Synthesis and structure-activity relationship of vicenistatin, a cytotoxic 20-membered macrolactam glycoside

Article abstract of DOI:10.1002/asia.201200615

We have developed two syntheses of vicenistatin and its analogues. Our first-generation strategy included the rapid and sequential assembly of the macrocyclic lactam by using an intermolecular Horner-Wadsworth-Emmons reaction between the C3-C13 fragment and the C1-C2, C14-C19 fragment, followed by an intramolecular Stille coupling reaction. The second-generation strategy utilized a ring-closing metathesis of a hexaene intermediate to generate the desired 20-membered macrolactam. This second-generation strategy made it possible to prepare synthetic analogues of vicenistatin, including the C20- and/or C23-demethyl analogues. Evaluation of the cytotoxic effect of these analogues indicated the importance of the fixed conformation of aglycon for determining the biological activity of the vicenistatins. An enantioselective total synthesis of vicenistatin (1) was accomplished by using an intermolecular Horner-Wadsworth-Emmons reaction and a ring-closing metathesis as key steps. This strategy made it possible to prepare synthetic analogues of vicenistatin, including the C20- and/or C23-demethyl vicenistatins. Evaluation of their cytotoxicity indicated the importance of the fixed conformation of aglycon in determining biological activity. Copyright

Full text of DOI:10.1002/asia.201200615

DOI: 10.1002/asia.201200615
Synthesis and Structure–Activity Relationship of Vicenistatin, a Cytotoxic 20-
Membered Macrolactam Glycoside
Hayato Fukuda,[a] Yuko Nishiyama,[b] Shiina Nakamura,[a] Yutaro Ohno,[a]
Tadashi Eguchi,[c] Yoshiharu Iwabuchi,[a] Takeo Usui,[b] and Naoki Kanoh*[a]
Abstract: We have developed two syn-
theses of vicenistatin and its analogues.
Our first-generation strategy included
the rapid and sequential assembly of
the macrocyclic lactam by using an in-
second-generation strategy utilized
a ring-closing metathesis of a hexaene
intermediate to generate the desired
second-generation strategy made it
possible to prepare synthetic analogues
of vicenistatin, including the C20- and/
or C23-demethyl analogues. Evaluation
of the cytotoxic effect of these ana-
logues indicated the importance of the
fixed conformation of aglycon for de-
termining the biological activity of the
vicenistatins.
20-membered
macrolactam.
This
termolecular
Horner–Wadsworth–
Keywords: antitumor
agents
·
Emmons reaction between the C3–C13
fragment and the C1–C2, C14–C19
fragment, followed by an intramolecu-
lar Stille coupling reaction. The
macrocycles · natural products ·
structure–activity relationships
total synthesis
·
Introduction
finally established by the first total synthesis.[4] The biosyn-
thetic pathway of vicenistatin (1) has been well-character-
ized by using feeding experiments with isotope-labeled in-
gredients[5] and by utilizing information from functional
analysis of the biosynthetic gene cluster.[6] Moreover, an
Macrocyclic lactams have been widely found in nature;[1]
these molecules are characteristic in terms of their structure,
in which a hydrophilic cluster is present in the macrocyclic
skeleton, especially at a remote site from an amide group
(Figure 1). There are two subgroups that belong to the
family of macrocyclic lactams: the first contains a sugar
moiety and the other contains only hydroxy groups. Vicenis-
tatin (1)[2] belongs to the former of these subgroups, which
are named macrolactam glycosides.
Vicenistatin (1), which is a conjugate of vicenisamine and
vicenilactam, was first isolated in 1993 by Shindo et al. from
the culture broth of Streptomyces sp. HC34.[2] Its stereo-
structure, including its absolute configuration, was originally
proposed based on the results of degradation studies[2–3] and
[a] Dr. H. Fukuda, S. Nakamura, Y. Ohno, Prof. Dr. Y. Iwabuchi,
Prof. Dr. N. Kanoh
Graduate School of Pharmaceutical Sciences
Tohoku University
Aobayama, Sendai 980-9578 (Japan)
Fax : (+81)22-795-6847
E-mail: nkanoh@m.tohoku.ac.jp
[b] Y. Nishiyama, Prof. Dr. T. Usui
Graduate School of Life and Environmental Sciences
University of Tsukuba
Tennodai, Tsukuba 305-8572 (Japan)
[c] Prof. Dr. T. Eguchi
Department of Chemistry and Material Sciences
Tokyo Institute of Technology
2-12-1 O-okayama, Meguro-ku, Tokyo 152-8551 (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201200615.
Figure 1. Representative macrocyclic lactams that are found in nature.
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
&
&
&
These are not the final page numbers! ÞÞ

FULL PAPER
enzyme named VinC in the gene cluster has been utilized to
create unnatural glycoside molecules, each of which contain
a vicenisamine unit.[7]
Several biological activities of vicenistatin (1) have been
documented. It was reported that vicenistatin (1) not only
showed growth-inhibition activity against several human
cancer cell-lines (human promyelocytic leukemia HL-60 and
human colon cancer COLO205 cells) in vitro, but also ex-
hibited antitumor activity against human colon carcinoma
Co-3 cells in the mouse xenograft model.[2] Thus, these re-
sults indicated that vicenistatin has potential for use as
a new anticancer drug.
Interestingly, a naturally occurring analogue, named vice-
nistatin M (2), which contains d-mycarose instead of vicenis-
amine, did not show any significant cytotoxic activity against
cancer cells, thereby suggesting the importance of vicenisa-
mine for exerting the cytotoxicity of vicenistatin (1).[8] Re-
cently, vicenistatin (1) was re-discovered as a compound
that inhibits the growth of budding yeast Saccharomyces cer-
evisiae from a high-throughput yeast-halo assay.[9]
Intrigued by these observations, we embarked on the
chemical synthesis of not only the natural product itself but
also of its analogues and derivatives that can be used for the
study of the structure–activity relationships in these com-
pounds and for molecular-target identification. Herein, we
report two syntheses and a study of the structure–activity re-
lationships in vicenistatin (1). The first-generation synthesis
was highly convergent and finally culminated in the second
total synthesis of vicenistatin (1),[10] yet the efficiency of the
macrolactam-ring-formation step was unsatisfactory. To
overcome this problem, we developed a second-generation
synthesis of vicenilactam, which featured a ring-closing-
metathesis reaction of the hexaene precursor. This second-
generation strategy made it possible to prepare several ana-
logues, including C20- and/or C23-demethyl compounds. By
using these derivatives and analogues, we performed a struc-
ture–activity-relationship study and demonstrated the im-
portance of the fixed conformation of aglycon in determin-
ing the biological activity of these compounds.
Scheme 1. Retrosynthetic analysis. PG=protecting group, X=halogen
atom, M=metal atom.
lactam II could be obtained through successive coupling re-
actions; that is, a Horner–Wadsworth–Emmons (HWE) re-
action between fragments III and IV followed by cross cou-
pling, or vice versa. This highly convergent strategy was ex-
pected to facilitate a study of its structure–activity relation-
ship.
First-Generation Synthesis of Protected Vicenilactam
Synthesis of iodoenal (ꢀ)-7 began from the known homoall-
yl alcohol 4 (Scheme 2).[4a,c] Carboalumination of compound
4 in wet CH2Cl2, followed by iodination, gave vinyliodide 6,
which was oxidized with Dess–Martin periodinane[13] and
Results and Discussion
Retrosynthetic Analysis
To gain insight into the mode-of-action and structure–activi-
ty relationship of vicenistatin (1), we wanted to develop
a method for its chemical synthesis that could also be uti-
lized to synthesize its analogues and derivatives. Our syn-
thetic strategy is summarized in Scheme 1.
Retrosynthetically, vicenistatin (1) was first divided into
vicenisamine fragment I and protected vicenilactam frag-
ment II. Vicenisamine fragment I was expected to be pre-
pared from known epoxyalcohol 3[11] by using the cyclic-car-
bamate formation developed by Roush and James.[12] Frag-
ment II was further divided into haloaldehyde III and phos-
phonate IV, which could be prepared from known homoallyl
alcohol 4[4a,c] and 5-hexyn-1-ol (5), respectively. Thus, macro-
Scheme 2. Synthesis of iodoolefin (ꢀ)-7. Reagents and conditions: a)
Me3Al (3 equiv), Cp2ZrCl2 (0.5 equiv), compound 4, wet CH2Cl2, 08C to
RT, 3 h, then I2 (1.2 equiv), 08C to RT, 2 h; b) Dess–Martin periodinane
(1.1 equiv), CH2Cl2, RT, 1 h; c) tBuOK (2.0 equiv), cis-2-butene (excess),
nBuLi (1.7 equiv), THF, ꢀ788C to ꢀ508C, 10 min, then (+)-Ipc2BOMe
(1.7 equiv), ꢀ788C, 30 min, then BF3·OEt2 (1.7 equiv), crude aldehyde,
ꢀ788C to 08C, then MeOH, sat. aq. NaHCO3, 30% aq. H2O2, overnight.
Ipc=isopinocampheyl.
&
&
&2
&
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Vicenistatin, a Cytotoxic 20-Membered Macrolactam Glycoside
then subjected to Brown asymmetric crotylboration[14] to
afford homoallyl alcohol (ꢀ)-7 in 30% overall yield and
with high enantio- and diastereoselectivity (100% de, 94%
ee). It should be noted that the carboalumination reaction
did not give reproducible results in the absence of water.[15]
Next, the introduction of the C3–C5 enal moiety by using
a cross-metathesis reaction was examined (Table 1). Among
the various metathesis partners (methyl acrylate (10),
Table 1. Cross-metathesis reactions of compound (ꢀ)-7 with several a,b-
unsaturated olefins.
Scheme 3. Synthesis of stannyl phosphate (ꢀ)-17. Reagents and condi-
tions: a) K2CO3 (2 equiv), MeOH, RT, 6 h; b) nBu3SnH (1.3 equiv),
iPr2NEt (0.2 equiv), [Pd2ACTHUNRTGNE(UGN dba)3]·CHCl3 (1 mol%), Cy3P·HBF4 (5 mol%),
CH2Cl2, RT, 2 h; c) Ph3P (2 equiv), DIAD (2 equiv), THF, 08C, 20 min,
then compound (ꢀ)-20, DPPA (2 equiv), 08C to RT, 2 h; d) H2 balloon,
5% Lindlar catalyst (20% w/w), quinoline (1.5 equiv), MeOH, RT, 2 h;
e) diethylphosphonoacetic acid (21, 2.5 equiv), EDCI (2.5 equiv), HOBt
(2.5 equiv), CH2Cl2, RT, 5 min, then Et3N (3 equiv), 08C to RT, 10 h.
dba=di(benzylidene)acetone, DIAD=diisopropyl azodicarboxylate,
DPPA=dipheynylphosphoryl azide, EDCI=1-ethyl-3-(3-dimethylamino-
propyl)carbodiimide hydrochloride, HOBt=1-hydroxylbenzotriazole.
Entry
Reactant[a]
Catalyst
Product
Yield [%][b]
1
2
3
10 (10)
11 (10)
12 (10)
12 (10)
12 (10)
13
13
13
14
13
9
9
8
8
8
trace
19
32–37 (61)
4[c]
5[d]
7
10
[a] The number of molar equivalents of the reactant used is given in pa-
rentheses. [b] The yield based on recovered starting material is given in
parentheses. [c] Heated at reflux. [d] No solvent.
methyl crotonate (11), and acrolein (12)) and catalysts
(Hoveyda–Grubbs second-generation catalyst (13) and
Grubbs second-generation catalyst (14)) that were examined
(Table 1), we found that the reaction with acrolein (12) in
CH2Cl2 in the presence of the Hoveyda–Grubbs second-gen-
eration catalyst (13) gave moderate but reproducible yield.
Several byproducts, one of which was cyclopentenol (15),
which was produced by a ring-closing metathesis (RCM) be-
tween the terminal- and C9–C10 olefins,[16] were obtained
along with the desired enal (ꢀ)-8. The enal (ꢀ)-8 was pro-
tected as its TMS ether to afford compound (ꢀ)-16 in 29%
yield in two steps from compound (ꢀ)-7 (see Scheme 4). Al-
though the overall yield was moderate, we decided to use
this reaction sequence because it quickly and reproducibly
provided the desired fragment, (ꢀ)-8, in only four steps
from compound 4.
Scheme 4. First-generation synthesis of protected vicenilactam fragment,
(+)-25. Reagents and conditions: a) TMSCl (3 equiv), 2,6-lutidine
(6 equiv), RT, 40 min; b) compound (ꢀ)-17 (1.1 equiv), KHMDS
(1.2 equiv), ꢀ788C, 5 min, then compound (ꢀ)-16, THF, ꢀ788C to 08C,
The preparation of phosphonate (ꢀ)-17 commenced with
the introduction of a chiral methyl group onto the C23
carbon atom (Scheme 3). According to the procedure devel-
oped by Paquette et al.,[17] chiral alcohol (ꢀ)-18 was pre-
pared in five steps from 5-hexyn-1-ol (5). Removal of a tri-
methylsilyl group in compound (ꢀ)-18, followed by palladi-
20 min; c) [Pd2ACHTUNRTGNE(UNG dba)3]·CHCl3 (20 mol%), Ph3As (20 mol%), iPr2NEt
(20 mol%), DMF, RT, 2 h. KHMDS=potassium hexamethyldisilazide.
um-catalyzed hydrostannation of the resulting terminal
alkyne, (ꢀ)-19, gave vinylstannane (ꢀ)-20 in 74% yield with
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
3
&
&
&
These are not the final page numbers! ÞÞ

Naoki Kanoh et al.
FULL PAPER
high regio- and stereoselectivity.[18] The minor gem-isomer
could be easily removed after the intramolecular Stille-cou-
pling reaction (see Scheme 4). Vinylstannane alcohol (ꢀ)-20
was converted into its corresponding azide under Mitsunobu
conditions, which was then reduced by using a Lindlar cata-
lyst and condensed with diethylphosphonoacetic acid (21) to
furnish fragment (ꢀ)-17.
the formation of the kinetically favored substituted cyclo-
pentene, as observed in the cross-metathesis of compound
(ꢀ)-7. Of RCM precursors VII, VIII, and IX, we selected
IX as an intermediate, because it was easily accessible from
the precursors of our first-generation synthesis.
Thus, tetraene (ꢀ)-26 was prepared by the Stille coupling
of compound (ꢀ)-8 with tri-n-butyl
ACHTUNGERTN(NUNG vinyl)tin, followed by
protection of the C7-hydroxy group in 82% overall yield
Having all of the fragments in hand, we then focused on
the construction of the macrocyclic lactam skeleton
(Scheme 4). After several experiments, we quickly recog-
nized that the intermolecular Stille coupling between TMS-
protected iodide (ꢀ)-16 and vinylstannane (ꢀ)-17 was prob-
lematic; these reactions either resulted in degradation of the
starting material or, surprisingly, produced a 1:1 inseparable
mixture of a regular coupling product (22) and an irregular
coupling product (23).[19] In contrast, coupling of these two
fragments by using a HWE reaction proceeded without diffi-
culty: phosphonate (ꢀ)-17 was treated with KHMDS in
THF at ꢀ788C and the resultant phosphonate carbanion
was successfully coupled with aldehyde (ꢀ)-16 to give pen-
taene (ꢀ)-24 in 71% yield. The intramolecular Stille cou-
pling of compound (ꢀ)-24 proceeded under the conditions
(Scheme 6). Phosphonate (ꢀ)-27 was prepared from com-
Scheme 6. Synthesis of
conditions: a) tributylACHTUNTGERN(NUNG vinyl)tin (1.7 equiv), Ph3P (0.33 equiv), [Pd2
tetraene
(ꢀ)-26.
Reagents
and
AHCTUNTGREGUN(NN dba)3]·CHCl3 (8 mol%), THF, RT, 1 h; b) TMSCl (3 equiv), 2,6-lutidine
(6 equiv), CH2Cl2, RT, 4 h.
reported by Nicolaou et al.[20] ([Pd
N
pound (ꢀ)-18 in the usual manner (Scheme 7). Coupling of
the two fragments was accomplished by using LiHMDS as
a base to give hexaene (ꢀ)-29 in 85% yield. Finally, Boc-
protection of the amide functionality furnished the desired
hexaene, (ꢀ)-30, in 91% yield (Scheme 8).
iPr2NEt, DMF) to give the desired macrolactam (+)-25 in
modest yield.
Second-Generation Synthesis of Protected Vicenilactam
Having the desired hexaene (ꢀ)-30 in hand, we next ex-
amined its RCM reaction of compound (ꢀ)-30 (Scheme 9).
Gratifyingly, the desired RCM reaction proceeded in the
presence of Grubbs first-generation catalyst (31) and p-qui-
none[21] to give a 5:1 inseparable mixture of compounds
(14E)-32 and (14Z)-32 in 65% combined yield. Grubbs- and
Hoveyda–Grubbs second-generation catalysts did not work
well in this reaction. It also should be noted that the Boc
group on the amide nitrogen atom was necessary for the de-
sired RCM reaction to proceed. Then, the mixture was
treated with methanolic HCl to remove the TMS group, fol-
lowed by reaction with p-bro-
Although we have accomplished the concise synthesis of
vicenilactam fragment (+)-25, the efficiency of the macro-
lactam-ring-formation step was quite unsatisfactory. To over-
come this problem, we revised our retrosynthetic analysis
and decided to develop a second-generation synthesis of
vicenilactam that featured a ring-closing metathesis (RCM)
of a hexaene precursor (Scheme 5). Of the five retrosynthet-
ꢀ
ically cleavable olefins, a disconnection of the C4 C5 (i.e.,
ꢀ
precursor V) or C9 C10 bonds (precursor VI) would not be
appropriate because these precursors would rather lead to
mobenzoyl chloride to give the
separable
p-bromobenzoates
(14E)-33 and (14Z)-33. Finally,
three protecting-group-manipu-
lation steps afforded the de-
sired compound (+)-25 in 88%
yield.
Synthesis of Protected
Vicenisamines
The synthesis of the vicenisa-
mine part is summarized in
Scheme 10. Starting from epox-
yalcohol (+)-3,[11] reaction with
benzoyl isocyanate in CH2Cl2
followed by treatment of the
product with 1,8-diazabicyclo-
Scheme 5. Ring-closing metathesis precursors that lead to compound (+)-25. PG: protecting group.
&
&
&4
&
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Vicenistatin, a Cytotoxic 20-Membered Macrolactam Glycoside
Scheme 7. Synthesis of phosphonate (ꢀ)-27. Reagents and condition-
s: a) Ph3P (2 equiv), DIAD (2 equiv), (ꢀ)-18, phthalimide (2 equiv),
THF, 08C to RT, 2 h; b) H2 balloon, Lindlar catalyst (20% w/w), quino-
line (1.5 equiv), EtOAc, RT, 2.5 h; c) 80% hydrazine (3 equiv), EtOH,
reflux, 3 h; d) diethylphosphonoacetic acid (21) (5 equiv), EDCI
(5 equiv), HOBt (5 equiv), CH2Cl2, RT, 5 min, then Et3N (15 equiv), RT,
10 h.
Scheme 9. Second-generation synthesis of protected vicenilactam (+)-25.
Reagents and conditions: a) Grubbs first-generation catalyst (31,
20 mol%), p-quinone (0.4 equiv), DCE, reflux, 1 h; b) 2% aq. HCl
(5 equiv), MeOH/Et2O (1:1), RT, 1 h; c) p-BrBzCl (5 equiv), pyridine,
RT, 5 h; d) chromatographic separation; e) compound (14E)-33,
TMSOTf (5 equiv), 2,6-lutidine (10 equiv), RT, 5 h; f) K2CO3 (29 equiv),
CHCl3/MeOH (1:1), RT, 37 h; g) TMSCl (1 equiv), Et3N (2 equiv),
CH2Cl2, RT, 2 h. DCE=1,2-dichloroethane, Bz=benzoyl, TMSOTf=tri-
methylsilyl trifluoromethanesulfonate.
of the resulting diol, gave glycosyl acetate 37 in 95% overall
yield. Alternatively, protection of a hydroxy group in com-
pound 39 as its triisopropylsilyl (TIPS) ether, followed by
acid treatment and fluorination with DAST, produced glyco-
syl fluoride 38 in 80% overall yield.
Scheme 8. Synthesis of a ring-closing metathesis precursor (ꢀ)-30. Re-
agents and conditions: a) compound (ꢀ)-27 (1.1 equiv), LiHMDS
(1.2 equiv), THF, ꢀ788C, 5 min, then compound (ꢀ)-26, ꢀ788C to 08C,
20 min; b) Boc2O (40 equiv), Et3N (40 equiv), DMAP (4 equiv), CH2Cl2,
RT, 11 h. DMAP=4-(dimethylamino)pyridine, Boc=tert-butoxycarbon-
yl.
Total Synthesis of Vicenistatin
A
First, we examined the glycosylation reaction of protected
vicenilactam (+)-25 with glycosyl acetate 37 under Mukaiya-
ma conditions[24] (SnCl4, AgClO4, 4ꢁ M.S., CH2Cl2). Al-
though Kakinuma and co-workers have accomplished this
glycosylation reaction (59%),[4c] it could not be reproduced
in our hands. Thus, we turned our attention to an alternative
glycosylation strategy in which glycosyl fluorides 36 and 38
were used as glycosyl donors (Scheme 11). Because the solu-
bility of vicenilactam itself in organic solvents was quite low,
TMS-protected vicenilactam (+)-25 was used as a glycosyl
acceptor throughout this study. Of the glycosylation condi-
tions that were examined, we found that the procedure re-
ported by Noyori and co-workers[25] was suitable for our pur-
poses, that is, (+)-25 was treated with glycosyl fluoride 36 at
ꢀ208C in the presence of TMSOTf to give a mixture of cou-
pling products 39 in 56% yield. Deprotection of the cyclic
carbamate moiety in compound 39 under alkaline condi-
tions, followed by careful chromatographic separation, af-
forded vicenistatin (1) in 10% yield (2 steps) along with its
yield through N!O rearrangement of the benzoyl group.[12]
N-methylation of compound (ꢀ)-34 by using MeI and subse-
quent methanolysis gave the hydroxy olefin, which was sub-
jected to ozonolysis in MeOH followed by reductive work-
up and acid treatment to give protected vicenisamine 35 as
a 1:1 mixture of anomers in 86% yield (three steps). This
mixture of protected vicenisamine 35 was then converted
into its corresponding thioglycosides by using phenylthio tri-
methylsilyl sulfide (PhSTMS);[22] a substitution reaction with
N-bromosuccinimide (NBS) and (diethylamino)sulfur tri-
fluoride (DAST)[23] gave glycosyl fluoride 36 in 65% yield in
two steps.
We also synthesized differentially protected vicenisamines
37 and 38 as follows: Removal of the cyclic carbamate
group on compound 35 with 30% aq. NaOH, followed by
protection of the resultant amino group as its Fmoc carba-
mate, afforded alcohol 39 in 87% yield in two steps. Acid
treatment of compound 39 at 958C, followed by acetylation
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
5
&
&
&
These are not the final page numbers! ÞÞ

Naoki Kanoh et al.
FULL PAPER
Scheme 10. Preparation of glycosyl fluorides 36, 38, and acetate 37. Re-
agents and conditions: a) BzNCO (1.2 equiv), CH2Cl2, RT, 1 h; b) DBU
(4 mol%), CH2Cl2, reflux, 2 h; c) NaH (1.2 equiv), MeI (1.5 equiv), THF,
RT, 30 min; d) NaOMe (0.5 equiv), MeOH, RT, 1 h; e) O3, MeOH,
ꢀ208C, 15 min, then Me2S (1 equiv), p-TsOH·H2O (0.2 equiv), reflux,
3 h; f) PhSTMS (5 equiv), TMSOTf (1.7 equiv), CH2Cl2, 08C to RT,
30 min; g) DAST (6.5 equiv), NBS (2.2 equiv), CH2Cl2, ꢀ158C, 1 h;
h) 30% aq. NaOH (15 equiv), MeOH, reflux, 20 min; i) FmocCl
(3 equiv), K2CO3 (3 equiv), H2O/THF (7:15), 08C, 1 h; j) AcOH/H2O
(5:1), 958C, 30 min; k) AcCl (16 equiv), DMAP (1.6 equiv), pyridine,
08C to RT, 1 h; l) TIPSOTf (3 equiv), pyridine, RT, overnight; m) AcOH/
H2O (5:1), 908C, 2 h; n) DAST (1.5 equiv), CH2Cl2, 08C, 5 min. DBU=
Scheme 11. Total synthesis of vicenistatin. Reagents and conditions: a)
compound 36 (5 equiv), TMSOTf (0.6 equiv), 4ꢁ M.S., CH2Cl2, ꢀ788C
to RT, 20 min; b) 5m aq. KOH/MeOH (2:3), RT, 7 days; c) compound 38
(3 equiv), TMSOTf (2 equiv), CH2Cl2, ꢀ408C, 2.5 h; d) TBAF
(1.0 equiv), THF, RT, 2 h. M.S.=molecular sieves, TBAF=tetra-n-buty-
lammonium fluoride.
1,8-diazabicycloACHTUNGTRENNUNG[5.4.0]undec-7-ene, DAST=diethylaminosulfur trifluor-
ide, NBS=N-bromosuccinimide, Fmoc=9-fluorenylmethyloxycarbonyl,
TIPS=triisopropylsilyl.
anomer, a-vicenistatin (40), in 28% yield (2 steps). Alterna-
tively, the reaction between compounds 38 and (+)-25 in the
presence of TMSOTf, followed by treatment with tetra-n-
butylammonium fluoride (TBAF), produced vicenistatin (1)
and a-vicenistatin (40) in 12% and 19% yield, respectively,
in two steps.
Synthesis of Analogues of Vicenistatin
Having a concise synthetic route to vicenistatin in hand, we
planned to synthesize analogues of vicenistatin to study
their structure–activity relationships (Scheme 12). We were
especially interested in the effect of the C20- and C23-
methyl groups on the biological activity of vicenistatin; that
is, if the demethyl analogues retained the biological activity,
such easy-to-make analogues would be more useful for bio-
logical studies. These demethyl analogues could be prepared
from the combinatorial coupling of a methylated fragment
(26 or 27) with a demethyl fragment (41 or 42). We also
Scheme 12. Synthetic design of vicenistatin analogues.
planned to evaluate a-glycosides that would be generated
by using the Noyori glycosylation procedure.
The preparation of C20-demethyl enal (+)-41 proceeded
in a straightforward manner (Scheme 13). The common in-
termediate, 6, was oxidized as described previously and the
resulting aldehyde was subjected to Brown allylboration to
&
&
&6
&
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Vicenistatin, a Cytotoxic 20-Membered Macrolactam Glycoside
tions of these substrates with Grubbs first-generation cata-
lyst (31) and p-quinone gave E/Z mixtures of macrolactams
52, 53, and 54 in 49–58% yield, respectively. Fortunately,
the (14Z)-isomers of these macrolactams could be separated
after removal of the Boc group.
At this stage, we noticed that all three of these demethy-
lated macrolactams existed as mixtures of two respective in-
terconverting conformational isomers, as judged from
1
1H NMR experiments (Figure 2). For example, the H NMR
spectrum of compound (ꢀ)-57 at 258C showed two sets of
Scheme 13. Synthesis of C20-demethyl enal (+)-41 and C23-demethyl
phosphonate 42. Reagents and conditions: a) Dess–Martin periodinane
(1.2 equiv), CH2Cl2, RT, 1.5 h; b) (+)-Ipc2BOMe (1.5 equiv), allylmagne-
sium bromide (1.5 equiv), ꢀ788C, 40 min, then crude aldehyde, ꢀ788C to
08C, then MeOH, sat. aq. NaHCO3, 30% aq. H2O2, ꢀ788C to RT, over-
night; c) acrolein (10 equiv), Hoveyda–Grubbs second-generation catalyst
(13, 2 mol%), CH2Cl2, RT, 3 h, 84%; d) tributyl
ACHTUNGERTN(NUNG vinyl)tin (2.5 equiv),
Ph3P (0.5 equiv), [Pd2A(dba)3]·CHCl3 (13 mol%), THF, RT, 1 h; e) TMSCl
CHTUNGTRENNUNG
(2.5 equiv), 2,6-lutidine (5 equiv), CH2Cl2, RT, 4 h; f) Ph3P (1.5 equiv),
phthalimide (1.5 equiv), DIAD (1.5 equiv), ꢀ208C to RT, 10 min;
g) 80% hydrazine (5 equiv), EtOH, reflux, 2 h; h) diethylphosphonoace-
tic acid (21, 4 equiv), EDCI (4 equiv), HOBt (4 equiv), CH2Cl2, RT,
5 min, then Et3N (12 equiv), RT, 15 h.
Figure 2. 1H NMR spectra of macrolactam (ꢀ)-57 at several tempera-
tures.
generate allyl alcohol (ꢀ)-43 in
76% yield. The cross-metathe-
sis reaction of compound (ꢀ)-
43 and acrolein under the opti-
mized conditions in the synthe-
sis of compound (ꢀ)-8 afforded
enal (+)-44 in 84% yield. This
result indicated that the cross-
metathesis reaction was signifi-
cantly influenced by the pres-
ence of the C20-methyl group
(Table 1).[16] The introduction of
a vinyl group and protection of
the C7-hydroxy group proceed-
ed to give C20-demethyl enal
(+)-41 in 85% yield. C23-de-
methyl phosphonate 42 was
also synthesized from 5-hexen-
1-ol (45) without any problems
(92% in three steps).
Combinatorial fragment-as-
sembly by using Horner–Wads-
worth–Emmons olefination fol-
Scheme 14. Synthesis of demethylated macrolactams (ꢀ)-55, (+)-56, and (ꢀ)-57. Reagents and conditions: a)
compounds 42 or 27 (1.1–1.2 equiv), LiHMDS or KHMDS (1.1–1.2 equiv), THF, ꢀ788C to 08C, 20 min;
b) Boc2O (10–35 equiv), Et3N (20–60 equiv), DMAP (1–5 equiv), CH2Cl2, RT, 8–15 h; c) Grubbs first-genera-
tion catalyst (31, 20 mol%), p-quinone (0.4 equiv), DCE, reflux, 1 h; d) TMSOTf (3–5 equiv), 2,6-lutidine (6–
lowed by Boc-protection deliv-
ered hexaenes (ꢀ)-49, (ꢀ)-50,
and (ꢀ)-51 in good yield
(Scheme 14). The RCM reac- 10 equiv), CH2Cl2, RT, 2–9 h.
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
7
&
&
&
These are not the final page numbers! ÞÞ

Naoki Kanoh et al.
FULL PAPER
Table 2. Structure–activity relationships of analogues of vicenistatin.[a]
signals in a ratio of about 3:1; these sets coalesced into
a single set of signals on heating. Interestingly, the ratio of
conformational isomers of these three demethylated macro-
lactams at 258C was almost the same (about 3:1). This sort
of conformational isomers was not observed in the case of
(+)-25, thus suggesting that the C20- and C23-methyl
groups play a role in fixing the conformation of macrolac-
tam to a similar extent.
These three aglycons, (ꢀ)-55, (+)-56, and (ꢀ)-57, were
glycosylated by using glycosyl fluoride 38 to give vicenistatin
analogues 58–63, as shown in Scheme 15. As expected, all of
these analogues existed as a mixture of two conformational
isomers, as shown in Figure 2.
IC50 on cytotoxicity [mm]
Compound
U87
HeLa
3Y1
0.52
1.76
1.69
5.88
4.00
12.8
13.4
11.8
1
1.02
3.51
3.56
6.51
4.81
1.24
12.8
10.3
1.59
3.30
2.45
10.0
5.25
11.8
11.9
13.2
58
59
60
40
61
62
63
[a] Cytotoxicity was evaluated by a WST-8 assay after treatment of the
indicated cells with each compound for 48 h.
effect of the C20- and C23-methyl groups, we hypothesize
that the fixed major conformer of aglycon is the biologically
active conformer of vicenistatin.
Conclusions
We have developed two methods for the synthesis of viceni-
lactam. The first of these syntheses features a rapid assem-
bly of vicenilactam through consecutive HWE and intramo-
lecular-Stille-coupling reactions, but resulted in a low overall
yield of vicenilactam. Based on this result, we designed and
executed
a second-generation synthesis, which utilized
a RCM reaction of hexaene precursor (ꢀ)-30. This second-
generation approach worked well to give not only the de-
sired vicenilactam fragment, (+)-25, but also demethyl de-
rivatives (ꢀ)-55, (+)-56, and (ꢀ)-57. Notably, the RCM re-
actions of polyolefinic compounds, such as (ꢀ)-30, proceed-
ed to give macrolactam 32 in acceptable yield. The vicenisa-
mine part was synthesized by using the Roush procedure.
Although not very efficient, coupling of the vicenisamine
fragments and the vicenilactam fragments gave vicenistatin
(1) and its analogues (40, 58–63) in reproducible yields.
From our NMR analysis and structure–activity-relationship
studies on these compounds, we concluded that the C20-
and C23-methyl groups are important for fixing the confor-
mation of the macrolactam and the fixed major conformer is
hypothesized to be the biologically active conformer. Fur-
ther structure–activity-relationship studies and target-iden-
tification studies on vicenistatin are currently underway.
Scheme 15. Synthesis of demethylated analogues of vicenistatin. Reagents
and conditions: compound 38 (2 equiv), TMSOTf (2 equiv), CH2Cl2,
ꢀ408C, 40 min–2 h; b) TBAF (3 equiv), THF, RT, 1–2 h.
Structure–Activity Relationship of Analogues of
Vicenistatin
The antiproliferative activities of vicenistatin (1) and its syn-
thesized analogues (40, 58–63) were evaluated against sever-
al mammalian cell-lines, including human glioblastoma U87
cells, human cervical cancer HeLa cells, and rat 3Y1 fibro-
blast cells. The results are summarized in Table 2.
Vicenistatin (1) inhibited the proliferation of mammalian
cell-lines with an IC50 value of 0.52–1.59 mm. The removal of
the C20- or C23-methyl groups caused a similar degree of
decrease in the cytotoxicity of the analogues. The effect of
this methyl-group removal on cytotoxic activity was found
to be cumulative: the IC50 value of didemethyl analogue 60
was several times larger than those of dimethyl analogues 58
and 59. The stereochemistry of the anomeric position of
vicenisamine was found to be important—but not crucial—
for determining the biological activities of the products. The
cytotoxic activity of a-anomer 40 was almost the same as
those of demethylated analogues 58 and 59. Overall, these
findings suggest that the methyl groups and stereochemistry
of the vicenisamine anomeric position are important for de-
termining the biological activity of vicenitatin (1). Further-
more, based on the observation of the conformation-fixing
Experimental Section
General Remarks
All reactions were carried out under an argon atmosphere in flame-
dried- or oven-dried glassware (unless otherwise noted) and monitored
by using analytical TLC (Merck Kieselgel 60 F254). Dehydrated THF
and CH2Cl2 were purchased from Kanto Chemical Co., Inc. Other sol-
vents were dehydrated and distilled according to standard procedures.
Reagents were obtained from commercial suppliers and used without fur-
ther purification unless otherwise noted. NMR spectroscopy was per-
formed on JEOL JNM-AL400 (400 MHz), JEOL ECP-500 (500 MHz),
and JEOL ECA-600 spectrometers (600 MHz). Chemical shifts (d) are
&
&
&8
&
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

Vicenistatin, a Cytotoxic 20-Membered Macrolactam Glycoside
reported relative to tetramethylsilane (TMS) for 1H nuclei (CDCl3) and
Synthesis of Vicenistatin (1) and a-Vicenistatin (40) from Compounds 38
and (+)-25
1
to residual CHCl3 (d=7.26 ppm for H and d=77.00 ppm for 13C nuclei),
pyridine (d=7.19 ppm for 1H and d=123.5 ppm for 13C nuclei), and
DMSO (d=2.49 ppm for 1H and d=39.7 ppm for 13C nuclei) as an inter-
nal reference. The following abbreviations are used to indicate the multi-
plicity of the signals: s=singlet, d=doublet, t=triplet, q=quartet, m=
multiplet, sept=septet, br=broad. IR spectroscopy was performed on
JASCO FT/IR-410 and JASCO IR-700. MS was performed on JEOL
JMS-DX303 (EI), JEOL JMS-700 (EI), JEOL JMS-T 100GC (EI), and
JEOL JMS 700 (FAB). Melting points were recorded on a Yazawa BY-2
and are uncorrected. Optical rotations were determined on a JASCO
DIP-370 Digital Polarimeter at room temperature by using the sodium D
line. Column chromatography on silica gel and flash column chromatog-
raphy were carried out with silica gel 60N (Kanto Chemical Co., Inc.,
spherical, neutral, 63–210 mm) and silica gel (Kanto Chemical Co., Inc.,
spherical, neutral, 40–50 mm), respectively.
To a stirring solution of compound (+)-25 (10.5 mg, 24.5 mmol) and com-
pound 38 (40 mg, 73.9 mmol) in CH2Cl2 (1 mL) was added TMSOTf
(8.9 mL, 49 mmol) at ꢀ408C. The reaction mixture was stirred at that tem-
perature for 2.5 h and then quenched with a saturated aqueous solution
of NaHCO3. The resulting mixture was extracted twice with EtOAc. The
combined organic extracts were washed with brine, dried (MgSO4), and
concentrated. This material was used without further purification. To
a stirring solution of the crude glycosyl adduct in THF (2 mL) was added
TBAF (1.0m in THF, 73 mL, 73 mmol) at RT. After 2 h, the mixture was
directly purified by column chromatography on silica gel (MeOH/CHCl3,
1:20) to give compound 1 (1.5 mg, 3.0 mmol, 12% in 2 steps) as a white
powder and compound 40 (2.3 mg, 4.6 mmol, 19% in 2 steps) as a white
powder. Analytical data for these compounds were in good agreement
with those that we reported previously.[10]
Synthesis of Phosphonate (ꢀ)-27
Cytotoxicity Assay
A solution of compound (ꢀ)-28 (178 mg, 733 mmol) and 80% hydrazine
(133 mL, 2.20 mmol) in EtOH (7 mL) was stirred at reflux for 3 h and fil-
tered through Celite. The filtrate was added to 10% aqueous HCl and
stirred at RT for 30 min. The mixture was concentrated and freeze-dried
to give the crude amine hydrochloride salt. To a stirring solution of the
crude amine hydrochloride salt and compound 21 (718 mg, 3.66 mmol) in
CH2Cl2 (15 mL) were added EDCI (702 mg, 3.66 mmol) and HOBt
(495 mg, 3.66 mmol) at RT and the mixture was stirred for 5 min. The so-
lution was supplemented with triethylamine (1.81 mL, 11.0 mmol), stirred
for 10 h, then quenched with water. The resulting mixture was extracted
twice with Et2O. The combined organic extracts were washed with brine,
dried (MgSO4), and concentrated. The residue was purified by column
chromatography on silica gel (EtOAc) to give compound (ꢀ)-27 (208 mg,
0.531 mmol, 72% in 2 steps) as a colorless oil.
(ꢀ)-27: ½aꢁ2D7 =ꢀ1.07 (c=0.48, CHCl3); 1H NMR (400 MHz, CDCl3): d=
6.83 (brs, 1H), 5.79 (ddt, J=17.1, 10.2, 6.6 Hz, 1H), 5.00 (dd, J=17.1,
1.5 Hz, 1H), 4.94 (d, J=10.2 Hz, 1H), 4.14 (quint., J=7.3 Hz, 4H), 3.22
(m, 1H), 3.11 (m, 1H), 2.84 (d, J=20.2 Hz, 2H), 2.16–2.01 (m, 2H), 1.68
(m, 1H), 1.51–1.42 (m, 1H), 1.34 (t, J=7.1 Hz, 6H), 1.28–1.20 (m, 1H),
0.93 ppm (d, J=6.6 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=163.5,
137.8, 113.7, 76.7, 61.9, 44.9, 35.1, 33.8, 32.8, 32.1, 30.5, 16.8, 15.8 ppm; IR
(neat): n˜ =3294, 3076, 2978, 2929, 2872, 1659, 1556, 1443, 1392, 1300,
1245 cmꢀ1; MS (FAB): m/z (%) 292 (100) [M+H]+; HRMS (FAB): m/z
calcd for C13H27NO4P: 292.1678 [M+H]+; found: 292.1684.
3Y1, U87, and HeLa cells were seeded in 96-well culture plates (3ꢂ
103 cells per well) that contained DMEM medium (100 mL per well).
After 24 h, the drugs were added to the wells (final concentration: 100,
30, 10, 3, 1, 0.3, 0.1 mm). After 48 h, WST-8 (10 mL) was added to each
well and the absorbance at 450 nm was measured by using a microplate
reader.
Acknowledgements
This work was supported by a Grant-in-Aid for Scientific Research on
Innovative Areas “Chemical Biology of Natural Products” from the Min-
istry of Education, Culture, Sports, Science and Technology, Japan.
[1] a) K. Kojiri, S. Nakajima, H. Suzuki, H. Kondo, H. Suda, J. Antibiot.
1992, 45, 868–874; b) I. Takahashi, Y. Oda, Y. Nishiie, K. Ochiai, T.
Mizukami, J. Antibiot. 1997, 50, 186–188; c) S. S. Mitchell, B. Nich-
olson, S. Teisan, K. S. Lam, B. C. M. Potts, J. Nat. Prod. 2004, 67,
1400–1402; d) D. W. Udwary, L. Zeigler, R. N. Asolkar, V. Singan,
A. Lapidus, W. Fenical, P. R. Jensen, B. S. Moore, Proc. Natl. Acad.
Sci. USA 2007, 104, 10376–10381; e) Y. Futamura, R. Sawa, Y.
Umezawa, M. Igarashi, H. Nakamura, K. Hasegawa, M. Yamasaki,
E. Tashiro, Y. Takahashi, Y. Akamatsu, M. Imoto, J. Am. Chem.
Soc. 2008, 130, 1822–1823; f) R. Raju, A. M. Piggott, M. M. Conte,
R. J. Capon, Org. Biomol. Chem. 2010, 8, 4682–4689; g) H. Jorgen-
sen, K. F. Degnes, A. Dikiy, E. Fjaervik, G. Klinkenberg, S. B.
Zotchev, Appl. Environ. Microbiol. 2010, 76, 283–293.
Synthesis of Pentaene 32
A solution of compound (ꢀ)-30 (60 mg, 108 mmol), p-quinone (4.7 mg,
43.1 mmol), and Grubbs first-generation catalyst (31; 17.7 mg, 21.5 mmol)
in DCE (108 mL) was stirred at reflux for 1 h. The mixture was cooled to
RT and purified directly by column chromatography on silica gel
(CHCl3) to give compound 32 (37 mg, 69.9 mmol, 65% yield, 14E/14Z
5:1) as a colorless oil.
[2] K. Shindo, M. Kamishohara, A. Odagawa, M. Matsuoka, H. Kawai,
J. Antibiot. 1993, 46, 1076–1081.
[3] H. Arai, Y. Matsushima, T. Eguchi, K. Shindo, K. Kakinuma, Tetra-
hedron Lett. 1998, 39, 3181–3184.
ACHTUNGTRENNUNG
(14E)-32: 1H NMR (500 MHz, CDCl3): d=7.25 (dd, J=14.6, 11.4 Hz,
1H), 6.60 (d, J=14.6 Hz, 1H), 6.31 (dd, J=14.8, 11.0 Hz, 1H), 6.18 (dd,
J=15.0, 11.4 Hz, 1H), 5.87 (dd, J=15.0, 9.2 Hz, 1H), 5.68 (d, J=11.0 Hz,
1H), 5.41 (ddd, J=14.8, 8.7, 5.7 Hz, 1H), 4.97 (dd, J=7.2, 7.2 Hz, 1H),
4.03 (dd, J=13.6, 9.6 Hz, 1H), 3.34 (dt, J=2.0, 8.4 Hz, 1H), 3.26 (dd, J=
13.8, 5.4 Hz, 1H), 2.63 (d, J=15.0 Hz, 1H), 2.55 (d, J=15.0 Hz, 1H),
2.31–2.19 (m, 3H), 2.10 (m, 1H), 2.00 (m, 1H), 1.85 (m, 1H), 1.78 (s,
3H), 1.57–1.35 (m, 2H), 1.51 (s, 12H), 1.03 (d, J=6.8 Hz, 3H), 0.89 (d,
J=6.8 Hz, 3H), 0.16 ppm (s, 9H); 13C NMR (125 MHz, CDCl3): d=
169.3, 153.7, 145.5, 143.8, 134.9, 133.5, 131.5, 128.2, 127.9, 127.0, 123.4,
120.8, 82.4, 77.1, 49.3, 48.6, 47.2, 37.8, 32.6, 31.9, 29.7, 28.1, 19.4, 17.6,
17.3, 17.0, 0.4 ppm; IR (neat): n˜ =2958, 2926, 1725, 1677, 1368, 1251,
1145, 1078, 841 cmꢀ1; MS (EI): m/z 529 [M]+, 429 (100); HRMS (EI): m/
z calcd for C31H51NO4Si: 529.3585 [M]+; found: 529.3609.
[4] a) Y. Matsushima, H. Itoh, T. Eguchi, K. Kakinuma, J. Antibiot.
1998, 51, 688–691; b) Y. Matsushima, T. Nakayama, S. Tohyama, T.
Eguchi, K. Kakinuma, J. Chem. Soc., Perkin Trans. 1 2001, 569–577;
c) Y. Matsushima, H. Itoh, T. Nakayama, S. Horiuchi, T. Eguchi, K.
Kakinuma, J. Chem. Soc., Perkin Trans. 1 2002, 949–958.
[5] a) M. Otsuka, T. Eguchi, K. Shindo, K. Kakinuma, Tetrahedron Lett.
1998, 39, 3185–3188; b) M. Otsuka, M. Fujita, Y. Matsushima, T.
Eguchi, K. Shindo, K. Kakinuma, Tetrahedron 2000, 56, 8281–8286;
c) H. Nishida, T. Eguchi, K. Kakinuma, Tetrahedron 2001, 57, 8237–
8242.
[6] a) Y. Ogasawara, K. Katayama, A. Minami, M. Otsuka, T. Eguchi,
K. Kakinuma, Chem. Biol. 2004, 11, 79–86; b) Y. Ogasawara, K. Ka-
kinuma, T. Eguchi, J. Antibiot. 2005, 58, 468–472; c) F. Kudo, T. Ki-
tayama, K. Kakinuma, T. Eguchi, Tetrahedron Lett. 2006, 47, 1529–
1532; d) S. Tohyama, K. Kakinuma, T. Eguchi, J. Antibiot. 2006, 59,
44–52; e) Y. Shinohara, F. Kudo, T. Eguchi, J. Am. Chem. Soc. 2011,
133, 18134–18137.
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
9
&
&
&
These are not the final page numbers! ÞÞ

Naoki Kanoh et al.
FULL PAPER
[7] a) A. Minami, K. Kakinuma, T. Eguchi, Tetrahedron Lett. 2005, 46,
6187–6190; b) A. Minami, R. Uchida, T. Eguchi, K. Kakinuma, J.
Am. Chem. Soc. 2005, 127, 6148–6149; c) A. Minami, T. Eguchi, J.
Am. Chem. Soc. 2007, 129, 5102–5107.
[8] Y. Matsushima, T. Nakayama, M. Fujita, R. Bhandari, T. Eguchi, K.
Shindo, K. Kakinuma, J. Antibiot. 2001, 54, 211–219.
[9] N. C. Gassner, C. M. Tamble, J. E. Bock, N. Cotton, K. N. White, K.
Tenney, R. P. St. Onge, M. J. Proctor, G. Giaever, C. Nislow, R. W.
Davis, P. Crews, T. R. Holman, R. S. Lokey, J. Nat. Prod. 2007, 70,
383–390.
[18] A. Darwish, A. Lang, T. Kim, J. M. Chong, Org. Lett. 2008, 10, 861–
864.
[19] This type of cine-substitution reaction was observed in the Stille
coupling between 1-substituted vinylstannane and various unsaturat-
ed halides/triflates; see: Y. Peng, W.-D. Z. Li, Eur. J. Org. Chem.
2010, 6703.
[20] K. C. Nicolaou, F. Murphy, S. Barluenga, T. Ohshima, H. Wei, J. Xu,
D. L. F. Gray, O. Baudoin, J. Am. Chem. Soc. 2000, 122, 3830–3838.
[21] S. H. Hong, D. P. Sanders, C. W. Lee, R. H. Grubbs, J. Am. Chem.
Soc. 2005, 127, 17160–17161.
[10] H. Fukuda, S. Nakamura, T. Eguchi, Y. Iwabuchi, N. Kanoh, Synlett
2010, 2589–2592.
[22] K. C. Nicolaou, S. P. Seitz, D. P. Papahatjis, J. Am. Chem. Soc. 1983,
105, 2430–2434.
[11] W. R. Roush, D. Gustin, Tetrahedron Lett. 1994, 35, 4931–4934.
[12] W. R. Roush, R. A. James, Aust. J. Chem. 2002, 55, 141–146.
[13] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155–4156.
[14] H. C. Brown, K. S. Bhat, R. S. Randad, J. Org. Chem. 1989, 54,
1570–1576.
[23] K. C. Nicolaou, R. E. Dolle, D. P. Papahatjis, J. L. Randall, J. Am.
Chem. Soc. 1984, 106, 4189–4192.
[24] T. Mukaiyama, T. Takashima, M. Katsurada, H. Aizawa, Chem. Lett.
1991, 533–536.
[25] S. Hashimoto, M. Hayashi, R. Noyori, Tetrahedron Lett. 1984, 25,
1379–1382.
[15] P. Wipf, S. Lim, Angew. Chem. 1993, 105, 1095–1097; Angew. Chem.
Int. Ed. Engl. 1993, 32, 1068–1071.
[16] M. Lautens, M. L. Maddess, Org. Lett. 2004, 6, 1883–1886.
[17] L. A. Paquette, M. Duan, I. Konetzki, C. Kempmann, J. Am. Chem.
Soc. 2002, 124, 4257–4270.
Received: July 10, 2012
Revised: August 14, 2012
Published online: && &&, 0000
&
&
&10
&
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 0000, 00, 0 – 0
ÝÝ These are not the final page numbers!

FULL PAPER
Antitumor Agents
Hayato Fukuda, Yuko Nishiyama,
Shiina Nakamura, Yutaro Ohno,
Tadashi Eguchi, Yoshiharu Iwabuchi,
Takeo Usui, Naoki Kanoh* &&&& — &&&&
An enantioselective total synthesis of
vicenistatin (1) was accomplished by
using an intermolecular Horner–Wads-
worth–Emmons reaction and a ring-
closing metathesis as key steps. This
strategy made it possible to prepare
synthetic analogues of vicenistatin,
including the C20- and/or C23-
demethyl vicenistatins. Evaluation of
their cytotoxicity indicated the impor-
tance of the fixed conformation of
aglycon in determining biological
activity.
Synthesis and Structure–Activity Rela-
tionship of Vicenistatin, a Cytotoxic
20-Membered Macrolactam Glycoside
Chem. Asian J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
11
&
&
&
These are not the final page numbers! ÞÞ