Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

cDNA Cloning and Biochemical Characterization of S-Adenosyl-l-Methionine: 2,7,4′-Trihydroxyisoflavanone 4′-O-Methyltransferase, a Critical Enzyme of the Legume Isoflavonoid Phytoalexin Pathway

cDNA Cloning and Biochemical Characterization of S-Adenosyl-l-Methionine:... Abstract Formononetin (7-hydroxy-4′-methoxyisoflavone, also known as 4′-O-methyldaidzein) is an essential intermediate of ecophysiologically active leguminous isoflavonoids. The biosynthetic pathway to produce 4′-methoxyl of formononetin has been unknown because the methyl transfer from S-adenosyl-l-methionine (SAM) to 4′-hydroxyl of daidzein has never been detected in any plants. A hypothesis that SAM: daidzein 7-O-methyltransferase (D7OMT), an enzyme with a different regiospecificity, is involved in formononetin biosynthesis through its intracellular compartmentation with other enzymes recently prevails, but no direct evidence has been presented. We proposed a new scheme of formononetin biosynthesis involving 2,7,4′-trihydroxyisoflavanone as the methyl acceptor and subsequent dehydration. We now cloned a cDNA encoding SAM: 2,7,4′-trihydroxyisoflavanone 4′-O-methyltransferase (HI4′OMT) through the screening of functionally expressed Glycyrrhiza echinata (Fabaceae) cDNAs. The reaction product, 2,7-dihydroxy-4′-methoxyisoflavanone, was unambiguously identified. Recombinant G.echinata D7OMT did not show HI4′OMT activity, and G. echinata HI4′OMT protein free from D7OMT was partially purified. HI4′OMT is thus concluded to be distinct from D7OMT, and their distant phylogenetic relationship was further presented. HI4′OMT may be functionally identical to (+)-6a-hydroxymaackiain 3-OMT of pea. Homologous cDNAs were found in several legumes, and the catalytic function of the Lotus japonicus HI4′OMT was verified, indicating that HI4′OMT is the enzyme of formononetin biosynthesis in general legumes. (Received September 26, 2002; Accepted December 25, 2002) Introduction Isoflavonoids are distributed mainly in the leguminous plants, and have critical physiological functions in the interaction with environmental microorganisms (Dixon 1999, Aoki et al. 2000). They are the most abundant antimicrobial phytoalexins (Dewick 1993), while they act as host signal molecules to the rhizobial bacteria (Phillips 1992, Spaink 1995) and the putative regulator of local auxin levels for the nodule organogenesis (Mathesius 2001) during symbiotic nitrogen fixation. The self-defense roles of isoflavonoids are also represented by their deterrent activity against insect feeding (Dewick 1993) and phytoestrogenic activity in higher animals, which causes infertility in sheep (Shutt 1976), although the estrogen activity in the diet is beneficial to human health (Stavric 1997). A methoxy group at C-4′ and a 3′,4′-methylenedioxy group are additional characteristic structural features of many isoflavonoids (Tahara and Ibrahim 1995, Clemens and Barz 1996): ca. 50% of naturally occurring isoflavonoids are 4′-methoxylated or 3′,4′-methylenedioxylated (Dewick 1993). The isoflavonoids possessing these substitutions constitute the major phytoalexins, e.g. medicarpin of alfalfa, pisatin of pea and vestitol of Lotus japonicus (Dixon 1999). These compounds originate from an isoflavone, formononetin (7-hydroxy-4′-methoxyisoflavone, also known as 4′-O-methyldaidzein) (Fig. 1). The biosynthetic process of the introduction of 4′-methoxyl to the precursor(s) to produce formononetin has, however, not been known. The methyl transfer from S-adenosyl-l-methionine (SAM) to 4′-hydroxyl of daidzein has never been detected in any plants that produce formononetin (Fig. 1). The O-methyltransferase (OMT) toward daidzein so far characterized produces only the 7-O-methylated isoflavone isoformononetin (4′-hydroxy-7-methoxyisoflavone) (Wengenmayer et al. 1974, Hagmann and Grisebach 1984, Edwards and Dixon 1991, Barz and Welle 1992, He and Dixon 1996, He et al. 1998), but the 7-O-methylated isoflavonoids are seldom plant constituents. Characterization of the 4′-O-methylation step is important both for the understanding of the biosynthetic mechanism of ecophysiologically active isoflavonoids and from the viewpoint of engineering a potential branching point in isoflavonoid synthesis to either methylated phytoalexins or the health-promoting phytoestrogens of soybean. The isoflavonoid skeleton is produced from a flavanone with 4′-hydroxyl through the action of a cytochrome P450 (P450), 2-hydroxyisoflavanone synthase (IFS), that catalyzes the hydroxylation at C-2 accompanied by the 1,2-aryl migration from C-2 to C-3 (Fig. 1) (Kochs and Grisebach 1986, Hashim et al. 1990). Then, the acid-labile IFS product is dehydrated to form isoflavone. An interesting hypothesis that the methyl transfer to 4′-hydroxyl is integrated into the aryl migration through the formation of 4′-methoxylated spiro-dienone intermediate was proposed (Crombie and Whiting 1992). However, the recent cloning of IFS cDNA (Akashi et al. 1999, Steele et al. 1999, Jung et al. 2000, Shimada et al. 2000) and the detailed biochemical study of the reaction (Akashi et al. 1999, Sawada et al. 2002) have shown that 2,7,4′-trihydroxyisoflavanone without 4′-O-methyl is the product from a flavanone, liquiritigenin (7,4′-dihydroxyflavanone). On the other hand, the hypothesis that a compartmentation of daidzein 7-OMT (D7OMT) in subcellular sites causes a change in the regiospecificity of the enzyme to produce formononetin was proposed (Dixon et al. 1996, Dixon 1999). This enzyme has been referred to as isoflavone OMT (IOMT) (He and Dixon 1996, He et al. 1998), but the name D7OMT more clearly indicating the substrate specificity and regiospecificity is used in the present article. This interesting hypothesis has been supported by the observations that transgenic alfalfa overexpressing D7OMT displayed enhanced production of 4′-O-methylated isoflavonoids on elicitation and that D7OMT fused with green fluorescent protein is located on the endoplasmic reticulum of alfalfa where the P450s are co-located (He and Dixon 2000, Dixon et al. 2001, Dixon 2001, Liu and Dixon 2001). However, no direct biochemical evidence through the in vitro demonstration of 4′-OMT activity has been provided. Cultured Glycyrrhiza echinata L. (Fabaceae) cells produce medicarpin by treatment with an elicitor, and constitutively accumulate a large quantity of formononetin (Nakamura et al. 1999). Recently, we demonstrated for the first time the in vitro activity of formononetin production from 2,7,4′-trihydroxyisoflavanone and SAM in the crude extract of G. echinata (Akashi et al. 2000). From this observation, which was made possible only by the use of a sufficient quantity of acid-labile 2-hydroxyisoflavanone substrate prepared by the heterologously expressed IFS, we suggested a new scheme of formononetin biosynthesis in which the substrate of the OMT is 2,7,4′-trihydroxyisoflavanone rather than daidzein and 4′-O-methylated 2-hydroxyisoflavanone subsequently undergoes another enzymatic dehydration to form formononetin (Fig. 1). We report here the molecular cloning of a cDNA encoding 2,7,4′-trihydroxyisoflavanone 4′-OMT (HI4′OMT) of G. echinata using a functional expression cloning method. The direct product of the OMT reaction, 2,7-dihydroxy-4′-methoxyisoflavanone, was isolated and unambiguously identified. Homologous cDNAs were found in other legumes, and the catalytic function of L. japonicus HI4′OMT was also verified. HI4′OMT and D7OMT are clearly distinct proteins encoded by different genes. These results provide the conclusive solution to the long-standing problem of isoflavonoid biosynthesis. Results Functional expression cloning of G. echinata HI4′OMT cDNA The formononetin-forming activity from 2,7,4′-trihydroxyisoflavanone and SAM in the G. echinata cells was activated by elicitation with yeast extract (Akashi et al. 2000). For the cloning of cDNA encoding HI4′OMT, G. echinata cDNAs were functionally expressed in Escherichia coli cells, and an E. coli clone displaying 2,7,4′-trihydroxyisoflavanone OMT activity was isolated as follows. λZapII cDNA library of elicited G. echinata was converted to the phagemid form by in vivo excision, and phagemids were then introduced into E. coli (Fig. 2). Five E. coli pools (ca. 10,000 clones pool–1) were prepared, and recombinant proteins in these pools were induced by isopropyl β-d-thiogalactopyranoside (IPTG). When crude extracts of the E. coli pools were reacted with 2,7,4′-trihydroxyisoflavanone and [14C]SAM, a new radioactive compound was produced in four pools. This product was assumed to be [14C]-2,7-dihydroxy-4′-methoxyisoflavanone because it was easily converted to [14C]formononetin by acid treatment. One pool was arbitrarily selected from the positive pools and subdivided into ten pools of smaller size (ca. 1,000 clones pool–1) for the second round of screening. Expression of proteins and the assay were performed again, and nine positive pools were identified. Subdivision of a randomly selected positive pool into ten pools of 1/10 size was performed, and the assay was repeated. Four positive pools (ca. 100 clones pool–1) and one positive pool (10 clones pool–1) were obtained from the third and the fourth screenings, respectively. Finally, a clone that showed 2,7,4′-trihydroxyisoflavanone OMT activity was isolated. The cDNA of the enzyme contained 1,349 bp nucleotides and encoded a polypeptide of 367 amino acid residues. The cDNA was named HI4′OMT. The deduced amino acid sequence of HI4′OMT had the conserved sequence motifs recorded for SAM-dependent OMTs of higher plants (Fig. 3) (Ibrahim and Muzac 2000, Schröder et al. 2002). HI4′OMT shared the highest identity (83%) with (+)-6a-hydroxymaackiain 3-OMT (accession no. U69554) of pea (Wu et al. 1997), 50% identity with D7OMT (IOMT) (accession no. U97125) of alfalfa (He et al. 1998), 30% identity with isoliquiritigenin/licodione 2′-OMT (accession no. D88742) of G. echinata (Haga et al. 1997) and 31% identity with caffeic acid 3-OMT (accession no. M63853) of alfalfa (Gowri et al. 1991). Product identification and the substrate specificity of HI4′OMT of G. echinata The product from the incubation of the recombinant HI4′OMT linking six histidine residues at the N-terminus with non-labeled SAM and 2,7,4′-trihydroxyisoflavanone was recovered and chemico-physically analyzed. It gave a single reverse-phase HPLC peak at retention time (Rt) 17.3 min (Fig. 4A, Chart 1) and the same Rf on thin-layer chromatography (TLC) as that of the product of radiolabeled assay during the functional expression cloning. It was readily converted to formononetin (Rt 37.5 min) by acid treatment (Fig. 4A, Chart 2). The UV spectrum of the product (λmax 279 nm and a shoulder at 315 nm) coincided well with the spectrum of 2,7,4′-trihydroxyisoflavanone (λmax 275 nm and a shoulder at 313 nm), indicating that it has a 2-hydroxyisoflavanone skeleton. The electron-impact mass spectrum of the compound exhibited a molecular ion peak at m/z 286 (C16H14O5), [M-H2O]+ peak at m/z 268 and also a retro-Diels-Alder fragment (C9H8O) peak at m/z 132 matching the O-methylated B-ring (Fig. 4B). The chemical structure of the direct reaction product was thus unambiguously identified as 2,7-dihydroxy-4′-methoxyisoflavanone, establishing the enzyme regiospecificity to be 4′-hydroxyl of 2,7,4′-trihydroxyisoflavanone. Table 1 summarizes the specific activity of HI4′OMT protein toward candidates for the isoflavonoid substrate. Daidzein was incubated with the recombinant HI4′OMT and [14C]SAM, but no reaction product was detected, indicating that HI4′OMT is distinct from D7OMT. When (±)-medicarpin (3-hydroxy-9-methoxypterocarpan), an analog of 6a-hydroxymaackiain, was reacted with the recombinant HI4′OMT and [14C]SAM, a radioactive compound assumed to be [14C]-3,9-dimethoxypterocarpan was produced. The specific activity of HI4′OMT to (±)-medicarpin (14 pkat mg–1) is in the same order as that of (+)-6a-hydroxymaackiain 3-OMT of pea to (+)-medicarpin (ca. 70–100 pkat mg–1; Wu et al. 1997), indicating very similar catalytic functions of these enzymes. cDNA cloning and biochemical characterization of G. echinata D7OMT To examine the 4′-O-transmethylation reaction to 2,7,4′-trihydroxyisoflavanone from SAM by D7OMT protein, a cDNA homologous to alfalfa D7OMT was screened from the G. echinata cDNA library using alfalfa D7OMT as a probe. The deduced amino acid sequence of G. echinata D7OMT showed 78% identity with alfalfa D7OMT (IOMT) and 50% identity with G. echinata HI4′OMT. As shown in Table 1, a crude extract of E. coli expressing D7OMT of G. echinata mediated the formation of [14C]isoformononetin from daidzein and [14C]SAM. However, neither HI4′OMT nor medicarpin OMT activities were detected. Partial purification of HI4′OMT and separation of HI4′OMT and D7OMT from G. echinata cells Partial purification of G. echinata HI4′OMT protein free from D7OMT was accomplished. The purification procedure is summarized in Table 2. D7OMT activity was observed in the crude extract (10,000×g supernatant) and ammonium sulfate fraction, but the activity was not recovered in the chromatofocusing step. The affinity-purified protein catalyzed SAM-dependent 4′-O-methylation of 2,7,4′-trihydroxyisoflavanone to produce 2,7-dihydroxy-4′-methoxyisoflavanone, but did not employ daidzein as the substrate, clearly indicating the distinct nature of HI4′OMT and D7OMT proteins. The specific activity (820 pkat mg–1) for 2,7,4′-trihydroxyisoflavanone was roughly in the same order as that of recombinant HI4′OMT (Table 1, 2). The purified protein, like recombinant HI4′OMT, displayed the OMT activity toward (±)-medicarpin in addition to 2,7,4′-trihydroxyisoflavanone (data not shown). Further, the protein eluted like the recombinant HI4′OMT at pH 4.8 on a chromatofocusing column. The biochemical properties of the purified protein thus coincided well with those of the recombinant HI4′OMT. Distribution of HI4′OMT in leguminous plants HI4′OMT-like sequences (>80% identity with that of G. echinata at the amino acid level) were found in the expressed sequence tag (EST) databases of leguminous plants, Lotus japonicus (accession no. AV407445, http://www.kazusa.or.jp/en/plant/lotus/EST/) (Asamizu et al. 2000), Medicago truncatula (accession no. TC28631, http://www.tigr.org/tdb/mtgi/) and Glycine max (accession no. TC101829, http://tigrblast.tigr.org/tgi/). The seedling of L. japonicus accumulates phytoalexin vestitol after treatment with reduced-glutathione as an elicitor, and we recently cloned and examined the elicitor responses of cDNAs encoding the enzymes of the pathway (Shimada et al. 2000). The L. japonicus EST clone AV407445, which has the predicted initiation codon, was obtained from Kazusa DNA Research Institute, and the full-length HI4′OMT cDNA was sequenced. It had 1,349 bp nucleotides, and the deduced amino acid sequence shared the highest identity (83%) with G. echinata HI4′OMT (Fig. 3). Crude extract of E. coli expressing L. japonicus HI4′OMT catalyzed the formation of [14C]-2,7-dihydroxy-4′-methoxyisoflavanone from 2,7,4′-trihydroxyisoflavanone and [14C]SAM. Like G. echinata HI4′OMT, no reaction product was formed by the incubation of the HI4′OMT with daidzein and [14C]SAM. Thus, L. japonicus HI4′OMT is functionally the same as G. echinata HI4′OMT. Discussion The hypothesis that D7OMT is the enzyme of formononetin biosynthesis was proposed by Dixon and his co-workers using alfalfa as the experimental system (Dixon et al. 1996, Dixon 1999). Initially, D7OMT contained in the subcellular compartment was hypothesized to catalyze the methyl transfer to the 4′-hydroxyl of daidzein. Recently, indirect biochemical evidence incorporating 2,7,4′-trihydroxyisoflavanone in, but excluding daidzein from, the intermediate of formononetin biosynthesis in alfalfa was presented (Liu and Dixon 2001). Still, however, in the alfalfa system, the HI4′OMT activity is attributed to D7OMT protein (Liu and Dixon 2001), supported by a three-dimensional crystal structure of D7OMT that can bind 2,7,4′-trihydroxyisoflavanone on a computer-generated illustration (Zubieta et al. 2001). Therefore, the major remaining problem was to determine whether HI4′OMT protein is identical to or distinct from D7OMT. In this study, a cDNA encoding HI4′OMT was successfully cloned from G. echinata cells. We clearly demonstrated that the G. echinata HI4′OMT is distinct from D7OMT (Table 1). Partial purification of HI4′OMT free from D7OMT further supported the distinct nature of these proteins. The biochemical evidence suggested that the purified protein is the product of HI4′OMT. In addition, transient accumulations of HI4′OMT mRNA in both elicited G. echinata cells and L. japonicus seedlings prior to the induction of formononetin-producing activity (G. echinata) (Akashi et al. 2000) and vestitol accumulation (L. japonicus) (Shimada et al. 2000) confirmed its involvement in formononetin biosynthesis (data not shown). We thus clarified here the missing link of isoflavonoid biosynthesis that has been a mystery for nearly three decades (Wengenmayer et al. 1974, Hagmann and Grisebach 1984, Barz and Welle 1992). Interestingly, HI4′OMT showed the highest amino acid identity (83%) with (+)-6a-hydroxymaackiain 3-OMT of pea (Preisig et al. 1989), the final enzyme in the biosynthesis of the phytoalexin (+)-pisatin, acting on the hydroxyl at the C-3 of (+)-pterocarpan skeleton that corresponds to C-7 in isoflavone numbering (Wu et al. 1997). G. echinata HI4′OMT is active on a pterocarpan, (±)-medicarpin, which has only one hydroxyl at C-3 as the methyl acceptor (Table 1), and (+)-6a-hydroxymaackiain 3-OMT was reported to be also active with (+)-medicarpin (Wu et al. 1997). It is thus reasonably assumed that pea (+)-6a-hydroxymaackiain 3-OMT is functionally identical to HI4′OMT. L. japonicus HI4′OMT was also functionally identified in this study, and HI4′OMT-like sequences were further found in soybean and M. truncatula. Because none of these plants except pea are known to produce pisatin, this family of proteins should primarily act as HI4′OMT. A close phylogenetic relationship between alfalfa D7OMT (IOMT) and pea (+)-6a-hydroxymaackiain 3-OMT among the higher plant OMTs has been demonstrated (Schröder et al. 2002, Gang et al. 2002). Now, the phylogenic tree composed of OMTs of the flavonoid pathway indicates that HI4′OMT and (+)-6a-hydroxymaackiain 3-OMT belong to the same branch, which is different from that of D7OMT (Fig. 5). Recently, a successful homology modeling, based on the D7OMT stereostructure, of the active sites of phenylpropane OMTs that are more distantly related than HI4′OMT to D7OMT, was reported (Gang et al. 2002). In a hypothetical binding of 2,7,4′-trihydroxyisoflavanone into the active site of D7OMT, the best fit was observed with (2S, 3R)-2,7,4′-trihydroxyisoflavanone among the four possible stereoisomers (Zubieta et al. 2001). In contrast, from the homology model of CYP93C2 protein (G. echinata IFS), the direct IFS reaction product from (2S)-liquiritigenin is predicted to be the opposite enantiomer (2R, 3S)-2,7,4′-trihydroxyisoflavanone (Sawada et al. 2002). Whether this isomer and the (+)-pterocarpan molecule can be accommodated in the active site of HI4′OMT protein in conformations favorable for the respective 4′- and 3-O-methylation should be examined by homology modeling and, in the future, from the crystallography data of HI4′OMT. While this study makes the hypothesis that D7OMT in a subcellular compartment is involved in formononetin biosynthesis unnecessary, the localization of HI4′OMT in the cell and its possible interaction with other enzymes of the (iso)flavonoid pathway are interesting. Although HI4′OMT as well as D7OMT are almost exclusively recovered in the soluble fraction (160,000×g supernatant) of G. echinata (data not shown), association of ‘soluble’ enzymes of flavonoid biosynthesis (Hrazdina 1992, Ibrahim et al. 1987) including alfalfa D7OMT (Dixon et al. 1996, Liu and Dixon 2001) with endoplasmic reticulum has been repeatedly described. In addition, interactions among the enzymes of flavonoid biosynthesis have been demonstrated in Arabidopsis thaliana (Burbulis and Winkel-Shirley 1999, Saslowsky and Winkel-Shirley 2001, Winkel-Shirley 2001). Because 2-hydroxyisoflavanones are not stable, 4′-O-methylation of 2,7,4′-trihydroxyisoflavanone should take place prior to the non-enzymatic dehydration in the cells, favoring the compartmentation of IFS, HI4′OMT, specific dehydratase and possibly P450 reductase on the endoplasmic reticulum membrane. The intrinsic activity of D7OMT remains to be investigated. The induction of D7OMT activity in elicited G. echinata cells (Akashi et al. 2000) in addition to the coincidence of D7OMT expression and phytoalexin production in alfalfa (He et al. 1998, He and Dixon 2000) strongly suggests its involvement in the host defense responses. Broad and flexible substrate specificities of plant OMTs sharing similar sequences (Ibrahim and Muzac 2000, Frick et al. 2001) and even changes of specificities through the formation of heterodimers of similar OMT proteins have been reported (Frick and Kutchan 1999). Indirect or regulatory roles for D7OMT in formononetin biosynthesis can be envisaged. Finally, the functional expression cloning used in this study is a rapid and convenient method when a high sensitivity assay, e.g. by the use of radioisotope-labeled substrates, is available. To our knowledge, this is the first successful cloning of a cDNA encoding the enzyme of plant-specific (secondary) metabolism using this method. In preliminary experiments, D7OMT, chalcone synthase and caffeic acid 3-OMT activities were also detected in the cDNA expression library of G. echinata (data not shown). This method will be a useful tool for the isolation of cDNA encoding enzymes in the diverse and complicated plant secondary metabolic pathway. To examine the possible interaction with enzymes of the formononetin pathway, cloning of the cDNA encoding 2,7-dihydroxy-4′-methoxyisoflavanone dehydratase is underway. Materials and Methods Buffers The following buffers were used: A, 100 mM potassium phosphate (pH 7.5) containing 10% (w/v) sucrose and 14 mM 2-mercaptoethanol; B, 25 mM histidine-HCl containing 1.4 mM 2-mercaptoethanol (pH 6.0); C, polybuffer 74 (Amersham Biosciences, Buckinghamshire, U.K.)-HCl with 1.4 mM 2-mercaptoethanol (pH 4.0); D, 50 mM Tris-HCl (pH 7.5) containing 14 mM 2-mercaptoethanol. Assays 2,7,4′-Trihydroxyisoflavanone was prepared by the incubation of the yeast microsome expressing IFS (CYP93C2) with (RS)-liquiritigenin and NADPH (Akashi et al. 1999). 2,7,4′-Trihydroxyisoflavanone (ca. 0.4 nmol) dissolved in 2 µl 2-methoxyethanol was incubated with an enzyme preparation of E. coli or G. echinata in the presence of 0.4 nmol S-adenosyl-l-[methyl-14C]methionine ([14C]SAM, 2.26 GBq mmol–1, Amersham Biosciences) in the total volume of 0.2 ml at 30°C for 10 min. After termination of the reaction with 5 µl acetic acid, the ethyl acetate extract of the mixture was subjected to silica-gel TLC [LK6DF (Whatman, Maidstone, U.K.); solvent, chloroform : acetone : 25% aq. ammonia solution = 70 : 29 : 1; 2,7,4′-trihydroxyisoflavanone (Rf 0.08), 2,7-dihydroxy-4′-methoxyisoflavanone (Rf 0.15), formononetin (Rf 0.30), isoformononetin (Rf 0.62)] and analyzed by a Typhoon 8600 image analyzer (Amersham Biosciences). Acid-catalyzed conversion of 2-hydroxyisoflavanone into isoflavone was performed as described (Akashi et al. 2000). Specific activity was determined by the incubation of the purified recombinant HI4′OMT (ca. 15 ng protein) or crude extract of E. coli expressing D7OMT (ca. 10 µg protein) with 2,7,4′-trihydroxyisoflavanone (ca. 0.8 nmol) and [14C]SAM (0.4 nmol). Incubation time was 6 min or shorter when the reaction rate was linear, and the reaction mixture was subjected to silica-gel TLC as above. [14C]-2,7-Dihydroxy-4′-methoxyisoflavanone was separated from TLC, and radioactivity was measured by liquid scintillation counting. D7OMT and medicarpin OMT were assayed by the same method as HI4′OMT using 10 nmol daidzein (Extrasynthése, Genay, France) or 0.5 nmol medicarpin (Plantech, Reading, U.K.). Prolonged incubation (20 min) was carried out because of low activity. The TLC condition for the medicarpin OMT product was: plate, LK6DF; solvent, toluene : ethyl acetate = 4 : 1; Rfs, medicarpin (0.58) and 3,9-dimethoxypterocarpan (0.86). For the assay with non-labeled substrates, the reaction mixture (2 ml) contained 1 µmol SAM, 200 nmol 2,7,4′-trihydroxyisoflavanone and the purified recombinant HI4′OMT (ca. 50 µg). HPLC was performed using a CAPCELL PAK C18 MG column (4.6×150 mm; Shiseido, Tokyo, Japan) at 40°C with a flow rate of 0.8 ml min–1 with a linear gradient elution for 35 min from 35% to 55% methanol in 3% aq. acetic acid. The eluate was monitored by a multiwavelength detector (MD-2010, Jasco, Tokyo, Japan). For mass spectrometric identification of the product, the ethyl acetate extract of the reaction mixture was applied to a silica-gel TLC [Kieselgel F254 (Merck, Darmstadt, Germany); solvent, toluene : ethyl acetate : methanol : light petroleum (6 : 4 : 1 : 3)], and a product (Rf 0.30) was collected and further purified by HPLC. The electron impact mass spectrum was recorded on a JEOL SX-102A mass spectrometer at the ionization voltage of 70 eV. Cloning of HI4′OMT cDNA from G. echinata cells In vivo excision of G. echinata λZapII cDNA library (Akashi et al. 1999) was performed using E. coli DH5α F′IQ (Invitrogen) with ExAssist helper phage. The excised phagemids were introduced into DH5α F′IQ, and the E. coli cells were grown on a Luria-Bertani (LB)/ampicillin (50 µg ml–1) agar plate. About 10,000 E. coli transformants were scraped from a plate and inoculated in 20 ml of LB/ampicillin medium in a 50 ml conical tube. The fraction was considered as one pool, and five independent pools were prepared. E. coli pools were incubated at 30°C until OD600 = 0.6. A small amount of the culture from each pool was placed on an LB/ampicillin agar plate for the second round screening. IPTG was added to the remaining pools to a final concentration of 5 mM, and the cultures were incubated for 12 h at 30°C. The medium was removed by centrifugation, and E. coli cells suspended in buffer A. The crude extract was prepared by a vigorous shaking with glass beads and centrifugation (10,000×g, 10 min), and HI4′OMT activity was assayed. A positive pool was arbitrarily selected and subdivided into 10 pools of smaller size (ca. 1,000 clones pool–1) for the next round screening. The fractionation of a positive pool and the assays were repeated, and a transformant showing 2,7,4′-trihydroxyisoflavanone OMT activity was isolated. Cloning of a cDNA homologous to alfalfa D7OMT from G. echinata cells D7OMT cDNA was cloned from alfalfa cells (Kirikae et al. 1993) by reverse transcription-PCR with specific primers prepared from the nucleotide sequence information of IOMT8 cDNA (He et al. 1998). Plaques (2×105) of the G. echinata λZapII cDNA library were screened with alfalfa D7OMT using an ECL Direct Nucleic Acid Labeling System (Amersham Biosciences). Positive clones were converted to pBluescript SK(–) by in vivo excision. The nucleotide sequence of a cDNA with a length of ca. 1,400 bp was determined. Heterologous expression of HI4′OMT and D7OMT in E. coli Coding regions of HI4′OMT and D7OMT of G. echinata were amplified by PCR with KOD polymerase (Toyobo, Tokyo, Japan) and cDNA clones as templates. NdeI and EcoRI sites were introduced upstream of the initiation codon and downstream of the stop codon of HI4′OMT, respectively. For D7OMT, the NheI site upstream of the initiation codon and HindIII sites downstream of the stop codon were introduced. The NdeI–EcoRI fragment of the PCR product from HI4′OMT was cloned into pET28a (Novagen) to express HI4′OMT as the fusion protein with six histidine residues. The NheI–HindIII fragment of the PCR product of D7OMT was cloned into pET21a to express D7OMT as non-tagged protein. E. coli BL21 (DE3) cells transformed with each vector were cultivated at 30°C in 200 ml LB medium supplemented with 50 µg ml–1 kanamycin or 50 µg ml–1 ampicillin until OD600 = 0.4. IPTG was added to a final concentration of 1 mM, and the cultures were incubated for 6 h at 30°C. HI4′OMT was purified using HisTrap Kit (Amersham Biosciences) from the crude extract of E. coli expressing HI4′OMT. An EST clone (accession no. AV407445) of L. japonicus accession ‘Miyakojima’ (Asamizu et al. 2000) was transformed in E. coli DH5α, and the protein was expressed by the addition of 5 mM IPTG and incubation for 12 h at 30°C. Purification of HI4′OMT from G. echinata cells From the cell-free extract of cultured G. echinata cells (Ak-1 line, 10 g) elicited with 0.2% (w/v) yeast extract (Difco, MI, U.S.A.) for 24 h (Nakamura et al. 1999), ammonium sulfate (30–80% saturation) precipitate was prepared as described (Akashi et al. 2000). The precipitate was dissolved in buffer B and desalted on a Sephadex G-25 column (Amersham Biosciences). The sample was subjected to chromatofocusing on a PBE94 column (5 ml, 6×200 mm, Amersham Biosciences) which had been equilibrated with buffer B. Elution was achieved with buffer C. The positive fraction was applied onto an adenosine-agarose gel (2 ml, 10×30 mm) (Attieh et al. 1995, Rakwal et al. 2000) that had been equilibrated with buffer D. The column was washed with buffer D containing 2 M NaCl. The protein was eluted with buffer D containing 4 mM of SAM, and the buffer was changed to buffer A by an ultrafiltration (Microcon YM-10, Millipore). Acknowledgments This work was supported by a Grant-in-Aid for Encouragement of Young Scientists (B) (no. 13780474) from the Japan Society for the Promotion of Science, and by a Grant-in-Aid for Scientific Research on Priority Areas (A) (no. 1302470) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan. EST clone (accession no. AV407445) of L. japonicus was obtained from Kazusa DNA Research Institute. 1 Corresponding author: E-mail, ayabe@brs.nihon-u.ac.jp; Fax, +81-466-80-1141. View largeDownload slide Fig. 1 Biosynthesis of O-methylated isoflavonoids. The open arrow indicates the pathway described in this study. The dashed arrow indicates the biosynthetic pathway that is not detected in G. echinata cells. Abbreviations used are: D7OMT, daidzein 7-OMT; IFS, 2-hydroxyisoflavanone synthase; IOMT, isoflavone OMT; HI4′OMT, 2,7,4′-trihydroxyisoflavanone 4′-OMT. View largeDownload slide Fig. 1 Biosynthesis of O-methylated isoflavonoids. The open arrow indicates the pathway described in this study. The dashed arrow indicates the biosynthetic pathway that is not detected in G. echinata cells. Abbreviations used are: D7OMT, daidzein 7-OMT; IFS, 2-hydroxyisoflavanone synthase; IOMT, isoflavone OMT; HI4′OMT, 2,7,4′-trihydroxyisoflavanone 4′-OMT. View largeDownload slide Fig. 2 Method of cloning G. echinata HI4′OMT cDNA. A λZapII cDNA library of G. echinata was converted to phagemid form to prepare the cDNA expression library. Recombinant proteins in E. coli pools were induced by IPTG, and HI4′OMT was assayed to identify the positive pools. After several rounds of screening, a single clone was isolated. The stars indicate putative clones of HI4′OMT. The gray test tubes contain positive pools. View largeDownload slide Fig. 2 Method of cloning G. echinata HI4′OMT cDNA. A λZapII cDNA library of G. echinata was converted to phagemid form to prepare the cDNA expression library. Recombinant proteins in E. coli pools were induced by IPTG, and HI4′OMT was assayed to identify the positive pools. After several rounds of screening, a single clone was isolated. The stars indicate putative clones of HI4′OMT. The gray test tubes contain positive pools. View largeDownload slide Fig. 3 Amino acid sequence of G. echinata HI4′OMT aligned with pea (+)-6a-hydroxymaackiain 3-OMT (accession no. U69554), L. japonicus HI4′OMT, G. echinata D7OMT and alfalfa D7OMT (IOMT) (accession no. U97125). The amino acid residues of which at least three sequences are identical are in reverse type. Gaps (–) are inserted to optimize alignment. Conserved sequence motifs in plant SAM-dependent OMT (arrow) (Ibrahim and Muzac 2000) and substrate preference motifs (broken line) (Schröder et al. 2002) are shown. Abbreviation: HMOMT, (+)-6a-hydroxymaackiain 3-OMT. View largeDownload slide Fig. 3 Amino acid sequence of G. echinata HI4′OMT aligned with pea (+)-6a-hydroxymaackiain 3-OMT (accession no. U69554), L. japonicus HI4′OMT, G. echinata D7OMT and alfalfa D7OMT (IOMT) (accession no. U97125). The amino acid residues of which at least three sequences are identical are in reverse type. Gaps (–) are inserted to optimize alignment. Conserved sequence motifs in plant SAM-dependent OMT (arrow) (Ibrahim and Muzac 2000) and substrate preference motifs (broken line) (Schröder et al. 2002) are shown. Abbreviation: HMOMT, (+)-6a-hydroxymaackiain 3-OMT. View largeDownload slide Fig. 4 HPLC (A) and electron impact mass spectrum (B) of the product of recombinant HI4′OMT reaction. (A) 1, direct reaction products; 2, product after the acid-treatment of the reaction mixture; 3, substrate (2,7,4′-trihydroxyisoflavanone, HI; Rt 6.2 min); 4, standard samples [D, daidzein (Rt 20.6 min); IF, isoformononetin (Rt 34.0 min); F, formononetin (37.5 min)]. The eluents were monitored at 285 nm. The product at Rt 17.3 min was purified by TLC and HPLC and analyzed by a mass spectrometer (B). The proposed fragmentation scheme is shown (C). View largeDownload slide Fig. 4 HPLC (A) and electron impact mass spectrum (B) of the product of recombinant HI4′OMT reaction. (A) 1, direct reaction products; 2, product after the acid-treatment of the reaction mixture; 3, substrate (2,7,4′-trihydroxyisoflavanone, HI; Rt 6.2 min); 4, standard samples [D, daidzein (Rt 20.6 min); IF, isoformononetin (Rt 34.0 min); F, formononetin (37.5 min)]. The eluents were monitored at 285 nm. The product at Rt 17.3 min was purified by TLC and HPLC and analyzed by a mass spectrometer (B). The proposed fragmentation scheme is shown (C). View largeDownload slide Fig. 5 Phylogenetic relationship of OMTs involved in flavonoid biosynthesis. Note that HI4′OMT and D7OMT belong to the different branches. The protein sequences were aligned with the CLUSTAL W program, and phylogeny was established by the neighbor-joining method (Thompson et al. 1994). The tree was displayed using TreeView software (Page 1996). View largeDownload slide Fig. 5 Phylogenetic relationship of OMTs involved in flavonoid biosynthesis. Note that HI4′OMT and D7OMT belong to the different branches. The protein sequences were aligned with the CLUSTAL W program, and phylogeny was established by the neighbor-joining method (Thompson et al. 1994). The tree was displayed using TreeView software (Page 1996). Table 1 Specific activities of heterologously expressed HI4′OMT and D7OMT of G. echinata Catalytic activity  Specific activity a (pkat mg–1)  HI4′OMT b  D7OMT c  2,7,4′-Trihydroxyisoflavanone 4′-OMT  4.8×103  0  Daidzein 7-OMT  0  23  Medicarpin OMT  14  0  Catalytic activity  Specific activity a (pkat mg–1)  HI4′OMT b  D7OMT c  2,7,4′-Trihydroxyisoflavanone 4′-OMT  4.8×103  0  Daidzein 7-OMT  0  23  Medicarpin OMT  14  0  Specific activity was calculated from the rate of radiolabeled isoflavonoid production from nearly identical amounts of [14C]SAM and the methyl acceptor (for details, see Materials and Methods). a Average of two independent experiments (maximum deviation was ca. 5%). b Purified HI4′OMT protein was used for the assays. c Crude extract of E. coli (10,000×g supernatant) expressing D7OMT was used for the assays. View Large Table 2 Partial purification of HI4′OMT protein from G. echinata cells Purification step  Total protein (mg)  Specific activity (pkat mg–1)  Total activity (pkat)  Relative purity (fold)  Recovery (%)  10,000×g sup.  4.4  3.47  15.3  1  100  (NH4)2SO4 ppt.  2.3  3.87  8.9  1.1  58.2  Chromatofocusing  0.12  59.3  7.1  17.1  46.4  Adenosine-agarose  0.001  820  0.66  236.3  4.3  Purification step  Total protein (mg)  Specific activity (pkat mg–1)  Total activity (pkat)  Relative purity (fold)  Recovery (%)  10,000×g sup.  4.4  3.47  15.3  1  100  (NH4)2SO4 ppt.  2.3  3.87  8.9  1.1  58.2  Chromatofocusing  0.12  59.3  7.1  17.1  46.4  Adenosine-agarose  0.001  820  0.66  236.3  4.3  G. echinata cells (10 g) treated with yeast extract for 24 h (Nakamura et al. 1999) were used for the purification. D7OMT activity was present in both the 10,000×g supernatant (1.0 pkat mg–1) and the ammonium sulfate fraction (2.0 pkat mg–1), but the activity was not detected in any fractions of the chromatofocusing. View Large Abbreviations D7OMT daidzein 7-O-methyltransferase EST expressed sequence tag HI4′OMT 2,7,4′-trihydroxyisoflavanone 4′-O-methyltransferase IFS 2-hydroxyisoflavanone synthase IPTG isopropyl β-d-thiogalactopyranoside OMT O-methyltransferase Rf retardation factor Rt retention time SAM S-adenosyl-l-methionine TLC thin-layer chromatography. The nucleotide sequences reported in this paper have been submitted to the DDBJ, GenBank and EMBL databases under accession numbers: AB091684 (HI4′OMT of G. echinata), AB091685 (D7OMT of G. echinata), AB091686 (HI4′OMT of L. japonicus). References Akashi, T., Aoki, T. and Ayabe, S. ( 1999) Cloning and functional expression of a cytochrome P450 cDNA encoding 2-hydroxyisoflavanone synthase involved in biosynthesis of the isoflavonoid skeleton in licorice. Plant Physiol.  121: 821–828. Google Scholar Akashi, T., Sawada, Y., Aoki, T. and Ayabe, S. ( 2000) New scheme of the biosynthesis of formononetin involving 2,7,4′-trihydroxyisoflavanone but not daidzein as the methyl acceptor. Biosci. Biotechnol. Biochem.  64: 2276–2279. Google Scholar Aoki, T., Akashi, T. and Ayabe, S. ( 2000) Flavonoids of leguminous plants: structure, biological activity, and biosynthesis. J. Plant Res.  113: 475–488. Google Scholar Asamizu, E., Nakamura, Y., Sato, S. and Tabata, S. ( 2000) Generation of 7137 non-redundant expressed sequence tags from a legume, Lotus japonicus. DNA Res.  7: 127–130. Google Scholar Attieh, J.M., Hanson, A.D. and Saini, H.S. ( 1995) Purification and characterization of a novel methyltransferase responsible for biosynthesis of halomethanes and methanethiol in Brassica oleracea. J. Biol. Chem.  270: 9250–9257. Google Scholar Barz, W. and Welle, R. ( 1992) Biosynthesis and metabolism of isoflavones and pterocarpan phytoalexins in chickpea, soybean and phytopathogenic fungi. In Recent Advances in Phytochemistry. Vol. 26. Phenolic Metabolism in Plants. Edited by Stafford, H.A. and Ibrahim, R.K. pp. 139–164, Plenum Press, New York. Google Scholar Burbulis, I.E. and Winkel-Shirley, B. ( 1999) Interactions among enzymes of the Arabidopsis flavonoid biosynthetic pathway. Proc. Natl. Acad. Sci. USA  96: 12929–12934. Google Scholar Clemens, S. and Barz, W. ( 1996) Cytochrome P450-dependent methylenedioxy bridge formation in Cicer arietinum. Phytochemistry  41: 457–460. Google Scholar Crombie, L. and Whiting, D.A. ( 1992) The mechanism of the enzymic induced flavanone-isoflavone change. Tetrahedron Lett.  33: 3663–3666. Google Scholar Dewick, P.M. ( 1993) Isoflavonoids. In The Flavonoids. Advances in Research since 1986 Edited by Harborne, J.B. pp. 117–238, Chapman and Hall, London. Google Scholar Dixon, R.A. ( 1999) Isoflavonoids: biochemistry, molecular biology, and biological functions. In Comprehensive Natural Products Chemistry. Volume 1. Polyketides and other secondary metabolites including fatty acids and their derivatives. Edited by Sankawa, U. pp. 773–823. Elsevier, Amsterdam. Google Scholar Dixon, R.A. ( 2001) Natural products and plant disease resistance. Nature  411: 843–847. Google Scholar Dixon, R.A., Chen, F., He, X.Z., Noel, J.P. and Zubieta, C. ( 2001) Properties and metabolic engineering of alfalfa phenylpropanoid pathway O-methyltransferases. In Recent Advances in Phytochemistry. Vol. 35. Regulation of Phytochemicals by Molecular Techniques. Edited by Romeo, J.T., Saunders, J.A. and Matthews, B.F. pp. 131–154. Pergamon, New York. Google Scholar Dixon, R.A., Lamb, C.J., Masoud, S., Sewalt, V.J.H. and Paiva, N.L. ( 1996) Metabolic engineering prospects for crop improvement through the genetic manipulation of phenylpropanoid biosynthesis and defense responses—a review. Gene  179: 61–71. Google Scholar Edwards, R. and Dixon, R.A. ( 1991) Isoflavone O-methyltransferase activities in elicitor-treated cell suspension cultures of Medicago sativa. Phytochemistry  30: 2597–2606. Google Scholar Frick, S. and Kutchan, T.M. ( 1999) Molecular cloning and functional expression of O-methyltransferases common to isoquinoline alkaloid and phenylpropanoid biosynthesis. Plant J.  17: 329–339. Google Scholar Frick, S., Ounaroon, A. and Kutchan, T.M. ( 2001) Combinatorial biochemistry in plants; the case of O-methyltransferases. Phytochemistry  56: 1–4. Google Scholar Gang, D.R., Lavid, N., Zubieta, C., Chen, F., Beuerle, T., Lewinsohn, E., Noel, J.P. and Pichersky, E. ( 2002) Characterization of phenylpropene O-methyltransferases from sweet basil: facile change of substrate specificity and convergent evolution within a plant O-methyltransferase family. Plant Cell  14: 505–519. Google Scholar Gowri, G., Bugos, R.C., Campbell, W.H., Maxwell, C.A. and Dixon, R.A. ( 1991) Stress responses in alfalfa (Medicago sativa L.) X. Molecular cloning and expression of S-adenosyl-l-methionine: caffeic acid 3-O-methyltransferase, a key enzyme of lignin biosynthesis. Plant Physiol.  97: 7–14. Google Scholar Haga, M., Akashi, T., Aoki, T. and Ayabe, S. ( 1997) A cDNA for S-adenosyl-l-methionine: isoliquiritigenin/licodione 2′-O-methyltransferase (accession no. D88742) from cultured licorice (Glycyrrhiza echinata L.) cells (PGR97-014). Plant Physiol.  113: 663. Google Scholar Hagmann, M. and Grisebach, H. ( 1984) Enzymatic rearrangement of flavanone to isoflavone. FEBS Lett.  175: 199–202. Google Scholar Hashim, M.F., Hakamatsuka, T., Ebizuka, Y. and Sankawa, U. ( 1990) Reaction mechanism of oxidative rearrangement of flavanone in isoflavone biosynthesis. FEBS Lett.  271: 219–222. Google Scholar He, X.Z. and Dixon, R.A. ( 1996) Affinity chromatography, substrate/product specificity, and amino acid sequence analysis of an isoflavone O-methyltransferase from alfalfa (Medicago sativa L.). Arch. Biochem. Biophys.  336: 121–129. Google Scholar He, X.Z. and Dixon, R.A. ( 2000) Genetic manipulation of isoflavone 7-O-methyltransferase enhances biosynthesis of 4′-O-methylated isoflavonoid phytoalexins and disease resistance in alfalfa. Plant Cell  12: 1689–1702. Google Scholar He, X.Z., Reddy, J.T. and Dixon, R.A. ( 1998) Stress responses in alfalfa (Medicago sativa L). XXII. cDNA cloning and characterization of an elicitor-inducible isoflavone 7-O-methyltransferase. Plant Mol. Biol.  36: 43–54. Google Scholar Hrazdina, G. ( 1992) Compartmentation in aromatic metabolism. In Recent Advances in Phytochemistry. Vol. 26. Phenolic Metabolism in Plants. Edited by Stafford, H.A. and Ibrahim, R.K. pp. 1–23, Plenum Press, New York. Google Scholar Ibrahim, R.K., De Luca, V., Khouri, H., Latchinian, L., Brisson, L. and Charest, P.M. ( 1987) Enzymology and compartmentation of polymethylated flavonol glucosides in Chrysosplenium americanum. Phytochemistry  26: 1237–1245. Google Scholar Ibrahim, R.K. and Muzac, I. ( 2000) The methyltransferase gene superfamily: a tree with multiple branches. In Recent Advances in Phytochemistry. Vol. 34. Evolution of Metabolic Pathways. Edited by Romeo, J.T., Ibrahim, R.K., Varin, L. and De Luca, V. pp. 349–384. Pergamon, New York. Google Scholar Jung, W., Yu, O., Lau, S.M., O’Keefe, D.P., Odell, J., Fader, G. and McGonigle, B. ( 2000) Identification and expression of isoflavone synthase, the key enzyme for biosynthesis of isoflavones in legumes. Nat. Biotechnol.  18: 208–212. Google Scholar Kirikae, Y., Sakurai, M., Furuno, T., Takahashi, T. and Ayabe, S. ( 1993) Biosynthesis of a dibenzoylmethane, licodione, in cultured alfalfa cells induced by yeast extract. Biosci. Biotechnol. Biochem.  57: 1353–1354. Google Scholar Kochs, G. and Grisebach, H. ( 1986) Enzymic synthesis of isoflavones. Eur. J. Biochem.  155: 311–318. Google Scholar Liu, C.J. and Dixon, R.A. ( 2001) Elicitor-induced association of isoflavone O-methyltransferase with endomembranes prevents the formation and 7-O-methylation of daidzein during isoflavonoid phytoalexin biosynthesis. Plant Cell  13: 2643–2658. Google Scholar Mathesius, U. ( 2001) Flavonoids induced in cells undergoing nodule organogenesis in white clover are regulators of auxin breakdown by peroxidase. J. Exp. Bot.  52: 419–426. Google Scholar Nakamura, K., Akashi, T., Aoki, T., Kawaguchi, K. and Ayabe, S. ( 1999) Induction of isoflavonoid and retrochalcone branches of the flavonoid pathway in cultured Glycyrrhiza echinata cells treated with yeast extract. Biosci. Biotechnol. Biochem.  63: 1618–1620. Google Scholar Page, R.D. ( 1996) TreeView: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci.  12: 357–358. Google Scholar Phillips, D.A. ( 1992) Flavonoids: plant signals to soil microbes. In Recent Advances in Phytochemistry. Vol. 26. Phenolic Metabolism in Plants. Edited by Stafford, H.A. and Ibrahim, R.K. pp. 201–231. Plenum Press, New York. Google Scholar Preisig, C.L., Matthews, D.E. and Van Etten, H.D. ( 1989) Purification and characterization of S-adenosyl-l-methionine: 6a-hydroxymaackiain 3-O-methyltransferase from Pisum sativum. Plant Physiol.  91: 559–566. Google Scholar Rakwal, R., Agrawal, G.K., Yonekura, M. and Kodama, O. ( 2000) Naringenin 7-O-methyltransferase involved in the biosynthesis of the flavanone phytoalexin sakuranetin from rice (Oryza sativa L.). Plant Sci.  29: 213–221. Google Scholar Saslowsky, D. and Winkel-Shirley, B. ( 2001) Localization of flavonoid enzymes in Arabidopsis roots. Plant J.  27: 37–48. Google Scholar Sawada, Y., Kinoshita, K., Akashi, T., Aoki, T. and Ayabe, S. ( 2002) Key amino acid residues required for aryl migration catalyzed by the cytochrome P450 2-hydroxyisoflavanone synthase. Plant J.  31: 555–564. Google Scholar Schröder, G., Wehinger, E. and Schröder, J. ( 2002) Predicting the substrates of cloned plant O-methyltransferases. Phytochemistry  59: 1–8. Google Scholar Shimada, N., Akashi, T., Aoki, T. and Ayabe, S. ( 2000) Induction of isoflavonoid pathway in the model legume Lotus japonicus: molecular characterization of enzymes involved in phytoalexin biosynthesis. Plant Sci.  160: 37–47. Google Scholar Shutt, D.A. ( 1976) The effects of plant oestrogens on animal reproduction. Endeavour  75: 110–113. Google Scholar Spaink, H.P. ( 1995) The molecular basis of infection and nodulation by rhizobia: the ins and outs of sympathogenesis. Annu. Rev. Phytopathol.  33: 345–368. Google Scholar Stavric, B. ( 1997) Chemopreventive agents in foods. In Recent Advances in Phytochemistry. Vol. 31. Functionality of Food Phytochemicals. Edited by Johns, T. and Romeo, J.T. pp. 53–87. Plenum Press, New York. Google Scholar Steele, C.L., Gijzen, M., Qutob, D. and Dixon, R.A. ( 1999) Molecular characterization of the enzyme catalyzing the aryl migration reaction of isoflavonoid biosynthesis in soybean. Arch. Biochem. Biophys.  367: 146–150. Google Scholar Tahara, S. and Ibrahim, R.K. ( 1995) Prenylated isoflavonoids—an update. Phytochemistry  38: 1073–1094. Google Scholar Thompson, J.D., Higgins, D.G. and Gibson, T.J. ( 1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucl. Acids Res.  22: 4673–4680. Google Scholar Wengenmayer, H., Ebel, J. and Grisebach, H. ( 1974) Purification and properties of a S-adenosylmethionine: isoflavone 4′-O-methyltransferase from cell suspension cultures of Cicer arietinum L. Eur. J. Biochem.  50: 135–143. Google Scholar Winkel-Shirley, B. ( 2001) Flavonoid biosynthesis. A colorful model for genetics, biochemistry, cell biology, and biotechnology. Plant Physiol.  126: 485–493. Google Scholar Wu, Q., Preisig, C.L. and VanEtten, H.D. ( 1997) Isolation of the cDNAs encoding (+)6a-hydroxymaackiain 3-O-methyltransferase, the terminal step for the synthesis of the phytoalexin pisatin in Pisum sativum. Plant Mol. Biol.  35: 551–560. Google Scholar Zubieta, C., He, X.Z., Dixon, R.A. and Noel, J.P. ( 2001) Structures of two natural product methyltransferases reveal the basis for substrate specificity in plant O-methyltransferases. Nat. Struct. Biol.  8: 271–279. Google Scholar http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Plant and Cell Physiology Oxford University Press

cDNA Cloning and Biochemical Characterization of S-Adenosyl-l-Methionine: 2,7,4′-Trihydroxyisoflavanone 4′-O-Methyltransferase, a Critical Enzyme of the Legume Isoflavonoid Phytoalexin Pathway

Download PDF
10 pages

Loading next page...
 
/lp/oxford-university-press/cdna-cloning-and-biochemical-characterization-of-s-adenosyl-l-RjUkIBiXcW

References (58)

Publisher
Oxford University Press
ISSN
0032-0781
eISSN
1471-9053
DOI
10.1093/pcp/pcg034
Publisher site
See Article on Publisher Site

Abstract

Abstract Formononetin (7-hydroxy-4′-methoxyisoflavone, also known as 4′-O-methyldaidzein) is an essential intermediate of ecophysiologically active leguminous isoflavonoids. The biosynthetic pathway to produce 4′-methoxyl of formononetin has been unknown because the methyl transfer from S-adenosyl-l-methionine (SAM) to 4′-hydroxyl of daidzein has never been detected in any plants. A hypothesis that SAM: daidzein 7-O-methyltransferase (D7OMT), an enzyme with a different regiospecificity, is involved in formononetin biosynthesis through its intracellular compartmentation with other enzymes recently prevails, but no direct evidence has been presented. We proposed a new scheme of formononetin biosynthesis involving 2,7,4′-trihydroxyisoflavanone as the methyl acceptor and subsequent dehydration. We now cloned a cDNA encoding SAM: 2,7,4′-trihydroxyisoflavanone 4′-O-methyltransferase (HI4′OMT) through the screening of functionally expressed Glycyrrhiza echinata (Fabaceae) cDNAs. The reaction product, 2,7-dihydroxy-4′-methoxyisoflavanone, was unambiguously identified. Recombinant G.echinata D7OMT did not show HI4′OMT activity, and G. echinata HI4′OMT protein free from D7OMT was partially purified. HI4′OMT is thus concluded to be distinct from D7OMT, and their distant phylogenetic relationship was further presented. HI4′OMT may be functionally identical to (+)-6a-hydroxymaackiain 3-OMT of pea. Homologous cDNAs were found in several legumes, and the catalytic function of the Lotus japonicus HI4′OMT was verified, indicating that HI4′OMT is the enzyme of formononetin biosynthesis in general legumes. (Received September 26, 2002; Accepted December 25, 2002) Introduction Isoflavonoids are distributed mainly in the leguminous plants, and have critical physiological functions in the interaction with environmental microorganisms (Dixon 1999, Aoki et al. 2000). They are the most abundant antimicrobial phytoalexins (Dewick 1993), while they act as host signal molecules to the rhizobial bacteria (Phillips 1992, Spaink 1995) and the putative regulator of local auxin levels for the nodule organogenesis (Mathesius 2001) during symbiotic nitrogen fixation. The self-defense roles of isoflavonoids are also represented by their deterrent activity against insect feeding (Dewick 1993) and phytoestrogenic activity in higher animals, which causes infertility in sheep (Shutt 1976), although the estrogen activity in the diet is beneficial to human health (Stavric 1997). A methoxy group at C-4′ and a 3′,4′-methylenedioxy group are additional characteristic structural features of many isoflavonoids (Tahara and Ibrahim 1995, Clemens and Barz 1996): ca. 50% of naturally occurring isoflavonoids are 4′-methoxylated or 3′,4′-methylenedioxylated (Dewick 1993). The isoflavonoids possessing these substitutions constitute the major phytoalexins, e.g. medicarpin of alfalfa, pisatin of pea and vestitol of Lotus japonicus (Dixon 1999). These compounds originate from an isoflavone, formononetin (7-hydroxy-4′-methoxyisoflavone, also known as 4′-O-methyldaidzein) (Fig. 1). The biosynthetic process of the introduction of 4′-methoxyl to the precursor(s) to produce formononetin has, however, not been known. The methyl transfer from S-adenosyl-l-methionine (SAM) to 4′-hydroxyl of daidzein has never been detected in any plants that produce formononetin (Fig. 1). The O-methyltransferase (OMT) toward daidzein so far characterized produces only the 7-O-methylated isoflavone isoformononetin (4′-hydroxy-7-methoxyisoflavone) (Wengenmayer et al. 1974, Hagmann and Grisebach 1984, Edwards and Dixon 1991, Barz and Welle 1992, He and Dixon 1996, He et al. 1998), but the 7-O-methylated isoflavonoids are seldom plant constituents. Characterization of the 4′-O-methylation step is important both for the understanding of the biosynthetic mechanism of ecophysiologically active isoflavonoids and from the viewpoint of engineering a potential branching point in isoflavonoid synthesis to either methylated phytoalexins or the health-promoting phytoestrogens of soybean. The isoflavonoid skeleton is produced from a flavanone with 4′-hydroxyl through the action of a cytochrome P450 (P450), 2-hydroxyisoflavanone synthase (IFS), that catalyzes the hydroxylation at C-2 accompanied by the 1,2-aryl migration from C-2 to C-3 (Fig. 1) (Kochs and Grisebach 1986, Hashim et al. 1990). Then, the acid-labile IFS product is dehydrated to form isoflavone. An interesting hypothesis that the methyl transfer to 4′-hydroxyl is integrated into the aryl migration through the formation of 4′-methoxylated spiro-dienone intermediate was proposed (Crombie and Whiting 1992). However, the recent cloning of IFS cDNA (Akashi et al. 1999, Steele et al. 1999, Jung et al. 2000, Shimada et al. 2000) and the detailed biochemical study of the reaction (Akashi et al. 1999, Sawada et al. 2002) have shown that 2,7,4′-trihydroxyisoflavanone without 4′-O-methyl is the product from a flavanone, liquiritigenin (7,4′-dihydroxyflavanone). On the other hand, the hypothesis that a compartmentation of daidzein 7-OMT (D7OMT) in subcellular sites causes a change in the regiospecificity of the enzyme to produce formononetin was proposed (Dixon et al. 1996, Dixon 1999). This enzyme has been referred to as isoflavone OMT (IOMT) (He and Dixon 1996, He et al. 1998), but the name D7OMT more clearly indicating the substrate specificity and regiospecificity is used in the present article. This interesting hypothesis has been supported by the observations that transgenic alfalfa overexpressing D7OMT displayed enhanced production of 4′-O-methylated isoflavonoids on elicitation and that D7OMT fused with green fluorescent protein is located on the endoplasmic reticulum of alfalfa where the P450s are co-located (He and Dixon 2000, Dixon et al. 2001, Dixon 2001, Liu and Dixon 2001). However, no direct biochemical evidence through the in vitro demonstration of 4′-OMT activity has been provided. Cultured Glycyrrhiza echinata L. (Fabaceae) cells produce medicarpin by treatment with an elicitor, and constitutively accumulate a large quantity of formononetin (Nakamura et al. 1999). Recently, we demonstrated for the first time the in vitro activity of formononetin production from 2,7,4′-trihydroxyisoflavanone and SAM in the crude extract of G. echinata (Akashi et al. 2000). From this observation, which was made possible only by the use of a sufficient quantity of acid-labile 2-hydroxyisoflavanone substrate prepared by the heterologously expressed IFS, we suggested a new scheme of formononetin biosynthesis in which the substrate of the OMT is 2,7,4′-trihydroxyisoflavanone rather than daidzein and 4′-O-methylated 2-hydroxyisoflavanone subsequently undergoes another enzymatic dehydration to form formononetin (Fig. 1). We report here the molecular cloning of a cDNA encoding 2,7,4′-trihydroxyisoflavanone 4′-OMT (HI4′OMT) of G. echinata using a functional expression cloning method. The direct product of the OMT reaction, 2,7-dihydroxy-4′-methoxyisoflavanone, was isolated and unambiguously identified. Homologous cDNAs were found in other legumes, and the catalytic function of L. japonicus HI4′OMT was also verified. HI4′OMT and D7OMT are clearly distinct proteins encoded by different genes. These results provide the conclusive solution to the long-standing problem of isoflavonoid biosynthesis. Results Functional expression cloning of G. echinata HI4′OMT cDNA The formononetin-forming activity from 2,7,4′-trihydroxyisoflavanone and SAM in the G. echinata cells was activated by elicitation with yeast extract (Akashi et al. 2000). For the cloning of cDNA encoding HI4′OMT, G. echinata cDNAs were functionally expressed in Escherichia coli cells, and an E. coli clone displaying 2,7,4′-trihydroxyisoflavanone OMT activity was isolated as follows. λZapII cDNA library of elicited G. echinata was converted to the phagemid form by in vivo excision, and phagemids were then introduced into E. coli (Fig. 2). Five E. coli pools (ca. 10,000 clones pool–1) were prepared, and recombinant proteins in these pools were induced by isopropyl β-d-thiogalactopyranoside (IPTG). When crude extracts of the E. coli pools were reacted with 2,7,4′-trihydroxyisoflavanone and [14C]SAM, a new radioactive compound was produced in four pools. This product was assumed to be [14C]-2,7-dihydroxy-4′-methoxyisoflavanone because it was easily converted to [14C]formononetin by acid treatment. One pool was arbitrarily selected from the positive pools and subdivided into ten pools of smaller size (ca. 1,000 clones pool–1) for the second round of screening. Expression of proteins and the assay were performed again, and nine positive pools were identified. Subdivision of a randomly selected positive pool into ten pools of 1/10 size was performed, and the assay was repeated. Four positive pools (ca. 100 clones pool–1) and one positive pool (10 clones pool–1) were obtained from the third and the fourth screenings, respectively. Finally, a clone that showed 2,7,4′-trihydroxyisoflavanone OMT activity was isolated. The cDNA of the enzyme contained 1,349 bp nucleotides and encoded a polypeptide of 367 amino acid residues. The cDNA was named HI4′OMT. The deduced amino acid sequence of HI4′OMT had the conserved sequence motifs recorded for SAM-dependent OMTs of higher plants (Fig. 3) (Ibrahim and Muzac 2000, Schröder et al. 2002). HI4′OMT shared the highest identity (83%) with (+)-6a-hydroxymaackiain 3-OMT (accession no. U69554) of pea (Wu et al. 1997), 50% identity with D7OMT (IOMT) (accession no. U97125) of alfalfa (He et al. 1998), 30% identity with isoliquiritigenin/licodione 2′-OMT (accession no. D88742) of G. echinata (Haga et al. 1997) and 31% identity with caffeic acid 3-OMT (accession no. M63853) of alfalfa (Gowri et al. 1991). Product identification and the substrate specificity of HI4′OMT of G. echinata The product from the incubation of the recombinant HI4′OMT linking six histidine residues at the N-terminus with non-labeled SAM and 2,7,4′-trihydroxyisoflavanone was recovered and chemico-physically analyzed. It gave a single reverse-phase HPLC peak at retention time (Rt) 17.3 min (Fig. 4A, Chart 1) and the same Rf on thin-layer chromatography (TLC) as that of the product of radiolabeled assay during the functional expression cloning. It was readily converted to formononetin (Rt 37.5 min) by acid treatment (Fig. 4A, Chart 2). The UV spectrum of the product (λmax 279 nm and a shoulder at 315 nm) coincided well with the spectrum of 2,7,4′-trihydroxyisoflavanone (λmax 275 nm and a shoulder at 313 nm), indicating that it has a 2-hydroxyisoflavanone skeleton. The electron-impact mass spectrum of the compound exhibited a molecular ion peak at m/z 286 (C16H14O5), [M-H2O]+ peak at m/z 268 and also a retro-Diels-Alder fragment (C9H8O) peak at m/z 132 matching the O-methylated B-ring (Fig. 4B). The chemical structure of the direct reaction product was thus unambiguously identified as 2,7-dihydroxy-4′-methoxyisoflavanone, establishing the enzyme regiospecificity to be 4′-hydroxyl of 2,7,4′-trihydroxyisoflavanone. Table 1 summarizes the specific activity of HI4′OMT protein toward candidates for the isoflavonoid substrate. Daidzein was incubated with the recombinant HI4′OMT and [14C]SAM, but no reaction product was detected, indicating that HI4′OMT is distinct from D7OMT. When (±)-medicarpin (3-hydroxy-9-methoxypterocarpan), an analog of 6a-hydroxymaackiain, was reacted with the recombinant HI4′OMT and [14C]SAM, a radioactive compound assumed to be [14C]-3,9-dimethoxypterocarpan was produced. The specific activity of HI4′OMT to (±)-medicarpin (14 pkat mg–1) is in the same order as that of (+)-6a-hydroxymaackiain 3-OMT of pea to (+)-medicarpin (ca. 70–100 pkat mg–1; Wu et al. 1997), indicating very similar catalytic functions of these enzymes. cDNA cloning and biochemical characterization of G. echinata D7OMT To examine the 4′-O-transmethylation reaction to 2,7,4′-trihydroxyisoflavanone from SAM by D7OMT protein, a cDNA homologous to alfalfa D7OMT was screened from the G. echinata cDNA library using alfalfa D7OMT as a probe. The deduced amino acid sequence of G. echinata D7OMT showed 78% identity with alfalfa D7OMT (IOMT) and 50% identity with G. echinata HI4′OMT. As shown in Table 1, a crude extract of E. coli expressing D7OMT of G. echinata mediated the formation of [14C]isoformononetin from daidzein and [14C]SAM. However, neither HI4′OMT nor medicarpin OMT activities were detected. Partial purification of HI4′OMT and separation of HI4′OMT and D7OMT from G. echinata cells Partial purification of G. echinata HI4′OMT protein free from D7OMT was accomplished. The purification procedure is summarized in Table 2. D7OMT activity was observed in the crude extract (10,000×g supernatant) and ammonium sulfate fraction, but the activity was not recovered in the chromatofocusing step. The affinity-purified protein catalyzed SAM-dependent 4′-O-methylation of 2,7,4′-trihydroxyisoflavanone to produce 2,7-dihydroxy-4′-methoxyisoflavanone, but did not employ daidzein as the substrate, clearly indicating the distinct nature of HI4′OMT and D7OMT proteins. The specific activity (820 pkat mg–1) for 2,7,4′-trihydroxyisoflavanone was roughly in the same order as that of recombinant HI4′OMT (Table 1, 2). The purified protein, like recombinant HI4′OMT, displayed the OMT activity toward (±)-medicarpin in addition to 2,7,4′-trihydroxyisoflavanone (data not shown). Further, the protein eluted like the recombinant HI4′OMT at pH 4.8 on a chromatofocusing column. The biochemical properties of the purified protein thus coincided well with those of the recombinant HI4′OMT. Distribution of HI4′OMT in leguminous plants HI4′OMT-like sequences (>80% identity with that of G. echinata at the amino acid level) were found in the expressed sequence tag (EST) databases of leguminous plants, Lotus japonicus (accession no. AV407445, http://www.kazusa.or.jp/en/plant/lotus/EST/) (Asamizu et al. 2000), Medicago truncatula (accession no. TC28631, http://www.tigr.org/tdb/mtgi/) and Glycine max (accession no. TC101829, http://tigrblast.tigr.org/tgi/). The seedling of L. japonicus accumulates phytoalexin vestitol after treatment with reduced-glutathione as an elicitor, and we recently cloned and examined the elicitor responses of cDNAs encoding the enzymes of the pathway (Shimada et al. 2000). The L. japonicus EST clone AV407445, which has the predicted initiation codon, was obtained from Kazusa DNA Research Institute, and the full-length HI4′OMT cDNA was sequenced. It had 1,349 bp nucleotides, and the deduced amino acid sequence shared the highest identity (83%) with G. echinata HI4′OMT (Fig. 3). Crude extract of E. coli expressing L. japonicus HI4′OMT catalyzed the formation of [14C]-2,7-dihydroxy-4′-methoxyisoflavanone from 2,7,4′-trihydroxyisoflavanone and [14C]SAM. Like G. echinata HI4′OMT, no reaction product was formed by the incubation of the HI4′OMT with daidzein and [14C]SAM. Thus, L. japonicus HI4′OMT is functionally the same as G. echinata HI4′OMT. Discussion The hypothesis that D7OMT is the enzyme of formononetin biosynthesis was proposed by Dixon and his co-workers using alfalfa as the experimental system (Dixon et al. 1996, Dixon 1999). Initially, D7OMT contained in the subcellular compartment was hypothesized to catalyze the methyl transfer to the 4′-hydroxyl of daidzein. Recently, indirect biochemical evidence incorporating 2,7,4′-trihydroxyisoflavanone in, but excluding daidzein from, the intermediate of formononetin biosynthesis in alfalfa was presented (Liu and Dixon 2001). Still, however, in the alfalfa system, the HI4′OMT activity is attributed to D7OMT protein (Liu and Dixon 2001), supported by a three-dimensional crystal structure of D7OMT that can bind 2,7,4′-trihydroxyisoflavanone on a computer-generated illustration (Zubieta et al. 2001). Therefore, the major remaining problem was to determine whether HI4′OMT protein is identical to or distinct from D7OMT. In this study, a cDNA encoding HI4′OMT was successfully cloned from G. echinata cells. We clearly demonstrated that the G. echinata HI4′OMT is distinct from D7OMT (Table 1). Partial purification of HI4′OMT free from D7OMT further supported the distinct nature of these proteins. The biochemical evidence suggested that the purified protein is the product of HI4′OMT. In addition, transient accumulations of HI4′OMT mRNA in both elicited G. echinata cells and L. japonicus seedlings prior to the induction of formononetin-producing activity (G. echinata) (Akashi et al. 2000) and vestitol accumulation (L. japonicus) (Shimada et al. 2000) confirmed its involvement in formononetin biosynthesis (data not shown). We thus clarified here the missing link of isoflavonoid biosynthesis that has been a mystery for nearly three decades (Wengenmayer et al. 1974, Hagmann and Grisebach 1984, Barz and Welle 1992). Interestingly, HI4′OMT showed the highest amino acid identity (83%) with (+)-6a-hydroxymaackiain 3-OMT of pea (Preisig et al. 1989), the final enzyme in the biosynthesis of the phytoalexin (+)-pisatin, acting on the hydroxyl at the C-3 of (+)-pterocarpan skeleton that corresponds to C-7 in isoflavone numbering (Wu et al. 1997). G. echinata HI4′OMT is active on a pterocarpan, (±)-medicarpin, which has only one hydroxyl at C-3 as the methyl acceptor (Table 1), and (+)-6a-hydroxymaackiain 3-OMT was reported to be also active with (+)-medicarpin (Wu et al. 1997). It is thus reasonably assumed that pea (+)-6a-hydroxymaackiain 3-OMT is functionally identical to HI4′OMT. L. japonicus HI4′OMT was also functionally identified in this study, and HI4′OMT-like sequences were further found in soybean and M. truncatula. Because none of these plants except pea are known to produce pisatin, this family of proteins should primarily act as HI4′OMT. A close phylogenetic relationship between alfalfa D7OMT (IOMT) and pea (+)-6a-hydroxymaackiain 3-OMT among the higher plant OMTs has been demonstrated (Schröder et al. 2002, Gang et al. 2002). Now, the phylogenic tree composed of OMTs of the flavonoid pathway indicates that HI4′OMT and (+)-6a-hydroxymaackiain 3-OMT belong to the same branch, which is different from that of D7OMT (Fig. 5). Recently, a successful homology modeling, based on the D7OMT stereostructure, of the active sites of phenylpropane OMTs that are more distantly related than HI4′OMT to D7OMT, was reported (Gang et al. 2002). In a hypothetical binding of 2,7,4′-trihydroxyisoflavanone into the active site of D7OMT, the best fit was observed with (2S, 3R)-2,7,4′-trihydroxyisoflavanone among the four possible stereoisomers (Zubieta et al. 2001). In contrast, from the homology model of CYP93C2 protein (G. echinata IFS), the direct IFS reaction product from (2S)-liquiritigenin is predicted to be the opposite enantiomer (2R, 3S)-2,7,4′-trihydroxyisoflavanone (Sawada et al. 2002). Whether this isomer and the (+)-pterocarpan molecule can be accommodated in the active site of HI4′OMT protein in conformations favorable for the respective 4′- and 3-O-methylation should be examined by homology modeling and, in the future, from the crystallography data of HI4′OMT. While this study makes the hypothesis that D7OMT in a subcellular compartment is involved in formononetin biosynthesis unnecessary, the localization of HI4′OMT in the cell and its possible interaction with other enzymes of the (iso)flavonoid pathway are interesting. Although HI4′OMT as well as D7OMT are almost exclusively recovered in the soluble fraction (160,000×g supernatant) of G. echinata (data not shown), association of ‘soluble’ enzymes of flavonoid biosynthesis (Hrazdina 1992, Ibrahim et al. 1987) including alfalfa D7OMT (Dixon et al. 1996, Liu and Dixon 2001) with endoplasmic reticulum has been repeatedly described. In addition, interactions among the enzymes of flavonoid biosynthesis have been demonstrated in Arabidopsis thaliana (Burbulis and Winkel-Shirley 1999, Saslowsky and Winkel-Shirley 2001, Winkel-Shirley 2001). Because 2-hydroxyisoflavanones are not stable, 4′-O-methylation of 2,7,4′-trihydroxyisoflavanone should take place prior to the non-enzymatic dehydration in the cells, favoring the compartmentation of IFS, HI4′OMT, specific dehydratase and possibly P450 reductase on the endoplasmic reticulum membrane. The intrinsic activity of D7OMT remains to be investigated. The induction of D7OMT activity in elicited G. echinata cells (Akashi et al. 2000) in addition to the coincidence of D7OMT expression and phytoalexin production in alfalfa (He et al. 1998, He and Dixon 2000) strongly suggests its involvement in the host defense responses. Broad and flexible substrate specificities of plant OMTs sharing similar sequences (Ibrahim and Muzac 2000, Frick et al. 2001) and even changes of specificities through the formation of heterodimers of similar OMT proteins have been reported (Frick and Kutchan 1999). Indirect or regulatory roles for D7OMT in formononetin biosynthesis can be envisaged. Finally, the functional expression cloning used in this study is a rapid and convenient method when a high sensitivity assay, e.g. by the use of radioisotope-labeled substrates, is available. To our knowledge, this is the first successful cloning of a cDNA encoding the enzyme of plant-specific (secondary) metabolism using this method. In preliminary experiments, D7OMT, chalcone synthase and caffeic acid 3-OMT activities were also detected in the cDNA expression library of G. echinata (data not shown). This method will be a useful tool for the isolation of cDNA encoding enzymes in the diverse and complicated plant secondary metabolic pathway. To examine the possible interaction with enzymes of the formononetin pathway, cloning of the cDNA encoding 2,7-dihydroxy-4′-methoxyisoflavanone dehydratase is underway. Materials and Methods Buffers The following buffers were used: A, 100 mM potassium phosphate (pH 7.5) containing 10% (w/v) sucrose and 14 mM 2-mercaptoethanol; B, 25 mM histidine-HCl containing 1.4 mM 2-mercaptoethanol (pH 6.0); C, polybuffer 74 (Amersham Biosciences, Buckinghamshire, U.K.)-HCl with 1.4 mM 2-mercaptoethanol (pH 4.0); D, 50 mM Tris-HCl (pH 7.5) containing 14 mM 2-mercaptoethanol. Assays 2,7,4′-Trihydroxyisoflavanone was prepared by the incubation of the yeast microsome expressing IFS (CYP93C2) with (RS)-liquiritigenin and NADPH (Akashi et al. 1999). 2,7,4′-Trihydroxyisoflavanone (ca. 0.4 nmol) dissolved in 2 µl 2-methoxyethanol was incubated with an enzyme preparation of E. coli or G. echinata in the presence of 0.4 nmol S-adenosyl-l-[methyl-14C]methionine ([14C]SAM, 2.26 GBq mmol–1, Amersham Biosciences) in the total volume of 0.2 ml at 30°C for 10 min. After termination of the reaction with 5 µl acetic acid, the ethyl acetate extract of the mixture was subjected to silica-gel TLC [LK6DF (Whatman, Maidstone, U.K.); solvent, chloroform : acetone : 25% aq. ammonia solution = 70 : 29 : 1; 2,7,4′-trihydroxyisoflavanone (Rf 0.08), 2,7-dihydroxy-4′-methoxyisoflavanone (Rf 0.15), formononetin (Rf 0.30), isoformononetin (Rf 0.62)] and analyzed by a Typhoon 8600 image analyzer (Amersham Biosciences). Acid-catalyzed conversion of 2-hydroxyisoflavanone into isoflavone was performed as described (Akashi et al. 2000). Specific activity was determined by the incubation of the purified recombinant HI4′OMT (ca. 15 ng protein) or crude extract of E. coli expressing D7OMT (ca. 10 µg protein) with 2,7,4′-trihydroxyisoflavanone (ca. 0.8 nmol) and [14C]SAM (0.4 nmol). Incubation time was 6 min or shorter when the reaction rate was linear, and the reaction mixture was subjected to silica-gel TLC as above. [14C]-2,7-Dihydroxy-4′-methoxyisoflavanone was separated from TLC, and radioactivity was measured by liquid scintillation counting. D7OMT and medicarpin OMT were assayed by the same method as HI4′OMT using 10 nmol daidzein (Extrasynthése, Genay, France) or 0.5 nmol medicarpin (Plantech, Reading, U.K.). Prolonged incubation (20 min) was carried out because of low activity. The TLC condition for the medicarpin OMT product was: plate, LK6DF; solvent, toluene : ethyl acetate = 4 : 1; Rfs, medicarpin (0.58) and 3,9-dimethoxypterocarpan (0.86). For the assay with non-labeled substrates, the reaction mixture (2 ml) contained 1 µmol SAM, 200 nmol 2,7,4′-trihydroxyisoflavanone and the purified recombinant HI4′OMT (ca. 50 µg). HPLC was performed using a CAPCELL PAK C18 MG column (4.6×150 mm; Shiseido, Tokyo, Japan) at 40°C with a flow rate of 0.8 ml min–1 with a linear gradient elution for 35 min from 35% to 55% methanol in 3% aq. acetic acid. The eluate was monitored by a multiwavelength detector (MD-2010, Jasco, Tokyo, Japan). For mass spectrometric identification of the product, the ethyl acetate extract of the reaction mixture was applied to a silica-gel TLC [Kieselgel F254 (Merck, Darmstadt, Germany); solvent, toluene : ethyl acetate : methanol : light petroleum (6 : 4 : 1 : 3)], and a product (Rf 0.30) was collected and further purified by HPLC. The electron impact mass spectrum was recorded on a JEOL SX-102A mass spectrometer at the ionization voltage of 70 eV. Cloning of HI4′OMT cDNA from G. echinata cells In vivo excision of G. echinata λZapII cDNA library (Akashi et al. 1999) was performed using E. coli DH5α F′IQ (Invitrogen) with ExAssist helper phage. The excised phagemids were introduced into DH5α F′IQ, and the E. coli cells were grown on a Luria-Bertani (LB)/ampicillin (50 µg ml–1) agar plate. About 10,000 E. coli transformants were scraped from a plate and inoculated in 20 ml of LB/ampicillin medium in a 50 ml conical tube. The fraction was considered as one pool, and five independent pools were prepared. E. coli pools were incubated at 30°C until OD600 = 0.6. A small amount of the culture from each pool was placed on an LB/ampicillin agar plate for the second round screening. IPTG was added to the remaining pools to a final concentration of 5 mM, and the cultures were incubated for 12 h at 30°C. The medium was removed by centrifugation, and E. coli cells suspended in buffer A. The crude extract was prepared by a vigorous shaking with glass beads and centrifugation (10,000×g, 10 min), and HI4′OMT activity was assayed. A positive pool was arbitrarily selected and subdivided into 10 pools of smaller size (ca. 1,000 clones pool–1) for the next round screening. The fractionation of a positive pool and the assays were repeated, and a transformant showing 2,7,4′-trihydroxyisoflavanone OMT activity was isolated. Cloning of a cDNA homologous to alfalfa D7OMT from G. echinata cells D7OMT cDNA was cloned from alfalfa cells (Kirikae et al. 1993) by reverse transcription-PCR with specific primers prepared from the nucleotide sequence information of IOMT8 cDNA (He et al. 1998). Plaques (2×105) of the G. echinata λZapII cDNA library were screened with alfalfa D7OMT using an ECL Direct Nucleic Acid Labeling System (Amersham Biosciences). Positive clones were converted to pBluescript SK(–) by in vivo excision. The nucleotide sequence of a cDNA with a length of ca. 1,400 bp was determined. Heterologous expression of HI4′OMT and D7OMT in E. coli Coding regions of HI4′OMT and D7OMT of G. echinata were amplified by PCR with KOD polymerase (Toyobo, Tokyo, Japan) and cDNA clones as templates. NdeI and EcoRI sites were introduced upstream of the initiation codon and downstream of the stop codon of HI4′OMT, respectively. For D7OMT, the NheI site upstream of the initiation codon and HindIII sites downstream of the stop codon were introduced. The NdeI–EcoRI fragment of the PCR product from HI4′OMT was cloned into pET28a (Novagen) to express HI4′OMT as the fusion protein with six histidine residues. The NheI–HindIII fragment of the PCR product of D7OMT was cloned into pET21a to express D7OMT as non-tagged protein. E. coli BL21 (DE3) cells transformed with each vector were cultivated at 30°C in 200 ml LB medium supplemented with 50 µg ml–1 kanamycin or 50 µg ml–1 ampicillin until OD600 = 0.4. IPTG was added to a final concentration of 1 mM, and the cultures were incubated for 6 h at 30°C. HI4′OMT was purified using HisTrap Kit (Amersham Biosciences) from the crude extract of E. coli expressing HI4′OMT. An EST clone (accession no. AV407445) of L. japonicus accession ‘Miyakojima’ (Asamizu et al. 2000) was transformed in E. coli DH5α, and the protein was expressed by the addition of 5 mM IPTG and incubation for 12 h at 30°C. Purification of HI4′OMT from G. echinata cells From the cell-free extract of cultured G. echinata cells (Ak-1 line, 10 g) elicited with 0.2% (w/v) yeast extract (Difco, MI, U.S.A.) for 24 h (Nakamura et al. 1999), ammonium sulfate (30–80% saturation) precipitate was prepared as described (Akashi et al. 2000). The precipitate was dissolved in buffer B and desalted on a Sephadex G-25 column (Amersham Biosciences). The sample was subjected to chromatofocusing on a PBE94 column (5 ml, 6×200 mm, Amersham Biosciences) which had been equilibrated with buffer B. Elution was achieved with buffer C. The positive fraction was applied onto an adenosine-agarose gel (2 ml, 10×30 mm) (Attieh et al. 1995, Rakwal et al. 2000) that had been equilibrated with buffer D. The column was washed with buffer D containing 2 M NaCl. The protein was eluted with buffer D containing 4 mM of SAM, and the buffer was changed to buffer A by an ultrafiltration (Microcon YM-10, Millipore). Acknowledgments This work was supported by a Grant-in-Aid for Encouragement of Young Scientists (B) (no. 13780474) from the Japan Society for the Promotion of Science, and by a Grant-in-Aid for Scientific Research on Priority Areas (A) (no. 1302470) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan. EST clone (accession no. AV407445) of L. japonicus was obtained from Kazusa DNA Research Institute. 1 Corresponding author: E-mail, ayabe@brs.nihon-u.ac.jp; Fax, +81-466-80-1141. View largeDownload slide Fig. 1 Biosynthesis of O-methylated isoflavonoids. The open arrow indicates the pathway described in this study. The dashed arrow indicates the biosynthetic pathway that is not detected in G. echinata cells. Abbreviations used are: D7OMT, daidzein 7-OMT; IFS, 2-hydroxyisoflavanone synthase; IOMT, isoflavone OMT; HI4′OMT, 2,7,4′-trihydroxyisoflavanone 4′-OMT. View largeDownload slide Fig. 1 Biosynthesis of O-methylated isoflavonoids. The open arrow indicates the pathway described in this study. The dashed arrow indicates the biosynthetic pathway that is not detected in G. echinata cells. Abbreviations used are: D7OMT, daidzein 7-OMT; IFS, 2-hydroxyisoflavanone synthase; IOMT, isoflavone OMT; HI4′OMT, 2,7,4′-trihydroxyisoflavanone 4′-OMT. View largeDownload slide Fig. 2 Method of cloning G. echinata HI4′OMT cDNA. A λZapII cDNA library of G. echinata was converted to phagemid form to prepare the cDNA expression library. Recombinant proteins in E. coli pools were induced by IPTG, and HI4′OMT was assayed to identify the positive pools. After several rounds of screening, a single clone was isolated. The stars indicate putative clones of HI4′OMT. The gray test tubes contain positive pools. View largeDownload slide Fig. 2 Method of cloning G. echinata HI4′OMT cDNA. A λZapII cDNA library of G. echinata was converted to phagemid form to prepare the cDNA expression library. Recombinant proteins in E. coli pools were induced by IPTG, and HI4′OMT was assayed to identify the positive pools. After several rounds of screening, a single clone was isolated. The stars indicate putative clones of HI4′OMT. The gray test tubes contain positive pools. View largeDownload slide Fig. 3 Amino acid sequence of G. echinata HI4′OMT aligned with pea (+)-6a-hydroxymaackiain 3-OMT (accession no. U69554), L. japonicus HI4′OMT, G. echinata D7OMT and alfalfa D7OMT (IOMT) (accession no. U97125). The amino acid residues of which at least three sequences are identical are in reverse type. Gaps (–) are inserted to optimize alignment. Conserved sequence motifs in plant SAM-dependent OMT (arrow) (Ibrahim and Muzac 2000) and substrate preference motifs (broken line) (Schröder et al. 2002) are shown. Abbreviation: HMOMT, (+)-6a-hydroxymaackiain 3-OMT. View largeDownload slide Fig. 3 Amino acid sequence of G. echinata HI4′OMT aligned with pea (+)-6a-hydroxymaackiain 3-OMT (accession no. U69554), L. japonicus HI4′OMT, G. echinata D7OMT and alfalfa D7OMT (IOMT) (accession no. U97125). The amino acid residues of which at least three sequences are identical are in reverse type. Gaps (–) are inserted to optimize alignment. Conserved sequence motifs in plant SAM-dependent OMT (arrow) (Ibrahim and Muzac 2000) and substrate preference motifs (broken line) (Schröder et al. 2002) are shown. Abbreviation: HMOMT, (+)-6a-hydroxymaackiain 3-OMT. View largeDownload slide Fig. 4 HPLC (A) and electron impact mass spectrum (B) of the product of recombinant HI4′OMT reaction. (A) 1, direct reaction products; 2, product after the acid-treatment of the reaction mixture; 3, substrate (2,7,4′-trihydroxyisoflavanone, HI; Rt 6.2 min); 4, standard samples [D, daidzein (Rt 20.6 min); IF, isoformononetin (Rt 34.0 min); F, formononetin (37.5 min)]. The eluents were monitored at 285 nm. The product at Rt 17.3 min was purified by TLC and HPLC and analyzed by a mass spectrometer (B). The proposed fragmentation scheme is shown (C). View largeDownload slide Fig. 4 HPLC (A) and electron impact mass spectrum (B) of the product of recombinant HI4′OMT reaction. (A) 1, direct reaction products; 2, product after the acid-treatment of the reaction mixture; 3, substrate (2,7,4′-trihydroxyisoflavanone, HI; Rt 6.2 min); 4, standard samples [D, daidzein (Rt 20.6 min); IF, isoformononetin (Rt 34.0 min); F, formononetin (37.5 min)]. The eluents were monitored at 285 nm. The product at Rt 17.3 min was purified by TLC and HPLC and analyzed by a mass spectrometer (B). The proposed fragmentation scheme is shown (C). View largeDownload slide Fig. 5 Phylogenetic relationship of OMTs involved in flavonoid biosynthesis. Note that HI4′OMT and D7OMT belong to the different branches. The protein sequences were aligned with the CLUSTAL W program, and phylogeny was established by the neighbor-joining method (Thompson et al. 1994). The tree was displayed using TreeView software (Page 1996). View largeDownload slide Fig. 5 Phylogenetic relationship of OMTs involved in flavonoid biosynthesis. Note that HI4′OMT and D7OMT belong to the different branches. The protein sequences were aligned with the CLUSTAL W program, and phylogeny was established by the neighbor-joining method (Thompson et al. 1994). The tree was displayed using TreeView software (Page 1996). Table 1 Specific activities of heterologously expressed HI4′OMT and D7OMT of G. echinata Catalytic activity  Specific activity a (pkat mg–1)  HI4′OMT b  D7OMT c  2,7,4′-Trihydroxyisoflavanone 4′-OMT  4.8×103  0  Daidzein 7-OMT  0  23  Medicarpin OMT  14  0  Catalytic activity  Specific activity a (pkat mg–1)  HI4′OMT b  D7OMT c  2,7,4′-Trihydroxyisoflavanone 4′-OMT  4.8×103  0  Daidzein 7-OMT  0  23  Medicarpin OMT  14  0  Specific activity was calculated from the rate of radiolabeled isoflavonoid production from nearly identical amounts of [14C]SAM and the methyl acceptor (for details, see Materials and Methods). a Average of two independent experiments (maximum deviation was ca. 5%). b Purified HI4′OMT protein was used for the assays. c Crude extract of E. coli (10,000×g supernatant) expressing D7OMT was used for the assays. View Large Table 2 Partial purification of HI4′OMT protein from G. echinata cells Purification step  Total protein (mg)  Specific activity (pkat mg–1)  Total activity (pkat)  Relative purity (fold)  Recovery (%)  10,000×g sup.  4.4  3.47  15.3  1  100  (NH4)2SO4 ppt.  2.3  3.87  8.9  1.1  58.2  Chromatofocusing  0.12  59.3  7.1  17.1  46.4  Adenosine-agarose  0.001  820  0.66  236.3  4.3  Purification step  Total protein (mg)  Specific activity (pkat mg–1)  Total activity (pkat)  Relative purity (fold)  Recovery (%)  10,000×g sup.  4.4  3.47  15.3  1  100  (NH4)2SO4 ppt.  2.3  3.87  8.9  1.1  58.2  Chromatofocusing  0.12  59.3  7.1  17.1  46.4  Adenosine-agarose  0.001  820  0.66  236.3  4.3  G. echinata cells (10 g) treated with yeast extract for 24 h (Nakamura et al. 1999) were used for the purification. D7OMT activity was present in both the 10,000×g supernatant (1.0 pkat mg–1) and the ammonium sulfate fraction (2.0 pkat mg–1), but the activity was not detected in any fractions of the chromatofocusing. View Large Abbreviations D7OMT daidzein 7-O-methyltransferase EST expressed sequence tag HI4′OMT 2,7,4′-trihydroxyisoflavanone 4′-O-methyltransferase IFS 2-hydroxyisoflavanone synthase IPTG isopropyl β-d-thiogalactopyranoside OMT O-methyltransferase Rf retardation factor Rt retention time SAM S-adenosyl-l-methionine TLC thin-layer chromatography. The nucleotide sequences reported in this paper have been submitted to the DDBJ, GenBank and EMBL databases under accession numbers: AB091684 (HI4′OMT of G. echinata), AB091685 (D7OMT of G. echinata), AB091686 (HI4′OMT of L. japonicus). References Akashi, T., Aoki, T. and Ayabe, S. ( 1999) Cloning and functional expression of a cytochrome P450 cDNA encoding 2-hydroxyisoflavanone synthase involved in biosynthesis of the isoflavonoid skeleton in licorice. Plant Physiol.  121: 821–828. Google Scholar Akashi, T., Sawada, Y., Aoki, T. and Ayabe, S. ( 2000) New scheme of the biosynthesis of formononetin involving 2,7,4′-trihydroxyisoflavanone but not daidzein as the methyl acceptor. Biosci. Biotechnol. Biochem.  64: 2276–2279. Google Scholar Aoki, T., Akashi, T. and Ayabe, S. ( 2000) Flavonoids of leguminous plants: structure, biological activity, and biosynthesis. J. Plant Res.  113: 475–488. Google Scholar Asamizu, E., Nakamura, Y., Sato, S. and Tabata, S. ( 2000) Generation of 7137 non-redundant expressed sequence tags from a legume, Lotus japonicus. DNA Res.  7: 127–130. Google Scholar Attieh, J.M., Hanson, A.D. and Saini, H.S. ( 1995) Purification and characterization of a novel methyltransferase responsible for biosynthesis of halomethanes and methanethiol in Brassica oleracea. J. Biol. Chem.  270: 9250–9257. Google Scholar Barz, W. and Welle, R. ( 1992) Biosynthesis and metabolism of isoflavones and pterocarpan phytoalexins in chickpea, soybean and phytopathogenic fungi. In Recent Advances in Phytochemistry. Vol. 26. Phenolic Metabolism in Plants. Edited by Stafford, H.A. and Ibrahim, R.K. pp. 139–164, Plenum Press, New York. Google Scholar Burbulis, I.E. and Winkel-Shirley, B. ( 1999) Interactions among enzymes of the Arabidopsis flavonoid biosynthetic pathway. Proc. Natl. Acad. Sci. USA  96: 12929–12934. Google Scholar Clemens, S. and Barz, W. ( 1996) Cytochrome P450-dependent methylenedioxy bridge formation in Cicer arietinum. Phytochemistry  41: 457–460. Google Scholar Crombie, L. and Whiting, D.A. ( 1992) The mechanism of the enzymic induced flavanone-isoflavone change. Tetrahedron Lett.  33: 3663–3666. Google Scholar Dewick, P.M. ( 1993) Isoflavonoids. In The Flavonoids. Advances in Research since 1986 Edited by Harborne, J.B. pp. 117–238, Chapman and Hall, London. Google Scholar Dixon, R.A. ( 1999) Isoflavonoids: biochemistry, molecular biology, and biological functions. In Comprehensive Natural Products Chemistry. Volume 1. Polyketides and other secondary metabolites including fatty acids and their derivatives. Edited by Sankawa, U. pp. 773–823. Elsevier, Amsterdam. Google Scholar Dixon, R.A. ( 2001) Natural products and plant disease resistance. Nature  411: 843–847. Google Scholar Dixon, R.A., Chen, F., He, X.Z., Noel, J.P. and Zubieta, C. ( 2001) Properties and metabolic engineering of alfalfa phenylpropanoid pathway O-methyltransferases. In Recent Advances in Phytochemistry. Vol. 35. Regulation of Phytochemicals by Molecular Techniques. Edited by Romeo, J.T., Saunders, J.A. and Matthews, B.F. pp. 131–154. Pergamon, New York. Google Scholar Dixon, R.A., Lamb, C.J., Masoud, S., Sewalt, V.J.H. and Paiva, N.L. ( 1996) Metabolic engineering prospects for crop improvement through the genetic manipulation of phenylpropanoid biosynthesis and defense responses—a review. Gene  179: 61–71. Google Scholar Edwards, R. and Dixon, R.A. ( 1991) Isoflavone O-methyltransferase activities in elicitor-treated cell suspension cultures of Medicago sativa. Phytochemistry  30: 2597–2606. Google Scholar Frick, S. and Kutchan, T.M. ( 1999) Molecular cloning and functional expression of O-methyltransferases common to isoquinoline alkaloid and phenylpropanoid biosynthesis. Plant J.  17: 329–339. Google Scholar Frick, S., Ounaroon, A. and Kutchan, T.M. ( 2001) Combinatorial biochemistry in plants; the case of O-methyltransferases. Phytochemistry  56: 1–4. Google Scholar Gang, D.R., Lavid, N., Zubieta, C., Chen, F., Beuerle, T., Lewinsohn, E., Noel, J.P. and Pichersky, E. ( 2002) Characterization of phenylpropene O-methyltransferases from sweet basil: facile change of substrate specificity and convergent evolution within a plant O-methyltransferase family. Plant Cell  14: 505–519. Google Scholar Gowri, G., Bugos, R.C., Campbell, W.H., Maxwell, C.A. and Dixon, R.A. ( 1991) Stress responses in alfalfa (Medicago sativa L.) X. Molecular cloning and expression of S-adenosyl-l-methionine: caffeic acid 3-O-methyltransferase, a key enzyme of lignin biosynthesis. Plant Physiol.  97: 7–14. Google Scholar Haga, M., Akashi, T., Aoki, T. and Ayabe, S. ( 1997) A cDNA for S-adenosyl-l-methionine: isoliquiritigenin/licodione 2′-O-methyltransferase (accession no. D88742) from cultured licorice (Glycyrrhiza echinata L.) cells (PGR97-014). Plant Physiol.  113: 663. Google Scholar Hagmann, M. and Grisebach, H. ( 1984) Enzymatic rearrangement of flavanone to isoflavone. FEBS Lett.  175: 199–202. Google Scholar Hashim, M.F., Hakamatsuka, T., Ebizuka, Y. and Sankawa, U. ( 1990) Reaction mechanism of oxidative rearrangement of flavanone in isoflavone biosynthesis. FEBS Lett.  271: 219–222. Google Scholar He, X.Z. and Dixon, R.A. ( 1996) Affinity chromatography, substrate/product specificity, and amino acid sequence analysis of an isoflavone O-methyltransferase from alfalfa (Medicago sativa L.). Arch. Biochem. Biophys.  336: 121–129. Google Scholar He, X.Z. and Dixon, R.A. ( 2000) Genetic manipulation of isoflavone 7-O-methyltransferase enhances biosynthesis of 4′-O-methylated isoflavonoid phytoalexins and disease resistance in alfalfa. Plant Cell  12: 1689–1702. Google Scholar He, X.Z., Reddy, J.T. and Dixon, R.A. ( 1998) Stress responses in alfalfa (Medicago sativa L). XXII. cDNA cloning and characterization of an elicitor-inducible isoflavone 7-O-methyltransferase. Plant Mol. Biol.  36: 43–54. Google Scholar Hrazdina, G. ( 1992) Compartmentation in aromatic metabolism. In Recent Advances in Phytochemistry. Vol. 26. Phenolic Metabolism in Plants. Edited by Stafford, H.A. and Ibrahim, R.K. pp. 1–23, Plenum Press, New York. Google Scholar Ibrahim, R.K., De Luca, V., Khouri, H., Latchinian, L., Brisson, L. and Charest, P.M. ( 1987) Enzymology and compartmentation of polymethylated flavonol glucosides in Chrysosplenium americanum. Phytochemistry  26: 1237–1245. Google Scholar Ibrahim, R.K. and Muzac, I. ( 2000) The methyltransferase gene superfamily: a tree with multiple branches. In Recent Advances in Phytochemistry. Vol. 34. Evolution of Metabolic Pathways. Edited by Romeo, J.T., Ibrahim, R.K., Varin, L. and De Luca, V. pp. 349–384. Pergamon, New York. Google Scholar Jung, W., Yu, O., Lau, S.M., O’Keefe, D.P., Odell, J., Fader, G. and McGonigle, B. ( 2000) Identification and expression of isoflavone synthase, the key enzyme for biosynthesis of isoflavones in legumes. Nat. Biotechnol.  18: 208–212. Google Scholar Kirikae, Y., Sakurai, M., Furuno, T., Takahashi, T. and Ayabe, S. ( 1993) Biosynthesis of a dibenzoylmethane, licodione, in cultured alfalfa cells induced by yeast extract. Biosci. Biotechnol. Biochem.  57: 1353–1354. Google Scholar Kochs, G. and Grisebach, H. ( 1986) Enzymic synthesis of isoflavones. Eur. J. Biochem.  155: 311–318. Google Scholar Liu, C.J. and Dixon, R.A. ( 2001) Elicitor-induced association of isoflavone O-methyltransferase with endomembranes prevents the formation and 7-O-methylation of daidzein during isoflavonoid phytoalexin biosynthesis. Plant Cell  13: 2643–2658. Google Scholar Mathesius, U. ( 2001) Flavonoids induced in cells undergoing nodule organogenesis in white clover are regulators of auxin breakdown by peroxidase. J. Exp. Bot.  52: 419–426. Google Scholar Nakamura, K., Akashi, T., Aoki, T., Kawaguchi, K. and Ayabe, S. ( 1999) Induction of isoflavonoid and retrochalcone branches of the flavonoid pathway in cultured Glycyrrhiza echinata cells treated with yeast extract. Biosci. Biotechnol. Biochem.  63: 1618–1620. Google Scholar Page, R.D. ( 1996) TreeView: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci.  12: 357–358. Google Scholar Phillips, D.A. ( 1992) Flavonoids: plant signals to soil microbes. In Recent Advances in Phytochemistry. Vol. 26. Phenolic Metabolism in Plants. Edited by Stafford, H.A. and Ibrahim, R.K. pp. 201–231. Plenum Press, New York. Google Scholar Preisig, C.L., Matthews, D.E. and Van Etten, H.D. ( 1989) Purification and characterization of S-adenosyl-l-methionine: 6a-hydroxymaackiain 3-O-methyltransferase from Pisum sativum. Plant Physiol.  91: 559–566. Google Scholar Rakwal, R., Agrawal, G.K., Yonekura, M. and Kodama, O. ( 2000) Naringenin 7-O-methyltransferase involved in the biosynthesis of the flavanone phytoalexin sakuranetin from rice (Oryza sativa L.). Plant Sci.  29: 213–221. Google Scholar Saslowsky, D. and Winkel-Shirley, B. ( 2001) Localization of flavonoid enzymes in Arabidopsis roots. Plant J.  27: 37–48. Google Scholar Sawada, Y., Kinoshita, K., Akashi, T., Aoki, T. and Ayabe, S. ( 2002) Key amino acid residues required for aryl migration catalyzed by the cytochrome P450 2-hydroxyisoflavanone synthase. Plant J.  31: 555–564. Google Scholar Schröder, G., Wehinger, E. and Schröder, J. ( 2002) Predicting the substrates of cloned plant O-methyltransferases. Phytochemistry  59: 1–8. Google Scholar Shimada, N., Akashi, T., Aoki, T. and Ayabe, S. ( 2000) Induction of isoflavonoid pathway in the model legume Lotus japonicus: molecular characterization of enzymes involved in phytoalexin biosynthesis. Plant Sci.  160: 37–47. Google Scholar Shutt, D.A. ( 1976) The effects of plant oestrogens on animal reproduction. Endeavour  75: 110–113. Google Scholar Spaink, H.P. ( 1995) The molecular basis of infection and nodulation by rhizobia: the ins and outs of sympathogenesis. Annu. Rev. Phytopathol.  33: 345–368. Google Scholar Stavric, B. ( 1997) Chemopreventive agents in foods. In Recent Advances in Phytochemistry. Vol. 31. Functionality of Food Phytochemicals. Edited by Johns, T. and Romeo, J.T. pp. 53–87. Plenum Press, New York. Google Scholar Steele, C.L., Gijzen, M., Qutob, D. and Dixon, R.A. ( 1999) Molecular characterization of the enzyme catalyzing the aryl migration reaction of isoflavonoid biosynthesis in soybean. Arch. Biochem. Biophys.  367: 146–150. Google Scholar Tahara, S. and Ibrahim, R.K. ( 1995) Prenylated isoflavonoids—an update. Phytochemistry  38: 1073–1094. Google Scholar Thompson, J.D., Higgins, D.G. and Gibson, T.J. ( 1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucl. Acids Res.  22: 4673–4680. Google Scholar Wengenmayer, H., Ebel, J. and Grisebach, H. ( 1974) Purification and properties of a S-adenosylmethionine: isoflavone 4′-O-methyltransferase from cell suspension cultures of Cicer arietinum L. Eur. J. Biochem.  50: 135–143. Google Scholar Winkel-Shirley, B. ( 2001) Flavonoid biosynthesis. A colorful model for genetics, biochemistry, cell biology, and biotechnology. Plant Physiol.  126: 485–493. Google Scholar Wu, Q., Preisig, C.L. and VanEtten, H.D. ( 1997) Isolation of the cDNAs encoding (+)6a-hydroxymaackiain 3-O-methyltransferase, the terminal step for the synthesis of the phytoalexin pisatin in Pisum sativum. Plant Mol. Biol.  35: 551–560. Google Scholar Zubieta, C., He, X.Z., Dixon, R.A. and Noel, J.P. ( 2001) Structures of two natural product methyltransferases reveal the basis for substrate specificity in plant O-methyltransferases. Nat. Struct. Biol.  8: 271–279. Google Scholar

Journal

Plant and Cell Physiology – Oxford University Press

Published: Feb 15, 2003

There are no references for this article.