Botanical Studies (2006) 47: 71-82.

*

Corresponding author: E-mail: kyto@gate.sinica.edu.tw;

Fax: 886-2-26515600; Tel: 886-2-26533161.

Cosuppression of tobacco chalcone synthase using

Petunia

chalcone synthase construct results in white

flowers

Chen-Kuen WANG, Po-Yen CHEN, Hsin-Mei WANG, and Kin-Ying TO*

Institute of BioAgricultural Sciences, Academia Sinica, Taipei 115, TAIWAN

(Received March 1, 2005; Accepted August 11, 2005)

ABSTRACT.

Chalcone synthase (CHS; EC 2.3.1.74) is a key enzyme in anthocyanin biosynthesis. In

order to understand the molecular mechanism controlling flower color, tobacco plants were transformed with

a chimeric construct containing expression cassettes for neomycin phosphotransferase II (nptII) selection

marker and CaMV 35S promoter-driven Petunia chsA cDNA, via Agrobacterium-mediated method. Four

transformants produced white flowers, while three transformants produced pink flowers similar to the

untransformed parent. Thin layer chromatography analysis revealed the absence of cyanidin in all white-

flowered transformants. Northern blot analysis showed that total chs mRNA levels were greatly decreased

in white-flowered transformants. By contrast, chs mRNA expression was induced in pink-flowered

transformants. RT-PCR analysis showed that the relative level of endogenous tobacco chs mRNA was less

than that of the transgenic Petunia chsA mRNA in white-flowered lines. In addition, plant/T-DNA junction

sequence analysis excluded the possibility that insertion of T-DNA into anthocyanin genes had inactivated

the anthocyanin biosynthetic pathway in white-flowered tobacco plants. Taken together, these results indicate

that cosuppression of the tobacco chs gene can occur using the equivalent Petunia gene, and demonstrate a

linkage between the expression level of chs mRNA, cyanidin content, and flower color in transgenic tobacco

plants.

Keywords: Anthocyanin; Chalcone synthase; Cosuppression; Flower color modification; Nicotiana tabacum;

Transgenic plants; Transgene silencing.

Abbreviations: CaMV, cauliflower mosaic virus; CHS, chalcone synthase; GUS, β-glucuronidase; NOS,

nopaline synthase; NPTII, neomycin phosphotransferase II; RT-PCR, reverse transcription-polymerase chain

reaction; T-DNA, transfer DNA; TLC, thin layer chromatography.

INTRODUCTION

Flower color is largely determined by two classes

of pigments: flavonoids, which contribute to a range

of colors from yellow to red to blue to purple; and

carotenoids, which are responsible for the red, orange

and yellow lipid-soluble pigments found embedded in

the membranes of chloroplasts and chromoplasts (Bartley

and Scolnik, 1995). Anthocyanins are a major colored

class of flavonoids that are responsible for the pink,

red, violet and blue colors of flowers and other tissues.

Anthocyanins perform diverse roles like attracting

pollinators and dispersing fruits and seeds. They also play

key roles in the signaling that takes place between plants

and microbes, in the male fertility of some species, in

defense as antimicrobial agents and feeding deterrents,

and in UV protection (Dixon and Steele, 1999; Forkmann

and Martens, 2001; Winkel-Shirley, 2001). Three common

anthocyanins are the pelargonidin-based (brick red to

orange), cyanidin-based (pink to red), and delphinidin-

based (blue) pigments.

The anthocyanin biosynthetic pathways of higher

plants—including those in Petunia, maize, snapdragon,

and recently Arabidopsis—are all well established

(Holton and Cornish, 1995; Mol et al., 1998; Winkel-

Shirley, 2001). Briefly, chalcone synthase (CHS; EC

2.3.1.74) catalyzes condensation of one molecule of

p-coumaroyl-coenzyme A (CoA) and 3 molecules of

malonyl-CoA, resulting in one molecule of 4, 2’, 4’,

6’-tetrahydroxychalcone (chalcone), which is a key

intermediate in the formation of flavonoids. The CHS

GENETICS

72

Botanical Studies, Vol. 47, 2006

enzyme, usually found in plant epidermal cells, has

a molecular weight of about 42,000 Da, requires no

cofactors, and has been isolated from several plant species,

including French bean, parsley, and carnation (Seigler,

1998). More recently, three-dimensional structure and

functional studies of alfalfa CHS enzyme have shown that

four residues (Cys164, Phe215, His303, Asn336) in the

catalytic site are responsible for its decarboxylation and

condensation reactions, and are highly conserved among

different species (Ferrer et al., 1999; Jez et al., 2000).

The first chs gene to be cloned was isolated from

parsley by an immunological approach (Kreuzaler

et al., 1983). The maize c2 gene, encoding a CHS

enzyme expressed in the aleurone layer of the kernel,

was isolated using transposon tagging (Wienand et al.,

1986). Additional chs genes have been isolated using

hybridization to cDNA libraries or genomic libraries with

the previously isolated chs clones. In Petunia hybrida

inbred line V30, 8 to 10 copies of chs genes, with high

homology (approximately 80%) at the DNA level, have

been identified (Koes et al., 1989). RNase protection

analysis with gene-specific probes have shown that only

two members, chsA and chsJ, are expressed in anthers

and corollas (Koes et al., 1989; Quattrocchio et al.,

1993) and that chsJ transcripts are 5% to 20% that of

chsA levels (Koes et al., 1989; Gutterson, 1995; Napoli

et al., 1999). In soybean, a total of eight chs genes have

recently been reported, but molecular characterization for

each member has not yet been done (Tuteja et al., 2004).

Expression of chs has been well studied in a number

of other plant species. In early developmental stages of

oat plants this enzyme is present in leaf tissue (Knogge

et al., 1986), in contrast to adult Petunia plants, where

its presence is limited to floral tissue (Koes et al., 1986;

Koes et al., 1989). Environmental stress, such as UV

light, phytopathogens and elicitors, or wounding, leads

to an induction of chs gene expression (Koes et al., 1989;

Winkel-Shirley, 2002). Developmental, tissue-specific and

inducible regulation of chs genes makes it an interesting

system with which to study the molecular mechanisms

underlying plant gene regulation.

Flower color is one of the most important

characteristics in ornamental plant breeding. By

controlling expression levels of genes involved in or

regulating the anthocyanin biosynthesis pathway, novel

varieties with respect to flower color have been obtained

from several plants (Forkmann and Martens, 2001;

Schijlen et al., 2004). The gene encoding the CHS enzyme

is one target that can be manipulated for just such a

purpose. Here we report the characterization of flower

color through introduction of a Petunia chsA cDNA

into tobacco, a model species for plant transformation.

We found that four out of seven transgenic plants

possessed flowers that were altered from pink to white.

Genetic analysis and molecular characterization of these

transformants were carried out, and they suggested a role

for cosuppression in these white flowers.

MATERIALS AND METHODS

Plant materials and growth conditions

Seeds of tobacco (Nicotiana tabacum cv. W38) and

Petunia hybrida cv. Ultra Blue were sown in a mixture

of peat and vermiculite and grown in a growth chamber

under a cycle of 14-h (6:00 a.m. to 8:00 p.m., 25°C)

illumination (250 μmol/m

2

/s) and 10-h (8:00 p.m. to 6:00

a.m., 20°C) darkness. After 1 month, plantlets were moved

into a green house and grown until maturation.

Plasmid construction and plant transformation

The transforming vector pCHS carrying expression

cassettes for neomycin phosphotransferase II (nptII)

selection marker and cauliflower mosaic virus (CaMV)

35S promoter-driven Petunia chsA was constructed as

described (Wang and To, 2004). This vector was used

to transform tobacco via Agrobacterium tumefaciens

strain LBA4404 as described (Horsch et al., 1985). After

selection on medium containing 100 μg/ml kanamycin

sulfate, regenerated plants were transferred to pots and

grown in a green house. Integration of the construct into

the tobacco genome was confirmed by PCR analysis as

described (Wang and To, 2004).

Genomic blot analysis

Genomic DNA from green leaves of tobacco was

isolated as previously described (To et al., 1999). DNA

restriction enzyme digests, Southern blotting, the synthesis

of probes containing CaMV 35S promoter and full-length

Petunia chsA cDNA, hybridization and detection were

described by Wang and To (2004).

Northern blot analysis

Total RNA was extracted from different tissues

with acid guanidinium thiocyanate-phenol-chloroform,

Northern analysis was performed and probed with full-

length Petunia chsA cDNA as previously described (To

et al., 1999; Huang et al., 2001; Wang and To, 2004). The

mRNA levels were quantified using X-ray films analyzed

with a chemiluminescent analyzer (IR LAS-100 Lite,

Fujifilm, Japan).

Extraction and TLC analysis of anthocyanins

Four flowers from each transgenic line and wild-type

tobacco were pooled and used for anthocyanin extraction

as described (Oud et al., 1995) with modification. Petals

were crushed with a mortar and pestle in 2 ml of 1%

(v/v) HCl in methanol. The methanol-HCl solution was

lyophilized. Anthocyanin pigments were dissolved in 0.8

ml of 2 N HCl and incubated for 30 min at 100°C. The

solution was cooled on ice and centrifuged for 2 min at

13,000 rpm. The supernatant was transferred into a new

Eppendorf tube. One hundred μl of isoamylalcohol was

added, and the mixture was vortexed vigorously for 2 min.

The organic phase was separated and stored at 4°C until

use.

WANG et al. — Genetic engineering of flower color

73

For TLC analysis, 2 μl of isoamylalcohol extract

was spotted onto a cellulose plate (TLC Plastic Sheets

Cellulose, Merck) using a micro capillary. The mobile

phase was acetic acid-water-HCl (30:10:3) (v/v). Pigment

standards of pelargonidin, cyanidin, and delphinidin were

purchased from Apin Chemicals (Oxfordshire, UK).

Seedling assay for kanamycin resistance

T1 seeds from self-fertilized transgenic plants were

sterilized in 1% sodium hypochloride for 20 min, and

washed thoroughly with sterile distilled water. They were

then germinated on a selection medium containing MS

salts, 3% sucrose, 0.8% agar, and 100 μg/ml kanamycin

sulfate. The cultures were incubated at 25°C under 14-h

illumination for 3 to 4 weeks. Seedlings with white

cotyledons, no true leaf development, and inhibition of

root extension were considered to be kanamycin sensitive

while seedlings with green cotyledons and healthy

development of leaves and roots were considered to be

kanamycin resistant.

RT-PCR analysis

Total RNA from different tissues was treated with

DNase I and 100 ng of treated total RNA was used to

perform reverse transcription-polymerase chain reaction

(RT-PCR) analysis, using a one-step RT-PCR kit (Qiagen).

For amplification of species-specific chsA regions,

primers P1 (5’ -CAGTGAGCACAAGACTGAT-3’ )

and P2 (5’-GAGATGGCCATCAATAGCA-3’) specific

for Petunia (accession no. AF233638), and primers

T1 (5’ -ACGGTACTCCGGATGGCT-3’) and T2 (5’

-GAAATTCCCAAAGGTTGG-3’) specific for tobacco

(accession no. AF311783), were employed simultaneously

in the same PCR tube (Figure 5C). Reverse transcription

was carried out at 50°C for 30 min. Initial PCR activation

was carried out at 95°C for 15 min. DNA amplification

was carried out for 30 cycles (denaturation at 94°C for

1 min, annealing at 55°C for 1 min, extension at 72°C

for 1 min). Final extension was performed at 72°C for

10 min. For amplification of consensus the chsA region

for both Petunia and tobacco plants, consensus primers

C1 (5’-ACAACAAGGGCGCTCGAG-3’) and C2 (5’-

CAAGCCCTTCACCAGTAG-3’) were employed (Figure

5F). Protocols for RT-PCR analysis were the same as

mentioned above, with the exception that the annealing

temperature for PCR amplification was changed from 55

°C to 57°C. Following amplification, 5 μl of PCR product

was analyzed on a 1% agarose gel. Quantification of

PCR products from RT-PCR analysis was conducted by

analyzing photographs with a chemiluminescent analyzer

(IR LAS-100 Lite, Fujifilm, Japan).

Cloning of plant/T-DNA insertion sequence

The strategy used for cloning plant/T-DNA insertion

sequences by an inverse PCR-based method has been

described elsewhere (Chen et al., 2003).

RESULTS

Transgenic plant verification

Seven transgenic plants were obtained. Genomic PCR

analysis was carried out to confirm integration of the

chimeric cassette into the tobacco chromosomal genome

from in vitro plantlets using sequence-specific primers for

nptII and Petunia chsA sequences (Figure 1A). No PCR

products were detected in wild-type tobacco (W38) using

nptII-specific or Petunia chs primers for amplification. A

unique band of approximately 0.8 kb, which is the nearly

expected size (795 bp) of the kanamycin resistant gene

(nptII), was observed in all transgenic plants using the

nptII-specific primers for amplification. A unique band

of approximately 1.2 kb, the expected size (1,170 bp) of

Petunia chsA cDNA, was also observed in all transgenic

plants using the Petunia chsA primers for amplification,

except for transformant CHS21 (right panel in Figure 1A).

Figu re 1. Confirmation of transgene in trans ge nic tobacco

plants. (A) Genom ic PCR analysis from wild-type tobacco

(W38) as well as transgenic plants. (B) Southern blot analysis

with non-radi oact ive DIG probe contai ning a CaMV 35S

promoter and Petunia chsA cDNA. Labels "p" and "w" in this

panel, as well as in Figure 3, Figure 4 and Figure 5, represent

the flower phenotype "pink" and "white".

74

Botanical Studies, Vol. 47, 2006

In vitro plantlets were transferred and grown in a

greenhouse. Large-scale isolation of genomic DNA from

green leaves and Southern blot analysis were used to

estimate the transgene copy number of each plant (Figure

1B). No EcoRI site is present in Petunia chsA or tobacco

chsA sequences. No hybridization band was detected in

wild-type DNA digested with EcoRI and probed with

the 2-kb DNA fragment containing CaMV 35S promoter

and Petunia chsA cDNA, probably due to the presence

of non-homologous CaMV 35S promoter (835 bp) in our

probe and to washing conditions. After hybridization,

the membrane was washed twice in washing solution

(0.2% SSPE; 0.1% SDS) at 65°C for 10 min and once in

buffer 1 (0.1 M maleic acid, pH 7.5; 0.15 M NaCl; 0.3%

Tween 20) at room temperature for 20 min, as described

previously (Wang and To, 2004). One hybridization band

was detected in CHS7, CHS11 and CHS21 transformants,

suggesting that only 1 copy of transgenic cassette was

integrated into their genomes. Two hybridization bands

were detected in CHS8, CHS9, CHS10 and CHS22

transformants, suggesting that there were two insertions in

these transgenic lines.

Alteration of flower color in transgenic plants

The seven transgenic tobacco plants were grown

to maturity. White flowers were observed in four

transformants (CHS7, CHS9, CHS10 and CHS22),

and pink flowers were observed in the remaining three

(CHS8, CHS11 and CHS21) as well as in the wild-type

tobacco control plant (W38) (Figure 2). No apparent

phenotypes other than flower color were affected by the

transformation.

Figure 2. Alteration of flower color in transgenic tobacco plants by overexpression of sense Petunia chalcone synthase construct.

Seven transgenic plants were obtained, and four transformants (CHS7, CHS9, CHS10, CHS22) carried white flowers. Similar pink

floral pigmentation as wild-type tobacco (W38) was observed in transformants CHS8, CHS11 and CHS21.

WANG et al. — Genetic engineering of flower color

75

To further investigate the relationship between

the accumulation of anthocyanidins and phenotypic

variation in flower color of transgenic plants, thin layer

chromatography (TLC) analysis was conducted (Figure

3). The major component of anthocyanins extracted from

petals of wild-type pink flowers as well as from petals

of three transformants carrying pink flowers (CHS8,

CHS11, CHS21) was cyanidin.

In contrast, transformants

with white flowers (CHS7, CHS9, CHS10, CHS22)

accumulated a very low to undetectable level of cyanidin,

suggesting blockage of the cyanidin biosynthetic pathway.

Total

chs

mRNA is largely suppressed in white-

flowered transgenic plants

Overexpression of chimeric sense or antisense chs

constructs reducing floral anthocyanin accumulation has

been reported elsewhere (Napoli et al., 1990; Courtney-

Gutterson et al., 1994; Jorgensen et al., 1996; Metzlaff

et al., 2000; Suzuki et al., 2000). To examine chs mRNA

in transgenic plants, total RNA was isolated from petals

and probed with the Petunia chsA sequence (Figure 4).

A strong signal was detected in wild-type tobacco (W38)

with the relative chs mRNA level in this sample set as

indicating 100% chs mRNA expression. A similar chs

mRNA expression level was detected in a white-flowered

tranformant CHS9 (105%). chs mRNA was reduced in

other white-flowered transformants (CHS7, 38%; CHS10,

50%; CHS22, 70%), but chs mRNA level was induced

in pink-flowered transformants (CHS8, 351%; CHS11,

461%; CHS21, 143%).

Relative levels of transgenic and endogenous

chs

mRNA in transgenic plants

To further characterize and distinguish endogenous

chs expression from foreign Petunia chsA mRNA in

transgenic tobacco plants, nucleotide sequences of the

chs coding region between Petunia and tobacco were

compared (Figure 5A), and high similarities (91%

identity) were observed. Two primer sets, specific for

tobacco (T1 and T2 primers) or Petunia (P1 and P2

primers), were designed accordingly (Figure 5B). Total

RNA was isolated from petals in different transformants as

well as wild-type plants, and species-specific primer sets

were employed simultaneously to examine the relative

levels of transgenic and endogenous chs mRNA by RT-

PCR analysis (Figure 5C). Only tobacco-specific but

not Petunia-specific chs mRNA was detected in wild-

type tobacco plants, and only Petunia-specific but not

tobacco-specific chs mRNA was detected in wild-type

Petunia plants (Figure 5C). Two different transcripts

representing the transgenic and endogenous chs mRNA

were detected in all transgenic plants, with the exception

that only endogenous tobacco-specific chs mRNA was

detected in a pink-flowered transformant CHS21 (Figure

5C). Comparison of the relative percentages of transgenic

and endogenous chs mRNA within each transgenic

plant clearly showed that the intensity of a PCR band

(approximately 0.4 kb) corresponding to tobacco-specific

chs mRNA was obviously weaker in all white-flowered

transformants (CHS7, 19%; CHS9, 15%; CHS10, 20%;

CHS22, 15%) than in those pink-flowered transformants

(CHS8, 37%; CHS11, 41%; CHS21, 100%). Higher levels

of transgenic Petunia-specific chs mRNA, as revealed by

Petunia-specific PCR product (approximately 0.65 kb),

were detected in the pink-flowered transformants (CHS8,

CHS11) (Figure 5C).

We also tested whether this gene construct was

expressed in other tissues of transgenic plants. In brief,

total RNA isolated from leaf tissue of transgenic plants

and Petunia-specific P1/P2 primers were used in RT-

PCR analysis (Figure 5D). A unique PCR product with

the expected size (approximately 0.65 kb) was detected in

six of the seven transformants. No PCR amplification was

observed in transformant CHS21 or wild-type tobacco

(W38) (Figure 5D).

Figure 3. T hin layer chrom atography analysis showing the

absence of cyanidin in white-flowering transgenic plants.

Figure 4. Northern blot analysis of the total chs mRNA level in

transgenic plants.

76

Botanical Studies, Vol. 47, 2006

Figure 5. Sequence comparison of tobacco and Petunia chs cDNA and RT-PCR analysis showing cosuppression of endogenous chs

mRNA in white-flowered transformants. (A) Nucleotide sequences of tobacco (accession no. AF311783) and Petunia (accession no.

AF233638) cDNAs encoding chalcone synthase. Different nucleotides between chs sequence of these two plant species are highlighted

in gray, and locations of primers are indicated by arrows. The recognition sequence for NcoI is boxed. (B) Strategy for identification

of foreign Petunia chs mRNA expression in transgenic tobacco plants by RT-PCR analysis. (C) Relative levels of transgenic (640

bp) and endogenous (374 bp) chs mRNA by RT-PCR analysis. Endogenous chs mRNA was largely reduced in those white-flowered

transformants (CHS7, CHS9, CHS10, CHS22). Abbreviations "Tob" and "Pet" in this panel, as well as in Figure 5D, Figure 5F, and

Figure 5G represent "wild-type tobacco W38" and "wild-type Petunia". (D) Detection of foreign Petunia chs mRNA in the leaf tissue

of transformants by RT-PCR analysis. (E) Strategy for confirmation of foreign Petunia chs mRNA expression in transgenic tobacco

plants by RT-PCR analysis combined with restriction enzyme digestion. (F) Total RNA from petals of each sample was isolated, and

the primer set C1/C2 was used to amplify chs transcripts from both tobacco and Petunia chs cDNA by RT-PCR analysis. NcoI was

added to digest the Petunia chs PCR fragments. (G) 100 ng of DNase-treated RNA from each sample prior to RT-PCR was run as a

loading control on a 1% RNA agarose gel.

WANG et al. — Genetic engineering of flower color

77

To exclude the possibility that the differential intensity

of RT-PCR products (Figure 5C) resulted from different

Tm values for each primer set (Tm for P1 and P2 is 56°C

and 58°C, respectively whereas Tm for T1 and T2 is 58°C

and 52°C, respectively), another primer set C1/C2 (Tm for

C1 and C2 is 58°C and 56°C, respectively) was designed

to amplify the 558 bp fragment from both chs cDNA of

tobacco and Petunia (Figure 5E). To differentiate between

the two chs cDNAs, we used the fact that the Petunia-

specific chs fragment contained a unique NcoI restriction

site not found in the tobacco fragment. Digestion of the

Petunia PCR fragment resulted in two smaller segments of

212 bp and 366 bp while the tobacco-specific chs fragment

remained uncut at 558 bp due to the absence of a NcoI

restriction site. Petal RNA samples were prepared from

different transformants as well as the wild-type tobacco

and Petunia plants. The consensus C1/C2 primer set was

employed to examine the relative levels of transgenic and

endogenous chs mRNA by RT-PCR analysis (Figure 5F).

A unique band of approximately 558 bp was detected in

all uncut samples examined. After purification, the PCR

products were subjected to NcoI digestion. Two smaller

fragments were observed in an RNA sample isolated from

control Petunia plants; by contrast and as expected, only

one band of approximately 558 bp was detected in the

RNA sample isolated from the wild-type tobacco plants.

We further examined the percentage of endogenous and

transgenic chs gene expression in individual transgenic

plants based on NcoI digestion (Figure 5F). In those

white-flowered transformants, the relative mRNA level of

endogenous chs gene (CHS7, 17%; CHS9, 7%; CHS10,

17%; CHS22, 22%) was found much lower than the

transgenic Petunia-specific chs gene (CHS7, 83%; CHS9,

93%; CHS10, 83%; CHS22, 78%). By contrast, the

relative mRNA level of the endogenous chs gene (CHS8,

37%; CHS11, 30%) was found to be around 50% the level

of the transgenic Petunia-specific chs gene (CHS8, 63%;

CHS11, 70%) in each pink-flowered transformant.

Inheritance in T1 progeny of transgenic plants

The stable integration of the chimeric construct into

the tobacco chromosomal genome was investigated by

germinating T1 seeds from self-fertilized transgenic plants

on MS medium containing 100 μg/ml kanamycin sulfate

(Table 1). Kanamycin-resistant and kanamycin-sensitive

T1 seedlings segregated in a 3:1 (resistant:sensitive) ratio

in three transformants (CHS7, CHS10, CHS22) and a

15:1 ratio in the other two transformants (CHS8, CHS11),

suggesting the presence of single copy of nptII transgene

in the nuclear genomes of three transformants (CHS7,

CHS10, CHS22) and two copies of nptII transgene in the

nuclear genomes of two transformants (CHS8, CHS11).

More than two copies of the nptII transgene were found

in transformant CHS21. No nptII transgene was found

in transformant CHS9 or wild-type tobacco W38. In

summary, transformant CHS9 was considered to be

kanamycin sensitive, and the other six transformants were

considered to be kanamycin resistant (Table 1). Since the

transforming vector pCHS contains expression cassettes

for the selection marker gene (nptII) and Petunia chs

cDNA within the T-DNA region (Wang and To, 2004),

it is reasonable to predict the existence of the same copy

number of the nptII transgene and Petunia chs transgene

in each transformant (Table 1).

Molecular cloning of plant flanking sequence

The Agrobacterium-mediated transformation has

been useful for introducing new genes into plants and

for inactivation of plant genes by T-DNA insertion

mutagenesis (Tinland, 1996). We now employed an

inverse PCR-based method (Chen et al., 2003) to

determine the T-DNA insertion sequences in all white-

flowered transformants (CHS7, CHS9, CHS10, CHS22)

and one pink-flowered transformant CHS21, which did

not express Petunia chs mRNA (Figure 5). A unique

plant DNA sequence, near the right border of T-DNA,

Table 1. Alteration of flower color and inheritance in T

1

progeny assay of transgenic tobacco plants.

a

Transformant line Flower’s color

No. of resistant and sensitive

seedlings from T

0

selfed seeds Test ratio

(R:S) χ

2

P Predicted copy no.

of nptII transgene

R

S

CHS7

White

125

43

3:1 0.0317 0.95~0.80

1

CHS8

Pink

277

18

15:1 0.0093 0.95~0.80

2

CHS9

White

0

198

0

CHS10

White

408

138

3:1 0.0165 0.95~0.80

1

CHS11

Pink

513

36

15:1 0.0899 0.8~0.7

2

CHS21

Pink

374

12

>2

CHS22

White

301

95

3:1 0.2155 0.7~0.5

1

Wild type W38

Pink

0

248

0

a

Seeds germinated on medium containing 100 μg ml

-1

kanamycin for 3 to 4 weeks.

78

Botanical Studies, Vol. 47, 2006

was obtained for transformants CHS7, CHS9 and

CHS10 (Figure 6). The flanking DNA sequence (266

bp) in transformant CHS7 had no sequence similarity

to any public domain nucleotide or expressed sequence

tag (EST) databases; however, the flanking sequence

(287 bp) in transformant CHS10, which also had no

sequence similarity to nucleotide databases, showed a

high similarity to several EST sequences including DD6-4

(tobacco ESTs characterized by hypersensitive response

specific pattern; E-value=1E-58), EST606585 (mixed

potato tissue cDNA clone; E-value=2E-23), EST498621

(P. infestans-challenged leaf Solanum tuberosum cDNA

clone; E-value=2E-23), EST580272 (potato roots

Solanum tuberosum cDNA clone; E-value=1E-21) and

EST895187 (Lycopersicon esculentum maturing fruit

cDNA clone; E-value=4E-15). For transformant CHS9,

a partial region (79 bp) of the plant insertion sequence

(156 bp) was identical to the transgenic Petunia chs gene

(complementary to nucleotide positions 1092 to 1170 in

Figure 5A), indicating that the right border of one T-DNA

construct carrying the chimeric expression cassette for

nptII and Petunia chsA genes had been integrated into

the 5’ end of the Petunia chsA gene in another T-DNA

construct likely carrying the same chimeric expression

cassette.

Three different inserts were detected in transformant

CHS21. Surprisingly, the insert sequence (55 bp) in clone

CHS21-1 was a partial sequence in the T-DNA left border

region of the transforming vector pCHS while the insert

sequence (321 bp) in clone CHS21-3 was also a partial

vector sequence near the 5’ terminus of the T-DNA right

border. The insert sequence (400 bp) in clone CHS21-2

showed no sequence similarity with transforming

vector pCHS and was considered to be a plant DNA

sequence. However, no sequence similarity was found

in public databases. Two different inserts were detected

in transformant CHS22. The insert sequence (194 bp) in

clone CHS22-1 was a partial T-DNA sequence between

the NOS terminator and the T-DNA left border. The

insert sequence (55 bp) in clone CHS22-2 showed 100%

identity to the tobacco intergenic spacer of the rRNA gene

(GenBank accession no. D76443). No sequence similarity

was found among the insertion sequences we obtained,

supporting the hypothesis of random integration of T-DNA

into plant chromosomal genomes (Tinland, 1996; Hellens

et al., 2000).

Figure 6. Inverse PCR analysis of selected transgenic tobacco plants showing plant/T-DNA insertion and relevant sequences. The

nptII fragment (red), right border region (blue) and NOS promoter region (yellow), are derived from the transforming vector pCHS.

The obtained tobacco sequence is indicated in green color, and the first TaqI restriction site within the tobacco sequence is underlined.

WANG et al. — Genetic engineering of flower color

79

Taken together, partial sequences corresponding

to T-DNA left border or right border were detected in

transformants CHS9, CHS21 and CHS22, suggesting the

presence of direct or invert repeats in their genomes. This

is consistent with the observation that multiple T-DNA

frequently integrate at the same position in plant genomes

during Agrobacterium-mediated transformation, resulting

in formation of inverted and direct repeats (De Buck et

al., 1999). No sequence similarity was found between

obtained plant sequences and any well-known pigment-

related genes in databases. We thus conclude that loss of

flower pigmentation in the white-flowered transformants

we examined was not due to the inactivation of pigment-

synthesizing genes in the host plant genome by T-DNA

insertion.

DISCUSSION

In this study, a Petunia chsA transgene was introduced

into tobacco, and its effect on flower colors in transgenic

tobacco plants was investigated. Four out of seven

transgenic plants showed phenotypic alteration from pink

flowers to pure white flowers. It has been well documented

that introduction of additional copies of chs gene into

Petunia and Arabidopsis plants frequently results in

events of chs post-transcriptional gene silencing (PTGS)

or "cosuppression" (Depicker and Van Montagu, 1997;

Metzlaff et al., 2000). Because CHS is the key enzyme in

the anthocyanin biosynthetic pathway, silencing of this

gene can be easily monitored by loss of pigmentation in

flowers. Multiple models of PTGS have been proposed,

including roles for RNA thresholds and DNA repeats

(Baulcombe, 1996). In parallel, PTGS has been suggested

to be due to involvement of a form of aberrant or double

stranded RNA (Vaucheret et al., 1998; Chicas and Macino,

2001; Vaucheret et al., 2001).

Introduction of sense or antisense chs constructs has

been carried out to modify flower color in various species

(see below). However, to the best of our knowledge,

introduction of sense chimeric chs into tobacco plants has

not been reported. van der Krol et al. (1988) reported that

when antisense Petunia chs cDNA was introduced into

tobacco plants, 36 out of 40 had flowers indistinguishable

from wild-type tobacco flowers; three plants gave

flowers with sectored pigmentation; and one plant gave

completely white flowers. The effect of overexpressing

a sense Petunia chs cDNA in Petunia plants on flower

color patterns has also been studied extensively. van der

Krol et al. (1990) reported that six transgenic plants were

obtained from P. hybrida var. VR (violet) using sense

Petunia chs construct whilst no phenotypic changes were

observed in any of the transgenic flowers; however, two

of these plants, when grown under higher light conditions,

produced flowers with patches of white floral tissue. In

contrast, Napoli et al. (1990) reported that three out of six

transformants from a Petunia commercial hybrid variety

(Pink Cascade) produced pure white flowers. A progeny

assay from one of these white-flowering transformants

demonstrated that the novel color phenotype co-segregated

with introduced chs whereas progeny plants without

the transformed gene showed a wild-type phenotype.

RNase protection analysis of petal RNAs isolated from

white flowers showed that the level of the chs mRNA

was reduced 50-fold from wild-type plants (Napoli et

al., 1990). Molecular breeding to generate white flowers

has been attempted in the high-value medicinal plant

Echinacea purpurea (Wang and To, 2004) and in several

ornamental plants, including Dendranthema grandiflora

(Courtney-Gutterson et al., 1994), Eustoma grandiflorum

(Deroles et al., 1998), Gerbera hybrida (Elomaa et

al., 1993), Torenia fournieri (Aida et al., 2000), and T.

hybrida (Suzuki et al., 2000) using sense or antisense

chs constructs; however, only D. grandiflora (florist’s

chrysanthemum), E. grandiflorum (lisianthus flowers), and

T. hybrida have produced white-flowered transformants.

These experiments and our present study show that

genetic modification of flower color can be accomplished

by transforming chimeric or endogenous chs constructs

into plants. However, effects on color modification are not

easily predicted and can differ between plant species and

varieties of the same species.

The kanamycin-sensitive phenotype in the white-

flowered transformant CHS9 in this study is interesting.

The kanamycin-resistant gene (nptII) is present in this

transformant (Figure 1A). Moreover, integration of the

chimeric cassette into tobacco genome was demonstrated

by Southern blot analysis (Figure 1B) and the plant/

T-DNA insertion site was characterized (Figure 6).

Petunia-specific chs mRNA was also detected by RT-

PCR analysis of this transformant (Figures 5C and

5F). An explanation for this observation may include

preferential silencing of the foreign nptII gene but not the

foreign Petunia chs gene within the chimeric construct

of this transformant. Since Petunia-specific chs mRNA

could be detected in both flower and leaf tissues from

this transformant (Figure 5), it represents an excellent

resource for studying the characteristics and mechanisms

of transgene-induced gene silencing in plants (Gelvin,

1998; Vaucheret et al., 1998; De Wilde et al., 2000;

Chicas and Macino, 2001; Vaucheret et al., 2001; Han

and Grierson, 2002). Another transformant CHS21 is

also interesting. This line shows normal levels of the

endogenous chs transcription (Figure 5) and cyanidin

content (Figure 3) and produces normal pink flowers. In

addition, this line had detectable levels of the transgenic

nptII gene (left panel in Figure 1A) and was resistant to

kanamycin (Table 1) but showed no detectable levels of

the transgenic Petunia chs gene (right panel in Figure 1A)

and its corresponding transgenic chs transcript (Figure 5),

suggesting that only a partial Petunia chs gene may have

been transferred into the plant genome during the T-DNA

integration process.

Northern blot analysis of total chs mRNA levels (Figure

4) together with RT-PCR analysis (Figure 5) demonstrated

cosuppression of chs mRNA occurred in white-flowered

transformants while overexpression of chs mRNA

80

Botanical Studies, Vol. 47, 2006

occurred in pink-flowered transformants. Overexpression

o f chs sense or chs antisense suppression resulting

in white-flowered transformants has been reported

previously (Napoli et al., 1990; Courtney-Gutterson et

al., 1994; Jorgensen et al., 1996; Metzlaff et al., 2000;

Suzuki et al., 2000). TLC analysis of anthocyanidin

accumulation (Figure 3) showed that loss of flower

color in our transgenic plants (CHS7, CHS9, CHS10,

CHS22) was linked to a dramatic reduction in chs mRNA

levels (Figure 4). In contrast, the content of cyanidin in

transformants with pink flowers (CHS8, CHS11, CHS21)

with overexpressed chs mRNA levels, was found to be

similar (CHS8, CHS21) or even lower (CHS11) than wild-

type plants. Since the deduced amino acid sequence of

the Petunia and tobacco CHS shows 95% identity, it is

possible that Petunia CHS enzyme could be catalytically

active in transgenic tobacco. Chalcone (the product of

the CHS reaction) and naringenin (the product of the

CHI reaction), are not only important for synthesizing

different kinds of anthocyanins, but they are also key

intermediates for synthesizing different flavonoids (Dixon

and Steele, 1999; Forkmann and Martens, 2001; Winkel-

Shirley, 2001). It has been reported that overexpression

of Petunia chalcone isomerase (the enzyme catalyzes

chalcone to naringenin) in tomato produced an increase

of up to 78-fold in fruit peel flavonols, mainly due to

an accumulation of rutin; however, no gross phenotypic

differences were observed between high-flavonol

transgenic and control lines (Muir et al., 2001). It is

possible that the overexpression of the Petunia chs gene

in transgenic tobacco plants may also lead to the synthesis

and then accumulation of other types of flavonoids

(most of which are colorless) while maintaining steady

amounts of anthocyanins and, as a result, could generate

flowers with a similar (pink) color. Further experiments

of flavonoid profiling and the use of species-specific

antibodies which can distinguish Petunia and tobacco

CHS enzymes will be useful in understanding flower color

modification by chimeric chs gene expression.

Acknowledgements. We thank Dr. Kenrick J. Deen for

critical reading of the manuscript. We are also grateful to

the Institute of Plant and Microbial Biology, Academia

Sinica, for providing greenhouse and transgenic core

facilities. This work was financially supported by a grant

from Academia Sinica of the Republic of China to Dr.

Kin-Ying To.

LITERATURE CITED

Aida, R., S . Ki shimoto, Y. Tanaka, and M. S hibata. 2000.

Modification of flower color in torenia (Torenia fournieri

Lind.) by genetic transformation. Plant Sci. 153: 33-42.

Bartl ey, G.E. a nd P.A. S colni k. 1995. Pl ant c arote noids :

pigments for photoprotection, visual attraction, and human

health. Plant Cell 7: 1027-1038.

Baulcombe, D.C. 1996. RNA as a target and a initiator of post-

transcriptional gene silencing in trans genic plants. P lant

Mol. Biol. 32: 79-88.

C hen, P.Y., C .K. Wan g, S .C. S oong, and K.Y. To. 20 03.

Complete sequence of the binary vector pBI121 and its

application in cloning T-DNA insertion from transgenic

plants. Mol. Breed. 11: 287-293.

Chica s, A. and G. Macino. 2001. Charac teris tics of post-

transcriptional gene silencing. EMBO Rep. 2: 992-996.

Courtney-Gutterson, N., C. Napoli, C. Lemieux, E. Firoozabady,

and K.E.P. Robinson. 1994. Modification of flower color in

florist’s chrysanthemum: production of a white-flowering

variety through molecular genetics. Bio/Technology 12:

268-271.

De Buck, S., A. Jacobs , M. Van Montagu, and A. Depicker.

1999. The DNA sequences of T-DNA junctions suggest

that complex T-DNA loci are formed by a recombination

process resembling T-DNA integration. Plant J. 20:

295-304.

De Wilde, C., H. Van Houdt, S. De Buck, G. Angenon, G. De

Jaeger, and A. Depicker. 2000. Plants as bioreactors for

protein production: avoiding the problem of transgene

silencing. Plant Mol. Biol. 43: 347-359.

Depicker, A. and M. Van Montagu. 1997. Post-transcriptional

gene silencing in plants. Curr. Opin. Cell Biol. 9: 373-382.

Deroles, S.C., J.M. Bradley, K.E. Schwinn, K.R. Markham, S.

Bloor, D.G. Manson, and K.M. Davies. 1998. An antisense

chalcone synthase gene leads to novel flower patterns in

lisianthus (Eustoma grandiflorum). Mol. Breed. 4: 59-66.

D i xo n , R .A . a nd C . L . S t e e l e . 1 99 9 . F l a v on o i ds a n d

isoflavonoids-a gold mi ne for m etabolic enginee ri ng.

Trends Plant Sci. 4: 394-400.

Elomaa, P., J. Honkanen, R. Puska, P. Seppanen, Y. Helariutta,

M. Mehto, M. Kotilainen, L. Nevalainen, and T.H. Teeri.

1993. Agrobacterium-mediated transfer of antisense

chalc one synthas e cDNA to Ger ber a hy brida inhibits

flower pigmentation. Bio/Technology 11: 508-511.

Ferrer, J.L., J.M. Jez, M.E. Bowman, R.A. Dixon, and J.P. Noel.

1999. Structure of chalcone synthase and the molecular

basis of plant polyketide biosynthesis. Nat. Struct. Biol. 6:

775-784.

F orkmann, G. and S. Martens. 2001. Metabolic engineering

and applications of flavonoids. Curr. Opin. Biotechnol. 12:

155-160.

Gel vi n, S .B . 199 8. T he int rod uct io n a nd e xpr es s io n of

transgenes in plants. Curr. Opin. Biotechnol. 9: 227-232.

Gutterson, N. 1995. Anthocyanin biosynthetic genes and their

application to flower color modification through sense

suppression. HortScience 30: 964-966.

Han, Y. and D. Grierson. 2002. Relationship between s mall

an tis ens e RNAs a nd abe rrant RNAs a ss oc iat ed wit h

transgene mediated gene silencing in tomato. Plant J. 29:

509-519.

Hellens , R., P. Mull ine aux, and H. Klee. 2000. A gui de to

Agrobacterium binary Ti ve ctors . Trends Pla nt S ci. 5:

446-451.

WANG et al. — Genetic engineering of flower color

81

Holton, T.A. and E.C. Cornish. 1995. Genetics and biochemistry

of anthocyanin biosynthesis. Plant Cell 7: 1071-1083.

Horsch, R.B., J.F. Fry, N.L. Hoffman, D. Eicholts, S.G. Rogers,

and R.T. F raley. 1985. A simple and general method for

transferring genes into plants. Science 227: 1229-1231.

Huang, Y.J., K.Y. To, M.N. Yap, W.J. Chiang, D.F. Suen, and

S.C.G. Chen. 2001. Cloning and characterization of leaf

senescence up-regulated genes in sweet potato. Phys iol.

Plant. 113: 384-391.

Jez, J.M., J.L. Ferrer, M.E. Bowman, R.A. Dixon, and J.P. Noel.

2000. Dissection of malonyl-coenzyme A decarboxylation

from polyketide formation in reaction mechanism of a plant

polyketide synthase. Biochemistry 39: 890-902.

J orgensen, R.A., P.D. Cluster, J. English, Q. Que, and C.A.

Napoli. 1996. Chalcone synthase cosuppression phenotypes

in petunia flowers: c om parison of sens e vs. antis ens e

constructs and single-copy vs. complex T-DNA. Plant Mol.

Biol. 31: 957-973.

Knogge, W., E. Schmelzer, and G. Weissenbock. 1986. The

role of chalcone synthase in the regulation of flavonoid

biosynt he si s in deve loping oat prim ary lea ves. Arch.

Biochem. Biophys. 250: 364-372.

Koes, R.E., C.E. Spelt, H.J. Reif, P.J.M. van den Elzen, E.

Veltkamp, and J.N. Mol. 1986. Floral tiss ue of Petunia

hybrida (V30) expresses only one member of the chalcone

synthase multigene family. Nucl. Acids Res. 14: 5229-5239.

Koes , R.E ., C.E . S pelt, P.J .M. van de n E lzen, and J .N.M.

Mol. 1989. Cloning and molecular characterization of the

chalcone synthase multigene family of Petunia hybrida.

Gene 81: 245-257.

Kreuzaler, F., H. Ragg, E. Fautz, D.N. Kuhn, and K. Hahlbrock.

1983. UV-induction of chalcone synthas e mRNA in cell

suspension cultures of Petroselinum hortense. P roc. Natl.

Acad. Sci. USA 80: 2591-2593.

Metzlaff, M., M. O’Dell, R. Hellens, and R.B. Flavell. 2000.

Developmental and transgene regulated nuclear processing

of primary transcripts of chalcone synthase A in petunia.

Plant J. 23: 63-72.

Mol, J., E. Grotewold, and R. Koes. 1998. How genes paint

flowers and seeds. Trends Plant Sci. 3: 212-217.

Muir, S.R., G.J. Collins, S. Robinson, S. Hughes, A. Bovy, C.H.

Ric De Vos, A.J. van Tunen, and M.E. Verhoeyen. 2001.

Overexpress ion of petunia chalcone isomeras e in tomato

results in fruit containing dramatically increase levels of

flavonols. Nat. Biotechnol. 19: 470-474.

Napoli, C.A., D. Fahy, H.Y. Wang, and L.P. Taylor. 1999. white

anther: a petunia mutant that abolis hes pollen flavonol

accumulation, induces male sterility, and is complemented

by a chalcone synthase transgene. P lant Physiol. 120:

615-622.

Napoli, C., C. Lemieux, and R. Jorgensen. 1990. Introduction

of a chimeric chalcone synthase gene into petunia results

in reversible cosuppression of homologous genes in trans.

Plant Cell 2: 279-289.

Oud, J .S .N., H. Schne iders, A.J. Kool, and M.Q.J.M. van

Grinsven. 1995. Breeding of transgenic orange Petunia

hybrida varieties. Euphytica 85: 403-409.

Quattrocchio, F., J.F. Wing, H.T.C. Leppen, J.N.M. Mol, and

R.E. Koes. 1993. Regulatory genes controlling anthocyanin

pigm entation are functionally conse rved a mong pl ant

species and have distinct sets of target genes. Plant Cell 5:

1497-1512.

Schijlen, E.G.W.M., C.H. Ric de Vos, A.J. van Tunen, and A.G.

Bovy. 2004. Modification of flavonoid biosynthesis in crop

plants. Phytochemistry 65: 2631-2648.

S eigle r, D.S . 1998. P lant S econda ry Met abolis m. Kl uwe r

Academic, Boston.

Suzuki, K.I., H.M. Xue, Y. Tanaka, Y. F ukui, M. F ukuchi-

Mizutani, Y. Murakami, Y. Katsumoto, S. Tsuda, and T.

Kusumi. 2000. Flower color m odifica tions of Tore nia

hybrida by cos uppress ion of anthocyanin bios ynthesis

genes. Mol. Breed. 6: 239-246.

Tinland, B. 1996. The integration of T-DNA into plant genomes.

Trends Plant Sci. 1: 178-184.

To, K.Y., D.F. Suen, and S.C.G. Chen. 1999. Mol ec ula r

characterization of ribulose-1,5-bisphosphate carboxylase/

oxygenase activase in rice leaves. Planta 209: 66-76.

Tuteja, J.H., S.J. Clough, W.C. Chen, and L.O. Vodkin. 2004.

Tiss ue-specific gene silencing mediated by a naturally

occurring chalcone synthase gene cluster in Glycine max.

Plant Cell 16: 819-835.

Van der Krol, A.R., P.E. Lenting, J. Veenstra, I.M. van der Meer,

R.E. Koes, A.G.M. Gerats, J.N.M. Mol, and A.R. Stuitje.

1988. An anti-sense chalcone synthase gene in transgenic

plants inhibits flower pigmentation. Nature 333: 866-869.

Van der Krol, A.R., L.A. Mur, M. Beld, J.N.M. Mol, and A.R.

S tuitje. 1990. Flavonoid genes in petunia: addition of a

limited number of gene copies may lead to suppression of

gene expression. Plant Cell 2: 291-299.

Vaucheret, H., C. Beclin, T. Elmayan, F. Feuerbach, C. Godon,

J-B. Morel, P. Mourrain, J-C. Palauqui, and S. Vernhettes.

1998. Transgene-induced gene silencing in plants. Plant J.

16: 651-659.

Vaucheret, H., C. Beclin, and M. Fagard. 2001. Post-

transcriptional gene silencing in plants. J Cell Sci. 114:

3083-3091.

Wang, H.M. and K.Y. To. 2004. Agrobacterium-me diated

transformation in the high-value medicinal plant Echinacea

purpurea. Plant Sci. 166: 1087-1096.

Wie na nd, U., U. We ydemann, U. Ni esba ch-Klos gen, P.A.

Peterson, and H. Saedler. 1986. Molecular cloning of the c2

locus of Zea mays, the gene coding for chalcone synthase.

Mol. Gen. Genet. 203: 202-207.

Winkel-S hirley, B. 2001. Flavonoid biosynthesis: a colorful

m ode l fo r gen et ics , b ioc hem is t ry, c el l bi olog y, a nd

biotechnology. Plant Physiol. 126: 485-493.

Winkel-Shirley, B. 2002. Biosynthesis of flavonoids and effects

of stress. Curr. Opin. Plant Biol. 5: 218-223.

82

Botanical Studies, Vol. 47, 2006

利用矮牽牛苯基苯乙烯酮合成.對轉殖菸草苯基苯乙烯酮合

成.的共同抑制作用以產生白色花.

王貞觀　陳柏言　王幸美　陶建英

中央研究院生物農業科學研究所籌備處

　　苯基苯乙烯酮合成.（chalcone synthase，簡稱 Chs；EC 2.3.1.74）是花色素 (anthocyanin) 生合成路

徑的關鍵酵素。欲了解花.顏色的分子½控機制，我們利用農桿菌 (Agrobacterium) 轉殖基因的方法，

將鑲嵌½體含 CaMV 35S 啟動子及矮牽牛 chsA cDNA 轉殖至菸草。所獲得轉殖菸草中，4 株轉殖株產

生白色花.，3 株轉殖株產生與野生型植株相同的粉紅色花.。薄層色層分析的結果顯示，所有白色花

.的轉殖株皆缺乏花青素 (cyanidin)。北方氏墨點分析法顯示，白色花.轉殖株內 chs mRNA 的總量大

幅下降；相反的，花色屬粉紅色的轉殖株，chs mRNA 的總量卻顯著增加。RT-PCT 分析結果顯示，花

色屬白色的轉殖株，其內生型菸草 chs mRNA 的相對量少於轉殖矮牽牛 chs mRNA。此外，植物/T-DNA

接合序列分析亦排除產生白色花色的原因是由於 T-DNA 插入植物色素合成相關基因導致該基因喪失活

性的可能性。綜合而言，本研究提供了廣泛的實驗證據，說明菸草 chs 基因的共同抑制作用是可以透過

別種植物如矮牽牛的相似基因而發揮作用，導致轉殖菸草產生白色花.；並清楚顯示 chs mRNA 表現

量、花青素含量及轉殖菸草花.顏色的關聯性。

關鍵詞：花色素；苯基苯乙烯酮合成.；共同抑制作用；花色之修飾；Nicotiana tabacum；轉殖植物；

轉殖基因導致的基因靜默。