Trans-o-Hydroxycinnamic Acid Glucosylation in Cell-Free Extracts of Melilotus alba

The gIucosylation of trans-o-hydroxycinnamic acid has been demonstrated in cell-free extracts of MeliIotus alba. The reaction requires UDPG, a sulfhydryl compound, and an active extract from sweet clover leaves. The product of the reaction has been tentatively identified as trans-/3-D-glucosyl-o-hydroxycinnamic acid. I N T R O D U C T I O N Low molecular weight phenols which occur in living cells are normally present in a combined form.' Synthesis of the combined form often involves the linking of the hydroxyl group of the phenol to a D-glucose molecule. Limited information regarding the enzymes that catalyze this type of conjugation is available. Yamaha and Cardini2 obtained an enzyme from wheat germ which catalyzed the formation of arbutin (hydroquinone-/?-D-glucoside) from hydroquinone and uridine diphosphate glucose (UDPG). The enzyme was relatively nonspecific since a large number of diand tri-phenols served as substrates, but it did not attack monophenols. Insect tissues also are a source of enzymes which can catalyze the formation of arbutin from hydroquinone and UDPG.3,4 Barber 5 , described the enzymatic glycosylation of quercetin (3,3',4',5,7-pentahydroxyflavone) to form rutin (quercetin-3-O-a-~-rhamnosyl-(I-t6)-&~-glucoside). Extensive in vivo and in vitro experiments led Pridham and Saltmarsh to conclude that the glucosylation of mono-, di-, and tri-hydric phenols produces primarily the corresponding mono-/3-glucosides. They, however, did not exclude the formation of oligoglucosides or glucosylation of more than one hydroxyl group. Pridham8 examined the ability of various plant tissues to convert quinol and resorcinol to the corresponding mono-/?-D-glucopyranosides in vivo. This activity was prominent in the angiosperms and gymnosperms, but appeared entirely absent in the algae and fungi. More recently, Corner and Swain9 reported the formation of glucose esters ofp-coumaric, * Cooperative investigations of the Crops Research Division, Agricultural Research Service, United States Department of Agriculture, and the Nebraska Agricultural Experiment Station. Supported in part by The National Science Foundation (Grant No. GB-1148). Published with the approval of the Director as Paper No. 2065, Journal Series, Nebraska Agricultural Experiment Station. 1 J. B. HARBORNE ( ditor), Biochemistry ofPhenolic Compounds, Chap. 4. Academic Press, New York (1964). 2 T. YAMAHA and C. E. CARDINI, Arch. Biochem. Biophys. 86, 127 (1960). 3 G. J. DUTTON and A. M. DUNCAN, Biochem. J. 77,18P (1960). 4 J. C. TRIVELLONI, Arch. Biochem. Biophys. 89, 149 (1960). 5 G. A. BARBER, Biochemistry 1,463 (1962). 6 G. A. BARBER, Biochem. Biophys. Res. Cornmun. 8,204 (1962). 7 J. B. PRIDHAM and M. J. SALTMARSH, Biochem. J. 87,218 (1963). 8 J. B. PRIDHAM, Phytochem. 3,493 (1964). 9 J. J. CORNER and T. SWAIN, Nature 207,634 (1965). 83 1313 1314 A. KLEINHOFS, F. A. HASKINS and H. J. GORZ caffeic, ferulic, and sinapic acids upon incubation of an acetone powder prepared from leaves of geranium (Geranium zonale cv. "Paul Crampel") with UDPG at pH 8.4 and 37". No esters were detected at pH 5.7 or 7.4, but esters were formed in small quantities at pH 9.4 and 10.4. No glucosides were formed from the hydroxycinnamic acids in this experiment. Sweet clover (Melilotus alba Desr.) plants of the CuCu genotype contain large amounts of o-coumaric acid glucoside and coumarinic acid glucoside, the transand cis-isomers, respectively, of /I-D-glucosyl-o-hydroxycinnamic acid.1° Experiments with sweet clover 1 1 3 l2 and Hierochloe odorata l3 have shown that when o-coumaric acid is administered to these plants in vivo it is rapidly converted to its P-glucoside. Kosuge and Connll referred to unpublished observations which indicated that the synthesis of o-coumaric acid glucoside occurred in reaction mixtures containing o-coumaric acid, UDPG, and a cell-free extract from white sweet clover shoots. Further information about the enzymatic catalysis of this reaction has not been found in the literature to date. RESULTS AND DISCUSSION Zdentz2cation of the Product of the Reaction Crude enzyme preparation (0.4 ml) was incubated with 0-756 pmoles o-coumaric acid+ 14C (sp. act. 0.091 pclpmole), 2.0 pmoles UDPG, 50 pmoles KH2P04 buffer, and 25 pmoles jl-mercaptoethanol in a total volume of 2.0 ml at pH 7.4 for 4 hr at 30'. The reaction was stopped by adding 2 ml methanol. The resulting suspension was centrifuged to remove the precipitate and the supernatant fraction was reduced to a small volume under a stream of air at room temperature. This concentrated solution was examined by paper chromatography and found to contain, in addition to free o-coumaric acid, a radioactive spot with Rfidentical to authentic o-coumaric acid glucoside in 2% acetic acid (Solvent A), n-propanol-conc, NH40H-H20 (8 : 1 : 1, v/v/v) (Solvent B), andn-butanol-glacial acetic acid-H20 (4: 1 : 5, v/v/v, upper phase) (Solvent C). When the chromatogram was sprayed with emulsin (5 mg/ml), a fluorescent spot appeared corresponding exactly with the radioactive spot. The color of the fluorescence appeared identical to that of o-coumaric acid. The residual o-coumaric acid, and the spot tentatively identified as o-coumaric acid glucoside, were the only radioactive peaks detected on the chromatogram with the exception of one very small peak. This small peak was later shown to occur as a result of a nonenzymatic reaction between o-coumaric acid an4 probably, P-mercaptoethanol. Under the standard assay conditions the nonenzyrnatic reaction was not detected. To determine whether the glucose ester of o-coumaric acid was a product of the reaction, mixtures with increased o-coumaric acid, UDPG and enzyme levels were incubated at pH 7.5 and 8.3. The reaction products were examined by paper chromatography. Spraying with emulsin revealed only one spot with the fluorescence of o-coumaric acid. Emulsin would be expected to hydrolyze both the glucose ester and the /3-glucoside of o-coumaric acid.14 To determine if the ester could be a minor product that did not separate from the glucoside with the solvents used, portions of the reaction mixtures were streaked on Whatrnan No. 3 paper and developed in solvent B alongside authentic o-coumaric acid glucoside. The band corresponding to the glucoside was eluted and the absorption spectrum determined over a 10 F. A. HA~KINS and H. J. GORZ, Biochem. Biophys. Res. Commun. 6,298 (1961). 11 T. KOSUGE and E. E. CONN, J. Biol. Chem. 234,2133 (1959). 12 F. A. HASKINS and T. KOSUGE, Genetics 52,1059 (1965). 13 S. A. BROWN, G. H. N. TOWERS and D. WRIGHT, Can. J. Biochem. Physiol. 38,143 (1960). 14 J. B. HARBORNE and J. J. CORNER, Biochem. J. 81,242 (1961). Trans-o-hydroxycinnamic acid glucosylation in cell-free e x t r a c t s of Melilotus alba 1315 range of 230-350 nm in 0.05 N HCI and 0.05 N NaOH. Although there was considerable contamination as evidenced by high absorption at the lower wavelengths (250 nm and shorter), the A,,, of the eluates corresponded exactly with that of authentic o-coumaric acid glucoside. Furthermore, the spectra of the eluates showed the same characteristic hypsochromic shift as the authentic o-coumaric acid glucoside sample with change from acid to alkali conditions. There was no indication of the bathochromic shift typical of cinnamic acid esters on changing from acid to alkali conditions.14 The participation of coumarinic acid or its glucoside was investigated by using an incubation mixture in which standard assay volumes were multiplied by four. For this experiment the crude enzyme preparation was precipitated with (NH4)2S04 (40-60 % sat.), and the precipitate was washed with (NH4),S04 (70 % sat.) to remove the endogenous o-hydroxycinnamic acid compounds. The samples were assayed for o-coumaric and coumarinic acids and their glucosides. No free or glucosidically bound coumarinic acid was detected over a 4-hr incubation period. Decrease in content of o-coumaric acid corresponded almost exactly with the increase in the content of o-coumaric acid glucoside (Fig. 1).