Plant Cell Physiol. 48(2): 243–251 (2007)
doi:10.1093/pcp/pcl060, available online at www.pcp.oxfordjournals.org
ß The Author 2006. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists.
All rights reserved. For permissions, please email: [email protected]
Perianth Bottom-Specific Blue Color Development in Tulip cv. Murasakizuisho
Requires Ferric Ions
Kazuaki Shoji
Kumi Yoshida
1,
*, Naoko Miki 2, Noriyuki Nakajima 3, Kazumi Momonoi
2,
*
1, 2
, Chiharu Kato
1
and
1
Agricultural Experiment Station, Toyama Agricultural Research Center, Yoshioka, Toyama, 939-8153 Japan
Graduate School of Information Science, Nagoya University, Chikusa, Nagoya, 464-8601 Japan
3
Biotechnology Research Center, Toyama Prefectural University, Kurokawa, Imizu, 939-0398 Japan
2
(Dp) chromophore, but recently bluish flowers have been
produced by modification of flower color using metabolic
engineering techniques to introduce a heterologous Dp
synthesis gene, flavonoid 30 ,50 -hydroxylase (Tanaka et al.
1998, Fukui et al. 2003).
The tulip shows a wide variation in flower color that is
imparted by certain pigments; red, orange, pink and purple
colors are attributed to anthocyanin, and yellow color
to carotenoid (Van Eijk et al. 1987, Nieuwhof et al. 1989,
Nieuwhof et al. 1990). Some investigations revealed
cyanidin (Cy), pelargonidin (Pg) and Dp as the anthocyanidin chromophores in red to purple colored tulips (Shibata
1956, Halevy and Asen 1959, Shibata and Ishikura 1959,
Shibata and Ishikura 1960, Torskangerpoll et al. 1999).
In a recent study using 17 species and 25 cultivars,
Torskangerpoll et al. (2005) summarized that the
tulip anthocyanins comprised five species, namely, 3-O(6-O--rhamnopyranosyl--glucopyranoside) of Dp, Cy
and Pg, and 3-O-[6-O-(2-O-acetyl--rhamnopyranosyl)-glucopyranoside] of Cy and Pg. In spite of the presence
of the Dp chromophore in tulips and a lot of attempts
at breeding, an entirely blue colored tulip has not been
produced so far.
In many tulip cultivars, the color of the inner bottom
of the perianth differs from that of the perianth itself. These
colors of the inner bottom constitute an important feature
to distinguish between tulip cultivars. Tulip cv.
Murasakizuisho has a purple perianth and a distinct blue
color at the bottom (Fig. 1A). Although some tulip
anthocyanins have been analyzed, there are few reports
concerning blue coloration. Since the blue coloration is
observed only at the bottom of the perianth, some physiological difference may exist in distinct regions of the petals.
The mechanism of blue color development in flower
petals has been investigated with considerable interest
because the anthocyanins in vitro show a red or purple
color but not blue under the same weakly acidic condition
as in vivo. Kondo et al. (1992) revealed the structure
of commelinin from blue dayflower, Commelina communis,
to be a metalloanthocyanin responsible for the blue color.
The entire flower of Tulipa gesneriana cv.
Murasakizuisho is purple, except the bottom, which is blue.
To elucidate the mechanism of the different color development
in the same petal, we prepared protoplasts from the purple
and blue epidermal regions and measured the flavonoid
composition by HPLC, the vacuolar pH by a proton-selective
microelectrode, and element contents by the inductively
coupled plasma (ICP) method. Chemical analyses revealed
that the anthocyanin and flavonol compositions in both purple
and blue colored protoplasts were the same; delphinidin 3-Orutinoside (1) and major three flavonol glycosides, manghaslin
(2), rutin (3) and mauritianin (4). The vacuolar pH values of
the purple and blue protoplasts were 5.5 and 5.6, respectively,
without any significant difference. However, the Fe3þ content
in the blue protoplast was 9.5 mM, which was 25 times
higher than that in the purple protoplasts. We could reproduce
the purple solution by mixing 1 with two equimolar
concentrations of flavonol with vismax ¼ 539 nm, which was
identical to that of the purple protoplasts. Furthermore,
addition of Fe3þ to the mixture of 1–4 gave the blue solution
with vismax ¼ 615 nm identical to that of the blue protoplasts.
We have established that Fe3þ is essential for blue color
development in the tulip.
Keywords: Anthocyanin — Blue color — Ferric ion —
Protoplast — Tulip — Vacuolar pH.
Abbreviations: CH3CN, acetonitrile; Cy, cyanidin; DMSO,
dimethylsulfoxide; Dp, delphinidin; Em, membrane potential; ICP,
inductively coupled plasma; LMCT, ligand to metal charge
transfer;
NMR,
nuclear
magnetic
resonance;
MES,
2-(N-morpholino)ethanesulfonic acid; Pg, pelargonidin; TFA,
trifluoroacetic acid.
Introduction
Producing a novel blue flower in tulip (Tulipa
gesneriana) has been a dream-like aim for many breeders
since the 16th century. Carnation and rose do not have
flowers with blue petals due to a lack of the delphinidin
*Corresponding authors: Kazuaki Shoji, E-mail, [email protected]; Fax, þ81-76-429-2701;
Kumi Yoshida, E-mail, [email protected]; Fax, þ81-52-789-5638.
243
244
Blue color development in tulip flower
A
C
B
D
Recently, cyanidin, which is generally red, was shown
to have blue coloration with Fe3þ and Mg2þ in Centaurea
cyanus (Kondo et al. 1994, Kondo et al. 1998, Shiono et al.
2005, Takeda et al. 2005) and in the Himalayan blue poppy
Meconopsis grandis (Yoshida et al. 2006). Yoshida et al.
(1995, 2005) also revealed that the mechanism underlying
blue color development in the morning glory, Ipomoea
tricolor cv. Heavenly Blue, during flower opening was
an increase in vacuolar pH from 6.6 to 7.7. The intra- or
intermolecular stacking of co-pigments such as flavonols
or aromatic organic acids also leads to the blue shift of
coloration (Goto and Kondo 1991, Yoshida et al. 2000).
In order to elucidate the cause of inner bottom-specific
blue color development in some tulips, we prepared colored
protoplasts from the epidermal cell layer and investigated
the composition of anthocyanin and co-pigments, vacuolar
pH and metal ion content. Herein, we show that a specific
accumulation of ferric ions at the bottom of the perianth
induced the blue color development in tulips. The effect of
flavonols on the blue color variation of cells is also
discussed.
E
Fig. 1 Flower color of Tulipa
gesneriana cv. Murasakizuisho
and its colored cells. (A) Purple
perianth and blue perianth bottom.
(B) Transverse section of a petal
shows that the colored cells are
located only on the epidermis. (C)
Purple epidermal cells of the
middle region of the petal. (D)
Chimeric phenotype of the purple
and blue cells at the border region
of the perianth and perianth
bottom. (E) Blue epidermal cells
of the perianth bottom. Bar ¼ 2 cm
in A. Bar ¼ 100 mm in B–D.
contained a mixture of blue and purple cells. To analyze
the vacuolar pH and the components in each type of
colored cell, we prepared protoplasts from blue and purple
petal regions separately. We successfully prepared protoplasts with only a small amount of debris from the purple
and blue epidermal cells (Fig. 2). The mean diameter
of purple protoplasts was 43.3 7.8 mm (n ¼ 100) and that
of blue cells was 46.3 9.4 mm (n ¼ 100).
Results
Reflection spectra of petals and absorption spectra of
protoplasts
We recorded the absorption spectra of both the
purple and blue protoplasts by microspectrophotometry
(Yoshida et al. 2003a, Yoshida et al. 2003b, Yoshida et al.
2006) and compared them with the reflection spectra
of the petals. The reflection spectra of both the purple
and blue regions of the petals showed characteristic
curves with vismax at 544 and 610 nm, respectively
(Fig. 3). The absorption spectra of purple and blue
protoplasts showed the same spectra as those of petals
with vismax at 541 and 615 nm, respectively (Fig. 3). These
results showed that no color change occurred during
protoplast preparation.
Preparation of colored protoplasts
As shown in Fig. 1B, the colored cells are situated only
in the epidermis of the petals. The epidermal cell color at the
bottom of the flowers was blue and the upper part was
purple (Fig. 1C–E). Interestingly, the border between the
purple and blue regions was not a distinct line and
Anthocyanin and co-pigment analyses of colored protoplasts
To reveal the composition of anthocyanins and
co-pigments in the colored protoplasts, the protoplast
suspension was diluted with 10% aqueous acetonitrile
(CH3CN) containing 0.5% trifluoroacetic acid (TFA)
to prepare test solutions (1.7 104 cells ml1), then the
Blue color development in tulip flower
A
245
B
A
1.2
A
2.0
Abs
Fig. 2 Purple and blue protoplasts prepared from the epidermis
of the adaxial perianth and
perianth bottom. Bar ¼ 100 mm.
1.0
1
Purple protoplast
Purple perianth
2
3
Abs
0.8
Anthocyanin
Co-pigment
0.4
4
0.0
0
10
20
30
Rt (min)
0
400
500
600
700
800
B
1.5
Abs
1.0
1
Blue protoplast
Blue perianth
bottom
0.5
Abs
B
Anthocyanin
Co-pigment
1
Wavelength (nm)
2
0.5
3
0.0
0
10
4
20
30
Rt (min)
0
400
500
600
700
800
Wavelength (nm)
Fig. 3 Reflection spectra of petals and absorption spectra of
protoplasts. (A) Reflection spectrum of the purple region of the petal
(vismax ¼ 544 nm) and absorption spectrum of purple protoplast
(vismax ¼ 541 nm). (B) Reflection spectrum of the blue region of the
petal (vismax ¼ 610 nm) and absorption spectrum of blue protoplast
(vismax ¼ 615 nm). No difference was observed between the
reflection spectra of petals and absorption spectra of protoplasts
except the absorbances showing that no color change occurred
during the enzyme treatment for protoplast preparation.
solution was analysed by HPLC. In the HPLC chromatograms, only one peak (1) was detected at 530 nm in both
protoplasts with the same retention time (Fig. 4).
Simultaneous detection at 360 nm revealed that three
major peaks of flavonols (2, 3 and 4) were observed in the
purple and blue protoplasts with some minor peaks. The
Fig. 4 HPLC chromatograms of the sample solution of the purple
(A) and blue (B) protoplasts obtained by a photodiode array
detection HPLC. Dashed lines show the chromatogram detected at
530 nm for anthocyanin and solid lines show that detected at
360 nm for co-pigments. One anthocyanin (1) and three major
co-pigments (2–4) were observed in both purple and blue
protoplasts.
purple and blue protoplasts were similar in composition
and content (Fig. 4).
The four major components (1–4) were purified by
various chromatographic techniques and the structure was
analyzed by mass spectrometry (MS) and nuclear magnetic
resonance (NMR). As a result, 1 was identified to be
delphinidin 3-O-(6-O--rhamnopyranosyl--glucopyranoside) (delphinidin 3-O-rutinoside; Shibata 1956, Shibata and
Ishikura 1959, Nakayama et al. 1999, Fig. 5). The structures
of 2, 3 and 4 were identified to be quercetin 3-O-(2,6-di-O-rhamnopyranosyl--glucopyranoside) (manghaslin; Buttery
and Buzzell 1975, Sakushima et al. 1980), quercetin
3-O-(6-O--rhamnopyranosyl--glucopyranoside)
(rutin;
246
Blue color development in tulip flower
OH
R1
HO
A
+
O
C
B
O
HO
OH
OH
3 O HO
OH
Table 1 Concentration of delphinidin 3-O-rutinoside
(Dp3rut, 1), flavonols (2–4) and elements in colored
protoplasts
OH
OH
OH
OH
O
OH
O
3 O R
2
OH
O
H3C
HO
1
H 3C
HO
O
OH
O
OH
OH
OH
2: R1 = OH, R2 = α-rhamnosyl
3: R1 = OH, R2 = OH
4: R1 = H, R2 = α-rhamnosyl
Fig. 5 Structure of anthocyanin and three major co-pigments
present in the purple and blue protoplasts. 1, delphinidin 3-Orutinoside; 2, manghaslin; 3, rutin; 4, mauritianin.
Vacuolar pH
6.5
Purple protoplast
Blue protoplast
6
5.5
5
4.5
500
Component
O
OH
550
600
Wavelength (nm)
650
Fig. 6 Correlation of cell color and vacuolar pH of purple (open
circles) and blue (closed squares) protoplasts. Cell color is shown
as the vismax of the absorption spectrum.
Budzianowski 1991) and kaempferol 3-O-(2,6-di-O--rhamnopyranosyl--glucopyranoside) (mauritianin; Yasukawa
et al. 1987, El-Sayed 1991), respectively (Fig. 5).
Quantitative analysis of protoplast extracts revealed
the amount of delphinidin 3-O-rutinoside (1) in vacuoles to
be about 10 mM and that of the three flavonols (2–4) to be
approximately 20 mM (Table 1). These results indicated
that blue color development is not attributed to the
compositional difference of anthocyanin and flavonols,
but to other factors, such as vacuolar pH and metal
contents. Although the co-existing flavonol glycosides
may not have a direct effect on blue color development,
they may play some roles in influencing the visible
absorption spectrum of the cell by a co-pigmentation
effect, i.e. a bathochromic effect and stabilization of
the anthocyanin (Asen et al. 1972, Goto and Kondo
1991, Brouillard and Dangles 1994). The involvement of
manghaslin (2) and mauritianin (4) in tulip flowers is not
known, and this is the first report on these flavonols.
Concentration of components (mM)
Purple protoplast
Flavonoid
Dp3rut (1)
Manghaslin (2)
Rutin (3)
Mauritianin (4)
Metal element
Ag
B
Bi
Ca
Cd
Cu
Fe
Li
Mg
Mn
Ni
Zn
Blue protoplast
14.3 0.8
11.2 0.8
8.9 0.6
8.8 1.1
8.8 1.3
8.0 1.2
5.3 1.4
5.7 1.3
50.1
4.5 1.4
0.1 0.1
2.1 0.5
50.1
0.3 0.1
0.4 0.1
0.3 0.3
27.4 3.5
0.3 0.1
0.1 0.1
0.7 0.1
50.1
8.8 2.9
0.2 0.1
6.2 1.0
50.1
0.3 0.1
9.5 0.8
0.8 1.8
56.8 7.3
0.5 0.2
0.2 0.1
1.3 0.3
In 3 ml of the preparation solvent, 5.1 104 cells were dissolved.
The concentration (SE) of Dp3rut (1) and flavonols (2–4) in the
colored protoplasts was determined from peak area by HPLC
analysis and that of metal elements by the ICP method. Values for
a single cell calculated from the cell number and the average cell
volume are shown.
Measurement of vacuolar pH of colored protoplasts
After recording the visible absorption spectra of the
purple and blue colored protoplasts by microspectrophotometry, the vacuolar pH of the cells was measured by
the proton-selective microelectrode method (Yoshida
et al. 2003a, Yoshida et al. 2003b). The mean pH
value of the purple colored protoplasts was 5.5 0.1
(n ¼ 9, Em ¼ 5.6 9.2 mV) and that of blue protoplasts
was 5.6 0.3 (n ¼ 8, Em ¼ 8.8 8.1 mV). This does not
represent a significant difference between the pH values of
purple and blue cells (Fig. 6). Therefore, it was concluded
that vacuolar alkalization was not the reason for the
different color development in tulip petal cells.
Element analysis of the colored protoplasts
Chelation of some divalent and trivalent metal ions
by anthocyanins can lead to blue flower color (Goto and
Kondo 1991, Kondo et al. 1992, Brouillard and Dangles
1994, Kondo et al. 1994, Kondo et al. 1998, Shiono et al.
2005, Andersen 2006). Therefore, we analyzed the composition of elements in purple and blue protoplasts using the
Blue color development in tulip flower
Concentration
(mM)
—
Mg2þ
Ca2þ
Mn2þ
Zn2þ
Fe3þ
Fe3þ
Fe3þ
Fe3þ
Fe3þ
Fe3þ
—
1.0
1.0
1.0
1.0
0.1
0.5
1.0
3.0
5.0
10.0
Color
vismax (nm)
Abs.
P
P
P
P
V
P
V
B
B
B
Br
533
533
533
533
567
534
564
589
588
582
545
1.04
1.07
1.11
1.04
1.27
0.92
0.94
1.01
1.33
1.45
1.21
A stock solution of 1 (20 mM) and each salt solution were mixed in
a buffer solution at pH 5.6 to adjust the final concentration of 1 to
1 mM. The absorption spectrum was measured immediately after
mixing at 258C. The color of the mixture was purple (P), violet (V),
blue (B) or brown (Br).
1.2
0.6
0.3
0
400
B
Reproduction of protoplast colors
Using the purified delphinidin 3-O-rutinoside (1), we
tried to generate protoplast colors in vitro. Because of
practical limitations, the concentration of 1 was 1 mM and
the absorption spectra were recorded in a quartz cuvette
with a 1 mm pathlength. Table 2 shows the color of the
mixture of 1 and various metal ions in a buffer solution at
pH 5.6. Among the detected elements (Mg2þ, Ca2þ, Mn2þ,
Zn2þ and Fe3þ) in colored protoplasts, only Fe3þ gave a
blue solution. Therefore, Fe3þ is considered to play an
important role in blue color development by forming a
chelate compound with 1. However, the spectra of the
mixture of 1 and Fe3þ were not similar to that of
500
600
700
Wavelength (nm)
800
Blue protoplast
Exp. 3
Exp. 4
Exp. 5
Exp. 6
1.6
1.2
0.8
0.4
0
400
inductively coupled plasma (ICP) instrument. The analysis
of 23 elements identified 12 elements, Ag, B, Bi, Ca, Cd, Cu,
Fe, Li, Mg, Mn, Ni and Zn, while the others remained
undetected (Table 1). The concentration of Mg was the
highest in the blue colored cells and those of Fe, B and Ca
were around 10–5 mM. The concentration of the elements in
purple cells was almost half that of those in blue cells with
the exception of Fe. The content of Fe in blue protoplast
was almost 10 mM, which was approximately 25 times
higher than that in purple protoplasts (0.4 mM). These
results strongly suggest the involvement of Fe in the blue
color development. Since there is also a difference in the
concentration of the flavonoids between the purple and blue
cells, there is generally a higher level of ions per flavonoid
molecule in blue cells.
Purple protoplast
Exp. 1
Exp. 2
0.9
Abs
Ion
A
Abs
Table 2 Effect of metal ions on the color development of
delphinidin 3-O-rutinoside (Dp3rut, 1)
247
500
600
700
Wavelength (nm)
800
Fig. 7 Reproduction of the purple and blue petal colors. (A)
Absorption spectra of purple protoplast (vismax ¼ 541 nm), 1
(Exp. 1, vismax ¼ 533 nm) and a mixture of 1 (1 mM) and 3
(2 mM) (Exp. 2, vismax ¼ 539 nm). (B) Absorption spectra of blue
protoplast (vismax ¼ 615 nm), a mixture of 1 (1 mM), 3 (2 mM) and
Fe3þ (1 mM) (Exp. 3, vismax ¼ 612 nm); 1 (1 mM), 2 (2 mM)
and Fe3þ (1 mM) (Exp. 4, vismax ¼ 622 nm); 1 (1 mM), 4
(2 mM) and Fe3þ (1 mM) (Exp. 5, vismax ¼ 542 nm); and 1 (1 mM),
2 (0.8 mM), 3 (0.6 mM), 4 (0.6 mM) and Fe3þ (1 mM) (Exp. 6,
vismax ¼ 615 nm). All the experiments were carried out by mixing
the components in 100 mM MES-Tris buffer at pH 5.6 at 258C and
the absorption spectra of the solutions were measured immediately
(path length: 1.0 mm) after mixing.
protoplasts, indicating that other components must be
essential for development of the same blue color.
Then, three components: delphinidin 3-O-rutinoside
(1), rutin (3) as co-pigment and Fe3þ were mixed. The
visible absorption spectrum of the solution of 1 had
a vismax at 533 nm showing a purple color (Fig. 7A,
Exp. 1). However, the color was unstable and gradually
disappeared. By addition of 3 to the solution
(1 : 3 ¼ 1 mM : 2 mM), the color became purple and stable.
The obtained spectrum showed an identical curve to that
of the purple protoplast with vismax at 539 nm (Fig. 7A,
Exp. 2).
Furthermore, we could reproduce the blue color
identical to that of the blue protoplasts by adding one
equivalent of Fe3þ to the purple solution obtained from
1 and 3 (Fig. 7B, Exp. 3). The solution mixed at a ratio of
1 : 3 : Fe3þ ¼ 1 mM: 2 mM : 1 mM showed the vismax at
248
Blue color development in tulip flower
612 nm with a broad absorption. When manghaslin (2) was
added instead of rutin (3), a similar blue solution was
obtained (vismax ¼ 622 nm) (Fig. 7B, Exp. 4), although with
the addition of mauritian (4, 2 mM) as a flavonol to 1
(1 mM) and Fe3þ (1 mM) the solution did not show a blue
color but a purple color (vismax ¼ 542 nm) (Fig. 7B, Exp. 5).
Finally, we could reproduce the same blue solution
(vismax ¼ 615 nm) as the cells by mixing 1, 2, 3, 4 and
Fe3þ (1 mM, 0.8 mM, 0.6 mM, 0.6 mM and 1 mM, respectively), which is almost the same as the composition in blue
protoplasts, at pH 5.6 (Fig. 7B, Exp. 6). The addition of
Mg2þ (5 mM) to the solutions of Exp. 3 did not show any
color change.
These results indicated that the purple coloration in the
tulip cells is developed by a mixture of 1 and two equivalent
flavonols, and the blue color depends on ferric ion
combined with flavonols, 2–4.
Discussion
In many flowers, different petal colors are developed
by different anthocyanin chromophores. For example, the
chromophore of blue and pink Gentiana petals are Dp
(Goto et al. 1982) and Cy (Hosokawa et al. 1995),
respectively, although the other substitution patterns of
glycosylation and acylation were the same. In the case
of morning glory, blue petals (I. tricolor cv. Heavenly blue)
contained a peonidin nucleus (Kondo et al. 1987) while red
petals (Ipomoea nil) had a Pg nucleus (Lu et al. 1992).
The phenomenon whereby the same anthocyanin shows
different colors, which is common in hydrangea (Hayashi
and Abe 1953, Takeda et al. 1985a, Takeda et al. 1985b,
Takeda et al. 1990, Yoshida et al. 2003b), is very rare in
other plants. We have found an interesting phenomenon
whereby different colors in regions of the same petal
developed with the same anthocyanin and co-pigment
composition in tulip cv. Murasakizuisho, but required
different concentrations of ferric ions, clarifying the
mechanism of blue color development.
First we established the preparation procedure to
obtain pure colored cells from petal epidermis. We analyzed
the composition of anthocyanin and flavonols (Table 1,
Fig. 4) and measured vacuolar pH. The vacuolar pH in
purple and blue cells was similar with no significant
difference (Fig. 6). The composition of anthocyanin
and flavonols of each cell was also similar. We identified
the structure of the anthocyanin as delphinidin 3-O-rutinoside (1) and the three major flavonol glycosides as
manghaslin (2), rutin (3) and mauritianin (4) (Fig. 5).
Rutin was previously reported in tulip flowers (Kawase and
Shibata, 1963); however, the other two flavonols, manghaslin and mauritianin, were identified in this investigation for
the first time.
Metal element analysis of colored cells by the ICP
method revealed that blue cells had a much higher Fe
content than purple cells (Table 1). The in vitro reproduction experiments (Fig. 7) clarified that ferric ion is essential
for development of blue color; the visible absorption
spectrum identical to that of blue cells was obtained by
mixing 1 (1 mM), 2–4 (2 mM in total) and Fe3þ (1 mM)
(Fig. 7B, Exp. 6) at which the ratio of the components is
in agreement with the in vivo ratio. The mixture of 1 and
flavonols gave a stable purple solution, which is the same
color as purple cells, suggesting that two equivalent
flavonols are enough to produce the co-pigment effect.
Interestingly, quercetin glycosides, manghaslin (2) and rutin
(3), showed blue color co-existing with 1 and Fe3þ (Fig. 7B,
Exps. 3, 4); however, a kaempferol glycoside, mauritianin
(4), did not gave blue solution but a purple one (Fig. 7B,
Exp. 5). These results strongly suggest that the substitution
pattern of the B-ring of flavonols also affects the blue
coloration. In these experiments, the concentration
of Mg2þ, which is the most abundant element in cells,
did not influence the absorption spectra. In blue colored
vacuoles of tulip petals, Fe3þ may chelate to the B-ring
of the Dp nucleus and the B-ring of flavonols may also
be complexed with Fe3þ. The chromophore of 1 and the
flavonol nucleus of 2 and 3, with ferric ion, may be
stacked together to develop the stable blue color, which
may due to LMCT (ligand to metal charge transfer;
Kondo et al. 1998).
In many tulip cultivars, the bottom of the perianth
shows different coloration from that of the perianth,
e.g. white (perianth) and blue (bottom), purple (perianth)
and yellow (bottom) or red (perianth) and black (bottom).
These phenotypes may be caused by different expression
of pigment synthetic genes between the perianth and
perianth bottom. In the present study, the quantitative
analysis of metal elements revealed that the Fe content of
blue cells was approximately 10 mM, being 25 times higher
than that in purple cells. Furthermore, the concentration of
other metals in blue cells was twice or more compared with
that in purple cells. These results suggest the presence of
an inner bottom-specific cation transport and/or storage
system. Specific ion storage may have some role in circadian
flower opening and closing of tulip (Azad et al. 2004).
The molar equivalent of Fe to anthocyanin in the blue
protoplasts was approximately 1 eq. (Table 1). This ratio is
essential for blue color development because when
the concentration of Fe3þ was lower than that of the
anthocyanin, the color shifted to purple, while an excessive
amount of Fe3þ led to the loss of blue color (Table 2).
Therefore, tulip may regulate the concentration of ferric ion
in relation to that of anthocyanin strictly by some unknown
mechanism.
Blue color development in tulip flower
In conclusion, the different coloration of purple
perianth and blue perianth bottom of the petals of tulip
cv. Murasakizuisho is attributed only to the difference in
Fe3þ concentration. The lower concentration of Fe3þ gave
purple cells and the higher concentration of Fe3þ gave blue
cells, in which delphinidin 3-O-rutinoside (1) and quercetin
glycosides (2 and 3) may chelate to ferric ion and the two
aromatic rings stack intermolecularly to inhibit hydration
to the anthocyanidin chromophore (Goto and Kondo
1991). Chimeric phenotypes of purple and blue cells at the
border regions (Fig. 1D) and intensity difference in blue
cells (Fig. 1E) might be caused by the different content of
Fe3þ and/or 2–4 in each cell (Fig. 7).
Materials and Methods
Plant material and growth conditions
Tulip plants (T. gesneriana cv. Murasakizuisho) were grown
in a greenhouse at 208C/158C (day/night) under a 12 h light/12 h
dark cycle. The plants were grown in individual pots filled with
standard horticultural soil. The flowers were used immediately
after anthesis for the preparation of protoplasts. For isolation of
anthocyanin and co-pigments, the flowers were stored at 208C
until extraction.
Microscopic observation of transverse section
Fresh petals were cut using a razor blade into approximately
10 10 mm squares; the petal was then placed between the two
halves of a split elder stem. The pith was set in a plant microtome
(MT-3, Nippon Ikakikai, Tokyo, Japan) and 50 mm sections
were cut, placed on glass slides, and the transverse section was
observed under a microscope (BX50WI, Olympus, Tokyo, Japan).
The epidermal cells were also observed after removing them from
the petals by using tweezers.
Preparation of protoplasts from colored epidermis
Fresh petals were cut using a razor and then sorted into
purple and blue segments. From each segment, epidermal tissues
were removed using tweezers and dipped into a buffer [20 mM
MES-Tris (pH 5.8), 0.6 M mannitol] containing 2.0% (w/v)
Sumizyme C (Shin Nihon Chemical, Anjo, Japan) and 0.2%
(w/v) Macerozyme R-10 (Yakult Honsha, Osaka, Japan). The
suspension was evacuated for 2 min, incubated at 308C for 3 h,
filtered through a nylon mesh (50 mm), washed with buffer [20 mM
MES-Tris (pH 5.8), 0.6 M mannitol] and centrifuged (70g)
for 5 min. This washing process was repeated three times.
All treatments after filtration were conducted at 48C.
249
et al. 2003a, Yoshida et al. 2003b). All data having a drift of55 mV
5 s1 were adopted.
HPLC analysis of anthocyanin and co-pigments
The protoplasts from purple and blue petal parts were
suspended in a solution at a concentration of 5.1 104 cells ml1.
After precipitating the protoplasts by centrifugation (100g,
3 min) and washing with buffer, the precipitated cells were
suspended in solvent (A): 10% aqueous CH3CN containing 0.5%
TFA at 1.7 104 cells ml1. Analytical HPLC with a photodiode
array detector (Alliance 2695, Waters, Milford, MA, USA) was
performed on a Mightysil RP-18 GP Aqua column (Kanto
Chemical, Tokyo, Japan; 150 4.6 mm, 5 mm) at a flow rate of
1.0 ml min1 at 408C. The column was eluted with a linear gradient
elution from 100% solvent (A) to 100% solvent (B) (30% aqueous
CH3CN containing 0.5% TFA) in 30 min. For anthocyanin
detection, a wavelength of 530 nm, and for co-pigments a
wavelength of 360 nm was adopted.
Purification of anthocyanin and co-pigments
Flavonoids of petals of T. gesneriana cv. Murasakizuisho
(500 g) were extracted with 800 ml of solvent (A), and the extract
was applied onto an Amberlite XAD-7 column (Mori et al. 2006).
After washing with 10 vols. of the same solution, anthocyanin
and the co-pigment were eluted using solvent (B). The eluent
was freeze-dried and dissolved in a small amount of solvent (A).
Further, HPLC purification was carried out on a Mightysil RP-18
GP column (Kanto Chemical Co., Tokyo, Japan; 250 10 mm,
5 mm) by using solvents (A) and (B). Elution was performed using
a linear gradient of 0 to 100% B in 30 min at a flow rate of
2.0 ml min1 at 258C. Each peak was collected by a fraction
collector and freeze-dried. These purification procedures were
repeated until a single peak was obtained to give pure 1 as
a dark red TFA salt (195.1 mg), and 2 (16.9 mg), 3 (4.6 mg) and
4 (22.4 mg) as a colorless mass.
Structural analysis by NMR and MS
1
H and 13C NMR spectra were measured using a JEOL
JNMLA400 spectrometer at room temperature with approximately
2 mg of the samples that were dissolved in TFAd-CD3OD for 1
and CD3OD for 2–4. The MS spectra were obtained using an
LC/MS–MS system [Agilent 1100/QTrap system, ESI(þ), Applied
Biosystems, Tokyo, Japan].
Measurement of the reflection spectrum of petals and visible
absorption spectrum of protoplasts
The reflection spectrum of fresh petals was recorded using an
integral sphere apparatus and the visible absorption spectrum
of protoplasts with a microspectrophotomeric system as described
by Yoshida et al. (2003a, 2003b, 2006).
ICP analysis
The protoplast suspension adjusted to a concentration of
5.1 104 cells ml1 was centrifuged (100g, 3 min) and washed
with buffer. To the precipitated protoplasts, 1% (v/v) aqueous
nitric acid (metal analysis grade) was added at a ratio of
1.7 104 cells ml1. After filtration, the sample solution was
analyzed using an ICP instrument (Vista-Pro, Seiko Instruments/
Varian Instruments). The plastic tubes used in this experiment were
dipped in 5% (v/v) aqueous nitric acid for several days and rinsed
three times with ultrapure water. For quantitative analysis, the ICP
multi-element standard solution IV (Merck, Germany) containing
23 elements (Ag, Al, B, Ba, Bi, Ca, Cd, Co, Cr, Cu, Fe, Ga, In, K,
Li, Mg, Mn, Na, Ni, Pb, Sr, Tl and Zn) was used as the standard.
Preparation of proton-selective microelectrodes and measurement of
vacuolar pH
Proton-selective microelectrodes were prepared and the
vacuolar pH was measured as described previously (Yoshida
Reproduction of petal color from the cell components
Stock solutions of 1 (20 mM) and metal ion solutions [20 mM
MgCl2, CaCl2, MnCl2, ZnCl2 and NH4Fe(SO4)2; Wako Chemicals,
Osaka, Japan] were prepared by dissolving them in a buffer
250
Blue color development in tulip flower
(100 mM MES-Tris, pH 5.6). Each stock solution of 2–4 (20 mM)
was prepared by dissolving it in dimethylsulfoxide (DMSO).
Compound 1 and/or flavonols (2–4) and/or each metal ion was
mixed in the buffer solution (100 mM MES-Tris, pH 5.6) in a total
volume of 200 ml to adjust the final concentration of 1 to 1 mM.
The visible absorption spectrum of the mixture was immediately
measured by spectrophotometry (JASCO V-560) in a quartz
cuvette (path length: 1.0 mm) at 258C.
Acknowledgments
We are grateful to Professor Tadao Kondo (Nagoya
University) for valuable advice and Mrs. Michie Sakai for her
assistance in protoplast preparation. We thank Shin Nihon
Chemical Co. Ltd. for providing Sumizyme C. This work was
financially supported by the Ministry of Education, Culture,
Sports, Science and Technology, Japan [(B) No. 16370021, The
21st Century COE Program No.14COEB01-00, Creative Scientific
Research No. 16GS0206 and Priority Areas No. 17035041].
References
Andersen, O.M. and Jordheim, M. (2006) The anthocyanins. In Flavonoids
Chemistry, Biochemistry and Applications. Edited by Andersen, O.M.
and Markham, K.R. pp. 471–551. CRC Press, Boca Raton, FL.
Asen, S., Stewart, R.N. and Norris, K.H. (1972) Co-pigmentation of
anthocyanins in plant tissues and its effect on color. Phytochemistry 11:
1139–1144.
Azad, A.K., Sawa, Y., Ishikawa, T. and Shibata, H. (2004) Phosphorylation
of plasma membrane aquaporin regulates temperature-dependent opening of tulip petals. Plant Cell Phisiol. 45: 608–617.
Brouillard, R. and Dangles, O. (1994) Flavonoids and flower colour. In The
Flavonoids: Advances in Research Since 1986. Edited by Harborne, J.B.
pp. 565–588. Chapman & Hall, London.
Budzianowski, J. (1991) Six flavonol glucuronides from Tulipa gesneriana.
Phytochemistry 30: 1679–1692.
Buttery, B.R. and Buzzell, R.I. (1975) Soybean flavonol glycosides:
identification and biochemical genetics. Can. J. Bot. 53: 219–224.
El-Sayed, N.H. (1991) A rare kaempferol trisaccharide anti-tumor promoter
from Sesbania sesban. Pharmazie 46: 679–680.
Fukui, Y., Tanaka, Y., Kusumi, T., Iwashita, T. and Nomoto, K. (2003) A
rationale for the shift in colour towards blue in transgenic carnation
flowers expressing the flavonoid 30 , 50 -hydroxylase gene. Phytochemistry
63: 15–23.
Goto, T. and Kondo, T. (1991) Structure and molecular stacking of
anthocyanins—flower color variation. Angew. Chem. Int. Ed. Engl. 30:
17–33.
Goto, T., Kondo, T., Tamura, H., Imagawa, H., Iino, A. and Takeda, K.
(1982) Structure of gentidelphin, an acylated anthocyanin isolated from
Gentiana makinoi, that is stable in diluted aqueous solution. Tetrahedron
Lett. 23: 3695–3698.
Hayashi, K. and Abe, Y. (1953) Studien uber anthocyane. XXIII. Papierchromatographische ubersicht der anthocyane im pflanzeneich. I. Misc.
Rep. Res. Inst. Nat. Resour. 29: 1–8.
Halevy, A.H. and Asen, S. (1959) Identification of the anthocyanins in
petals of tulip varieties Smiling Queen and Pride of Haarlem. Plant
Physiol. 34: 494–499.
Hosokawa, K., Fukushi, E., Kawabata, J., Fujii, C., Ito, T. and
Yamamura, S. (1995) Three acylated cyanidin glucoside in pink flowers
of gentiana. Phytochemistry 40: 941–944.
Kawase, Y. and Shibata, M. (1963) A glycoside of flavonol (rutin) isolated
from tulip flowers. Bot. Mag. Tokyo 76: 89–91.
Kondo, T., Kawai, T., Tamura, H. and Goto, T. (1987) Heavenly blue
anthocyanin IV. Structure determination of heavenly blue anthocyanin,
a complex monomeric anthocyanin from the morning glory Ipomoea
tricolor, by means of the negative NOE method. Tetrahedron Lett. 28:
2273–2276.
Kondo, T., Ueda, M. and Isobe, M. (1998) A new molecular mechanism
of blue color development with protocyanin, a supramolecular
pigment from cornflower, Centaurea cyanus. Tetrahedron Lett. 49:
8307–8310.
Kondo, T., Ueda, M., Tamura, H., Yoshida, K., Isobe, M. and Goto, T.
(1994) Composition of protocyanin, a self-assembled supramolecular
pigment form the blue cornflower Centaurea cyanus. Angew. Chem. Int.
Ed. Engl. 33: 978–979.
Kondo, T., Yoshida, K., Nakagawa, A., Kawai, T., Tamura, H. and
Goto, T. (1992) Structural basis of blue-colour development in flower
petals from Commelina communis. Nature 358: 515–518.
Lu, T.S., Saito, N., Yokoi, M., Shigihara, A. and Hondo, T. (1992)
Acylated pelargonidin glycosides in the red-purple flowers of Pharbitis
nil. Phytochemistry 31: 289–295.
Mori, M., Kondo, T., Toki, K. and Yoshida, K. (2006) Structure of
anthocyanin from the blue petals of Phacelia campanularia and its blue
flower color development. Phytochemistry 67: 622–629.
Nakayama, M., Yamaguchi, M., Urashima, O., Kan, Y., Fukui, Y.,
Yamaguchi, Y. and Koshioka, M. (1999) Anthocyanins in the dark
purple anthers of Tulipa gesneriana: Identification of two novel
delphinidin 3-O-(6-O-(acetyl--rhamnopyranosyl)--glucopyranosides).
Biosci. Biotechnol. Biochem. 63: 1509–1511.
Nieuwhof, M., Van Eijik, J.P. and Eikelboom, W. (1989) Relation between
flower colour and pigment composition of tulip (Tulipa L.). Neth. J.
Agric. Sci. 37: 365–370.
Nieuwhof, M., van Raamsdonk, L.W.D. and van Eijk, J.P. (1990) Pigment
composition of flowers of Tulipa species as a parameter for biosystematic
research. Biochem. Syst. Ecol. 18: 399–404.
Sakushima, A., Hisada, S. and Ogihara, Y. (1980) Studies on the
constituents of apocynaceae plants. Gas chromatography–mass spectrometric determination of flavonoid triglycosides from the leaves of
Cerbera manghas L. (2). Chem. Pharm. Bull. 28: 1219–1223.
Shibata, M. (1956) Uber das Anthocyanin in den dunkelpurpurnen Bluten
von Tulipa gesneriana L. Studien uber die Phisiologie von Liliaceen I.
Bot. Mag. Tokyo 69: 462–468.
Shibata, M. and Ishikura, N. (1959) On anthocyanins in tulip flowers.
Naturwissenschaften 46: 601–602.
Shibata, M. and Ishikura, N. (1960) Paper chromatographic survey of
anthocyanin in tulip flowers. I. Jpn. J. Bot. 17: 230–238.
Shiono, M., Matsugaki, N. and Takeda, K. (2005) Phytochemistry:
structure of the blue cornflower pigment. Nature 436: 791.
Takeda, K., Kariuda, M. and Ito, H. (1985a) Blueing of sepal colour of
Hydragea macrophylla. Phytochemistry 24: 2251–2254.
Takeda, K., Kubota, R. and Yagioka, C. (1985b) Copigments in
bluing of sepal color of Hydrangea macrophylla. Phytochemistry 24:
1204–1209.
Takeda, K., Osakabe, A., Saito, S., Furuyama, D., Tomita, A., Kojima, Y.,
Yamada, M. and Sakuta, M. (2005) Components of protocyanin, a blue
pigment from the blue flowers of Centaurea cyanus. Phyotochemistry 66:
1607–1613.
Takeda, K., Yamashita, T., Takahashi, A. and Timberlake, C.F. (1990)
Stable blue complexes of anthocyanin–aluminium-3-p-coumaroyl- or
3-caffeoyl-quinic acid involved in the bluing of Hydrangea flower.
Phytochemistry 29: 1089–1091.
Tanaka, Y., Tsuda, S. and Kusumi, T. (1998) Metabolic engineering to
modify flower color. Plant Cell Physiol. 39: 1119–1126.
Torskangerpoll, K., Fossen, T. and Andersen, O.M. (1999) Anthocyanin
pigments of tulips. Phytochemistry 52: 1687–1692.
Torskangerpoll, K., Norbaek, R., Nodland, E., Ovstedal, D.O. and
Andersen, M. (2005) Anthocyanin content of Tulipa species and cultivars
and its impact on tepal colours. Biochem. Syst. Ecol. 33: 499–510.
Van Eijk, J.P., Nieuwhof, M., van Keulen, H.A. and Keijzer, P.
(1987) Flower colour analyses in tulip (Tulipa L.). The
occurrence of carotenoids and flavonoids in tulip tepals.
Euphytica 36: 855–862.
Yasukawa, K. and Takido, M. (1987) A flavonol glycoside from Lysimachia
mauritiana. Phytochemistry 26: 1224–1226.
Blue color development in tulip flower
Yoshida, K., Kawachi, M., Mori, M., Maeshima, M., Kondo, M.,
Nishimura, M. and Kondo, T. (2005) The involvement of tonoplast
proton pumps and Naþ(Kþ)/Hþ exchangers in the change of petal color
during flower opening of morning glory, Ipomoea tricolor cv. Heavenly
Blue. Plant Cell Physiol. 46: 407–415.
Yoshida, K., Kitahara, S., Ito, D. and Kondo, T. (2006) Ferric ions
involved in the flower color development of the Himalayan blue poppy,
Meconopsis grandis. Phytochemistry 67: 992–998.
Yoshida, K., Kondo, T., Okazaki, Y. and Katou, K. (1995) Cause of blue
petal colour. Nature 373: 291.
251
Yoshida, K., Osanai, M. and Kondo, T. (2003a) Mechanism of dusky
reddish-brown ‘kaki’ color development of Japanese morning glory,
Ipomoea nil cv. Danjuro. Phytochemistry 63: 721–726.
Yoshida, K., Toyama, Y., Kameda, K. and Kondo, T. (2000) Contribution
of each caffeoyl residue of the pigment molecule of gentiodelphin to blue
color development. Phytochemistry 54: 85–92.
Yoshida, K., Toyama-Kato, Y., Kameda, K. and Kondo, T. (2003b)
Sepal color variation of Hydrangea macrophylla and vacuolar pH
measured with a proton-selective microelectrode. Plant Cell Physiol. 44:
262–268.
(Received November 7, 2006; Accepted December 12, 2006)