Flavone synthases (FNSs) catalyze the oxidation of flavanones to flavones, i.e. the formation of apigenin from (2S)‐naringenin. While many plants express a microsomal‐type FNS II, the soluble FNS I appears to be confined to a few species of the Apiaceae and was cloned recently from parsley plants. FNS I belongs to the FeII/2‐oxoglutarate‐dependent dioxygenases characterized by short conserved sequence elements for cofactor binding, and its evolutionary context and mode of action are under investigation. Using a homology‐based reverse transcription polymerase chain reaction approach, two additional flavonoid‐specific dioxygenases were cloned from immature parsley leaflets, which were identified as flavanone 3β‐hydroxylase (FHT) and flavonol synthase (FLS) after expression in yeast cells. Sequence alignments revealed marginal differences among the parsley FNS I and FHT polypeptides of only 6%, while much less identity (about 29%) was observed with the parsley FLS. Analogous to FNS I, FLS oxidizes the flavonoid γ‐pyrone by introducing a C2, C3 double bond, and (2R,3S)‐dihydrokaempferol (cis‐dihydrokaempferol) was proposed recently as the most likely intermediate in both FNS I and FLS catalysis. Incubation of either FNS I or FLS with cis‐dihydrokaempferol exclusively produced kaempferol and confirmed the assumption that flavonol formation occurs via hydroxylation at C3 followed by dehydratation. However, the lack of apigenin in these incubations ruled out cis‐dihydrokaempferol as a free intermediate in FNS I catalysis. Furthermore, neither (+)‐trans‐dihydrokaempferol nor unnatural (−)‐trans‐dihydrokaempferol and 2‐hydroxynaringenin served as a substrate for FNS I. Overall, the data suggest that FNS I has evolved uniquely in some Apiaceae as a paraphyletic gene from FHT, irrespective of the fact that FNS I and FLS catalyze equivalent desaturation reactions.

Flavonol synthase was classified as a 2-oxoglutarate-dependent dioxygenase converting natural (2R,3R)-dihydroflavonols, i.e. dihydrokaempferol, to the corresponding flavonols (kaempferol). Flavonol synthase from Citrus unshiu (Satsuma mandarin), expressed in Escherichia coli and purified to homogeneity, was shown to accept also (2S)-naringenin as a substrate, producing kaempferol in high yield and assigning sequential flavanone 3β-hydroxylase and flavonol synthase activities to the enzyme. In contrast, dihydrokaempferol was identified as the predominant product from assays performed with the unnatural (2R)-naringenin as substrate. The product which was not converted any further on repeated incubations was identified by 1H NMR and CD spectroscopies as (−)-trans-dihydrokaempferol. The data demonstrate that Citrus flavonol synthase encompasses an additional non-specific activity trans-hydroxylating the flavanones (2S)-naringenin as well as the unnatural (2R)-naringenin at C-3.

Flavonols are produced by the desaturation of flavanols catalyzed by flavonol synthase. The enzyme belongs to the class of intermolecular dioxygenases which depend on molecular oxygen and FeII/2-oxoglutarate for activity, and have been in focus of structural studies recently. Flavonol synthase cDNAs were cloned from six plant species, but none of the enzymes had been studied in detail. Therefore, a cDNA from Citrus unshiu (Satsuma mandarin) designated as flavonol synthase was expressed in Escherichia coli, and the purified recombinant enzyme was subjected to kinetic and mutational chacterizations. The integrity of the recombinant synthase was revealed by a molecular ion from MALDI-TOF mass spectrometry at m/z 37888 +/- 40 (as compared to 37899 Da calculated for the translated polypeptide), and by partial N-terminal sequencing. Maximal flavonol synthase activity was observed in the range of pH 5-6 with dihydroquercetin as substrate and a temperature optimum at about 37 degrees C. Km values of 272, 11 and 36 micro m were determined for dihydroquercetin, FeII and 2-oxoglutarate, respectively, with a sixfold higher affinity to dihydrokaempferol (Km 45 micro m). Flavonol synthase polypeptides share an overall sequence similarity of 85% (47% identity), whereas only 30-60% similarity were apparent with other dioxygenases. Like the other dioxygenases of this class, Citrus flavonol synthase cDNA encodes eight strictly conserved amino-acid residues which include two histidines (His221, His277) and one acidic amino acid (Asp223) residue for FeII-coordination, an arginine (Arg287) proposed to bind 2-oxoglutarate, and four amino acids (Gly68, His75, Gly261, Pro207) with no obvious functionality. Replacements of Gly68 and Gly261 by alanine reduced the catalytic activity by 95%, while the exchange of these Gly residues for proline completely abolished the enzyme activity. Alternatively, the substitution of Pro207 by glycine hardly affected the activity. The data suggest that Gly68 and Gly261, at least, are required for proper folding of the flavonol synthase polypeptide.

Flavone synthase I, a soluble 2-oxoglutarate-dependent dioxygenase catalyzing the oxidation of flavanones to flavones in several Apiaceae species, was induced in parsley cell cultures by continuous irradiation with ultraviolet/blue light for 20 h. The enzyme was extracted from these cells and purified by a revised purification protocol including the fractionation on hydroxyapatite, Fractogel EMD DEAE, and Mono Q anion exchangers, which resulted in an apparently homogeneous flavone synthase at approximately 10-fold higher yield as compared to the previous report. The homogeneous enzyme was employed to raise an antiserum in rabbit for partial immunological characterization. The specificity of the polyclonal antibodies was demonstrated by immunotitration and Western blotting of the crude ammonium sulfate-fractionated enzyme as well as of the enzyme at various stages of the purification. High titer cross-reactivity was observed toward flavone synthase I, showing two bands in the crude extract corresponding to molecular weights of 44 and 41 kDa, respectively, while only the 41 kDa was detected on further purification. The polyclonal antiserum did not cross-react with recombinantly expressed flavanone 3beta-hydroxylase from Petunia hybrida or flavonol synthase from Citrus unshiu, two related 2-oxoglutarate-dependent dioxygenases involved in the flavonoid pathway.

Flavanone 3beta-hydroxylase catalyzes the Fe(II)/oxoglutarate-dependent hydroxylation of (2S)-flavanones to (2R,3R)-dihydroflavonols in the course of flavonol/anthocyanin or catechin biosynthesis. The enzyme from Petunia hybrida consists of a 41,655-Da polypeptide that is prone to rapid proteolysis in crude plant extracts as well as on expression in Escherichia coli, and commercial protease inhibitors were inefficient in stopping the degradation. To pinpoint the primary site of proteolysis and to improve the activity yields, two revised schemes of purification were developed for the recombinant polypeptides. Applying a four-step protocol based on extraction and ion-exchange chromatography at pH 7.5, the primary, catalytically inactive proteolytic enzyme fragment (1.1 mg) was isolated and shown to cross-react on Western blotting as one homogeneous band of about 38 kDa. Mass spectrometric analysis assigned a mass of 37,820 +/- 100 Da to this fragment, and partial sequencing revealed an unblocked amino terminus identical to that of the native 3beta-hydroxylase. Thus, the native enzyme had been degraded by proteolysis of a small carboxy-terminal portion, and the primary site of cleavage must be assigned most likely to the Glu 337-Leu 338 bond, accounting for a loss of about 3800 Da. Alternatively, the enzyme degradation was greatly reduced when the extraction of recombinant bacteria was carried out with phosphate buffer at pH 5.5 followed by size exlusion and anion-exchange chromatography. This rapid, two-step purification resulted in a homogeneous 3beta-hydroxylase of high specific acitivity (about 32 mkat/kg) at roughly 5% yield, and the procedure is a major breakthrough in mechanistic investigations of this class of labile dioxygenases.

Two different hydrophilic materials with large pores, Superose 6 and Fractogel EMD BioSec (S), which are designed for size-exclusion chromatography (SEC) of plasma proteins with high molecular masses, are tested for their performance on a preparative scale. The model mixtures are preparations of the clotting factors VIII (FVIII) and IX (FIX). A combination of a Fractogel EMD BioSec (S) column and a Superose 6 column has proved to be particularly effective for separations in a wide molecular size range, from several millions down to about 20,000. Superose 6 showed good results on a small scale as well as on a large scale, even in the molecular mass range over 1,000,000. However, recovery of FVIII clotting activity was less than 70% with this material and therefore not satisfactory. Fractogel did not perform well in terms of separation on a small scale. However, in the case of biopolymers with high molecular masses, separation was improved by using larger columns. With Fractogel, recovery of activity of the two clotting factors FVIII and FIX was satisfactory, above 80%. On a large scale, the active fraction in the clotting factor concentrate was successfully separated from the non-active fraction with either size exclusion (SE) material. In the preparation under investigation, the clotting factor VIII is found in a complex with the von Willebrand factor (vWF). The FVIII-vWF complex has a molecular mass of several millions. It dissociates in the presence of high concentrations of Ca2+ ions. Under such conditions FVIII and vWF were successfully separated with both SEC columns.

Flavanone 3beta-hydroxylase, involved in the biosynthesis of flavonoids, catechins, and anthocyanidins, is a non-heme iron enzyme, dependent on Fe2+, molecular oxygen, 2-oxoglutarate, and ascorbate, the typical cofactors of the class of 2-oxoglutarate-dependent dioxygenases. Sequence alignment analysis of various 2-oxoglutarate-dependent dioxygenases and related enzymes revealed eight amino acid residues that seem to be strictly conserved within this group of enzymes. Among these residues, two histidines (His220 and His278) and one aspartic acid (Asp222) were identified as part of the putative iron-binding site and an arginine residue (Arg288) as part of the 2-oxoglutarate binding site, by site-directed mutagenesis and functional analysis of the mutated recombinant enzyme. The mutant genes were expressed in Escherichia coli to give soluble proteins whose molecular masses were in excellent agreement with the wild-type enzyme. Four out of nine mutant enzymes, [Gln78]FHT, [Gln121]FHT, [Gln264]FHT and [Gln266]FHT, were enzymatically active with activities reduced to 26-57%, implying that the mutated amino acid residues are not essential for catalysis. Replacement of His220 by glutamine and Asp222 by asparagine remarkably reduced the catalytic activity to about 0.15% and 0.4%, respectively. The [Gln220]FHT and [Asn222]FHT enzymes showed a slightly increased Km value with respect to iron binding, as compared to the wild-type enzyme. The most drastic effect on the reaction rate of flavanone 3beta-hydroxylase was achieved by mutating His278 to glutamine. The mutant had no detectable enzyme activity, indicating that His278 was essential for the catalytic reaction. The observed protection of purified enzyme from inactivation by diethylpyrocarbonate after the addition of cofactors provided further independent confirmation for the involvement of histidine residues in the active site. The substitution of Arg288 by lysine or glutamine induced a precipitous decrease in catalytic activity and a fivefold and 160-fold increase in the Michaelis constants for 2-oxoglutarate, respectively. In addition, the enzymatic activities of the latter two mutant enzymes showed a strong pH dependence in the weakly acidic as well as in the neutral pH range, unlike the wild-type enzyme. These results clearly indicate that Arg288 probably contributes to the specific binding of 2-oxoglutarate at the active site of the enzyme, most likely by providing a positive charge, properly located in order to interact with the delta-carboxyl function of 2-oxoglutarate. Furthermore, we conclude that His220, His278 and Asp222 constitute three of the possible ligands for iron binding in the active site of flavanone 3beta-hydroxylase.

A heterologous cDNA probe from Petunia hybrida was used to isolate flavanone-3 beta-hydroxylase-encoding cDNA clones from carnation (Dianthus caryophyllus), china aster (Callistephus chinensis) and stock (Matthiola incana). The deduced protein sequences together with the known sequences of the enzyme from P. hybrida, barley (Hordeum vulgare) and snapdragon (Antirrhinum majus) enabled the determination of a consensus sequence which revealed an overall 84% similarity (53% identity) of flavanone 3 beta-hydroxylases from the different sources. Alignment with the sequences of other known enzymes of the same class and to related non-heme iron-(II) enzymes demonstrated the strict genetic conservation of 14 amino acids, in particular, of three histidines and an aspartic acid. The conservation of the histidine motifs provides strong support for the possible conservation of structurally similar iron-binding sites in these enzymes. The putative role of histidines as chelators of ferrous ions in the active site of flavanone 3 beta-hydroxylases was corroborated by diethyl-pyrocarbonate modification of the partially purified recombinant Petunia enzyme.

A heterologous cDNA probe from Petunia hybrida was used to isolate flavanone-3beta-hydroxylase-encoding cDNA clones from carnation (Dianthus caryophyllus), china aster (Callistephus chinensis) and stock (Matthiola incana). The deduced protein sequences together with the known sequences of the enzyme from P. hybrida, barley (Hordeum vulgare) and snapdragon (Antirrhinum majus) enabled the determination of a consensus sequence which revealed an overall 84% similarity (53% identity) of flavanone 3beta-hydroxylases from the different sources. Alignment with the sequences of other known enzymes of the same class and to related non-heme iron-(II) enzymes demonstrated the strict genetic conservation of 14 amino acids, in particular, of three histidines and an aspartic acid. The conservation of the histidine motifs provides strong support for the possible conservation of structurally similar iron-binding sites in these enzymes. The putative role of histidines as chelators of ferrous ions in the active site of flavanone 3beta-hydroxylases was corroborated by diethyl-pyrocarbonate modification of the partially purified recombinant Petunia enzyme.

Chalcone synthase (CHS) and stilbene synthase (STS) are closely related polyketide synthases which are key enzymes in the biosynthesis of flavonoids and stilbenes. Scots pine (Pinus sylvestris) is an interesting plant for a direct comparison of the enzymes. It not only contains the usual flavonoids, but also an unusual chalcone derivative (pinocembrin), and it synthesizes stilbenes of the pinosylvin type. We analysed a CHS and a STS by molecular cloning and functional expression in Escherichia coli. The CHS was active not only with 4-coumaroyl-CoA (to naringenin chalcone), but also with cinnamoyl-CoA (leading to pinocembrin). The STS was identified as dihydropinosylvin synthase, because it preferred dihydrocinnamoyl-CoA to cinnamoyl-CoA. The protein deviated in 47 positions from the CHS consensus. It had 73.2% identity with the CHS from P. sylvestris and only 65.3% with a STS from peanut (Arachis hypogaea). We also investigated the regulation of both enzyme types in P. sylvestris plantlets exposed to stress. CHS was present in non-stressed plantlets, and induction led to a transient increase with a peak after 16 h. STS1 type activities were regulated differently and were absent in non-stressed plantlets. Increases were observed after a lag period of at least 6 h, and highest activities were obtained after 30 h. The analysis of the reactions in the plant extracts and the substrate specificity of the cloned STS indicate that the plants contain at least two different types of STS: the cloned dihydropinosylvin synthase and a pinosylvin synthase which preferentially utilizes cinnamoyl-CoA as substrate.

A cDNA encoding flavanone 3 beta-hydroxylase was isolated from petals of Petunia hybrida. The open reading frame of the nearly full length cDNA coded for a 369-amino acid polypeptide with a calculated Mr of 41,466. The function of this nucleotide sequence was verified by comparison with amino acid sequence of the amino terminus and tryptic peptides from purified plant enzyme, by Northern blotting with RNA from wild type and mutant plants, and by prokaryotic expression yielding an enzymatically active hydroxylase. Computer-aided sequence analysis revealed high similarity (73.5%) to flavanone 3 beta-hydroxylase from barley. Genomic Southern blot analysis showed the presence of only one gene for flavanone 3 beta-hydroxylase in P. hybrida.

Flavanone 3β-hydroxylase from Petunia hybrida has been purified to apparent homogeneity utilizing an improved purification protocol including chromatography of the partially purified enzyme on hydroxyapatite, chromatofocusing, and hydrophobic interaction chromatography on phenyl-Superose. The specificity of mouse and rabbit polyclonal antisera directed to flavanone 3β-hydroxylase was demonstrated by Western blotting and immunotitration with crude extracts from wild-type flowers of P. hybrida and with the purified enzyme. Cross-reactivity was observed with flavanone 3β-hydroxylase from extracts of illuminated parsley cells. A Petunia mutant with white flowers, previously shown to lack 3β-hydroxylase activity and to accumulate flavanone glycosides, showed complete absence of the enzyme protein.

Soluble flavone synthase I from illuminated parsley cells was purified to near homogeneity by a six-step procedure. A molecular mass of 48 +/- 2 kDa was determined by gel permeation chromatography and denaturing polyacrylamide gel electrophoresis. A single protein with an isoelectric point at pH 4.8 +/- 0.1 was detected on isoelectric focusing gels, which catalyzed the overall conversion of 2S-flavanones into the corresponding flavones in the presence of molecular oxygen, 2-oxoglutarate, ferrous ion, and ascorbate. Apparent Michaelis constants for 2S-naringenin, 2S-eriodictyol, and 2-oxoglutarate were determined as 5, 8, and 16 microM, respectively. (+)-Dihydrokaempferol and 2R-naringenin were not accepted as substrates. The enzyme was strongly inhibited by Cu2+ and Zn2+. Potent competitive inhibition with respect to 2-oxoglutarate was observed with 2,4-pyridinedicarboxylate (Ki = 1.8 microM). With crude extracts as well as with the purified enzyme neither the hypothetical intermediate 2-hydroxyflavanone nor a dehydratase activity capable of converting the chemically synthesized compound to flavone could be observed. Moreover, the introduction of the double bond into the substrate naringenin was not altered by addition of chemically synthesized 2-hydroxynaringenin into the reaction mixture. Therefore, 2-hydroxyflavanones are apparently not freely dissociable intermediates in the biosynthesis of flavones in parsley and are not capable of entering the active site of the enzyme to compete with the flavanone. It is postulated that flavone synthase I catalyzes double-bond formation by direct abstraction of vicinal hydrogen atoms at C-2 and C-3 of the substrate. Thus, flavone synthase I is a member of a novel subgroup within the 2-oxoglutarate-dependent dioxygenases that can be referred to as 2-oxoglutarate-dependent desaturases.

Previous indirect evidence suggested that the biosynthesis of flavonoids in Citrus may not proceed via the usual chalcone synthase reaction and that glycosylation occurs during chalcone formation and not afterward, as has been reported in other species. We detected chalcone-synthase and UDP-glucose:flavanone-7-O-glucosyl-transferase activities in cell-free extracts of Citrus. The glucosylated flavanone was further rhamnosylated when exogenous UDP-glucose and NADPH were added to the extract. Chalcone-synthase activity was detected in cell-free extracts derived from young leaves and fruits. Young fruits (2 millimeter diameter) had the highest chalcone synthase activity. UDP-glucose:flavanone-7-O-glucosyl-transferase activity was measured in cell-free extracts derived from young leaves and fruits of Citrus mitis and Citrus maxima. The highest UDP-glucose:flavanone-7-O-glucosyl-transferase activity was found in young C. maxima leaves. These data indicate that Citrus contains a flavonoid pathway similar to that studied in other species.

Glutamate dehydrogenase (GDH, E.C. 1.4.1.3) of mustard cotyledons was investigated during the first 4 days of seedling development. The enzyme was found to be composed of seven catalytically active isoforms (each with a molecular mass of 270 kDa) which exhibited a charge heterogeneity when investigated by isoelectric focusing. Antibodies against the purified isoform 7, raised in rabbits, cross-reacted with each of the isoforms in Western blotting experiments. In addition, each of the isoforms was composed of four immunopositive reacting polypeptides with 19, 21, 23 and 25 kDa. During development of the seedlings, a shift in the isoform pattern towards the more acidic forms was found which was more pronounced when the seedlings were supplied with 15 mM NH4Cl. The time course of changes in total GDH level can be correlated with the time course of disappearance of storage proteins. Both parameters are negatively regulated by light possibly via the photoreceptor, phytochrome. There are some indications that GDH in young mustard cotyledons mainly acts in the deaminating direction.

Previous indirect evidence suggested that the biosynthesis of flavonoids in Citrus may not proceed via the usual chalcone synthase reaction and that glycosylation occurs during chalcone formation and not afterward, as has been reported in other species. We detected chalcone-synthase and UDP-glucose:flavanone-7-O-glucosyl-transferase activities in cell-free extracts of Citrus. The glucosylated flavanone was further rhamnosylated when exogenous UDP-glucose and NADPH were added to the extract. Chalcone-synthase activity was detected in cell-free extracts derived from young leaves and fruits. Young fruits (2 millimeter diameter) had the highest chalcone synthase activity. UDP-glucose:flavanone-7-O-glucosyl-transferase activity was measured in cell-free extracts derived from young leaves and fruits of Citrus mitis and Citrus maxima. The highest UDP-glucose:flavanone-7-O-glucosyl-transferase activity was found in young C. maxima leaves. These data indicate that Citrus contains a flavonoid pathway similar to that studied in other species.

Individual flowers from inflorescences of Dahlia variabilis (cv Scarlet Star) in young developmental stages contained relatively high activity of (+)-dihydroflavonol (DHF) 4-reductase. The DHF reductase was purified from such flowers to apparent homogeneity by a five-step procedure. This included affinity adsorption on Blue Sepharose and elution of the enzyme with NADP+. By gel filtration and by sodium dodecyl sulfatepolyacrylamide gel electrophoresis it was shown that DHF reductase contains only one polypeptide chain with a Mr of about 41,000. The reductase required NADPH as cofactor and catalyzed transfer of the pro-S hydrogen of NADPH to the substrate. Flavanones and dihydroflavonols (3-hydroxyflavanones) were substrates for DHF reductase with pH optima of about 6.0 for flavanones and of about 6.8 for dihydroflavonols. Flavanones were reduced to the corresponding flavan-4-ols and (+)-dihydroflavonols to flavan-3,4-cis-diols. Apparent Michaelis constants determined for (2S)-naringenin, (2S)-eriodictyol, (+)-dihydrokaempferol, (+)-dihydroquercetin, and NADPH were, respectively, 2.3, 2,10,15, and 42 μm, solVKm values were higher for dihydroflavonols than for flavanones. Conversion of dihydromyricetin to leucodelphinidin was also catalyzed by the enzyme at a low rate, whereas flavones and flavonols were not accepted as substrates. DHF reductase was not inhibited by metal chelators.

The existence of an enzymatic activity catalyzing the NADPH-dependent reduction of (+)-dihydroquercetin to (+)-2,3-trans-3,4-cis-leucocyanidin in crude extracts of wild type Zea mays is described. Expression of this dihydroquercetin reductase (DQR) activity depends on the genotype of mutants for anthocyanin biosynthesis. Protein translated in vitro from the RNA product of a full-length A1 cDNA clone exhibits DQR activity; therefore, the A1 locus is the structural gene for DQR of maize.

(2S)-Flavanone 3-hydroxylase from flowers of Petunia hybrida catalyses the conversion of (2S)-naringenin to (2R,3R)-dihydrokaempferol. The enzyme could be partially stabilized under anaerobic conditions in the presence of ascorbate. For purification, 2-oxoglutarate and Fe2+ had to be added to the buffers. The hydroxylase was purified about 200-fold by a six-step procedure with low recovery. The Mr of the enzyme was estimated by gel filtration to be about 74,000. The hydroxylase reaction has a pH optimum at pH 8.5 and requires as cofactors oxygen, 2-oxoglutarate, Fe2+ and ascorbate. With 2-oxo[1-14C]glutarate in the enzyme assay dihydrokaempferol and 14CO2 are formed in a molar ratio of 1:1. Catalase stimulates the reaction. The product was unequivocally identified as (+)-(2R,3R)-dihydrokaempferol. (2S)-Naringenin, but not the (2R)-enantiomer is a substrate of the hydroxylase. (2S)-Eriodictyol is converted to (2R,3R)-dihydroquercetin. In contrast, 5,7,3',4',5'-pentahydroxy-flavanone is not a substrate. Apparent Michaelis constants for (2S)-naringenin and 2-oxoglutarate were determined to be respectively 5.6 mumol X l-1 and 20 mumol X l-1 at pH 8.5. The Km for (2S)-eriodictyol is 12 mumol X l-1 at pH 8.0. Pyridine 2,4-dicarboxylate and 2,5-dicarboxylate are strong competitive inhibitors with respect to 2-oxoglutarate with Ki values of 1.2 mumol X l-1 and 40 mumol X l-1, respectively.