Identification of Fatty Acid Glucose Esters as ... - ACS Publications

Oct 19, 2015 - The Os9BGlu31 gene is most strongly expressed in senescing flag leaf and developing seed and is induced in rice seedlings in response t...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/JAFC

Identification of Fatty Acid Glucose Esters as Os9BGlu31 Transglucosidase Substrates in Rice Flag Leaves Juthamath Komvongsa,†,‡ Bancha Mahong,§ Kannika Phasai,†,‡ Yanling Hua,‡,∥ Jong-Seong Jeon,§ and James R. Ketudat Cairns*,†,‡,⊥ †

School of Biochemistry, Institute of Science, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand Center for Biomolecular Structure, Function and Application, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand § Graduate School of Biotechnology and Crop Biotech Institute, Kyung Hee University, Yongin 446-701, Korea ∥ Center for Scientific and Technological Equipment, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand ⊥ Laboratory of Biochemistry, Chulabhorn Research Institute, Bangkok 10210, Thailand ‡

S Supporting Information *

ABSTRACT: Rice Os9BGlu31 transglucosidase transfers glucosyl moieties between various carboxylic acids and alcohols, including phenolic acids and flavonoids, in vitro. The role of Os9BGlu31 transglucosidase in rice plant metabolism has only been suggested to date. Methanolic extracts of rice bran and leaves were found to contain oleic acid and linoleic acid to which Os9BGlu31 could transfer glucose from the 4-nitrophenyl β-D-glucoside (4NPGlc) donor to form 1-O-acyl glucose esters. Os9BGlu31 showed higher activity with oleic acid (18:1) and linoleic acid (18:2) than with stearic acid (18:0) and had both a higher kcat and a higher Km for linoleic than oleic acid in the presence of 8 mM 4NPGlc donor. Os9BGlu31 knockout mutant rice lines were found to have significantly larger amounts of fatty acid glucose esters than wild-type control lines. Because the transglucosylation reaction is reversible, these data suggest that fatty acid glucose esters act as glucosyl donor substrates for Os9BGlu31 transglucosidase in rice. KEYWORDS: fatty acid, transglycosidase, flag leaf, rice mutant, tandem mass spectrometry

■

INTRODUCTION Glucosyl conjugates comprise glucosides and glucosyl esters, including glycolipids, glycoproteins, polysaccharides, and a variety of secondary metabolites. Many glucosyl conjugates are found in plants, in which they play roles as structural components, either active forms or inactive storage forms of hormones, chemical defense agents and other bioactive compounds, nonvolatile storage forms of aromatic scent components, and metabolic intermediates, among others.1 The production and recycling of these compounds are catalyzed by glucosyltransferases, glycoside hydrolases, and transglucosidases. Glycoside hydrolase family 1 (GH1) enzymes include βglucosidases (EC 3.2.1.21) with a range of specificities, as well as other β-glycosidases and transglycosidases.2 These enzymes work through a retaining mechanism, in which a β-glucosidic bond is broken with acid catalysis to form a covalent intermediate with the enzyme nucleophile residue in the first (glycosylation) step and the enzyme is displaced by an external nucleophile in the second (deglycosylation) step with basic assistance. If the nucleophile for the deglycosylation step is water, hydrolysis occurs, whereas if another nucleophile displaces the catalytic nucleophile, transglycosylation occurs. Many GH1 β-glucosidases catalyze transglycosylation as well as hydrolysis reactions, and recently, GH1 transglycosidases, which catalyze transglycosylation with little or no hydrolysis, have been reported.3−5 © XXXX American Chemical Society

GH1 transglycosidases that have been described include transgalactosidases and transglucosidases.3,4 Galactolipid-galactolipid galactosyl transferase, a chloroplast outer membrane enzyme that transfers galactose from monogalactosyl diacyl glyceride (MGDG) to another MGDG or digalactosyl diacyl glyceride (DGDG) to make glycolipids with longer oligosaccharide chains, was found to be the GH1 protein Sensitive-toFreezing-2 (SFR2).3 GH1 enzymes were also found to be acyl glucose-dependent anthocyanin glucosyl transferases (AAGT, a type of transglucosidase), responsible for the addition of a glucosyl moiety onto cyanidin 3-O-glucoside in carnation and delphinium flowers.4 These AAGTs were found to catalyze transglycosylation with no detectable hydrolysis. AAGTs were demonstrated to work with serine carboxypeptidase-like (SCPL) acyl glucose-dependent acyl transferases to build up large anthocyanin complexes by a series of sugar and acyl transfers to the anthocyanin backbone.6 The Arabidopsis GH1 enzyme AtBGLU10 was proposed to play a similar role in anthocyanin synthesis.7 Rice (Oryza sativa) Os9BGlu31 is a GH1 transglycosidase that acts to transfer glucose among phenolic acids, phytohormones, and flavonoids.5 The highest activity was observed with the donors feruloyl glucose, 4-coumaroyl Received: August 23, 2015 Revised: October 15, 2015 Accepted: October 19, 2015

A

DOI: 10.1021/acs.jafc.5b04105 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

mM fatty acids in 140 μL of 50 mM citrate (pH 4.5) at 30 °C for 1 h. The product of a larger reaction with oleic acids was purified by silica gel TLC, eluted from the resin in 50% methanol, and submitted to 1H NMR to confirm the structure of 1-O-oleoyl β-D-glucose ester. A similar preparation of 1-O-linoleoyl β-D-glucose was not possible, because of the instability of this product. 1H NMR for oleic acid glucosyl ester was performed on a 500 MHz NMR spectrometer (Bruker AVANCE III HD) with a CPP BBO 500 Cryoprobe. CD3OD was used as the solvent, and the 1H NMR spectra were collected at a frequency of 500.366 MHz. The 1H NMR experiment data were processed with BRUKER TOPSPIN version 3.2. The 1H peaks (shown in Figure S1 of the Supporting Information) were assigned as follows: δ 0.90 (3H, t, J = 6.9 Hz, H18), 1.28 (20H, m, H4−H7, H12− H17), 1.61 (2H, m, H3), 2.03 (4H, m, H8, H11), 2.36 (2H, t, J = 7.9 Hz, H2), 5.35 (2H, t, J = 5.0 Hz, H9 and H10), 3.35 (1H, m, H2′), 3.57 (1H, m, H4′), 3.67 (1H, m, H3′), 3.84 (1H, m, H5′), 4.06 (1H, m, H6′), 4.14 (1H, m, H6′), 5.48 (1H, J = 8.5 Hz, H1′). To confirm the production of 1-O-linoleoyl β-D-glucose, enzymatic and control reaction mixtures were prepared in D2O and monitored by 1 H NMR. The 500 μL reaction mixtures contained 20 mM 4NPGlc, 20 mM linoleic acid, and 50 mM citrate (pH 4.5). To start the enzymatic reaction, 280 μg of Os9BGlu31 was added and the reaction mixtures were incubated at 30 °C for 3 and 24 h before being assayed with 1H NMR. The NMR was as described for oleoyl glucose, but the water peak was suppressed. The spectra are shown in Figure S2 of the Supporting Information. Kinetic Parameters of Os9BGlu31 Transglucosylation of Fatty Acids. The relative activity toward 4NPGlc donor substrates was determined by incubating 10 μg of Os9BGlu31 with 20 mM 4NPGlc and 0.25 mM fatty acids in 140 μL of 50 mM citrate (pH 4.5) at 30 °C for 1 h. The apparent Km and Vmax values of 4NPGlc in the presence of various acceptors were determined by varying the concentration of 4NPGlc in the range of 0.2−10 mM with 0.25 mM fatty acid acceptors and 10 μg of Os9BGlu31 in 50 mM citrate (pH 4.5). The apparent Km and Vmax values of the fatty acid acceptors were determined by varying their concentrations between 0.05 and 2 mM in the presence of 8 mM 4NPGlc with 10 μg of Os9BGlu31 in 50 mM sodium citrate (pH 4.5). The reaction was stopped by adding 100 μL of 2 M Na2CO3, and the 4NP released was quantified by the absorbance at 405 nm. The kinetic parameters were determined by nonlinear regression of the Michaelis−Menten plots with GraFit version 5.0 (Erithacus Software, Horley, U.K.). Growth and Extraction of Os9BGlu31 Mutant and Wild-Type Rice Lines. Rice plants were grown in an experimental field plot of Kyung Hee University under natural environmental conditions in the summer.9 The os9bglu31 knockout mutant alleles were identified from the population of T-DNA-tagged mutants10,11 and from the Tos17 insertion mutant collection.12 The japonica cultivars Dongjin and Nipponbare are the background genotypes of os9bglu31-1 and os9bglu31-4, and os9bglu31-2 and os9bglu31-3, respectively. Homozygous lines for insertions in the Os9BGlu31 gene and wild-type segregant lines from the same heterozygous parents were compared. Homozygous mutants for the insertions were identified by polymerase chain reaction (PCR) analysis of genomic DNA isolated from mature rice leaves using the following gene-specific primers: for os9bglu31-1, 5′-CCTCGCTCCCCAATACTATTTTGA-3′ and 5′-GAAGACGATGGAAACAAACACATC-3′; for os9glu31-2 and os9bglu31-3, 5′CTGGAGCACTGTCAATGAGCCTAA-3′ and 5′-TCAACCCGAGTGGGAACTGTTATT-3′; and for os9bglu31-4, 5′-GCTCTTGTGAAATTATACTGTCTG-3′ and 5′-ATTCTATATCTATCTTGGACCACC-3′. The T-DNA-specific primer 5′-ATCCAGACTGAATGCCCACAGG-3′ for os9bglu31-1 and os9bglu31-4 and the Tos17-specific primer 5′-CTGGACATGGGCCAACTATACAGT-3′ for os9bglu31-2 and os9bglu31-3 were used to determine mutation sites of each os9bglu31 mutant allele. RNA Isolation and RT-PCR Analysis. Total RNA from leaves of 2-month-old mutants and their respective wild types was prepared using Trizol reagent (Invitrogen, Carlsbad, CA). The isolated RNA extracts were reverse-transcribed with an oligo-dT primer and a First Strand cDNA Synthesis Kit (Roche, Mannheim, Germany). The first-

glucose, and sinapoyl glucose, which are known to serve as donors in acyl and glucosyl transfer reactions in the vacuole in plant cells,6 where Os9BGlu31 is localized. The free acids of these compounds also served as the best acceptors, suggesting that Os9BGlu31 may equilibrate the levels of phenolic acids and carboxylated phytohormones and their esters.5 In the presence of glucosyl acceptor substrates, no hydrolysis product (glucose) was detected, although low levels of hydrolysis of the donor substrate were detected in the absence of a viable acceptor substrate other than water. The Os9BGlu31 gene is most strongly expressed in senescing flag leaf and developing seed and is induced in rice seedlings in response to drought stress, and treatment with phytohormones, including ABA, ethephon, methyljasmonate, 2,4-D, and kinetin. Mutagenesis revealed that Trp243 is a critical residue for determining substrate specificity, with the Trp243Asn variant providing higher activity for both transglycosylation and hydrolysis than wild-type Os9BGlu31.8 Nevertheless, the natural substrates of the native enzyme in the plant remain unknown. In this work, we sought to identify natural Os9BGlu31 substrates and products in rice. First, we identified substrates in rice bran extracts that Os9BGlu31 could transglycosylate with 4-nitrophenyl β-D-glucopyranoside (4NPGlc) as a donor. The kinetics of transglycosylation for the identified free fatty acid substrates were determined and compared to those of previously characterized substrates. The relative concentrations of the free fatty acids and their glucosyl esters in os9bglu31 knockout mutant and wild-type rice leaf extracts were determined by quantitative ultraperformance liquid chromatography−triple quadrupole tandem mass spectrometry (UPLC− MS/MS QQQ) to show that oleic acid and linoleic acid glucose esters are likely to be glucose donor substrates for Os9BGlu31 in the rice plants.

■

MATERIALS AND METHODS

Extraction, Purification, and Identification of Substrates from Rice Bran. Initially, 1 g of rice bran (cultivar KDML105, from a mill in Nakhon Ratchasima, Thailand) was extracted in 50% ethanol overnight. The supernatant was separated via silica gel thin layer chromatography (TLC) on silica gel plates developed with a 7:2.8:0.2 chloroform/methanol/30% ammonia solvent system, and component spots were detected by staining with 10% sulfuric acid in ethanol and heating at 110 °C. The extract was separated via TLC without staining, and the regions corresponding to the major stained spots were scraped, collected separately, and eluted in 100% methanol overnight. The crude extract and TLC-purified components were used as substrates in a reaction with 10 μg of Os9BGlu31 (produced from recombinant Escherichia coli as previously described),5 10 mM 4NPGlc, and 10% ethanol extract in 50 mM sodium citrate (pH 4.5). The reaction mixtures were separated via TLC, as described above, to identify the component serving as the substrate for Os9BGlu31. To purify the substrate, 20 g of rice bran was extracted with 50% ethanol in water (HPLC grade) and the mixture stirred overnight at room temperature. The mixed extract was filtered through a 0.45 μm filter. Ten milliliters of supernatant was loaded into a Waters Sep-Pak tC18 resin column (Waters, Milford, MA). The column was washed with 10 mL of water, and 20 mL each of 5, 50, 60, 70, 80, 90, and 100% methanol. Each fraction was evaluated via silica gel TLC developed with a chloroform/methanol/30% ammonia [7:2.8:0.2 (v/ v/v)] solvent, stained with 10% sulfuric acid in ethanol, and charred at 110 °C. Fractions eluted in 70, 80, and 90% methanol that appeared to act as substrates were evaluated by 1H NMR (proton nuclear magnetic resonance spectroscopy). To determine whether oleic acid and linoleic acid were substrates, 10 μg of Os9BGlu31 was incubated with 20 mM 4NPGlc and 0.25 B

DOI: 10.1021/acs.jafc.5b04105 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry strand cDNAs served as templates in the RT-PCR with the Os9BGlu31 gene-specific primers, 5′-CTCAATCATCTCCAAGAGAAGT-3′ and 5′-AACAGGTACTCGAACACGTC-3′, as previously described.13 The rice Ubiquitin5 gene (OsUBQ5)-specific primers 5′-GACTACAACATCCAGAAGGAGTC-3′ and 5′-TCATCTAATAACCAGTTCGATTTC-3′ were used to amplify the control. Characterization and Identification of Fatty Acids in Rice Flag Leaves as Os9BGlu31 Substrates. Flag leaves from 3-monthold plants were collected, freeze-dried, and ground to powder. Samples (50 mg) were extracted with 500 μL of 70% ethanol for UPLC−MS/ MS QQQ analysis. The amounts of fatty acids and their glucose esters in rice flag leaf extracts were quantified by triple quadrupole LC−ESIMS/MS and NMR, with oleic acid and linoleic acid standards (CalBiochem, La Jolla, CA). Ten micrograms of Os9BGlu31 was used to catalyze the reaction of 20 mM 4NPGlc with 5 mM oleic acid or linoleic acid in 50 mM sodium citrate buffer (pH 4.5) at 30 °C for 1 h. The reactions were stopped with 1% formic acid and the mixtures filtered through a 0.22 μm filter before injection into an Agilent 1290 LC system (Agilent) in line with an electrospray ionization (ESI) Agilent 6400 series QQQ mass spectrometer with MassHunter software. Separation was conducted through an Agilent SB-C18 RRHD 1.8 μm, 2.1 mm × 150 mm column (Agilent). The mobile phase consisted of 0.2% formic acid in water (solvent A) and 0.2% formic acid in acetonitrile (solvent B). The gradient varied linearly from 40 to 60% B (v/v) over 13 min and from 60 to 100% B (v/v) over 15 min with a flow rate of 0.3 mL/min. The mass spectrometry analysis was performed in negative ion mode, with a capillary voltage of 3 kV and a gas temperature of 300 °C. The gas flow was set at 16 L/ min, nebulizer at 45 psi with the sheath gas heater at 300 °C, and the sheath gas flow at 11 L/min. The fatty acid product ions were identified for each precursor, and the fragmentation voltage was optimized to yield the highest product ion abundance. Finally, the specific compounds were detected by multiple-reaction monitoring (MRM). The abundance of oleic acid and linoleic acid was monitored with a collision energy of 15 V. The product ion at m/z 185.1 generated from the precursor ion at m/z 281 was monitored for oleic acid, while the m/z 211 product ion of the m/z 279 precursor was monitored for linolenic acid. The selected product ions for monitoring the abundance of oleoyl glucose ester and linoleoyl glucose ester were generated with a collision energy of 20 V and were at m/z 281.2 (precursor ion at m/z 489 with formate) and m/z 279 (precursor ion at m/z 441), respectively. Statistical Analysis. Analysis of variance (ANOVA) was performed using SPSS software (SPSS 17.0 for Windows, SPSS Inc., Chicago, IL). Triplicate determinations were evaluated on each test, and averages were used in the experiment. The differences among mean values were established using Duncan’s multiple-range test at a 95% significance level.

Figure 1. Thin layer chromatography of transglycosylation of rice bran oleic acid, linoleic acid, and rice bran substrate. The products of reactions of 4NPGlc with oleic acid, linoleic acid, and purified rice bran substrate catalyzed by Os9BGlu31 were separated via silica gel TLC in a 7:2.8:0.2 (v/v/v) chloroform/methanol/30% ammonia solvent system and detected by charring with sulfuric acid. The lanes show 4-nitrophenyl β-D-glucoside standard (4NPGlc), glucose standard (Glc), linoleic acid standard (LA), oleic acid standard (OA), and purified rice bran substrate (Rice bran), followed by reactions with (+) and without (−) Os9BGlu31 enzyme with linoleic acid, oleic acid, and purified rice bran substrate.

Os9BGlu31-catalyzed reactions of commercial oleic acid and linoleic acid with 4NPGlc yielded products that migrated at the same position as the crude rice bran extract and semipurified fraction (Figure 1). The reaction with oleic and 4NPGlc showed a new product (M − H + HCO2−)− at m/z 489, and that with linoleic acid at m/z 441 (M − H)− on UPLC−MS/ MS QQQ in negative ion mode as shown in Figure S3, matching the expected masses of oleoyl glucose ester and linoleoyl glucose ester within the error of the machine. Examples of the selected ion abundance chromatograms for all compounds are shown in Figure S4. The 1H NMR spectrum of the purified oleic acid product was consistent with its assignment as 1-O-oleoyl β-D-glucose (see Materials and Methods), with the anomeric proton (H1) chemical shift of 5.48 ppm and the coupling constant of 8.5 Hz confirming the compound’s β-configuration as shown in Figure S1. 1-O-Linoleoyl β-D-glucose could not be purified because of the instability and complete loss of the compound after drying steps. Therefore, the Os9BGlu31-catalyzed reaction of 4NPGlc and linoleic acid was monitored by 1H NMR to verify the structure of 1-O-linoleoyl β-D-glucose. After reaction for 3 h, βanomeric peak H1 of glucose was observed at 5.57 ppm with a coupling constant of 8.2 Hz, similar to the peaks observed for the oleoyl β-D-glucose anomeric proton (Figure S2). A set of multiple peaks with an average chemical shift of 5.52 ppm that appeared in the reaction were assigned as the hydrogen atoms on the two double bonds of the acyl group and were similar to those observed with 1-O-oleoyl β-D-glucose. After overnight reaction (at which point the enzyme had lost activity), the intensity of the peak of the acyl glucose anomeric doublet had decreased, while that of the acyl double bond peaks had not, suggesting a shift of the linoleoyl group away from the anomeric position.17

■

RESULTS AND DISCUSSION Identification of Fatty Acids as a Substrate of Os9BGlu31 in Rice. Initially, seedling and bran extracts were tested for potential acceptor substrates to which Os9BGlu31 could transfer a glucose, but bran was found to give a clearer result (data not shown). Silica gel TLC showed that a product was formed from the crude extract, and that the TLC-purified band corresponding to a light brown spot (Rf = 0.53) migrating ahead of 4NPGlc (Rf = 0.50) in the chloroform/methanol/ammonia solvent system gave a more intense product spot. The glucosyl conjugate product migrated faster than the substrate in the ammonia-containing solvent system (Figure 1) but slower in a system in which water replaced ammonia, suggesting the substrate was a carboxylic acid that would be ionized in the ammonia solvent system. Purification of this component by reversed phase chromatography gave a product with NMR peaks consistent with a mixture containing oleic acid and linoleic acid. C

DOI: 10.1021/acs.jafc.5b04105 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

and 1-O-linoleoyl glucose as donors. The kcat/Km values for the fatty acids as acceptor substrates were 0.5−1% of the values previously determined for ferulic acid.5 Nevertheless, we proceeded to investigate the levels of the substrates and products in os9bglu31 mutant rice lines. Production of Os9BGlu31 Mutant Rice Lines. To examine whether the disruption of the Os9BGlu31 gene interferes with any transglucosidase function in rice leaves, we identified null mutants for the Os9BGlu31 gene (Figure 3A).

In the initial report of brassins from rape (Brassica napus L.), the presence of fatty acid 1-O-acyl glucose esters was noted, although the active component was later identified as brassenosteroids.14−16 These authors suggested the fatty acyl glucose esters may have a variety of structural and signaling roles, although none were demonstrated. Tanaka et al.17 synthesized a 1-O-oleoyl β-D-glucose/α-D-glucose mixture chemically, although with low yield due in part to a poor purification yield. He found that it had no activity in the bean second-internode bioassay. When attempting to similarly produce 1-O-linoleoyl β-D-glucose, he obtained 3-O-linoleoyl D-glucose, apparently because of acyl migration during the synthesis. Notably, the 3-O-linoleoyl D-glucose did show bioactivity in the pollen germination and pollen tube extension assay. The migration of linoleic acid from the 1-O- to 3-Oposition is consistent with our NMR time course result, although the spectrum in the sugar region was too crowded to interpret this with the small amount of product produced. Kinetics of Os9BGlu31 Transglucosylation of Fatty Acids. Comparison of the relative rates of glucosyl transfer to stearic acid (18:0), oleic acid (18:1Δ9), and linoleic acid (18:2Δ9,12) at a 4NPGlc donor substrate concentration of 20 mM showed that the unsaturated fatty acids were glucosylated at ∼2 times the rate of stearic acid and that linoleic acid had the highest rate (Figure 2). No significant glucose release was seen

Figure 3. Isolation and characterization of the Os9BGlu31 mutants. (A) Schematic diagrams of the rice Os9BGlu31 gene and the insertion positions of T-DNA (os9bglu31-1 and os9bglu31-4) and Tos17 retrotransposon (os9bglu31-2 and os9bglu31-3). The 12 OsBGlu31 exons are indicated by boxes. (B) RT-PCR analysis of os9bglu31 mutant alleles. Os9BGlu31 mRNA is not detectable in the mutant alleles. OsUBQ5 transcripts serve as a PCR control. Samples are: DJ, Dongjin wild type; NB: Nipponbarewildtype; 31-1 to 31-4: os9bglu31 knockout lines.

The mutant alleles os9bglu31-1 and os9bglu31-4 were found to contain T-DNA insertions in the 5′-untranslated region (UTR) and the 11th intron, respectively. The additional two alleles, os9bglu31-2 and os9bglu31-3, harbored Tos17 insertions in the eighth exon. Genomic DNA PCR amplification of the gene insertion regions with Os9BGlu31-specific and T-DNA- or Tos17-specific primers allowed isolation of rice lines homozygous for the insertions from segregating progeny (Figure 3A). RT-PCR analysis using the gene-specific primers indicated that the transcripts for endogenous Os9BGlu31 are absent from all the isolated homozygous mutants (Figure 3B), indicating that they are all null knockout mutants. We did not observe any obvious visible phenotype from the mutants under our paddy field growth condition. Comparison of Fatty Acids and Esters in Flag Leaf Extracts of Wild-Type and Os9BGlu31 Knockout Rice. To test whether free fatty acids could act as natural substrates of Os9BGlu31 in planta, we quantified oleic acid, linoleic acid, and their glucose esters by UPLC−MS/MS QQQ. No consistent pattern of change in the free fatty acid concentrations was seen in os9bglu31 knockout lines, although levels of oleic acid (10% increase; P < 0.05) and linoleic acid (60% increase; P < 0.05)

Figure 2. Relative rates of glycosylation of the fatty acids oleic acid, linoleic acid, and stearic acid. Reaction mixtures contained 10 μg of Os9BGlu31, 0.25 mM fatty acid acceptors, and 20 mM 4NPGlc in 50 mM citrate buffer (pH 4.5) at 30 °C for 30 min. The reaction rates were measured as release of 4-nitrophenol from the donor substrate. Under the conditions tested, the hydrolysis of 4NPGlc was negligible.

under these and similar conditions,5 as evidenced by the lack of glucose product seen via TLC in Figure 1. However, the kcat/ Km for oleic acid was >2-fold higher than that for linoleic acid, because of the lower Km for oleic acid (Table 1). Because purification of the 1-O-fatty acyl glucose esters yielded very low recoveries and they could not be recovered after drying, we were unable to test the activities of the enzyme with 1-O-oleoyl

Table 1. Kinetics for Os9BGlu31 Utilization of 4NPGlc in the Presence of Fatty Acid Acceptor Substratesa acceptor substrate kinetics

a

−1

acceptor

Km (mM)

kcat (s )

oleic acid linoleic acid ferulic acidb

0.149 ± 0.014 0.57 ± 0.09 0.05 ± 0.004

0.0436 ± 0.0015 0.074 ± 0.004 1.21 ± 0.06

donor substrate (4NPGlc) kinetics kcat/Km (mM

−1

−1

s )

0.291 ± 0.004 0.128 ± 0.028 25.4

Km (mM)

kcat (s−1)

kcat/Km (mM−1 s−1)

0.619 ± 0.006 1.14 ± 0.13 9.33 ± 0.62

0.0269 ± 0.0006 0.034 ± 0.0018 1.21 ± 0.20

0.0434 ± 0.005 0.0298 ± 0.004 0.13

Ferulic acid data from ref 5 are included for comparison. bPreviously published values for reactions with a ferulic acid acceptor.5 D

DOI: 10.1021/acs.jafc.5b04105 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 4. Fatty acids for Os9BGlu31 knockout and WT rice flag leaf extracts. The dark bars indicate the measured specific abundances of ions for the designated compound in the knockout lines, and the light gray bars indicate the amount in the corresponding wild-type lines. Shown are data for (A) oleic acid, (B) linoleic acid, (C) oleoyl glucose ester, and (D) linoleic acid glucose ester.

appeared to be significantly elevated in the os9bglu31-1 line (Figure 4A,B). On the other hand, levels of oleoyl glucose ester (Figure 4C) and linoleoyl glucose ester (Figure 4D) increased in all knockout lines compared to those of their paired wildtype segregant lines. The levels of fatty acyl glucose esters are higher in knockout than wild-type lines, with 4.1−9.2-fold higher abundances for oleoyl glucose (P < 0.05) and 4.8−13.6fold higher abundances for linoleoyl glucose (P < 0.05). The similar effects in four independent mutant lines prove that this buildup is specific to the knockout of Os9BGlu31 and does not result from effects on another gene. The levels of 1-O-fatty acyl glucose esters are very low in the wild-type line, suggesting that increasing the levels of Os9BGlu31 is likely to have little effect on the 1-O-oleoyl and 1-O-linoleoyl glucose ester levels in the plant. The fact that the os9bglu31 knockout rice plants showed no obvious differences in their morphological phenotypes, despite the changes in the 1-O-acyl glucose fatty acid ester concentrations, suggests that the effects of these compounds are subtle under normal growth conditions. The discovery of free fatty acids in rice bran extracts acting as glucosyl acceptor substrates in vitro suggested that they might act as substrates in the plant to produce the 1-O-acyl β-Dglucose esters. However, the levels of 1-O-acyl glucose esters of oleic acid and linoleic acid were increased at least 4-fold in each os9bglu31 knockout line, suggesting these 1-O-acyl glucose esters serve primarily as substrates rather than products of Os9BGlu31 in rice flag leaves. Although Os9BGlu31 is expressed in seeds, which led us to explore bran as a convenient source of possible products, Os9BGlu31 is most strongly expressed in flag leaves,5 so the increase in the 1-O-acyl glucose fatty acid esters in the flag leaves appears to indicate a significant function for the enzyme. In vitro, Os9BGlu31 acts primarily as a transglucosidase when carboxylic acids and phenolic acids are present and no hydrolysis products can be

detected, even at low concentrations of these acceptor substrates.5 There are numerous compounds available to act as acceptors in the plant cell vacuoles. For instance, benzoic acid, a strong acceptor substrate for Os9BGlu31 in vitro,5 along with several other carboxylic acids was reported in barley mesophyll protoplast vacuoles.18 Nonetheless, we cannot exclude the possibility that Os9BGlu31 hydrolyzes the glucose esters to produce free glucose and fatty acids in the plant, because other factors in the vacuole may affect its activity. Because the previous observation of the fatty acid 1-O-acyl glucose esters in rape was not substantiated in later work,19 our observation helps to prove the presence of 1-O-oleoyl glucose and 1-O-linoleoyl glucose in plant species. The facts that the reverse phase elution position and parent and product ion masses observed in the plant are the same as the standards generated by in vitro reactions suggest that they are at least the fatty acid acyl glucoses, although whether they linked to glucose carbon 1 could be debated. However, the buildup of these compounds in mutant plants that do not express the Os9BGlu31 enzyme, which acts exclusively on β-D-glucosyl conjugates at the anomeric carbon,5 confirms that the compounds are 1-O-oleoyl and 1-O-linoleoyl β-D-glucose esters. It has been shown that short chain and branched chain fatty acid 1-O-acyl glucose esters are generated by UDP-glucosedependent glucosyltransferases in tomato trichomes.20 It is anticipated that the 1-O-fatty acyl glucose esters are generated by a similar mechanism in rice leaves and can then serve as glucosyl donors for Os9BGlu31. The role of these 1-O-fattyacyl glucose molecules and that of Os9BGlu31 in modulating their amounts will require further investigation. E

DOI: 10.1021/acs.jafc.5b04105 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

■

(7) Miyahara, T.; Sakiyama, R.; Ozeki, Y.; Sasaki, N. Acyl-glucosedependent glucosyltransferase catalyzes the final step of anthocyanin formation in Arabidopsis. J. Plant Physiol. 2013, 170, 619−624. (8) Komvongsa, J.; Luang, S.; Marques, J. V.; Phasai, K.; Davin, L. B.; Lewis, N. G.; Ketudat Cairns, J. R. Active site cleft mutants of Os9BGlu31 transglucosidase modify acceptor substrate specificity and allow production of multiple kaempferol glycosides. Biochim. Biophys. Acta, Gen. Subj. 2015, 1850, 1405−1414. (9) Eom, J. S.; Cho, J. I.; Reinders, A.; Lee, S. W.; Yoo, Y.; Tuan, P. Q.; Choi, S. B.; Bang, G.; Park, Y. I.; Cho, M. H.; Bhoo, S. H.; An, G.; Hahn, T.-R.; Ward, J. M.; Jeon, J.-S. Impaired function of the tonoplast-localized sucrose transporter in rice, OsSUT2, limits the transport of vacuolar reserve sucrose and affects plant growth. Plant Physiol. 2011, 157, 109−119. (10) Jeon, J. S.; Jang, S.; Lee, S.; Nam, J.; Kim, C.; Lee, S.-H.; Chung, Y.-Y.; Kim, S.-R.; Lee, Y. H.; Cho, Y.-G.; An, G. T-DNA insertional mutagenesis for functional genomics in rice. Plant J. 2000, 22, 561− 570. (11) Jeong, D. H.; An, S. Y.; Kang, H. G.; Moon, S.; Han, J. J.; Park, S.; Lee, H. S.; An, K. S. An G. H. T-DNA insertional mutagenesis for activation tagging in rice. Plant Physiol. 2002, 130, 1636−1644. (12) Miyao, A.; Tanaka, K.; Murata, K.; Sawaki, H.; Takeda, S.; Abe, K.; Shinozuka, Y.; Onosato, K.; Hirochika, H. Target site specificity of the Tos17/2 retrotransposon shows a preference for insertion within genes and against insertion in retrotransposon-rich regions of the genome. Plant Cell 2003, 15, 1771−1780. (13) Cho, J. I.; Lee, S. K.; Ko, S.; Kim, H. K.; Jun, S. H.; Lee, Y. H.; Bhoo, S. H.; Lee, K. W.; An, G.; Hahn, T. R.; Jeon, J. S. Molecular cloning and expression analysis of the cell-wall invertase gene family in rice (Oryza sativa L.). Plant Cell Rep. 2005, 24, 225−236. (14) Goffman, F. D.; Pinson, S.; Bergman, C. Genetic diversity for lipid content and fatty acid profile in rice bran. J. Am. Oil Chem. Soc. 2003, 80, 485−490. (15) Mitchell, J. W.; Mandava, N.; Worley, J. F.; Plimmer, J. R.; Smith, M. V. Brassins-a new family of plant hormones from rape pollen. Nature 1970, 225, 1065−1066. (16) Mandava, N. B. Plant growth-promoting brassinosteroids. Annu. Rev. Plant Physiol. Plant Mol. Biol. 1988, 39, 23−52. (17) Tanaka, A. K.; Kobayashi, A.; Yamashita, K. Synthesis and biological activities of 1-O-glucopyranosyl fatty acid esters. Agric. Biol. Chem. 1979, 43, 2537−2542. (18) Tohge, T.; Ramos, M. S.; Nunes-Nesi, A.; Mutwil; Giavalisco, P.; Steinhauser, D.; Schellenberg, M.; Willmitzer, L.; Persson, S.; Martinoia, E.; Fernie, A. R. Towaed the storage metabolome: profiling the barley vacuole. Plant Physiol. 2011, 157, 1469−1482. (19) Grove, M. C.; Spencer, G. F.; Pfeffer, P. E.; Mandava, N.; Warthen, J. D., Jr.; Worley, J. F. 6-D-Glucopyranosyl fatty acid esters from Brassica napus pollen. Phytochemistry 1978, 17, 1187−1189. (20) Ghangas, G. S.; Steffens, J. C. UDP-glucose:fatty acid transglucosylation and transacylation in triacylglucose biosynthesis. Proc. Natl. Acad. Sci. U. S. A. 1993, 90, 9911−9915.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jafc.5b04105. 1 H NMR spectra of purified 1-O-oleoyl β-D-glucose in CD3OD (Figure S1), 1H NMR spectra of the production of 1-O-linoleoyl β-D-glucose from linoleic acid and 4NPGlc catalyzed by Os9BGlu31 (Figure S2), mass spectra of 1-O-oleoyl β-D-glucose and 1-O-linoleoyl β-Dglucose (Figure S3), and selected ion chromatogram of flag leaf extracts from wild-type and knockout Os9BGu31 rice lines (Figure S4) (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Telephone: +66 44 224304. Author Contributions

J.K. and B.M. should be considered co-first authors. Funding

This work was supported by grants from Suranaree University of Technology, the Commission on Higher Education National Research University Project Grant to SUT, and the Royal Golden Jubilee Ph.D. Program of the Thailand Research Fund grant to J.K. and J.R.K.C. (Grant PHD/0001/2552) and from the Mid-Career Researcher Program of the National Research Foundation (NRF-2013R1A2A2A01068887 to J.-S.J.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Sukanya Luang is thanked for conducting preliminary extraction experiments. Prasat Kittakoop is thanked forNMR analysis of rice bran substrate fractions. We are also grateful to Rice Genome Resource Center, National Institute of Agrobiological Sciences, Japan, for providing Tos17 rice mutants.

■

REFERENCES

(1) Ketudat Cairns, J. R.; Esen, A. β-glucosidase. Cell. Mol. Life Sci. 2010, 67, 3389−3405. (2) Ketudat Cairns, J. R.; Pengthaisong, S.; Luang, S.; Sansenya, S.; Tankrathok, A.; Svasti, J. Protein-carbohydrate interactions leading to hydrolysis and transglycosylation in plant glycoside hydrolase family 1 enzymes. J. Appl. Glycosci. 2012, 458, 2011−2022. (3) Moellering, E. R.; Benning, C. Galactoglycerolipid metabolism under stress: a time for remodeling. Trends Plant Sci. 2011, 16, 98− 107. (4) Matsuba, Y.; Sasaki, N.; Tera, M.; Okamura, M.; Abe, Y.; Okamoto, E.; Nakamura, H.; Funabashi, H.; Takatsu, M.; Saito, M.; Matsuoka, H.; Nagasawa, K.; Ozeki, Y. A novelglucosylationreaction on anthocyanins catalyzed by acyl-glucose-dependent glucosyltransferase in the petals of carnation and delphinium. Plant Cell 2010, 22, 3374−3389. (5) Luang, S.; Cho, J. I.; Mahong, B.; Opassiri, R.; Akiyama, T.; Phasai, K.; Komvongsa, J.; Sasaki, N.; Hua, Y.; Matsuba, T.; Ozeki, Y.; Jeon, J. S.; Ketudat Cairns, J. R. Rice Os9BGlu31 is a transglucosidase with the capacity to equilibrate phenylpropanoid, flavonoid, and phytohormone glycoconjugates. J. Biol. Chem. 2013, 288, 10111− 10123. (6) Nishizaki, Y.; Yasunaga, M.; Okamoto, E.; Okamoto, M.; Hirose, Y.; Yamaguchi, M.; Ozeki, Y.; Sasaki, N. p-Hydroxybenzoyl-glucose is a zwitter donor for the biosynthesis of 7-poly acylated anthocyanin in delphinium. Plant Cell 2013, 25, 4150−4165. F

DOI: 10.1021/acs.jafc.5b04105 J. Agric. Food Chem. XXXX, XXX, XXX−XXX