Plant Cell Rep (2006) 25: 741–746
DOI 10.1007/s00299-006-0119-4
PHYSIOLOGY AND BIOCHEMISTRY
Jae Hyung Ko · Bong Gyu Kim · Hor-Gil Hur ·
Yoongho Lim · Joong-Hoon Ahn
Molecular cloning, expression and characterization
of a glycosyltransferase from rice
Received: 1 September 2005 / Revised: 13 January 2006 / Accepted: 14 January 2006 / Published online: 14 February 2006


C Springer-Verlag 2006
Abstract Secondary plant metabolites undergo several
modification reactions, including glycosylation. Glyco-
sylation, which is mediated by UDP-glycosyltransferase
(UGT), plays a role in the storage of secondary metabolites
and in defending plants against stress. In this study, we
cloned one of the glycosyltransferases from rice, RUGT-5
resulting in 40–42% sequence homology with UGTs from
other plants. RUGT-5 was functionally expressed as a glu-
tathione S-transferase fusion protein in Escherichia coli
and was then purified. Eight different flavonoids were used
as tentative substrates. HPLC profiling of reaction products
displayed at least two peaks. Glycosylation positions were
located at the hydroxyl groups at C-3, C-7 or C-4 flavonoid
positions. The most efficient substrate was kaempferol, fol-
lowed by apigenin, genistein and luteolin, in that order.
According to in vitro results and the composition of rice
flavonoids the in vivo substrate of RUGT-5 was predicted
to be kaempferol or apigenin. To our knowledge, this is
the first time that the function of a rice UGT has been
characterized.
Keywords Flavonoids . Glycoslytransferase . Oryza
sativa
Communicated by I. S. Chung
J. H. Ko · B. G. Kim · Y. Lim · J.-H. Ahn (
)
Bio/Molecular Informatics Center, Department of Molecular
Biotechnology, Konkuk University,
Seoul 143-701, South Korea
e-mail: jhahn@konkuk.ac.kr
Tel.: +82-2-45-3764
Fax: +82-2-3437-6106
e-mail: hydekjh@msn.com
e-mail: dkimbk@hanmail.net
e-mail: yoongho@konkuk.ac.kr
H.-G. Hur
Department of Environmental Science and Engineering,
Gwangju Institute of Science and Technology,
Gwangju 500-712, South Korea
e-mail: hghur@gist.ac.kr
Abbreviations IPTG: Isopropyl-β-d-thiogalactopyran-
oside . PCR: Polymerase chain reaction . UGT: Uridine
diphosphate dependent glycosyltransferase
Introduction
Plants produce many types of secondary metabolites, in-
cluding flavonoids, alkaloids and terpenoids (Wink 1999).
These compounds undergo modification reactions such as
methylation, hydroxylation and glycosylation, which lead
to the structural diversity of secondary metabolites (Schwab
2003). Glycosylation is one of the major modification reac-
tions that often occurs in the final step of the natural com-
pound biosynthesis. The primary roles of glycosylation in
plants are: the stabilization of pigments, enhancement of
solubility, storage of secondary metabolites and regulation
of plant growth regulators. Enzymes leading to glycoside
formation known as glycosyltransferases (UGTs), transfer
nucleotide-diphosphate-activated sugars to low molecular
weight substrates. The activated sugar form is typically
UDP-glucose, but UDP-galactose and UDP-rhamnose may
also be found. In plants, sugar acceptors include all major
classes of secondary metabolites, such as phenolics, ter-
penoids, cyanohydrins and alkaloids (Vogt and Jones 2000;
Jones and Vogt 2001; Bowles et al. 2005).
One of the most widely studied classes of plant glyco-
sides is the large and heterogenic group of polyphenols.
To date, an overwhelming number of polyphenolic glyco-
sides including flavonoid glycosides have been identified.
Flavonoids are an important group of polyphenolic natural
products and exhibit a wide range of biological activities
including antioxidant and estrogenic properties (Cornwell
et al. 2004). Moreover, xenobiotics, defined as foreign com-
pounds and man-made chemicals, may also be glycosylated
by the plant (Jones and Vogt 2001). UGTs are so diverse
that Arabidopsis has at least 120 UGTs, half of which are
theoretically involved in secondary plant metabolism (Li
et al. 2001; Ross et al. 2001; Arabidopsis Genome Initia-
tive 2000).
742
According to the CAZY database (http://www. afmb.
cnrs-mrs.fr/CAZY/), UGTs can be classified into 78 fami-
lies on the basis of their substrate specificity and sequence
similarity (Mackenzie et al. 1997). A total of 193 rice gly-
cosyltransferases belong to Family 1, which is known to
be involved in the modification of small compounds such
as plant secondary metabolites and hormones. The total
number of UGTs in rice is expected to outnumber those
found in Arabidopsis. To date, no UGTs from rice have
been functionally characterized. This study represents the
initial step in the functional analysis of rice UGTs and
reports the characterization of RUGT-5, a UGT found in
rice.
Materials and methods
Cloning of RUGT-5
Total RNA from a rice plant was isolated using Qia-
gen plant total RNA isolation (Qiagen, Germany, http://
www.qiagen.com). cDNA was synthesized as described by
Kim et al. (2003). RUGT-5 was cloned by polymerase chain
reaction (PCR) using cDNA as a template. The sequence
5 -gcagcttagctcggagtgac-3 was used as the forward primer,
while 5 -gccacctcatacgtcaacag-3 was used as the reverse
primer. PCR was performed with Hotstart Taq polymerase
(Qiagen, Germany) under the following conditions: 40 cy-
cles of 1 min denaturation at 94◦ C, 1 min annealing at
55◦ C, and 2 min amplification at 72◦ C. The PCR prod-
uct was subcloned into pGEMT-Easy vector (Promega,
USA, http://www.promega.com) and both strands were se-
quenced.
Expression and purification of RUGT-5 in E. coli
To express RUGT-5 in E. coli, the full length cDNA
of RUGT-5 was amplified with the forward primer con-
taining an EcoRI site and the first 20 nucleotides from
the initiation codon of RUGT-5, and with the reverse
primer containing the last 20 nucleotides from the stop
codon. Pfu DNA polymerase (Intron Biotechnology, Korea,
http://www.intron.co.kr) was used for this PCR reaction.
The PCR product was digested using EcoRI and sub-cloned
into pGEX 5X-1 EcoRI/SmaI sites. The resulting construct
was transformed into E. coli BL21(DE3). For the protein
induction, the transformant was grown overnight with shak-
ing in 2 ml LB medium containing ampicillin at 37◦ C. The
culture was then inoculated in 5 ml fresh LB medium con-
taining ampicillin. It was grown at 37◦ C until absorbance at
600 nm reached to 0.8. IPTG (0.1 mM) was then added. The
culture was incubated at 20◦ C with shaking for 20 h. The
resulting cell was harvested by centrifugation, and lysed
by sonification. The soluble protein was used for purifi-
cation with the Glutathione Separose 4B column (Amer-
shambiosciences, USA, http://www.amershambiosciences
.com).
Analysis of reaction products
The reaction mixture for UDP-glycosyltransferase
contained 25 µg of the purified RUGT-5, 10 mM
KH2 PO4 (pH 7.4), 5 mM MgCl2 , 500 µM UDP-glucose,
and 70 µM of substrates. The flavonoids were pur-
chased from Indofinechemicals (New Jersey, USA,
http://www.indofinechemical.com). The reaction mixture
was incubated at 37◦ C for 30 min and extracted twice with
ethylacetate. The ethylacetate was then evaporated com-
pletely. Metabolites were analysed using an Agilent HPLC
(series 100, USA, http://www.agilent.com) equipped with
a photo diode array (PDA) detector and a Waters Symmetry
C18 column (3.5 µm particle size, 4.6 mm × 250 mm,
Milford, MA, USA, http://www.waters.com). For the
analytical scale, the mobile phase consisted of 0.1%
formic acid (pH 3.0) and was programmed as follows:
20% acetonitrile at 0 min, 40% acetonitrile at 10 min, 70%
acetonitrile at 20 min, 90% acetonitrile at 30 min, 90%
acetonitrile at 35 min and 20% acetonitrile at 40 min. The
flow rate was 1 ml/min and UV detection was performed
at 270 nm. Quantification of the metabolites and the
parent material was monitored using HPLC in duplicate
experiments. Several different concentrations of each
substrate were analysed with HPLC, and the resulting
value was used as a standard for the analysis of remaining
reaction product after enzymatic conversion of substrates.
One unit of enzyme was defined as the amount of enzyme
that produced 1 pmole of product in a minute.
Results and discussion
Cloning and expression of RUGT-5
The rice genome was searched with the plant sec-
ondary product UGT consensus sequence (called PSPG
motif; Hughes and Hughes 1994), WAPQVELAAH-
PAVGCFVTHCGWNSTLSESISAGVPMVAWPFFADQ.
In addition, the CAZY database was examined and
193 rice Family 1 GTs were found. Among these, we
cloned 30 UGTs from rice that showed a high degree of
homology with flavonoid glycosyltransferases. Among
the 30 UGT genes studied, one UGT (RUGT-5; GenBank
accession number XM 463383) was further characterized.
RUGT-5 was cloned by RT-PCR and sequenced. RUGT-5
consists of a 1,485 bp open reading frame encoding the
53.5 kDa protein. The predicted protein sequence had
40–42% identity with glycosyltransferases from Catha-
ranthus roseus, Nicotiana tabacum and Dorotheanthus
bellidiformis. A phylogenetic tree of flavonoid UGTs
showed three groups; the first group contained flavonoid
3-O-glycosyltransferase, the second group consisted of
flavonoid 5-O-glucosyltransfrerase and the third group
was composed of UGTs displaying diverse regioselectivity
(Fig. 1). RUGT-5 was most similar to the third group, with
strong correlation for its regioselectivity (see below). In
order to determine the substrate, RUGT-5 was sub-cloned
into a pGEX E. coli expression vector. The recombinant
 743

Zea 3GT
Hordeum 3GT
Gentiana 3GT
Petunia 3GT Forsythia
hi 3GT T
Vigna 3GaT Perilla 3GT
Gentiana 3'GT
Aralia
ia 3GaT
Petunia 3GaT
Dorotheanthus 5GT
Vitis 3GT
Malus 3GT
Citrus GT

Scutellaria 7GT

Brassica
a GT

RUGT-5

Iris 5GT

Solanum GT

Medicago
di GT Nicotiana
a GT

Verbena 5GT Petunia 5GT

0.1
Perilla 5GT Torenia 5GT

Fig. 1 Phylogenetic tree of UDP-dependent glycosyltransferase.
3-O-Glycosyltransferases form one group, 5-O-glycosyltransferases
form second group, and UGTs showing diverse regioselectivity
form third group. GenBank accession numbers of the UGTs:
Medicago GT(AY747627), Solanum GT(STU82367), Scutellaria
7GT(AB031274), Dorotheanthus 5GT(Y18871), Gentiana 3 GT
(AB076697), Hordeum 3GT(X15694), Zea 3GT(AY167672),
Vigna 3GaT(AB009370), Petunia 3GT(AB027454), Gen-
tiana 3GT(D85186), Forsythia 3GT(AF127218), Perilla 3GT
(AB002818), Aralia 3GaT(AB103471), Petunia 3GaT(AF165148),
Vitis 3GT(AB047092), Malus 3GT(AF117267), Citrus GT(AB
033758), Brassica GT(A62529), Iris 5GT(AB113664), Nico-
tiana GT(AB072919), Petunia 5GT(AB027455), Torenia 5GT(AB
076698), Perilla 5GT(AB013596) and Verbena 5GT(AB013598)

RUGT-5 was detected in the soluble fraction of E. coli showed that all substrates tested, except catechol and caf-
lysate and purified as a gluthatione S-transferase (GST) feic acid, produced new reaction product(s). This indicated
fusion protein (Fig. 2). The resulting molecular weight that RUGT-5 might transfer a glucose group to flavonoids
of the purified recombinant RUGT-5 was approximately as well as to isoflavonoids. According to Meβner et al.
79.5 kDa, which is consistent with the combined molecular
weights of GST (26 kDa) and RUGT-5.
Expression of RUGT-5 was investigated using real-time
quantitative reverse transcriptase-polymerase chain reac-
tion by methods described in Kim et al. (2005). RUGT-
5 was expressed in seed, root, stem and leaf tissue, al-
though the expression in leaves was found to be approxi-
mately 3-fold higher than in other tissues (data not shown).
Flavonoids profiling of different rice tissues also showed
that leaves contained higher levels of flavonoids than roots
(unpublished data), consistent with levels observed in Ara-
bidopsis thaliana (Tohge et al. 2005).

Determination of the substrate of RUGT-5

Based on protein sequence homology, RUGT-5 was as-

sumed to use phenylpropanoid compounds as substrates.
Six flavonoids (apigenin, eriodictyol, kaempferol, luteolin,
naringenin and quercetin), two isoflavonoids (daidzein and
genistein) and three other phenolic compounds (catechol, Fig. 2 Expression and purification of the recombinant RUGT-5.
caffeic acid and esculetin) were tested as possible substrates M—molecular weight marker, 1—uninduced E. coli lysate, 2—
for the recombinant RUGT-5 protein. The reaction prod- induced E. coli total lysate, 3—soluble fraction of induced E. coli
ucts were analysed by either TLC or HPLC. TLC analysis lysate, 4—the purified recombinant RUGT-5
744

P1 (2003), Arabidopsis glucosyltransferases that showed ac-

140 A
120 tivity towards endogenous flavonoids also transferred a glu-
100 cose residue into xenobiotics 2,4,5-trichlorophenol (TCP).
80 P2 HPLC analysis of the reaction products revealed at least
60 P3 two peaks different from substrates. For example, reac-
Absorbance (mAU)

40
20 tion products of kaempferol showed three peaks while
0 those of luteolin, quercetin and naringenin showed two
0 5 10 15 (Fig. 3). LC/MS analysis of the reaction products from
B
each flavonoids demonstrated that molecular weight was
175 S1 100
increased by 164 Da compared to the weight of the sub-
Absorbance (mAU)
80
150 strate, which corresponded with the mass of glucose. These
60
125
100 40
P1
results indicated that one glucose molecule was transferred
75 20 S1
into each substrate.
50 0
240 280 320 360 400 Regiospecificity of RUGT-5 was examined by compari-
25 Wave length (nm)

0 son of HPLC retention time and UV spectra of reaction

0 5 10 15 products with the corresponding authentic glycosylated
compounds. In addition, the hypsochromic shift of UV-
C absorbances between substrates and reaction products was
175 P5 examined due to the observation of a hypsochromic shift
150
125 when the glycosylated positions are at either 3 or 4 in
100 the flavone or flavonol. In contrast, glycoslation at the 7-
75 P4
50 hydroxyl group does not demonstrate a hypsochromic shift
Absorbance (mAU)

25 (Vogt et al. 1997; Kramer et al. 2003).

0 The reaction products of kaempferol displayed three
0 5 10 15 peaks (P1–P3 in Fig. 3A). The observed retention time
D
and the UV-spectrum of P1 that appeared as the major
100
120 S2 peak were indistinguishable from those of the authentic
Absorbance (mAU)

80
100 60 3-O-kaempferol glucoside, indicating that P1 is a 3-O-
80 40 P5 kaempferol glucoside. P2 showed a hypsochromic shift,
60 S2

40
20
while P3 did not (Table 1). This suggests that P2 is likely
to be a 4 -O-glucoside of kaempferol and P3 was likely to
0

20 240 280 320 360 400

Wave length (nm)
0 be a 7-O-glucoside of kaempferol. RUGT-5 yielded two
0 5 10 15 products with luteolin (P4 and P5 in Fig. 3C). P5 had a
retention time and UV-spectrum similar to those of the
175 P6 E luteolin 4 -O-glucoside and P4 was tentatively assigned
150
125
100 Table 1 HPLC retention time, absorbance maximum and structure
Absorbance (mAU)

75 P7 assignment of RUGT-5 reaction products

50
25 Compounds Retention UV (nm) Structure assignment
0 time (min)
0 5 10 15 Kaempferol 13.9 266, 366 Kaempferol aglycon
F
P1a 6.7 266, 348 Kaempferol-3-O-glucoside
S3 100

200 P2b 7.2 266, 366 Kaempferol-7-O-glucoside

Absorbance (mAU)

80

150
60 P3b 7.7 266, 362 Kaempferol-4 -O-glucoside
40
P6 Luteolin 11.0 256, 348 Luteolin aglycon
100 20 S3
0 P4a 5.8 256, 348 Luteolin-7-O-glucoside
50 240 280 320 360
Wave length (nm)
400
P5a 7.2 268, 338 Luteolin-4 -O-glucoside
0 Apigenin 13.9 268, 338 Apigenin aglycon
0 5 10 15
P6a 7.4 268, 338 Apigenin-7-O-glucoside
Retention time (min) P7b 7.7 268, 324 Apigenin-4 -O-glucoside
Fig. 3 HPLC profile of reaction products with RUGT-5. Quercetin 11.0 256, 370 Quercetin aglycon
A Kaempferol reaction products. B Authentic kaempferol-3-O- Minorb 5.7 256, 356 Quecetin-3-O-glucoside
glucoside. C Luteolin reaction products. D Authentic luteolin-4 -
O-glucoside. E Apigenin reaction products. F Authentic apigenin Majora 7.3 254, 366 Quercetin-4 -O-glucoside
-7-O-glucoside. Right insets in B, D and F are UV spectra for au- a
These reaction products were assigned according to comparison of
thentic compounds and major reaction products HPLC retention time and UV spectra with authentic compounds
b
These reaction products were assigned according to the hyp-
sochromic shift
 745

Table 2 Substrate specificity Substrate Km (µM) Vmax (pkat/mg) Vmax /Km (pkat mg−l
µM−l) Kcat /Km (µM−l
s−l)
of the purified recombinant
protein RUGT-5 Kaempferol 239.5 1666.7 6.96 0.37
Apigenin 327 2000 6.11 0.32
Genistein 120.7 733.3 6.07 0.32
Enzyme assays were carried out using 25 µg of the purified RUGT-5, 10–200 µM of each substrates and
500 µM of UDP-glucose

as a luteolin 7-O-glucoside with a hypsochromic shift Table 3 Relative activity of RUGT-5 towards several substrates
(Table 1). In the case of quercetin, two reaction products Compound Conversion rate (%) Structure
were also produced: the major product was determined
to be a quercetin 4 -O-glucoside after comparison with
the authentic compound (data not shown) and the minor
Kaempferol 100a
product was likely a quercetin-3-O-glucoside based on
hypsochromic shift (Table 1). Likewise, the two reaction
products of apigenin were determined to be apigenin 7-O-
glucoside (P6) and apigenin 4 -O-glucoside (P7) (Fig. 3E
and 3F; Table 2). Two reaction products from naringenin
were also observed (data not shown). The major reaction Apigenin 95
product was labelled naringenin 7-O-glucoside, based on
comparison of the retention time and UV spectra with the
authentic compound, and the minor product labelled nari-
genin 4 -O-glucoside because narigenin has three hydroxyl
groups at C-5, C-7 and C-4 and narignenin 4 -O-glucoside
Genistein 94
is common in nature. In summary, RUGT-5 transferred a
glucose molecule to the hydroxyl group of either C-3, C-7
or C-4 . When both C-3 and C-4 hydroxyl groups were
present as seen in quercetin and luteolin, 4 -O-glucoside
was a major product of RUGT-5. In case where only the
C-4 hydroxyl group was present (as in kaempferol, narin-
genin and apigenein), the major product was determined by Luteolin 84
the presence of C-3 hydroxyl group. That is, when the C-3
hydroxyl group was present, the C-3 glycosylated product
appeared as the major product, while in the absence of
a C-3 hydroxyl group, the C-7 glycosylated product was
the major product. UGTs from plants also displayed broad
regioselectivity. UGT73G1 from Allium cepa transferred Eriodictyol 80
glucose(s) into hydroxyl groups at the C-3, C-7 and C-4
positions of the quercetin (Kramer et al. 2003). 5-GT from
Dorotheanthus bellidiformis glycosylated quercetin at
either the 4 -hydroxyl or the 7 -hydroxyl group, at a ratio
of 3 to 1, respectively (Vogt et al. 1997).
The kinetic parameters Km and Vmax for apigenin, genis-
Quercetin 70
tein and kaempferol were determined using Lineweaver–
Burk plots (Table 2). According to Kcat /Km ratio that reflects
the enzyme catalytic efficiency, RUGT-5 used kaempferol
most efficiently although a high enzymatic affinity towards
genistein was observed.
The relative activity of RUGT-5 was examined with eight Naringenin 58
different (iso) flavonoids. As shown in Table 3, the best
substrates were keampferol and apigenin, followed by and
genistein.
Several glycosylated phenolic compounds such as
flavonoids and phenylpropanoids were also found in rice. a
100% is equivalent to 500 pkat/mg
Apigenin, kaempferol and luteolin were most likely to be b
100 µM of substrate was used and the reaction mixture was incu-
present, based on UV spectrum and HPLC retention time bated at 37◦ C for 20 min
(unpublished data). These flavonoids exist in glycosylated
forms even though the dominant forms of sugar observed
746
in rice several glycosylated flavonoids are not known. Ac-
cording to in vitro results and the known composition of
flavonoids in rice, the in vivo substrate of RUGT-5 is likely
to be kaempferol. Thus, it is likely that RUGT-5 has a role
in storing kaempferol after conversion to its glycoside.
Currently, several UGTs from plants have been cloned
and characterized. With the completion of the genome
project for Arabidopsis thaliana, UGTs have been stud-
ied extensively due to the availability of several mutant
lines. In contrast, rice whose genome project have been
completed is an important model crop plant whose UGTs
have not been studied. Therefore, RUGT-5 is the first UGT
for rice to have a characterized function.
The glycosylation pattern of flavonoids has an effect on
bioavailability (Aziz et al. 1998), but chemical glycosyla-
tion is difficult and expensive (Arend et al. 2001). Due to its
broad range of substrates and low regioselectivity, RUGT-
5 may be an attractive enzyme for engineering flavonoid
diversity. It will also be interesting to determine if RUGT-5
can glycosylate other structurally-related compounds, in-
cluding antibiotics.
Acknowledgements This work was supported by a grant of Bi-
ogreen 21 Program, Rural Development Administration, Republic of
Korea and in part by KRF2004-F00019 and the R&D Program for
NBT Fusion Strategy of Advanced Technologies (Korea Ministry of
Commerce, Industry and Energy).
References
Arabidopsis Genome Initiative (2000) Analysis of the genome se-
quence of the flowering plant Arabidopsis thaliana. Nature
408:796–815
Arend J, Waszecha H, Hefner T, Stockigt J (2001) Utilizing genet-
ically engineered bacteria to produce plant-specific glucosides.
Biotechnol Bioeng 76:125–131
Aziz AA, Edwards CAM, Lean MEJ, Crozier A (1998) Absorption
and excretion of conjugated flavonols, including quercetin-4 -
O-β-glucoside and isorhamnetin-4 -O-β-glucosides by human
volunteers after the consumption of onions. Free Radic Res
29:257–269
Bowles D, Lsayenkova J, Lim E-K, Poppenberger B (2005) Glyco-
syltransferases: managers of small molecules. Curr Opin Plant
Biol 8:254–263
Cornwell T, Cohick W, Raskin I (2004) Dietary phytoestrogens and
health. Phytochemistry 65:995–1016
Hughes J, Hughes MA (1994) Multiple secondary plant product
UDP-glucose glucosyltransferase genes expressed in cassava
(Manihot esculenta Crantz) cotyledons. DNA Seq 5:41–49
Jones P, Vogt T (2001) Glycosyltransferases in secondary plant
metabolism: tranquilizers and stimulant controllers. Planta
213:164–174
Kim BG, Kim SY, Song HS, Lee C, Hur HG, Kim SI, Ahn J-H
(2003) Cloning and expression of the isoflavone synthase gene
from Trifolium pratense. Mol Cells 15:301–306
Kim BG, Ko JH, Min SY, Ahn J-H (2005) Classification and expres-
sion profiling of putative R2R3 MYB genes in rice. Agric Chem
Biotechnol 48:127–132
Kramer CM, Prata RTN, Willits MG, De Luca V, Steffens JC,
Graser G (2003) Cloning and regiospecificity studies of two
flavonoid glucosyltransferases from Allium cepa. Phytochem-
istry 64:1069–1076.
Li Y, Baldauf S, Lim E-K, Bowles DJ (2001) Phylogenetic analysis
of the UDP-glycosyltransferase multigene family of Arabidopsis
thaliana. J Biol Chem 276:4338–4343
Mackenzie PI, Owens IS, Burchell B, Bock KW, Bairoch A, Be-
langer A, Fournel-Gigleux S, Green M, Hum DW, Iyanagi
T, Lancet D, Louisor P, Magdalow J, Chowdhury JR, Ritter
JK, Schachter H, Tephly TR, Tipton KF, Nebert DW (1997)
The UDP glycosyltransferase gene superfamily: recommended
nomenclature update based on evolutionary divergence. Phar-
macogenetics 7:255–269
Meβner B, Thulke O, Schäffner AR (2003) Arabidopsis glucosyl-
transferases with activities toward both endogenous and xeno-
biotic substrates. Planta 217:138–146
Ross J, Li Y, Lim E-K, Bowles D (2001) Higher plant glycosyltrans-
ferases. Genome Biol 2:3004.1–3004.6
Schwab W (2003) Metabolome diversity: too few genes, too many
metabolites? Phytochemistry 62:837–849
Tohge T, Nishiyama Y, Hirai MY, Yano M, Nakajima J, Awazuhara
M, Inoue E, Takahashi H, Goodenowe DB, Kitayama M, Noji M,
Yamazaki M, Saito K (2005) Functional genomics by integrated
analysis of metabolome and transcriptome of Arabiopsis plants
over-expressing an MYB transcription factor. Plant J 42:218–
235
Vogt T, Zimmermann E, Grimm R, Meyer M, Stack D (1997) Are
the characteristics of betanidin glucosyltransferases from cell-
suspension cultures of Dorotheanthus bellidiformis indicative
of their phylogenetic relationship with flavonoid glucosyltrans-
ferases? Planta 203:349–361
Vogt T, Jones P (2000) Glycosyltransferases in plant natural product
synthesis: characterization of a supergene family. Trends Plant
Sci 5:380–386
Wink N (1999) Biochemistry of plant secondary metabolism. In:
Annual plant reviews, vol 2. CRC Press, Boca Raton