Asymmetric total synthesis of 6-tuliposide B and its biological activities against tulip pathogenic fungi

Article abstract of DOI:10.1271/bbb.100845

The structure-activity relationship was investigated to evaluate the antifungal activities of tuliposides and tulipalins against tulip pathogenic fungi. 6-Tuliposide B was effectively synthesized via the asymmetric Baylis-Hillman reaction. Tuliposides and

Full text of DOI:10.1271/bbb.100845

Biosci. Biotechnol. Biochem., 75 (4), 718–722, 2011
Asymmetric Total Synthesis of 6-Tuliposide B and Its Biological Activities
against Tulip Pathogenic Fungi
y
1;
Kengo SHIGETOMI,1 Shoko OMOTO,1 Yasuo KATO,2 and Makoto UBUKATA
1Division of Applied Bioscience, Graduate School of Agriculture, Hokkaido University, Sapporo 060-8589, Japan
2Department of Biotechnology, Faculty of Engineering, Toyama Prefectural University,
5180 Kurokawa, Imizu, Toyama 939-0398, Japan
Received November 30, 2010; Accepted January 20, 2011; Online Publication, April 22, 2011
[doi:10.1271/bbb.100845]
The structure-activity relationship was investigated to
evaluate the antifungal activities of tuliposides and
tulipalins against tulip pathogenic fungi. 6-Tuliposide B
was effectively synthesized via the asymmetric Baylis-
Hillman reaction. Tuliposides and tulipalins showed
antifungal activities against most of the strains tested at
high concentrations (2.5 mM), while Botrytis tulipae was
resistant to tuliposides. Tulipalin formation was involved
in the antifungal activity, tulipalin A showed higher
inhibitory activity than 6-tuliposide B and tulipalin B.
Both the tuliposides and tulipalins showed pigment-
inducing activity against Gibberella zeae and inhibitory
activity against Fusarium oxysporum f. sp tulipae. These
activities were induced at a much lower concentration
(0.05 mM) than the antifungal MIC values.
It is also known that tulip cultivars can be damaged by
pathogenic fungi; for example, Fusarium oxysporum f.
sp tulipae and Botrytis tulipae respectively cause bulb
rot disease and Botrytis blight disease. We presumed
that tuliposides and tulipalins would have comparable
growth-inhibiting activities against tulip pathogenic
fungi. We therefore tested the biological activities of
6-tuliposide A, 6-tuliposide B, tulipalin A and tulipalin
B against these natural enemies, focusing on their
antifungal activities. We first explored the asymmetric
synthesis of 6-tuliposide B. We had already established
the total synthesis of 6-tuliposide B via chiral column
separation of protected 6-tuliposide B and its 30R-
epimer,10) and Katuski-Sharpless kinetic resolution of
the (S)-enantiomer and (R)-enantiomer of methyl
3-hydroxy-4-(p-methoxybenzyloxy)-2-methylenebuta-
noate.11) However, approximately half of the synthetic
compounds would be wasted by using this separation
method when the naturally occurring type of 6-tulipo-
side B was needed for further study. We therefore
developed a superior synthetic method for preparing the
naturally occurring type of 6-tuliposide via an asym-
metric Baylis-Hillman reaction to prepare these mo-
lecular probes more efficiently. After that, we evaluated
the biological activities of tuliposides and tulipalins
against several tulip pathogenic fungi, and investigated
SAR by using synthetic analogues of tuliposides to gain
insight into the action mechanism of these compounds.
Key words: tuliposide; tulipalin; Baylis-Hillman reaction
Tuliposides are the secondary metabolites mainly
occurring in Liliaceae and Alstroemeriaceae.1–3) Among
them, 6-tuliposide A (6-O-(40-hydroxy-20-methylenebu-
tanoyl)-D-glucose) and 6-tuliposide B (6-O-[(30S)-30,40-
dihydroxy-20-methylenebutanoyl]-D-glucose) are present
in tulip cultivars at high concentrations (0.2–2% w/w
fresh weight in all parts of the plant).4) These tuliposides
can be respectively converted into tulipalin A (2-
methylene-ꢀ-butyrolactone) and (ꢀ)-tulipalin B ((S)-3-
hydroxy-2-methylene-ꢀ-butyrolactone).5) It has re-
mained unclear whether or not this conversion in plant
tissues proceeded in an enzymatic manner, because
tulipalin formation also occurs in a pH-dependent
manner (above pH 7). An enzyme responsible for
converting tuliposides into tulipalins has recently been
purified from tulip bulbs and characterized.6) Among
tuliposides and tulipalins, tulipalin A is known to be
allergic7,8) for humans and to have insecticidal activity
against Thrips palmi.9)
6-Tuliposide B has been found to have potent
antimicrobial activity against a broad panel of bacteria,4)
prompting us to synthesize 6-tulipoisde B.10) We have
clarified that the formation of tulipalin B from 6-
tuliposide B was involved in the antibacterial activity
and that its target molecule could be MurA from the
results of a SAR study.11) These results also demon-
strated that 6-tuliposide B and tulipalin B would be
strategic weapons against bacterial infection.12,13)
Results and Discussion
To establish a stereoselective synthesis of the natural
type of 6-tuliposide B (30S-epimer), we first tried the
ꢁ-isocupreidine-catalyzed asymmetric Baylis-Hillman
reaction of 1,1,1,3,3,3-hexafluoroisopropyl acrylate
(HFIPA) with protected 2-hydroxyacetaldehyde.14) We
found that the Baylis-Hillman reaction of 3 eq. of
HFIPA and 1 eq. of 2-(tert-butyldimethylsilyloxy)-
acetaldehyde (1a) in the presence of 1 eq. of ꢁ-
isocupreidine gave the corresponding adduct (63%
e.e.) in a 16% yield. The reaction was slow (120 h),
and using the 2-(p-methoxybenzyloxy)-acetaldehyde
(1b) instead of 1a did not give any desired product
under the conditions tested (data not shown). Brzezinski
et al. have reported that N-acyl camphor sultam (2)
promoted the asymmetric Baylis-Hillman reaction. This
y
To whom correspondence should be addressed. Tel/Fax: +81-11-706-3638; E-mail: m-ub@for.agr.hokudai.ac.jp
Abbreviations: SAR, structure-activity relationship; MIC, minimum inhibitory concentration; MurA, UDPGlcNAc enolpyruvyltransferase

Asymmetric Synthesis of 6-Tuliposide B and SAR
719
Table 1. Asymmetric Synthesis of the 6-Tuliposide B Side-Chain
(3b) with outstanding enantiomeric excess in a good
yield (90%, >99% e.e., entry 5). Acidic silica gel
column chromatography (Wakogel C200, EtOAc:hexane,
1:4 to 1:2) was effective for purifying 3b which was
easily converted into a butanoate (4b) by treating with
triethylamine.15) Protection of the hydroxyl group of 4b
and subsequent alkaline hydrolysis of the ester gave acid
6 which was converted into 8 by condensation with 1-O-
(2-trimethylsilylethyl)-2,3,4-tri-O-(trimethylsilyl)-ꢁ-D-
glucopyranoside (7). The resulting sugar ester was
deprotected by trifluoroacetic acid to give the natural
type of 6-tuliposide B (9) as shown in Scheme 1.
Entry Reactant Solvent Time (h) Yield (%) e.e. (%)a
1
2
3
4
5
1a
1a
1b
1b
1b
CH2Cl2
DMF
24
24
0
7
—
97
CH2Cl2
DMF
DMF
24
24
0
—
98
This synthesized 6-tuliposide B (9) was evaluated for
its antifungal activity, together with its synthetic 30R-
epimer (epi-9), (ꢀ)-tulipalin B (10), (þ)-tulipalin B
(ent-10), natural 6-tuliposide A (11) and commercially
available tulipalin A (14). Several tuliposide B ana-
logues and their 30-epimers were also synthesized as
described,11) and their antifungal activity was analyzed.
The structures of all the compounds are depicted in
Table 2. Four species of tulip pathogenic fungi, namely
Pythium ultimum, Rhizoctonia solani, Botrytis tulipae,
Fusarium oxysporum f. sp tulipae and non-tulip patho-
genic Gibberella zeae were employed as test organisms.
All assay results are expressed as minimum inhibitory
concentration (MIC) values against each fungus in
Table 2. The growth of P. ultimum, R. solani and
G. zeae was inhibited by all naturally occurring 6-
tuliposide B (9), 6-tuliposide A (11), (ꢀ)-tulipalin B
(10) and tulipalin A (14) compounds. P. ultimum was
highly sensitive to these compounds, especially so to
(ꢀ)- and (þ)-tulipalin B (10 and ent-10). Synthetic (þ)-
tulipalin B (ent-10) exhibited the same activities as
natural 10 against these three strains. 6-Tuliposide A
(11) and 6-tuliposide B (9) showed equal antifungal
activity. In the case of the lactones, however, tulipalin A
(14) showed higher activity than tulipalin B (10), except
against P. ultimum. This characteristic was contrary to
that of the antibacterial activity, for which (ꢀ)- and (þ)-
tulipalin B (10 and ent-10) showed higher antibacterial
activity than tulipalin A (14). The higher antibacterial
activity of tulipalin Bs has been attributed to its putative
target, MurA.11) In contrast, the high antifungal activity
of tulipalin A (14) may be ascribed to its cell perme-
ability being enhanced by the absence of a ꢁ-hydroxyl
20
90
120
>99
aThe enantiomeric excesses are shown as those of butanoates (4a and 4b)
prepared from corresponding dioxanones (3a and 3b) by triethylamine
treatment.
coupling reaction has afforded 1,3-dioxan-4-ones in high
selectivity from a variety of aliphatic aldehydes, and
subsequent ring opening gave (3S)-3-hydroxy-2-meth-
ylene esters in moderate to good yields (33–98%) with
high selectivity (>99% e.e.).15) We therefore next
examined whether this method could be applied for
asymmetric synthesis of the naturally occurring type of
6-tuliposide B by using two types of protected 2-
hydroxyacetaldehydes (1a and 1b) as shown in Table 1.
We first used CH2Cl2 as a solvent in the reaction of
TBDMS-protected aldehyde (1a) and sultam (2) because
of its suitability for 1a, although the reaction provided
no desired adduct (entry 1). However, the use of DMF,
which is poor solvent for 1a, afforded the desired adduct
(3a) with superior selectivity in a poor yield (7%,
97% e.e., entry 2). We have previously reported that
ꢂ-silyloxy aldehyde afforded a 1,2-silyl-migrated adduct
in the Baylis-Hillman reaction,16) which could lead to
the formation of undesired by-products (e.g., ꢀ-lactone).
2-(p-Methoxybenzyloxy)-acetaldehyde (1b) was exam-
ined to prevent this undesired migration of the protect-
ing group, resulting in the cyclic adduct (3b) being
obtained in better yield (20%) with satisfactory selec-
tivity (98% e.e., entry 4). The reaction in CH2Cl2 did not
afford any product (entry 3). Finally, a prolonged
reaction time (120 h) afforded the Baylis-Hillman adduct
4b
Scheme 1.
a) Et3N, MeOH, 90%; b) TBDMSOTf, Et3N, CH2Cl2, quant.; c) LiOH (1.1 eq), MeCN:H2O ¼ 1:1, 60 ꢁC, 92%;
d) 6, diisopropyl carbodiimide, DMAP, CH2Cl2, 0 ꢁC, 61%; e) TFA:CH2Cl2 ¼ 2:1, 83%.

720
K. SHIGETOMI et al.
Table 2. Structures of the 6-Tuliposide B and Tulipalin B Analogues and Their Antifungal Activities
MIC (mM)
Pythium ultimum
Rhizoctonia solani
Botrytis tulipae
Fusarium oxysporum t.
Gibberella zeae
6-tuliposide B (9/epi-9)
tulipalin B (10/ent-10)
6-tuliposide A (11)
0.25/0.25
0.05/0.05
0.25
2.5/2.5
2.5/2.5
2.5
n.d./n.d.
n.d./n.d.
n.d.
2.5a/2.5a
n.d.a/n.d.a
2.5a
2.5b/2.5b
0.25b/0.25b
2.5b
tulipalin A (14)
0.25
0.5
2.5
0.5a
0.25b
ꢂ-Me glucoside (15/epi-15)
ꢁ-Me glucoside (16/epi-16)
1,2-dideoxy type (17/epi-17)
0.25/0.25
0.25/0.25
0.25/0.25
2.5/2.5
2.5/2.5
2.5/2.5
n.d./n.d.
n.d./n.d.
n.d./n.d.
n.d.a/n.d.a
n.d.a/n.d.a
n.d.a/n.d.a
2.5b/2.5b
2.5b/2.5b
2.5b/2.5b
12, 13, 18, 19, epi-12, epi-13, epi-18 and ent-19 did not show any activities at 2.5 mM.
n.d., No antifungal activity was detected at 2.5 mM.
aThe color of culture medium did not change to pink at 0.05 mM.
bThe color of culture medium changed to pale yellow at 0.05 mM.
group. P. ultimum has recently been classified in the
chromalveolata kingdom, rather than fungi kingdom, so
this strain might have shown different behavior toward
the tulipalins. Tulipalin A (14) showed higher activity
than 6-tuliposide A (11) in most cases and was the only
growth inhibitor against B. tulipae. It is interesting to
note that B. tulipae had enzyme systems capable of
converting 6-tuliposide A (11) into the hydroxyacid
form of tulipalin A that had no antifungal activity (data
not shown). Further studies are in progress to clarify its
mode of resistance against tulip secondary metabolites
from enzymatic and genetic points of view. Tuliposides
and tulipalins exhibited growth inhibitory activities
against fungi, but higher MICs (2.5 mM) than those
against bacteria (0.1–0.3 mM11)) were needed. We would
be interested to learn about the detailed mechanism for
their antifungal activities.
In addition to their antifungal activities, tuliposides
and tulipalins also exhibited ‘‘unexpected’’ acceleration
or inhibition of fungal pigment production at a low
concentration (0.05 mM). The culture medium for
G. zeae (a teleomorph of Fusarium graminearum)
turned pale yellow in an early phase (12 h) of the
culture period after adding tuliposides and tulipalins,
whereas the pigment was only induced after a week
without adding these compounds. In contrast, the culture
medium of F. oxysporum, which normally turns pink
during the culture, did not show any particular color by
adding these compounds. Moreover, the non-natural
type of (ꢀ)-tulipalin B (ent-10) and 6-tuliposide B (epi-
9) that did not show any antifungal activity against
F. oxysporum also inhibited production of the pigment.
The F. oxysporum species are known to produce such
naphthoquinone-type pigments as fusarubins17) and
javanicins,18) and G. zeae secretes such pigments as
aurofusarin,19) zearalenone20) and deoxynivalenol.21) It
is very interesting that these tulip metabolites affected
the secondary metabolism of pathogenic F. oxysporum
and non-pathogenic G. zeae. Although tuliposides and
tulipalins actually showed antifungal properties as we
speculated, they were unexpectedly efficient as pigment-
inducers or -inhibitors rather than as phytoanticipins
against some fungi.
Antifungal SAR was subsequently examined by using
the synthetic analogues,11) 40-deoxy type (12), hexanoyl
type (13), ꢂ-methyl glucoside (15), ꢁ-methyl glucoside
(16), 1,2-dideoxy type (17), amide-type (18), and their
30-epimers (epi-12, epi-13, epi-15, epi-16, epi-17 and
epi-18). Hydrogenated tulipalin Bs (19 and ent-19) were
also tested as shown in Table 2. Among the compounds
tested, 15, 16, 17 and their epimers in which the sugar
part of 6-tuliposide B was modified exhibited antifungal
activities. All the active analogues gave the same MIC
values against each fungus irrespective of the sugar
modification. Meanwhile, 12, 13, 18 and their epimers in
which the side-chain moiety of 6-tuliposide B was
modified were all inactive. The 40-deoxy-type (12 and
epi-12) and hexanoyl-type (13 and epi-13) analogues are
unable to form tulipalin B, and the amide-type analogues
(18 and epi-18) are difficult to lactonize because of
robust amide bonds. On the other hand, 6-tuliposide A
(11), which can be regarded as a side-chain analogue of
6-tuliposide B (9), retained its antifungal activity. These
results indicate that tulipalin formation played a key role
in the mechanism for antifungal action, like it did in the
antibacterial activities. Tuliposides can spontaneously or
enzymatically lactonize to release their sugar moieties,
and resulting tulipalin A (14) or tulipalin B (10) would
behave as an active principle. In most cases, the
configuration of the ꢁ-hydroxyl group did not affect
the antifungal activity, and this characteristic was
analogous to that for the antibacterial activity. Hydro-
genated tulipalin Bs (19 and ent-19) did not show any
activity, indicating that the unsaturated ꢂ-methylene
moiety was necessary for antifungal activity. The results
of the SAR study also demonstrate that tulipalin
formation was involved in the pigment-inducing or
-inhibiting activity. Side-chain-modified 12, 13 and 18

Asymmetric Synthesis of 6-Tuliposide B and SAR
721
OMe), 4.52 (2H, s, benzyl CH2), 4.56 (2H, s, benzyl CH2), 4.79 (1H,
m, ꢁ-CH), 5.45 (1H, t, J ¼ 4:6 Hz, acetalic CH), 5.73 (1H, d,
J ¼ 2:3 Hz, =CH2), 6.51 (1H, d, J ¼ 2:3 Hz, =CH2), 6.87 (4H, d,
J ¼ 8:6 Hz, aromatic H), 7.25 (2H, d, J ¼ 8:6 Hz, aromatic H), 7.27
(2H, d, J ¼ 8:6 Hz, aromatic H). 13C-NMR (67.5 MHz, CDCl3): 55.2
(PhOMe), 69.8 (CH2), 71.4 (ꢀ-CH2), 73.2 (benzyl CH2), 73.5 (benzyl
CH2), 99.5 (acetalic CH), 113.8 (aromatic C), 127.3 (=CH2), 129.3
(aromatic C), 129.39 (aromatic C), 129.44 (aromatic C), 129.6
(aromatic C), 133.8 (ꢁ-C), 159.3 (aromatic C), 159.4 (aromatic C),
163.0 (carbonyl C).
and hydrogenated 19 did not show these activities, while
the analogues which could form tulipalins showed the
same activities irrespective of their configuration at
the 30-position. Tulipalins would have been the active
principles, and tuliposides their precursors for exhibiting
both antifungal and pigment-inducing or -inhibiting
activities.
Conclusion
Synthesis of 6-O-[(30S)-30-(tert-butyldimethylsilyloxy)-40-(p-metho-
xybenzyloxy)-20-methylenebutanoyl]-2,3,4-tri-O-(trimethylsilyl)-1-O-
We established the asymmetric total synthesis of
6-tuliposide B, using N-acyl camphor sultam as a chiral
template, to selectively prepare the natural type of 6-
tuliposide B without monotonous HPLC separation to
remove the unwanted 30R-epimer. We also clarified its
biological activities against tulip pathogenic fungi.
Tuliposides and tulipalins had the potential to inhibit
both bacterial and fungal growth. Tulipalins showed
stronger antifungal activities than tuliposides, with
tulipalin A being more effective than tulipalin B.
However, those activities were lower than the antibac-
terial activities. Tuliposides and tulipalins perturbed the
production of fungal pigments at a lower concentration
than MIC in their antifungal activities against certain
pathogenic fungi. The results of the SAR study proved
tulipalins to have been the active principles in all these
activities. Further molecular biological studies are
needed to clarify details of the mechanism of action
for these activities.
(2-trimethylsilylethyl)-ꢁ-D-glucopyranoside (8). To
a solution of
carboxylic acid 6 (140 mg, 0.38 mmol) in 4.0 mL of CH2Cl2, N,N-
dimethylaminopyridine (23.2 mg, 0.19 mmol) and diisopropylcabodii-
mide (75 mL, 0.46 mmol) were added at ꢀ20 ꢁC. Protected sugar 7
(150 mg, 0.30 mmol) was then added, and the mixture stirred overnight
at 0 ꢁC. The reaction mixture was diluted with Et2O and was washed
with a saturated NH4Cl aqueous solution. The organic layer was
washed with brine and dried over anhydrous Na2SO4. Removal of the
solvent by evaporation and subsequent purification by silica gel
column chromatography (EtOAc:hexane, 1:19) afforded 154 mg of
24
sugar ester 8 (0.18 mmol, 61%) as a syrup, ½ꢂꢃD ¼ þ9:2 (c 3.01,
CHCl3). HR-FD-MS m=z [M]þ: calcd. for C39H76O16Si5, 844.4285;
found, 844.4280. 1H-NMR (270 MHz, CDCl3): 0.02–0.18 (42H, m,
SiMe), 0.91 (9H, s, t-Bu), 0.95–1.04 (2H, m, CH2CH2SiMe3), 3.30–
3.50 (6H, m, H-2, H-3, H-4, H-5, H-40a and CH2CH2SiMe3), 3.56 (1H,
dd, J ¼ 10:2, 3.0 Hz, H-40b), 3.84 (3H, s, OMe), 3.84–3.97 (1H, m,
CH2CH2SiMe3), 4.06 (1H, dd, J ¼ 11:7, 5.2 Hz, H-6a), 4.18 (1H, d,
J ¼ 7:6 Hz, H-1), 4.42–4.60 (3H, m, H-6b and benzyl CH2), 4.83 (1H,
br dd, H-30), 6.06 (1H, s, =CH2), 6.37 (1H, s, =CH2), 6.86 (2H,
d, J ¼ 8:2 Hz, aromatic H), 7.25 (2H, d, J ¼ 8:2 Hz, aromatic H).
13C-NMR (67.5 MHz, CDCl3): ꢀ5:0 and ꢀ4:8 (Me2Si), ꢀ1:5
(CH2CH2SiMe3), 0.9–1.3 (OSiMe3), 18.1 and 18.2 (CH2CH2SiMe3
and CMe3), 25.8 (CMe3), 55.2 (PhOMe), 63.9 (C-6), 66.7
(CH2CH2SiMe3), 70.4 (C-40), 72.2 (C-4), 72.7 (benzyl CH2), 73.7
(C-30), 74.8 (C-5), 76.0 (C-2), 78.5 (C-3), 102.7 (C-1), 113.5 (aromatic
C), 126.5 (=CH2), 129.1 (aromatic C), 130.6 (aromatic C), 140.8
(C-20), 158.9 (aromatic C), 165.7 (C-10).
Experimental
Chemicals of the highest commercial purity were used without
further purification unless otherwise stated. Thin-layer chromatography
was performed with Merck 60 F254 silica gel. Silica gel column
chromatography was performed with Kanto Chemicals 60 N silica gel
(spherical, neutral), except for purifying dioxanone 3b which used
C200 Wakogel (Wako Co.). Chiral HPLC was performed with a Daicel
Chiralpakꢀ IA column (ꢃ 4:6 mm ꢂ 250 mm) and a Hitachi L-7455
photodiode array detector at 30 ꢁC. 1H- and 13C-NMR spectra were
measured in CDCl3 with a Jeol JNM-EX270 instrument and in D2O
with a Jeol JNM-LA400. Chemical shifts are reported in ꢄ ppm, using
tetramethylsilane as the internal standard, and coupling constants (J)
are given in Hertz. All the FD-mass spectra were measured with a
JMS-SX102A instrument by the GC-MS and NMR Laboratory of
Faculty of Agriculture at Hokkaido University. Optical rotation data
were determined with a Jasco P-2200 polarimeter in ꢃ 3:4 mm ꢂ
50 mm cells at 24 ꢁC. Dichloromethane was distilled from phosphorous
oxide. All of the compounds having an ꢂ,ꢁ-unsaturated carbonyl
structure were supplemented with 200 ppm of hydroquinone to prevent
polymerization after being purified. Tulipalin A (14) was purchased
from Aldrich Chemical Company. The synthetic procedures and
spectral data for compounds 5, 6, 7, 9, epi-9, 10, ent-10, 12, epi-12, 13,
epi-13, 15, epi-15, 16, epi-16, 17, epi-17, 18, epi-18, 19 and ent-19
have been presented in our previous reports.10,11)
Preparation of 6-tuliposide A (11). To cut petals (30 g fresh weight)
of the murasaizuishou tulip was added 100 mL of cold 50% MeOH,
and the mixture was vigorously stirred at 4 ꢁC for 1 h. The extract was
filtered through cheesecloth and centrifuged (2;000 ꢂ g, 4 ꢁC, 5 min).
The resulting supernatant was defatted by washing with CHCl3, and
the aqueous layer was lyophilized. The extract was dissolved in a
minimum amount of water, applied to an ODS-column (ꢃ 4:5 mm ꢂ
150 mm) and eluted with water. The fractions containing tuliposide A
were combined, concentrated in vacuo, and lyophilized. Crude
tuliposide A was further purified by preparative HPLC at 225 nm in
an RP-18GPAqua column (ꢃ 10 mm ꢂ 250 mm), using
a linear
gradient of MeOH (0 to 75% for 17 min) at a flow rate of 3 mL/min.
Pure tuliposide A (13.7 mg) was obtained as an amorphous powder
after lyophilization. 1H-NMR (400 MHz, D2O): 2.58 (2H, t, J ¼
6:3 Hz, H-30), 3.74 (2H, t, J ¼ 6:3 Hz, H-40), 3.2–4.0 (4H, m, H-2, H-3,
H-4 and H-5), 4.3–4.5 (2H, m, H-6), 4.68 (0.5H, d, J ¼ 8:0 Hz, H-1ꢁ),
5.23 (0.5H, d, J ¼ 3:9 Hz, H-1ꢂ), 5.81 (1H, s, =CH2), 6.32 (1H, d,
J ¼ 3:4 Hz, =CH2). 13C-NMR (100 MHz, D2O): 34.9 (C-30), 60.7
(C-40), 129.7 (=CH2), 137.0 (C-20), 169.4 (C-10); ꢂ-D-glucosyl: 64.3
(C-6), 70.4 (C-4), 70.5 (C-5), 72.2 (C-2), 73.3 (C-3), 92.4 (C-1);
ꢁ-D-glucosyl: 64.4 (C-6), 74.2 (C-5), 70.4 (C-4), 74.9 (C-2), 76.3
(C-3), 96.8 (C-1).
Synthesis of (6S)-2,6-Di-(p-methoxybenzyloxy)methyl-5-methylene-
1,3-dioxan-4-one (3b). Acylated camphor sultam 2 (47.8 mg, 0.178
mmol) and DABCO (1.98 mg, 0.018 mmol) were dissolved in 0.2 mL
of DMF at room temperature. The mixture was cooled at 0 ꢁC for 1 h,
and PMB-protected aldehyde 1b (467 mg, 2.59 mmol) was added.
After 120 h, the reaction mixture was diluted with Et2O and washed
with brine. The organic layer was dried over anhydrous Na2SO4 and
evaporated to remove the solvent. The obtained crude product was
purified by silica gel (Wakogel C200) column chromatography
(EtOAc:hexane, 1:4 to 1:2). Evaporation of the corresponding fractions
Antifungal assay. The antifungal activities were evaluated by a
microscopic analysis during the growth of each fungus. The fungal
strains were maintained on potato dextrose agar (PD agar; Difco,
Detroit, WI, USA). The tested compounds were each diluted with Milli
Q-water or DMSO as described in our previous report.11) The
minimum inhibitory concentration (MIC) is defined as the lowest
concentration at which no growth was apparent. Each fungus was pre-
cultured on a PD agar plate for several days until the hyphae had grown
well at 25 ꢁC. The mycelia were scratched off from the surface of the
24
gave 66.1 mg of 3b (90%), ½ꢂꢃD ¼ þ25:4 (c 2.15, CHCl3). HR-FD-
MS m=z [M]þ: calcd. for C23H26O7, 414.1679; found, 414.1655.
1H-NMR (270 MHz, CDCl3): 3.65–3.69 (4H, m, CH2), 3.80 (6H, s,

722
K. SHIGETOMI et al.
plate and suspended in sterilized saline. To a 24-well tissue culture
plate (Corning, NY, USA) containing 200 mL of the PD medium
supplemented with 0.05–2.5 mM of each compound was inoculated
25 mL of the mycelial suspension of a fungal strain. The morphological
changes of the fungal hyphae were observed under a CK2 optical
microscope (Olympus, Japan) during incubation at 20 ꢁC for several
days. The fungal strains used were Fusarium oxysporum f. sp. tulipae
MAFF 235110 (tulip bulb rot disease), Pythium ultimum MAFF
235799 (tulip root rot disease), Rhizoctonia solani MAFF 235845
(tulip leaf rot disease), Botrytis tulipae MAFF 237887 (tulip botrytis
blight disease), and Gibberella zeae MAFF 305135 (non-pathogenic
against tulip).
6) Kato Y, Shoji K, Ubukata M, Shigetomi K, Sato Y, Nakajima N,
and Ogita S, Biosci. Biotechnol. Biochem., 73, 1895–1897 (2009).
7) Marks JE, Arch. Dermatol., 124, 914–917 (1988).
8) Gette MT and Marks JE, Arch. Dermatol., 126, 202–205
(1990).
9) Kim C-S, Datta PK, Hara T, Itoh E, and Horiike M, Biosci.
Biotechnol. Biochem., 63, 152–154 (1999).
10) Shigetomi K, Kishimoto T, Shoji K, and Ubukata M, Tetrahe-
dron: Asymmetry, 19, 1444–1449 (2008).
11) Shigetomi K, Shoji K, Mitsuhashi S, and Ubukata M,
Phytochemistry, 71, 312–324 (2009).
12) Saaltink GJ and Maas Geesteranus HP, Neth. J. Pl. Path., 75,
123–128 (1969).
13) Nishiyama K, Kusaba T, Ohta K, Nahata K, and Ezuka A, Ann.
Phytopath. Soc. Japan, 45, 668–674 (1979).
Acknowledgment
14) Iwabuchi Y, Nakatani M, Yokoyama N, and Hatakeyama S,
J. Am. Chem. Soc., 121, 10219–10220 (1999).
This work was supported in part by grant-aid for
scientific research (A) provided by the Japan Society for
the Promotion of Science of Japan.
15) Brzezinski LJ, Rafel S, and Leahy JW, J. Am. Chem. Soc., 119,
4317–4318 (1997).
16) Shigetomi K, Kishimoto T, Shoji K, and Ubukata M, Hetero-
cycles, 69, 63–67 (2006).
References
17) Ruelius HW and Gauche A, Ann. Chem., 569, 38–59 (1950).
18) Arsenault P, Tetrahedron Lett., 6, 4033–4037 (1965).
19) Ashley JN, Hobbs BC, and Raistrick H, Biochem. J., 31, 385–
397 (1937).
1) Slob A, Phytochemistry, 12, 811–815 (1973).
2) Slob A, Jekel B, and Jong BD, Phytochemistry, 14, 1997–2005
(1975).
3) Christensen LP, Phytochemistry, 38, 1371–1373 (1995).
4) Shoji K, Ubukata M, Momonoi K, Tsuji T, and Morimatsu T,
J. Jpn. Soc. Hort. Sci., 74, 469–475 (2005).
20) Urry WH, Wehrmeister HL, Hodge EB, and Hidy PH,
Tetrahedron Lett., 7, 3109–3114 (1966).
21) Yoshizawa T and Morooka N, Agric. Biol. Chem., 37, 2933–
2934 (1973).
5) Tschesche R, Kammerer FJ, and Wulff G, Tetrahedron Lett., 6,
¨
701–706 (1968).