Mandelonitrile Lyase from Ximenia Americana L.: Stereospecificity and Lack of Flavin Prosthetic Group (Cyanogenesis/Cyanohydrin) GARY W
Total Page：16
File Type：pdf, Size：1020Kb
Proc. Natl. Acad. Sci. USA Vol. 86, pp. 6978-6981, September 1989 Botany Mandelonitrile lyase from Ximenia americana L.: Stereospecificity and lack of flavin prosthetic group (cyanogenesis/cyanohydrin) GARY W. KUROKI* AND ERIC E. CONNt Department of Biochemistry and Biophysics, University of California, Davis, CA 95616 Contributed by Eric E. Conn, June 29, 1989 ABSTRACT A mandelonitrile lyase (EC 4.1.2.10) that been purified to apparent electrophoretic homogeneity and catalyzes the dissociation of (S)-(-)-mandelonitrile to benzal- its properties have been examined. dehyde and hydrogen cyanide has been purified to apparent homogeneity from leaves ofXimenia americana L. (Olacaceae). MATERIALS AND METHODS The lyase was purified 122-fold with 38% yield by chromatog- raphy on carboxymethyl-celiulose and chromatofocusing. The Materials. Leaves of X. americana were collected at Fort enzyme had a pH optimum of 5.5, with a K. value of 280 ,uM. Desoto Park (Saint Petersburg, FL) and shipped to Davis, Activity toward 4-hydroxy-(R,S)-mandelonitrile was 77% of California within 24 hr of harvesting. Samples employed in that observed with the endogenous substrate; no activity was cyanide analysis were lyophilized prior to shipment to Davis. observed toward the aliphatic substrate acetone cyanohydrin. CM-cellulose was purchased from Whatman. Gel filtration The enzyme was stable at 40C and at room temperature for at media, Polybuffer 74, and Polybuffer Exchanger were ob- least 1 month. Native and subunit molecular weights of 38,000 tained from Pharmacia. (R,S)-Mandelonitrile, almond and 36,500, respectively, suggest the enzyme is a monomer. emulsin, fluorescein isothiocyanate-labeled concanavalin A, Schiff's reagent, Amberlite XAD-4, gel filtration, and SDS/ The isoelectric point was pH 3.9 as determined by isoelectric PAGE protein standards were purchased from Sigma. Iso- focusing. Staining with periodic acid-Schiff and fluorescein- electric focusing (IEF) PAGplates were obtained from LKB. labeled concanavalin A reagents indicate this enzyme is a Acrylamide, N, N '-methylenebisacrylamide, SDS, and glycoprotein. In contrast to (R)-mandelonitrile lyases isolated N, N, N', N'-tetramethylethylenediamine (TEMED) were from Prunus species, the Ximenia lyase does not appear to be purchased from Bio-Rad. Thiol reagants, metal salts, and a flavoprotein. A second enzyme that eluted from the chro- metabolites were from our laboratory collection. matofocusing column at pH 4.0 was also active toward man- Cyanide Analysis. Cyanide determinations were performed delonitrile. However, this form accounted for less than 10% of as described (13). the total activity, and its specific activity was only 6% of that Enzyme Purification. Leaves ofX. americana (100 g) were of the major component. Additional physical and kinetic homogenized in an ice-cold blender by addition of 2 vol of studies suggested this activity may be due to a nonspecific cold (-200C) acetone, followed by filtration through a Buch- enzyme that is active toward mandelonitrile. ner funnel. This procedure was repeated twice and the final powder was stored at -20'C. The following steps were (R)-(+)-Mandelonitrile, produced by hydrolysis of cyano- carried out at 40C. Protein was extracted by adding acetone genic glucosides that occur in members of the family Ro- powder (30 g) to 35 ml of0.1 M Mes-KOH (pH 6.3) containing saceae, is reversibly dissociated into benzaldehyde and HCN 25% (vol/vol) glycerol and 3 g of Amberlite XAD-4. The by the enzyme mandelonitrile lyase, an a-hydroxynitrile slurry was filtered through two layers of cheesecloth and Iyase (EC 4.1.2.10) (1). Pfeil and associates (2-4) reported centrifuged for 20 min at 17,000 x g. Subsequent procedures that these enzymes, when isolated from the genus Prunus, were carried out at room temperature. The supernatant was were flavoproteins and speculated on the role of the flavin decanted and chromatographed on a Sephadex G-25 column cofactor in a reaction not involving oxidation/reduction. (2.5 x 41 cm) equilibrated with 50 mM Mes-KOH (pH 6.0) Other workers (5-8) have examined the role of the flavin containing 25% glycerol. Fractions (7.0 ml) were monitored moiety and this has been the subject of considerable debate. for protein by absorbance at 280 nm. Those containing Although the (R)-(+)-mandelonitrile lyases of rosaceous protein were pooled and chromatographed on a CM-cellulose species appear to be flavoproteins (1), the analogous enzyme column (1.5 x 5 cm) equilibrated with 50 mM Mes-KOH (pH in Sorghum bicolor that utilizes 4-hydroxy-(S)-mande- 6.0) containing 25% glycerol. Fractions (5 ml) were assayed lonitrile as a substrate does not contain a flavin cofactor (9). for Iyase activity. The active fractions were pooled and dialyzed overnight against 4 liters of25 mM histidine-HCI (pH Moreover, lyases in cassava (10) and flax (11), which utilize 6.2). The dialyzed protein preparation was chromatographed the aliphatic substrate acetone cyanohydrin, do not appear to on Polybuffer Exchanger 94 (1.2 x 56 cm), equilibrated with be flavoproteins. Thus the (R)-(+)-mandelonitrile lyase of 25 mM histidine hydrochloride (pH 6.2). Material was eluted rosaceous species is atypical in its requirement for a flavin in Polybuffer 74, pH 3.8. Fractions (6.0 ml) were collected cofactor. and assayed for Iyase activity. Active fractions were pooled The occurrence of the glucoside of (S)-(-)-mandelonitrile and concentrated to 1 ml by ultrafiltration and stored at 40C. in Ximenia americana (family Olacaceae) (12) provided an Mandelonitrile Lyase Assays. Lyase assays were performed opportunity to examine whether the lyase that acts on the by monitoring the decomposition of (R,S)-mandelonitrile. enantiomer of (R)-(+)-mandelonitrile is a flavoprotein. Ac- The amount of benzaldehyde produced was measured by cordingly the a-hydroxynitrile Iyase of X. americana has Abbreviation: IEF, isoelectric focusing. The publication costs of this article were defrayed in part by page charge *Present address: Department of Botany and Plant Sciences, Uni- payment. This article must therefore be hereby marked "advertisement" versity of California, Riverside, CA 92521. in accordance with 18 U.S.C. §1734 solely to indicate this fact. tTo whom reprint requests should be addressed. 6978 Downloaded by guest on October 2, 2021 Botany: Kuroki and Conn Proc. Natl. Acad. Sci. USA 86 (1989) 6979 Table 1. Purification of mandelonitrile lyase from X. americana Volume, Protein, Activity, Specific activity, Purification Yield, Fraction ml mg units units/mg factor % Sephadex G-25 89 48.1 178 3.7 100 CM-cellulose 144 1.9 137 72 19 77 Chromatofocusing Lyase 1 25 0.21 5.8 28 8 3 Lyase 2 45 0.15 68 453 122 38 CM, carboxymethyl. recording the absorbance at 249.6 nm (e = 13.2 X 103 serum albumin (Mr, 66,000), egg albumin (Mr, 45,000), car- M-1-cm-l). Standard reaction mixtures contained 100 mM bonic anhydrase (Mr, 29,000), and ribonuclease (Mr, 13,700). Mes-KOH (pH 5.5), 395 AM (R,S)-mandelonitrile, and up to 100 ng of protein in a final volume of 1.0 ml. Reactions were AND initiated by adding enzyme and the increase in absorbance RESULTS DISCUSSION was then followed. Since (R,S)-mandelonitrile also decom- Cyanide Analysis. The cyanide content of leaves of X. poses slowly at this pH in the absence of enzyme, the americana was determined to be 52 ,umol of HCN per g (dry nonenzymic rate was determined for each assay and sub- weight). This compares favorably with the value [115 ,mol of tracted from the total rate to obtain the true enzymic rate. The HCN per g (dry weight)] found in the first report of cyano- assay was linear for reaction times of 1-3 min and in the range genesis in this species (12). of 10-100 ng of protein. One unit of lyase activity is defined Purification of Mandelonitrile Lyase. Attempts to extract as the amount of enzyme that catalyzes the production of 1 mandelonitrile lyase from leaves of X. americana proved to ,umol of benzaldehyde per min. be difficult due to a high concentration of phenolic com- Substrate specificity studies with acetone cyanohydrin and pounds. Homogenization with aqueous buffers containing a 4-hydroxy-(R,S)-mandelonitrile were performed as de- variety of phenolic adsorbents and antioxidants consistently scribed (14, 15). resulted in extremely low yields. However, it was possible to Studies to determine the stereospecificity of the Ximenia obtain highly active extracts from acetone powders. The and almond Iyases were performed by adding 153 units of inclusion of 25% glycerol in extraction and CM-cellulose Ximenia to 50 of and buffers was also critical in maintaining enzyme activity, but lyase gmol (R,S)-mandelonitrile allow- the presence ofglycerol resulted in reduced flow rates at 4°C. ing the reaction to proceed until the reaction rate was Therefore, the purification was carried out at room temper- equivalent to the nonenzymic rate. An equal amount of ature. Stability studies demonstrated the lyase activities to be Ximenia lyase was then added to ensure the reaction was stable for up to 1 month at room temperature. Such thermo- indeed complete. The amount of benzaldehyde released was stability has also been observed for the Sorghum lyase (9). then determined. Almond lyase (150 units, obtained from Upon application ofthe desalted Ximenia homogenate to a almond emulsin) was then added and the reaction was CM-cellulose column, the lyase activity did not bind and was allowed to proceed as described above. The amount of eluted in the wash fraction (data not shown). This procedure additional benzaldehyde formed was then determined. Re- resulted in 19-fold purification with 77% recovery of enzyme ciprocal assays were performed in which the almond lyase activity (Table 1). Subsequent purification by chromatofo- was added first and then Ximenia lyase was added.