Probing the Deoxyflavonoid Biosynthesis: Naringenin Chalcone Is a Substrate for the Reductase Synthesizing Isoliquiritigenin

Abstract

Isoliquiritigenin formation is a key reaction during deoxyflavonoid biosynthesis, which is catalyzed by two enzymes, chalcone synthase (CHS) and reductase (CHR). The substrates for CHS are established. However, the substrate for CHR is unknown. In this study, an in vitro reaction was performed to confirm whether naringenin chalcone can be a substrate. Naringenin chalcone was used as a substrate during the CHR reaction. Analyzing the product revealed that isoliquiritigenin was produced from naringenin chalcone, indicating that naringenin chalcone is a substrate. This study is the first to identify a substrate for CHR, reveals that deoxyflavonoid biosynthesis diverges from naringenin chalcone, endorses the term “chalcone reductase,” and answers the long-standing questions about doubly-labeled acetic acid uptake pattern in deoxyflavonoid biosynthesis.

INTRODUCTION

Flavonoids are widely distributed in higher plants. In addition to flavonoids, deoxyflavonoids such as daizein (10 in Fig. 1), which lack one oxygen atom, are produced in certain plants, such as legumes.¹⁾ The flavonoid skeleton is formed by a reaction producing naringenin chalcone (5) from one 4-coumaroyl CoA molecule (2) and three malonyl CoA molecules (1), which is catalyzed by chalcone synthase (CHS),²⁾ belonging to type III polyketide synthase.³⁾

Fig. 1. Flavonoid and Deoxyflavonoid Biosynthesis

Biosynthetic pathway of flavone (left panel) and deoxyisoflavone (right panel). Blue arrows indicate possible reactions by chalcone reductase (CHR). Red bonds indicate incorporated intact acetate units from [1,2-¹³C₂]-acetate. CHS, chalcone synthase; CHI, chalcone isomerase

The first route-specific product in deoxyflavonoid biosynthesis is isoliquiritigenin (6); however, the details of the enzymes that produce isoliquiritigenin are unknown. Enzymatic activity was first detected in stress-induced cultured cells from Glycyrrhiza echinata⁴⁾ and the enzyme was named “deoxychalcone synthase (DOCS).” DOCS activity was also detected in Pueraria lobata cell cultures treated with an endogenous elicitor.⁵⁾ Enzyme purification from the cell culture of Glycine max was performed, revealing that DOCS comprised two enzymes, CHS and chalcone reductase (CHR).⁶⁾ Isoliquiritigenin synthesis requires CHS and CHR; however, the CHR substrate is unidentified.⁷⁾ CHR reacts with neither naringenin chalcone (5) nor naringenin (7) as substrates (routes c and d in Fig. 1), and the activity of CHR can only be detected when CHS and its substrates co-exist in a CHR reaction solution.⁶⁾ Because the substrate for CHR remains unknown, it was referred to as “nicotinamide adenine dinucleotide phosphate (NADPH)-dependent reductase that co-acts with CHS.”⁸⁾

In 1995, two papers^9,10) reported cloning CHR from Medicago sativa and described it as “chalcone reductase” for the first time. Despite being referred to as “chalcone reductase,” its function was still described as “reductase which synthesizes 4,2′,4′-trihydroxychalcone in co-action with CHS.” After that, CHR has been conventionally called “chalcone reductase.” However, calling it “chalcone reductase” is incorrect because naringenin chalcone is unacceptable for CHR. CHR was called later “chalcone ketide reductase”¹¹⁾ or “polyketide reductase.”³⁾

The three-dimensional structure of CHR was elucidated in 2005,¹²⁾ revealing the active site has no space to accept CoA, and therefore strongly suggesting the polyketomethylene intermediate attached to CoA (3) is not a substrate (route a in Fig. 1). Since CHR is a member of the aldo-keto reductase (AKR) superfamily,^12–14) 4-coumaroyl-trione (4), a ring-closed compound with carbonyl group, was proposed to be a CHR substrate¹²⁾ (route b in Fig. 1). This 4-coumaroyl-trione is currently the most accepted substrate.¹⁵⁾ However, when 4-coumaroyl-trione is transferred from CHS to CHR, it must be released into the aqueous solution and aromatized because it is extremely labile¹⁶⁾; therefore, 4-coumaroyl-trione is possibly not a CHR substrate. Moreover, naringenin chalcone produced from 4-coumaroyl-trione through aromatization is spontaneously and rapidly converted to naringenin under physiological conditions.^17,18)

In addition to the in vitro reaction,⁶⁾ the old feeding experiments of doubly-labeled acetate indicate that naringenin chalcone is not a substrate for CHR.^{19–22) 13}C-NMR analyses of deoxyflavonoids biosynthesized from [1,2-¹³C₂]-acetate demonstrated a single specific assembly of three units in the A-ring (daizein (10) in Fig. 1). Reduction must occur before naringenin chalcone formation because free rotation of the naringenin chalcone A-ring results in randomization of the label pattern as indicated in apigenin (9) in Fig. 1. After that, reviews have described that neither naringenin chalcone nor naringenin is a substrate for CHR.^{4,7,11,23,24)}

CHS was originally named “flavanone synthase” because the product detected in an in vitro enzyme reaction was naringenin, a flavanone.²⁾ However, “flavanone synthase” was corrected to “chalcone synthase” because naringenin chalcone was proved to be the original product existing in the enzyme reaction solution and rapidly decreased by spontaneous isomerization to naringenin.¹⁸⁾ This instability of naringenin chalcone might be a key point in deoxyflavonoid biosynthesis. Whether naringenin chalcone is unstable to be a substrate for CHR is unknown. Naringenin chalcone has been isolated from tulip petals²⁵⁾ and tomato pericarp²⁶⁾ and may be relatively stable in some parts of the plant, likely under non-physiological conditions. Naringenin chalcone production has previously been confirmed in in vitro CHS experiments.^15,18) Naringenin chalcone exists for a short period under certain physiological conditions and possibly becomes a substrate for CHR.

Although CHR is responsible for branching deoxyflavonoid biosynthesis from flavonoid, which compound deoxyflavonoid biosynthesis diverges from is unknown. Since a substrate for CHR cannot be determined, a virtual compound, polyketomethylene intermediate or 4-coumaroyl-trione is considered to be a substrate. The actual substrate for CHR needs to be identified to know the branching point. The reason for the substrate being unknown is possibly due to an insufficiency of the previous in vitro reaction experiment. The substrate for CHR must exist in the solution of CHS and CHR reaction, where compounds existing are naringenin chalcone and naringenin. Therefore, in this study, the in vitro reaction is re-examined to confirm whether naringenin chalcone or naringenin can be a substrate for CHR.

MATERIALS AND METHODS

Naringenin Chalcone Synthesis

Naringenin chalone was synthesized from naringenin, as previously described.¹⁷⁾ Naringenin (0.69 g) was heated with 4 g of potassium hydroxide and 4 mL of water in a boiling water bath for 4 min. After cooling, the mixture was acidified with 20% hydrochloric acid. The resulting orange-yellow precipitate was recrystallized from water–ethanol, yielding 51.8 mg. Naringenin chalcone was confirmed using ¹H-NMR spectroscopy (Bruker AVANCE III HD 400 NanoBay spectrometer, Bruker, Billerica, MA, U.S.A.). The NMR spectrum shown in the Supplementary S2 is identical to the reported data.²⁶⁾ The crystals contained a small quantity of apigenin (9), an impurity in purchased naringenin. Apigenin was removed using HPLC under the same conditions described in the assay procedure before the enzyme assay.

Davidigenin Synthesis

Davidigenin was synthesized from phloretic acid and resorcinol as previously described.^27,28) Phloretic acid (6 mmol, 990 mg) and resorcinol (6 mmol, 661 mg) were stirred in 4 mL boron trifluoride–ethyl ether complex (32 mmol) at 90 °C for 120 min. The mixture was poured into 200 mL of 10% aqueous sodium acetate and stirred for 30 min at room temperature. The solution was extracted with 200 mL ethyl acetate. The extract was washed with water and brine before drying over sodium acetate. The solvent was evaporated under reduced pressure, and the residue was chromatographed over a silica gel column using hexane–ethyl acetate mixtures. Fractions containing davidigenin were combined and evaporated. Finally, davidigenin (white powder, 420 mg) was obtained by recrystallization with hexane–ethyl acetate and confirmed using ¹H-NMR spectroscopy (Bruker AVANCE III HD 400 NanoBay spectrometer). The NMR spectrum shown in Supplementary S3 is identical to the reported data.²⁷⁾

Cloning CHS and CHR from G. max

In the soybean genome, at least 11 CHR-related homologs exist, three homologs (CHR1,2,6) of which have enzymatic activity.¹⁵⁾ The protein purified from cell culture is CHR1 (Glyma.14G005700 in Phytozome v.10.3), which is induced by pathogenic attacks and referred to as cDNA (accession number X55730 in EMBL data library).^6,8) cDNA of CHR and CHS were first cloned into the pYES2 plasmid (Invitrogen, Waltham, MA, U.S.A.) and then transferred to pET-15b or pET-21a (Novagen, Pretoria, South Africa), respectively. Four oligonucleotides were synthesized to clone the cDNA of CHR⁸⁾ into pYES2. RNA was obtained from 7-d-old G. max seedlings using a NucleoSpin™ RNA Plant kit (Macherey-Nagel, Allentown, PA, U.S.A.). RT-PCR was performed using a SuperScript™ IV One-Step RT-PCR System (Invitrogen) and PrimeStar™ Max DNA polymerase (TaKaRa, San Jose, CA, U.S.A.) to amplify CHR cDNA, which was cloned into pYES using the In-Fusion™ HD Cloning Kit (TaKaRa). PCR was performed to amplify the cDNA of CHR to introduce it into the Xho I site of pET-15b using the obtained plasmid as a template. The plasmid was named pET-CHR. Similarly, CHS (locus name: Glyma.11G011500 in Phytozome v.10.3) was cloned between the Nde I and Eco RI sites of pET-21a to construct pET-CHS. Primers are listed in the Supplementary S1.

Protein Expression and Purification

A portion (1/50) of overnight Escherichia coli BL21(DE3) culture harboring pET-CHR or pET-CHS was inoculated in Luria–Bertani medium containing 100 µg/mL of ampicillin. Then, the culture was incubated at 37 °C for 2.5 h. After the culture was cooled on ice, 1 mM isopropyl thio-β-D-galactoside was added to induce protein expression, and the culture was incubated at 30 °C for 20 h. Cells were collected through centrifugation (at 1100 × g for 10 min) and resuspended in xTractor™ buffer (TaKaRa) to lyse cells. Cell lysis was performed by gently shaking the cells for 15 min at 25 °C. The lysate was loaded onto a Capturem™ His-Tagged Purification Maxiprep (TaKaRa). After elution from the column, the eluate was passed through the Zeba™ Spin Desalting Column (Thermo Fisher Scientific, Waltham, MA, U.S.A.) to remove histidine using 0.1 M potassium phosphate buffer (pH 7.0).

Reaction Mixture of Enzyme Assay

CHS and CHR

The reaction mixture comprised purified CHS (6.9 µg) and CHR (11.4 µg) proteins, 0.185 mM 4-coumaroyl CoA (laboratory stock²⁹⁾ synthesized as previously described³⁰⁾), 0.30 mM malonyl CoA, and 1 mM NADPH in 50 µL of 0.1 M potassium phosphate buffer (pH 6.0).

CHR with Naringenin Chalcone

The reaction mixture comprised purified CHR protein (11.4 µg), 0.147 mM naringenin chalcone, and 1 mM NADPH in 50 µL of 0.1 M potassium phosphate buffer (pH 6.0). To make spontaneous isomerization of naringenin chalcone to naringenin reduced and enzyme activity maintained, pH was set to pH 6.0. Spontaneous isomerization of naringenin chalcone by alterations of pH was previously investigated.¹⁷⁾

CHR with Phloretin

The reaction mixture comprised purified CHR protein (11.4 µg), 0.29 mM phloretin (FUJIFILM Wako Pure Chemical Corporation, Richmond, VA, U.S.A.), and 1 mM NADPH in 50 µL of 0.1 M potassium phosphate buffer (pH 6.0).

Enzyme Assay

Each mixture was incubated at 30 °C for 30 min except as described in the figure legends. The mixture was then extracted with butanol. After evaporating the solvent using nitrogen gas flow, the residue was dissolved in methanol and separated using reverse-phase HPLC on a Kinetex™ C18 column (4.6 mm ID × 150 mm, Phenomenex, Torrance, CA, U.S.A.) maintained at 40 °C using SHIMADZU LC-20A HPLC equipped with photodiode-array detector (Kyoto, Japan). Gradient elution was performed with water and methanol containing 1% acetic acid. The gradient profile was 0–5 min, 42% methanol; 5–20 min, linear gradient from 42–45% methanol. The flow rate was 1.0 mL/min. The eluate was monitored at 254 nm.

RESULTS

Reduction of Naringenin Chalcone by CHR

In this study, CHR was expressed in E. coli, and the enzymatic activity was investigated. Isoliquiritigenin production by CHR co-existing with CHS and its substrates, malonyl CoA and 4-coumaroyl CoA was confirmed (Fig. 2B). This CHR was used in the reactions with naringenin chalcone or naringenin as a substrate. No radioactive compounds as a tracer were used to ensure substrate and product detection by HPLC. Although CHR did not produce liquiritigenin (8) from naringenin (Fig. 2C), CHR reacted with naringenin chalcone to produce isoliquiritigenin (Fig. 2D), indicating that naringenin chalcone is a substrate for CHR. Approximately 85% of naringenin chalcone isomerized to naringenin during 30 min; however, naringenin chalcone remained at approximately 15%, which might be enough to function as a substrate for CHR. However, this 85% decrease in substrate during initial 30 min makes calculating Km and Vmax very difficult. Since the substrate concentration is decreasing every moment, the exact substrate concentration cannot be obtained. Accurate substrate concentration is essential to obtain accurate Km and Vmax.

Fig. 2. HPLC Analyses of CHR Products

(A) Standard compounds: 1; liquiitigenin, 2; naringenin, 3; naringenin chalcone, 4; isoliquiitigenin. (B) Products by CHR co-existing with CHS and its substrates, malonyl CoA and 4-coumaroyl CoA. (C) Products by CHR from naringenin. (D) Products by CHR from naringenin chalcone. (E) Products without CHR from naringenin chalcone. The eluate was monitored at 254 nm using a photodiode-array detector.

Co-incubating CHS and CHR produced a large amount of naringenin in early reports^6,8) because naringenin chalcone is quickly isomerized to naringenin during the transfer from CHS to CHR via the buffer. However, the amounts of naringenin and naringenin chalcone detected in the CHS and CHR reaction in this study were lower compared to the amount of isoliquiritigenin (Fig. 2B). This finding is likely due to the interaction between CHS and CHR,³¹⁾ which efficiently delivers naringenin chalcone from CHS to CHR without isomerization to naringenin. The naringenin:isoliquiritigenin ratio produced in in vitro reactions was previously investigated.^5,15)

Reduction of Phloretin to Davidigenin by CHR

Phloretin (11) in Fig. 3, the naringenin chalcone analog, was tested to confirm the reactivity to naringenin chalcone. The difference between naringenin chalcone and phloretin is that the double bond in naringenin chalcone is hydrogenated. Isomerization of naringenin is caused by attacking the phenolic hydroxyl of the A-ring to the C3 carbon in naringenin chalcone. Phloretin cannot isomerize to flavanone due to the lack of a double bond. As naringenin chalcone and phloretin structures are the same except for the bond between C2 and C3 carbons, phloretin was expected to be reduced to davidigenin (12) by CHR. As expected, davidigenin was produced as a product of phloretin (Fig. 3B).

Fig. 3. Reduction of Phloretin to Davidigenin by CHR

(A) Standards of phloretin and davidigenin. (B) Products by CHR with phloretin. The peak at approximately 16 min is apigenin, contained in purchased phloretin as an impurity. The eluate was monitored at 254 nm using a photodiode-array detector.

Analyses of the Time Course Changes of Reduction by CHR

In order to see the effects of the instability of naringenin chalcone on isoliquiritigenin production, the time course changes were analyzed. A rapid decrease in naringenin chalcone and a rapid increase in naringenin were observed, with a small increase in isoliquiritigenin production (Fig. 4). The amount of non-enzymatically produced naringenin was greater than that of enzymatically produced isoliquiritigenin (Fig. 4).

Fig. 4. Time–Course Analyses of CHR Products

Naringenin chalcone (0.185 mM) and phloretin (0.29 mM) reacted with CHR as substrates. The initial amount of naringenin chalcone or phloretin was 1. Samples were prepared at five points: 15, 30, 60, 90, and 120 min. Red: naringenin chalcone; blue: naringenin; purple: isoliquiritigenin; orange: phloretin; green: davidigenin.

The timecourse increase of davidigenin production continued linearly until 120 min (Fig. 5). In contrast, that of isoliquiritigenin flattened out after 30 min, and overtaken by davidigenin. Since phloretin does not undergo spontaneous ring formation into flavanones, the stable presence of the substrate is the reason for the linear increase of davidigenin. In other words, the flattening of isoliquiritigenin synthesis is due to the large decrease of naringenin chalcone.

Fig. 5. Enlarged View of the Time Course of Isoliquiritigenin and Davidigenin Production

Purple: isoliquiritigenin; green: davidigenin.

DISCUSSION

Naringenin chalcone was previously reported as not being a substrate for CHR due to two experimental results. First, CHR reduces neither naringenin nor naringenin chalcone in an in vitro reaction using a purified enzyme from soybean cell culture.⁶⁾ Second, the labeling pattern of deoxyflavonoids in feeding experiments with doubly-labeled acetate cannot be explained if reduction happens to naringenin chalcone in which the bond between C-1″ and C-1 can freely rotate.^19–22) However, the findings of this study in which CHR reduced naringenin chalcone contradict these two experimental results. The contradictions are resolved as follows:

(1) The result that CHR reduces neither naringenin chalcone nor naringenin was obtained using purified enzymes from plant cell culture.⁶⁾ Since the available amounts of plant-derived enzymes were limited, their experiments used tracer compounds. Their substrate was radio-labeled naringenin chalcone enzymatically prepared from 4-coumaroyl CoA and [¹⁴C]-labeled malonyl CoA using purified CHS. TLC analysis of the obtained CHS products showed two peaks of radioactivity. One peak is described as naringenin, and the other as naringenin chalcone. Although the product of CHR co-acting with CHS, isoliquiritigenin, was identified in their study, there was no description to identify naringenin chalcone, a product of CHS. Therefore, the radioactive peak that the authors assigned as naringenin chalcone might not be naringenin chalcone because CHS produces several by-products, such as p-coumaroyltriacetic acid lactone and bisnoryangonin.²⁹⁾ The authors did not separate naringenin chalcone, and directly used a mixture of naringenin and naringenin chalcone as CHR substrates. Even if naringenin chalcone exists there, the amount is drastically decreased during the CHS reaction through isomerization to naringenin since the authors continued an in vitro enzyme reaction for 30 min at 30 °C. They possibly used “a mixture of naringenin and naringenin chalcone” containing a too-small amount of naringenin chalcone to become a substrate for CHR, and could detect no isoliquiritigenin due to low isoliquiritigenin amount and low detection sensitivity to radioactivity, and concluded that neither naringenin nor naringenin chalcone could function as a substrate for CHR.

(2) Since the bond between C-1″ and C-1 of naringenin chalcone freely rotates, the acetate-derived doubly-labeling pattern must undergo randomizing at the stage of naringenin chalcone, resulting in a mixture of labeling patterns in oxy-type flavonoids such as apigenin.^20,21) In contrast, in deoxyflavonoid biosynthesis, such as pisatin, a metabolite of isoliquiritigenin, only one type of doubly-labeling pattern derived from doubly-labeled acetate is observed.^19,20,22) Therefore, the proposal that the reduction occurs in the polyketomethylene chain intermediate before naringenin chalcone formation,⁷⁾ or 4-coumaroyl-trione just before aromatization,^3,12) has been accepted. If reduction occurs to naringenin chalcone in which the bond between C-1″ and C-1 can rotate, randomizing of labeling pattern in the deoxyflavonoid cannot be explained. Figure 6 shows the proposed mechanism of naringenin and isoliquiritigenin formation from naringenin chalcone.

Fig. 6. Proposed Mechanism of Naringenin and Isoliquritigenin Formation from Naringenin Chalcone

Left panel; formation of isoliquiritigenin by CHR. Upper panel; non-enzymatic formation of naringenin. Middle panel; formation of naringenin by chalcone isomerase (CHI) type II. Lower panel; formation of liquiritigenin by CHI type II. Red bonds indicate incorporated intact acetate units from [1,2-¹³C₂]-acetate. The hydrogen bonds between amino acid residues and naringenin chalcone are based on Ref. 34.

The hydroxyl group at C-2″ of the chalcone is hydrogen-bonded with the adjacent carbonyl group. Thus, the rotation they propose for randomizing requires energy exceeding this hydrogen-bond energy. For example, factors such as time length or enzyme activation are required. The hydrogen bond prevents the bond between C-1″ and C-1 from rotating during very short periods. If the hydrogen bond can retain this form for a while after being released into an aqueous solution and immediately reacts as a substrate for CHR, the product labeled single-pattern is produced (a). That is, randomizing does not occur in deoxyflavonoids.

The strength of this hydrogen bond explains the isomerization from isoliquiritigenin to liquiritigenin well, which needs chalcone isomerase (CHI).³²⁾ When isoliquititgenin is formed, the C1″–C1 bond rotates, and the hydrogen bond between hydroxyl and carbonyl groups is formed again in the aqueous solution because it is more stable than when the hydroxyl group is free (b). Therefore, the hydroxyl group of the resulting isoliquiritigenin is distanced from the double bond, resulting in no spontaneous isomerization from isoliquiritigenin to liquiritigenin, explaining why isomerization requires CHI (c).

In contrast, if the hydrogen bond is too strong to fix conformation for a long time, randomizing caused in oxy-type flavonoids such as apigenin cannot be explained. This randomizing in oxy-type flavonoids can be explained as follows for each of (1) the non-enzymatic and (2) enzymatic processes.

(1) Carbocation formed by the rise of a carbonyl group at C-1 (d) can have several resonance structures (f), which are relatively stable and exist for a relatively long period. During this period, the carbonyl becomes enol hydroxyl, leading to a decrease in the strength of the hydrogen bond to hydroxyl at C-2,″ rotates around the bond between C-1″ and C-1 (g), and randomizes the labeling pattern on the A-ring. Although the possibility that carbocation at C-1 returns to carbonyl again (e) is not denied, the hydroxyl group at C-6″ or C-2,″ which are not distinguished, preferentially attacks the carbocation at C-3 (h) to form the third ring, which is possibly the reason for the short lifetime of naringenin chalcone in aqueous solution. As naringenin can never return to naringenin chalcone in physiological conditions, naringenin chalcone with randomizing patterns did not form, thus never becomes a substrate for CHR.

(2) CHI produces flavanones from naringenin chalcone. Two types of CHI exist, type I and -II.³³⁾ CHI type I only reacts to naringenin chalcone, whereas type II reacts to isoliquiritigenin and naringenin chalcone. The CHI in G. max is type II. The reaction mechanism of these CHIs has been elucidated based on three-dimensional protein structures.³⁴⁾ In the active site of CHI type I, tyrosine and serine hydroxyls are hydrogen-bonded to two hydroxyls (4″- and 6″-) on the A-ring, and threonine hydroxyl is hydrogen-bonded to the carbonyl at C-1.³⁴⁾ Three-point fixing does not make the C1″–C1 bond rotate in CHI type I. In contrast, in the CHI type II active site, tyrosine leaves the hydroxyl (6″-) and forms a hydrogen bond network to the carbonyl. Thus, only one (4″-) of the three hydroxyls in the A-ring is hydrogen-bonded to the enzyme. As 4″-hydroxyl is on the C4″-, C1″- and C1 carbon axes, the C1″–C1 bond can rotate freely (i). CHI type II makes the C1″–C1 bond actively rotate because it is necessary to isomerize deoxychalcone to flavanone (j). This force to make the C1″–C1 bond actively rotate also happens to oxy-type chalcone (i). Thus, besides randomizing by non-enzymatic isomerization, this force of CHI type II is another cause of randomizing in oxy-type flavonoids. Since CHI in most plants is type II, randomizing in the A-ring in oxy-type flavonoid is explained by the reaction of CHI type II.

Naringenin chalcone was revealed to be a substrate in this study, indicating that CHR removes a phenolic hydroxyl group. No enzymes are known that remove phenolic hydroxyl groups. The reaction mechanism can be well explained by considering that the phenolic hydroxyl group once becomes a carbonyl group, and then the generated carbonyl group is reduced, as shown in a. It is presumed that for C2″ to become a carbonyl group, a carbonyl group is required at the adjacent position. To clarify the reaction mechanism, it is necessary to examine reactivity using several other analogs as substrates.

CONCLUSION

A substrate for CHR has not been identified. This study is the first to identify a substrate for CHR. Polyketomethylene intermediates and 4-coumaroyl-trione, not proven to exist, have been proposed to be a substrate for CHR. The basis for the proposal is the results of an in vitro reaction,⁶⁾ feeding experiments,^19–22) and homology of CHR to AKR.¹²⁾ Among them, the in vitro reaction experiment was insufficient and lead to the misunderstanding that naringenin chalcone is not a substrate.

This study proves that naringenin chalcone, an unstable existing compound, functions as a CHR substrate. That is, deoxyflavonoid biosynthesis is revealed to diverge from naringenin chalcone. In addition, identification of substrate now endorses the term “chalcone reductase” to CHR, which had not been proven to accept chalcone as a substrate before, and answers the long-standing questions about doubly-labeled acetic acid uptake pattern in deoxyflavonoid biosynthesis.

CHR is a member of the AKR superfamily,^12–14) and belongs to AKR4 subfamily composing several plant oxidoreductase of poorly defined function.³⁵⁾ CHR was revealed to reduce naringenin chalcone to isoliquiritigenin through removing a phenolic hydroxyl group. No AKR reducing a phenolic hydroxyl group is reported. The results of this study would contribute to not only the further research on the reaction mechanism of CHR, but also identifying the functions of other plant AKRs.

Conflict of Interest

The author declares no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1) Hegnauer R, Gpayer-Barkmeijer RJ. Relevance of seed polysaccharides and flavonoids for the classification of the leguminosae: a chemotaxonomic approach. Phytochemistry, 34, 3–16 (1993).
• 2) Kreuzaler F, Hahlbrock K. Enzymatic synthesis of aromatic compounds in higher plants: formation of naringenin (5,7,4′-trihydroxyflavanone) from p-coumaroyl coenzyme A and malonyl coenzyme A. FEBS Lett., 28, 69–72 (1972).
• 3) Austin MB, Noel JP. The chalcone synthase superfamily of type III polyketide synthases. Nat. Prod. Rep., 20, 79–110 (2003).
• 4) Ayabe S, Udagawa A, Furuya T. NAD(P)H-dependent 6′-deoxychalcone synthase activity in Glycyrrhiza echinata cells induced by yeast extract. Arch. Biochem. Biophys., 261, 458–462 (1988).
• 5) Hakamatsuka T, Noguchi H, Ebizuka Y, Sankawa U. Deoxychalcone synthase from cell suspension cultures of Pueraria lobata. Chem. Pharm. Bull., 36, 4225–4228 (1988).
• 6) Welle R, Grisebach H. Isolation of a novel NADPH-dependent reductase which coacts with chalcone synthase in the biosynthesis of 6′-deoxychalcone. FEBS Lett., 236, 221–225 (1988).
• 7) Schröder J, Schröder G. Stilbene and chalcone synthases: related enzymes with key functions in plant-specific pathways. Z. Naturforsch. C J. Biosci., 45, 1–8 (1990).
• 8) Welle R, Schröder G, Schiltz E, Grisebach H, Schröder J. Induced plant responses to pathogen attack: analysis and heterologous expression of the key enzyme in the biosynthesis of phytoalexins in soybean (Glycine max L. Merr. cv. Harosoy 63). Eur. J. Biochem., 196, 423–430 (1991).
• 9) Ballance GM, Dixon RA. Medicago safiva cDNAs encoding chalcone reductase. Plant Physiol., 107, 1027–1028 (1995).
• 10) Sallaud C, El-Turk J, Bigarré L, Sevin H, Welle R, Esnault R. Nucleotide sequences of three chalcone reductase genes from alfalfa. Plant Physiol., 108, 869–870 (1995).
• 11) Springob K, Nakajima J, Yamazaki M, Saito K. Recent advances in the biosynthesis and accumulation of anthocyanins. Nat. Prod. Rep., 20, 288–303 (2003).
• 12) Bomati EK, Austin MB, Bowman ME, Dixon RA, Noel JP. Structural elucidation of chalcone reductase and implications for deoxychalcone biosynthesis. J. Biol. Chem., 280, 30496–30503 (2005).
• 13) Hyndman D, Bauman DR, Heredia VV, Penning TM. The aldo-keto reductase superfamily homepage. Chem. Biol. Interact., 143–144, 621–631 (2003).
• 14) Penning TM. The aldo-keto reductases (AKRs). Overview. Chem. Bio.l Interact., 234, 236–246 (2015).
• 15) Mameda R, Waki T, Kawai Y, Takahashi S, Nakayama T. Involvement of chalcone reductase in the soybean isoflavone metabolon: identification of GmCHR5, which interacts with 2-hydroxyisoflavanone synthase. Plant J., 96, 56–74 (2018).
• 16) Capponi M, Gut IG, Hellrung B, Persy G, Wirz J. Ketonization equilibria of phenol in aqueous solution. Can. J. Chem., 77, 605–613 (1999).
• 17) Shimokoriyama M. Interconversion of chalcones and flavanones of a phloroglucinol-type structure. J. Am. Chem. Soc., 79, 4199–4202 (1957).
• 18) Heller W, Hahlbrock K. Highly purified “flavanone synthase” from parsley catalyzes the formation of naringenin chalcone. Arch. Biochem. Biophys., 200, 617–619 (1980).
• 19) Stoessl A, Stothers JB. The incorporation of [1,2-¹³C₂]-acetate into pisatin to establish the biosynthesis of its polyketide moiety. Z. Naturforsch. C J. Biosci., 34, 87–89 (1979).
• 20) Dewick PM, Steele MJ, Dixon RA, Whitehead IM. Biosynthesis of isoflavonoid phytoalexins: incorporation of sodium [1,2-¹³C₂]-acetate into phaseollin and kievitone. Z. Naturforsch. C J. Biosci., 37, 363–368 (1982).
• 21) Light RJ, Hahlbrock K. Randomization of the flavonoid A ring during biosynthesis of kaempferol from [1,2-¹³C₂] acetate in cell suspension cultures of parsley. Z. Naturforsch. C J. Biosci., 35, 717–721 (1980).
• 22) Ayabe S, Furuya T. Studies on plant tissue cultures. Part 36. Biosynthesis of a retrochalcone, echinatin, and other flavonoids in the cultured cells of Glycyrrhiza echinata. A new route to a chalcone with transposed A- and B-rings. J. Chem. Soc., Perkin Trans. 1, 2725–2734 (1982).
• 23) Forkmann G, Heller W. Biosynthesis of flavonoids. Comprehensive Natural Products Chemistry, Chapter 1, Elsevier, Amsterdam, pp. 713–748 (1999).
• 24) Ayabe S, Uchiyama H, Aoki T, Akashi T. Plant Phenolics: Phenylpropanoids. Comprehensive Natural Products Chemistry II, Chapter 1, Elsevier, Amsterdam, pp. 929–976 (2010).
• 25) Sütfeld R, Wiermann R. Chalcone synthesis with enzyme extracts from tulip anther tapetum using a biphasic enzyme assay. Arch. Biochem. Biophys., 201, 64–72 (1980).
• 26) Yamamoto T, Yoshimura M, Yamaguchi F, Kouchi T, Tsuji R, Saito M, Obata A, Kikuchi M. Anti-allergic activity of naringenin chalcone from a tomato skin extract. Biosci. Biotechnol. Biochem., 68, 1706–1711 (2004).
• 27) Siddaiah V, Rao CV, Venkateswarlub S, Subbarajub GV. A concise synthesis of polyhydroxydihydrochalcones and homoisoflavonoids. Tetrahedron, 62, 841–846 (2006).
• 28) Gutmann A, Bungaruang L, Weber H, Leypold M, Breinbauer R, Nidetzky B. Towards the synthesis of glycosylated dihydrochalcone natural products using glycosyltransferase-catalysed cascade reactions. Green Chem., 16, 4417–4425 (2014).
• 29) Akiyama T, Shibuya M, Liu HM, Ebizuka Y. p-Coumaroyltriacetic acid synthase, a new homologue of chalcone synthase, from Hydrangea macrophylla var. thunbergia. Eur. J. Biochem., 263, 834–839 (1999).
• 30) Stöckigt J, Zenk MH. Chemical syntheses and properties of hydroxycinnamoyl-coenzyme A derivatives. Z. Naturforsch. C J. Biosci., 30, 352–358 (1975).
• 31) Tropf S, Kärcher B, Schröder G, Schröder J. Reaction mechanisms of homodimeric plant polyketide synthase (stilbenes and chalcone synthase): a single active site for the condensing reaction is sufficient for synthesis of stilbenes, chalcones, and 6′-deoxychalcones. J. Biol. Chem., 270, 7922–7928 (1995).
• 32) Boland MJ, Wong E. Purification and kinetic properties of chalcone-flavanone isomerase from soya bean. Eur. J. Biochem., 50, 383–389 (1975).
• 33) Shimada N, Aoki T, Sato S, Nakamura Y, Tabata S, Ayabe S. A cluster of genes encodes the two types of chalcone isomerase involved in the biosynthesis of general flavonoids and legume-specific 5-deoxy(iso)flavonoids in Lotus japonicus. Plant Physiol., 131, 941–951 (2003).
• 34) Burke JR, La Clair JJ, Philippe RN, Pabis A, Corbella M, Jez JM, Cortina GA, Kaltenbach M, Bowman ME, Louie GV, Woods KB, Nelson AT, Tawfik DS, Kamerlin SCL, Noel JP. Bifunctional substrate activation via an arginine residue drives catalysis in chalcone isomerases. ACS Catal., 9, 8388–8396 (2019).
• 35) Krishnamurthy P, Pothiraj R, Suthanthiram B, Somasundaram SM, Subbaraya U. Phylogenomic classification and synteny network analyses deciphered the evolutionary landscape of aldo–keto reductase (AKR) gene superfamily in the plant kingdom. Gene, 816, 146169 (2022).