Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Plant nutrition wikipedia , lookup
Silencer (genetics) wikipedia , lookup
Expression vector wikipedia , lookup
Genomic imprinting wikipedia , lookup
Ridge (biology) wikipedia , lookup
Endogenous retrovirus wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
Genetic engineering wikipedia , lookup
Gene regulatory network wikipedia , lookup
Genetically modified organism wikipedia , lookup
Plant Cell Physiol. 39(11): 1119-1126 (1998) JSPP © 1998 Mini Review Metabolic Engineering to Modify Flower Color Yoshikazu Tanaka, Shinzo Tsuda and Takaaki Kusumi Institute for Fundamental Research, Suntory Ltd., 1-1-1 Wakayamadai, Shimamoto-cho, Mishima-gun, Osaka, 618-8503 Japan Thanks to the rapid progress in molecular biology of flavonoid biosynthesis and plant transformation, it has become feasible to modify the pathway and flower color through genetic engineering. One of the advantages of molecular breeding is that flower color can be specifically modified without changing the other characteristics of the targeted variety. Novel flower color varieties such as brickred petunias and violet carnations have been successfully made by expression of heterologous flavonoid genes. Flavonoid metabolic engineering has and will give new perspectives in plant molecular biology besides its industrial application. Key words: Anthocyanin — Flavonoid — Flower color — Metabolic engineering — Transgenic plants. Flower colors are mainly produced by flavonoids, carotenoids and betalains. Carotenoids are largely responsible for the production of yellow and orange flowers such as sun flower and tomato (Bartley and Scolnik 1995). Betalains are yellow to red nitrogenous compounds derived from tyrosine and are distributed only in Caryophyllales (Stafford 1994). Flavonoids have a wide range of colors from pale yellow to red, purple and blue (Goto 1987, Goto and Kondo 1991). The flavonoid biosynthetic pathway has been the choice to genetically engineer flower color to obtain novel colors. Such engineering has also helped us to understand transgene expression in plants. This review summarizes the achievements made in engineering flavonoid biosynthesis to modify flower color. The field has been reviewed several times (Mol et al. 1995, Davies and Schwinn 1997, Tanaka et al. 1998). Flavonoids and flower color—Anthocyanins are a colored class of flavonoids and accumulate in vacuoles. There are six major anthocyanidins (chromophores of anthocyanins); pelargonidin, cyanidin, peonidin (3' O-methyl cyanidin), delphinidin, petunidin (3' O-methy delphinidin) and malvidin (3',5' O-dimethy delphinidin). As the number of hydroxyl groups in the B-ring is increased, the visible absorption maximum is shifted longer. The maxima in 0.01% Abbreviations: CaMV, cauliflower cultivar. mosaic virus; cv., HCl-methanol solution of pelargonidin, cyanidin and delphinidin are 520, 535 and 546 nm, respectively (Goto 1987). Modification of the anthocyanidins mainly with glycosylation and acylation results in a wide variety of anthocyanins. Hundreds of anthocyanins have been purified and their structures determined (Strack and Wray 1994). Besides the structures of anthocyanins, changes of vacuolar pH, intermolecular stackings (self-association of anthocyanins and co-pigmentation of anthocyanins with polyphenols), intramolecular stacking of aromatically modified anthocyanins, metal complexation and cell shapes give almost infinite flower colors (Goto 1987, Goto and Kondo 1991, Brouillard and Dangles 1994, Mol et al. 1998). Anthocyanins are red and stable at a low pH. They are bluer but often unstable at a vacuolar pH which is usually weakly acidic. The stackings and metal complexation contribute to stabilize anthocyanins and flower color, especially blue color (Goto 1987, Goto and Kondo 1991, Brouillard and Dangles 1994). The major role of anthocyanin production in petals is to attract pollinators. Flavonoids and anthocyanins also play important roles in photoprotection, reproduction, pathogenesis and symbiosis (Shirley 1996). Flavonoid biosynthesis—The flavonoid biosynthetic pathway shown in figure 1 has been extensively reviewed (Heller and Forkmann 1994, Forkmann 1994, Holton and Cornish 1995). The pathway leading to anthocyanidin 3glucosides is generally conserved among plant species and is well understood. Anthocyanidin 3-glucosides are further modified with sugars and aliphatic and aromatic acids. There are both species and variety differences in the extent of modification and the types of glycosyl and acyl groups attached (Fig. 1). The first enzyme committed to flavonoid production is chalcone synthase (CHS). This enzyme catalyzes the stepwise condensation of three acetate units starting from malonyl-CoA with p-coumaroyl-CoA to yield 4,2',4',6tetrahydroxychalcone. CHS shares homology with some condensing enzymes such as arabidopsis FA El and /?ketoacyl-acyl carrier protein synthase (James et al. 1995). Flavanone 3-hydroxylase (F3H), a kind of 2-oxoglutarate dependent dioxygenases, catalyzes hydroxylation of naringenin to dihydrokaempferol (DHK). DHK is hydroxylated to dihydroquercetin (DHQ) and to dihydromyricetin (DHM) by flavonoid 3-hydroxylase (F3'H) and flavonoid 1119 Metabolic engineering to modify flower color 1120 COOH COS-CoA PAL OH OH " • OH Phenylalanine Cinnamic acid 4-Coumaric acid 4-Coumaroyl-CoA 3 x Malonyj^CoA 4-Caffeoyl-CoA JCHS OH 0 4,2',4',6'-Tetrahydroxychalcone 1 CHI FNS OH OH O FLS Kaempferol o Naringenin F3H OH OH O Dihydroauercetin OH OH O OH b Dihydromyricetin Dihydrokaempferol I D F R , ANS, 3GT " Hl I DFR, ANS, 3GT DFR, ANS, 3GT PH -OH OH OGIc OGIc OH Delphinidin 3-Glucoside OH Cyanidin 3-Glucoside OGIc OGIc-Malonate OH A chrysanthemum anthocyanin R2 OGIc Rose anthocyanins OH R1 = H or OH or OCH3 R2 = OH or OGIc OGIc-Caf OCH3 A gentian anthocyanin (gentiodelphin) I OH HO. OCH3 Carnation anthocyanins R1 = H or OH R2 = OH or OGIc OGIc A petunia anthocyanin Fig. 1 Generalized pathway leading to anthocyanidin 3-glucosides (Holton and Cornish 1995). Anthocyanins of non-blue flowers (rose, chrysanthemum and carnation) and blue flowers (gentian and petunia) are also shown. Blue flowers usually have delphinidin modified with one or more aromatic acyl group(s) such as caffeic acid (Caf) and coumaric acid. PAL, phenylalanine ammonia-lyase; C4H, cinnamate4-hydroxylase; 4CL, 4-coumarate:CoA ligase; CHS, chalcone synthase; CHI, chalcone isomerase; F3H, flavanone 3-hydroxylase; Fill, flavonoid-3'-hydroxylase; F3'5"H, flavonoid 3\5'-hydroxylase; DFR, dihydroflavonol 4-reductase; ANS, anthocyanidin synthase; 3GT, flavonoid 3-glucosyl transferase; CC3H, 4-coumaroyl-CoA 3-hydroxylase; FNS, flavone synthase; FLS, flavonol synthase; Glc, glucose; Rha, rhamnose. Metabolic engineering to modify flower color 3',5'-hydroxylase (F3'5'H), respectively (Fig. 1). These two hydroxylases belong to the same cytochrome P-450 family (CYP 75, Holton et al. 1993a, Brugliera et al. 1997). Both are key enzymes in determining flower color because they determine the hydroxylation patterns of the B-ring of the dihydroflavonols and this hydroxylation eventually determines the structure of the anthocyanidin leading to flower color (Holton and Tanaka 1994). F3'5"H can also convert DHQ to DHM. Dihydroflavonol 4-reductase (DFR) catalyzes the reduction of dihydroflavonols to leucoanthocyanidins and is proposed to belong to the 3/?-hydroxysteroid dehydrogenase/DFR superfamily (Baker and Blasco 1992). Leucoanthocyanidins are converted to anthocyanidins by another 2-oxoglutarate-dependent dioxygenase, anthocyanidin synthase (ANS), and then to the anthocyanidin 3-glucosides by UDP-glucose: flavonoid 3-O-glucosyltransferase (3GT). UDP-glucose: anthocyanin 5-O-glucosyltransferase (5GT) genes have been recently cloned (Yamazaki et al. 1998). Both 3GT and 5GT belong to the distinct families of glycosyltransferase superfamily. Petunia UDP-rhamnose: anthocyanin rhamnosyltransferase also has homology to glycosyltransferases (Brugliera et al. 1994, Kroon et al. 1994). Genes of aromatic acyl transferase (AAT) catalyzing the transfer of aromatic group to 3- or 5-glucose of anthocyanins have been cloned (Tanaka et al. 1997). Many of these enzymes have been biochemically characterized by expressing cloned cDNAs in heterologous systems (Holton et al. 1993b, Tanaka et al. 1996, 1997, Yamazaki et al. 1998). Peonidin-type anthocyanins are synthesized from the cyanidin-type anthocyanins and petunidin- and malvidin-type anthocyanins are from the delphinidin-type anthocyanins by the transfer of a methyl group from S-adenosylmethionine catalyzed by anthocyanin methyltransferases (Heller and Forkmann 1994). Anthocyanins are reversibly modified with glutathione (Marrs et al. 1995, Alfenito et al. 1998) and postulated to be transported into vacuoles by ATP-binding cassette (ABC) transporters (Lu et al. 1997). A 24 kDa vacuolar protein has been suggested to be involved in intravacuolar pigmented structures in sweet potato (Nozue et al. 1997). These processes are not clearly understood. Flavonols are derived from dihydroflavonols by a 2-oxoglutarate-dependent dioxygenase, flavonol synthase (FLS, Holton et al. 1993b). Apigenin (a flavone) is synthesized from naringenin by flavone synthase (FNS). Interestingly there are two kinds of FNS, a cytochrome P-450 and a dioxygenase (Heller and Forkmann 1994). Regulation of flavonoid biosynthetic pathway—Spatial and/or temporal expressions of structural genes or enzymes of flavonoid biosynthesis in flowers, leaves and seedlings have been well studied in many plants such as petunia (Brugliera et al. 1994), snapdragon (Jackson et al. 1992), gerbera (Helariutta et al. 1993), carnation (Stich et al. 1992), rose (Tanaka et al. 1995), eggplant (Toguri et al. 1121 1993), Arabidopsis (Pelletier et al. 1997), grape (Boss et al. 1996), perilla (Gong et al. 1997) and lisianthus (Nielsen and Podivinsky 1997). Transcription of these genes is spatially and developmentally regulated in a well-coordinated way paralleling flavonoid synthesis. In maize, two gene families, R (Myc type) and Cl (Myb type) encoding transcriptional factors, regulate the expression of the structural genes in the pathway. Snapdragon and petunia have similar transcriptional factors that regulate the structural genes (Holton and Cornish 1995, Mol et al. 1998). Regulatory genes controlling anthocyanin biosynthesis are functionally conserved among plant species and some of them function in heterologous species. They have distinct sets of target genes (Quattroccio et al. 1993, Martin 1996), which allows regulatory diversity in the pathway depending on species. In petunia flowers, Anl (a Myc gene), An2 (a Myb gene) and Anil are regulatory genes of anthocyanin biosynthesis in the flower. Anil encodes a novel Trp-Asp repeat protein and is a component of a signal transduction cascade that modulates An2 function (de Vetten et al. 1997). Molecular tools for flower color modification—The first step in metabolic engineering is to clone the structural genes. Cloning structural genes in the flavonoid pathway is not difficult because all but a few structural genes in the pathway (Fig. 1) have been cloned. The homologous genes can be isolated by the hybridization approach at least from closely related species. This approach is very effective for conserved genes such as CHS and is less effective for variable genes such as CHI (Gutterson 1995). In our experience, most structural genes can be readily cloned by low stringent screening with heterologous counterpart genes within dicots. For less conserved genes such as CHI and sometimes DFR, a polymerase chain reaction approach using conserved amino acid sequences has been successfully used (Gutterson 1995, Tanaka et al. 1996). Cloned genes in the pathway from various flowers are listed (Davies and Schwinn 1997). Constitutive promoters including cawliflower mosaic virus (CaMV) 35S promoter, an enhanced CaMV 35S promoter (Mistuhara et al. 1996) and Mac promoter (Comai et al. 1990) have been successfully used to express genes in petals. Constitutive expression of the structural genes does not have deleterious effect on the transgenic plant. Promoters derived from genes in the flavonoid pathway should be useful for temporal and spatial expression. Suppression of a gene is more challenging than expression of a gene in transgenic plants. Plant gene expression can be suppressed by introducing either an antisense (Mol et al. 1990) or sense gene of interest or its close homologs. The mechanism of sense suppression (cosuppression) is not yet fully understood (Taylor 1997, Gallie 1998) although the phenomenon is a post-transcriptional event. Doublestranded RNA is proposed to be a mediator in sequence- 1122 Metabolic engineering to modify flower color specific genetic silencing and cosuppression (Montgomery and Fire 1998). The frequency and degree of cosuppression depend on transgene promoter strength and an intact full length cDNA may be preferable for efficient suppression (Que et al. 1997) although it is not essential to use full length cDNA to obtain phenotypes of antisense or sense suppression (Gutterson 1995). The frequency and degree of antisense and sense suppression also depend on the targeted gene and species. The frequency varies from less than 1% to 40% in our laboratory. Transformation of floricultural crops—Plant cells have can detect the transgenes in the genome and suppress the expression unless the transgenes are integrated in "correct positions" (Kumpatla et al. 1998, Matzke and Matzke 1998, Gallie 1998). It is, however, not possible to integrate a transgene into the correct position artificially. It is necessary to obtain many (dozens to hundreds) transgenic plants and select a few transgenic lines whose phenotypes are desirable and stable enough to be commercialized. Therfore, a highly efficient transformation system for the targeted species or variety is essential. Interestingly, a high frequency of shutdown of transgene expression during growth is often observed in chrysanthemum (Deroles et al. 1997). Transformation of floricultural crops has been reviewed (Deroles et al. 1997, Tanaka et al. 1998). Rose, carnation, tulip and chrysanthemum that occupy more than 50% of the cut flower market are claimed to be routinely transformed (Mol et al. 1995) but there is much variation in extent of success among cultivars within a species. Problems and strategies of plant transformation have already been discussed (Birch 1997). Plant transformation generally consists of three steps. The first step is to transfer foreign DNA into plant cells (infection), the second step is to select transgenic cells (selection) and the third step is to regenerate them into complete plants (regeneration). Each step needs careful optimization of various procedures and parameters. To deliver foreign DNA into plant cells, Agrobacte/•/wm-mediated gene transfer has been widely used especially for dicots and some monocots. Particle bombardment is often used for monocots resistant to Agrobacterium (Deroles et al. 1997). For selection, the neomycin phosophotransferase gene that confers kanamycin resistance has been most widely and successfully used as a selection marker. Herbicide-resistant genes encoding enzymes which are insensitive to the herbicide or detoxify it can be used as selectable markers. It is important to select only the cells with the introduced transgenes. Weak selection often results in chimerical problems. Most dicot species regenerate well via organogenesis on media containing balanced hormone supplementation, usually cytokinin and auxin. An efficient transformation system has been established for petunias and torenias and 50 to 100 independent transgenic plants per gene construct are routinely obtained in our laboratory. Embryogenesis is the preferred route to transform roses in our laboratory (Fig. 2A, Katsumoto et al. 1995). A rose cultivar (cv.) Royalty was also transformed via embryogenic callus (Firoozababy et al. 1994). Modification of anthocyanin amount—White is not a novel color because of abundant occurrence. Still, molecular breeding of a white variety is industrially useful because only flower color can be modified without sacrificing the other desirable characteristics. Such a sacrifice is often made during hybridization breeding. It is possible to obtain white flowers from anthocyanin producing flowers by suppressing the expression of one of many structural or regulatory genes in the pathway. The reported results are summarized in Table 1. It is interesting that novel color patterns were obtained in petunia and lisianthus while only uniform suppression was observed in rose, gerbera, and chrysanthemum. This may be relevant to the fact that petunia and lisianthus have natural white-sectored patterns (Deroles et al. 1998). Suntory Ltd. markets Surfinia™ (a petunia) and Summerwave™ (a torenia) that are superior to conventional petunias and torenias with their creeping and vigorous characters. We have been working on widening their flower color variation by metabolic engineering. Antisense CHS-A in a petunia cv. Surfinia Mini Purple resulted in a few plants with pale color or pure white flowers (Katsumoto et al. unpublished results, Figure 2B). White and blue/white (two out of four petals are white and the other two petals are blue) transgenic torenia plants were successfully generated from a blue torenia cv. Summerwave Blue by sense suppression of CHS or DFR genes (Suzuki et al. 1997). In both cases the transgenic plants retain the superior characteristics of the host and only flower color was modified. The amount of anthocyanins was increased using the transcriptional factors of the pathway. The flower color of the tobacco was changed from pink to intense red by expressing the maize R gene (Lc allele) with CaMV 35S promoter. Cl driven by the same promoter had no effect. Arabidopsis root, petal and stamen accumulate anthocyanins by constitutive expression of both Cl and R (Lloyd et al. (1992)). Transgenic petunia plants that had a higher anthocyanin content in floral and vegetative tissues were obtained by constitutive expression of Lc. Leaves of the transgenic petunia are purple due to accumulation of anthocyanins (Bradley et al. 1998). Delila (an R homolog controlling flavonoid synthesis in snapdragon) increased anthocyanins in tobacco flower and tomato flowers and vegetative tissues (Mooney et al. 1995). Because dihydroflavonols are the common precursors of anthocyanins and colorless flavonols, DFR and FLS compete with each other for dihydroflavonols. Antisense Metabolic engineering to modify flower color 1123 B Fig. 2 A. Embryogenic rose calli under kanamycin selection. White parts will regenerate into plantlets and brown parts are dead cells. B. Transgenic plants with white or pale colored flower were obtained from a petunia cv. Surfinia™ Mini Purple (up) with antisense suppression of chalcone synthase gene. C. Suppression of flavonol synthase gene in a petunia cv. Surfinia Pink (right) resulted in a redder flower (left) that hardly contained flavonols. D. A pale pink petunia accumulating dihydrokaempferol (right) was transformed with rose dihydroflavonol 4-reductase cDNA and a novel flower color (left) derived from pelargonidin that petunias hardly produce was obtained. E. Flowers of a torenia cv. Summerwave™ Blue (right) and its sport cv. Pinlc and White (right). Pink and White is deficient in flavonoid 3'5'-hydroxylase and aromatic acyl transferase. F. Transgenic violet carnations (left). Moondust™ (paler ones) and Moonshadow™ (darker ones). They produce delphinidin type anthocyanins in their petals and have a novel blue hue. A transgenic carnation plant in tissue culture (right). Efficient carnation transformation has been reported previously (Lu et al. 1991). 1124 Metabolic engineering to modify flower color Table 1 Suppression of flavonoid pathway Host species (colour) Petunia (purple) Petunia (violet) Petunia (purple) Chrysanthemum (pink) Gerbera (red) Rose (red) Carnation (pink) Torenia (blue) Torenia (blue) Carnation (red) Lisianthus (purple) Gene construct Antisense CHS-A Sense CHS-A Sense CHS-A Sense CHS Antisense CHS, DFR Sense CHS Sense CHS Sense CHS, DFR Sense CHS, DFR Antisense CHS, DFR Antisense F3H Antisense CHS Phenotype Reference White, pattern White, pattern White, pattern White, uniform Pink, uniform Pink, uniform Pale pink, uniform White, pattern Pale blue, pattern van der Krol et al. (1988) Napoli et al. (1990) van der Krol et al. (1990) Courtney-Gutterson et al. (1994) Elomaa et al. (1993) Gutterson (1995) Gutterson (1995) Suzuki et al. (1997) Aida et al. (1997) White, uniform White, pattern Zucker et al. (1998) Deroles et al. (1998) Phenotypes of transgenic plants depend on each transgenic event. The strongest phenotype of each report is shown. Some species showed uniform suppression and some formed novel color patterns. suppression of the FLS gene in petunia and tobacco resulted in a higher content of anthocyanin and more intense flower color (Holton et al. 1993b). A similar result was obtained with sense suppression of FLS gene in a petunia cv. Surfinia Pink. Flower color changed from pink to red purple and the transgenic plant flower contained more anthocyanins and less flavonol than the host (Fig. 2C. Suzuki et al., unpublished results). Making brick-red flowers—Petunia flowers rarely contain pelargonidin-type anthocyanin and are not brick-red because DHK can not be reduced by petunia DFR that has high activity for DHM (Forkmann and Ruhau 1987) and is suiable for delphinidin production. Transgenic brickred petunias accumulating pelargonidin-type anthocyanins have been obtained from DHK accumulating petunia (deficient in F3'5H, F3"H and FLS) with using maize, gerbera and rose DFR cDNAs (Meyer et al. 1987, Helariutta et al. 1993, Tanaka et al. 1995 (Fig. 2D), respectively). Stability of the phenotype and the transgene of the petunias containing maize DFR cDNA has been extensively studied (Linn et al. 1990. Meyer and Heidmann 1994). The gerbera DFR gene showed a stronger and more consistent expression than maize DFR. The choice of the gene source is a factor to be considered for engineering plants even if the gene encodes the same enzyme activity (Elomaa et al. 1995). When the rose DFR gene was introduced to a petunia cv. Surfinia Purple that did not accumulate DHK due to its dominance in F3"H and FS'S'H genes was transformed, no flower color changes were observed (unpublished results). This indicates that it is difficult for a transgene to compete with endogenous genes. Molecular breeding of blue flowers—Structures of anthocyanin, presence of co-pigments and relatively high vacuolar pH (more than about 5.0, Holton and Tanaka 1994) are key factors for flowers to have a bluish hue (Goto and Kondo 1991, Holton and Tanaka 1994, Mol et al. 1998). Most blue flowers contain aromatically acylated delphinidin derivatives (Fig. 1). The absorbance maximum of anthocyanin shifts towards longer wavelengths (bluer color) by about 10 nm with one hydroxylation of B-ring, and by about 4 nm with one aromatic acylation in \% trifluoroacetic acid solution (Fujiwara et al. 1997). Figure 2E shows the effect of hydroxylation of the B-ring and aromatic acylation of anthocyanins to flower color. Blue petals of a torenia cv. Summerwave Blue mainly contain malvidin 5-(p-coumaroyl)-glucoside 3-glucoside while petals of its pink counterpart contain peonidin 3,5-diglucoside. This indicates that F3'5'H and AAT activities are necessary for the production of blue flowers in vivo. There is little or no change in color of anthocyanins with glycosylation and aliphatic acylation. Rose, chrysanthemum and carnation only have pelargonidin- and cyanidin-type anthocyanins not modified with aromatic acyl groups. Genes encoding F3'5'H have been isolated from petunia (Holton et al. 1993a), eggplant (Toguri et al. 1993), gentian (Tanaka et al. 1996), lisianthus (Nielsen and Podivinsky 1997) and many others (unpublished results). Petunia F3'5'H cDNAs could complement their deficiency in petunia that was also deficient in F3'H and low in pH (see below). The flower color changed from pale pink to reddish purple (Holton et al. 1993a). Florigene Ltd, (Australia) and Suntory Ltd. successfully developed transgenic violet carnations by introduction of a petunia F3'5'H and DFR genes into a DFR-deficient white carnation (Holton et al., unpublished results). The petals of the carnations predominantly contain delphinidin, which is not produced by the native carnations. Such bluish flowers have not been produced by traditional Metabolic engineering to modify flower color breeding of carnation (Fig. 2F). The transgenic violet carnations named Moondust™ have been marketed in Australia and Japan and became the first transgenic floricultural crop to be sold. Darker versions, Moonshadow™, have been obtained and will be marketed soon. Application of AAT genes (Tanaka et al. 1997) to modify flower color is interesting. When anthocyanins make a complex with polyphenols such as flavonol and flavone glycosides (co-pigmentation), their absorption is increased by up to about 35 nm in wavelength (bathocromic shift, Goto 1987). Co-pigmentation can also result in a large increase in absorptivity. The amount of flavonols can be modified by genetic engineering of FLS genes (Holton et al. 1993b). Although intramolecular stacking of anthocyanins may be achieved by expression of multiple AAT and GT genes in heterologous plants, only some of the genes have been cloned. Anthocyanins are bluer at a higher pH (Goto 1987, Goto and Kondo 1991). Even an peonidin derivative can make blue flowers as in blue petals of morning glory when it is highly modified with six glucose and three caffeic acid molecules and its vacuolar pH is high, 7.7 (Yoshida et al. 1995). Because non-blue flowers, however, tend to have a lower vacuolar pH, it is important to increase the pH to produce blue flowers. Petunia is known to have seven loci controlling vacuolar pH (phl-phl, Mol et al. 1998). When one of them is homozygous recessive, the vacuolar pH increases in the petal. Only Ph6 has been cloned (Chuck et al. 1993). Anl and Ph6 turned out to be alleles of the same locus (Mol et al. 1998). Mutations in An2 and Anil also affect pH (Mol et al. 1998). Cloning of Ph genes and their functional analysis may be useful to elevate flower vacuolar pH (Mol et al. 1998). However, it is not certain if pH regulation by petunia Ph genes is applicable to other species. Further biochemical studies on the mechanism of pH control in vacuoles or bluing phenomena of senescent flowers often observed in some species may help us to engineer vacuolar pH. The three dimensional structure of Commelina blue pigment has been determined. The complex is made of six anthocyanins (malonylawobanin), six flavones (flavocommelin) and two magnesium ions (Kondo et al. 1992). The biochemical process of formation of the complex is yet to be revealed. A specific co-pigment and a metal ion are necessary for a specific anthocyanin to form a stable metalloanthocyanin. Tulip, impatiens, cyclamen and pelargonium produce delphinidin but lack true blue colors (Davies and Schwinn 1997). A species-specific strategy is required to obtain blue varieties. Many genes must be introduced and properly expressed to synthesize highly modified anthocyanins or form metal complexes and/or elevate vacuolar pH in order to make truly blue flowers from non-blue flowers. This is a 1125 very challenging task at the moment and substantial improvement of plant biotechnology is essential. Further chemical and biological understanding of flower color and flavonoid pathway together with improved plant transformation systems should make modification of flower color more feasible. The authors thank Florigene Ltd. for providing the photo of Figure 2 and permission of presenting unpublished results. References Aida, R., Kishimoto, S., Tanaka, Y. and Shibata, M. (1997) J. Japan. Soc. Hort. Aci. 66 suppl. 2: 486-487. Alfenito, M.R., Souer, E., Goodman, C D . , Buell, R., Mol, J., Koes, R. and Walbot, V. (1998) Plant Cell 10: 1135-1149. Baker, M.E. and Blasco, R.E. (1992) FEBS Lett. 301: 89-93. Bartley, G.E. and Scolnik, P.A. (1995) Plant Cell 7: 1027-1038. Birch, R.G. (1997) Annu. Rev. Plant Physiol. Plant Mol. Biol. 48: 297326. Boss, P.K., Davies, C. and Robinson, S.P. (1996) Plant Mol. Biol. 21: 565-569. Bradley, J.M., Davies, K.M., Deroles, S.C., Bloor, S.J. and Lewis, D.H. (1998) Plant J. 13: 381-392. Brouillard, R. and. Dangles, 0. (1994) In The Flavonoids: Advances in Research Since 1986. Edited by Harborne, J.B. pp. 565-588. Chapman and Hall, London. Brugliera, F., Barri-Rewell, G., Holton, T.A. and Mason, J.G. (1997) In Abst. 5th Intl. Cong. Plant Mol. Biol. p. 118. Brugliera, F., Holton, T.A., Stevenson, T.W., Farcy, E., Lu, C.-Y. and Cornish, E. (1994) Plant J. 5: 81-92. Chuck, G., Robbins, T., Nijjar, C , Ralston, E. and Courtney-Gutterson, N. (1993) Plant Cell 5: 371-378. Comai, L., Moran, P. and Maslyar, D. (1990) Plant Mol. Biol. 15: 373381. Courtney-Gutterson, N., Napoli, C , Lemieux, C , Morgan, A., Firoozababy, E. and Robinson, K.E.P. (1994) Bio/Technology 12: 268271. Davies, K.M. and Scwinn, K.E. (1997) In Biotechnology of Ornamental Plants. Edited by Geneve, R.L., Preece, J.E. and Merkle, S.A. pp. 259294. CAB International, Wallingford. de Vetten, N., Quattrocchio, F., Mol. J. and Koes, R. (1997) Gene Dev. 11: 1422-1434. Deroles, S.C., Boase, M.R. and Konczak, I. (1997) In Biotechnology of Ornamental Plants. Edited by Geneve, R.L., Preece, J.E. and Merkle, S.A. pp. 87-120. CAB International, Wallingford. Deroles, S., Bradley, J.M., Schwinn, K.E., Markham, K.R., Bloor, S., Manson, D.G. and Dvies K.M. (1998) Mol. Breed. 4: 59-66. Elomaa, P., Helariutta, Y., Griesbach, R.J., Kotilainen, M., Seppanen, P. and Teeri, T.H. (1995) Mol. Gen. Genet. 248: 649-656. Elomaa, P., Honkanen, J., Puska, R., Seppanen, P., Helariutta, Y., Mehto, M., Kotilainen, M., Nevalainen, L. and Teeri, T.H. (1993) Bio/ Technology 11: 508-511. Firoozababy, E., Moy, Y., Courtney-Gutterson, N. and Robinson, K. (1994) Bio/Technology 12: 609-613. Forkmann, G. (1994) In The Flavonoids: Advances in Research Since 1986. Edited by Harborne, J.B. pp. 537-564. Chapman and Hall, London. Forkmann, G. and Rehnau, B. (1987) Z. Naturforsch. 42c: 1146-1148. Fujiwara, H., Tanaka, Y., Fukui, Y., Nakao, M., Ashikari, T. and Kusumi, T. (1997) fur. /. Biochem. 249: 45-51. Gong, Z., Yamazaki, M., Sugiyama, M., Tanaka, Y. and Saito, K. (1997) Plant. Mol. Biol. 35: 915-927. Goto, T. (1987) Prog. Chem. Org. Nat. Prod. 52: 114-158. Goto, T. and Kondo, T. (1991) Angrew. Chem. Int. Ed. Engl. 30: 17-33. Gallie, D.R. (1998) Curr. Opin. Plant Biol. 1: 166-172. Gutterson, N. (1995) Hort Sci. 30: 964-966. 1126 Metabolic engineering to modify flower color Helariutta, Y., Elomaa, P., Kotilainen, M., Seppanen, P. and Teeri, T.H. (1993) Plant Mol. Biol. 22: 183-193. Heller, W. and Forkmann, G. (1994) In The Flavonoids: Advances in Research Since 1986. Edited by Harborne, J.B. pp. 499-536. Chapman and Hall, London. Holton, T.A., Brugliera, F., Lester, D.R., Tanaka, Y., Hyland, C D . , Menting, J.G.T., Lu, C.-Y., Farcy, E., Stevenson, T.W. and Cornish, E.C. (1993a) Nature 366: 276-279. Holton, T.A., Brugliera, F. and Tanaka, Y. (1993b) Plant J. 4: 1003-1010. Holton, T.A. and Cornish, E.C. (1995) Plant Cell 7: 1071-1083. Holton, T.A. and Tanaka, Y. (1994) Trends Biotech. 12: 40-42. Jackson, D., Roberts, K. and Martin, C. (1992) Plant J. 2: 425-434. James, D.W., Lim, E., Keller, J., Plooy, I., Ralston, E. and Dooner, H.K. (1995) Plant Cell 7: 309-319. Katsumoto, Y., Tsuda, S., Nakamura, N., Kusumi, T. and Fukui, H. (1995) In Abst. 14th Plant Tissue Culture Conference Jpn. Assoc. Plant Tissue Cult. pp. 103. (in Japanese). Kondo, T., Yoshida, Y., Nakagawa, A., Kawai, T., Tamura, H. and Goto, T. (1992) Nature 358: 515-518. Kroon, J., Souer, E., de Graaff, A., Xue, Y., Mol, J. and Koes, R. (1994) Plant J. 5: 69-80. Kumpatla, S., Chandrasekharan, M.B., Iyer, L.M., Li, G. and Hall, T.C. (1998) Trends Plant Sci. 3: 97-104. Linn, F., Heidman, I., Saedler, H. and Meyer, P. (1990)Mol. Gen. Genet. 222: 329-336. Lloyd, A.M., Walbot, V. and Davis, R.W. (1992) Science258: 1773-1775. Lu, C , Nugent, G., Wardley-Richardson, T., Chandler, S.F., Young, R. and Dalling, M.J. (1991) Bio/Technology 9: 864-868. Lu, Y-P., Li, Z-S. and Rea, P.A. (1997) Proc. Natl. Acad. Sci. USA 94: 8243-8248. Marrs, K.A., Alfenito, M.R., Lloyd, A.M. and Walbot, V. (1995) Nature 375: 397-400. Martin, C. (1996) Curr. Opin. Biotech. 7: 130-138. Matzke, A.J.M. and Matzke, M.A. (1998) Curr. Opin. Plant Biol. 1: 142148. Meyer, P. and Heidmann, I. (1994) Mol. Gen. Genet. 243: 390-399. Meyer, P., Heidemann, I., Forkmann, G. and Saedler, H. (1987) Nature 330: 677-678. Mitsuhara, I., Ugaki, M., Hirochika, H., Ohshima, M., Murakami, T., Gotoh, Y., Katayose, Y., Nakamura, S., Honkura, R., Nishimiya, S., Ueno, K., Mochizuki, A., Tanimoto, H., Tsugawa, H., Otsuki, Y. and Ohashi, Y. (1996) Plant Cell Physiol. 37 49-59. Mol, J., Grotewold, E. and Koes, R. (1998) Trends Plant Sci. 3: 212-217. Mol, J.N.M., Holton, T.A. and Koes, R.E. (1995) Trends Biotech. 13: 350-355. Mol, J.N.M., van der Krol, A.R., van Tunen, A.J., van Blokland, R., de Lange, P. and Stuitje, A.R. (1990) FEBS Lett. 268: 427-430. Montgomery, M.K. and Fire, A. (1998) Trends Genet. 14: 255-258. Mooney, M., Desnos, T., Harrison, K., Jones, J,, Carpenter, R. and Coen, E. (1995) Plant J. 7: 333-339. Napoli, C , Lemieux, C. and Jorgensen, R. (1990) Plant Cell 2: 279-289. Nielsen, K.M. and Podivinsky, E. (1997) Plant Sci. 167-174. Nozue, M., Yamada, K., Nakamura, T., Kubo, H., Kondo, M. and Nishimura, M. (1997) Plant Physiol. 115: 1065-1072. Pelletier, M.K., Murrell, J.R. and Shirley, B.W. (1997) Plant Physiol. 113: 1437-1445. Quattroccio, F., Wing, J.F., Leppen, H.T.C., Mol, J.N.M. and Koes, R. (1993) Plant Cell 5: 1497-1512. Que, Q., Wang, H-Y., English, J.J. and Jorgensen, R.A. (1997) Plant Cell 9: 1257-1368. Shirley, B.W. (1996) Trends Plant Sci. 1: 377-381. Stafford, H.A. (1994) Plant Sci. 101: 91-98. Stich, K., Eidenberger, T., Wurst, F. and Forkmann, G. (1992) Z. Naturforsch. 47C: 553-560. Strack, D. and Wray, V. (1994) In The Flavonid: Advances in Research Since 1986. Edited by Harborne, J.B. pp. 1-22. Chapman and Hall, London. Suzuki, K., Zue, H., Tanaka, Y., Fukui, Y., Mizutani, M. and Kusumi, T. (1997) Plant Cell Physiol. 38: s38. Tanaka, Y., Fukuchi-Mizutani, M., Sakakibara, K., Fujiwara, H., Fukui, Y., Nakao, M., Ashikari, T. and Kusumi, T. (1997) In Absrtact 5th Intl. Cong. Plant Mol. Biol. p. 736. Tanaka, Y., Fukui, Y., Fukuchi-Mizutani, M., Holton, T.A., Higgins, E. and Kusumi, T. (1995) Plant Cell Physiol. 36: 1023-1031. Tanaka, Y., Tsuda, S. and Kusumi, T. (1998) In Applied Plant Biotechnology. Edited by Chopra, V.L. pp. 177-231. Oxford & IBH, New Delhi. Tanaka, Y., Yonekura, K., Fukuchi-Mizutani, M., Fukui, Y., Fujiwara, H., Ashikari, T. and Kusumi, T. (1996) Plant Cell Physiol. 37: 711-716. Taylor, C.B. (1997) Plant Cell 9: 1245-1249. Toguri, T., Umemoto, N., Kobayashi, O. and Ohtani, T. (1993) Plant Mol. Biol. 23: 933-946. van der Krol, A.R., Lenting, P.E., Veenstra, J., van der Meer, I.M., Koes, R.E., Gerats, A.G.M., Mol, J.N.M. and Stuitje, A.R. (1988) Nature 333: 866-869. van der Krol, A.R., Mur, L.A., Beld, M., Mol, J.N.M. and Stuitje, A.R. (1990) Plant Cell 2: 291-299. Yamazaki, M., Gong, Z., Mizutani, M., Fujiwara, H., Tanaka, Y., Kusumi, T. and Saito, K. (1998) In Abst. Japanese Society for Plant Cell and Molecular Biology. Yoshida, K., Kondo, T., Okazaki, Y. and Katou, K. (1995) Nature 373: 291. Zuker, A., Ahroni, A., Tzfira, T., Ovadis, M., Itzhaki, H., Scklarman, E., Ben-Meir, H. and Vainstein, A. (1998) In Abst. IX. International Congress on Plant Tissue and Cell Culture 35. (Received August 26, 1998; Accepted October 9, 1998)