Total synthesis and biological evaluation of (+)-gambieric acid A and its analogues

Article abstract of DOI:10.1002/chem.201204303

In this study, we report the first total synthesis and complete stereostructure of gambieric acid A, a potent antifungal polycyclic ether metabolite, in detail. The A/B-ring exocyclic enol ether 32 was prepared through a Suzuki-Miyaura coupling of the B-ring vinyl iodide 18 and the alkylborate 33 and subsequent closure of the A-ring by using diastereoselective bromoetherification as the key transformation. Suzuki-Miyaura coupling of 32 with acetate-derived enol phosphate 49, followed by ring-closing metathesis of the derived diene, produced the D-ring. Subsequent closure of the C-ring through a mixed thioacetalization completed the synthesis of the A/BCD-ring fragment 8. The A/BCD- and F′GHIJ-ring fragments (i.e., 8 and 9) were assembled through Suzuki-Miyaura coupling. The C25 stereogenic center was elaborated by exploiting the intrinsic conformational property of the seven-membered F′-ring. After the oxidative cleavage of the F′-ring, the E-ring was formed as a cyclic mixed thioacetal (i.e., 70) and then stereoselectively allylated by using glycosylation chemistry. Ring-closing metathesis of the diene 3 thus obtained closed the F-ring and completed the polycyclic ether skeleton. Finally, the J-ring side chain was introduced by using a Julia-Kocienski olefination in the presence of CeCl₃ to complete the total synthesis of gambieric acid A (1), thereby unambiguously establishing its complete stereostructure. The present total synthesis enabled us to evaluate the antifungal and antiproliferative activities of 1 and several synthetic analogues. Copyright

Full text of DOI:10.1002/chem.201204303

DOI: 10.1002/chem.201204303
Total Synthesis and Biological Evaluation of (+)-Gambieric Acid A
and Its Analogues
Kazuya Ishigai,[a] Haruhiko Fuwa,*[a] Keisuke Hashizume,[a] Ryo Fukazawa,[a]
Yuko Cho,[b] Mari Yotsu-Yamashita,[b] and Makoto Sasaki*[a]
Abstract: In this study, we report the
first total synthesis and complete ster-
eostructure of gambieric acid A, a
potent antifungal polycyclic ether me-
tabolite, in detail. The A/B-ring exocy-
clic enol ether 32 was prepared through
a Suzuki–Miyaura coupling of the B-
ring vinyl iodide 18 and the alkylborate
33 and subsequent closure of the A-
ring by using diastereoselective bro-
moetherification as the key transforma-
tion. Suzuki–Miyaura coupling of 32
with acetate-derived enol phosphate
49, followed by ring-closing metathesis
of the derived diene, produced the D-
ring. Subsequent closure of the C-ring
through a mixed thioacetalization com-
pleted the synthesis of the A/BCD-ring
fragment 8. The A/BCD- and F’GHIJ-
ring fragments (i.e., 8 and 9) were as-
sembled through Suzuki–Miyaura cou-
pling. The C25 stereogenic center was
elaborated by exploiting the intrinsic
conformational property of the seven-
membered F’-ring. After the oxidative
cleavage of the F’-ring, the E-ring was
formed as a cyclic mixed thioacetal
(i.e., 70) and then stereoselectively al-
lylated by using glycosylation chemis-
try. Ring-closing metathesis of the
diene 3 thus obtained closed the F-ring
and completed the polycyclic ether
skeleton. Finally, the J-ring side chain
was introduced by using a Julia–Ko-
cienski olefination in the presence of
CeCl3 to complete the total synthesis
of gambieric acid A (1), thereby unam-
biguously establishing its complete ster-
eostructure. The present total synthesis
enabled us to evaluate the antifungal
and antiproliferative activities of 1 and
several synthetic analogues.
Keywords: natural products · poly-
cyclic ethers
relationships
·
·
structure–activity
Suzuki–Miyaura
coupling · total synthesis
Introduction
Moreover, Hirama and co-workers have reported that GAA
only weakly inhibits the binding of tritiated dihydrobreve-
toxin B ([3H]PbTx-3) to site 5 of the voltage-gated sodium
ion channels of excitable membranes.[6] Instead, the Nagai/
Yasumoto group has shown that GAs exhibit potent antifun-
gal activity against Aspergillus niger[5] and also suggested the
possible role of GAA as an endogenous growth regulator of
G. toxicus.[7]
Gambieric acids (GAs) A–D were isolated as antifungal
substances from a culture medium of the ciguatera causative
dinoflagellate, Gambierdiscus toxicus, by Nagai, Yasumoto,
and co-workers.[1] The skeletal structure of GAs consists of
a nonacyclic polyether core involving six-, seven-, and nine-
membered cyclic ethers, arranged with an isolated tetrahy-
drofuran ring (A-ring). GAs share the structural characteris-
tics common to those of polycyclic ether neurotoxins, which
are produced by G. toxicus, for example, ciguatoxin, maito-
toxin, and gambierol.[2–4] However, Nagai and co-workers
have reported that gambieric acid A (GAA, 1) shows no
toxicity in mice at a dose of 1 mgkgÀ1 (ip injection).[1,5]
The gross structure and relative stereochemistry of the
polycyclic ether domain of GAs were established by 2D
NMR spectroscopic analysis.[1] The complete stereostructure
was later proposed on the basis of degradation experiments,
application of chiral anisotropic reagents, chiral HPLC anal-
ysis, and NMR-based conformational analysis on gambieric
acid B.[8] Satake and co-workers have determined the abso-
lute configuration of the C9 stereogenic center by using the
modified Mosherꢀs method,[9] and the relative stereochemis-
try of C7/C9 and C9/C11 stereogenic centers was proposed
[a] K. Ishigai, Prof. Dr. H. Fuwa, K. Hashizume, R. Fukazawa,
Prof. Dr. M. Sasaki
Graduate School of Life Sciences
Tohoku University, 2-1-1 Katahira
Aoba-ku, Sendai 980-8577 (Japan)
Fax : (+81)22-217-6214
E-mail: hfuwa@m.tohoku.ac.jp
masasaki@m.tohoku.ac.jp
3
on the basis of JH,H values and nuclear Overhauser effect
(NOE) correlations. However, our previous studies on the
synthesis and NMR spectroscopic analysis of the A/B-ring
model compounds of GAs strongly suggested that the rela-
tive stereochemical relationship between the C9 and C11
stereogenic centers was erroneously assigned in the original-
ly proposed structure, and the absolute configuration of the
polycyclic ether domain of GAs must be opposite to that of
the proposed structure.[10] However, our revised structure of
GAs must be verified unambiguously by total synthesis.
[b] Dr. Y. Cho, Prof. Dr. M. Yotsu-Yamashita
Graduate School of Agricultural Science
Tohoku University, 1-1 Tsutsumidori-amamiya
Aoba-ku, Sendai 981-8555 (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201204303.
5276
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5276 – 5288

FULL PAPER
Stimulated by the structural and biological aspects of
GAs, we embarked on the total synthesis of GAA (1), the
representative congener of GAs.[11,12] Herein, we describe in
detail the first total synthesis of GAA (1), culminating in
the establishment of its complete stereostructure
(Figure 1).[13] The salient features of our total synthesis of 1
ploiting the intrinsic conformational bias of the “unnatural”
F’-ring; 3) stereoselective allylation of the E-ring cyclic
mixed thioacetal using glycosylation chemistry; and 4) modi-
fied Julia–Kocienski olefination[15] to form the trisubstituted
olefin within the J-ring side chain. Furthermore, we evaluat-
ed the antifungal and antiproliferative activities of GAA
and its synthetic analogues, providing preliminary informa-
tion on the structure–activity relationships of this intriguing
natural product.
Results and Discussion
Synthetic route toward GAA (1): Our synthetic route
toward GAA (1) is illustrated in a retrosynthetic format in
Scheme 1. The target molecule 1 could be obtained from
polycyclic ether 2 through introduction of the J-ring side
chain by using the Julia–Kocienski olefination[15] in a later
stage. Considering the extraordinarily complex molecular ar-
chitecture, we were completely aware from the outset that it
was necessary to devise a convergent synthetic approach ca-
pable of building the entire polycyclic ether skeleton of 1,
preferably by assembling two polycyclic ether fragments of
approximately equal complexity at the central domain of
the molecule. Because it was evident that ring-closing meta-
thesis (RCM)[16] is a suitable method for the construction of
the nine-membered F-ring, the most significant challenge
was how to synthesize the RCM precursor, an O-linked
diene 3, in a convergent fashion.
Figure 1. Structure of gambieric acid A.
include 1) convergent union of the two advanced polycyclic
ether fragments of comparable structural complexity (the A/
BCD- and F’GHIJ-ring fragments, 8 and 9, respectively;
Scheme 1) through Suzuki–Miyaura coupling;[14] 2) stereo-
controlled elaboration of the C25 stereogenic center by ex-
Scheme 1. Synthetic route toward 1. NAP = 2-Naphthylmethyl, TMS=trimethylsilyl.
Chem. Eur. J. 2013, 19, 5276 – 5288
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5277

H. Fuwa, M. Sasaki et al.
Scheme 2. Our earlier attempts at the construction of fused seven- and nine-membered ring systems. AIBN = 2,2’-Azobisisobutyronitril, Bn=benzyl,
MOM = methoxymethyl.
In our previous studies, we have successfully demonstrat-
ed the synthesis of the fused seven- and nine-membered
ring system of ciguatoxins, which is closely related to the
central ring system of 1, through radical cyclization of acryl-
ate-tethered O,S- or O,Se-acetal (10 or 11, respectively; see
Scheme 2A).[17] However, our model studies on applying the
radical cyclization chemistry to GAs were not fruitful be-
cause of difficulties in controlling the C25 and C26 stereo-
genic centers (Scheme 2B).[12a,c]
Accordingly, we established a new strategy for the synthe-
sis of the central domain of GAs by exploiting Suzuki–
Miyaura coupling as the key transformation.[12e,f] The diene
3 could be obtained from the ester 5 through Yamaguchi
lactonization[18] and stereoselective allylation at the C26 po-
sition. We envisioned that the ester 5 would be obtained
from the ketone 6 through oxidative cleavage of the F’-ring,
which is absent in the natural product. Here the seven-mem-
bered F’-ring was introduced for the following two purposes:
1) to control the stereochemistry at the C25 position by
taking advantage of the intrinsic conformational property of
seven-membered cyclic ethers and 2) to facilitate the union
of cyclic ether fragments by Suzuki–Miyaura coupling.[19–22]
Thus, the ketone 6 should be available from the endocyclic
enol ether 7 through a sequence of hydroboration and oxi-
dation. In turn, the compound 7 could be synthesized from
the A/BCD-ring fragment 8 and F’GHIJ-ring fragment 9
through Suzuki–Miyaura coupling.
Synthesis of the B-ring fragment 18: The synthesis of the B-
ring fragment 18 started with the known alcohol 19[23]
(Scheme 3). A three-step sequence involving acidic cleavage
of the acetonide, benzylidene-acetal protection of the 1,3-
diol moiety of the resultant triol, and subsequent hydrobora-
tion of the terminal olefin gave the diol 20. Oxidative lacto-
Scheme 3. Synthesis of the B-ring fragment 18. CSA=(Æ)-10-Camphorsulfonic acid, Cy=cyclohexyl, MS=molecular sieves, NBSH=o-nitrobenzenesul-
fonyl hydrazide, Tf=trifluoromethanesulfonyl, thexyl=2,3-dimethyl-2-butyl.
5278
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5276 – 5288

(+)-Gambieric Acid A and Its Analogues
FULL PAPER
nization of 20 with 2,2,6,6-tetramethylpiperidin-1-oxyl
[24]
(TEMPO)/PhIACHTUNGTRENNUNG(OAc)2 directly gave the seven-membered
lactone 21. Treatment of 21 with potassium bis(trimethylsily-
l)amide (KHMDS) in the presence of (PhO)2P(O)Cl[25] gave
the corresponding enol phosphate, whose palladium-cata-
lyzed methoxycarbonylation ([Pd
MeOH/DMF, 508C) gave the a,b-unsaturated ester 22
(86%, two Diisobutylaluminum hydride
steps).[26]
ACHTUGNTRNE(UNNG PPh3)4], CO, Et3N,
(DIBALH) reduction of 22, followed by silylation of the re-
sultant hydroxy group, gave the tert-butyldimethylsilyl
(TBS) ether 23 (71%, two steps). Hydroboration of 23 with
thexylborane proceeded in a highly stereoselective manner
to provide the alcohol 24 in 80% yield (diastereomer ratio
(d.r.)>20:1). A two-step protective group manipulation se-
quence gave the alcohol 25 (82%, two steps). Oxidation and
Wittig methylenation yielded the olefin 26 (90%, two
steps). Hydroboration of the olefin 26, followed by one-pot
oxidation/Wittig reaction, gave an a,b-unsaturated ester,
which was reduced with DIBALH to afford the allylic alco-
hol 27 (86%, three steps). Sharpless asymmetric epoxidation
of 27 by using (+)-diethyl tartrate (DET) gave the epoxy al-
cohol 28 (94%, d.r. 10:1). Chlorination of 28 with N-chloro-
succinimide (NCS)/Ph3P and subsequent treatment of the
derived chloro-epoxide 29 with lithium diisopropylamide
(LDA)[27] afforded the propargylic alcohol 30 in 89% yield.
At this stage, the minor C9 epimer was removed by flash
column chromatography on silica gel. Iodination of 30 with
NIS (N-iodosuccinimide)/AgNO3 (98%), followed by dii-
mide reduction[28] of the resultant iodoalkyne 31, gave the
(Z)-vinyl iodide 18 in 83% yield. The stereochemistry of the
C9 position was unambiguously determined by derivatiza-
tion of the alcohol 30 and the application of the modified
Mosherꢀs method.[9,29]
Scheme 4. Synthesis of the A/B-ring fragment 32. 9-BBN=9-Borabicyclo-
AHCTUNGERTG[NNUN 3.3.1]nonyl, TBAI=tetra-n-butylammonium iodide, TBDPS=tert-butyl-
diphenylsilyl.
Here we masked the C1, C9, and C12 hydroxy groups as
their NAP ethers; it has been shown that the NAP group
tolerates strong acids and bases but can be selectively re-
moved under mild conditions by using DDQ.[31] The primary
alcohol 40 thus obtained was smoothly converted to the A/
B-ring exocyclic enol ether 32 by iodination and base treat-
ment.
Synthesis of the A/B-ring fragment 32: Next, we performed
a Suzuki–Miyaura coupling of the alkylborate 33 and the
vinyl iodide 18 (Scheme 4). The alkylborate 33, generated in
situ from the iodide 34[29] (tBuLi, B-methoxy-9-borabicyclo-
A
Synthesis of the D-ring enol phosphate 42 and enol triflate
43: Synthesis of the D-ring fragments 42 and 43 is summar-
ized in Scheme 5. Reductive removal of the benzyl group
from the known diol 44[12h] with lithium naphthalenide[32]
gave the triol 45 in 99% yield. Protection of the 1,3-diol
moiety of 45 as its benzylidene acetal and subsequent treat-
ment with tetra-n-butylammonium fluoride (TBAF) afford-
ACHTUNGTRENNUNG
no)ferrocene), Ph3As, Cs2CO3, aqueous THF/DMF, 508C) to
give the olefin 35 quantitatively. The p-methoxyphenylmeth-
yl (MPM) group of 35 was removed by 2,3-dichloro-5,6-di-
cyano-1,4-benzoquinone (DDQ) oxidation (99%), and the
resultant alcohol 36 was treated with N-bromosuccinimide
(NBS) in acetonitrile, which gave the bromide 37 in 97%
yield (d.r.>20:1). The stereochemical outcome of the bro-
moetherification was in accordance with that observed in
our previous model study.[10b] Reduction of 37 with Ph3SnH/
AIBN delivered the compound 38 in 95% yield. At this
stage, the relative stereochemistry of the A-ring tetrahydro-
furan was unambiguously confirmed by NOE experiments
as shown.
ed the diol 46 (55%, two steps). Oxidative lactonization of
[33]
46 with TEMPO/PhIACHTNUGTRNEUNG(OAc)2 gave the lactone 47 (72%),
which was transformed into the enol phosphate 42 under
standard conditions (KHMDS, (PhO)2P(O)Cl, hexamethyl-
phosphoramide (HMPA), THF, À788C).[25] The enol triflate
43 was readily prepared from the lactone 47 (KHMDS,
PhNTf2, HMPA, THF, À788C).[34]
Suzuki–Miyaura coupling of the A/B-ring exocyclic enol
ether 32 and the lactone-derived D-ring enol phosphate 42
or enol triflate 43: We investigated Suzuki–Miyaura cou-
pling of the A/B-ring exocyclic enol ether 32 and the D-ring
Having established the A-ring tetrahydrofuran with com-
plete stereocontrol, we performed routine protective group
manipulations before the elaboration of the C- and D-rings.
Chem. Eur. J. 2013, 19, 5276 – 5288
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5279

H. Fuwa, M. Sasaki et al.
tone-derived enol triflates ([Pd2ACHTNUGTRENUNG(dba)3] (dba=tris(dibenzyli-
deneacetone)/Ph3As, Cs2CO3, aqueous THF/DMF, RT)[20a]
gave 48 in only 15% yield (Table 1, entry 5). Changing the
base to NaHCO3 slightly improved the yield of 48, but the
result remained unsatisfactory (Table 1, entry 6). The
[PdCl2ACHTUNGTRENUNG(dppf)·CH2Cl2]/Ph3As-catalyst system was completely
ineffective (Table 1, entry 7). Thus, we could not synthesize
the endocyclic enol ether 48 in synthetically useful quanti-
ties.
Synthesis of endocyclic enol ether 48 via Suzuki–Miyaura
coupling/RCM sequence: We have previously reported the
synthesis of endocyclic enol ethers based on a Suzuki–
Miyaura coupling/RCM sequence, which utilized alkylbor-
anes and acetate-derived enol phosphates as coupling part-
ners.[35,36] This methodology has been shown to be particular-
ly useful when it is difficult to prepare the corresponding
lactone-derived enol triflates/phosphates, and/or they are in-
sufficiently reactive toward palladium-catalyzed reactions.
As shown in Scheme 6, this complementary methodology
may enable the synthesis of the endocyclic enol ether 48
from the A/B-ring fragment 32 and the acetate-derived enol
phosphate 49.
Scheme 5. Synthesis of the D-ring enol phosphate 42 and enol triflate 43.
Table 1. Investigation of the Suzuki–Miyaura coupling.[a]
Entry
D-ring
Pd catalyst
Ligand
Base
Yield
[%]
1[b]
2[b]
3[b]
4[b]
5[c]
6[c]
7[c]
42
42
42
42
43
43
43
[Pd
[Pd
A
U
none
none
none
Ph3As
Ph3As
Ph3As
Ph3As
Cs2CO3
NaHCO3
Cs2CO3
Cs2CO3
Cs2CO3
NaHCO3
NaHCO3
21
25
0
E
N
E
0
E
E
15
26
0
T
CHTUNGTRENNUNG
A
CHTUNGTRENNUNG
[a] All reactions were performed by using 2.6 equivalents of 9-BBN-H, a
Pd catalyst (10 mol%), 2 equivalents of 42 or 43, and 3 equivalents of
base in THF/DMF. Yields are based on the exo-olefin 32. [b] Reaction
performed at 508C. [c] Reaction performed at RT.
Scheme 6. Alternative synthetic strategy for the endocyclic enol ether 48.
enol phosphate 42 in detail (Table 1). We initially examined
the Suzuki–Miyaura coupling by using Cs2CO3 as a base and
[PdACHTUNGTRENNUNG(PPh3)4] as a catalyst in aqueous THF/DMF at 508C. Un-
fortunately, the desired endocyclic enol ether 48 was isolated
in only 21% yield (Table 1, entry 1). Changing the base to
NaHCO3 did not improve the yield of 48 (Table 1, entry 2),
The synthesis of the enol phosphate 49 is shown in
Scheme 7. Removal of the acetonide group of the known al-
cohol 51,[37] protection of the 1,3-diol moiety of the resultant
triol with benzylidene acetal, and subsequent acylation of
the resulting alcohol led to the acetate 52 in 68% yield for
the three steps. Enolization of 52 with KHMDS in presence
of (PhO)2P(O)Cl (HMPA, THF, À788C) gave the enol
phosphate 49.
Hydroboration of the A/B-ring exocyclic enol ether 32
with 9-BBN-H, followed by in situ reaction of the derived
alkylborane with the enol phosphate 49 ([PdACHTUNGTRENNUNG(PPh3)4],
Cs2CO3, aqueous THF/DMF, 508C), proceeded smoothly to
afford the diene 50. This diene was immediately exposed to
the Grubbs second-generation catalyst (G-II)[38] in toluene
whereas the use of [PdCl2ACHTNUTRGNE(UNG dppf)·CH2Cl2] alone or in combi-
nation with Ph3As did not afford 48 (Table 1, entries 3 and
4). From these results, we considered that the low reactivity
of the enol phosphate 42 might be the reason for the unsuc-
cessful outcome. Hence, we examined the use of the more
reactive enol triflate 43 as the coupling partner. However,
we were disappointed to find that Suzuki–Miyaura coupling
using 43 under the conditions specifically optimized for lac-
5280
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5276 – 5288

(+)-Gambieric Acid A and Its Analogues
FULL PAPER
Scheme 7. Synthesis of the endocyclic enol ether 48. DMAP=4-Dimethy-
laminopyridine, Ts=p-toluenesulfonyl.
at 708C to give the desired enol ether 48 in 73% yield for
the two steps.
Synthesis of the A/BCD-ring fragment 8: Stereoselective hy-
droboration of 48 with BH3·SMe2, followed by alkaline oxi-
dative workup, gave an alcohol, which was oxidized by
tetra-n-propylammonium perruthenate (TPAP)/N-methyl-
morpholine N-oxide (NMO)[39] to provide the ketone 53
(80%, two steps, d.r.>20:1) (Scheme 8). The stereochemis-
try of the C18 stereogenic center of 53 was confirmed by an
NOE experiment, as shown. After the removal of the TBS
Scheme 8. Synthesis of the A/BCD-ring fragment 8. NaHMDS=Sodium
bis(trimethylsilyl)amide.
treatment with LiEt3BH (THF, 08C), gave the tertiary alco-
hol 59 in 97% yield for the two steps. The relative stereo-
chemistry of the C21 stereogenic center was determined by
a rotating-frame Overhauser effect (ROE) experiment, as
shown. Finally, a four-step sequence that involved removal
of the TIPS group, oxidation of the resultant primary alco-
hol, Wittig methylenation, and silylation of the tertiary alco-
hol gave the A/BCD-ring fragment 8 (76%, four steps).
group with HF·pyridine, treatment of the resultant hemiace-
[40a]
tal with EtSH/Zn
G
promoted the mixed thioacetali-
zation and spontaneous cleavage of the benzylidene acetal.
Subsequent acetylation of the liberated hydroxy groups gave
the diacetate 54 (76%, three steps). The C19 axial methyl
group was stereoselectively installed by the one-pot oxida-
tion/methylation protocol (meta-chloroperoxybenzoic acid
(mCPBA), CH2Cl2, À788C; then excess Me3Al, 08C)[40b] to
give the tetracyclic ether 55 in 91% yield (d.r.>20:1). The
stereochemistry of the C19 stereogenic center was establish-
ed by an NOE experiment, as shown.
Toward the formation of the A/BCD-ring fragment 8, the
C21 axial methyl group was introduced according to the
methodology previously developed in our laboratory.[40a]
After removal of the acetyl groups of 55, the primary hy-
droxy group was selectively protected as the triisopropylsilyl
(TIPS) ether 56 (95%, two steps). Oxidation of 56 with
TPAP/NMO[39] and Wittig methylenation of the derived
ketone gave the exo-olefin 57. Dihydroxylation of 57 with
OsO4/NMO exclusively gave the diol 58 in 86% yield for
the three steps (d.r.>20:1). Tosylation of 58, followed by
Synthesis of the F’GHIJ-ring fragment 9: The synthesis of
the F’GHIJ-ring enol phosphate 9 started with the previous-
ly reported tetracyclic ether 60[12g,h] (Scheme 9). A sequence
of routine protective group manipulations gave the alcohol
62. This alcohol was then converted to the methyl ketone 63
in three steps (82% overall yield). Wittig methylenation of
63, followed by stereoselective hydroboration of the derived
olefin with dicyclohexylborane, gave the alcohol 64 (87%,
two steps, d.r.>20:1).[41] Oxidation of 64, Wittig reaction of
the derived aldehyde by using Ph3P=CHCO2Et, and hydro-
genation of the resultant enoate delivered the ester 65
(92%, three steps). Reduction of 65 with DIBALH, removal
of the TBS group, and TEMPO/PhIACHTUNRGTNEUNG(OAc)2-mediated oxida-
Chem. Eur. J. 2013, 19, 5276 – 5288
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5281

H. Fuwa, M. Sasaki et al.
Scheme 9. Synthesis of the F’GHIJ-ring fragment 9. Piv=Pivaloyl, PPTS=pyridinium p-toluenesulfonate.
tive lactonizaton[24] afforded the seven-membered lactone 66
(76%, three steps). At this stage, the configuration of the
C30 stereogenic center was determined by a ROE experi-
ment, as shown. Finally, treatment of 66 with KHMDS/
(PhO)2P(O)Cl gave the F’GHIJ-ring enol phosphate 9,
which was immediately used in subsequent Suzuki–Miyaura
coupling.
sider that the energy difference between TS-A and TS-B is
likewise small. Consequently, the hydroboration would pro-
duce a mixture of the desired and undesired alcohols. Mean-
while, two conformers C and D can similarly be considered
for 23, but in this case the energy difference between these
conformers was estimated to be quite large (12 kJmolÀ1 by
ab initio calculations). Accordingly, TS-C must be signifi-
cantly lower in energy than TS-D, and it is most conceivable
that the hydroboration of 23 would preferentially proceed
via TS-C to give the desired alcohol. Despite the disappoint-
ing diastereoselectivity of the hydroboration of 7, oxidation
of the resultant diastereomeric mixture of alcohols with
Dess–Martin periodinane (DMP)[46] and ensuing treatment
Convergent synthesis of the polycyclic ether skeleton 2:
After completing the formation of the A/BCD-ring frag-
ment 8 and F’GHIJ-ring fragment 9, we focused on their as-
sembly and the construction of the entire polycyclic ether
skeleton of GAA (1). Hydroboration of 8 with 9-BBN-H
gave the corresponding alkylborane, which was coupled in
situ with the enol phosphate 9 under our previously opti-
with excess 1,8-diazabicycloACHTNUTRGNE[UNG 5.4.0]undec-7-ene (DBU) in tol-
uene heated at reflux allowed us to enrich the thermody-
namically favored 6 in 73% yield from 7 (d.r. 18:1).[12e,f]
Having successfully secured the C25 stereogenic center,
the F’-ring was oxidatively cleaved prior to the elaboration
of the E- and F-rings. Enolization of the ketone 6 with lithi-
um bis(trimethylsilyl)amide (LHMDS) in the presence of
trimethylsilyl chloride (TMSCl)/Et3N (THF, À78 to 08C)
gave the corresponding enol silyl ether, which was immedi-
ately oxidized with OsO4/NMO to afford the a-hydroxy
ketone 67 as a 1:1 mixture of inconsequential diastereomers
mized conditions ([PdCl2ACHTUNGTRENUNG(dppf)·CH2Cl2], Cs2CO3, aqueous
THF/DMF, 508C)[21a] to give the endocyclic enol ether 7 in
95% yield (Scheme 10). Hydroboration of 7 with BH3·SMe2,
followed by oxidative workup under basic conditions, gave a
2:1 mixture of diastereomeric alcohols. The disappointingly
low stereoselectivity of hydroboration was in sharp contrast
to that observed for the B-ring endocyclic enol ether 23
(Scheme 3). Our plausible explanation for the observed ster-
eoselectivity of the hydroboration of 7 and 23 is summarized
in Figure 2. In the case of 7, two possible conformers A and
B can be considered, and the energy difference between
these conformers is likely to be small (2.9 kJmolÀ1 as esti-
mated by ab initio calculations on a model compound).[42]
By avoiding unfavorable steric repulsion,[43] hydroboration
would occur from the upper face of the conformer A or
from the lower face of the conformer B, leading to the tran-
sition state TS-A or TS-B, respectively. Since it is known
that hydroboration proceeds via a four-membered transition
state that occurs very early along the reaction coordinate,[44]
the transition state of hydroboration must reflect the ground
state conformation of a starting alkene.[45] Thus, we can con-
(Scheme 10). Treatment of 67 with PbACTHNUTRGENUN(G OAc)4 in MeOH/ben-
zene at 08C resulted in the cleavage of the F’-ring,[47] and
the resultant aldehyde ester was immediately exposed to
Ph3P=CH2 to give the ester 5 (55%, four steps). The
1H NMR analysis of 5 indicated that no epimerization occur-
red at the C25 position during the four-step transformation
of 6. After removal of the TMS group with TBAF/AcOH,
saponification of the ester provided the hydroxy acid 68
(97%, two steps). Lactonization of 68 under Yamaguchi
conditions[18] gave the seven-membered lactone 4 in 85%
yield. Reduction of 4 with DIBALH and in situ acetyla-
tion[48] gave the a-acetoxy ether 69 as a 10:1 mixture of dia-
5282
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5276 – 5288

(+)-Gambieric Acid A and Its Analogues
FULL PAPER
Scheme 10. Completion of the polycyclic ether skeleton 2. RSM=Recovered starting material.
Figure 2. Rationale for the stereoselective hydroboration of seven-membered endocyclic enol ethers.
stereomers in 85% yield, which was converted to the mixed
TMSOTf and 2,6-di-tert-butylpyridine (DTBP; 94%, d.r.
thioacetal 70 by treatment with TMSSPh in the presence of
10:1, stereochemistry not determined).
Chem. Eur. J. 2013, 19, 5276 – 5288
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5283

H. Fuwa, M. Sasaki et al.
Stereoselective allylation of the C26 position required ex-
tensive model studies. We learned from our previous studies
that the stereoselective allylation of lactone-derived a-ace-
toxy ethers promoted by a Lewis acid and allyltrimethylsi-
lane required harsh conditions and gave the corresponding
allylated products only in moderate yields.[12e,f] Hence, we
strove to develop a new methodology for the stereoselective
allylation, which can tolerate a wide range of functionalities.
Hirama and co-workers have reported the allylation of a
seven-membered cyclic mixed thioacetal by using NIS/
AgOTf as an activator although the scope of the substrate
remained elusive.[49] Accordingly, we investigated the appli-
cation of glycosylation chemistry to cyclic mixed thioacetals
in greater detail. We used the cyclic mixed thioacetal 71[29]
as the DE-ring model compound to investigate suitable re-
action conditions for allylation (Table 2). We first attempted
Figure 3. Stereochemistry of the allylated product 72.
Table 2. Stereoselective allylation of the mixed thioacetal 71.[a]
Entry
Activator
(equiv)
Triflate
(0.1 equiv)
Additive
T
[8C]
Yield
[%]
G
1
2
3
4
5
6
7
NIS (1.3)
NIS (1.3)
NIS (1.3)
NBS (1.3)
NBS (1.3)
NBS (1.3)
NBS (3.0)
none
None
None
À78 to 0
À78 to 0
À78 to 0
À78
0
Figure 4. Rationale for the stereoselective allylation of cyclic mixed thio-
acetals.
TMSOTf
TMSOTf
TMSOTf
TfOH
MeOTf
TMSOTf
19
51
49
61
56
72
4 ꢂ MS
4 ꢂ MS
4 ꢂ MS
4 ꢂ MS
4 ꢂ MS
À40
À40
generated oxocarbenium cation 73, and allyltrimethylsilane
approaches from the a-face to maximize orbital overlapping.
In contrast, the nucleophilic attack of allyltrimethylsilane
from the b-face would cause an unfavorable interaction be-
tween the pseudoaxial lone pair of the ether oxygen and the
HOMO of allyltrimethylsilane. Notably, the stereoselectivity
of the allylation of 71 was opposite to that expected for the
real substrate, i.e., the cyclic mixed thioacetal 70. However,
as previously demonstrated with a model compound,[12e,f] we
considered that in the real substrate, the C25 bulky substitu-
ent must be pseudoaxially oriented, which makes the a-face
sterically crowded and prevents the approach of allyltrime-
thylsilane from the a-face of oxocarbenium cation 74 (Fig-
ure 4B). Consequently, allyltrimethylsilane would be forced
to access from the less-hindered b-face so as to give the ally-
lated product with the correct stereochemistry at C26.
In the event, stereoselective allylation of 70 was per-
formed under optimized conditions (allyltrimethylsilane,
NBS, TMSOTf, 4 ꢂ MS, CH2Cl2, À40 to 08C) to afford the
diene 3 in 74% yield (d.r.>20:1), along with the recovered
starting material (16%) (Scheme 10). RCM of the diene 3
by its exposure to G-II[38] in CH2Cl2 at 408C gave the nona-
cyclic polyether core 2 quantitatively. The relative configura-
tion of the stereogenic centers along the E- and F-rings was
established at this stage by ROE experiments, as shown,
thereby supporting our proposed allylation model, illustrat-
ed in Figure 4B.
À40
[a] All reactions were performed in the presence of excess allyltrimethyl-
silane in CH2Cl2. In all cases, product 72 was isolated as a single stereo-
isomer.
to activate 71 with NIS, but found that the desired reaction
did not occur (Table 2, entry 1). On the other hand, the ad-
dition of TMSOTf was found to be beneficial[50] although 72
was isolated only in 19% yield (Table 2, entry 2). The yield
of 72 was considerably improved to 51% when the reaction
was performed in the presence of 4 ꢂ molecular sieves
(Table 2, entry 3). By using NBS instead of NIS,[51] the reac-
tion proceeded even at À788C to give 72 in 49% yield
(Table 2, entry 4). Changing the activator of NBS resulted in
moderate yields of 72 (Table 2, entries 5 and 6). Finally, we
found that increasing the molar amount of NBS was effec-
tive for achieving an optimal result (72%, Table 2,
entry 7).[52] The acid-sensitive benzylidene acetal was not af-
fected under these conditions.
The configuration of the newly formed C26 stereogenic
center was determined by an NOE experiment, as shown in
Figure 3, and the stereoselectivity of the allylation of 71 was
in accordance with that observed for the O-glycosylation of
septanosides.[53] The stereochemical outcome can be ex-
plained by a stereoelectronic effect (Figure 4A). The pseu-
doaxial lone pair of the ether oxygen would stabilize the
5284
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5276 – 5288

(+)-Gambieric Acid A and Its Analogues
FULL PAPER
Scheme 11. Total synthesis of GAA (1).
[54]
tion of the primary alcohol of 79 with TEMPO/PhIACTHNUTRGNEUNG(OAc)2
Completion of the total synthesis of (+)-GAA (1): Having
completed the synthesis of the polycyclic ether skeleton 2,
we next developed the J-ring side chain (Scheme 11). Re-
moval of the acetonide group of 2 with acidic methanol was
followed by selective acylation of the primary alcohol
(PivCl, pyridine) to deliver a pivaloyl ester (88%, two
steps). Dess–Martin oxidation, Wittig methylenation of the
resultant ketone, and reductive removal of the pivaloyl
group gave the alcohol 75 (86%, three steps). This alcohol
was converted to the methyl ketone 76 by using a three-step
sequence involving oxidation, methylation, and a second ox-
idation (82%, three steps). Our earlier model studies have
demonstrated that methyl ketones such as 76 are challenging
substrates for Julia–Kocienski olefination because of their
propensity for enolization or retro-oxa-Michael reac-
tion.[12g,h] However, as we previously reported, these unde-
sired side reactions could be suppressed by using CeCl3 as
an additive.[12g,h] Thus, the Julia–Kocienski reaction of 76
with the sulfone 77[12g,h] in the presence of CeCl3 (LDA,
THF, À788C) afforded the trisubstituted (E)-olefin 78 in
53% yield along with the corresponding (Z)-isomer in 21%
yield. The stereochemistry of 78 was confirmed by a ROE
experiment, as shown. Removal of all the NAP groups with
DDQ[31] gave the tetraol 79 in 80% yield. Selective oxida-
proceeded smoothly to deliver the corresponding aldehyde,
which was immediately oxidized with NaClO2 and then es-
terified with TMSCHN2 to provide the methyl ester 80
(66%, three steps). Finally, cleavage of the TBS group with
acetic acid and saponification of the derived methyl ester 81
with aqueous LiOH gave GAA (1) in 96% overall yield.
1
The H and 13C NMR, IR, and HRMS data of our synthetic
material were identical with those of the natural product.
Furthermore, the specific rotation value of synthetic (+)-1
([a]2D5 = +22.5 (c=0.40 in MeOH)) was in accordance with
that of the authentic sample ([a]2D0 = +33 (c=0.488 in
MeOH)).[1] Thus, we concluded that our revised complete
stereostructure of GAA, as shown structure 1, was accu-
rate.[10]
Evaluation of the antifungal and antiproliferative activities
of synthetic GAA and its analogues: We investigated the bi-
ological activity of 1 and its synthetic analogues by taking
advantage of our total synthesis. We prepared five synthetic
analogues; that is, the GAA methyl ester 81 (Scheme 11),
(46Z)-GAA (82), J-ring side chain truncated analogue 83,
truncated A/BCD-ring analogue 84, and truncated GHIJ-
ring analogue 85 (Figure 5). The preparation of analogues
82–85 is included in the Supporting Information.
Chem. Eur. J. 2013, 19, 5276 – 5288
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5285

H. Fuwa, M. Sasaki et al.
proximately five-fold) than GAA. Interestingly, the antipro-
liferative activity of the GAA methyl ester 81 was only two-
fold less active than that of the parent GAA. On the other
hand, the J-ring side chain truncated analogue 83 was inac-
tive at a concentration of 30 mm. These results raised the
possibility that the antifungal and antiproliferative activities
of GAA might not be correlated. Truncated analogues 84
and 85 were completely inactive at 30 mm.
Conclusion
In this study, we described the first total synthesis and com-
plete stereostructure of (+)-GAA (1) in detail. The synthe-
sis of the A/BCD-ring fragment 8 featured Suzuki–Miyaura
coupling of the B-ring vinyl iodide 18 and the alkylborate
33, stereoselective bromoetherification to forge the A-ring
tetrahydrofuran, and application of our Suzuki–Miyaura
coupling/RCM sequence for the convergent assembly of the
BCD-ring domain. The synthetically challenging central
domain of 1 was constructed in a convergent fashion by
using the “unnatural” seven-membered F’-ring. Thus, the A/
BCD- and F’GHIJ-ring fragments (8 and 9, respectively)
were assembled through a Suzuki–Miyaura coupling. Al-
though the hydroboration of the coupling product, the endo-
cyclic enol ether 7, provided a mixture of diastereomeric al-
cohols, the C25 stereogenic center was successfully formed
by the subsequent oxidation and base-promoted equilibra-
tion under thermodynamic control to give the ketone 6.
After oxidative cleavage of the F’-ring, stereoselective ally-
lation of the cyclic mixed thioacetal 70, based on the glyco-
sylation chemistry, was utilized for the elaboration of the E-
ring. Subsequently, the F-ring was formed through RCM.
Stereoselective introduction of the J-ring side chain was ach-
ieved by using our modified Julia–Kocienski olefination to
complete the total synthesis of 1. Our total synthesis pro-
ceeded in 84 steps (longest linear sequence) from commer-
cially available l-gulonolactone in 0.012% overall yield. The
spectroscopic properties and specific rotation value of 1
matched those of the natural product, which resulted in an
unambiguous determination of the absolute configuration of
1. Thus, our total synthesis of 1 successfully confirmed that
our previously revised structure was actually correct and
demonstrated the vital role of organic synthesis in the struc-
ture elucidation of complex natural products.[56] From a bio-
synthetic perspective, the stereostructure of gambieric acids
B–D should also be revised similarly. Moreover, we evaluat-
ed the antifungal and antiproliferative activities of 1 and
several synthetic analogues and investigated the preliminary
structure–activity relationships of 1.
Figure 5. Synthetic analogues of GAA (1).
Table 3. Antifungal and antiproliferative activities of synthetic GAA (1)
and its analogues.[a]
Compound
Antifungal activity
Antiproliferative activity
[b]
[ngdiskÀ1
]
IC50 [mM][c]
Sporulation
inhibition
Growth
inhibition
GAA (1)
10
1000
10
500
>10000
500
10000
>10000
>10000
5.1
11
1.0
>30
>30
>30
81
82
83
84
85
100
>10000
>10000
[a] For details of assay procedures, see the Supporting Information.
[b] Against A. niger by disk diffusion. [c] Against P388 cells.
We first evaluated the antifungal activity of synthetic
GAA (1) and its analogues against A. niger by disk diffusion
assay, and the results are summarized in Table 3. GAA (1)
inhibited the sporulation of A. niger at 10 ng diskÀ1, and the
total growth was inhibited at a higher dose (500 ng diskÀ1).
These results were in accordance with those reported for the
natural product.[5] The (46Z)-GAA (82) was found to be
equipotent to GAA, showing that the geometry of the tri-
substituted olefin within the J-ring side chain is not relevant
to the antifungal activity. On the other hand, the activities
of the J-ring side chain truncated analogue 83 and GAA
methyl ester 81 were 10- to 100-fold lower than that of
GAA, indicating the importance of the C1 carboxy group
and the length of the J-ring side chain. Truncated analogues
84 and 85 were completely inactive at a dose of 10 mg diskÀ1.
In addition, we evaluated the antiproliferative activity of
GAA (1) and the synthetic analogues in cultured P388
mouse lymphoma cells by using the WST-8 assay.[55] GAA
inhibited the growth of P388 cells with an IC50 value of
5.1 mm. The (46Z)-GAA (82) was slightly more potent (ap-
Acknowledgements
We thank Professor Michio Murata (Osaka University) for providing us
with copies of 1H and 13C NMR spectra of the authentic sample of
(+)-gambieric acid A and Professor Hiroshi Nagai (Tokyo University of
5286
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5276 – 5288

(+)-Gambieric Acid A and Its Analogues
FULL PAPER
Marine Science and Technology) for his useful suggestions. This work
was financially supported in part by Grants-in-Aid (Nos. 21241050,
23681045, 23102016, and 24102517) from MEXT, Japan, and a grant from
ERATO Murata Lipid Active Structure Project from JST. The Astellas
Foundation for Research on Metabolic Disorders and the Banyu Life Sci-
ence Foundation International are acknowledged for financial support.
K.I. is a recipient of the Research Fellowship for Young Scientists from
JSPS.
Chem. Int. Ed. 2005, 44, 4442–4489; c) A. Deiters, S. F. Martin,
Chem. Rev. 2004, 104, 2199–2238.
[17] a) M. Sasaki, T. Noguchi, K. Tachibana, Tetrahedron Lett. 1999, 40,
1337–1340; b) M. Sasaki, T. Noguchi, K. Tachibana, J. Org. Chem.
2002, 67, 3301–3310.
[18] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull.
Chem. Soc. Jpn. 1979, 52, 1989–1993.
[19] For a review on “B-alkyl” Suzuki–Miyaura coupling, see: S. R.
Chemler, D. Trauner, S. J. Danishefsky, Angew. Chem. 2001, 113,
4676–4701; Angew. Chem. Int. Ed. 2001, 40, 4544–4568.
[20] a) M. Sasaki, H. Fuwa, M. Inoue, K. Tachibana, Tetrahedron Lett.
1998, 39, 9027–9030; b) M. Sasaki, H. Fuwa, M. Ishikawa, K. Tachi-
bana, Org. Lett. 1999, 1, 1075–1077; c) M. Sasaki, M. Ishikawa, H.
Fuwa, K. Tachibana, Tetrahedron 2002, 58, 1889–1911.
[21] For the total synthesis of marine polycyclic ether natural products
based on a Suzuki–Miyaura coupling, see: a) H. Fuwa, N. Kainuma,
K. Tachibana, M. Sasaki, J. Am. Chem. Soc. 2002, 124, 14983–
14992; b) C. Tsukano, M. Sasaki, J. Am. Chem. Soc. 2003, 125,
14294–14295; c) C. Tsukano, M. Ebine, M. Sasaki, J. Am. Chem.
Soc. 2005, 127, 4326–4335; d) H. Fuwa, M. Ebine, M. Sasaki, J. Am.
Chem. Soc. 2006, 128, 9648–9650; e) H. Fuwa, M. Ebine, A. J. Bour-
delais, D. G. Baden, M. Sasaki, J. Am. Chem. Soc. 2006, 128, 16989–
16999; f) M. Ebine, H. Fuwa, M. Sasaki, Org. Lett. 2008, 10, 2275–
2278; g) M. Ebine, H. Fuwa, M. Sasaki, Chem. Eur. J. 2011, 17,
13754–13761.
[22] For reviews on the total synthesis of marine polycyclic ether natural
products based on Suzuki–Miyaura coupling, see: a) M. Sasaki, Bull.
Chem. Soc. Jpn. 2007, 80, 856–871; b) M. Sasaki, H. Fuwa, Nat.
Prod. Rep. 2008, 25, 401–426; c) H. Fuwa, Bull. Chem. Soc. Jpn.
2010, 83, 1401–1420.
[23] Compound 19 was available in three steps from l-gulonolactone.
See: J. S. Clark, J. G. Kettle, Tetrahedron Lett. 1997, 38, 127–130.
[24] M. Ebine, Y. Suga, H. Fuwa, M. Sasaki, Org. Biomol. Chem. 2010,
8, 39–42.
[1] a) H. Nagai, K. Torigoe, M. Satake, M. Murata, T. Yasumoto, H.
Hirota, J. Am. Chem. Soc. 1992, 114, 1102–1103; b) H. Nagai, M.
Murata, K. Torigoe, M. Satake, T. Yasumoto, J. Org. Chem. 1992,
57, 5448–5453.
[2] a) M. Murata, A. M. Legrand, Y. Ishibashi, T. Yasumoto, J. Am.
Chem. Soc. 1989, 111, 8929–8931; b) M. Murata, A. M. Legrand, Y.
Ishibashi, M. Fukui, T. Yasumoto, J. Am. Chem. Soc. 1990, 112,
4380–4386.
[3] a) M. Murata, T. Iwashita, A. Yokoyama, M, Sasaki, T. Yasumoto,
J. Am. Chem. Soc. 1992, 114, 6594–6596; b) M. Murata, H. Naoki,
T. Iwashita, S. Matsunaga, M. Sasaki, A. Yokoyama, T. Yasumoto, J.
Am. Chem. Soc. 1993, 115, 2060–2062; c) M. Murata, S. Matsunaga,
M. Satake, T. Yasumoto, J. Am. Chem. Soc. 1994, 116, 7098–7107;
d) M. Satake, S. Ishida, T. Yasumoto, M. Murata, H. Utsumi, T. Hi-
nomoto, J. Am. Chem. Soc. 1995, 117, 7019–7020.
[4] a) M. Satake, M. Murata, T. Yasumoto, J. Am. Chem. Soc. 1993, 115,
361–362; b) A. Morohashi, M. Satake, T. Yasumoto, Tetrahedron
Lett. 1999, 40, 97–100.
[5] H. Nagai, Y. Mikami, K. Yazawa, T. Gonoi, T. Yasumoto, J. Anti-
biot. 1993, 46, 520–522.
[6] M. Inoue, M. Hirama, M. Satake, K. Sugiyama, T. Yasumoto, Toxi-
con 2003, 41, 469–474.
[7] B. Sakamoto, H. Nagai, Y. Hokama, Phycologia 1996, 35, 350–353.
[8] A. Morohashi, M. Satake, H. Nagai, Y. Oshima, T. Yasumoto, Tetra-
hedron 2000, 56, 8995–9001.
[9] I. Ohtani, T. Kusumi, Y. Kashman, H. Kakisawa, J. Am. Chem. Soc.
1991, 113, 4092–4096.
[25] a) K. C. Nicolaou, G.-Q. Shi, J. L. Gunzner, P. Gꢃrtner, Z. Yang, J.
Am. Chem. Soc. 1997, 119, 5467–5468; b) K. C. Nicolaou, G.-Q. Shi,
K. Namoto, F. Bernal, Chem. Commun. 1998, 1757–1758.
[26] M. Sasaki, S. Honda, T. Noguchi, H. Takakura, K. Tachibana, Synlett
2000, 838–840.
[27] a) S. Takano, K. Samizu, T. Sugihara, K. Ogasawara, J. Chem. Soc.
Chem. Commun. 1989, 1344–1345; b) J. S. Yadav, P. K. Deshpande,
G. V. M. Sharma, Tetrahedron 1990, 46, 7033–7046.
[10] a) H. Fuwa, T. Goto, M. Sasaki, Org. Lett. 2008, 10, 2211–2214;
b) H. Fuwa, K. Ishigai, T. Goto, A. Suzuki, M. Sasaki, J. Org. Chem.
2009, 74, 4024–4040.
[11] a) I. Kadota, N. Oguro, Y. Yamamoto, Tetrahedron Lett. 2001, 42,
3645–3647; b) I. Kadota, H. Takamura, Y. Yamamoto, Tetrahedron
Lett. 2001, 42, 3649–3651; c) J. S. Clark, T. C. Fessard, C. Wilson,
Org. Lett. 2004, 6, 1773–1776; d) J. S. Clark, M. C. Kimber, J. Rob-
ertson, C. S. P. McErlean, C. Wilson, Angew. Chem. 2005, 117, 6313–
6318; Angew. Chem. Int. Ed. 2005, 44, 6157–6162; e) S. W. Roberts,
J. D. Rainier, Org. Lett. 2007, 9, 2227–2230; f) T. Saito, T. Nakata,
Org. Lett. 2009, 11, 113–116.
[12] a) K. Sato, M. Sasaki, Org. Lett. 2005, 7, 2441–2444; b) K. Sato, M.
Sasaki, Angew. Chem. 2007, 119, 2570–2574; Angew. Chem. Int. Ed.
2007, 46, 2518–2522; c) K. Sato, M. Sasaki, Tetrahedron 2007, 63,
5977–6003; d) H. Fuwa, A. Suzuki, K. Sato, M. Sasaki, Heterocycles
2007, 72, 139–144; e) H. Fuwa, S. Noji, M. Sasaki, Chem. Lett. 2009,
38, 866–867; f) H. Fuwa, S. Noji, M. Sasaki, J. Org. Chem. 2010, 75,
5072–5082; g) K. Tsubone, K. Hashizume, H. Fuwa, M. Sasaki, Tet-
rahedron Lett. 2011, 52, 548–551; h) K. Tsubone, K. Hashizume, H.
Fuwa, M. Sasaki, Tetrahedron 2011, 67, 6600–6615.
[28] A. G. Myers, B. Zheng, M. Movassaghi, J. Org. Chem. 1997, 62,
7507–7507.
[29] See the Supporting Information for details.
[30] a) J. A. Marshall, B. A. Johns, J. Org. Chem. 1998, 63, 7885–7892;
b) G. Seidel, A. Fꢄrstner, Chem. Commun. 2012, 48, 2055–2070.
[31] a) M. J. Gaunt, J. Yu, J. B. Spencer, J. Org. Chem. 1998, 63, 4172–
4173; b) M. Inoue, H. Uehara, M. Maruyama, M. Hirama, Org. Lett.
2002, 4, 4551–4554.
[32] H.-J. Liu, J. Yip, K.-S. Shia, Tetrahedron Lett. 1997, 38, 2253–2256.
[33] T. M. Hansen, G. J. Florence, P. Lugo-Mas, J. Chen, J. N. Abrams,
C. J. Forsyth, Tetrahedron Lett. 2003, 44, 57–59.
[34] K. Fujiwara, M. Tsunashima, D. Awakura, A. Murai, Tetrahedron
Lett. 1995, 36, 8263–8266.
[35] H. Fuwa, M. Sasaki, Org. Lett. 2008, 10, 2549–2552.
[36] For a review on the Suzuki–Miyaura coupling of acetate- and acet-
amide-derived enol phosphates, see: H. Fuwa, Synlett 2011, 6–29.
[37] G. Bhaskar, A. R. Kumar, E. Ramu, B. V. Rao, Synthesis 2008,
3061–3064.
[38] M. Scholl, S. Ding, C. W. Lee, R. H. Grubbs, Org. Lett. 1999, 1, 953–
956.
[39] a) W. P. Griffith, S. V. Ley, G. P. Whitcombe, A. D. White, J. Chem.
Soc. Chem. Commun. 1987, 1625–1627; b) S. V. Ley, J. Norman,
W. P. Griffith, S. P. Marsden, Synthesis 1994, 639–666.
[13] A part of this work has been communicated: H. Fuwa, K. Ishigai, K.
Hashizume, M. Sasaki, J. Am. Chem. Soc. 2012, 134, 11984–11987.
[14] For representative reviews on Suzuki–Miyaura coupling, see: a) N.
Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457–2483; b) A. Suzuki,
Angew. Chem. 2011, 123, 6854–6869; Angew. Chem. Int. Ed. 2011,
50, 6722–6737.
[15] P. R. Blakemore, W. J. Cole, P. J. Kocienski, A. Morley, Synlett 1998,
26–28.
[40] a) H. Fuwa, M. Sasaki, K. Tachibana, Tetrahedron 2001, 57, 3019–
3033; b) K. C. Nicolaou, C. V. C. Prasad, C.-K. Hwang, M. E.
Duggan, C. A. Veale, J. Am. Chem. Soc. 1989, 111, 5321–5330.
[16] For selected reviews on olefin metathesis, see: a) A. H. Hoveyda,
A. R. Zhugralin, Nature 2007, 450, 243–251; b) K. C. Nicolaou, P. G.
Bulger, D. Sarlah, Angew. Chem. 2005, 117, 4516–4563; Angew.
Chem. Eur. J. 2013, 19, 5276 – 5288
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5287

H. Fuwa, M. Sasaki et al.
[41] K. Toshima, T. Jyojima, H. Yamaguchi, Y. Noguchi, T. Yoshida, H.
Murase, M. Nakata, S. Matsumura, J. Org. Chem. 1997, 62, 3271–
3284.
[42] For brevity, we performed ab initio calculations on previously syn-
thesized model compound I[12e,f] in place of 7. We considered that I
is a suitable model of 7 because
[50] a) G. H. Veeneman, J. H. van Boom, Tetrahedron Lett. 1990, 31,
1331–1334; b) G. H. Veeneman, S. H. van Leeuwen, H. Zuurmond,
J. H. van Boom, J. Carbohydr. Chem. 1990, 9, 783–796; c) H. Zuur-
mond, S. C. van der Laan, G. A. van der Marel, J. H. van Boom, Car-
bohydr. Res. 1991, 215, C1–C3; d) P. Smid, G. A. de Ruitar, G. A.
van der Marel, F. M. Rombouts, J. H. van Boom, J. Carbohydr.
Chem. 1991, 10, 833–849.
the diastereoselectivity of the hy-
[51] M. Sasaki, K. Tachibana, H. Nakanishi, Tetrahedron Lett. 1991, 32,
6873–6876.
droboration of I with BH3·SMe2 in
THF at room temperature was
[52] We have also examined the NIS/AgOTf conditions reported by
Hirama and co-workers (ref. [49]) and found that the reaction tem-
perature must be raised from À408C to room temperature for com-
plete consumption of the starting material. The product yield was
satisfactory (69%) and comparable to our best result (72%). How-
ever, our optimized NBS/TMSOTf conditions were considered to be
more suitable for the real system because of the higher reactivity of
70 under these conditions.
same as that observed for 7 (d.r.
2:1 favoring the desired diastereo-
mer).[12e,f] For NMR-based confor-
mational analyses on I, see the
Supporting Information.
[43] a) U. Majumder, J. M. Cox, H. W. B. Johnson, J. D. Rainier, Chem.
Eur. J. 2006, 12, 1736–1746; b) A. M. Orendt, S. W. Roverts, J. D.
Rainier, J. Org. Chem. 2006, 71, 5565–5573.
[53] S. Castro, W. S. Fyvie, S. A. Hatcher, M. W. Peczuh, Org. Lett. 2005,
7, 4709–4712.
[44] a) D. J. Pasto, F. M. Klein, J. Org. Chem. 1968, 33, 1468–1476;
b) D. J. Pasto, B. Lepeska, T.-C. Cheng, J. Am. Chem. Soc. 1972, 94,
6083–6090; c) K. N. Houk, N. G. Rondan, Y.-D. Wu, J. T. Metz,
M. N. Paddon-Row, Tetrahedron 1984, 40, 2257–2274.
[45] G. S. Hammond, J. Am. Chem. Soc. 1955, 77, 334–338.
[46] a) D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155–4156;
b) D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277–7287.
[47] a) C. Tsukano, M. Sasaki, Tetrahedron Lett. 2005, 46, 4617–4619;
b) K. Fujiwara, A. Goto, D. Sato, Y. Ohtaniuchi, H. Tanaka, A.
Murai, H. Kawai, T. Suzuki, Tetrahedron Lett. 2004, 45, 7011–7014.
[48] V. H. Dahanukar, S. D. Rychnovsky, J. Org. Chem. 1996, 61, 8317–
8320.
[54] a) A. De Mico, R. Margarita, L. Parlanti, A. Vescovi, G. Piancatelli,
J. Org. Chem. 1997, 62, 6974–6977; b) M. F. Semmelhack, C. S.
Chou, D. A. Cortes, J. Am. Chem. Soc. 1983, 105, 4492–4494.
[55] H. Tominaga, M. Ishiyama, F. Ohseto, K. Sasamoto, T. Hamamoto,
K. Suzuki, M. Watanabe, Anal. Commun. 1999, 36, 47–50.
[56] a) K. C. Nicolaou, S. A. Snyder, Angew. Chem. 2005, 117, 1036–
1069; Angew. Chem. Int. Ed. 2005, 44, 1012–1044; b) M. E. Maier,
Nat. Prod. Rep. 2009, 26, 1105–1124; c) Y. Usami, Mar. Drugs 2009,
7, 314–330; d) T. L. Suyama, W. H. Gerwick, K. L. McPhail, Bioorg.
Med. Chem. 2011, 19, 6675–6701.
Received: December 4, 2012
[49] S. Kobayashi, M. Hori, M. Hirama, Carbohydr. Res. 2008, 343, 443–
452.
Revised: February 9, 2013
Published online: March 28, 2013
5288
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5276 – 5288