the four cDNA clones using the gene-specific primer pairs given in Table S3. The resulting amplicons were digested with a number of restriction enzymes (Takara) and ligated into the pET32a vector (Novagen, Darmstadt, Germany) digested with the respective restriction enzymes. The constructs were transformed into E. coli strain BL21 (DE3), and the transgenic cultures were ◦ incubated at 37 C until the OD600 reached 0.5. Expression of the transgene was induced by exposure to 1mM (80 µL) isopropyl β-D-1-thiogalactopyranoside (IPTG) for 16 h and a drop in the temperature Molecules 2017, 22, 759 9 of 11 to 18 ◦C. N-terminal hexahistidine-tagged proteins were purified by passing through a Ni-NTA Sefinose His-bind column (Bio Basic Inc., Markham, ON, Canada), and then were exchanged through an Ultrafiltration tube (Millipore, MA, USA) in the presence of binding buffer (20 mM Tris–HCl, 500 mM NaCl, pH 8.0). The homogeneity of the heterologously expressed protein was determined by SDS-PAGE and its concentration was assessed using Bradford’s reagent (Beyotime, Shanghai, China) employing bovine serum albumin as the standard.

4.4. Enzyme Assays Enzyme assays were performed in 50 µL volume reactions, each containing 1mg/mL purified protein (2 µL), 400mM DTT (0.5 µL), 50mM SAM(0.5 µL), and 10mM substrate (1 µL), 400 mM Tris-HCl buffer (pH 7.5, 25µL) and double distilled water (25 µL). The reactions were incubated at 37 ◦C for 30 min and terminated by the addition of 50 µL of acetonitrile. When caffeoyl CoA was employed as substrate, enzyme assays were performed using 4-fold reaction volumes and stopped by adding 5 M NaOH (12 µL) to hydrolyze its thioester bond. To neutralize the mixture, 6 M HCl (28 µL) was added, and this mixture was afterwards extracted by an equal volume of ethyl acetate. The resulting extract was dried and then redissolved in 50 µL of methanol. Proteins extracted from E. coli BL21 cells harboring an empty pET32a plasmid were used as the negative control. Reaction products were subjected to LC-MS analysis, using an Agilent 1100 system (Agilent Technologies, Santa Clara, CA, USA) equipped with a reversed phase C18 column (Agilent) and an electron spray ionization mass spectrometer. The two mobile phases were 0.1% aqueous acetic acid and methanol. The flow rate was set to 1.0 mL/min, applying a linear gradient from 20% methanol to 50% methanol over 20 min for the phenylpropanoids, and from 35% methanol to 65% methanol for the flavonoids. The absorbance of the reaction product was monitored at 320 nm and 346 nm. The effect of varying the pH and temperature on enzyme activity was determined by comparisons with the standard assay conditions described above. The range of pH was 6.0–9.0 and the temperature range investigated was 25–50 ◦C. 2+ The impact of the presence of Mg cations was assessed by replacing MgCl2 with either ethylene diamine tetraacetic acid (EDTA) or water. Kinetic assays were performed by varying the concentration of apigenin or luteolin from 5–400 µM at the optimal pH and temperature and allowing a 10 min reaction time. The quantity of reaction product generated was estimated using a standard calibration curve. Subsequently, Vmax, Km, and Ki values were calculated using the Michaelis-Menten equation implemented in Graphpad Prism 5 software (GraphPad Software, La Jolla, CA, USA).

4.5. Production of Apigenin 40-O-Methoxides in E. coli E. coli BL21 cells harboring pET32a-PaCOMT1 were pre-cultured overnight at 37 ◦C in 3 mL of LB liquid medium; the following day, 300 µL of the culture was seeded into 15 mL of LB liquid medium ◦ in the presence of the ampicillin antibiotic, and the culture was maintained at 37 C until the OD600 had reached 0.6. IPTG was then added to a concentration of 0.4 mM, and the cells were held at 16 ◦C for 5 h. Then, apigenin dissolved in dimethyl sulfoxide was added at a concentration of 100 µM. After 12–48 h incubation at 16 ◦C, a 500 µL aliquot was collected and extracted in an equal volume of ethyl acetate, the extracts air-dried at room temperature, and the residue dissolved in 100 µL of methanol. For the subsequent HPLC analysis, extracts from E. coli BL21 cells carrying an empty pET32a plasmid were used as the negative control.

4.6. Accession Number The Genbank accession numbers used are as followed: Plagiochasma appendiculatum transcriptome sequence (SRP073827), Mesembryanthemum crystallinum PFOMT (AY145521), PaF40OMT (KY977687), PaCOMT2 (KY977688), PaCOMT3 (KY977689), PaCOMT4 (KY977690), Plagiochasma appendiculatum OMT1 (KP729179), Medicago sativa CCoAOMT (AAC28973.1), Populus tomentosa CCoAOMT (ACE95173.1), Eucalyptus gunnii CCoAOMT (CAA72911.1), Broussonetia papyrifera CCoAOMT (AAT37172.1), Nicotiana tabacum CCoAOMT (AAC49913.1), Sorghum bicolor COMT (AAL57301), Molecules 2017, 22, 759 10 of 11

Zea mays COMT (Q06509), Arabidopsis thaliana COMT (NP-200227), Prunus dulcis COMT (Q43609), Populus tomentosa COMT (AAF63200), Medicago sativa COMT (AAB46623), Stylosanthes humilis COMT (2119166A), Medicago sativa IOMT (AAC49927), Glycyrrhiza echinata D7OMT (AB091685), Medicago truncatula IOMT4 (DQ419912), Hordeum vulgare F7OMT (X77467), Medicago truncatula IOMT7 (DQ419914), and Glycyrrhiz aechinata HI40OMT (AB091684).

4.7. Catalog Number Catalog numbers are as follows: caffeic acid (A15950), SAM (A4377), apigenin (A0113), luteolin (A0108), scutellarein (A0779), daidzein (A0008), baicalein (A0018), acacetin (A0763), eriodictyol (A0402), naringenin (A0147), quercetin (A0083), kaempferol (A0129), and genistein (A0009).

5. Conclusions In summary, PaF40OMT is characterized as able to preferentially methylate the apigenin 40-OH group to form acacetin. It is also the first Class II OMT to be isolated from a liverwort species. The enzyme was confirmed to be a highly site-specific OMT, able to exclusively methylate the flavone 40-OH group in the presence of a methyl donor. This high selectivity provides a simple synthetic pathway for the bio-based transformation of apigenin and other flavones.

Supplementary Materials: Supplementary Materials are available online. Acknowledgments: We are grateful for the support of the National Natural Science Foundation of China (No. 31370330). Author Contributions: AXC designed the experiments. HL, RXX, and SG performed the experiments. HL and AXC analyzed the data and wrote the paper. All the authors analyzed the results and edited and approved the final version of the paper. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Kumar, S.; Pandey, A.K. Chemistry and biological activities of flavonoids: An overview. Sci. World J. 2013, 2013, 162750. [CrossRef][PubMed] 2. Patel, D.; Shukla, S.; Gupta, S. Apigenin and cancer chemoprevention: Progress, potential and promise (review). Int. J. Oncol. 2007, 30, 233–245. [CrossRef][PubMed] 3. Shukla, S.; Gupta, S. Apigenin: A promising molecule for cancer prevention. Pharm. Res. 2010, 27, 962–978. [CrossRef][PubMed] 4. Shukla, S.; Bhaskaran, N.; Babcook, M.A.; Fu, P.; Maclennan, G.T.; Gupta, S. Apigenin inhibits prostate cancer progression in TRAMP mice via targeting pi3k/Akt/FoxO pathway. Carcinogenesis 2014, 35, 452–460. [CrossRef][PubMed] 5. Walle, T. Methylation of dietary flavones increases their metabolic stability and chemopreventive effects. Int. J. Mol. Sci. 2009, 10, 5002–5019. [CrossRef][PubMed] 6. Wen, X.; Walle, T. Methylated flavonoids have greatly improved intestinal absorption and metabolic stability. Drug Metab. Dispos. 2006, 34, 1786–1792. [CrossRef][PubMed] 7. Wu, H.J.; Wu, W.; Sun, H.Y.; Qin, G.W.; Wang, H.B.; Wang, P.; Yalamanchili, H.K.; Wang, J.; Tse, H.F.; Lau, C.P. Acacetin causes a frequency-and use-dependent blockade of hKv1.5 channels by binding to the S6 domain. J. Mol. Cell.Cardiol. 2011, 51, 966–973. [CrossRef][PubMed] 8. Carballo-Villalobos, A.I.; González-Trujano, M.E.; López-Muñoz, F.J. Evidence of mechanism of action of anti-inflammatory/antinociceptive activities of acacetin. Eur. J. Pain 2014, 18, 396–405. [CrossRef][PubMed] 9. Hsu, Y.L.; Kuo, P.L.; Lin, C.C. Acacetin inhibits the proliferation of Hep G2 by blocking cell cycle progression and inducing apoptosis. Biochem. Pharmacol. 2004, 67, 823–829. [CrossRef][PubMed] 10. Singh, R.P.; Agrawal, P.; Yim, D.; Agarwal, C.; Agarwal, R. Acacetin inhibits cell growth and cell cycle progression, and induces apoptosis in human prostate cancer cells: Structure-activity relationship with linarin and linarin acetate. Carcinogenesis 2005, 26, 845–854. [CrossRef][PubMed] Molecules 2017, 22, 759 11 of 11

11. Edwards, R.; Dixon, R.A. Purification and characterization of S-adenosyl-L-methionine: Caffeic acid 3-O-methyltransferase from suspension cultures of alfalfa (Medicago sativa L.). Arch. Biochem. Biophys. 1991, 287, 372–379. [CrossRef] 12. Kim, B.G.; Jung, B.R.; Lee, Y.; Hur, H.G.; Lim, Y.; Ahn, J.H. Regiospecific flavonoid 7-O-methylation with Streptomyces avermitilis O-methyltransferase expressed in Escherichia coli. J. Agric. Food Chem. 2006, 54, 823. [CrossRef][PubMed] 13. Liu, J.; Chen, L.; Cai, S.; Wang, Q. Semisynthesis of apigenin and acacetin-7-O-β-D-glycosides from naringin and their cytotoxic activities. Carbohydr. Res. 2012, 357, 41–46. [CrossRef][PubMed] 14. Hanamura, S.; Hanaya, K.; Shoji, M.; Sugai, T. Synthesis of acacetin and resveratrol 3, 5-di-O-β-glucopyranoside using lipase-catalyzed regioselective deacetylation of polyphenol glycoside peracetates as the key step. J. Mol. Catal. BEnzym. 2016, 128, 19–26. [CrossRef] 15. Fowler, Z.L.; Gikandi, W.W.; Koffas, M.A. Increased malonyl coenzyme A biosynthesis by tuning the Escherichia coli metabolic network and its application to flavanone production. Appl. Environ. Microbiol. 2009, 75, 5831–5839. [CrossRef][PubMed] 16. Kim, B.G.; Kim, H.; Hur, H.G.; Lim, Y.; Ahn, J.H. Regioselectivity of 7-O-methyltransferase of poplar to flavones. J. Biotechnol. 2006, 126, 241–247. [CrossRef][PubMed] 17. Zubieta, C.; He, X.Z.; Dixon, R.A.; Noel, J.P. Structures of two natural product methyltransferases reveal the basis for substrate specificity in plant O-methyltransferases. Nat. Struct. Mol. Biol. 2001, 8, 271–279. [CrossRef][PubMed] 18. Zhou, J.M.; Gold, N.D.; Martin, V.J.J.; Wollenweber, E.; Ibrahim, R.K. Sequential O-methylation of by a single gene product in wheat. BBA Gen. Subj. 2006, 1760, 1115–1124. [CrossRef][PubMed] 19. Srivastava, J.K.; Gupta, S. Extraction, characterization, stability and biological activity of flavonoids isolated from chamomile flowers. Mol. Cell. Pharmacol. 2009, 1, 138. [CrossRef][PubMed] 20. Liu, B.G.; Ning, Z.X.; Gao, J.H.; Xu, K.Y. Preparing apigenin from leaves of Adinandra nitida. Food Technol. Biotechnol. 2008, 46, 111–115. 21. Harbourne, N.; Jacquier, J.C.; O’Riordan, D. Optimisation of the extraction and processing conditions of chamomile (Matricaria.chamomilla L.) for incorporation into a beverage. Food Chem. 2009, 115, 15–19. [CrossRef] 22. Atsumi, S.; Cann, A.F.; Connor, M.R.; Shen, C.R.; Smith, K.M.; Brynildsen, M.P.; Chou, K.J.; Hanai, T.; Liao, J.C. Metabolic engineering of Escherichia coli for 1-butanol production. Metab. Eng. 2008, 10, 305–311. [CrossRef][PubMed] 23. Kim, S.J.; Kim, M.R.; Bedgar, D.L.; Moinuddin, S.G.; Cardenas, C.L.; Davin, L.B.; Kang, C.; Lewis, N.G. Functional reclassification of the putative cinnamyl alcohol dehydrogenase multigene family in Arabidopsis. Proc. Natl. Acad. Sci. USA 2004, 101, 1455–1460. [CrossRef][PubMed] 24. Sun, Y.; Wu, Y.; Zhao, Y.; Han, X.; Lou, H.; Cheng, A. Molecular cloning and biochemical characterization of two cinnamyl alcohol dehydrogenases from a liverwort Plagiochasma appendiculatum. Plant Physiol. Biochem. 2013, 70, 133–141. [CrossRef][PubMed] 25. Gao, S.; Yu, H.N.; Xu, R.X.; Cheng, A.X.; Lou, H.X. Cloning and functional characterization of a 4-coumarate CoA ligase from liverwort Plagiochasma appendiculatum. Phytochemistry 2015, 111, 48–58. [CrossRef][PubMed] 26. Gambino, G.; Perrone, I.; Gribaudo, I. A rapid and effective method for RNA extraction from different tissues of grapevine and other woody plants. Phytochem. Anal. 2008, 19, 520–525. [CrossRef][PubMed] 27. Tamura, K.; Dudley, J.; Nei, M.; Kumar, S. Mega4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 2007, 24, 1596–1599. [CrossRef][PubMed]

Sample Availability: Samples of the compounds are available from the authors.

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).