Characterization of a Vitis vinifera cv. Cabernet Sauvignon 3′,5′-O-methyltransferase showing strong preference for anthocyanins and glycosylated flavonols

Article abstract of DOI:10.1016/j.phytochem.2010.05.027

At ripening initiation in red grapevine (Vitis vinifera) berries, the exocarp turns color from green to red and then to purple due to the accumulation and extent of methylation of anthocyanins. The accumulation of transcripts encoding an O-methyltransferase was recently shown to be closely correlated with the onset of ripening and the degree of blue/purple pigmentation in grapevine berries; however, the biochemical function of this gene has remained uncharacterized. In this study, an O-methyltransferase cDNA that showed a distinct expression pattern when compared to closely related sequences was expressed in Escherichia coli and enzyme assays were carried out with a broad array of anthocyanin and other flavonoid substrates. We demonstrate that this enzyme carries out 3′,5′-O-methylation of anthocyanins and flavonol compounds in vitro, which are known to be present in grape berries, with a preference for glycosylated substrates. The highest relative specific activity for the enzyme was found with delphinidin 3-O-glucoside as substrate. The enzyme is not able to methylate flavan type skeletons with chiral centers, such as either catechins or dihydroquercetin. The enzyme showed negligible specific activity for caffeoyl-CoA, compared to flavonol and anthocyanin substrates. Phylogenetic analysis of the O-methyltransferase suggests that it may be a member of a distinct subclass of Type 2 bivalent metal-dependent S-adenosyl-methionine O-methyltransferases.

Full text of DOI:10.1016/j.phytochem.2010.05.027

Phytochemistry 71 (2010) 1474–1484
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Characterization of a Vitis vinifera cv. Cabernet Sauvignon 30,50-O-methyltransferase
showing strong preference for anthocyanins and glycosylated flavonols
Joost Lücker a,1, Stefan Martens b, Steven T. Lund a,
*
a Faculty of Land and Food Systems, Wine Research Centre, University of British Columbia, Vancouver, B.C., Canada V6T 1Z4
b Fondazione Edmund Mach, Istituto Agrario San Michele all’Adige – IASMA, Centro Ricerca e Innovazione, Area Alimentazione, Via E. Mach 1, 38010 San Michele all’Adige (TN), Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
At ripening initiation in red grapevine (Vitis vinifera) berries, the exocarp turns color from green to red
and then to purple due to the accumulation and extent of methylation of anthocyanins. The accumulation
of transcripts encoding an O-methyltransferase was recently shown to be closely correlated with the
onset of ripening and the degree of blue/purple pigmentation in grapevine berries; however, the bio-
chemical function of this gene has remained uncharacterized. In this study, an O-methyltransferase cDNA
that showed a distinct expression pattern when compared to closely related sequences was expressed in
Escherichia coli and enzyme assays were carried out with a broad array of anthocyanin and other flavo-
noid substrates. We demonstrate that this enzyme carries out 30, 50-O-methylation of anthocyanins
and flavonol compounds in vitro, which are known to be present in grape berries, with a preference for
glycosylated substrates. The highest relative specific activity for the enzyme was found with delphinidin
3-O-glucoside as substrate. The enzyme is not able to methylate flavan type skeletons with chiral centers,
such as either catechins or dihydroquercetin. The enzyme showed negligible specific activity for caffeoyl-
CoA, compared to flavonol and anthocyanin substrates. Phylogenetic analysis of the O-methyltransferase
suggests that it may be a member of a distinct subclass of Type 2 bivalent metal-dependent S-adenosyl-
methionine O-methyltransferases.
Received 16 September 2009
Received in revised form 12 May 2010
Available online 30 June 2010
Keywords:
Vitis vinifera
Vitaceae
Grapevine
Enzyme kinetics
Anthocyanin
Flavonol
O-methyltransferase
Delphinidin 3-O-glucoside
Cyanidin 3-O-glucoside
Malvidin 3-O-glucoside
Crown Copyright Ó 2010 Published by Elsevier Ltd. All rights reserved.
1. Introduction
Grape berry growth follows a double sigmoidal curve. The onset
of ripening (termed ‘veraison’ by viticulturists) marks the begin-
Red grapes (Vitis spp.) and their fermentation products are rich
sources of flavonoids that have important organoleptic and health-
promoting properties. Anthocyanins impart red, blue, and purple
coloration to grapes and wines, whereas flavonols in the grape
exocarp (berry skin tissue) contribute to bitter taste, as well as to
the stabilization of wine color through co-pigmentation with
anthocyanins (Boulton, 2001). Certain anthocyanins and flavonoids
were suggested to prevent the occurrence of skin cancer or act as
protectants against blood and arterial diseases (Lila, 2004). Red
wine anthocyanins have been shown to be absorbed by the human
gastrointestinal tract into the circulatory system as intact glyco-
sides (de Pascual-Teresa and Sanchez-Ballesta, 2008; Garcia-
Alonso et al., 2009). The colorful anthocyanins are also important
for the attraction of foraging animals and subsequent seed dis-
persal (Schaefer et al., 2008).
ning of the second stage of growth when berries begin to soften
and increase in size due to rapid cell enlargement. The sugar con-
tent of the berry increases rapidly and acidity decreases. During
ripening initiation, the exocarp loses chlorophyll and begins to
synthesize and accumulate phenolic compounds via the flavonoid
pathway that are responsible for the development of characteristic
colors, yellow-gold (flavonols), red (cyanidin-type anthocyanins),
and blue to purple (delphinidin-type anthocyanins) (Fig. 1) (Wat-
son, 2003). Flavonoid compounds all share the same basic skeleton,
the flavan-nucleus, having two aromatic rings with six carbon
atoms (rings A and B), connected by a heterocycle including three
carbon atoms (ring C) (Fig. 1). Modifications of the central C-ring
determine the specific structural class, such as flavanones, isoflav-
ones, flavones, flavonols, flavanols, and anthocyanins. Vitis spp.
berry exocarp does not contain pelargonidin-type anthocyanins,
but grape cultivars accumulate diverse delphinidin- and cyani-
din-type anthocyanins (Boss et al., 1996b). Flavonoids and antho-
cyanidins with a free hydroxyl group at the 3 position of the
heterocyclic C-ring are unstable under physiological conditions
and generally not found in nature (Forkmann and Heller, 1999).
The enzyme, UDP-glucose:flavonoid 3-O-glucosyltransferase
(3GT, EC 2.4.1.115), efficiently transfers a glucose moiety from
* Corresponding author. Address: 241-2205 East Mall, Vancouver, B.C., Canada
V6T 1Z4. Tel.: +1 604 822 5708; fax: +1 604 822 5143.
E-mail addresses: jlucker@gmail.com (J. Lücker), stefan.martens@iasma.it (S.
Martens), stlund@interchange.ubc.ca (S.T. Lund).
1
Present address: School of Biological Sciences, 427 John Hines Building (62), The
University of Queensland, St. Lucia, Qld 4072, Australia.
0031-9422/$ - see front matter Crown Copyright Ó 2010 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2010.05.027

J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
1475
Fig. 1. The flavonoid biosynthesis pathway and structures of compounds found in V. vinifera berries. Branch pathways to proanthocyanidins as well as additional glycosylated
and acylated anthocyanins and flavonoids are not shown. glc: glucoside.
UDP-glucose to position 3 of the C-ring, promoting stabilization.
The expression of transcripts encoding 3GT coincides with the
accumulation of anthocyanins in the exocarp of red grapes at
ripening initiation (Boss et al., 1996a,b; Ford et al., 1998). Activity
of native UDP-glucose:cyanidin 3-O-glucosyltransferase isolated
from grape cell suspension cultures, as with the heterologously ex-
pressed enzyme, was highest with delphinidin (16) and cyanidin
(15), and much lower with their respective methylated forms (Do
et al., 1995; Ford et al., 1998). O-methylation of anthocyanins at
available hydroxyl groups at the 30 and 50 positions of the B-ring
was shown to be more efficient with cyanidin 3-O-glucoside (18)
as substrate than with the respective aglycone by enzyme assays
using a native O-methyltransferase (OMT) partially purified from
V. vinifera cell cultures (Bailly et al., 1997). It is likely, therefore,
that in V. vinifera, the O-methylation step occurs after the glycosyl-
ation step in the pathway, as shown in Fig. 1.
Expression of a putative transcript for this anthocyanin O-meth-
yltransferase activity was first discovered when a caffeoyl-CoA O-
methyltransferase (cCoAomt)-like gene was found to be one of nine
transcripts consistently up-regulated coincident with the occur-
rence of red pigmentation in berry exocarp (Ageorges et al.,
2006). Subsequently, transcript (Castellarin and Di Gaspero,

1476
J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
2007; Lund et al., 2008) and protein (Lücker et al., 2009) accumu-
lation likely representing the same OMT was also correlated with
the expression of other flavonoid pathway genes or proteins and
the onset of ripening in several V. vinifera cultivars. The omt tran-
script abundance and degree of anthocyanin methylation was
highest in berries of grape cultivars with darker hues (Castellarin
and Di Gaspero, 2007). Despite the availability of these data, the
isolation of a functional full length cDNA for the respective OMT
and its biochemical characterization have not been completed thus
far.
Methyltransferases were recently divided into three different
families based upon a combination of sequence alignments and
structural studies (Noel et al., 2003). The three types show differ-
ences in structural folding. Catalysis is influenced by a different
organization of active site residues. Type 1 methyltransferases gen-
erally methylate hydroxyl moieties of phenylpropanoids, like caf-
feic acid (Fig. 4F, 30). Type 2 methyltransferases are found in all
lignin producing plants and are specific for coenzyme A-deriva-
tized phenylpropanoid compounds. Bacterial macrolide biosynthe-
sis methyltransferases and mammalian catechol OMTs are also
related to these (Noel et al., 2003). Type 3 methyltransferases have
been named the SABATH family of methyltransferases and consist
mainly of enzymes converting carboxylic acids to methyl ester
derivatives (D’Auria et al., 2003; Noel et al., 2003). Type 2 S-aden-
osylmethionine (SAM) and bivalent metal ion cofactor-dependent
enzymes are usually involved in lignin biosynthesis and are spe-
cific for caffeoyl-CoA (Fig. 4H, 32) and 5-hydroxyferuloyl-CoA. Re-
54.9% identity that was proposed to form a second subclass within
the family of Type 2 OMTs (Ibdah et al., 2003; Noel et al., 2003).
In a search for other V. vinifera paralogs that were closely re-
lated to pfomt or faomt in the Pinot Noir genome, three additional
distinct Type 2 cCoAomts were identified. For genomic clone,
CAO65398, with 67.1% identity to pfomt and 51.7% to faomt, an
EST was identified from V. shuttleworthii (CV093856) but no orthol-
ogous ESTs were found from V. vinifera. CAO69794 with 60.8%
identity to pfomt and 48.3% to faomt was found as an EST in Caber-
net Sauvignon petiole tissue (CF517412). For the genomic clone
CAO65399 (65.8% identity to pfomt, 52.5% to faomt), a correspond-
ing Vitis EST could not be retrieved from Genbank and was not
investigated further. For additional identity percentages between
related sequences, see Supplementary material Table S1.
Phylogeny analysis of these sequences, described above, in
combination with representatives of Type 1 OMTs showed that
FAOMT is a Type 2 methyltransferase family member (Fig. 2).
FAOMT falls outside the subclade of the true caffeoyl-CoA methyl-
transferases, containing a characterized Vitis cCoAomt paralog,
Genbank ID: Z54233 (Busam et al., 1997; Ibdah et al., 2003), and
does not form a branch in the second subclade of promiscuous
Type 2 OMTs (Ibdah et al., 2003). Fig. 2 shows that the latter
subclade with PFOMT does include the two newly identified
expressed paralogs from Vitis species.
2.2. FAOMT amino acid alignment
cently,
a
new subclass was shown to have preference for
The conservation between the Vitis FAOMT and other Type 2
flavonoids, as well as these respective phenylpropanoids as sub-
strates (Ibdah et al., 2003). We report here on the identification
and functional characterization of a SAM- and Mg-dependent
OMT activity that prefers flavonol- and anthocyanidin-glycosides
as substrates and performs methylation at the 30 and 50 positions
on the B-ring of the flavan skeleton; however, in contrast to the
previously characterized promiscuous enzymes of the second Type
2 OMT subclass (Ibdah et al., 2003), this Type 2 OMT shows only
negligible activity with caffeoyl-CoA (32) tested as a substrate.
OMTs is shown in an amino acid alignment (Fig. S1). The organiza-
tion and number of the consecutive a-helix and ß-sheet domains is
conserved for FAOMT (Kopycki et al., 2008). The amino acids
responsible for binding the SAM/SAH molecules and the bivalent
metal ion are mostly conserved between FAOMT and the biochem-
ically characterized Type 2 OMTs, Sl-OMT, Mc-PFOMT, At-OMT, Vv-
CCoAOMT, Ms-CCoAOMT, and Os-08g0498100 (rice OMT15). The
most striking differences between FAOMT and other OMTs are that
the FAOMT has a shorter N-terminal region, and a more polar
insertion loop domain. The CCoAOMTs involved in lignin biosyn-
thesis have a conserved Arg-206, which was suggested to interact
with the negatively charged 30-phosphate group of the adenosine
30-50-diphosphate moiety of CoA (Ferrer et al., 2005); this residue
is not present in PFOMT and FAOMT. As was determined by crystal-
lography for iceplant PFOMT and the Medicago sativa CCoAOMT,
the variable N-terminal region (Fig. S1) is likely involved in deter-
mining the position specificity and the kinetic properties of these
enzymes; for PFOMT, in particular, it was postulated that this
may affect the specificity for flavonoid substrates (Ferrer et al.,
2005; Kopycki et al., 2008; Vogt, 2004). The N-terminal region of
FAOMT is shorter and consists of four polar serines followed by
two positively charged amino acids that are not conserved in
PFOMT and SlOMT, but are present in the CCoAOMTs. The insertion
loop domain indicated in Fig. S1 shows that FAOMT has more polar
negatively charged amino acids (D, E) in the middle of this domain
than PFOMT and its related enzymes. The FAOMT insertion loop
domain shows even lower conservation with that of CCoAOMT
from M. sativa than with that of PFOMT. The M. sativa insertion
loop domain contains fewer charged amino acids but has, for in-
stance, three nonpolar neutral prolines that are not conserved in
FAOMT. The amino acid side chains of Tyr-51, Trp-184, and Phe-
198 in PFOMT that were predicted to be involved in sandwiching
the quercetagetin and quercetin (6) aromatic moieties and locating
the hydroxyl groups to be methylated on the flavonoid molecule
closely to the SAM (Kopycki et al., 2008) are conserved, but the
Tyr-51 is replaced by a leucine in FAOMT, similar to the isoleucine
in the M. sativa CCoAOMT, although another Tyr is present three
amino acids upstream from the Leu.
2. Results
2.1. Identification of faomt and relatedness to other sequences by
phylogeny
Correlation of gene and protein expression patterns with the
accumulation of methylated anthocyanins in grape exocarp identi-
fied a predicted omt sequence, Genbank ID: ABQ02272, from V.
vinifera cv. Nebbiolo (Castellarin and Di Gaspero, 2007; Lücker
et al., 2009). A tblastn search of Cabernet Sauvignon berry ex-
pressed sequence tags (ESTs) (Peng et al., 2007) with ABQ02272 re-
vealed a single clone with 100% sequence identity, which had a
predicted 48 amino acid N-terminal extension, as do the most sim-
ilar paralogs identified in the Pinot Noir genome sequence data,
Genbank IDs: CAO66180, CAO66182, and CAO66184 (The French-
Italian Consortium for Grapevine Genome Characterization,
2007). The 1001 bp full length cDNA from Cabernet Sauvignon that
we designated ‘faomt’ for ‘flavonol and anthocyanin 30, 50-O-meth-
yltransferase’, contains an open reading frame of 708 bp which en-
codes a protein of 235 amino acids, having a predicted molecular
mass of 26.4 kDa and a theoretical isoelectric point of 5.84. The
Genbank ID for this nucleotide sequence is HM142924.
The closest ortholog to faomt in the nr database was a Plantago
major cCoAomt (Genbank ID: CAJ43712) with 70.3% identity, thus
far without determined biochemical function. The closest charac-
terized omt to faomt was found to be a Mesembryanthemum crystall-
inum phenylpropanoid and flavonoid omt (pfomt; AY145521) with

J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
1477
Fig. 2. Phylogenetic analysis of FAOMT from V. vinifera with other OMTs. Bootstrap values with a minimum of 50% for maximum likelihood are displayed. Maximum
likelihood values represent percentages of 500
-corrected replicates (log L = ꢀ18792.64) analyzed using Phyml. FAOMT is underlined and displayed in a larger font.
c
Enzymatically characterized Type 2 OMTs are displayed in bold font. Type 1 OMTs (Noel et al., 2003): RhOOMT1 Rosa hybrida orcinol OMT (AF502433), HvF7OMT Hordeum
vulgare flavonoid 7-O-OMT (X77467), PtCOMT1 Pinus taeda caffeic acid OMT (COMT) (U39301), CrOMT2 Catharanthus roseus COMT2 (AY127568), ChaCOMT Chrysosplenium
americanum COMT (U16793), ChaFOMT C. americanum 30, 50 flavonoid OMT (U16794), TsISOMT Thalictrum tuberosum isoquinoline OMT (AF064696), TaOMT2 Triticum
aestivum 30,40,50 flavonoid OMT (DQ223971), MtIOMT4 Medicago truncatula isoflavonoid OMT (DQ419912), MtIOMT7 M. truncatula isoflavonoid OMT (DQ419914),
GeHI40OMT Glycyrrhiza echinata 20,7,40-hydroxyflavonone 40-OMT (AB091684), GeD7OMT G. echinata 7-OMT (AB091685), Ms7OMT Medicago sativa isoflavone 7-OMT
(U97125), PgCOMT Populus ꢁ generosa COMT (M73431), VvCOMT V. vinifera COMT (AF239740), VvCAO69811 V. vinifera PN40024 genomic DNA derived, related to IOMT. Type
2 OMTs: MsCCoAOMT M. sativa (U20736), McCCoAOMT Mesembryanthemum crystallinum CCoAOMT (AF053553), NtCCoAOMT Nicotiana tabacum CCoAOMT (Z82982),
putAtCCoAOMT Arabidopsis thaliana putative CCoAOMT (NM_119566), PtCCoAOMT Populus trichocarpa CCoAOMT (AJ224894), VvCCoAOMT V. vinifera CCoAOMT (Z54233),
ZmCCoAOMT Zea mays CCoAOMT (AJ242980), OsCCoAOMT Oryza sativa cv. japonica CCoAOMT (NM_001063445 Os06g0165800), MvCatOMT Mycobacterium vanbaalenii PYR-
1
(Mvan3280), Pppred1 Physcomitrella patens subsp. Patens predicted protein 1 (XM_001766915), SpSAMMT Schizosaccharomyces pombe SAM-dependent OMT
(NM_001020945), SpCatOMT S. pombe probable catechol OMT (SPBC119), RnCatOMT Rattus norvegius catechol OMT (M60753), HsCatOMT Homo sapiens catechol OMT
(NM_007310), Sc-OMT1 Streptomyces coelicolor A3(2) (SCO2338), Pppred2 P. patens subsp. Patens predicted protein 2 (XM_001754959), Selamoe: Selaginella moellendorfii EST
sequence (FE444786), Os08g498100; O. sativa CCoAOMT ROMT15 (Os08g498100), Os08g498400; O. sativa CCoAOMT ROMT17 (Os08g498400), AtCCoAOMT A. thaliana
CCoAOMT (NM_105468), PtEF146103 P. trichocarpa unknown (EF146103), PkCCoAOMT Populus kitakamiensis CCoA 3-OMT (AB000408), MdcvG Malus x domestica cv. Gala EST
sequence (EB129882), PmCCoAOMT Plantago major (first 22 N terminal amino acids removed from AM159091), AntmCCoAOMT Anthirrinum majus EST sequence (AJ792459),
FAOMT V. vinifera cv. Cabernet Sauvignon flavonoid and anthocyanin methyltransferase, VvCAO66182 V. vinifera PN41004 genomic DNA derived, McPFOMT M. crystallinum
phenylpropanoid and flavonoid OMT (AY145521), SlOMT Stellaria longipes CCoAOMT (L22203), AtOMT A. thaliana CCoAOMT-like (AY087244), VvCAO65398; V. vinifera
PN41004 genomic DNA derived, VvCAO69794; V. vinifera PN41004 genomic DNA derived.
2.3. Analysis of relative expression by real time RT-PCR
real time RT-PCR (Fig. 3). In the exocarp, faomt transcripts began
accumulating at ripening initiation and peaked shortly thereafter,
followed by a gradual decrease as the berry matured (Fig. 3A).
Expression in other tissues was lower, but strongest in tendrils
The four representative genes from the Type 2 omt family were
analyzed for expression in different tissues along ripening using

1478
J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
stages (Fig. 3C). Fig. 3D shows that the CAO65398 transcript was
detectable in flower, root, leaf, and seed.
A
2.4. Enzyme activity assays
The purified recombinant FAOMT (Fig. S2) was used in enzyme
assays with an array of relevant substrates with different func-
tional groups and chirality (Fig. 4). A pH optimum of 8.5 was deter-
mined for FAOMT using cyanidin 3-O-glucoside (18) and quercetin
3-O-glucoside (10) as substrates. FAOMT activity was tested in four
different buffers so that the pH ranged from 2.9 to 9.4. Activity was
detected at pH 6.0 and higher (Fig. S3). The substrate conversion
rate was maximal close to 49 °C, with half maximal conversion rate
around 40 °C, showing a sharp decline at temperatures higher than
51 °C (Fig. S4). The pH optimum curve became sharper as the tem-
perature of the assay was adjusted closer to the point of maximal
conversion rate, and half maximal activity was detected at pH 8
and pH 8.8 (Fig. S5).
The enzyme was dependent on the addition of a metal cofactor
with only traces of product detected in the absence of MgCl2. This
was likely due to traces of bivalent metals in the assay buffer. Opti-
mal activity was seen with MgCl2 used at a concentration of 7 mM.
With the addition of EDTA as a metal chelator to the assay, no
activity was detected (Fig. 5). FAOMT also required the addition
of S-adenosyl-methionine (SAM or AdoMet) as a methyl donor.
B
C
D
C
F
B
E
A
D
Fig. 3. Relative expression of the different Type 2 V. vinifera O-methyltransferases
in grape tissues. Expression is displayed as relative quantity (DDCt) with flower
tissue as the calibrator sample and glyceraldehyde 3-phosphate dehydrogenase as
the reference gene. (A) FAOMT flavonol and anthocyanin O-methyltransferase, (B)
Z54233 Vitis vinifera previously characterized true caffeoyl-CoA O-methyltransfer-
ase (Busam et al., 1997), (C) CAO69794 genomic clone of putative type 2 OMT and
detected as an expressed sequence tag in Cabernet Sauvignon petiole tissue
(CF517412), (D) CAO65398 genomic clone of putative type 2 OMT and detected as
an expressed sequence tag in V. shuttleworthii (CV093856). Flower
DCt for the
different genes are displayed in each panel. Exocarp and mesocarp stages are
defined by growing degree days (GDD), which are calculated by taking the average
of the daily high and low temperature of each day compared to a baseline (10 °C).
Fifty percent ripening initiation in the 2003 season was just prior to 941 GDD.
G
H
and stems. The true cCoAomt from V. vinifera (Z54233) (Busam
et al., 1997) was mainly expressed in the stem and to a lesser ex-
tent in the tendril, root, leaf, and exocarp tissues before ripening
initiation (Fig. 3B). Of the two other V. vinifera genes that grouped
with pfomt in the phylogenetic tree, CAO69794 was expressed
most highly in stem and tendril, and lower in leaf and green berry
Fig. 4. Presentation of structure skeletons tested as substrates for FAOMT. (A)
Anthocyanidin and anthocyanidin-glucosides with different R groups representing
various skeleton side chains. (B) Flavonols and flavonol glucosides,
R groups
representing various skeleton side chains. (C)–(E) Flavonoid skeletons with chiral
centers. (C) dihydroquercetin (7), (D) (+)-catechin (28), (E) (ꢀ)-epi-catechin (29), (F)
caffeic acid (30), (G) 4-hydroxybenzoic acid (31), (H) caffeoyl-CoA (32). glc:
glucoside, rut: rutinoside.

J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
1479
assay conditions using two different lengths of incubation at eight
different temperatures (Fig. S6). Results showed that relative to the
0 °C control, the flavonol glycosides were relatively stable even at
55 °C, whereas the flavonol aglycones were substantially degraded
after 18 min incubation at 55 °C. The cyanidin-derived glycosides
showed higher temperature stability than the delphinidin-derived
glycosides. Methylated anthocyanins were more stable under as-
say conditions than their non-methylated analogs. Delphinidin 3-
O-glucoside (19) was the least stable but was 82% intact after
18 min incubation at 35 °C (Fig. S6).
At 35 °C, FAOMT had the highest relative specific activity with
delphinidin 3-O-glucoside (19) (Table 1). There was no significant
difference in specific activity with cyanidin 3-O-glucoside (18)
and cyanidin 3-O-rutinoside (26). The relative activity measured
with cyanidin 3, 5-di-O-b-glucoside (27), however, was only 55%.
Relative specific activity rates for other substrates are shown in Ta-
ble 1. Myricetin (9) and quercetin (6), which also have three or two
vicinal hydroxyl groups on the B-ring of the flavan molecule, were
methylated as well. Quercetin (6) is converted less efficiently than
its glucoside (10) (Table 1). The anthocyanidins were also con-
verted to the predicted 30- and 50-methylated products, as identi-
fied by their respective authentic standards, but these aglycones
were highly unstable and their relative specific activities could
not be determined (Table 1).
Although the flavonols, flavonol glucosides, and anthocyanins
were all converted to their expected 30- and 30, 50-methylated prod-
ucts by FAOMT, dihydroquercetin (7), caffeic acid (30), and the chi-
ral flavanols, catechin (28) and epi-catechin (29), were not
methylated by this enzyme, not even when using increased protein
amounts of up to 500 ng and longer incubation times up to 30 min
(Fig. 4). Caffeoyl-CoA (32) methylation could not be detected under
the linear conditions for the flavonol and anthocyanin substrates,
but increased protein amounts and incubation time did result in
conversion to feruloyl-CoA. Substrates with vicinal hydroxyl
groups on the B-ring of the flavan structure and lacking chiral cen-
ters were converted by FAOMT. Substrates with vicinal hydroxyl
groups on the A ring were not tested. Pelargonidin 3-O-glucoside
(25) and 4-hydroxybenzoic acid (31), both used as controls to
A
B
C
Fig. 5. LC–DAD chromatograms of multiple substrate conversion by FAOMT.
Absorbance (mAU) is indicated on the left. (A) Ten
treated under assay conditions. (B) Ten M each of delphinidin 3-O-glucoside (19,
D3 g), cyanidin 3-O-glucoside (18, C3 g) and quercetin 3-O-glucoside (10, Q3 g)
added simultaneously as substrates to be converted by FAOMT (350 ng) with
addition of EDTA in the buffer and without MgCl2 added. (C) Same amounts of
substrates added to the assay as in (B), but with addition of MgCl2 and without
EDTA. Assays were incubated at 35 °C for 10 min. Other peaks indicated are: Pt3 g:
petunidin 3-O-glucoside (22), Peo3 g: peonidin 3-O-glucoside (21), M3 g: malvidin
3-O-glucoside (23), Ir3 g: isorhamnetin 3-O-glucoside (14).
lM of authentic standards
l
Enzyme assays were carried out under saturating concentrations of
0.5 mM SAM.
After initial experiments where relative specific activity was
compared for all substrates at 49 °C, we detected possible degrada-
tion of some of the substrates at this temperature in addition to
their methylation, as the resulting loss in substrate did not account
for the increase in product formed. Eleven different flavonoid sub-
strates and products were simultaneously tested for stability under
Table 1
Substrate conversion by FAOMT. Specific activity of FAOMT at 35 °C with 4 l
M cyanidin-3-O-glucoside (18) substrate of 10.25 (SE = 0.73) nkatal mgꢀ1 His-tag purified protein was
set to 100%.
Substrate
Rt (m/z
Rt (m/z product)
Authentic standard product(s) Rt (m/z)
Reaction type
Relative specific
substrate)
activity (std. dev.)
Cyanidin 3-O-
glucoside (18)
Delphinidin 3-O-
glucoside (19)
Petunidin 3-O-
glucoside (21)
Quercetin 3-O-
glucoside (10)
Quercetin (6)
Myricetin (9)
7.7 (449)
6.4 (465)
8.3 (479)
9.64 (463)
Peonidin 3-O-glucoside (21) 9.64 (463)
30-O-methylation
30,50-O-methylation
50-O-methylation
30-O-methylation
100 (12)
169 (18)
52 (5)
8.3 (479), 10.0 (493)
10.0 (493)
Petunidin 3-O-glucoside (22) 8.3 (479),
malvidin 3-O-glucoside (23) 10.0(493)
Malvidin 3-O-glucoside (23) 10.0 (493)
11.64
(465)
12.45 (479)
Isorhamnetin 3-O-glucoside (14), 12.45
(479)
Isorhamnetin 15.15 (317)
Not used
76 (5)
13.8 (303) 15.15 (317)
12.5;(319) 13.9 (333), 15.1 (347)
30-O-methylation
40 (11)
nd
Likely 30, 50-O-methylation
Cyanidin aglycone 10.7 (287) 11.9 (mostly 285, some 301)
Peonidin 11.9 (301)
30-O-methylation, then oxygen nd
(15)
loss in MS
Delphinidin
aglycone (16)
Petunidin
aglycone (24)
Cyanidin 3-O-
rutinoside (26)
Cyanidin-3,5-di-
O-b-glucoside
(27)
10.0 (303) 11.3 (317), 11.75 (349), 12.1
11.3 (317) 11.75 (349), 12.1 (315)
Petunidin (24) 11.3 (317), malvidin 12.1
(331), 12.1 (315, most abundant) (331)
30, 50-O-methylation, water
addition, loss of oxygen in MS
50-O-methylation, water
nd
Malvidin 12.1 (331)
Not used
nd
addition, loss of oxygen in MS
Likely 30-O-methylation
8.3 (595)
5.3 (611)
10.07 (609)
6.94 (625)
119 (11)
55 (9)
Not used
Likely 30-O-methylation
(+)-Catechin (28), caffeic acid (30), (ꢀ)-epi-catechin (29), 4-hydroxybenzoic acid (31), and pelargonidin 3-O-glucoside (25) were not converted by FAOMT. Caffeoyl-CoA (32)
conversion to feruloyl-CoA could not be detected under the above conditions. Dihydroquercetin (7) was not methylated but most likely isomerized, forming a second peak
with identical m/z 303). Rt = retention time in min., m/z = [M]+.

1480
J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
check for possible methylation on the 40 position, remained
unaltered.
substrate binding and site of methylation, as was shown by domain
swapping experiments between a true M. sativa CCoAOMT and the
iceplant PFOMT (Ferrer et al., 2005; Kopycki et al., 2008). For
CCoAOMT, the insertion loop domain was predicted to provide a
scaffold for the CoA moiety of the caffeoyl-CoA (32) during cataly-
sis (Ferrer et al., 2005). The N-terminal region in FAOMT is shorter
than the N-terminal region in PFOMT and in most other OMTs. A
shorter N-terminal region has been suggested to lower the speci-
ficity for caffeoyl-CoA (32) and caffeic acid (30) (Vogt, 2004). Since
the expression pattern for faomt was distinct from that of the true
V. vinifera cCoAomt (Busam et al., 1997), FAOMT is not likely to
function in the conversion of caffeoyl-CoA (32), which is in agree-
ment with the negligible activity found here. The insertion loop do-
main in FAOMT is more polar and negatively charged in relation to
that in PFOMT, perhaps providing a manifold for the glycosidic
forms and positively charged anthocyanins preferred by the
FAOMT enzyme compared to caffeoyl-CoA (32). The flavonol, quer-
cetin 3-O-glucoside (10), was also converted by the iceplant
PFOMT enzyme, but the PFOMT showed higher relative specific
activity, thus higher preference for its aglycone, whereas caf-
feoyl-CoA (32) was preferred by this enzyme in an equal manner
as to quercetin (6) (Ibdah et al., 2003).
Type 1 and Type 2 O-methyltransferases share some substrate
specificities, as both of these families have been shown to convert
flavonoid aglycones in plants (Cacace et al., 2003; Ibdah et al.,
2003). CrOMT2 from Catharanthus roseus, an enzyme belonging
to the Type 1 family, does not accept phenylpropanoids as sub-
strates, only flavonoids. While only non-glycosylated flavonoids
were tested, the enzyme not only could accept the flavonols,
myricetin (9) and quercetin (6), as substrates but also the
dihydroflavonols, dihydroquercetin (7) and dihydromyricetin (8)
(Cacace et al., 2003). These latter ones are compounds with the
B-ring protruding out of the plane in relation to the heterocycle,
connected to the A-ring of the flavan molecule. This is in contrast
to FAOMT which cannot methylate these types of compounds.
Whether this is unique to FAOMT or common to Type 2 OMTs is
not known, since the iceplant PFOMT was not tested for activity
with dihydroquercetin (7) (Ibdah et al., 2003). Although these
authors did detect some activity for the recombinant PFOMT using
another chiral flavonoid, catechin (28), as substrate, this activity
was absent for the native enzyme (Ibdah et al., 2003). The recom-
binant FAOMT enzyme did not show activity with catechins (28,
29), either.
Fig. 5 shows UV-DAD LC chromatograms of assays where three
substrates common to grape berries, delphinidin 3-O-glucoside
(19), cyanidin 3-O-glucoside (18), and quercetin 3-O-glucoside
(10) were presented to the enzyme simultaneously in a competi-
tion assay. The chromatograms showed that the retention times
of product authentic standards were identical to retention times
of the products produced by FAOMT. The m/z values for the parent
ions were also identical (Table 1). From integrations of the peak
areas, it was verified that delphinidin 3-O-glucoside (19) was con-
verted the most in this multi-substrate assay, followed by cyanidin
3-O-glucoside (18), and then quercetin 3-O-glucoside (10).
Enzyme kinetic data were determined using quercetin 3-O-glu-
coside (10) and cyanidin 3-O-glucoside (18) as substrates at 49 °C.
The affinity (Km) of FAOMT for quercetin 3-O-glucoside (10) was 3-
fold greater than for cyanidin 3-O-glucoside (18). The turnover
number (Kcat) for cyanidin 3-O-glucoside (18) was 3-fold higher
than for quercetin 3-O-glucoside (10), resulting in no significant
difference in catalytic efficiency (Table 2). The specific activity at
49 °C using 4 lM cyanidin 3-O-glucoside (18), which was also used
in the substrate efficiency determination experiments (Table 1),
was 28.8 nkatal per mg enzyme. For quercetin 3-O-glucoside
(10), the relative specific activity to cyanidin 3-O-glucoside (18)
was 66.7% at this temperature, which is a similar ratio as what
was found by the experiment conducted at the lower temperature
of 35 °C.
3. Discussion
We conclude that FAOMT methylates flavonol, anthocyanin,
and flavonol-glycoside substrates in the exocarp of grape. Expres-
sion of faomt (Ageorges et al., 2006; Castellarin and Di Gaspero,
2007; Castellarin et al., 2006; Lund et al., 2008) and its protein
(Lücker et al., 2009) are closely temporally and spatially correlated
with the accumulation of these methylated compounds. The two
expressed Type 2 omts present in the grape genome that are more
closely related to pfomt than to faomt showed an expression pat-
tern more like that of the true cCoAomt of V. vinifera than faomt,
and are likely not involved in the methylation of anthocyanins in
grape exocarp.
The high specific activity of FAOMT with the glycosides tested,
the low Kms, and the high catalytic efficiency ðKcatKꢀ1Þ compared
m
Considering the high pH optimum of 8.5 and the minimum pH
of 6.0 for FAOMT activity, the enzyme is likely localized to the cyto-
plasm, possibly associated with the endoplasmic reticulum along
with other components of the flavonoid biosynthetic pathway
(Hrazdina and Jensen, 1992; Poustka et al., 2007; Winkel-Shirley,
1999). The pH of the cytoplasm was shown to be around pH 7 in
other plant species (Felle, 2005). The pH optimum for FAOMT of
8.5 is similar to the pH optimum of the partially purified native
V. vinifera cyanidin 3-O-methyltransferase, although a broader pH
range from 7.75 to 9.75 was reported (Bailly et al., 1997).
with the lower specific activity for the aglycone quercetin (6) pro-
vide further evidence to place FAOMT activity after that of 3GT in
the anthocyanin pathway, supporting the current hypothesis held
for V. vinifera (Bailly et al., 1997; Castellarin and Di Gaspero,
2007; Do et al., 1995; Ford et al., 1998). The highest specific activity
for FAOMT was found for the most labile anthocyanin substrate,
delphinidin 3-O-glucoside (19), which is converted by this enzyme
in two subsequent methylation steps to malvidin 3-O-glucoside
(23) (Jackman and Smith, 1996; Sroka, 2005). It is not surprising,
then, that the more stable 30, 50-methylated malvidin 3-O-gluco-
side (23) is the most abundant anthocyanin in most red V. vinifera
grapes (Mori et al., 2007; Wang et al., 2003).
True Type 2 CCoAOMTs have been shown to convert certain fla-
vonol aglycones such as quercetin (6), although at a much lower
rate than with PFOMT (Ibdah et al., 2003; Lukacin et al., 2004). This
could indicate that the true CCoAOMTs and the more flavonoid-
In Type 2 O-methyltransferases, the N-terminal region and an
insertion loop domain near the C-terminus are important for
Table 2
Kinetic properties of the purified His-tagged FAOMT. Linear assay conditions were performed under optimal temp. of 49 °C.
km (lM)
Vmax (nkatal mgꢀ1
)
Kcat (sꢀ1
)
Kcat/km (Mꢀ1 sꢀ1
)
V at 4 l )
M (nkatal mgꢀ1
Cyanidin 3-O-glucoside (18)
Quercetin 3-O-glucoside (10)
7.60, SE = 0.07
2.70, SE = 0.05
83.38, SE = 4.43
31.17, SE = 1.3
2.3825
0.8976
313,487
332,442
28.8, SE = 0.8
19.16, SE = 0.37
Specific activity at 4
lM substrate is given for comparison to the conversion at 35 °C in Table 1. Parameters were determined for product accumulation and calculated using
calibration curves of products incubated under the same assay conditions.

J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
1481
specific methyltransferases share a common ancestor. It is interest-
ing to note that the more ancient species, Physcomitrella patens and
Selaginella moellendorfii, both possess at least one ortholog to the
Type 2 caffeoyl-CoA methyltransferases, as obtained from EST
database mining. These homologs do not group into any of the spe-
cific subclades, but show relatedness to all subclades of the Type 2
OMTs in the bootstrapped tree. Phylogenetic analysis of the FAOMT
amino acid sequence showed that it grouped together with other
uncharacterized Type 2 plant OMTs in a statistically significant,
distinct group from the subclade containing the PFOMT and other
characterized Type 2 OMTs. We suggest that FAOMT could be the
first characterized enzyme of a third subclass of the Type 2 family
of bivalent metal and SAM-dependent OMTs, based on FAOMT’s
high specific activity with glycosylated flavonols and anthocyanins,
its separate phylogenetic assignment, distinctions in the active site
domains, and its lack of significant activity with caffeoyl-CoA (32).
The two other uncharacterized, low abundance V. vinifera
CCoAOMT-like transcripts that are more closely related to PFOMT
are perhaps more promiscuous like PFOMT and their occurrence
in the grape genome could be an indication that these are ancestral
forms of both the true CCoAOMT and FAOMT. This postulate is
strengthened by the fact that orthologs in the ancient plant species
show resemblance to all Type 2 subclasses in their insertion loop
domain and by phylogenetic analysis.
A biological explanation for the maximal conversion rate of the
enzyme at the high temperature of 49 °C that we determined for
FAOMT may be that elevated ambient temperatures normally
cause degradation of UV-protecting flavonols and anthocyanins,
but methylation in addition to glycosylation can abate degradation,
as we showed using a number of flavonoid standards. Further, the
antioxidant capacity of anthocyanins has been suggested to be
greater at higher temperatures as a means of protecting plants
from excessive temperature injury and UV radiation damage (Shao
et al., 2007). Although total anthocyanin content was reported to
be lower at higher temperature conditions, the malvidin glycoside
content in the grape exocarp was not affected (Mori et al., 2007).
Cooling of sun exposed grapes resulted in increased total anthocy-
anin content of grapes in field experiments (Spayd et al., 2002). In
warm climates, such as in many of the Australian wine growing re-
gions, it was shown for Shiraz berries that heat induced the up-reg-
ulation of the fls gene, presumably underlying the accumulation of
the more UV-protecting quercetin 3-O-glucoside (10) and isorham-
netin 3-O-glucoside (14) close to the end of the season (Downey
et al., 2006, 2003). The abundance of these flavonol glucosides cor-
relating with higher seasonal growing temperatures is in agree-
ment with our finding that quercetin 3-O-glucoside (10) and
isorhamnetin 3-O-glucoside (14) were the most stable compounds
under high temperature conditions. FAOMT activity may have
evolved in plant tissues containing high amounts of flavonoids in
order to increase stabilization against oxidation and heat
degradation.
season. Tendril and stem tissues were collected from vines grown
in a greenhouse at UBC Vancouver. Primers for real time RT-PCR
are given in Table S2. Dissociation curve analyses and sequencing
of products were used to confirm gene specificity for each of the
amplicons (data not shown). Flower was used as the calibrator
sample because this was the tissue with the lowest differences in
D
Ct values among the different genes tested. Efficiencies of all pri-
mer pairs were above 1.8, and data were corrected accordingly.
4.2. Cloning and functional expression of FAOMT
Genbank accession number EC932040 is an EST derived from a
cDNA clone from V. vinifera cv. Cabernet Sauvignon (Peng et al.,
2007). The respective cDNA clone was sequenced in both direc-
tions using M13F and M13R primers and completely overlapping
bidirectional sequence reads were obtained. The full length cDNA
was cloned in-frame with an N-terminal 6x His tag using NdeI
and HindIII restriction enzyme sites of the pET28b + cloning vector
(Novagen, www.emdbiosciences.com) by the sticky end PCR meth-
od (Zeng, 1998). Polymerase chain reaction using the primers given
in Table S2 was carried out with Platinum Pfx DNA polymerase
(www.invitrogen.com) in a RobocyclerÒ 96 gradient cycler (Strata-
gene, La Jolla, CA, USA) with the following parameters: 94 °C for
4 min, followed by 40 cycles at 94 °C for 30 s, 55 °C for 30 s,
68 °C for 1 min. After transformation to E. coli TOP10 competent
cells (www.invitrogen.com) for sequencing purposes, the correct
plasmid was subsequently transformed to BL21DE3 protein
expression competent cells (Novagen). The pET28b + empty vector
was used as a control. After initial growth at 37 °C, IPTG-induced
cultures were transferred to 20 °C and 200 rpm overnight for pro-
tein expression. Bacterial pellets were washed twice using His-trap
column binding buffer (20 mM NaPO4, 500 mM NaCl, 30 mM imid-
azole, pH 7.4, 10% glycerol, CompleteÒ EDTA-free proteinase inhib-
itor (Roche, www.roche.com)). His tag expressed protein was
purified using His spin trap columns eluting the protein in two sep-
arate fractions according to the manufacturers’ protocol (GE
Healthcare, www.gehealthcare.com). The eluted proteins in
20 mM NaPO4, 500 mM NaCl, 500 mM imidazole, pH 7.4 and 10%
glycerol were buffer exchanged and further purified using an Ami-
con Ultra 10 kDa cutoff filter (Millipore, http://www.Milli-
pore.com). The buffer was exchanged to 10% glycerol, 50 mM
Bis–Tris propane, pH 8.5, and then the protein was aliquoted and
stored at ꢀ80 °C. The N-terminus contains a 20 amino acid long
His-tag, which was not removed because previous studies showed
no effect on turnover rate of flavonoids in iceplant PFOMT experi-
ments (Kopycki et al., 2008; Zubieta et al., 2001). Protein concen-
tration was determined at OD280 using the calculated extinction
coefficient for His-tagged FAOMT.
4.3. Enzyme purity analysis
Precast NuPAGE NovexÒ Bis–Tris 4–12% gels were run in 1X
MES buffer on the NuPAGE gel electrophoresis system under
reducing conditions and stained with Simply Blue Safe Stain (Invit-
4. Experimental
rogen). As a size marker, 5 ll Mark 12 protein ladder was used
4.1. Plant material and real time RT-PCR
(Invitrogen). Various enzyme purification stages indicated that
the His-tagged FAOMT was purified nearly to homogeneity, with
a molecular weight of approximately 29 kDa (Fig. S2A).
The isolated His tag protein was also analyzed for purity using
matrix-assisted laser desorption ionization time-of-flight (MAL-
DI-TOF) on a Voyager-DE STR 4311 system (Applied Biosystems,
www.appliedbiosystems.com) (Fig. S2B). After desalting for 3 h
RNA isolation of plant tissues, cDNA synthesis, primer design,
and real time RT-PCR analyses were performed essentially as pre-
viously described (Reid et al., 2006), but using Finnzymes
DyNAmo™ HS SYBRÒ Green qPCR kit (www.Finnzymes.com) and
ABI 7500 software v. 2.0.1 (www.appliedbiosystems.com). V. vinif-
era cv. Cabernet Sauvignon berry samples were collected from a
commercial vineyard near Osoyoos, B.C., and immediately snap
frozen. Exocarp, mesocarp (berry ‘flesh’ tissue), seed, root, leaf,
and flower tissue originated from vines collected during the 2003
using drop dialysis with a 0.025 lm pore 25 mm nitrocellulose
disk (Millipore), the protein was mixed in equal amounts with
sinapinic acid. Settings used were linear mode of operation, posi-
tive polarity, 25,000 V accelerating voltage, 93% grid voltage,

1482
J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
0.15% guide wire, and 1000 nsec extraction delay time. Mass acqui-
sition range was 2000–100,000 Da. 100 laser shots were carried
out to acquire the spectrum with a laser intensity of 2431 at 20 Hz.
www.home.agilent.com) at a rate of 0.85 ml minꢀ1 at 350 bar over
a Zorbax SB-C18 column (4.6 ꢁ 150 mm, 3.5 M) heated at 70 °C
l
(Agilent Technologies). HCO2H at 0.2% was used as solvent A, and
0.2% HCO2H in CH3CN (EMD Chemicals) as solvent B, with the fol-
lowing gradient: 5% solvent B to 14% solvent B after 7 min, then in-
creased to 60% solvent B after 15 min, and kept at 60% solvent B for
another 5 min. After each sample run, the column was set to cali-
brate for 5 min at 5% solvent B. The LC was connected to an MS
ESI set at 350 °C, positive ion scanning mode, nebulizer set at 60
psi, 12 l minꢀ1 N2 gas flow, ion trap scanning between 40 m/z to
700 m/z. The DAD recorded spectra from 190 to 600 nm. Anthocy-
anins and anthocyanidins were analyzed at 520 nm with a band-
width of 10 nm. Flavonoid glucosides and aglycones were
analyzed at 365 nm with a bandwidth of 16 nm. CoAs were sepa-
rated and identified as described using Nucleodur C18ec 250/4
(Macherey & Nagel, Germany) (Ibdah et al., 2003).
4.4. Enzyme assays using the heterologous FAOMT
Most anthocyanin, anthocyanidin, and flavonoid substrates
were purchased from Extrasynthese (www.extrasynthese.com);
petunidin 3-O-glucoside (22) was from Apin Chemicals
(www.apinchemicals.com), CoA esters (32) were from TransMIT
Flavonoidforschung (www.transmit.de), and keracyanin (cyanidin
3-O-rutinoside, 26), SAM, caffeic acid (30), myricetin (9), quercetin
(6), isorhamnetin, (+)-catechin (28), and (ꢀ)-epi-catechin (29)
were from Sigma–Aldrich (www.sigmaaldrich.com). Substrates
were weighed in volumetric flasks and dissolved using sonication
in MeOH-5 mM HCl (1:9). Flavonoid glucosides and dihydroqu-
ercetin (7) were dissolved in MeOH-5 mM HCl (1:1). Quercetin
(6) was dissolved in DMSO, and subsequently diluted to MeOH-
DMSO (1:1). Myricetin (9) was dissolved in MeOH-5 mM HCl
(7:3). Caffeic acid (30), 4-hydroxybenzoate (31), CoA esters (32)
and the catechins (28, 29) were dissolved in H2O. Glycosylated sub-
strates were aliquoted and stored at ꢀ80 °C in the dark for up to
two months.
For the determination of enzyme kinetics, assays were per-
formed under linear and optimized conditions (49 °C, pH 8.5) with
less than 10% substrate conversion for all substrates using 0.7 ng
protein per each 100 ll assay for quercetin 3-O-glucoside (10) or
using 0.875 ng per assay for cyanidin 3-O-glucoside (18) in freshly
prepared solutions. All assays were performed in triplicate with at
least 10 different substrate concentrations per Km experiment,
ranging up to 10 times the calculated Km. Calculations of product
formation were based on standard curves for the respective prod-
ucts, which were determined via assays under identical conditions
as used for the substrates. Peak areas of the optimal DAD spectra
were integrated manually and converted to product concentrations
based on the standard curves. The spreadsheet, Anemona.xlt, was
used for determining Michaelis–Menten kinetics using non-linear
regression for calculation of the curves (Hernandez and Ruiz,
1998).
Enzyme assays were performed in MOPSO, Citrate, MES, or Bis–
Tris propane buffers (Sigma–Aldrich) in their respective pH ranges,
using 33.6 ng His-tag purified protein. The pH for the enzyme anal-
yses ranged from pH 2.9 through pH 9.4, with pH 0.2 intervals be-
tween pH 7.6 and pH 9.1. Substrate concentrations used were
between 0.1 lM and 150 lM. Optimal assays were performed at
pH 8.5 in 50 mM Bis–Tris propane buffer at 49 °C, as tested using
quercetin 3-O-glucoside (10) and cyanidin 3-O-glucoside (18).
HCO2H was used to adjust the pH (Fluka, www.sigmaaldrich.com).
Prior to the assays, a freshly thawed enzyme aliquot was diluted to
an appropriate concentration in 10% glycerol containing BTP buffer
at pH 8.5 on ice. As controls, either His-tag purified and buffer ex-
changed pET28 empty vector for residual bacterial protein activity,
or boiled FAOMT enzyme was used. Temperature optima were
tested between 0 and 65 °C, with 2 degree intervals between 40
and 55 °C, using 84 ng purified protein per assay. Optimum MgCl2
(Sigma–Aldrich) concentration was tested using 0–10 mM MgCl2.
Optimum SAM concentration was determined between 0 and
4.5. Protein alignments and phylogenetic analysis
In order to find orthologs that are closely related to faomt in
plant species that are well known for accumulation of anthocya-
nins or other flavonoids, the non-human non-mouse EST database
(est_others) at NCBI was searched using tblastn. Two full length
omt sequences were assembled from apple and snapdragon ESTs
showing high similarity to faomt (Dong et al., 1998; Martin et al.,
1991).
1000 lM SAM. Assays were checked for linearity between 0 and
60 min and determined to be linear between 5 and 45 min. Tem-
perature stability of substrates was tested under assay conditions
but using eight different temperatures in between 0 °C and 55 °C.
Assays were performed in triplicate for either 18 or 36 min using
To determine sequence identity percentages between Type 2
methyltransferases, the program DNASTAR MegAlign was used,
incorporating the BLOSUM series protein weight matrix with Clu-
stalW alignment and settings of gap penalty at 10 and gap length
penalty at 0.2. Prediction of the secondary structure of the protein
was performed using APSSP2 by online submission of the full
length protein (http://www.imtech.res.in/raghava/apssp2/). The
output displays coil, helix, or strand prediction for each amino acid
and a probability score of 0.5–1.0 (Raghava, 2002). In the align-
ment only the probabilities of helix (alpha helix) or strand (beta
sheet) of 0.6 and higher are indicated.
Representative Types 1 and 2 methyltransferase proteins were
aligned using an anchored dialign online submission with Thresh-
old = 2 as the setting (http://dialign.gobics.de/anchor/submission.
php). The alignment was hand corrected in Genedoc. The aligned
sequences were subsequently submitted to Phyml online (http://
www.atgc-montpellier.fr/phyml/) which searches for the best
tree among five random trees using SPRs and the LG model
of amino acid substitution (Hordijk and Gascuel, 2005; Le and
Gascuel, 2008). The best BioNJ tree was automatically selected.
The proportion of invariant sites was estimated to be 0.022, the
substitution rate number was set at 4, and the gamma distribu-
tion pattern was estimated to be 1.9448. For subsequent maxi-
mum likelihood bootstrapping analyses, the initial results for
all substrates combined, each at a concentration of 5
For substrate specificity, assays were performed under linear
conditions with 8.75 ng purified protein, using 4 M of each sub-
strate and 10 min incubation at 35 °C. These assays were per-
formed in a total volume of 100 l using 50 mM Bis–Tris propane
pH 8.5, 0.1 mg mlꢀ1 BSA, 7 mM MgCl2, 500
M SAM. The substrate
lM.
l
l
l
conversion rate was determined by the amount of substrate lost
after incubation with the enzyme as long as a methylated product
could be determined for each substrate (Yoon et al., 2005). Control
assays with boiled enzymes were carried out simultaneously. Sub-
strates that did not show any conversion of substrate were also
tested with enzyme concentrations of up to 500 ng protein. The
reactions were transferred to ice and stopped by addition of
CHCl3:MeOH–HCl (2% v/v) (400 ll, 2:1), followed by immediate
vortexing for 10 s and subsequent centrifugation at 15,800g for
2 min, resulting in a Folch separation (Folch et al., 1957). The upper
phase (ꢂ200
ll) was collected and transferred to a 2 ml opaque
glass vial with glass insert. LC–MS analysis was performed on an
LC/MSD Trap XCT Plus with an Agilent 1100 Series front-equipped
with
a diode array detector (DAD) (Agilent Technologies,

J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
Felle, H.H., 2005. PH regulation in anoxic plants. Ann. Bot. 96, 519–532.
1483
the proportion of invariable sites and gamma distribution were
set and bootstrapping was done for 500 repetitions (Guindon
and Gascuel, 2003). The results of the bootstrapping were con-
verted to percentages and bootstrap values lower than 50% were
removed.
Ferrer, J.-L., Zubieta, C., Dixon, R.A., Noel, J.P., 2005. Crystal structures of alfalfa
caffeoyl coenzyme A 3-O-methyltransferase. Plant Physiol. 137, 1009–1017.
Folch, J., Lees, M., Stanley, G.H.S., 1957. A simple method for the isolation and
purification of total lipides from animal tissues. J. Biol. Chem. 226, 497–509.
Ford, C.M., Boss, P.K., Hoj, P.B., 1998. Cloning and characterization of Vitis vinifera
UDP-glucose:flavonoid 3-O-glucosyltransferase, a homologue of the enzyme
encoded by the maize bronze-1 locus that may primarily serve to glucosylate
anthocyanidins in vivo. J. Biol. Chem. 273, 9224–9233.
Forkmann, G., Heller, W., 1999. Biosynthesis of flavonoids. In: Barton, D., Nakanishi,
K., Meth-Cohn, O. (Eds.), Comprehensive Natural Products Chemistry, vol. 2.
Elsevier, Oxford, pp. 713–748.
Garcia-Alonso, M., Minihane, A.-M., Rimbach, G., Rivas-Gonzalo, J.C., de Pascual-
Teresa, S., 2009. Red wine anthocyanins are rapidly absorbed in humans and
affect monocyte chemoattractant protein 1 levels and antioxidant capacity of
plasma. J. Nutr. Biochem. 20, 521–529.
Guindon, S., Gascuel, O., 2003. A simple, fast, and accurate algorithm to estimate
large phylogenies by maximum likelihood. Syst. Biol. 52, 696–704.
Hernandez, A., Ruiz, M.T., 1998. An EXCEL template for calculation of enzyme
kinetic parameters by non-linear regression. Bioinformatics 14, 227–228.
Hordijk, W., Gascuel, O., 2005. Improving the efficiency of SPR moves in
phylogenetic tree search methods based on maximum likelihood.
Bioinformatics 21, 4338–4347.
Acknowledgements
We thank Tatiana Pirogovskaia and Merinda Deng for carry-
ing out the RNA isolations. We also thank Lina Madilao for help
in developing the LC–MS methods and Suzanne Perry for the
MALDI-TOF analysis. We gratefully acknowledge funding for this
research from Genome Canada and Genome British Columbia, as
well as additional funding for this work from a Natural Sciences
and Engineering Research Council of Canada (NSERC) Discovery
Grant award to STL. We thank the Province of British Columbia
and the BC wine industry for their continued support of
research activities in the University of British Columbia Wine
Research Centre. Materials described here can be provided upon
request.
Hrazdina, G., Jensen, R.A., 1992. Spatial organization of enzymes in plant metabolic
pathways. Ann. Rev. Plant Phys. 43, 241–267.
Ibdah, M., Zhang, X.-H., Schmidt, J., Vogt, T., 2003. A Novel Mg2+-dependent O-
methyltransferase in the phenylpropanoid metabolism of Mesembryanthemum
crystallinum. J. Biol. Chem. 278, 43961–43972.
Jackman, R.L., Smith, J.L., 1996. Anthocyanins and betalains. In: Hendry, G.A.F.,
Houghton, J.D. (Eds.), Natural Food Colorants. Chapman and Hall, London, pp.
244–309.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.phytochem.2010.05.027.
Kopycki, J.G., Rauh, D., Chumanevich, A.A., Neumann, P., Vogt, T., Stubbs, M.T., 2008.
Biochemical and structural analysis of substrate promiscuity in plant Mg2+
-
dependent O-methyltransferases. J. Mol. Biol. 378, 154–164.
Le, S.Q., Gascuel, O., 2008. An improved general amino acid replacement matrix.
Mol. Biol. Evol. 25, 1307–1320.
References
Lila, M.A., 2004. Anthocyanins and human health: an in vitro investigative
approach. J. Biomed. Biotechnol. 5, 306–313.
Lücker, J., Laszczak, M., Smith, D., Lund, S.T., 2009. Computational approach to the
generation of an improved predicted peptide database from EST data and
application to iTRAQ analyses in grape (Vitis vinifera cv. Cabernet Sauvignon)
berries at ripening initiation. BMC Genomics 10:50.
Lukacin, R., Matern, U., Specker, S., Vogt, T., 2004. Cations modulate the substrate
specificity of bifunctional class I O-methyltransferase from Ammi majus. FEBS
Lett. 577, 367–370.
Lund, S.T., Peng, F.Y., Nayar, T., Reid, K.E., Schlosser, J., 2008. Gene expression
analyses in individual grape (Vitis vinifera L.) berries during ripening initiation
reveal that pigmentation intensity is a valid indicator of developmental staging
within the cluster. Plant Mol. Biol. 68, 301–315.
Martin, C., Prescott, A., Mackay, S., Bartlett, J., Vrijlandt, E., 1991. Control of
anthocyanin biosynthesis in flowers of Antirrhinum majus. Plant J. 1, 37–49.
Mori, K., Goto-Yamamoto, N., Kitayama, M., Hashizume, K., 2007. Loss of
anthocyanins in red-wine grape under high temperature. J. Exp. Bot. 58,
1935–1945.
Noel, J.P., Dixon, R.A., Pichersky, E., Zubieta, C., Ferrer, J.L., 2003. Structural,
functional, and evolutionary basis for methylation of plant small molecules. In:
Romeo, J.T. (Ed.), Recent Advances in Phytochemistry, vol. 37. Elsevier Science,
Oxford, pp. 37–58.
Peng, F.Y., Reid, K.E., Liao, N., Schlosser, J., Lijavetzky, D., Holt, R., Martínez Zapater,
J.M., Jones, S., Marra, M., Bohlmann, J., Lund, S.T., 2007. Generation of ESTs in
Vitis vinifera wine grape (Cabernet Sauvignon) and table grape (Muscat
Hamburg) and discovery of new candidate genes with potential roles in berry
development. Gene 402, 40–50.
Ageorges, A., Fernandez, L., Vialet, S., Merdinoglu, D., Terrier, N., Romieu, C., 2006.
Four specific isogenes of the anthocyanin metabolic pathway are systematically
co-expressed with the red colour of grape berries. Plant Sci. 170, 372–383.
Bailly, C., Cormier, F., Bao Do, C., 1997. Characterization and activities of S-adenosyl-
l-methionine: cyanidin 3-glucoside 30-O-methyltransferase in relation to
anthocyanin accumulation in Vitis vinifera cell suspension cultures. Plant Sci.
122, 81–89.
Boss, P.K., Davies, C., Robinson, S.P., 1996a. Analysis of the expression of
anthocyanin pathway genes in developing Vitis vinifera L. Cv Shiraz grape
berries and the implications for pathway regulation. Plant Physiol. 111, 1059–
1066.
Boss, P.K., Davies, C., Robinson, S.P., 1996b. Anthocyanin composition and
anthocyanin pathway gene expression in grapevine sports differing in berry
skin colour. Aust. J. Grape Wine Res. 2, 163–170.
Boulton, R., 2001. The copigmentation of anthocyanins and its role in the color of
red wine: a critical review. Am. J. Enol. Vitic. 52, 67–87.
Busam, G., Junghanns, K.T., Kneusel, R.E., Kassemeyer, H.H., Matern, U., 1997.
Characterization and expression of caffeoyl-coenzyme A 3-O-methyltransferase
proposed for the induced resistance response of Vitis vinifera L. Plant Physiol.
115, 1039–1048.
Cacace, S., Schroder, G., Wehinger, E., Strack, D., Schmidt, J., Schroder, J., 2003. A
flavonol O-methyltransferase from Catharanthus roseus performing two
sequential methylations. Phytochemistry 62, 127–137.
Castellarin, S., Di Gaspero, G., 2007. Transcriptional control of anthocyanin
biosynthetic genes in extreme phenotypes for berry pigmentation of naturally
occurring grapevines. BMC Plant Biol. 7:46.
Castellarin, S., Di Gaspero, G., Marconi, R., Nonis, A., Peterlunger, E., Paillard, S.,
Adam-Blondon, A.-F., Testolin, R., 2006. Colour variation in red grapevines (Vitis
vinifera L.): genomic organisation, expression of flavonoid 30-hydroxylase,
flavonoid 30,50-hydroxylase genes and related metabolite profiling of red
cyanidin-/blue delphinidin-based anthocyanins in berry skin. BMC Genomics
7:12.
D’Auria, J.C., Chen, F., Pichersky, E., 2003. The SABATH family of methyltransferases
in Arabidopsis thaliana and other plant species. In: Romeo, J.T. (Ed.), Recent
Advances in Phytochemistry, vol. 37. Elsevier Science Ltd., Oxford, pp. 253–283.
de Pascual-Teresa, S., Sanchez-Ballesta, M., 2008. Anthocyanins: from plant to
health. Phytochem. Rev. 7, 281–299.
Do, C.B., Cormier, F., Nicolas, Y., 1995. Isolation and characterization of a UDP-
glucose: cyanidin 3-O-glucosyltransferase from grape cell suspension cultures
(Vitis vinifera L.). Plant Sci. 112, 43–51.
Dong, Y.H., Beuning, L., Davies, K., Mitra, D., Morris, B., Kootstra, A., 1998. Expression
of pigmentation genes and photo-regulation of anthocyanin biosynthesis in
developing Royal Gala apple flowers. Aust. J. Plant Physiol. 25, 245–252.
Downey, M.O., Dokoozlian, N.K., Krstic, M.P., 2006. Cultural practice and
environmental impacts on the flavonoid composition of grapes and wine: a
review of recent research. Am. J. Enol. Vitic. 57, 257–268.
Poustka, F., Irani, N.G., Feller, A., Lu, Y., Pourcel, L., Frame, K., Grotewold, E., 2007. A
trafficking pathway for anthocyanins overlaps with the endoplasmic reticulum-
to-vacuole protein-sorting route in Arabidopsis and contributes to the
formation of vacuolar inclusions. Plant Physiol. 145, 1323–1335.
Raghava, G.P.S., 2002.
A combination method for protein secondary structure
prediction based on neural network and example based learning. CASP V, A-
132.
Reid, K., Olsson, N., Schlosser, J., Peng, F., Lund, S., 2006. An optimized grapevine
RNA isolation procedure and statistical determination of reference genes for
real-time RT-PCR during berry development. BMC Plant Biol. 6:27.
Schaefer, H.M., McGraw, K., Catoni, C., 2008. Birds use fruit colour as honest signal of
dietary antioxidant rewards. Funct. Ecol. 22, 303.
Shao, L., Shu, Z., Sun, S.-L., Peng, C.-L., Wang, X.-J., Lin, Z.-F., 2007. Antioxidation of
anthocyanins in photosynthesis under high temperature stress. J. Integr. Plant
Biol. 49, 1341–1351.
Spayd, S.E., Tarara, J.M., Mee, D.L., Ferguson, J.C., 2002. Separation of sunlight and
temperature effects on the composition of Vitis vinifera cv. Merlot berries. Am. J.
Enol. Vitic. 53, 171–182.
Sroka, Z., 2005. Antioxidative and antiradical properties of plant phenolics. Z.
Naturforsch. (C) 60, 833–843.
Downey, M.O., Harvey, J.S., Robinson, S.P., 2003. Synthesis of flavonols and
expression of flavonol synthase genes in the developing grape berries of
Shiraz and Chardonnay (Vitis vinifera L.). Aust. J. Grape Wine Res. 9, 110–121.
The French–Italian Consortium for Grapevine Genome Characterization, 2007. The
grapevine genome sequence suggests ancestral hexaploidization in major
angiosperm phyla. Nature 449, 463–467.

1484
J. Lücker et al. / Phytochemistry 71 (2010) 1474–1484
Vogt, T., 2004. Regiospecificity and kinetic properties of a plant natural product O-
methyltransferase are determined by its N-terminal domain. FEBS Lett. 561,
159–162.
Yoon, Y., Yi, Y.S., Lee, Y., Kim, S., Kim, B.G., Ahn, J.-H., Lim, Y., 2005. Characterization
of O-methyltransferase ScOMT1 cloned from Streptomyces coelicolor A3(2).
Biochim. Biophys. Acta 1730, 85–95.
Wang, H., Race, E.J., Shrikhande, A.J., 2003. Anthocyanin transformation in Cabernet
Sauvignon wine during aging. J. Agric. Food Chem. 51, 7989–7994.
Watson, B., 2003. Evaluation of wine grape maturity. In: Hellman, E.W. (Ed.), Oregon
Viticulture. Oregon State University Press, Corvallis, OR, pp. 235–245.
Winkel-Shirley, B., 1999. Evidence for enzyme complexes in the phenylpropanoid
and flavonoid pathways. Physiol. Plantarum 107, 142–149.
Zeng, G., 1998. Sticky-end PCR: new method for subcloning. Biotechniques 25, 206–
208.
Zubieta, C., He, X.-Z., Dixon, R.A., Noel, J.P., 2001. Structures of two natural product
methyltransferases reveal the basis for substrate specificity in plant O-
methyltransferases. Nat. Struct. Biol. 8, 271–279.