IDENTIFICATION AND FUNCTIONAL EVALUATION OF ABSCISIC ACID UDP-GLUCOSYLTRANSFERASE GENE CANDIDATES IN GRAPEVINE (VITIS VINIFERA)byKEYU LIUB.Sc., Beijing Institute of Technology, 2009A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCEinTHE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES(Plant Science)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)November 2013?Keyu Liu, 2013iiAbstractIn grape and some other fruits used for wine making, abscisic acid (ABA) has been demonstrated to be involved in regulating fruit development. ABA metabolism has been characterized in the model plant Arabidopsis thaliana and shown to involve at least three catabolic pathways that inactivate free ABA. Abscisic acid UDP-Glycosyltransferase (ABA-UGT), a member of GlycosylTransferase(GTs) protein family 1, is a key enzyme in the abscisic acid metabolism which leads to the deactivation of ABA. ABA-UGT catalyzes the glucosylation process from abscisic acid to its conjugated metabolite abscisic acid glucose ester (ABA-GE) in planta. ABA-GE is a terminal product of the ABA metabolic pathway and does not appear to have any biological function. Thus, manipulation of ABA-UGT in wine-making fruits will offer a possibility to control the metabolism of ABA with the potential to optimize the ripening of fruit. In this study, genomic and biochemical approaches were used to attempt to identify ABA-UGT in Vitis vinifera, the European grapevine species most widely used in wine making. The identification of ABA-UGT can be a foundation for further research about how ABA is involved in the fruit development and other processes in grapevine.iiiPrefaceThis dissertation is an original, unpublished intellectual product of the author, Keyu Liu. The ABA metabolites content assay in Chapter 3 was done by Plant Biotechnology institute in Saskatoon.ivTable of contents Abstract??????????...????????????????.???.iiPreface???????????????????????????.???.iiiTable of contents?????????????????..??????..??..ivList of tables????....?????????..?????????????..viiList of figures??????????????????????.?????viii1. Introduction?????????????????????????...?.11.1 British Columbia wine industry background????????????11.2 Climacteric and non-climacteric fruit ripening??????..??.??.21.2.1 Climacteric fruit???????????????.???..??21.2.2 Non-climacteric fruit???????????????.??..?31.3 Plant hormones and their roles in the ripening process????...???31.3.1 Ethylene???????????????????..?.???..41.3.2 Auxin..?????????????????????????61.4 Abscisic acid??????????????????????.???61.4.1 Structure and role of ABA??????????????.??..61.4.2 ABA biosynthesis?????????????????..???81.4.3 ABA catabolism??????????????????.??.101.4.4 Enzymes involved in the ABA catabolism???????...??.121.4.5 The role of ABA in fruit ripening?????????????..131.4.6 ABA homeostasis?????????????????..??..141.4.7 ABA-glucose ester????????????????...???15v1.5 Glycosyltransferases??????????????????...??..161.5.1 Glycosyltransferase family 1..????????????..??.171.5.2 Arabidopsis ABA UDP-glucosyltransferase UGT71B6?????181.5.3 UDP-glucosyltransferases in grapevine???????????201.6 ABA-UGT and glucosylation in ABA homeostasis??????...??..211.7 Gateway cloning system?????????????????...??.221.8 Research objective????????????????????..?...232. VviABA-UGT candidate selection and the cloning, expression, purification and characteristic assay of candidate enzymes????????..?....?252.1 Vitis vinifera UDP-glucosyltransferase candidate selection????.?252.1.1 Candidate selection by sequence similarity????????.?.252.1.2 Candidate selection by evolutionary relationship??????252.1.3 Candidate selection by expression level trend?????.???282.2 Clone of VviUGT candidate genes and UGT71B6?????????..302.2.1 Genomic DNA isolation of Vitis vinifera and Arabidopsis?..?.....302.2.2 Polymerase chain reaction (PCR) to amplify candidate gene......................................................................................................342.2.3 Amplified candidate genes isolated from DNA electrophoresis gel?????????????????.?????????362.2.4 TOPO cloning reaction to generate entry clone????...??...382.2.5 LR recombination reaction to generate expression clone???422.3 Expression and purification of recombinant proteins?????...??.45vi2.3.1 Transformation of expression vector to Rosetta-gami (DE3) pLysS Competent cells??????????????????...??462.3.2 Optimization of expression conditions???????????..462.3.3 SDS-PAGE protein gel protocol??????????...???.502.3.4 Anti-GST western blot protocol??????????.?.?.?.522.3.5 Purification of recombinant proteins????????.?.??..532.3.6 Chaperone DnaK.???????????????..????.592.3.7 GST tag cleavage????????????????...???.602.4 Recombinant enzyme assay??????????????????..622.4.1 Setup enzyme assay reaction???????????????.622.4.2 ABA-glucose ester quantitative assay????????....??...633. General discussion????????????????????????66References?????????????????????????...???69viiList of tablesChapter 2Table 2.1 CTAB plant genomic DNA extraction protocol??????????...31Table 2.2 PCR reaction condition????????????????????.35Table 2.3 GeneJET gel extraction kit protocol???????????????.37Table 2.4 TOPO Cloning Reaction???????????????????...40Table 2.5 QIAprep Spin Miniprep Kit protocol??????????????...41Table 2.6 LR reaction procedures????????????????????.43Table 2.7 protein expression procedure??????????????????47Table 2.8 SDS-PAGE materials?????????????????????50Table 2.9 Anti-GST Western Blot buffers????????????????....52Table 2.10 GST fusion protein purification materials and buffers???????...54Table 2.11 TEV cleavage reaction????????????????????61Table 2.12 ABA-UGT enzyme assay protocol???????????????.63viiiList of figures Chaper 1Figure 1.1 Abscisic acid??????????????????????.??6Figure 1.2 Enantiomers of ABA, (+)-ABA and (-)-ABA?????????...??8Figure 1.3 ABA biosynthesis pathway????????????????...??9Figure 1.4 ABA catabolism??????????????????????..11Figure 1.5 ABA and ABA metabolites contents in different developmental stages of grape berry?????????????????????????15Figure 1.6 Gateway cloning system???????????????????.23Chapter 2Figure 2.1 phylogenetic analysis of Vitis vinifera ABA-UGT candidates and some Arabidopsis UGTs?????????????????????27Figure 2.2 RNA seq data of some Vitis vinifera glucosyltransferases??????.30Figure 2.3 Verification of Isolated genomic DNA?????????????...34Figure 2.4 PCR results of candidate genes????????????????...36Figure 2.5 Maps of pENTR/TRV/DTOPO vector and pDEST15 vector?????.39Figure 2.6 PCR verification of expression clone??????????????.45Figure 2.7 Optimization of inducing time and IPTG concentration??????.....49Figure 2.8 Protein purification res?Lt using recommended procedures?????..56 Figure 2.9 Western blot showing effects of 1M urea on protein purification???..59Figure 2.10 Purified recombinant proteins????????????????...60ixFigure 2.11 Quantitative enzyme assay result??????????????......6511 Introduction1.1 British Columbia wine industry backgroundOver the past quarter-century, Canadian vintners have increased their production of high-quality wines. In British Columbia, grape production mainly occurs in four viticultural areas: the Okanagan Valley, the Similkameen Valley, the Fraser Valley and on the coastal islands.In British Columbia, the climate is always the first issue that concerns viticulturists and wine scientists. It is a cool-climate grape growing environment, like New Zealand, northern France, Italy, and Germany. For many years it was believed that Vitis vinifera (the noble European grape varieties such as Chardonnay, Riesling, Cabernet Sauvignon, Merlot, Pinot noir) which are cultivated in the Mediterranean region, central Europe, and southwestern Asia, could not survive the rigours of Canadian winters and the freeze-thaw-freeze cycle of early spring . As a result, the majority of plantings here were historically the winter-hardy North American labrusca varieties (such as Concord and Niagara, now only used for processing in the food industry) and early-ripening, winter-resistant hybrids, such as Vidal, Seyval Blanc, Baco Noir and Marechal Foch. More recently, the BC wine industry has shifted to vinifera varieties for wine production and shown that its wines can be competitive internationally.21.2 Climacteric and non-climacteric fruit ripeningFruit ripening is a complicated physiological and biochemical process by which fruits attain their desirable flavour, quality, colour, palatable nature and other textural properties (Grierson et al., 1987). Throughout the ripening process, the fruit becomes sweeter as the starches are converted into simple sugars by amylases; the fruit changes from green to colorful as the chlorophyll is broken down by hydrolases and anthocyanins accumulate in the skin; the fruit becomes less tart as the acids are converted to neutral molecules by kinases; the fruit becomes softer as the amount of pectin is degraded by pectinases (Seymour et al., 1993). On the basis of ripening behaviour, fruit can be classified as climacteric and non-climacteric fruits. 1.2.1 Climacteric fruitClimacteric species are defined as fruit which have high respiration rate during the fruit?s ripening, and enter the ?climacteric phase? at ripening initiation. During the ripening initiation process in climacteric fruit, they emit a burst of ethylene along with an increased rate of respiration (Alexander et al., 2002). Ripe fruit are soft and delicate and generally cannot withstand the rigours of transport and repeated handling. These fruits are harvested hard and green, and are then ripened near distribution and consumption areas. Tomato, mango, banana and apple are typical climacteric fruits.3Several studies have indicated the role of ethylene in the induction of ripening in climacteric fruits. When specific ethylene biosynthesis inhibitorswhich block the action of ethylene were applied, climacteric fruits showed astrong inhibition of ripening but when the fruits in the green-ripe stage were exposed to exogenous ethylene, the ripening process could be activated (Gray et al., 1992). Small doses of ethylene can be used to induce the ripening process of climacteric fruits under controlled temperature and humidityconditions.1.2.2 Non-climacteric fruitNon-climacteric fruits once harvested do not ripen further. At ripening initiation in non-climacteric fruits, there is no characteristic increased rate of respiration (Itar et al., 2008). Non-climacteric fruits produce very small amounts of ethylene at ripening initiation and show little to no responses to exogenous ethylene treatment, depending on the species (Goldschmidt et al.,1997). Although in non-climacteric fruits ethylene evolution is very low and the ripening process seems to occur independently of ethylene production, some recent studies suggest that ethylene may still have some additive functions in the ripening process of grape (Chervin et al., 2004). Non-climacteric fruits include orange, cherry, strawberry, watermelon and the fruit which is the subject of this dissertation, grape.41.3 Plant hormones and their roles in the ripening processPlant hormones are endogenous signal molecules that effect growth, development, and/or plant defenses at micromolar or lower concentrations in plant cells (Davies et al,. 2004). They determine the formation of organs like flowers and leaves, and the development and ripening of fruits and even plant death. Hormones are produced in one part of a plant and may be transported to a different part where they exert a particular effect. There are five major classes of plant hormones, auxin, abscisic acid, cytokinins, ethylene and gibberellins. Abscisic acid and Ethylene have been demonstrated to regulated fruit ripening, while a potential role for auxins remains unclear and may be species-specific.1.3.1 EthyleneEthylene is unique among the plant hormones in that it is a gaseous hormone. Ethylene has been demonstrated to regulate the triple response in seedlings (inhibited stem elongation, increased stem thickening and increased lateral growth), wound responses, and defense against some pathogens (Abeles et al.,1992). Ethylene has long been regarded as the main regulator of ripening in climacteric fruits. Ethylene biosynthesis has been determined to be developmentally regulated by a coordinated increase in 1-amino-cycloproopane carboxylic acid oxidase (ACO) and 1-amino-cyclopropane carboxylate synthase (ACS) activities. Ethylene is 5perceived by specific ethylene receptors and signal transduction occurs through de-repression of a MAP kinase cascade, finally activating several transcription factors (TFs) which target dozens of genes directly or indirectly. The actions of those genes accumulate to bring about some changes in the fruit, such as chlorophyll degradation and pigment development, sugar accumulation, softening (cell wall degradation) and secondary metabolite accumulation (Argueso et al,. 2007).Before the 1980s, ethylene was thought to have a very limited role in the ripening process of non-climacteric fruits. More recent studies, however, have revealed that some aspects of non-climacteric fruit ripening may be associated with ethylene. It has been demonstrated that there is a fully functional pathway for ethylene biosynthesis in grape berry tissue, and this pathway is activated just before the ripening initiation stage (Chervin et al. 2004), referred to by viticulturists using the French term, veraison. This study also showed that ethylene perception is important for some changes in berries associated with ripening. A more recent study using a field treatment with 1-MCP (an inhibitor of ethylene action) at various times following full flowering, showed delayed common characteristics of berry ripening such as berry enlargement and anthocyanin synthesis, which provides further evidence for an additive role for ethylene in non-climacteric ripening regulation (Villarreal et al,. 2010). 1.3.2 AuxinAuxin was the first plant hormone discovered. Although the major function of auxin is to stimulate cell elongatroot initiation and phototropismthe ripening process; however, the function of auxin in ripening is complicated. In climacteric fruits, auxin may stimulate the ethylene prior to the initiation of ripening, but in grape, which is a non-climacteric fruit, the initiation of sugar accumulation was delayed and the rate of sugar accumulation was lower in an auxinal., 2012). Furthermore, in some fruits, auxin interacts with ethylene to regulate fruit growth and ripening.1.4 Abscisic acid1.4.1 Structure and role of ABAABA plays important roles in the control of growth, transpiration, stress ion and cell division, for example, lateral (Cleland et al,. 1995), it is also involved in biosynthesis of -treated fruit (Figure 1.1 Abscisic acid6Davies et 7tolerance and the maturation and repression of germination of seeds. Many of these processes seem to be linked to perception of water deficit (Zeevaart et al., 1988). There are several ABA-mutant Arabidopsis plants that have been identified and are available now to researchers. These mutants are either hypersensitive or insensitive to ABA and they reflect the importance of ABA in seed germination and early embryo development.ABA has a chiral centre at C1?, so there are two enantiomers, (+)-ABA and (-)-ABA (Figure 1.2). Previous studies have demonstrated that these two enantiomers show different activities towards growth, stomatal function, seed development and germination (Priest et al., 2005). These differences have been observed at the levels of transcription, cellular response, or whole plants. Furthermore, these two enantiomers are not always perceived and metabolized by plants in a same manner. (+)-ABA is the naturallyoccurring, biologically active form, whereas (-)-ABA is not found in plants and is inactive as a plant hormone. 8Figure 1.2 Enantiomers of ABA, (+)-ABA and (-)-ABA1.4.2 ABA biosynthesis The biosynthesis of ABA in plants occurs through the modification and cleavage of carotenoids to xanthoxin, which undergoes two oxidation steps to produce ABA. The ABA biosynthesis pathway is shown in Figure 1.3. (Nambara et al.,2005) Carotenoid zeaxanthin is converted to violaxanthin by zeaxanthin epoxidase (ZEP) via the intermediate antheraxanthin. A reverse reaction occurs in chloroplasts in high light conditions catalyzed by violaxanthin de-epoxidase (VDE). 9Figure 1.3 ABA biosynthesis pathway (Nambara et al.,2005)The synthesis of 9?-cis-neoxanthin and 9?-cis-violaxanthin from violaxanthin is not yet fully elucidated (Schwarts and Zeevaart, 2010). Two enzymes may be required in this step which are neoxanthin synthase (NSY) and an 10isomerase. After that, a cleavage reaction which is catalyzed by a family of 9-cis-epoxycarotenoid dioxygenases (NCED) produces the C15 product, xanthoxin. Then xanthoxin is converted into abscisic aldehyde by a short-chain dehydrogenase reductase. The final step in the biosynthesis of ABA involves the oxidation of the aldehyde at C1 to a carboxyl group by an aldehyde oxidase.1.4.3 ABA catabolismAs shown in Figure 1.4, there are four different pathways in ABA catabolism, which include one conjugation pathway leading to ABA-glucose ester and three oxidation pathways leading to three metabolites, DPA, neoPA and 7?OH-ABA (Nambara et al., 2005).11Figure 1.4 ABA catabolismCatabolism of ABA occurs through either conjugation to glucose(ABA-GE) or through the oxidation of methyl groups on the ABA ring(DPA, neoPA, 7?OH-ABA)In the three oxidation pathways, the one which form 8?OH-ABA which then rearranges to form phaseic acid (PA) and dihydrophaseic acid (DPA) is considered the predominant one, so PA and DPA are the most widespread and abundant ABA catabolites. ABA is biologically inactivated through catabolism. 8?OH-ABA contain solid biological activity (Arai S et al., 1999), but PA loses some activity following the cyclization process in which it is formed (Zhang et al,. 2001). 12DPA has been demonstrated to be biologically inactive (Walton et al., 1995), so the inactivation of ABA is completed by this stage. In addition, 7?OH-ABA and 9?OH-ABA and its isomer, neoPA, are considered as minor oxidation metabolites (Walton et al., 1995). 9?OH-ABA and neoPA have only been found in specific plant species (e.g., colza, pea, orange, Arabidopsis)but do not occur in all plants (Zhou et al., 2004).In the ABA conjugation pathway, ABA and all its oxidative catabolites (7?OH-ABA, 8?OH-ABA, 9?OH-ABA, PA, DPA, neoPA) are potential targets for conjugation with glucose by the process of glucosylation(Nambara et al., 2005). Since glucosylation and ABA-glucose ester are the subjects of this thesis, I will focus further review of them in the following chapter sections.1.4.4 Enzymes involved in ABA catabolismThe enzymes that involved in ABA metabolism are always crucial subjects for better understanding how ABA functions in plants. It is interesting that almost all of the biosynthesis genes were identified several years ago through the use of mutants (Nambara et al., 2005), but no catabolic genes wereidentified until recently.In the ABA oxidation catabolism pathways, cytochrome P450 (CYP) 13enzymes have been demonstrated to be responsible for the hydroxylation of ABA at the 8?-methyl position (Nambara et al., 2005). CYPs form a large multigene family, one member of which encodes an enzyme with 8?-hydroxylase activity in Arabidopsis. This is also the only gene involved in oxidative catabolism of ABA identified in Arabidopsis.There was also little progress until recently in identifying genes encoding the enzymes involved in the glucosylation of ABA. In 2005, ABA UDP-glucosyltransferases (ABA-UGT) was identified in Arabidopsis and demonstrated to be responsible for the glucosylation of ABA in ABA catabolism (Lim et al., 2005). ABA-UGT will be discussed in more detail later in this chapter.1.4.5 The role of ABA in fruit ripeningAbscisic acid has been shown to play a major role in positively regulating theonset of ripening in grape berry. Free ABA and ABA catabolite levels change dramatically along ripening. Free ABA levels are high early in berry development after which they decrease to relatively low levels just prior to veraison. They increase again at about the time of the initiation of sugar accumulation and reach a peak two to three weeks later, after which time they decline as the berry matures. Due to this pattern of accumulation, ABA has been suggested to be involved in ripening, perhaps controlling the initiation 14of ripening (Coombe et al., 1972); however, the mechanisms underlying how ABA regulates ripening of non-climacteric fruits is still unclear. Previous studies have suggested that ABA could be used to improve food colour and nutrition in pigmented fruits by increasing anthocyanin and other phenolic compound concentrations. ABA may also be involved in promoting cell wall hydrolase activities, fruit softening and sugar accumulation (Kondo et al., 1991). Giribaldi et al. (2009) performed a time course study on grape berry in which ABA and fluridone (an ABA biosynthesis inhibitor) were applied to fruit before veraison and at early- and mid-veraison. The resultsshowed that ABA treatments were very effective in increasing the concentration of skin ABA, and the treatment time is very important. So it suggested that the berry was very sensitive to ABA and the way ABA regulates fruit ripening is accurate (Giribaldi et al., 2009).1.4.6 ABA homeostasisThe fluctuation in cellular ABA level plays an important role in the regulation of ABA functions, so the homeostasis mechanism that regulatesthe ABA level in plant cells need to be studied. ABA homeostasis involvesABA biosynthesis, ABA catabolism, as well as ABA transport in the plant.Figure 1.5 shows the levels of ABA and its metabolites in different fruit developmental stages along ripening in pericarp. Several studies on ABA 15homeostasis have been done and revealed how ABA regulates plant stress adaptation responses and seed development through changes in ABA homeostasis. In addition, it is possible that ABA homeostasis is important in regulating non-climacteric fruit ripening.Figure 1.5 ABA and ABA metabolites contents in different developmental stages of grape berry (unpublished data from a separate project conducted by the Plant Biotechnology Institute, Saskatoon, SK)1.4.7 ABA-glucose esterAfter ABA and its metabolites are conjugated to ABA-GE by the process of glucosylation, the bulk of ABA-GE is compartmentalized and sequestered in the vacuole. This has been demonstrated by radiolabeling experiments(Cornish et al., 1984) and the measurement of ABA-GE in vacuoles in cell suspension cultures and protoplasts (Bray et al., 1985).16It has been reported by many authors that ABA-GE is not biologically active. In addition, some scientists also suggest that ABA-GE may be a storage form of ABA by the evidence that the free ABA level increases when ABA-GE is fed exogenously (Lehmann, 1983). There is no evidence, however, to support that endogenous ABA-GE can be cleaved to release free ABA, although there are many samples from different plants that have been studied. Therefore, the lack of evidence that the endogenous ABA-GE can be cleaved to form free ABA combined with the vacuolar location of ABA-GE led researchers to conclude that ABA-GE should be a dead-end product of ABA catabolism.In addition, it has been hypothesized that ABA-GE may be the long distance transport form of ABA; the hypothesis is based upon the finding that ABA-GE is present in the xylem of many species. Although there is no direct evidence to support this hypothesis, it has been cited many times in recent publications about ABA metabolism and requires further study.1.5 GlycosyltransferasesGlycosyltransferases (GTs) are enzymes that catalyze an important process in living cells which is called glycosylation. Glycosylation is involved in the transfer of an activated sugar molecule to an acceptor molecule. Nucleotide diphospate (NDP)-sugars are commonly the activated donor sugar, but it can 17also be a dolichol-phosphate-sugar, sugar-1-phosphate, or a nucleosidemonophosphate-sugar (Coutinho et al., 2003). Acceptor molecules range from small secondary metabolites to large molecules like proteins, lipids, or growing polysaccharide chains. Because the GTs are responsible for the majority of phytomass transfer, GTs have been described to be quantitatively the most important enzymes in plants (Campbell et al., 1997).Since 1997, more and more GT protein sequences have become available and they have been classified into different sub-families. All of these data can be viewed on the CAZy (Carbohydrate-Active enzymes) website(http://www.cazy.org/GlycosylTransferases). To date, there are 94 GT families in total that have been identified.1.5.1 Glycosyltransferase Family 1Family 1 GTs have been demonstrated to catalyze the glycosylation of small molecular weight molecules. There is a subset of family 1 GTs which are the UDP-glucosyltransferases (UGTs). Previous studies showed that UGTs contain a 44 amino acid consensus sequence and it has also been demonstrated that this is a UGT signature motif which forms the UDP binding site (Nishimura et al., 2010).According to the PROSITE database (http://prosite.expasy.org/) of protein 18families and domains, the consensus sequences in the UGT subfamily vary in length from 435-507 amino acids and have nine conserved regions. The sequence similarities between UGT amino acid sequences varies from lower than 30% to over 95%. Very little information is available from plants regarding the expression of UGT genes, likely due to the fact that some family members are expressed at low levels and transcripts are rare. Studies of tomato and tobacco UGTs have demonstrated that UGTs respond rapidly to wounds and pathogen attacks (Langlois-Meurinne et al., 2005).The Bowles group did a series of studies of UGT in plants especially in Arabidopsis thaliana (Arabidopsis) (Priest et al., 2006). In the Arabidopsisgenome, there are 107 GTs containing the 44 amino acid UGT signature motif. The phylogenetic relationship of these 107 GTs has been studied and they have been divided into 14 groups, labeled A to N.1.5.2 Arabidopsis ABA UDP-glucoslytransferase UGT71B6ABA UDP-glucosyltransferase (ABA-UGT) is the enzyme that catalyzes the conjugation of UDP-glucose to ABA to form ABA-GE. An ABA-UGT hasonly been identified in Arabidopsis to date (Lim et al, 2005). In this study, all of the Arabidopsis UGT multigene family members were expressed in E. coli and then purified as fusion proteins. These fusions were then each screened for activity in vitro against (?)-ABA. Eight UGTs showed positive results. 19Then they found out that of these eight UGTs, only UGT71B6 showed activity towards (+)-ABA but not (-)-ABA. The results demonstrated that UGT71B6 is the only UGT that recognizes the natural enantiomers, (+)-ABA, so they concluded that UGT71B6 is the Arabidopsis ABA-UGT.The substrate selectivity of UGT71B6 was next studied (Priest et al., 2005). Twenty substrates including (+)-ABA , (-)-ABA and other ABA structural analogues were analyzed. The results showed that UGT71B6 is highly selective towards the natural enantiomer, (+)-ABA, and is able to glucoslyate many different structural analogues of ABA to varying degrees. The analyses towards ABA analogues can provide a foundation for the further investigation of features of ABA molecules.UGT71B6 has also been used to disturb ABA homeostasis in Arabidopsis(Priest et al., 2006). Transgenic plants overexpressing UGT71B6 werestudied to analyze effects on levels of ABA and its metabolites (ABA-GE, PA, DPA, neoPA and 7?OH-ABA) in plants which were grown under standard conditions or subjected to wilt stress. The result showed that the overexpression of UGT71B6 led to massive quantities of ABA-GE relative to the wild type, but had very little impact on absolute levels of free ABA and limited effects on plant growth and development. This suggested that the plant can accommodate substantial disturbance in ABA glucosylation. A 20transgenic plant with a knockout of UGT71B6 was also analyzed in this study, but no significant changes were observed among the profiles of ABA and its metabolites. This may have been because UGT71B6 may be highly specific to particular cells or particular developmental stages. The manipulation study of UGT71B6 provides a foundation and method to undertake further study of the role of ABA-GE in planta.Further, a recent study in 2012 demonstrated that UGT71B6?s expression level increased when the plants were placed under cold stress but decreased under heat stress. This study also suggested that the expression of UGT71B6 is coordinate with that of an ABA transporter (ABCG25, ABCG40)(Baron et al. 2012).In summary, the identification of UGT71B6 provides an important tool to study ABA metabolism and how free ABA and its metabolites regulate various aspects of plant development and stress responses. In addition, it is also important for identifying ABA UDP-glucosyltransferase orthologs in other species, including grapevine.1.5.3 UDP-glucosyltransferases in grapevineIn grapevine, there are several UGTs that have been identified and characterized. In 1998, UDP-glucose:3-O-glucosyltransferase (UFGT) was 21identified (Ford et al., 1998). It was demonstrated that UFGT is involved in the glucosylation of anthocyanins in planta, but not flavonols. It is the key enzyme that forms malvidin 3-O-glucoside, which is the major anthocyanin pigment in grape berries. In addition, they also indicated that UFGT gene expression is crucial for anthocyanin accumulation in the berry pericarp of red grape, while this gene is not expressed in tissues which do not accumulate anthocyanins like in vegetative tissues and berries of white cultivars. These findings suggest that anthocyanin biosynthesis in grape is controlled by the expression of UFGT (Boss et al., 1996). The crystal structure of UFGT has been revealed in a recent study (Petit et al., 2007).In addition, 5-O-glucosyltransferase has been demonstrated to be required for the transformation from malvidin 3-O-glucoside to malvin(3,5-di-O-glucoside) in some non-vinifera Vitis species (Janvary et al., 2009).A functional 5-O-glucosyltrasferase gene and its non-functional allele have been used as a diagnostic tool for the classification of wines according to their varietal origin; this tool is used to determine whether non-vinifera grape species are used in red wine making because it has been demonstrated that anthocyanin 3,5-di-O-glucoside is lacking in V. vinifera.Beside these UGTs, there are also many other GTs that have been studied in grape, these studies each provide information which can be used to classify 22other grape GT family members of unknown functions.1.6 ABA-UGT and glucosylation in ABA homeostasisThe ABA level in the plant cell is regulated by homeostasis mechanismsinvolving biosynthesis, catabolism, redistribution and inactivation. Glucosylation of ABA is the major pathway of ABA inactivation (Walton et al., 1995). This suggests that ABA glucosylation in the conjugated catabolism pathway and the ABA-UGT enzyme which catalyzes this process should be important in ABA homeostasis and those physiological processes that controlled by ABA such as seed dormancy, stomatal movement, response to abiotic stress and non-climacteric fruit ripening.1.7 Gateway cloning systemGateway technology is a cloning system that was invented by Invitrogen scientists around 15 years ago. It is a molecular biological method that enablesresearchers to efficiently transfer DNA fragments between different cloning vectors while maintaining the reading frame. The first step in Gateway cloning is inserting the gene of interest to the entry vector to generate an entry clone (Figure 1.6). In the entry clone, there are attL restriction sites on both sides of the gene of interest, which are necessary 23because these sites are cut by the Gateway recombinant enzymes to form sticky ends. These sticky ends complement the sticky ends on a destination vector which contains attR restriction sites. The process in which the expression clone is formed is called the LR reaction. Figure 1.6 Gateway cloning system (Patton et al.,2000)1.8 Research objectivesABA homeostasis may be important for the ripening process of non-climacteric fruit like grape berry so it could be targeted using biotechnology methods to control the timing and degree of certain aspects of fruit ripening. Manipulation of endogenous ABA in grape berries could offer 24cool-climate viticulturists a means to accelerate ripening and improve the fruit and wine qualities in late harvest cultivars. The Bowles group has published a series of research data regarding the identification and characterization of Arabidopsis abscisic acid UDP-glucosyltransferase (ABA-UGT) UGT71B6. The methods they used and the availability of UGT71B6 sequence data provide a solid foundation to study ABA-UGT in the wine making grapevine Vitis vinifera. In my dissertation research, I selected and cloned a number of candidate genes of Vitis viniferaABA-UGT based on the sequence of AtABA-UGT UGT71B6 and the expression trend of some Family 1 VviGTs. I have further expressed and purified the proteins that these candidate genes encode as GST fusions. Finally, the activities of these candidate enzymes of ABA-UGT were functionally evaluated. The results are discussed in chapter 3.252 VviABA-UGT candidate selection and the cloning, expression, purification and characteristic assay of candidate enzymes2.1 VviUDP-glucosyltransferase candidate selection2.1.1 Candidate selection by sequence similarityThe first method used in Vitis vinifera ABA-UGT candidate selection is based on similarity to the amino acid sequence of UGT71B6, the Arabidopsis ABA-UGT. I used Basic Local Alignment Search Tool (BLAST) in the National Center for Biotechnology Information (NCBI) website to search for candidates which have high sequence similarity to UGT71B6. The type of BLAST used was TBLASTN (search translated nucleotide database using a protein query), and the database used was the Vitis vinifera nucleotide collection, and a e-100 e-value cutoff was applied The top two results based on E-values were selected as VviABA-UGT candidates. These were designated as CAN1 and CAN2, the two candidates that have the highest sequence similarity to UGT71B6. An additional 21 candidates with low E-values from the TBLASTN search were selected for phylogenetic analyses.2.1.2 Candidate selection by evolutionary relationshipThe 21 candidates from the TBLASTN search described in section 2.1.1, above, and 20 Arabidopsis glucosyltransferase Family 1 protein members were aligned with UGT71B6 to perform a phylogenetic analysis observe their evolutionary relationship. The analysis was performed based on the 26conserved 44 amino acid motif in GT Family 1 proteins. Phylogenetic analyses with bootstrapping were carried out using MEGA 4 software(http://www.megasoftware.net/mega4) to generate a phylogenetic tree in order to determine which of the 21 candidates had the closest evolutionary relationship to UGT71B6. Additionally, some characterized Arabidopsis UGTs that do not carry out ABA glucosylation were included as functional outliers in these analyses.As shown in Figure 2.1, I determined that the two GTs indicated by arrowshave closer evolutionary relationships to UGT71B6, so these two proteins were selected as candidates 3 and 4 (CAN3 and CAN4)27Figure 2.1 Phylogenetic analysis Phylogenetic analysis of V. vinifera ABA-UGT candidates and some Arabidopsis UGTs. V. vinifera candidates are designated by their chromosomal locations. UGT71B5 UGT71B8 UGT71B6 UGT71B1 Chr18 Chr19 Chr12 Chr6 UGT71C1 Chr9 Chr14 Chr17 UGT89B1 Chr4 UGT72B1 UGT72E1 Chr16 Chr3 Chr15 Chr2 Chr8 UGT73C6 Chr5 Chr13 UGT76C1 ChrUn Chr11 Chr10 UGT84A1 UGT85A1 UGT75C1 UGT75B1 UGT74F2 UGT84B1 Chr1 UGT74B1 Chr7 Chr7R94987786464048416120823617115382443546382617115341834331820391600.05282.1.3 Candidate selection by expression level trendI postulated that changes in ABA-UGT expression levels should show a similar trend with ABA glucose ester (ABA-GE) levels in pericarp from different developmental stages. Since both the ABA-GE data throughout ripening and RNA-seq data for some grapevine GT1 family transcripts were available, I compared these two data sets and selected as candidates those GT1 family members whose expression level trends were consistent with patterns of ABA-GE levels.As shown in Figure 1.5, It is clear that ABA-GE level are relatively high at the beginning of ripening, go down prior to the veraison stage, and then keep increasing until the mature stage. I speculated that the expression level of VviABA-UGT should be low at the beginning of ripening and then increasesthrough the whole developmental process. RNA-Seq data for some V. vinifera GTs became available recently (Zenoni et al., 2010). I generated a table, shown in Figure 2.2, that summarizes relevant data for several Family 1 GTs. According to these data, six of those VviGTs (1, 13, 14, 18, 25, 26) have expression trends consistent with patterns of ABA-GE accumulation in pericarp; two of them (14, 18) was ruled out because its predicted amino acid sequence length is more than two times 29longer than UGT71B6 and other candidates. Three of the other four GTs (highlighted in green) were excluded because I was unable to use PCR-based cloning, due to these three genes all having introns plus their expression levels were too low to be amplified by using cDNA as PCR templates. So finally, only one of these VviGTs, shown in yellow, was selected as CAN5.30Figure 2.2 RNA-Seq data for some V. vinifera glucosyltransferases.Gene ID refers to Vitis vinifera Genome version 12.0. RPKM stand for Reads Per Kilobase per Million mapped reads which is a method to quantify gene expression level from RNA-seq data.2.2 Cloning of the five VviABA-UGT candidate genes and UGT71B62.2.1 Genomic DNA isolation from V. vinifera and ArabidopsisGenomic DNA was extracted from V. vinifera leaves by the CTAB protocol.31(Sue et al., 1997) Table 2.1 shows materials and methods for the CTAB protocol.Table 2.1 CTAB plant genomic DNA extraction protocolMaterials CTAB buffer 100ml2.0 g CTAB (Hexadecyl trimethyl-ammonium bromide)10.0 mL 1 M Tris, pH 8.04.0 mL 0.5 M EDTA, pH 8.028.0 mL 5 M NaCl40.0 mL H2O1 g PVP 40 (polyvinyl pyrrolidone (vinylpyrrolidine homopolymer) Mw 40,000)Adjust to pH 5.0 with HCl and make up to 100 mL with H2O.1 M Tris, pH 8.0Dissolve 121.1 g of Tris base in 800 ml of H2O. Adjust pH to 8.0 by adding 42 mL of concentrated HCl. Allow the solution to cool to room temperature before making the final adjustments to the pH. Adjust the volume to 1 L with H2O. Sterilize using an autoclave.5x TBE buffer 54 g Tris base 27.5 g boric acid 32Table 2.1 CTAB plant genomic DNA extraction protocolMaterials 5x TBE buffer20 mL of 0.5M EDTA (pH 8.0) Make up to 1 L with deionized water. To make a 0.5X working solution, do a 1:10 dilution of the concentrated stock.Procedure - Grind 200 mg of plant tissue to a fine paste in approximately 500 ?L of CTAB buffer. - Transfer CTAB/plant extract mixture to a microfuge tube. - Incubate the CTAB/plant extract mixture for 15 min at 55 oCin a recirculating water bath. - After incubation, spin the CTAB/plant extract mixture at 12,000 x g for 5 min to spin down cell debris. Transfer the supernatant to clean microfuge tubes. - To each tube add 250 ?L of Chloroform:Isoamyl alcohol (24:1) and mix the solution by inversion. After mixing, spin the tubes at 13,000 xg for 1 min. - Transfer the upper aqueous phase only (contains the DNA) to a clean microfuge tube. - To each tube add 50 ?L of 7.5 M ammonium acetate followed by 500 ?L of ice cold absolute ethanol. - Invert the tubes slowly several times to precipitate the DNA. Generally the DNA can be seen to precipitate out of solution. Alternatively the tubes can be placed for 1 hr at -20 oC after the addition of ethanol to precipitate the DNA- Following precipitation, the DNA can be pipetted off by slowly rotating/spinning a tip in the cold solution. The 33Table 2.1 CTAB plant genomic DNA extraction protocolProcedure precipitated DNA sticks to the pipette and is visible as a clear thick precipitate. To wash the DNA, transfer the precipitate into a microfuge tube containing 500 ?L of ice cold 70% ethanol and slowly invert the tube. Repeat. (alternatively the precipitate can be isolated by spinning the tube at 13,000 rpm for 1 min to form a pellet). Remove the supernatant and wash the DNA pellet by adding two changes of ice cold 70% ethanol. - After the wash, spin the DNA into a pellet by centrifuging at 13,000 rpm for 1 min. Remove all of the supernatant and allow the DNA pellet to dry (approximately 15 min). Do not allow the DNA to over dry or it will be hard to re-dissolve. Resuspend the DNA in sterile DNase-free water (approximately 50-400 ?L H2O; the amount of water needed to dissolve the DNA can vary, depending on how much is isolated). RNaseA (10 ?g/ml) can be added to the water prior to dissolving the DNA to remove any RNA in the preparation (10 ?L RNaseA in 10mL H2O). - After resuspension, the DNA is incubated at 65 ? for 20 min to destroy any DNases that may be present and store at 4?. - Agarose gel electrophoresis of the DNA will show the integrity of the DNA.The concentration of isolated gDNA was quantified by ND-1000 NanoDropspectrophotometer (Thermo Scientific, Delaware USA) and then diluted the 34gDNA to 100ug/ml for PCR. The quality of gDNA was verified by agarose gel electrophoresis and the result is shown in Figure 2.3(a). Genomic DNA quality was also tested by PCR using primers designed to amplify actin and ubiquitin C (UBC). PCR results are showed in Figure 2.3(b). The agarose electrophoresis gel and PCR protocols can be found in 2.2.2 and 2.2.3.a bFigure 2.3 Verification of isolated genomic DNAAll the DNA markers used are GeneRuler 1 kb Plus DNA Ladder (ThermoScientific, Ottawa, Ontario)2.2.2 Polymerase chain reaction (PCR) to amplify candidate genes. Primer design: Depending on the pENTR/TEV/TOPO vector I used, 4 base pairs, CACC, were added to the 5? end of the forward primer in order to 35enable directional cloning. DNA Primers were made by Integrated DNA Technology (IDT, Iowa USA)Table 2.2 PCR reaction conditionPCR reaction mixPfu DNA Polymerase 10X Buffer with MgSO4 5?LdNTP mix, 10 mM each 1 ?Lforward primer 10 uM 1 ?Lreverse primer 10 uM 1 ?LgDNA template 10-100 ngPfu DNA polymerase 0.5 ?LNuclease-Free Water to final volume of 50 ?LPCR reactionconditionStep Temp Time Number of CycleInitial Denaturation 95? 1min 1 cycleDenaturationAnnealing*Extension95?45-60?72?30sec30sec3min30 cyclesFinal Extension 72? 5min 1 cycle*temperature gradient experiments were applied to optimize the yield of PCR product.The PCR results were verified by agarose gel electrophoresis (see Figure 2.3.3) After the temperature gradient PCR experiments were done to find out the most optimal annealing temperatures for each candidate gene, the reaction 36volumes were increased to 120 ?L to produce more PCR products for cloning. 2.2.3 Amplified candidate genes isolated agarose gelPCR products were separated by gel electrophoresis in a 1% (w/v) agarose gel made in 1 x TAE (40 mM Tris, 14% (v/v) acetic acid, 1 mM EDTA) and containing 0.05?L/mL SYBRSafe DNA gel stain (Invitrogen). Prior to loading, loading dye (50% (v/v) glycerol, 0.25% (w/v) bromophenol blue) was added at 1/10 the final volume. The gel was run at 90 V in a running buffer of 1 x TAE for 30 minutes. PCR products were visualized by UV light using a Kodak molecular imaging system. Figure 2.4 shows the final PCR results.Figure 2.4 PCR results of candidate genesPCR products were cut from agarose gel and extracted by using GeneJET Gel Extraction Kit (Thermo Scientific) M AT 1 2 3 4 51.5kb37Table 2.3 GeneJET gel extraction kit protocolStep Procedure1 Excise gel slice containing the DNA fragment. Place the gel slice into a pre-weighed 1.5 mL tube and weigh. Record the weight of the gel slice.2 Add 1:1 volume of Binding Buffer to the gel slice.3 Incubate the gel mixture at 50-60?C for 10 min. Mix the tube by inversion every few minutes to facilitate the melting process.4 Transfer up to 800 ?L of the solubilized gel solution to the GeneJET purification column. Centrifuge for 1 min (13,000 x g). Discard the flow-through and place the column back into the same collection tube.5 Add 700 ?L of Wash Buffer to the GeneJET purification column. Centrifuge for 1 min (13,000 x g). Discard the flow-through and place the column back into the same collection tube.6 Centrifuge the empty GeneJET purification column for an additional 1 min (15,000 x g) to completely remove residual wash buffer.7 Transfer the GeneJET purification column into a 1.5ml microfuge tube. Add 50 ?L of Elution Buffer to the center of the purification column membrane. Centrifuge for 1 min (13,000 x g).8 Discard the GeneJET purification column and store the purified 38Table 2.3 GeneJET gel extraction kit protocolStep Procedure8 DNA at -20 ?C. The concentration of purified PCR products were quantified using a NanoDrop spectrophotometer.2.2.4 TOPO cloning reactions to generate entry clonesPurified PCR products were cloned into the pENTR/TEV/TOPO vectors which is the vector I used to generate entry clones, Figure 2.5 shows the map of pENTR/TEV/D-TOPO vector Table 2.4 shows the TOPO cloning reaction protocol which was used to generate each entry clone containing the five VviABA-UGT candidates and the AtABA-UGT positive control. The strains of bacteria used to generate entry clone was One Shot TOP10 Chemically Competent Escherichia coli (Invitrogen) 39Figure 2.5 Maps of pENTR/TEV/D-TOPO vector and pDEST15 vector.40Table 2.4 TOPO Cloning ReactionReagents VolumeFresh PCR product 1 ?L (5-10 ng)Salt solution (supplied with kit) 1 ?L Sterile water add to final volume of 5 ?LpENTR/TEV/TOPO vector 1 ?LFinal volume 6 ?LReactions were gently mixed and incubated for 5 minutes at room temperature (22-23 ?) and then the reactions were transformed into One Shot TOP10 Chemically Competent E. coli.Two microlitres of TOPO cloning reaction was added into a vial of One Shot TOP10 Competent E. coli, and then cells were incubated on ice for 30 minutes. After that, cells were heat-shocked for 30 seconds at 42? without shaking and then samples were immediately transferred to to ice. Next, 250?L of room temperature S.O.C medium(2% w/v tryptone, 0.5% w/v Yeast extract, 8.56 mM NaCl, 2.5 mM KCl,10 mM MgCl2, 10 mM MgSO4, 20 mM glucose) was added. Transformed cells were shaken (200 rpm) at 37? for 1 hour and then 100 ?L from each transformation was spread on a pre-warmed selective plate and incubated overnight at 37 ?. The LB medium I used in each plate included 50 ?g/mL of the selective marker, kanamycin.41Positive clones were selected by colony PCR. Each positive colony was picked with a sterile pipette tip and then this was twirled in the PCR reaction buffer, so the colony worked as a template in each PCR. Primers used in colony PCR were the forward primer for each gene of interest and the M13R primer.Positive clones selected by colony PCR were cultured overnight in LB medium containing 50 ?g/mL kanamycin and then QIAprep Spin Miniprep Kit was used to isolate plasmid DNA. Samples were stored at -20?.Table 2.5 QIAprep Spin Miniprep Kit protocol1. Pellet 1?5 mL bacterial overnight culture by centrifugation at 8,000 rpm for 3 min at room temperature.2. Resuspend pelleted bacterial cells in 250 ?L Buffer P1 and transfer to a microcentrifuge tube.3. Add 250 ?L Buffer P2 and mix thoroughly by inverting the tube 4?6 times until the solution becomes clear. 4. Add 350 ?L Buffer N3 and mix immediately and thoroughly by inverting the tube 4?6 times.5. Centrifuge for 10 min at 13,000 rpm in a table-top microcentrifuge.42Table 2.5 QIAprep Spin Miniprep Kit protocol6. Apply the supernatant from step 5 to the QIAprep spin column by decanting or pipetting. Centrifuge for 1 min and discard the flow-through.7. Wash the QIAprep spin column by adding 0.75 mL Buffer PE. Centrifuge for 1 min at 13,000 rpm and discard the flow-through, wash twice.8. Centrifuge for 1 min at 15,000 rpm to remove residual wash buffer.9. Place the QIAprep column in a 1.5 mL microcentrifuge tube. To elute DNA, add 50 ?L Buffer EB to the center of the QIAprep spin column, let stand for 2 min, and centrifuge for 2 min. Collect the flow through and store at -20?.Isolated plasmids were double checked by PCR (using gene of interest forward primer and M13R primer).2.2.5 Perform LR recombination reaction to generate expression cloneAn LR recombination reaction was performed using Gateway LR Clonase enzyme mix to transfer each gene of interest from its entry construct to the destination vector, pDEST15, to generate an expression clone. The map of pDEST15 vector is showed in Figure 2.643Table 2.6 LR reaction procedures1. Add the following components to a 1.5 mL microcentrifuge tube at room temperature and mix:Entry clone (50-150 ng) 1-7 ?LDestination vector (150 ng/?L) 1 ?LTE buffer, pH 8.0 to 8 ?L2. Thaw the LR Clonase? II enzyme mix on ice for about 2 min. Vortex the LR Clonase? II enzyme mix briefly, twice. 3. To each sample, add 2 ?L of LR Clonase? II enzyme solution and mix well by vortex.4. Return LR Clonase? II enzyme mix to -20?C or -80?C storage.5. Incubate reactions at 25 ?C for 1 h.6. Add 1 ?L of the Proteinase K solution to each sample to terminate the reaction. Vortex briefly. Incubate samples at 37 ?C for 10 min.One ?L of each LR reaction was then transformed into 50 ?L of DH5? competent E. coli cells and incubated on ice for 30 minutes. Transformation was carried out by heat-shocking the cells by incubating at 42 ? for 30 seconds. Then 250 ?L of S.O.C medium was added and this was incubated at 37 ? for one hour with shaking. Fifty ?L of each transformation was plated onto selective media containing 100 ?g/mL of ampicillin and grown44overnight. The reason why I used DH5? cells here as an intermediate step and I did not directly transform the vectors to the expression E. coli strain, Rosetta-gami 2 (DE3) pLysS, is because Rosetta-gami 2 (DE3) pLysS can only produce very low amount of plasmid DNA and cannot be stored for long periods. DH5? cells can produce high amounts of plasmid DNA and they are hardy. So DH5? cells worked for the storage of expression clones long-term and were utilized to produce more plasmids for DNA sequencing. As described in section 2.2.4, positive expression clones of candidate genes were select by colony PCR and the isolated plasmids were verified by PCR (T7 forward and reverse primers which amplify flanking plasmid DNA) and DNA sequencing.Figure 2.6 shows PCR verification of the expression clones. Expression plasmids isolated from DH5? were stored at -20 ? for the next step plusglycerol stocks were prepared and stored at -80 ? (500 ?L bacterial culture, 500 ?L 30% glycerol).45Figure 2.6 PCR verification of expression cloneLanes from left to right are Marker (M), two controls, each UGT71B6 (C) and 5 candidate genes (1-5). The upper part shows the PCR results by using T7 primers and the lower part shows the PCR results by using T7R and the forward primers of candidate genes.2.3 Expression and purification of recombinant proteinsThe pDEST15 vector was used for expression of recombinant proteins (VviABA-UGT candidates and AtABA-UGT) as N-terminal fusions tagged with a 26 kD GST domain from Schistosoma japonicum. The GST tag allows the recombinant protein to be purified using glutathione HiCap matrix (Qiagen) and recovered by exchanging with reduced L-glutathione (Sigma-Aldrich).462.3.1 Transformation of expression vector to Rosetta-gami (DE3)pLysS competent cellsThe strain of E. coli used in the protein expression work was Rosetta-gami 2(DE3)pLysS Competent Cells (Novagen). Expression plasmids isolated from DH5? cells were transformed to Rosetta-gami 2 (DE3) pLysS cells. The transformation protocol used was similar to the transformation protocol used in generating the entry clone, the only difference being that Rosetta cells were grow much slower than normal E. coli cells. Colonies on selective plates started to appear after 24 hours, so the plates were incubated at 37 ?for 24-30 hours to produce enough colonies. Positive clone were selected by colony PCR and verified by plasmid PCR. Transformed Rosetta-gami 2 (DE3) pLysS cell cultures were kept at -80? with glycerol and the glycerol stocks were re-made after every time they were used. 472.3.2 Optimization of heterologous protein expression conditions Table 2.7 shows the standard procedure for protein expression Table 2.7. Protein expression procedureStep Procedure1 Use a sterile pipette tip to scrape some of the frozen bacteria, leave the pipette tip in a culture tube with 2 mL TB (1.2% w/v tryptone, 2.4% yeast extract, 72 mM K2HPO4, 17 mM KH2PO4, 0.4% glycerol) medium with 100 ?g/mL ampicillin. Grow the cultures for 24 hours at 37 ? on a rotational shaker set to 225rpm. 2 After 24 hours, transfer 1 mL of culture to a 500 mL flask with 150 mL TB medium with 100 ?g/mL ampicillin and 40 ?g/mL leucine). Grow the culture at 37 ?, 225 rpm, until an OD600 of 0.4.3 Add IPTG to medium and induce the bacterial at lower temperature. See text, below, for parameters tested.4. Collect 1 mL culture for SDS-PAGE and Western Blot analyses.Parameters that may affect protein expression are inducing temperature, inducing time, and IPTG concentration. In order to optimize these parameters, one grapevine candidate gene (CAN1) was used to carry out a series of 48experiments.First, three temperatures (18 , 25 , 37 ) were compared by a simple overnight expression using a standard concentration of IPTG (1 mM). One mL of cell culture was analyzed by anti-GST Western blot (see 2.4.3 and 2.4.4 for SDS-PAGE and Western Blot protocols). The results suggested that there was no significant difference on the yield of recombinant proteins in response to these three temperatures. So in order to minimum the possibility of getting inclusion bodies which will not have enzyme activity, 18  was the temperature I used for all subsequent expressions.A second set of experiments was applied to find out the most optimal IPTG concentration at an induction temperature of 18 . Three cell cultures were tested in this experiment. The first cell culture was induced without IPTG so it worked as a negative control, whereas the other two cell cultures were induced with 0.05 mM IPTG and 1 mM IPTG, respectively. These cultures were each induced at 18 for 24 hours, then one mL of cell culture was picked every 6 hours and analyzed by anti-GST Western blot. The result of this experiment is shown in Figure 2.7. It was obvious that 0.05 mM of IPTG was sufficient for protein expression. The cell culture which was induced with 0.05 mM IPTG had an earlier yield and smoother expression than the culture induced with 1 mM IPTG. In addition, the result also 49suggested that the protein was produced throughout the 24 hour induction period because the signal intensity on the Western blot became stronger and stronger.Figure 2.7 Optimization of inducing time and IPTG concentration.The first lane, C, shows a negative control in which no IPTG was added.I concluded from these experiments that the best condition for protein induction was 18  for 24 hours with 0.05 mM IPTG. And since the nature of all the five candidate proteins are similar, they are supposed to share similar expression conditions, and this was also demonstrated when I performed the expression of other four candidate proteins.502.3.3 SDS-PAGE protein gel protocolTable 2.8 SDS-PAGE materialsRunning Gel Solution1.5 mL H2O, 1.5 mL Acrylamide/Bis (30%/0.8% w/v) 0.5 mL 50 % glycerol, 1.5 mL 1 M Tris (pH 8.8)65 ?L 10% APS, 50 ?L 20% SDS, 3.5 ?L TEMEDStacking Gel Solution2.5 mL H2O, 320 ?L Acrylamide/Bis (30%/0.8% w/v) 390 ?L,1 M Tris (pH 6.8), 31 ?L10% APS, 31 ?L 20% SDS, 2 ?LTEMED10x Running Buffer (1 L)Tris 30 gGlycine 142 gSDS 10 gCoomassie Blue StainPageBlue Protein Staining Solution5x SDS Sample Buffer(40 mL)16 mL ddH2O, 5 mL 0.5M Tris (pH 6.8) 8 mL 50% Glycerol, 8 mL 10% SDS2 mL 2-?-mercaptoethanol, 20 mg bromophenol blueSample preparation:Add five ?L of 5X SDS sample buffer to 20 ?L of sample and then heat the sample in a 37  water bath for 10 minutes. If the sample is a cell culture, 51centrifuge one mL at 13000 x g for 10 minutes, discard the supernatant, and then use 200 ?L of 5X SDS sample buffer to resuspend the cell pellet. Boil the sample for five minutes, centrifuge at 13000 x g for 10 minutes, and retain the supernatant for SDS-PAGE.SDS-PAGE protocol:First add four mL running gel solution quickly into gel caster form and then pour two mL stacking gel solution on top of the running gel. Insert the combs and wait for 30 minutes until the gel polyperizes. After which, load samplesand run the gel at 14 mA for 30 minutes in 1X SDS Running Buffer. After that, raise the current to 60 mA and run for one hour. At this point, the gel can either be transferred to a membrane (see anti-GST Western blotting protocol) or stained with Coomassie depend on the purpose.If the gel was going to be transferred to a PVDF membrane for a Western blot analysis, see section 2.3.4 for more details on this procedure. If the gel was going to be stained with Coomassie blue, first wash the gel three times with deionized water, washing five minutes each time. Then transfer the gel to a box with Page Blue Protein Staining Solution (from Thermo Fisher),microwave the box for 30 seconds, and then put the box on a shaker and shake for one hour. Afterwards, wash the gel in deionized water with shaking until bands are discernible.522.3.4 Anti-GST western blot protocolAn SDS-PAGE gel can be transferred to a membrane to perform anti-GST Western Blot to detect GST-tagged recombinant protein.Table 2.9 Anti-GST Western blot buffers1X Transfer buffer (1L)14.4 g glycine, 3.02 g Tris base, 200 mL methanol Add ddH2O to a final volume of 2 L1X Blockingbuffer5% (w/v) non fat milk in PBS0.1% Tween 201X Washing Buffer1X PBS0.1% Tween 20The transfer procedure was carried out using the Semi-Dry Blotting Unit(Fisher Scientific, Ottawa Ontario). Two blotting pads and Polyvinylidene fluoride (PVDF, BIO-RAD, Mississauga, Ontario) transfer membrane were soaked in transfer buffer, the PVDF was then pre-wetted in 100% methanol. After that, one pad was placed on the bottom of the unit, the gel was put ontothe pad, and the PVDF transfer membrane was put on top of the gel. Then the other pad was placed on top of the PVDF membrane. The transfer condition was 60 mA for 18 hours (or 15V for 30 minutes).53After transfer, the PVDF membrane was moved to 50 mL of 1X blocking buffer immediately. The membrane was immersed in blocking buffer for one hour at room temperature with shaking.The membrane was then incubated with primary antibody (4 ?L of anti-GST polyclonal antibody in 4 mL of blocking buffer) for one hour at 37  without shaking. Then the membrane was washed three times with washing buffer for 10 minutes with shaking.Next, the membrane was incubated with secondary antibody (0.8 ?L of anti-rabbit IgG antibody-alkaline phosphate in 4 mL of blocking buffer) for one hour at 37  without shaking. Then the membrane was washed three times with washing buffer.At the final step, the membrane was stained with 5 mL BCIP/NBT substrate until the bands became visible.2.3.5 Purification of recombinant proteinsRecombinant fusion protein purification was carried out using a glutathione HiCap matrix (QIAGEN, Toronto, Ontario). The manufacturer?s recommended materials and protocols are shown, below.54Table 2.10 GST fusion protein purification materials and buffersPBS 50 mM NaH2PO4, pH 7.2150 mM NaClPBS-L(cell lysis buffer)1X PBS, 1 mg/mL lysozymeProtease inhibitor* 5 ?L/mLDNase I50 U/mL, 1 M MgCl2 1 ?L/mL1 mM DTT, 1 mM EDTA 1% (v/v) Triton X-100PBS-EW(equilibration and wash buffer)1X PBS1 mM DTT1 mM EDTATNGT (elution buffer)50 mM Tris, pH 8.0 0.4 M NaCl, 50 mM L-glutathione (reduced)0.1% Triton X-100, 1 mM DTT* Protease Inhibitor Cocktail (Sigma-Aldrich, Missouri, USA)2.5.4.1 Preparation of cleared E. coli cell lysateThaw the cell pellet derived from 100 mL cell culture on ice for 15 minutes and resuspend the pellet in five mL buffer PBS-L. Then incubate on an end-over-end shaker at room temperature for 30 minutes. Next, sonicate the cell lysate three times for 10 seconds each time. Centrifuge the cell lysate at 5515,000 x g for 20 minutes at 4  to pellet the cellular debris; repeat one to two times in order to make sure all of the cell debris that may stick in the purification column has been removed. Save the supernatant and add 5 ?L4X SDS-PAGE sample buffer to 20 ?L supernatant for SDS-PAGE and Western blot analysis.2.5.4.2 Purification of GST-tagged proteinAdd 1 mL of the 50% Glutathione HiCap Matrix slurry (500 ?L bed volume) into an empty column. Wash the settled resin with 10 column volumes of buffer PBS-EW twice. Load the lysate into the resin, let the cell lysate incubate with the resin for one hour at 4  and then save the column flow-through for SDS-PAGE and Western blot analysis. Wash the column three times with 2.5 mL buffer PBS-EW and save the wash fractions for SDS-PAGE and Western blot analysis. Add 500 ?L of elution buffer TNGT and collect the eluate containing the GST-tagged protein, repeating this step twice, and store eluted protein in -80 with 15% (v/v) glycerol and 1% (v/v) protease inhibitor.During the purification procedure, the cell lysate, the rest of the cell lysate after binding to column, the washing fractions and the final eluate were collected and analyzed by anti-GST Western blot. The Western blot result is shown in Figure 2.8. The data suggested that the recombinant GST-fusion 56proteins were not able to bind the GST resin, which means the conditionsneeded to be optimized.Figure 2.8 Protein purification result using manufacturer?s recommended procedures.Lanes from left to right are cell lysate (1), lysate after binding (2), washing flowthrough (3) and eluate (4). It was clear that an 80 kD band of GST fusion protein failed to bind with anti-GST column and not in the eluate.There are various factors that may affect the binding affinity, including binding temperature and time, binding pH, the concentration of dithiothreitol (DTT) and Triton X-100 in the lysis buffer, the sonication condition and the 1 2 3 4100kD70kD57elution condition. So these factors were adjusted to improve the binding. For binding and other conditions, the pH of the cell lysate was switched from 7.4 (standard) to 8.5-9, and two binding temperatures were tried, which were room temperature for one hour or 4  overnight. In addition, room temperature (one hour) and 4  (overnight) were also applied in the elution step. The results showed that the pH and the temperature conditions were not the reasons of low binding affinity (data not shown). Besides, I also tried to perform cell lysis without sonication because the heat produced by sonication may damage the proteins, but it also did not help the binding (data not shown).DTT is the reducing agent that is used in cell lysis buffer. The role of DTT in protein purification is to avoid oxidative damage to proteins. Oxidized GST-fusion protein would lose its binding affinity to GSH resin. So the concentration of DTT in the cell lysis buffer was increased to 5 mM to determine whether it could solve the problem. Note that the concentration of DTT cannot be higher than 5 mM because higher level of DTT would reduce the disulfide bonds of proteins, causing them to unfold and then lose their enzymatic activities.Triton X-100 is a mild non-ionic detergent which can increase solubilization 58of cell membranes and lysis and some proteins will be more soluble with detergents like Triton X-100; however, the detergent may also affect the activities of proteins and their binding affinities, so Triton X-100 was removed from the cell lysis buffer and elution buffer to see if it would help binding. The results of these changes in reducing agent (DTT) and detergent (Triton X-100) did not improve the binding affinity (data not shown).I then considered the possibility that the GST tag on the target proteins was not optimally exposed for binding. I tested a method that has been previously shown to alleviate the problem which was to add up to 4 M urea in the cell lysis buffer in order to destabilize the conformation of GST fusion proteins. Three urea concentrations were tested, 1 M, 2 M, and 4 M. The results showed that the addition of urea in the lysis buffer did improve the binding affinity and whereas the urea concentration tested did not show significant differences on the function. So finally 1 M urea was applied because higher concentration of urea will denature the proteins which means the enzymes may lose their activities. Figure 2.9 shows the western blot analysis result of each steps of purification after the application of 1M urea. 59Figure 2.9 Western blot gel showing effects of 1 M urea on protein purification.Lane from left to right are Marker (M; Protein MW Marker, Ladder, Prestained, 10-170 kD, Usbio, Burlington, Ontario), cell lysis (1), washing flowthrough (2) and eluate (3).2.3.6 Chaperone protein DnaKFigure 2.10 shows an SDS-PAGE gel with all the six recombinant proteins include the five grapevine candidates (see 2.1) and UGT71B6. A 65 kDprotein co-eluting with GST tagged proteins was consistently and clearly evident. This well-known contaminant was mostly likely DnaK protein. DnaK is a chaperone protein that is always expressed in some E. coli strainsand known to co-elute with GST fusion proteins. DnaK always binds with ATP, so an additional wash step with a buffer containing ATP and MgSO4 can be used to remove DnaK; however, this additional wash step must be performed in a 37  water bath for an hour. Since this temperature and time 80kD 60may cause the denaturation of the fusion proteins then lose their enzyme activities, and DnaK is not expected to affect the enzyme activities of my recombinant proteins, I did not apply this additional wash step.Figure 2.10 Purified recombinant proteins.Lanes from right to left are marker (M), UGT71B6 (C) and five candidate fusion proteins (1-5). The upper arrow around 80 kD shows the purified fushion protein, the lower arrow around 65 kD shows DnaK.2.3.7 GST tag cleavageAccording to the research of the Bowles group (Priest et al., 2005) about the substrate selectivity of UGT71B6, GST-tagged AtABA-UGT showed thesame glucosylation activity with or without a GST tag. So for my work, I performed the enzyme assays with GST-tagged enzyme. I also performed GST tag cleavages and stored the cleaved enzymes as backups.GST tags can be removed by the TEV protease because the entry vector 100 kD70 kD 5 4 3 2 1 C M80 kD65 kD61introduced a TEV site in the N-termini of the proteins. The protocol for GST tag removal using AcTEV protease (Invitrogen) is shown, below.First a TEV cleavage reaction was set up in a microcentrifuge tube: Table 2.11 TEV cleavage reaction mixFusion protein 20 ?g20X TEV Buffer 7.5 ?L0.1 M DTT 1.5 ?LAcTEV protease, (10 units) 1.0 ?LWater To 150 ?LThe reaction was incubated in a 30  water bath for 6 hours. Then the reaction was stored at -20 for next step.After the reaction with TEV protease, there are two methods to remove the free GST tag and the TEV protease remaining in the reaction mix. The first way is based on the chromatography, since there is a His binding site on Invitrogen?s TEV protease. So a two step chromatography can be used here, first using anti-GST tag to remove the free GST tag, then keeping the flow-through and using anti-His tag to remove the remaining TEV protease. The second way is much simpler. Since the molecular weight (MW) of the 62GST tag is around 26 kD and the MW of TEV protease is 27 kD, a 30 kDprotein cutoff filter can be used to remove both of them by centrifugation.2.4 Recombinant enzyme assayAn ABA-glucose ester quantitative analysis was performed to test the activities of my candidate grapevine ABA-UGTs toward (+)-ABA. I also tested UGT71B6 as a positive control for the assay and used (-)-ABA as substrate in negative controls to provide supporting evidence for the biological relevance of any grapevine candidates showing glucosyltransferase activity towards (+)-ABA.2.4.1 Enzyme assay reactionsTable 2.12 shows the protocol I used to functionally evaluate all the ABA-UGT candidate enzymes.63Table 2.12 ABA-UGT enzyme assay protocolStep Procedure 1 Set up reaction:1 ?g recombinant protein, 1 mM (+)-ABA or (-)-ABA5 mM UDP-glucose, 5 mM MgCl2, 10 mM DTTAdd 50 mM Tris-HCl (pH 6.95) to 100 ?L2 Mix the reaction and stay on 30  water bath for 1 hour3 Add 10 ?L 240mg/mL trichloroacetic acid (TCA), and stay 30 minutes in room temperature to stop the reaction4 Centrifuge the tube at 15,000 x g for 10 minutes.5 Save the supernatant and store it at -20 2.4.2 ABA-glucose ester quantitative assayThe reaction products from 2.4.1 were sent to Plant Biotechnology Institute (National Research Council (NRC), Saskatoon, SK) at which assays were performed to quantify the products of my enzyme assays with my five VviABA-UGT candidates and the AtABA-UGT control.According to the protocol that the NRC provided, the samples were subjected to ultra-performance liquid chromatography-electrospray tandem mass spectrometry (UPLC-ES-MS/MS) analysis and quantification, similar to that 64described in detail in Ross et al. (2004). Samples were injected onto an ACQUITY UPLC HSS C18 column (2.1x100 mm, 1.8 ?m) with an ACQUITY HSS C18 VanGuard Pre-column (2.1x5mm, 1.8 ?m) and separated by a gradient elution of water containing 0.025% acetic acid against an increasing percentage of acetonitrile containing 0.025% acetic acid.Briefly, the analysis utilizes the Multiple Reaction Monitoring (MRM) function of the MassLynx v4.1 (Waters Inc., Longueuil, Qu?bec) control software. The resulting chromatographic traces are quantified by the QuanLynx v4.1 software (Waters Inc., Longueuil, Qu?bec) off-line, in whereeach trace is integrated and the resulting ratio of signals (non-deuterated/internal standard) is compared with a calibration curvewhich is previously constructed to yield the amount of metabolite present (ng per sample). Figure 2.11 shows the quantitative levels of ABA and its metabolites (DPA, ABA-GE, PA, 7?OH-ABA, neo-PA and t-ABA) in all the six reaction samples. It suggested that only the control UGT71B6 (sample 1) successfully catalyzed the glycosylation and produced around 92ng/?l ABA-GE; however, each of the five candidate enzymes (samples 2-6) failed to catalyze the reaction and none of those ABA metabolites were produced.65Figure 2.11 Quantitative enzyme assay resultThis figure shows the quantitative level of ABA and its metabolites in each of the six samples. Sample 1 was the one with UGT71B6, sample 2-6 were CAN1-5.All the five reactions were also analyzed by HPLC in The Mass Spectrometry Laboratory (UBC Wine Research Centre, Vancouver, BC). The results showed that ABA-GE was not observed in all of the five reactions(data not shown). 663 General discussionThe identification of Vitis vinifera ABA-UGT will offer scientists and viticulturists a means to modify its expression in fruit, by which they can alter ABA catabolism in order to potentially control the pace of ripening. If the time required for ripening can be reduced, it will help to solve the climate issue that concerns the British Columbia wine industry. This dissertation research selected five VviABA-UGT candidates and cloned, expressed, purified and functionally evaluated each of them. The experimental details and results were shown in chapter 2.Since the recombinant enzyme assay result showed that each of the five VviABA-UGT candidates that I selected failed to catalyze the glucosylation of ABA to ABA-GE, I concluded that none of them is the Vitis vinifera ABA-UGT. Based onthe control recombinant enzyme assay result with the previously characterized Arabidopsis ABA-UGT, UGT71B6, the conjugation of ABA and UDP-glucose toABA-GE conclusively demonstrates that the methodology that I used in the assay was correct and the recombinant proteins that I purified did not lose enzyme activities in the process of protein purification or any other step. For future studies on the functional identification of VviABA-UGT, candidate selection should focus more on the expression level of gene and protein candidates at the early veraison stage. By doing BLAST searches, I found that there are over 80grapevine UGTs that have similar e-values (less than e-80), so it is very hard to base 67gene candidate hypotheses on closer sequence similarities toward UGT71B6 than the others. According to the level of ABA-GE in different berry developmental stages,grapevine ABA-UGT should have an evident expression at an early veraison stage, so it could be a wise way for the future identification of VviABA-UGT candidates.An alternate approach to selecting a small number of VviABA-UGT candidates based on amino acid sequence or expression profiling data would be to perform high throughput assays.According to the Genoscope Grape Genome Browser (v. 12.1) (http://www.genoscope.cns.fr/externe/GenomeBrowser/Vitis/), the grapevine genomeassembly is nearly complete and 361 putative Family 1 UGTs can be retrieved from the database. So a large scale of candidate screening is a possible method to identify VviABA-UGT, similar to what the Bowles group carried out in Arabidopsis (Lim et al., 2005), but it would be much more demanding to do it on grapevine because there are only 107 genes in the Arabidopsis UGT family.Furthermore, a PCR-based method is also a possible approach. This would entail, first,designing a pair of degenerative primers based on the UGT Family 1 protein motifand then cloning these using PCR on cDNA libraries prepared from RNA samples from early veraison fruit stages. Then, a series of gene which express at early veraison stage will be cloned, they could be separated by single colony isolation and the 68VviABA-UGT definitely should be one of them.69ReferencesAlexander L., Grierson D. (2002) Ethylene biosynthesis and action in tomato: a model for climacteric fruit ripening. J. Exp. Bot. 53: 2039-2055Arai S., Todoroki Y., Ibaraki S., Naoe Y., Hirai N., et al(1999). Synthesis and biological activity of 3-chloro, -bromo, and -iodoabscisic acids, and biological activity of 3-fluoro-8-hydroxyabscisic acid. Phytochem. 52:1185?93Argueso C., Hansen M., Kieber J. (2007) Regulation of ethylene biosynthesis. J. Plant Growth Reg. 26: 92?105Aspler T. (2012) Wine Industry. The Canadian EncyclopediaBaron, K.N., Schroeder, D.F. and Stasolla, C. (2012) Transcriptional response of abscisic acid (ABA) metabolism and transport to cold and heat stress applied at the reproductive stage of development in Arabidopsis thaliana. Plant Sci. 188?189: 48?59Boss, P. K., Davies, C. & Robinson, S. P. (1996) Analysis of the expression of anthocyanin pathway genes in developing Vitis vinifera L. cv. Shiraz grape berries and the implications for pathway regulation. Plant Physiol. 111: 1059? 1066Chervin C, Tira-Umphon A, Chatelet P, Jauneau A, Boss PK, Tesniere C (2009) Ethylene and other stimuli affect expression of the UDP glucose-flavonoid 3-O-glucosyltransferase in a non-climacteric fruit. Vitis 48:11?16Chervin C, El-Kereamy A, Roustan J.P (2004) Ethylene seems required for the berry development and ripening in grape, a non-climacteric fruit. Plant Sci. 167: 1301-1305Cornish, K. and Zeevaart, J.A.D. (1984) Abscisic acid metabolism in relation to water stress and leaf age in Xanthium strumarium. Plant Physiol. 76,1029?1035C.J.Brady (1987) Fruit ripening. Plant Physiol. 38:155-178Cleland, RS (1995) Auxin and Cell Elongation. In: P.J. Davies (Ed.) Plant Hormones: Physiology, Biochemistry and Molecular Biology, Kluwer Academic Publishers, Dordrecht, Netherlands, pp. 214?227Coombe BG (1972) The hormone content of ripening grape berries and the effects of growth substances treatments. Plant Physiol. 51: 629-63470Davies PJ (2004) The plant hormones: their nature, occurance, and functions. In: Davies PJ (ed) plant hormones. Kluwer, Dordrecht pp 1?15Esposito D, Garvey L.A., Chakiath C.S.(2009) Gateway cloning for protein expression. Methods Mol. Biol. 498: 31?54F.B Abeles, P.W Morgan, M.E Saltveit Jr. (1992) Ethylene in Plant Biology, Second EditionAcademic Press, Inc, New YorkFord CM et al. (1998) Cloning and characterization of Vitis vinifera UDP?glucose: flavonoid 3-O-glucosyltransferase, a homologue of the enzyme encoded by the maize bronze-1 locus that may primarily serve to glycosylate anthocyanidins in vivo J. Biol. Chem., 273, 9224?923rGiribaldi M, Hartung W, Schubert A (2009) The effects of abscisic acid on grape berry ripening are affected by the timing of treatment. J. Int. Sci. Vigne Vin 43:1?7G?khan S (1998) Effect of ethylene on climacteric and non-climacteric hortic?Ltural crops and possibilities of using ethylene antagonists in postharvest physiology. The Association of Food Technology 23(5):329-337Goldschmidt EE (1997) Ripening of citrus and other non-climacteric fruits: a role for ethylene. Acta Hort. 463: 335-340Gray J, Picton S, Shabbeer J, Schuch W, Grierson D. (1992) Molecular biology of fruit ripening and its manipulation with antisense genes. Plant Mol. Biol. 19(1): 69?87Grierson, D., Kader, A. (1987) Fruit ripening and quality. in: The Tomato Crop, Springer, pp. 241-280Hall D, De Luca V(2007) Mesocarp localization of a bi-functional resveratrol/ hydroxycinnamic acid glucosyltransferase of Concord grape (Vitis labrusca) Plant J., 49: 579?591Itai A, Fujita N (2008) Identification of climacteric and non-climacteric phenotypes of Asian pear cultivars by CAPS analysis of ACC synthase genes. HortScience 43:119?121Janvary L, Hoffmann T, Pfeiffer J, Hausmann L, Topfer R, Fischer TC, Schwab W (2009) A double mutation in the anthocyanin 5-O-glucosyltransferase gene disrupts enzymatic activity in Vitis vinifera L. J Agric. Food Chem. 57(9) : 3512-3518J.C.Pech, M. Bouzayan (2008) Climacteric fruit ripening: Ethylene-dependent and 71independent regulation of ripening pathways in melon fruit. Plant Sci. 175: 114-120Julie G, Steve P, Junaid S, Wolfgang S and Don G(1992) Molecular biology of fruit ripening and its manipulation with antisense genes. Plant Mol. Biol. 19: 69-87Lairson LL, Henrissat B, Davies GJ, Withers SG (2008) Glycosyltransferases: structures, functions, and mechanisms. Annu. Rev. Biochem., 77:521-555Li Y,Sandie B, Lim EK, and Bowles DJ (2001) Phylogenetic analysis of the UDP-glycosyltransferase multigene family of Arabidopsis thaliana. J. Biol. Chem. 276: 4338-4343Lehmann H. (1983) Time course of the biotransformation of abscisic acid and transabscisic acid conjugates in cell suspension cultures of Lycopersicon peruvianum. Biochem.Physiol Pflanzen. 178: 21-27Lelievre, J. M., Latche, A., Jones, B., Bouzayen, M. and Pech, J.-C. (1997) Ethyleneand fuit ripening. Plant Physiol. 101: 727-739Lim EK, Baldauf S, Li Y, Elias L, Worrall D, Spencer SP, Jackson RG, Taguchi G, Ross J, Bowles DJ (2003) Evolution of substrate recognition across a m?Ltigene family of glycosyltransferases inArabidopsis. Glycobiology 13: 139?145Lim EK, Doucet CJ, Hou B, Jackson RG, Abrams SR, Bowles DJ (2005) Resolution of (+) -abscisic acid using an Arabidopsis glycosyltransferase.Tetrahedron Asymmetry 16:143-47O?Donnell PJ, Truesdale MR, Calvert CM, Dorans A, Roberts MR, Bowles DJ (1998) A novel tomato gene that rapidly responds to wound- and pathogen-related signals. Plant J., 14:137-142LM Mathilde, CM.M Gachon, Saindrenan P (2005) Pathogen-responsive expression of glycosyltransferase genes UGT73B3 and UGT73B5 is necessary for resistance to pseudomonas syringae pv tomato in Arabidopsis. Plant Physiol. 139: 1890-1901Nambara E, Marion-Poll A (2005) Abscisic acid biosynthesis and catabolism. Annu. Rev. Plant Biol. 56:165?85Payasi A. and Sanwal G.G. (2008) Ripening of climacteric fruits and their control. J. Food Biochem. 34: 679-710Peat TS, Bottcher C, Newman J, Lucent D, Cowieson N, Davies C (2012) Crystal 72structure of an indole-3-acetic acid amido synthetase from grapevine involved in auxin homeostasis. Plant Cell 24, 4525?4538Pedro MC, Emeline D, Gideon JD, Bernard H (2003) An evolving hierarchical family classification for glycosyltransferases. J. Mol. Biol. 328: 307-317Petit P, Granier T, Langlois d'Estaintot B, Manigand C, Bathany K, Schmitter J.M., Lauvergeat V, Hamdi S, Gallois B(2007). Crystal structure of grape dihydroflavonol 4-reductase, a key enzyme in flavonoid biosynthesis J. Mol. Biol., 368, pp. 1345?1357Perin C, Gomez-Jimenez MC, Hagen L, Dogimont C (2002), Molecular and genetic characterization of a non-climacteric phenotype in melon reveals two loci conferring altered ethylene response in fruit. Plant Physiol. 129: 300-309Piotrowska A, Bajguz A (2011) Conjugates of abscisic acid, brassinosteroids, ethylene, gibberellins, and jasmonates. Phytochem. 72: 2097?2112Priest, D.M., Jackson, R.G., Ashford, D.A., Abrams, S.R. and Bowles, D.J. (2005) The use of abscisic acid analogues to analyse the substrate selectivity of UGT71B6, a UDP-glycosyltransferase of Arabidopsis thaliana. FEBS Lett. 579: 4454?4558Porebski S, Grant BL, Baum BR.(1997) Modification of a CTAB DNA extraction protocol for plants containing high polysaccharide and polyphenol components. Plant Mol. Biol., 15: 8-15Priest D.M., S.J. Ambrose (2006) Use of the glucosyltransferase UGT71B6 to disturb abscisic acid homeostasis in Arabidopsis thaliana. Plant J., 46, pp. 492?502Roberts MR, Warner SAJ, Darby R, Lim EK, Draper J, Bowles DJ (1999) Differential reg?Lation of a glucosyl transferase gene homologue during defence responses in tobacco. J. Exp. Bot. 50:407-410Ross J, Li Y, Lim EK, Bowles DJ (2001) Higher plant glycosyltransferases Genome Biol., 2: 3004Schwartz SH, Zeevaart JAD(2004). Abscisic acid biosynthesis and metabolism. In Plant Hormones, Biosynthesis, Signal Transduction, Action, ed. PJ Davies, pp. 137?55. Dordrecht, Netherlands: Kluwer73Scopes R.K. (2009) Protein Purification: Principles and Practice (3rd ed.), Springer, New York (1994). p.380-389Setha S. (2012). Roles of abscisic acid in fruit ripening. Walailak J. Sci. Technol. 9 297?308Seymour GB, Taylor JE and Tucker GA (1993) Biochemistry of Fruit Ripening. London: Chapman and HallTrainotti L, Tadiello A, Casadoro G (2007) The involvement of auxin in the ripening of climacteric fruits comes of age: the hormone plays a role of its own and has an intense interplay with ethylene in ripening peaches. J. Exp. Bot. 58:3299-3308Villarreal NM, Bustamante CA, Civello P M (2010) Effect of ethylene and 1-MCP treatments on strawberry fruit ripening. J. Sci. Food Agric. 90(4): 683-689Walton DC, Li Y (1995) Abscisic acid biosynthesis and metabolism. In Plant Hormones, ed. PJ Davis, pp. 140Yang S.F, Hoffman N.E (1984) Ethylene biosynthesis and its regulation in higher plant. Plant Physiol. 35: 155-189Zeevaart, J.A.D. Creelman, RA (1988) Metabolism and physiology of abscisic acid. Annu. Rev. Plant Physiol. Plant Mol. Biol. 39: 439?473Zenoni S, Ferrarini A, Giacomelli E, Xumerle L, Fasoli M, Malerba G, Bellin D, Pezzotti M, Delledonne M (2010) Characterization of transcriptional complexity during berry development in Vitis vinifera using RNA-Seq. Plant Physiol. 152: 1787?1795Zhang DP, Chen SW, Peng YB, Shen YY (2001) Abscisic acid-specific binding sites in the flesh of developing apple fruit. J. Exp. Bot. 52:2097 Zhou R, Cutler AJ, Ambrose SJ, Galka MM, Nelson KM (2004) A new abscisic acidcatabolic pathway. Plant Physiol. 134:361