60708 (183)

Biosci. Biotechnol. Biochem., 71, 60708-1–11, 2007

Chemical Profiling and Gene Expression Profiling during the Manufacturing Process of Taiwan Tea ‘‘Oriental Beauty’’

y Jeong-Yong CHO,1 Masaharu MIZUTANI,1; Bun-ichi SHIMIZU,1 Tomomi KINOSHITA,1 Miharu OGURA,2 Kazuhiko TOKORO,2 Mu-Lien LIN,3 and Kanzo SAKATA1

1Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan 2Central Research Laboratory, Takasago International Corporation, 1-4-11 Nishi-yawata, Hiratsuka-shi, Kanagawa 254-0073, Japan 3Tea Research and Extension Station, 324 Chunghsin Road, Yangmei, Taoyuan 326, Taiwan

Received December 18, 2006; Accepted March 15, 2007; Online Publication, June 7, 2007 [doi:10.1271/bbb.60708]

Oriental Beauty, which is made from tea leaves in- infested by the tea green (J. formosana). The fested by the tea green leafhopper (Jacobiasca formosa- infested leaves (Fig. 1B) are small and yellowish as na) in Taiwan, has a unique aroma like ripe fruits and compared to healthy leaves (Fig. 1A). It is traditionally honey.Advance To determine what occurs in the tea Viewleaves dur- known that attack promotes the tea quality, es- ing the oolong tea manufacturing process, the gene pecially the strength of the unique aroma of the tea.1) expression profiles and the chemical profiles were inves- Oolong tea is generally produced from fresh tea tigated. Tea samples were prepared from Camellia si- leaves via complicated processes of plucking, solar nensis var. sinensis cv. Chin-shin Dah-pang while the tea withering, indoor withering, turn over, panning, rolling, leaves were attacked by the insect. The main volatile and drying, although various processing methods are compounds, such as linalool-oxides, benzyl alcohol, 2- used for other types of oolong tea.3) The typical man- phenylethanol, and 2,6-dimethylocta-3,7-diene-2,6-diol, ufacturing process for Oriental Beauty is illustrated in increased during manufacture. The gene expression Fig. 1C. There are unusual points in the manufacturing profiles during manufacture were analyzed by differ- process for OrientalProofs Beauty as compared to those for the ential screening between fresh leaves and tea leaves of other types of oolong tea. In Oriental Beauty manufac- the first turn over. Many up-regulated transcripts were ture, the steps of solar withering and indoor withering found to encode various proteins homologous to stress are longer, and the additional step, wetting and soften- response proteins. Accordingly, the endogenous contents ing, is done before rolling. Thus the degree of fermen- of abscisic acid and raffinose increased during manu- tation in Oriental Beauty is higher, and that may also be facture. Thus the traditional manufacturing method is a responsible for the promotion of tea quality. unique process that utilizes plant defense responses to Tea aroma is one of the most important factors in elevate the production of volatile compounds and other determining the character and quality of each tea, es- metabolites. pecially oolong tea and black tea. The main aroma con- stituents of oolong tea are linalool, geraniol, benzyl al- Key words: Oriental Beauty (Pom-Fong tea); oolong cohol, 2-phenylethanol, and linalool oxides.2,4,5) Most of tea; tea green ; stress response these alcoholic aroma compounds are mainly present gene; volatile compound as disaccharide glycosides, such as -primeveroside, - acuminoside, and -vicianoside, in fresh leaves of tea Oriental Beauty is a famous Formosa oolong tea that cultivars for oolong tea (C. sinensis var. sinensis cv. has a unique aroma like ripe fruit and honey.1,2) Other Shuixian and Maoxue) and for green tea (cv. Yabu- names are Pom-Fong tea, Champagne oolong tea, Chan kita).6,7) These glycosides are hydrolyzed to release Pin oolong tea, and White Tip oolong tea, etc. Oriental alcoholic aroma compounds by the action of endogenous Beauty is mainly produced in counties of Hsinchu, enzymes (-glucosidase and -primeverosidase) during Miaoli, Taoyuan, and Taipei in the northern part of the tea manufacturing process, suggesting that the hy- Taiwan. In Oriental Beauty manufacture, one of the drolysis of glycosides is one of the key factors in alco- most characteristic factors is the use of tea leaves holic aroma formation during the processing of oolong

y To whom correspondence should be addressed. Tel: +81-774-38-3230; Fax: +81-774-38-3229; E-mail: [email protected] Abbreviations: ABA, abscisic acid; diol, 2,6-dimethylocta-3,7-diene-2,6-diol; EST, expressed sequencing tags; NCED, 9-cis-epoxycarotenoid dioxygenase; AMI, 2-O-(-L-arabinopyranosyl)-myo-inositol 60708-2 J.-Y. CHO et al. AB

Healthy tea leaves Infested tea leaves (Non-infested leaves) by tea green leafhoppers C (Jacobiasca formosana) Cultivation Turn over Plucking Solar withering with /indoor withering (F) (4 h, SW) (3 weeks) (every 2 h, 5 times, T1-T5)

Wetting Panning Rolling Drying Product /softening

Fig. 1. Tea Leaves (Healthy and Infested) and Sampling during the Manufacture of Oriental Beauty. Sampling of tea leaves was done at each step of the manufacturing process (F, SW, and T1-T5). For chemical profiling of volatile compounds and metabolites, the sample at each step was subjected to the processes of panning, wetting and softening, rolling, and drying. For gene expression profiling, each sample was immediately frozen without panning. teaAdvance6–10) as well as black tea.11,12) Furthermore, View the cells 104 U2 and tissues of tea leaves are still alive before panning U1 (Fig. 1C), and therefore it is likely that many aroma 103 compounds are produced by de novo biosynthesis in tea leaves. It has been reported that emission of volatiles and 102 expression of the genes involved in volatile biosynthesis is induced by stresses such as insect attack and wound- 101 13–15) ing in various plant species. In addition to increased Intensity of Cy5 (T1) production of volatiles, it is well known that abiotic 100 Proofs stresses trigger a wide variety of plant responses, in- 100 101 102 103 104 cluding alteration in gene expression, the accumulation Intensity of fluorescein (F) of phytohormones such as abscisic acid (ABA), jas- monic acid, and ethylene, and enhanced synthesis of Fig. 2. Differential Screening Analysis with Fresh Leaves (F) and specific proteins (e.g., heat shock proteins and enzymes the Tea Leaves of First Turn Over (T1) during the Oolong Tea that function in various metabolite biosynthesis, such as Manufacturing Process. sugars and osmoprotective raffinose-family oligosaccha- U1, highly induced transcripts; U2, moderately induced tran- scripts. rides).16,17) These plant responses lead to acclimation and tolerance to environmental stresses. In the same way, fluctuation of various metabolites in tea leaves must occur in response to stresses such as UV-irra- during the oolong tea manufacturing process should give diation, drought, and wounding during the manufactur- us valuable information about a vast number of induc- ing process. However, little is known about what occurs ible genes, including stress-response genes and biosyn- in tea leaves during processing. thetic genes of aroma compounds. Recently, molecular biological approaches such as In this study, we report gene expression profiles and differential screening, expressed sequencing tags (EST), chemical profiles to understand what occurs in tea leaves and DNA microarray have been used to identify in- during the manufacturing process of Oriental Beauty. ducible genes in response to various stresses from many First, the changes in the chemical constituents of the plant tissues, including nonwoody plants (Arabidopsis, volatile compounds were examined during manufacture. tomato, Brassica)18–20) and woody plants (poplar, grape- We also investigated the gene expression profiling in tea fruit).21,22) In tea leaves (C. sinensis), Chen et al.23) leaves during the earlier stages of manufacture by dif- performed EST analysis of spring tender shoots. Park et ferential screening analysis. Although we could not find al.24) also reported EST analysis of a subtractive cDNA the transcripts to be directly related to aroma compound library between young leaves and mature leaves, and biosynthesis, we identified various stress response genes identified several genes involved in catechin biosyn- and identified the accumulation of ABA and raffinose thesis. In this way, transcript profiling in tea leaves in response to the manufacturing process. Accordingly, Chemical and Gene Profiling in the Manufacture of Taiwan Oolong Tea 60708-3 we also discuss the relation between the oolong tea thesized by T7 DNA polymerase using Cy5-uridine manufacturing method and plant stress response. triphosphate (UTP) and fluorescein UTP. The probed microbeads (4:0 105 beads, a total of 8:0 105 beads) Materials and Methods for the tea samples (F and T1) were mixed and hy- bridized overnight at 50 C in digoxigenin (DIG) Easy Materials. Tea leaves (cv. Chin-shin Dah-pang) in- Hyb (Roche Diagnostics, Basel, Switzerland). The hy- fested by tea green leafhoppers (J. formosana) for about bridized microbeads were washed in 1 saline sodium 3 weeks were plucked in Hsinchu County of Taiwan citrate (SSC)/1% sodium dodecyl sulfate (SDS), and in June 2004. The manufacture of Oriental Beauty was then in 0:1 SSC/0.1% SDS at 65 C. The microbeads carried out in the same county according to the tra- were sorted with a MoFlo fluorescence-activated cell ditional method, as shown in Fig. 1C. After the infested sorter (DakoCytomation, Fort Collins, CO). The distri- tea leaves were exposed to sunlight for 4 h as a solar bution of cDNAs up- and down-regulated during oolong withering process, the leaves were indoor-withered at tea manufacture is shown in Fig. 2. The up-regulated room temperature, accompanied by 5 repetitions (every cDNAs were further fractioned into two zones, highly 2 h) of indoor withering, followed by turn over. Then the induced genes (U1 group) and moderately induced genes leaves were parched at 180–190 C for 3–5 min. The (U2 group). The cDNAs of U1 group were separately leaves obtained were wrapped in a piece of wet cloth collected 200 microbeads each and stored for DNA and held at room temperature for 30–40 min as a wetting sequence analysis. and softening process, followed by rolling for 3–5 min. Drying at 70 C for 1.5 h gave the oolong tea products. Sequencing analysis. The cDNA mixtures on the Sampling of tea leaves was carried out, as shown in microbeads were amplified by universal primers, and Fig. 1C. Tea leaves were collected at each step of fresh the PCR fragments were ligated into a TOPO vector teaAdvance leaves to the fifth turn over (fresh leaves, View F; solar (Invitrogen, Carlsbad, CA). The ligated mixture was withering, SW; the first turn over, T1; the second turn transformed into competent cells of Escherichia coli over, T2; the third turn over, T3; the forth turn over, T4; DH5-T1 (Invitrogen) on ice for 30 min. The trans- the fifth turn over, T5) during oolong tea manufacture. formed cells were plated onto LB agar containing 50 mg/ For chemical profiling of volatile compounds and ml ampicillin, 40 ml X-gal (5-bromo-3-indolyl--D-gal- metabolites, samples at each step were subjected to the actoside, 20 mg/ml), and 40 ml of 20% isopropylthio-- processes of panning, wetting and softening, rolling, and D-galactopyranoside (IPTG), and incubated at 37 C drying. For gene expression profiling, each sample was overnight to obtain white colonies, which contained the immediately frozen without panning. In this study, we inserted cDNAs in the TOPO vector. obtained 20 g of each of the dried tea samples for The selected white colonies containing cDNA inserts Proofs chemical analyses and 10 g of each of the undried leaves were incubated at 37 C at 150 rpm overnight in LB for gene expression analysis. broth containing 50 mg/ml ampicillin. Plasmid cDNAs were isolated using a MagExtractor (Toyobo, Osaka, Total RNA extraction. Total RNA was extracted from Japan). The sequencing reaction was performed by the the tea leaves (F-T5) at each step of the oolong tea ma- BigDye Terminator ver. 1.1 Cycle Sequencing kit nufacturing process by a modified protocol of CTAB (Applied Biosystems) with M13 universal primers fol- (cetyltrimethylammonium bromide) according to Chang lowing the manufacturer’s instructions. Cycle sequence et al.25) reaction was carried out GeneAmp PCR system 9700 Thermal Cycler (Applied Biosystems, Foster City, CA) Differential screening analysis. Differential screening and sequenced on an ABI Prism 377 DNA sequencer analysis was performed using a DNA microbeads array- (Applied Biosystems). Each sequence was qualified with Megaclone and Megasort technique (Takara-Bio, Otsu, removal of vector sequences. The qualified sequences Japan), according to Kamauchi et al.26) Briefly, the in- of cDNAs were subjected to BLASTN and BLASTX fested fresh tea leaves (F) and the infested tea leaves at (www.ncbi.nlm.nih.gov) homolog searches.27) In addi- the step of the first turn over (T1) were used to analyze tion, the sequences of cDNAs were grouped using a differentially expressed genes in tea leaves during oo- DNASIS program (Hitachi Software, Tokyo, Japan). long tea manufacture. The cDNAs were prepared and The consensus sequence of each cDNA group obtained purified using an Oligotex-dT30 mRNA purification kit was also screened by BLASTX homolog search. (Takara-Bio) using 1.8 mg of poly (A)þ RNA with 50- biotin-conjugated anchored (dT19) primer. After soni- RT-PCR analysis. RT-PCR was performed using a cation treatment, the cDNAs were fractioned to a range ReverTra Dash RT-PCR Kit (Toyobo) according to the of 400–600 bp by electrophoresis and purified. The manufacturer’s protocols. First-strand cDNA was syn- cDNA fragments were cloned onto microbeads. The thesized in a 20 ml reaction mixture containing 0.5 mgof microbeads contained the 30-end of DpnII digested total RNA with an oligo (dT) primer, RNase inhibitor, cDNA reverse transcribed with MMLV-RTase (Takara- and a ReverTra Ace reverse transcriptase (Toyobo). The Bio). The probes for each cDNA template were syn- RT reactions were carried out at 42 C for 60 min and 60708-4 J.-Y. CHO et al. Table 1. Primers Used in RT-PCR Analysis

Former primers (50----30) Reverse primers (50----30) TPS TTGAGACGATTGCAAGATCTGTCAC AGGTACACAGAATGTGTATTTTGTGG RSA CAGGTAGGGGTGAAGGGGACTGGT CCCCAATTAAAATTAAGGAGTTGGGT RSB CCATTCAATTTTGAGCTCATAAC-GG TCAACAATGGACAGACCTGAACATC CHI TCAACTGTCATGGAACTACAACTATG TAGTCAATGTAGTACTTAACACGAGC NTD GGAAAAGATGATGAAATTATTGTCGG CAAATGAGACCTAACAGTCTCCTAC AMY TTCGAGGGCTCCTAATGGTTTCAAGC GATCATCCAAATGTAATTCCGGC NCED TCGCTAAAGTCGACTTGTTTTCTGG TTCGTGCCGTATTGTTACAGCAGAG 26S rRNA ATGAGTAGGAGGGCGCGGCGGT GGAGGCACTCGGTCCTCCGGAT

RSA, Raffinose synthase A; RSB, raffinose synthase B; TSP, trehalose 6-phosphate synthase; CHI, chitinase; NTD, Ntdin.; NCED,9-cis-epoxycarotenoid dioxygenase; AMY, -amylase

99 C for 5 min, followed by chilling to 4 C. The re- area of volatile compound to that of 9-nonanone (as action mixture was diluted 4 times with deionized water, internal standard). and 1.0 ml of aliquot was used as a template for PCR A series of tea samples in this study was prepared amplification. The PCR reaction was carried out in a from tea leaves infested by tea green leafhoppers. The 10-ml reaction mixture containing 2.5 mM MgCl2,10 insect attack usually occurs in a limited area of tea mM Tris–HCl, 50 mM KCl, 0.25 units of KOD dash Taq gardens for only a few weeks before the rainy season in polymerase (Toyobo), 2 mM dNTP, and 1.0 mM of each Taiwan. Preparation of the tea samples was done by an gene-specific primer (Table 1). The RT sample was first expert on Oriental Beauty. In this way, we applied 5 g of denatured at 94 C for 5 min. Semi-quantitative PCR tea samples in the volatile analysis, and the remaining amplificationAdvance was performed with an appropriate View num- samples were used for the other chemical analyses. ber of cycles (18, 26, or 28 cycles) of denaturation at 94 C for 30 s, annealing at 60 C for 30 s, and extension Carbohydrates analysis. Each tea sample (1.0 g) was at 74 C for 30 s using Takara PCR Thermal Cycler extracted with boiling water (10 ml) for 10 min and (Dice TP 600, Takara-Bio). 26S rRNA encoded from filtered. The filtrate was supplemented with polyvinyl- C. sinensis was used as a control. The PCR products polypyrrolidone (PVPP, 0.5 g) and stirred for 20 min. were analyzed with a 1% (w/v) agarose gel containing The solution was centrifuged at 10,000 rpm for 10 min. 20 ng/ml of ethydium bromide (Nacalai Tesque, Kyoto, The supernatant was freeze-dried to afford water ex- Japan). tracts. The extracts (equivalent to 10 mg of dry tea leaf sample) was analyzedProofs by HPLC under the following Volatile compound analysis by GC–MS. Preparation conditions: column, YMC polyamine column II (4.6 of volatiles of each sample was carried out by the i.d. 250 mm, 5 mm, YMC, Tokyo, Japan); mobile 28) brewed-extraction method using 9-nonanone as an phase, MeCN–H2O = 7:3 (v/v); flow rate, 1.0 ml/min; internal standard to get reliable data. Briefly, each tea injection volume, 10 ml; column temperature, 40 C; de- sample (5.0 g) was brewed with deionized water (75 g) tection, RI-101 detector (Shodex, Tokyo, Japan); HPLC at 70–80 C for 10 min. After filtration, the filtrate was pump, LC-10AD (Shimadzu). A liquid chromatograph– filled to 50 ml, and supplemented with 50 ml of an eth- electron spray ionization–mass spectrometer (LC–ESI– anol solution containing 0.01% 9-nonanone as internal MS; positive ion mode; electron voltage, 100 eV; API- standard. The solution was saturated with sodium chlo- 3000, Applied Biosystems) system was used to identify ride and partitioned with 20 ml of dichloromethane. The carbohydrates under the HPLC conditions described extract was dried over anhydrous sodium sulfate for above. 12 h. After the sodium sulfate was filtrated off, the solu- tion was carefully concentrated with a Kuderna-Danish Abscisic acid analysis. Each tea sample (2.5 g) was evaporative concentrator. GC/MS analysis was per- soaked in MeOH (10 ml) overnight, filtered, and con- formed on a GCMS-QP-2010 (Shimadzu, Kyoto, Japan), centrated. The MeOH extract was partitioned between equipped with a BC-WAX fused silica capillary column 2% NaHCO3 (30 ml) and CH2Cl2 (30 ml, 2 times). The (50 m 0:2mm i.d., film thickness 0.15 mm, GL Sci- aqueous layer (2% NaHCO3) was adjusted to pH 3.0 ences, Tokyo, Japan). The oven temperature was held at with 2 N HCl and extracted with CH2Cl2 (30 ml, 2 70 C and programmed to 220 Cat4C/min. Helium times). The extracts (equivalent to 30 mg of dry tea leaf was used as a carrier gas, and the flow rate was 18 cm/s. sample) was analyzed by HPLC under the following The split ratio was 1:50. Mass spectra were obtained at conditions: column, Cosmosil 5C18 AR-II (4.6 i.d. 70 eV (EI) at an ion source temperature of 200 C. 250 mm, 4.5 mm, Nacalai Tesque); mobile phase, Identification of the components was made by compar- MeOH–H2O with 1% acetic acid = 37.5:62.5 (v/v); ison of the GC retention times and mass spectra to those flow rate, 1.0 ml/min; injection volume, 10 ml; column of authentic compounds. The relative quantities of each temperature, 40 C; detection, Photodiode array detector volatile compound are shown as the ratio of the GC peak (254 nm, Waters, Milford, MA); HPLC pump, Delta 600 Chemical and Gene Profiling in the Manufacture of Taiwan Oolong Tea 60708-5 Table 2. Volatile Compound Contents during the Oolong Tea Man- to be at the highest level at T5 and dramatically in- ufacturing Process creased, about 400-fold, at T5 as compared to those at F. Area/I.S. Areaa The production of monoterpene alcohols such as gera- Compounds niol and linalool was greatly promoted during manu- FT5 facture, and linalool oxides and linalool-derived com- 1-Penten-3-ol 0.003 0.047 pounds such as 2,6-dimethylocta-1,7-diene-3,6-diol and Isoamyl alcohol 0.002 0.200 Amyl alcohol NDb 0.177 2,6-dimethylocta-2,7-diene-1,6-diol also increased sig- Hexanol ND 0.177 nificantly. In contrast, diol increased only 1.7-fold and (Z)-3-Hexenol ND 0.706 furthermore, 3,7-dimethylocta-1,5,7-trien-3-ol (hotrie- (E)-2-Hexenol 0.002 0.130 nol) decreased a little during the manufacturing process. Linalool 0.022 0.236 Six-carbon (C6) volatiles such as hexanal, hexanol, Geraniol 0.036 0.873 trans-Linalool-3,6-oxide (linalool oxide I) 0.040 1.759 hexenols, and hexenoic acids were hardly found at F, cis-Linalool-3,6-oxide (linalool oxide II) 0.067 1.701 but increased greatly at T5. Methyl jasmonate was found trans-Linalool-3,7-oxide (linalool oxide III) 0.108 1.593 at F but decreased to a trace at T5. cis-Linalool-3,7-oxide (linalool oxide IV) 0.142 1.212 3,7-Dimethylocta-1,5,7-trien-3-ol (hotrienol) 0.202 0.159 Differential screening between F and T1 2,6-Dimethylocta-3,7-diene-2,6-diol (diol) 1.816 3.058 To obtain transcripts differentially induced in tea 2,6-Dimethylocta-1,7-diene-3,6-diol ND 0.219 2,6-Dimethylocta-2,7-diene-1,6-diol 0.073 0.445 leaves during manufacture, differential screening analy- Benzyl alcohol 0.114 4.335 sis was performed between F and T1. There are 30,094 2-Phenylethanol 0.105 4.352 beads in up-regulated transcripts at T1 (Fig. 2). The up- Hexanal ND 0.040 regulated transcripts at T1 were further divided into two Benzaldehyde 0.001 0.237 Caproic acid 0.043 0.604 fractions, highly induced transcripts (U1 group, 5,674 Advance Viewbeads) and moderately induced transcripts (U2 group, (Z)-3-Hexenoic acid ND 0.471 (E)-2-Hexenoic acid ND 0.681 24,420 beads). To identify transcripts induced in tea Methyl jasmonate 0.038 trace leaves during manufacture, the highly induced tran- aVolatile compound contents are expressed as ratios of volatile compounds scripts (U1 group) were sequenced. (GC peak area)/9-nonanone (GC peak area) of internal standard. bND, not detected. Identification of up-regulated transcripts More than 1200 beads up-regulated in the U1 group were selected randomly, cloned, and sequenced. After (Waters). Cinnamic acid (Nacalai Tesque) was used as removal of the vector sequence, the sequences of 584 an internal standard. clones were qualifiedProofs and annotated by BLASTX and BLASXN database searches. Among 584 clones, the Results deduced amino acid sequences of 421 clones showed significant homology to known functional proteins, Volatile compounds except for those of the other 163 clones, which showed To characterize the change in volatile compound con- uncharacterized proteins (unknown proteins, 16 clones; tents during the oolong tea manufacturing process of unnamed proteins, 14 clones; and hypothetical proteins, Oriental Beauty, samples were prepared from tea leaves 133 clones). The sequences of the clones were grouped infested by tea green leafhoppers according to the tra- by DNASIS program. Each consensus sequence of the ditional method (Fig. 1). Fresh infested tea leaves (F) grouped clones was annotated by BLASTX database re- and infested tea leaves of the fifth turn over (T5) were search. These results are summarized in Table 3. There collected and subjected to the processes on and after were several abundant transcripts coding for proteins panning to obtain dried materials. Comparison of homologous to heat shock proteins (66 clones), DnaJ volatiles between F and T5 is shown in Table 2. In the proteins (51 clones), -amylase (46 clones), senescence- final tea product (T5), 24 volatile compounds, including associated protein (55 clones, including Ntdin), small 18 alcohols, 2 aldehydes, 3 acids, and 1 ester, were rubber particle proteins (34 clones), 9-cis-epoxycarote- identified. Seventeen volatile compounds excluding noid dioxygenase (NCED) (25 clones), ATP synthase hexanal, amyl alcohol, hexanol, (Z)-3-hexenol, hexenoic (26 clones), raffinose synthase (17 clones), and so on. acids, and 2,6-dimethylocta-1,7-diene-3,6-diol (diol), Several clones among them showed significant sequence were detected in F before the manufacturing process, similarities to various enzymes related to carbohydrate although the levels were quite low. In contrast, diol, metabolism, such as raffinose synthase, -amylase, tre- which is known as one of the characteristic volatile halose 6-phosphate synthetase, and glyceraldehydes-3- compounds of Oriental Beauty,28) showed the highest phosphate dehydrogenase. levels in F. Furthermore, transcripts involved in the biosynthesis Most volatile compound contents increased during the of abscisic acid (ABA) were also identified in the U1 manufacturing process (Table 2). Aromatic alcohols group. There were several clones homologous to the such as benzyl alcohol and 2-phenylethanol were found non-coding region of NCED from Vitis vinifera, and 60708-6 J.-Y. CHO et al. Table 3. Up-Regulated Transcripts in Tea Leaves during Oolong Tea Manufacture

Accession Genebank No. of Gene functions Origin E value no. match clones BJ999390 Trehalose 6-phosphate synthase Ginkgo biloba AAX16014 8e-13 4 BJ999391 Raffinose synthase Cucumis sativus AAD02832 2e-31 17 BJ999392 Raffinose synthase Cucumis sativus AAD02832 7e-33 2 BJ999393 Jasmonic acid 2 Solanum tuberosum AAF04915 0.083 13 BJ999394 Jasmonic acid 2 Solanum tuberosum AAF04915 8e-05 3 BJ999395 9-cis-Epoxycarotenoid dioxygenase Solanum tuberosum AAT75151 5e-37 1 BJ999396 9-cis-Epoxycarotenoid dioxygenasea Vitis vinifera AY337614 2e-13 24 BJ999397 Glutamine synthase Camellia sinesis BAD11327 0.008 2 BJ999398 Chitinase Oryza sativa BAA19793 3e-41 2 BJ999399 Hexose transporter protein Solanum lycopersicum CAA52689 3e-43 12 BJ999400 Bax inhibitor Nicotiana tabacum AAK73102 7e-35 12 BJ999401 Bax inhibitor Lycopersicon esculentum AAR28754 3e-43 3 BJ999402 Small rubber particle protein Hevea brassiliensis AAO66433 8e-34 34 BJ999403 -Amylase Prunus armeniaca AAD38148 6e-24 40 BJ999404 -Amylase Nicotiana langsdorffii AAM20167 7e-28 6 BJ999405 Ntdin (senescence-associated protein) Nicotiana tabacum BAA88985 2e-32 25 BJ999406 Senescence-associated protein Pisum sativum BAB33421 9e-57 6 BJ999407 Senescence-associated protein Arabidopsis thaliana CAB36774 4e-19 15 BJ999408 Senescence-associated protein Oryza sativa BAD37413 2e-13 5 BJ999409 Senescence-associated protein Pisum sativum BAB33421 2e-46 2 BJ999410 Senescence-associated protein Zea mays AAV31120 4e-04 2 BJ999411 DnaJ protein ABC69274 4e-47 51 AdvanceBJ999412 ATP synthase ViewArabidopsis thaliana P92549 3e-67 24 BJ999413 ATP synthase Arabidopsis thaliana P60112 1e-09 2 BJ999414 ATPase Petunia axillaris AAB03873 6e-21 2 BJ999415 ATPase Veronica incana AAW331007 5e-57 2 BJ999416 ATPase Sorghum bicolor YP 762326 1e-53 1 BJ999417 PRL1-Interacting factor Oryza sativa BAD87678 4e-06 3 BJ999418 40S ribosome protein Solanum tuberosum ABA40437 1e-08 3 BJ999419 Heat shock protein Medicago truncatula ABE89016 5e-08 37 BJ999420 Heat shock protein 83 Hevea brasiliensis AAQ08597 6e-59 8 BJ999421 Heat shock protein HSP82 Zea mays AAB26482 3e-27 9 BJ999422 Heat shock protein Hevea brasiliensis AAQ08597 2e-34 2 BJ999423 Small heat shock protein HSP22.3 Glycine max AAB03097 5e-32 8 BJ999424 Heat shock protein 70 Medicago truncatula ProofsABD32895 5e-12 1 BJ999425 Little protein Oryza sativa AAS55470 8e-28 17 BJ999426 Hypothetical protein Glycine max AAG00940 7e-36 12 BJ999427 Hypothetical protein Aurantimonas sp. ZP 01227369 1e-24 55 BJ999428 Hypothetical protein Nicotiana tabacum YP 173374 1e-41 66 BJ999429 Molybdenum cofactor sulfurase Oryza sativa BAD45965 8e-10 1 BJ999430 GPDb Helianthus annuus ABG35250 8e-23 1 BJ999431 Chlorophyll a/b binding protein Pisum sativum CAA57492 2e-21 1 BJ999432 Cheperonin 60 alpha subnit Canavalia lineata AAC68501 2e-29 1 BJ999433 Putative NAC domain protein Solanum tuberosum CAC42087 2e-16 1 BJ999434 Glyoxysomal fatty acid oxidation Oryza sativa BAB63983 2e-28 1 BJ999435 Heat shock protein Solanum lycopersicum AAA86424 1e-19 1 BJ999436 Lactoylgluthathione lyase family protein Arabidopsis thaliana NP 563973 9e-31 3 BJ999437 Ring domain containing protein Capsicum annuum AAR99376 3e-13 3 BJ999438 Polyprotein Lycopersicon esculentum AAD13304 2e-33 6 BJ999439 Unknown protein Aurantimonas sp. ZP 0122736 2e-33 16 BJ999440 Dessication-related protein Arabidopsis thaliana AAM65140 3e-31 3

aThis datum was matched by a BLASTN database research. bGlyceraldehyde-3-phosphate dehydrogenase.

there was a single clone showing significant homology clone showing homology to molybdenum cofactor sul- to the coding region of this from Solanum tuberosum. furase from O. sativa. The final step of ABA biosyn- Gene expression of NCED, a key enzyme in ABA bio- thesis is catalyzed by ABA aldehyde oxidase, which re- synthesis, is known to be induced in response to envir- quires molybdenum cofactor.30) It has been reported that onmental stresses.29) In addition, there were 25 clones Ntdin functions in molybdenum cofactor biosynthesis31) showing similarity to Ntdin of a senescence-associated and hence, these clones might be involved in ABA protein from Nicotiana tabacum, and there was one biosynthesis. Chemical and Gene Profiling in the Manufacture of Taiwan Oolong Tea 60708-7 Manufacturing process was performed with a set of primers, a forward primer corresponding to the coding region of one clone and a Genes FSWT1T2T3Cycles reverse primer corresponding to the non-coding region of the other clone, and the amplified DNA fragments RSA 28 were cloned and sequenced. The DNA sequence of the RSB 28 PCR product obtained overlapped with those of the two NCED homologs, indicating that only one NCED is TSP 28 expressed during manufacture. The expression of NCED CHI 28 increased at the step of solar-withering (Fig. 3).

NTD 26 Effects of the manufacturing process on carbohy- NCED 28 drates and ABA contents The carbohydrate contents of each sample of dried tea AMY 28 leaves from F to T5 were quantitatively analyzed by 26S rRNA 18 HPLC. Six carbohydrates, fructose, glucose, sucrose, myo-inositol, 2-O-(-L-arabinopyranosyl)-myo-inositol Fig. 3. RT-PCR Analysis of Expressed Genes. (AMI), and raffinose, were detected and identified by Tea leaves were collected at each step of the manufacturing LC–ESI–MS analysis (data not shown). The presence of process (F, fresh leaves; SW, solar-withering; T1, the first turn over; these carbohydrates, except for myo-inositol, have been RSA T2, the second turn over; T3, the third turn over). , raffinose 32) synthase A; RSB, raffinose synthase B; TSP, trehalose 6-phosphate reported in various parts of tea plants (cv. Yabukita). synthase; CHI, chitinase; NTD, Ntdin.; NCED,9-cis-epoxycarote- AMI and sucrose were the main sugar components in the noid dioxygenase; AMY, -amylase. The expression of 26S rRNA tea samples (Table 4). During the oolong tea manufac- Advancewas used as a control. The number of PCR cycles used View for semi- turing process, the contents of fructose and glucose quantitative analysis is indicated on the right. gradually increased and that of sucrose decreased. myo- Inositol and AMI did not change significantly. In ac- Expression of selected genes during manufacture cordance with the gene expression analysis, raffinose Expression of up-regulated genes during the manu- greatly increased at solar withering and then decreased. facturing process was evaluated by RT-PCR. Seven Furthermore, the ABA contents of each sample of genes, coding for proteins homologous to -amylase dried tea leaves from F to T5 were quantitatively ana- (AMY), raffinose synthase (RSA; RSB), trehalose 6-phos- lyzed by HPLC, because we found increased expression phate synthetase (TPS), NCED, chitinase (CHI), and of ABA-related genes, as described above. As expected, Ntdin of senescence-associated protein (NTD), were se- the ABA contentsProofs dramatically increased at the step of lected, and primers specific to the genes were designed solar withering during the manufacturing process and on the basis of their nucleotide sequences. Expression of then gradually decreased (Fig. 4). these genes was confirmed to increase in T1, compared to that in F (Fig. 3). Moreover, the increased expression Discussion of these genes was generally observed even in the step of solar-withering. In the manufacture of Oriental Beauty, there are two Although two kinds of clones homologous to NCED characteristic points, the utilization of infested tea leaves have been identified in the U1 group, it is not known (Fig. 1B), and the higher fermentation degree due to whether these transcripts are derived from distinct genes the manufacturing process (Fig. 1C). In this study, we or a single gene. To clarify the possibilities, RT-PCR characterized the gene expression profiles and the

Table 4. Changes of Carbohydrate Contents during the Manufacturing Process

Contents of compounds (mg/gram of tea products)a Samplesb Fructose Glucose Sucrose myo-Inositol AMIc Raffinose F 0:12 0:04 0:39 0:13 1:68 0:39 0:44 0:04 3:91 0:25 NDd SW 0:23 0:07 0:33 0:11 1:95 0:54 0:39 0:08 2:37 0:40 0:36 0:04 T1 0:21 0:07 0:19 0:01 2:05 0:58 0:37 0:08 2:44 0:59 0:33 0:06 T2 0:32 0:06 0:28 0:10 1:62 0:51 0:46 0:13 2:53 0:18 0:49 0:10 T3 0:44 0:27 0:50 0:26 1:51 0:30 0:52 0:27 3:31 0:66 0:27 0:03 T4 0:79 0:21 0:84 0:13 1:05 0:27 0:51 0:03 2:96 0:05 0:24 0:05 T5 1:03 0:12 1:10 0:12 0:45 0:18 0:43 0:02 2:67 0:31 0:16 0:09

aData are shown as the mean S.D. (n ¼ 3). bSamples were collected at each step of the manufacturing process (F, fresh leaves; SW, solar-withering; T1, the first turn over; T2, the second turn over; T3, the third turn over; T4, the forth turn over; T5, the fifth turn over) and subjected to the processes of panning, rolling, and drying. cAMI, 2-O-(-L-arabinopyranosyl)-myo-inositol. dND, not detected. 60708-8 J.-Y. CHO et al.

20.0 differential screening, but no transcripts involved in the biosynthesis of volatile compounds were detected in 16.0 the sequenced clones (1,200 clones) among 5,674 beads in the U1 group on the basis of homology research 12.0 (Table 3). This result is probably due to lower expres- sion of genes related to biosynthesis of volatile com- g/g tea product) µ 8.0 pounds. Such transcripts may be found either in the remaining clones of the U1 group or in the U2 group re- 4.0 presenting moderately induced transcripts (Fig. 2). Iden- Content ( 0.0 tification of the genes encoding enzymes involved in the FSWT1T2T3T4T5 biosynthesis of the characteristic volatile compounds is Manufacturing process now in progress. Various transcripts up-regulated during manufacture Fig. 4. Change in Abscisic Acid Content during the Manufacturing were identified by differential screening. Most of the Process. clones showed significant sequence similarities to the Samples were collected at each step of the manufacturing process genes previously identified as stress-responsive genes (F, fresh leaves; SW, solar-withering; T1, the first turn over; T2, the second turn over; T3, the third turn over; T4, the forth turn over; T5, and senescence-related genes from other plant spe- the fifth turn over), and subjected to the processes of panning, cies.36–38) RT-PCR confirmed increased expression of rolling, and drying. seven genes selected from the identified clones during manufacture (Fig. 3). Thus, in this study, we identified a large number of transcripts induced in response to the chemical profiles during the manufacturing process of oolong tea manufacturing process. OrientalAdvance Beauty using the samples prepared View from tea Several transcripts identified in this screening were leaves infested by tea green leafhoppers. In previous associated with carbohydrate metabolism, such as raf- studies, alcoholic volatile compounds such as linalool finose synthase, trehalose 6-phosphate synthase, and - oxides, benzyl alcohol, 2-phenylethanol, and 2,6-dime- amylase (Table 3). It has been reported of many plants thylocta-3,7-diene-2,6-diol (diol) were reported to be that carbohydrate-related genes are expressed under en- the main volatile compounds of Oriental Beauty.28,33) vironmental stresses such as drought and coldness: raf- In this study, we too detected a great increase in these finose synthase in Cicer arietinum;39) trehalose 6-phos- compounds except for diol by comparing F and T5 phate synthase in Lycopersicon esculentum;40) -amy- (Table 2). On the other hand, we found high levels of lase in Cucumis sativus;41) and glyceraldehydes 3-phos- diol even in the fresh-leaf material (F) before manufac- phate dehydrogenaseProofs in Craterostigma plantagineum.42) turing, suggesting that the manufacturing process is not Hence, various stresses during manufacture probably involved in production of diol. The dramatic increase in affect carbohydrate metabolism in tea leaves. In partic- alcoholic volatile compounds together with the presence ular, gene expression of raffinose synthase was found to of diol and hotrienol appears to be responsible for the increase in tea leaves during oolong tea manufacture, promotion of tea quality, especially the strength of the and we confirmed by HPLC analysis that the raffinose unique aroma of the tea. Here we focused our research content dramatically increased at the step of solar with- interests on biochemical events during tea manufacture. ering during oolong tea manufacture (Table 4). It is During oolong tea manufacture, tea leaves are ex- known that various plants accumulate raffinose family posed to various stresses, such as plucking (wounding), oligosaccharides, such as sucrose, galactinol, and raffi- solar withering (drought, heat, and UV/light radiation), nose, in response to the stresses such as drought and indoor withering (drought), and turn over (wounding). It cold.43–45) has been reported that volatiles of monoterpenes such as Some of the transcripts responsible for the tea man- linalool and ocimene and of C6-compounds such as ufacturing process were identified as NCED, Ntdin of hexanal and hexenols are produced by insect attack and senescene-associated protein, and molybdenum cofactor wounding.13,14,34) Continuous mechanical wounding has sulfurase (Table 3), which are involved in ABA biosyn- also been reported to make leaves of the Lima plant emit thesis. ABA is well known as a plant defense-related a variety of volatiles.35) Hence it is most likely that var- hormone responding to environmental stresses such as ious stresses during manufacture also cause the produc- drought, cold, and saline. Previous reports state that tion of volatiles of the monoterpene alcohols and C6- NCED is induced in response to drought stress in cow- compounds. It would be interesting to study which pea,46) tomato,47) and bean plants,29) and that molybde- stresses are mainly related to the formation of these num cofactor sulfurase is induced under cold and os- volatiles in tea leaves. motic stresses in Arabidopsis.48) Therefore, the tran- The increased volatile compounds (Table 2) during scripts detected in this study are also probably involved oolong tea manufacture are probably produced by de in ABA biosynthesis, suggesting that various stress con- novo biosynthesis as well as the hydrolysis of glycoside ditions during oolong tea manufacture probably induce precursors. The infested tea samples were analyzed by ABA biosynthesis. Consequently, the ABA contents Chemical and Gene Profiling in the Manufacture of Taiwan Oolong Tea 60708-9 dramatically increased at the solar withering step during 6) Sakata, K., Watanabe, M., and Usui, T., Molecular basis the manufacturing process (Fig. 4). During oolong tea of alcoholic aroma formation during tea processing. In manufacture, the ABA contents (12.41 mg/g dry weight ‘‘Food for Health in the Pacific Rim,’’ Third International tea product) at step of solar withering increased over 25- Conference on Food Science and Technology, eds. Whitaker, J. R., Haard, N. F., Shoemaker, C. F., and fold as compared to that of fresh tea leaves (0.44 mg/g Singh, R. P., Food and Nutrition Press, Trumbull, pp. 93– dry weight of tea product). Drought conditions such as 105 (1999). solar withering and indoor withering during manufacture 7) Sakata, K., -Primeverosidase relationship with floral tea can play important roles in the accumulation of ABA. aroma formation during processing of oolong tea and It is noteworthy that most transcripts up-regulated black tea. In ‘‘Caffeinated Beverages, Health Benefits, in tea leaves during oolong tea manufacture are in good Physiological Effects, and Chemistry’’ ACS Symposium agreement with the ABA-inducible genes previously Series 754, American Chemical Society, eds. Parliament, reported from Arabidopsis.49) Therefore, it is possible T. H., Ho, C., and Schieberie, P., Washington, DC, that tea leaves during oolong tea manufacture first ac- pp. 327–335 (2000). cumulate ABA with the induction of ABA biosynthetic 8) Ma, S. J., Mizutani, M., Hiratake, J., Hayashi, K., Yagi, genes, and that the increased level of endogenous ABA K., Watanabe, N., and Sakata, K., Substrate specificity of induces various stress defense genes. Hence it would be -primeverosidase, a key enzyme in aroma formation during oolong tea and black tea manufacturing. Biosci. interesting to investigate whether the clones identified in Biotechnol. Biochem., 65, 2719–2729 (2001). this study are up-regulated by an exogenous application 9) Mizutani, M., Nakanishi, H., Ema, J., Ma, S. J., Noguchi, of ABA in tea leaves. Furthermore, increased ABA E., Inohara-Ochiai, M., Fukuchi-Mizutani, M., Nakao, levels due to the tea manufacturing process can enhance M., and Sakata, K., Cloning of -primeverosidase from the production of volatile compounds in tea leaves. Thus tea leaves, a key enzyme in tea aroma formation. Plant the traditional method of manufacture, one of the im- Physiol., 130, 2164–2176 (2002). portantAdvance factors in Oriental Beauty, utilizes plant View defense 10) Wang, D., Kubota, K., Kobayashi, A., and Juan, I. M., responses to elevate the production of volatile com- Analysis of glycosidically bound aroma precursors in tea pounds and other metabolites. As described above, the leaves. 3. Change in the glycoside content of tea leaves other characteristic factor in the oolong tea is the use of during the oolong tea manufacturing process. J. Agric. tea leaves infested by tea green leafhoppers as leaf Food Chem., 49, 5391–5396 (2001). 11) Wang, D., Yoshimura, T., Kubota, K., and Kobayashi, material. We have started to investigate the effects of A., Analysis of glycosidically bound aroma precursors in insect attack on the production of the volatile com- tea leaves. 1. Qualitative and quantitative analysis of 50) pounds. Gene expression profiling in response to glycosides with aglycons as aroma compounds. J. Agric. insect attack is also in progress. Food Chem., 48, 5411–5418 (2000). 12) Wang, D., Kurasawa,Proofs E., Yamaguchi, Y., Kubota, K., Acknowledgment and Kobayashi, A., Analysis of glycosidically bound aroma precursors in tea leaves. 2. Changes in the This work was supported in part by a Grant-in-Aid for glycoside contents and glycosidase activities in tea Scientific Research B (no. 15405023, to K.S.) from the leaves during the black tea manufacturing process. J. Japanese Society for the Promotion of Science, and Agric. Food Chem., 49, 1900–1903 (2001). by a grant (no. 1604158, to K.S.) from the Ministry of 13) Arimura, G., Kost, C., and Boland, W., Herbivore- induced, indirect plant defences. Biochim. Biophys. Acta, Education, Culture, Sports, Science and Technology of 1734, 91–111 (2005). Japan. We thank the Tea Research and Extension Station 14) Bartram, S., Jux, A., Gleixner, G., and Boland, S., in Taiwan for helping in the preparation of tea samples Dynamic pathway allocation in early terpenoid biosyn- used in this work. thesis of stress-induced lima bean leaves. Phytochemis- try, 67, 1661–1672 (2006). References 15) Howe, G. A., and Schilmiller, A. L., Oxylipin metab- olism in response to stress. Curr. Opin. Plant Biol., 5, 1) Kinoshita, T., and Sakata, K., The mysterious aroma of 230–236 (2002). Oriental Beauty. Koryo (in Japanese), 229, 113–120 16) Mahajan, S., and Tuteja, N., Cold, salinity and drought (2006). stresses: an overview. Arch. Biochem. Biophys., 444, 2) Hara, Y., Luo, S. J., Wickremasinghe, R. L., and 139–158 (2005). Yamanishi, T., Flavor of tea. Food Rev. Int., 11, 477– 17) Mittler, R., Abiotic stress, the field environment and 525 (1995). stress combination. Trends Plant Sci., 11, 15–19 (2006). 3) Hara, Y., Luo, S. J., Wickremasinghe, R. L., and 18) Dunaeva, M., and Adamska, I., Identification of genes Yamanishi, T., Processing of tea. Food Rev. Int., 11, expressed in response to light stress in leaves of 426–430 (1995). Arabidopsis thaliana using RNA differential display. 4) Kawakami, M., Topics and progress in tea flavor Eur. J. Biochem., 268, 5521–5529 (2001). science. Foods Food Ingredients J. Jpn. (in Japanese), 19) Fei, Z., Tang, X., Alba, R. M., White, J. A., Ronning, C. 197, 13–26 (2002). M., Martin, G. B., Tanksley, S. D., and Giovannoni, J. J., 5) Yamanishi, T., Tea aroma. Foods Food Ingredients J. Comprehensive EST analysis of tomato and comparative Jpn. (in Japanese), 168, 23–34 (1996). genomics of fruit ripening. Plant J., 40, 47–59 (2004). 60708-10 J.-Y. CHO et al. 20) Li, R., Rimmer, R., Buchwaldt, L., Sharpe, A. G., Aroma components of Pomfon oolong tea. Nippon Seguin-Swartz, G., Coutu, C., and Hegedus, D. D., No¯geikagaku Kaishi (in Japanese), 64, 1349–1354 Interaction of Sclerotinia sclerotiorum with a resistant (1990). Brassica napus cultivar: expressed sequence tag analysis 34) Pichersky, E., and Gershenzon, J., The formation and identifies genes associated with fungal pathogenesis. function of plant volatiles: perfumes for pollinator Fungal Genet. Biol., 41, 735–753 (2004). attraction and defense. Curr. Opin. Plant Biol., 5, 237– 21) Nanjo, T., Futamura, N., Nishiguchi, M., Igasaki, T., 243 (2002). Shinozaki, K., and Shinohara, K., Characterization of 35) Mithofer, A., Wanner, G., and Boland, W., Effects of full-length enriched expressed sequence tags of stress- feeding Spodoptera littoralis on lima bean leaves. II. treated poplar leaves. Plant Cell Physiol., 45, 1738–1748 Conditions of mechanical wounding resembling insect (2004). feeding are sufficient to elicit herbivory-related volatile 22) Shatters, R. G., Bausher, M. G., Hunter, W. B., emission. Plant Physiol., 137, 1160–1168 (2005). Chaparro, J. X., Dang, P. M., Niedz, R. P., Mayer, R. 36) Boominathan, P., Shukla, R., Kumar, A., Manna, D., T., McCollum, T. G., and Sinisterra, X., Putative Negi, D., Verma, P. K., and Chattopadhyay, D., Long- protease inhibitor gene discovery and transcript profiling term transcript accumulation during the development of during fruit development and leaf damage in grapefruit dehydration adaptation in Cicer arietinum. Plant Phys- (Citrus paradisi Macf.). Gene, 326, 77–86 (2004). iol., 135, 1608–1620 (2004). 23) Chen, L., Zhao, L. P., and Gao, Q. K., Generation and 37) Bohnert, H. J., Nelson, D. E., and Jensen, R. G., analysis of expressed sequence tags from the tender Adaptations to environmental stresses. Plant Cell, 7, shoots cDNA library of tea plant (Camellia sinensis). 1099–1111 (1995). Plant Sci., 168, 359–363 (2004). 38) Kore-eda, S. C. M. A., Akselrod, I., Bufford, D., 24) Park, J. S., Kim, J. B., Hahn, B. S., Kim, K. H., Ha, S. Fredrickson, M., Clark, E., and Cushman, J. C., Tran- H., Kim, J. B., and Kim, Y. H., EST analysis of genes script profiling of salinity stress responses by large-scale involved in secondary metabolism in Camellia sinensis expressed sequence tag analysis in Mesembryanthemum Advance(tea), using suppression subtractive hybridization. ViewPlant crystallinum. Gene, 341, 83–92 (2004). Sci., 166, 953–961 (2004). 39) Romo, S. L. E., and Dopico, B., Water stress-regulated 25) Chang, S., Puryear, J., and Cairney, J., A simple and gene expression in Cicer arietinum seedlings and plants. efficient method for isolation of RNA from pine trees. Plant Physiol., 39, 1017–1026 (2001). Plant Mol. Biol. Rep., 95, 379–383 (1993). 40) Cortina, C. C., and Francisco, A. C. M., Tomato abiotic 26) Kamauchi, S., Nakatani, H., Nakano, C., and Urade, R., stress enhanced tolerance by trehalose biosynthesis. Gene expression in response to endoplasmic reticulum Plant Sci., 169, 75–82 (2005). stress in Arabidopsis thaliana. FEBS J., 272, 3461–3476 41) Todaka, D., Matsushima, H., and Morohashi, Y., Water (2005). stress enhances -amylase activity in cucumber cotyle- 27) Altschule, S. F., Madden, T. L., Schaffer, A. A., Jzang, dons. J. Exp. Bot., 51, 739–745 (2000). J. H., Zhang, Z., Miller, W., and Lipman, D. J., Gapped 42) Velasco, R.,Proofs Salamini, F., and Bartels, D., Dehydration BLAST and PSI-BLAST: a new generation of protein and ABA increase mRNA levels and enzyme activity of database search programs. Nucleic Acids Res., 25, 3389– cytosolic GAPDH in the resurrection plant Craterostig- 3402 (1997). ma plantagineum. Plant Mol. Biol., 26, 541–546 (1994). 28) Kawakami, M., Ganguly, S. N., Banerjee, J., and 43) Albini, F. M. M. C., Finzi, P. V., Ferrarotti, M., Cantoni, Kobayashi, A., Aroma composition of oolong tea and B., Puliga, S., and Vazzana, C., Galactinol in the leaves black tea by brewed extraction method and characteriz- of the resurrection plant Boea hygroscopica. Phytochem- ing compounds of Darjeeling tea aroma. J. Agric. Food istry, 51, 499–505 (1999). Chem., 45, 200–207 (1995). 44) Peterbauer, T., Lahuta, L. B., Blochl, A., Mucha, J., 29) Qin, X., and Zeevaart, J. A., The 9-cis-epoxycarotenoid Jones, D. A., Hedley, C. L., Gorecki, R. J., and Richter, cleavage reaction is the key regulatory step of abscisic A., Analysis of the raffinose family oligosaccharide acid biosynthesis in water-stressed bean. Proc. Natl. pathway in pea seeds with contrasting carbohydrate Acad. Sci. USA, 96, 15354–15361 (1999). composition. Plant Physiol., 127, 1764–1772 (2001). 30) Zdunek-Zastocka, E., Omarov, R. T., Koshiba, T., and 45) Taji, T., Ohsumi, C., Iuchi, S., Seki, M., Kasuga, Lips, H. S., Activity and protein level of AO isoforms in M., Kobayashi, M., Yamaguchi-Shinozaki, K., and pea plants (Pisum sativum L.) during vegetative devel- Shinozaki, K., Important roles of drought- and cold- opment and in response to stress conditions. J. Exp. Bot., inducible genes for galactinol synthase in stress toler- 55, 1361–1369 (2004). ance in Arabidopsis thaliana. Plant J., 29, 417–426 31) Yang, S. H., Berberich, T., Miyazaki, A., Sano, H., and (2002). Kusano, T., Ntdin, a tobacco senescence-associated 46) Iuchi, S., Kobayashi, M., Yamaguchi-Shinozaki, K., and gene, is involved in molybdenum cofactor biosynthesis. Shinozaki, K., A stress-inducible gene for 9-cis-epoxy- Plant Cell Physiol., 44, 1037–1044 (2003). carotenoid dioxygenase involved in abscisic acid biosyn- 32) Suzuki, M., Watanabe, N., Kuboi, T., Konishi, S., thesis under water stress in drought-tolerant cowpea. Sakata, K., and Ina, K., Characteristic distribution of Plant Physiol., 123, 553–562 (2000). 2-O-(-L-arabinopyranosyl)-myo-inositol in tea plant 47) Thompson, A. J., Jackson, A. C., Parker, R. A., Morpeth, (Camellia sinensis). In ‘‘Proceedings of the International D. R., Burbidge, A., and Taylor, I. B., Abscisic acid Symposium on Tea Science,’’ Organizing Committee of biosynthesis in tomato: regulation of zeaxanthin epox- ISTS, Shizuoka, pp. 110–114 (1991). idase and 9-cis-epoxycarotenoid dioxygenase mRNAs 33) Takami, C., Shimotsukasa, A., and Kobayashi, A., by light/dark cycles, water stress and abscisic acid. Chemical and Gene Profiling in the Manufacture of Taiwan Oolong Tea 60708-11 Plant Mol. Biol., 42, 833–845 (2000). and Shinozaki, K., Monitoring the expression pattern of 48) Xiong, L., Ishitani, M., Lee, H., and Zhu, J. K., The around 7,000 Arabidopsis genes under ABA treatments Arabidopsis LOS5/ABA3 locus encodes a molybdenum using a full-length cDNA microarray. Funct. Integr. cofactor sulfurase and modulates cold stress- and Genomics, 2, 282–291 (2002). osmotic stress-responsive gene expression. Plant Cell, 50) Ogura, M., Kinoshita, T., Shimizu, B., Shirai, F., 13, 2063–2083 (2001). Tokoro, K., Lin, M. L., and Sakata, K., Identification 49) Seki, M., Ishida, J., Narusaka, M., Fujita, M., Nanjo, T., of aroma components during processing of the famous Umezawa, T., Kamiya, A., Nakajima, M., Enju, A., Formosa oolong tea ‘‘Oriental Beauty,’’ ACS Symposi- Sakurai, T., Satou, M., Akiyama, K., Yamaguchi- um, in press. Shinozaki, K., Carninci, P., Kawai, J., Hayashizaki, Y.,

Advance View

Proofs