Synthesis of aspergillide A via proline-catalyzed trans-to-cis isomerization of a substituted tetrahydropyran

Article abstract of DOI:10.1016/j.tet.2011.02.061

The transformation of a common synthetic intermediate of aspergillides B and C into aspergillide A, a cytotoxin produced by a marine-derived fungus, has been accomplished in an eleven-step sequence involving an efficient proline-catalyzed isomerization of

Full text of DOI:10.1016/j.tet.2011.02.061

Tetrahedron 67 (2011) 2882e2888
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis of aspergillide A via proline-catalyzed trans-to-cis isomerization of
a substituted tetrahydropyran
,
Tomohiro Nagasawa a, Tomoo Nukada b, Shigefumi Kuwahara a
*
a Laboratory of Applied Bioorganic Chemistry, Graduate School of Agricultural Science, Tohoku University, Tsutsumidori-Amamiyamachi, Aoba-ku, Sendai 981-8555, Japan
b Department of Fermentation Science, Faculty of Applied Bio-Science, Tokyo University of Agriculture, 1-1-1 Sakuragaoka, Setagaya-ku, Tokyo 156-8502, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The transformation of a common synthetic intermediate of aspergillides B and C into aspergillide A,
a cytotoxin produced by a marine-derived fungus, has been accomplished in an eleven-step sequence
involving an efficient proline-catalyzed isomerization of a 2,6-trans-substituted tetrahydropyran-2-ac-
etaldehyde intermediate into the corresponding cis isomer and the Yamaguchi macrolactonization as the
key steps.
Received 14 February 2011
Received in revised form 21 February 2011
Accepted 21 February 2011
Available online 26 February 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Aspergillide
Cytotoxic
Total synthesis
Isomerization
Macrolide
1. Introduction
O
O
H
O
O
H
O
O
H
Aspergillides AeC are fourteen-membered macrolides isolated
by Kusumi and co-workers from a bromine-modified 1/2PD culture
medium of the marine-derived fungus Aspergillus ostianus strain
01F313 and shown to exhibit significant cytotoxicity against mouse
lymphocytic leukemia cells (L1210).1 The proposed structure of
aspergillide C (3) (Fig. 1), including its absolute configuration, was
soon confirmed by our total synthesis.2 The structures of aspergil-
lides A and B initially assigned by the Kusumi group were, however,
found to need stereochemical revision through synthetic studies by
Hande and Uenishi as well as X-ray crystallographic analyses by Ooi
and co-workers, which concluded the genuine structures of asper-
gillides A and B to be 1 and 2, respectively.3,4 Their unique fourteen-
membered macrolide structures embedded with a tetrahydro- or
dihydropyran ring,5,6 coupled with their interesting biological ac-
tivity, prompted syntheticefforts toward 1e3 byquitea few research
groups including us, which have so far culminated in five total/for-
mal syntheses for 1,7 seven for 2,3,7c,eeg,8 and three for 3.2,9
O
O
O
H
H
H
3
7
HO
HO
HO
1
2
3
aspergillide A
aspergillide B
aspergillide C
Fig. 1. Structures of aspergillides A, B, and C.
iodolactonization of the major 3,7-trans isomer 5 for the stereo-
selective installation of the 4- -hydroxy group, and the Yamaguchi
a
macrolactonization to construct the macrocyclic structures.2,8c In
light of the previous syntheses shown in Scheme 1, we envisaged
three possibilities of obtaining aspergillide A (1), the C3 epimer of
aspergillide B with 3,7-cis stereochemistry, from a synthetic in-
termediate of 2 and 3: (1) epimerization of the stereochemistry of
macrocyclic intermediate 8 at the C3 position via a retro-oxy-Mi-
chael/oxy-Michael process (Route 1); (2) exploitation of the already
existing 3,7-cis stereochemistry of 6 obtained as the minor isomer
in the above-mentioned C-alkylation (Route 2); and (3) stereo-
chemical inversion of the C3 position of the major isomer 5 via
a retro-oxy-Michael/oxy-Michael sequence (Route 3). Our suc-
cessful synthesis of aspergillide A (1) according to Route 3 was
previously communicated together with some attempts in line with
The latter half stage of our reported syntheses of aspergillides B
(2) and C (3), both of which have a 3,7-trans stereochemical re-
lationship, featured the Lewis-acid promoted Ferrier-type C-alkyl-
ation of cyclic acetal 4 to form a mixture 5 and 6 (Scheme 1),
* Corresponding author. Tel./fax: þ81 22 717 8783; e-mail address: skuwahar@
biochem.tohoku.ac.jp (S. Kuwahara).
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.02.061

T. Nagasawa et al. / Tetrahedron 67 (2011) 2882e2888
2883
starting material 8 even at elevated temperatures. The use of KOt-
Bu as a stronger base also brought no fruitful outcome, giving only
the starting material at room temperature or a complex mixture at
45 ꢀC.10
2.2. Exploitation of the 3,7-cis stereochemistry of minor
isomer 6 (Route 2)
2.2.1. Macrolactonization of a 3,4-cis/3,7-cis seco acid. Faced with
the difficulty to epimerize the 3,7-trans macrolactone 8 into the
corresponding cis isomer 10, we next attempted to take advantage
of the 3,7-cis stereochemistry existing in the minor C-alkylation
product 6, the same stereochemistry as that present in aspergillide
A (1) (see Scheme 1). In line with our previous procedures
employed for the syntheses of aspergillides A and B,2,8c the olefinic
ester 6 was hydrolyzed with an aq NaOH solution, and the resulting
carboxylate intermediate was directly treated with KI3/NaHCO3 to
give iodolactone 11 (Scheme 3). Reductive removal of the iodine
atom from 11 proceeded smoothly, affording 12 in 97% overall yield
from 6. Hydrolysis of the lactone moiety of 12 with LiOH in aq THF
gave a mixture containing a hydroxy carboxylate intermediate. The
mixture was concentrated to dryness, dissolved in DMF, and then
treated with an excess amount of TBSOTf in the presence of imid-
azole and DMAP to give a bis-TBS-protected intermediate, the TBS
A
Scheme 1. The latter half stage of our reported synthesis of aspergillides B (2) and C
(3), and three potential routes from their synthetic intermediates to aspergillide A (1).
Route 1.7a In this article, we describe a full account of our synthetic
efforts toward 1 including those along Route 2.
2. Results and discussion
2.1. Epimerization of macrocyclic intermediate 8 (Route 1)
Since aspergillides A (1) and B (2) are epimeric to each other at
the alkoxy-bearing C3 position
b to the lactone carbonyl, it would
be most convenient if the direct epimerization of 2 to 1 via a retro-
oxy-Michael/oxy-Michael process could be effected. This possibility
was previously examined by Ooi et al.4 Their treatment of 2 with
either SiO2 or NaOMe, however, afforded no desired product 1,
giving only a methanolysis product in the latter case (Scheme 2).
These results as well as our concern about the possible formation of
a g-lactone through the intramolecular attack of the C4 hydroxy
group to the lactone carbonyl during basic or acidic treatment of 2
made us choose the TBS-protected form of 2 (compound 8) as
a suitable epimerization substrate. Thus, the 3,7-trans isomer 8 was
first exposed to DBU in toluene, hoping to obtain the corresponding
3,7-cis isomer 10 via alkoxy unsaturated ester 9. This treatment
was, however, unsuccessful, resulting only in the recovery of the
Scheme 2. Attempted epimerization of 3,7-trans isomer 8 to 3,7-cis isomer 10.
Scheme 3. Attempt for the conversion of 6 into aspergillide A (1).

2884
T. Nagasawa et al. / Tetrahedron 67 (2011) 2882e2888
ester group of which was then selectively hydrolyzed by directly
adding water to the reaction mixture to afford hydroxy-protected
carboxylic acid 13 in 91% yield for the two steps. Oxidative depro-
tection of the PMB group of 13 with DDQ furnished seco acid 14 in
82% yield, which set the stage for the pivotal macrolactonization
step.
The macrolactonization of 14 was attempted by using Yama-
guchi’s11 and Shiina’s12 protocols. Unfortunately, however, sub-
jection of 14 to the Yamaguchi lactonization conditions
(Cl3C6H2COCl, Et3N, DMAP, toluene, 110 ꢀC, 13 h) gave a complex
mixture. Despite our careful chromatographic separation, no de-
sired product 15 was detected from the mixture, but instead di-
meric macrodiolide 16 and 3-epi-15 were obtained in yields of 19%
and 34%, respectively. The structure of 3-epi-15 was confirmed by
its deprotection into alcohol 17 whose spectral (1H and 13C NMR)
and TLC data were identical with those of a sample derived from
aspergillide B (2) in two steps via ketone 18.13 The formation of
3-epi-15 in 34% yield came as a surprise to us. From the facts that no
desired product 15 was isolated and that the macrolactonization of
analogous 3,7-trans-substituted seco acids proceeded smoothly to
give precursors of aspergillides B and C in high yields,2,3,7e9 we are
speculating that 3-epi-15 would have been produced via cis-to-
trans epimerization at the stage of an activated ester derivative of
14 prior to macrolactonization.7e,14 Exposure of 14 to the Shiina
macrolactonization conditions [MNBA (2-methyl-6-nitrobenzoic
anhydride), DMAP, CH2Cl2], on the other hand, gave only the
macrodiolide 16 in 20% yield at room temperature (20 h) and in 16%
yield at reflux (18 h) along with some unidentified products.
2.2.2. Macrolactonization of
a
3,4-trans/3,7-cis seco acid.-
Considering the possibility that the presence of the TBSO group at
the C4 position of 14 oriented cis to the C3 substituent might have
encumbered its macrolactonization into 15 (see Scheme 3), we next
planned to macrolactonize a seco acid possessing a 3,4-trans/3,7-cis
stereochemical relationship, exactly the same stereochemistry as
that of natural aspergillide A (1). The preparation of the desired
seco acid (compound 23 in Scheme 4) began with reductive
opening of the lactone ring of 12 to give a diol intermediate. Se-
lective protection of its primary hydroxy group with TBSOTf in
CH2Cl2 in the presence of Et3N and DMAP afforded 19 in 79% yield
from 12. Oxidation of the alcohol 19 with DesseMartin’s period-
inane (DMP) proceeded smoothly to provide ketone 20 (97%
yield),15 which was then subjected to stereoselective reduction
with NaBH4 in MeOH to give an alcohol. The product was then
Scheme 4. Preparation of 3,4-trans/3,7-cis seco acid 23 and its macrolactonization.
C-alkylation product 5 with 3,7-trans stereochemistry as the
starting material.
2.3. Preparation of a 3,4-trans/3,7-cis seco acid from 3,7-trans
intermediate 5 and its transformation to aspergillide A (1)
(Route 3)
protected as its MOM ether 21 bearing an
a-oriented oxygen
functionality at the C4 position with a 3,4-trans relationship (72%
from 20). Unmasking of the TBS-protected alcohol and subsequent
DMP oxidation gave 3,7-cis aldehyde 22 (97% for the two steps),
which was then converted into the seco acid 23 by the Pinnick
oxidation followed by PMB-deprotection with DDQ (69% for the
two steps). For the macrolactonization of 23 we attempted three
methodologies. The Shiina protocol [MNBA, DMAP, CH2Cl2
(0.7 mM), rt, 22 h] only gave the corresponding dimeric macro-
diolide 25 in low yields and Gerlach’s modification of the Cor-
eyeNicolaou method [PySSPy, Ph3P, AgClO4, toluene (0.7 mM),
120 ꢀC, 23 h]16 resulted in the formation of a complex mixture. To
our delight, however, we could finally obtain the target macrocycle
24, albeit in low yield (14%, not optimized), by use of the Yamaguchi
macrolactonization conditions [Cl3C6H2COCl, Et3N, DMAP, toluene
(0.9 mM), 110 ꢀC, 8 h].17 Although the chemical yield of 24 was not
satisfactory and the formation of the undesired dimeric product 25
was unavoidable, it was suggested that a 3,4-trans/3,7-cis seco acid
like 23 would likely macrocyclize. At this point, we decided to re-
view the synthetic route to the seco acid 23, which commenced
with the minor C-alkylation product 6 (see Scheme 1), seeking for
a more efficient approach to 23 (or its equivalent) using the major
For the preparation of an appropriate macrocyclization pre-
cursor with a 3,4-trans/3,7-cis stereochemical relationship from the
major isomer 5 in the Ferrier-type C-alkylation of 4 (see Scheme 1),
the unsaturated ester 5 bearing a 3,7-trans stereochemical re-
lationship was first exposed to a one-pot hydrolysis/iodolactoni-
zation sequence, the product of which was then reduced with
Bu3SnH in toluene in the presence of Et3B and O2 to give lactone 26
in 90% yield for the two steps (Scheme 5).8c LiAlH4 reduction of 26
and subsequent treatment of the resulting diol afforded a bis-TBS-
protected intermediate. Selective removal of the TBS group on the
primary hydroxy function with camphorsulfonic acid in MeOH/
CH2Cl2 gave alcohol 27 in 86% yield from 26. Exposure of 27 to
DMP/NaHCO3 in CH2Cl2 cleanly afforded aldehyde 28, the substrate
we chose for the epimerization at its C3 position. Gratifyingly, on
treatment with a catalytic amount of
D-proline in MeOH, the 3a-
substituted aldehyde 28 underwent smooth equilibration to give
a 95:5 epimeric mixture of 29 and 28, favoring the desired 3b-
epimer 29 with 3,7-cis stereochemistry.18,19 After chromatographic
isolation in 81% yield, the aldehyde 29 was subjected to oxidative

T. Nagasawa et al. / Tetrahedron 67 (2011) 2882e2888
2885
Table 1
Macrolactonization of 31
OH
O
O
H
macrolactonization
HO2C
H
O
O
H
H
TBSO
TBSO
31
32
Entry
Reagents and conditions
Isolated yield (%)
32
33a
3-epi-32a
1
2
3
4
5
6
BrC5H4NEt$BF4, Et3N
MeCN, 90 ꢀC, 6 h
complex mixture
complex mixture
complex mixture
PySSPy, Ph3P, AgBF4
Toluene, 110 ꢀC, 54 h
MNBA, DMAP
CH2Cl2/THF, 50 ꢀC, 17 h
MNBA, DMAP
d
25
30
22
14
19
d
CH2Cl2, rt, 28 h
Cl3C6H2COCl, Et3N, DMAP
Toluene, 110 ꢀC, 5 h
Cl3C6H2COCl, Et3N, DMAP
Toluene, 80 ꢀC, 8 h
trace
trace
a
See Fig. 2 for the structure.
Fig. 2. Undesired products formed in the macrolactonization of 31.
Scheme 5. Preparsation of 3,4-trans/3,7-cis seco acid 31 via proline-catalyzed epi-
merization and its transformation to aspergillide A (1).
additional attempts to further improve the chemical yield, mainly
by varying the reaction temperature and concentration, were un-
successful. The modest yields of this macrolactonization could be
rationalized by considering that the seco acid 31, which would
surely adopt a stable chair conformation with its C3-, C4-, and C7-
substituents equatorially oriented, would need to change its
conformation into an energetically unfavorable one with all the
substituents axially arranged to macrocyclize into 32,21 as dis-
cussed by Fuwa and co-workers in their macrolactonization of
23.7e,17 Finally, deprotection of the TBS group furnished in 76% yield
aspergillide A (1) as a white crystalline solid (mp 64.5e65.5 ꢀC), the
1H and 13C NMR spectra as well as the specific rotation of which
showed good agreement with those of natural aspergillide A.
deprotection of the PMB group to provide hydroxy aldehyde 30
(80% yield), which was then oxidized by the Pinnick method to give
macrocyclization precursor 31 (97% yield).
Having secured the seco acid 31 possessing the same 3,4-trans/
3,7-cis stereochemistry as 23, we proceeded to its macro-
lactonization into 32. As shown in Table 1, besides the three
methods employed in this study so far (the Yamaguchi, Shiina, and
Gerlach protocols), we also examined macrolactonization condi-
tions using the Mukaiyama reagent, 2-bromo-1-ethylpyridinium
tetrafluoroborate.20 Exposure of 31 to the Mukaiyama, Gerlach, and
Shiina conditions (at 50 ꢀC) resulted in the formation of complex
mixtures (entries 1e3), while the implementation of the Shiina
macrolactonization at room temperature afforded dimeric macro-
diolide 33 in 22% yield along with minor products that were not
characterized (entry 4). Fortunately, however, treatment of the seco
acid 31 under Yamaguchi’s conditions at 110 ꢀC gave rise to the
desired lactone 32 in 25% yield together with dimeric byproduct 33
(14%) and a trace amount of 3-epi-32 (entry 5). The structure of 3-
epi-32 was confirmed by comparing its 1H and 13C NMR spectra
with those of an authentic sample (compound 8 in Scheme 1)
previously prepared in our total synthesis of aspergillide B (2).8c
Lowering the reaction temperature to 80 ꢀC brought some desir-
able effect, giving 32 in an improved yield of 30% (entry 6). Some
3. Conclusion
The conversion of our synthetic intermediate 26 of aspergillide
B (2) into aspergillide A (1) was accomplished in nine steps by using
the efficient proline-mediated epimerization of the 3,7-trans-
substituted intermediate 28 into the corresponding cis isomer 29
and the Yamaguchi macrolactonization as the key steps. This syn-
thesis constitutes the first synthesis of aspergillide A (1) and rep-
resents the successful completion of our efforts toward the total
synthesis of all the three aspergillides (A, B, and C) from the com-
mon intermediate 5.

2886
T. Nagasawa et al. / Tetrahedron 67 (2011) 2882e2888
4. Experimental
as a pale yellow oil. [
a
]
þ4.53 (c 1.18, CHCl3); IR: nmax 2720 (w),
D17
1726 (s), 1247 (s), 1087 (s), 836 (s); 1H NMR (400 MHz):
d 0.047 (3H,
4.1. General
s), 0.062 (3H, s), 0.88 (9H, s), 1.17 (3H, d, J¼6.1 Hz), 1.32e1.69 (6H,
m), 1.74e1.89 (2H, m), 1.96e2.06 (2H, m), 2.69 (1H, ddd, J¼16.2, 6.3,
2.3 Hz), 2.73 (1H, ddd, J¼16.2, 7.8, 2.6 Hz), 3.43e3.52 (1H, m), 3.79
(3H, s), 3.85 (1H, ddd, J¼9.0, 4.3, 4.3 Hz), 4.01e4.08 (1H, m), 4.37
(1H, d, J¼11.4 Hz), 4.37e4.44 (1H, m), 4.49 (1H, d, J¼11.4 Hz), 5.41
(1H, dd, J¼15.5, 5.7 Hz), 5.63 (1H, dtd, J¼15.5, 6.6, 0.8 Hz),
6.84e6.89 (2H, m), 7.23e7.28 (2H, m), 9.77 (1H, dd, J¼2.6, 2.3 Hz);
IR spectra were recorded by a Jasco FT/IR-4100 spectrometer
using an ATR (ZnSe) attachment. NMR spectra were recorded with
TMS as an internal standard in CDCl3 by a Varian MR-400 spec-
trometer (400 MHz for 1H and 100 MHz for 13C) or a Varian UNITY
plus-500 spectrometer (500 MHz for 1H and 125 MHz for 13C).
Optical rotation values were measured with a Jasco DIP-371 po-
larimeter, and the mass spectra were obtained with Jeol JMS-700
spectrometer operated in the EI or FAB mode. Melting points were
determined with a Yanaco MP-J3 apparatus and are uncorrected.
Merck silica gel 60 (7e230 mesh) was used for column chroma-
tography. Solvents for reactions were distilled prior to use: THF
from Na and benzophenone; CH2Cl2 and CH3CN from CaH2; MeOH
from Mg(OMe)2; toluene from LiAlH4. All air- or moisture-sensitive
reactions were conducted under a nitrogen atmosphere.
13C NMR (100 MHz):
d
ꢁ4.9, ꢁ4.7, 18.0, 19.6, 24.9, 25.7 (3C), 27.4,
28.2, 32.3, 36.1, 42.1, 55.2, 67.8, 69.9, 70.4, 71.4, 74.2,113.6 (2C),129.1
(2C), 129.5, 131.1, 132.9, 158.9, 201.2; HRMS (EI): m/z calcd for
C28H46O5Si, 490.3115; found, 490.3113 (Mþ).
4.1.3. {(2R,3S,6R)-3-tert-Butyldimethylsilyloxy-6-[(S)-6-(4-methox-
ybenzyloxy)-1-heptenyl]tetrahydropyran-2-yl}acetaldehyde (29). To
a stirred solution of 28 (71.0 mg, 0.145 mmol) in MeOH (0.6 mL)
was added D-proline (5.2 mg, 44 m
mol) at 0 ꢀC. The mixture was
stirred at 0 ꢀC for 1 h, warmed to 60 ꢀC, and then stirred for an
additional 3 h. The mixture was diluted with EtOAc and succes-
sively washed with satd NaHCO3 aq and brine, dried (MgSO4), and
concentrated in vacuo. The residue was purified by silica gel col-
umn chromatography (hexane/EtOAc¼10:1) to give 57.4 mg (81%)
4.1.1. 2-{(2S,3S,6S)-3-tert-Butyldimethylsilyloxy-6-[(S)-6-(4-methox-
ybenzyloxy)-1-heptenyl]tetrahydropyran-2-yl}ethanol
(27). To
a stirred suspension of LiAlH4 (11.9 mg, 0.314 mmol) in THF (1 mL)
was added dropwise a solution of 26 (91.1 mg, 0.243 mmol) in THF
(2 mL) at 0 ꢀC. After being stirred at room temperature for 1 h, the
D21
of 29 as a pale yellow oil. [
a
]
þ45.6 (c 1.13, CHCl3); IR: nmax 2733
mixture was quenched by successively addition of water (10
mL),
(w), 1728 (s), 1247 (s), 1092 (vs), 836 (s); 1H NMR (400 MHz):
15% NaOH aq (10 L), and water (30 L). The mixture was dried
m
m
d
0.050 (3H, s), 0.064 (3H, s), 0.87 (9H, s), 1.16 (3H, d, J¼6.2 Hz),
(MgSO4) and filtered, and the filtrate was concentrated in vacuo to
give a diol intermediate as a pale yellow oil (98.3 mg), which was
then taken up in CH2Cl2 (3 mL). To the solution was successively
1.32e1.61 (6H, m), 1.69e1.77 (1H, m), 1.94e2.09 (3H, m), 2.48 (1H,
ddd, J¼16.2, 8.6, 3.0 Hz), 2.76 (1H, ddd, J¼16.2, 3.9, 2.0 Hz), 3.35
(1H, ddd, J¼10.0, 9.0, 4.5 Hz), 3.43e3.52 (1H, m), 3.71 (1H, ddd,
J¼8.6, 8.6, 3.9 Hz), 3.80 (3H, s), 3.78e3.86 (1H, m), 4.37 (1H, d,
J¼11.3 Hz), 4.48 (1H, d, J¼11.3 Hz), 5.40 (1H, dd, J¼15.5, 5.8 Hz),
5.63 (1H, dtd, J¼15.5, 6.7, 0.8 Hz), 6.84e6.89 (2H, m), 7.23e7.28
added Et3N (110 mL, 0.789 mmol) and TBSOTf (160 mL, 0.683 mmol)
at 0 ꢀC, and the mixture was stirred at 0 ꢀC for 30 min and at room
temperature for an additional 30 min. The mixture was quenched
with satd NaHCO3 aq and extracted with CH2Cl2. The extract was
washed with brine, dried (MgSO4), and concentrated in vacuo. The
residue was purified by silica gel column chromatography (hexane/
EtOAc¼20:1) to give a bis-TBS-protected intermediate (155 mg) as
a colorless oil, which was then dissolved in a mixture of CH2Cl2 and
MeOH (1:1, 2 mL). To the solution was added camphorsulfonic acid
(2H, m), 9.78 (1H, dd, J¼3.0, 2.0 Hz); 13C NMR (100 MHz):
d
ꢁ4.8,
ꢁ4.0, 17.9, 19.6, 24.9, 25.7 (3C), 31.3, 32.3, 33.2, 36.1, 46.6, 55.2,
69.9, 70.8, 74.3, 77.8, 77.8, 113.7 (2C), 129.1 (2C), 130.0, 131.1, 132.1,
159.0, 201.8; HRMS (EI): m/z calcd for C28H46O5Si, 490.3115; found,
490.3118 (Mþ).
(12.4 mg, 52.8
m
mol) at 0 ꢀC, and the mixture was stirred at 0 ꢀC for
4.1.4. {(2R,3S,6R)-3-tert-Butyldimethylsilyloxy-6-[(S)-6-hydroxy-1-
1 h before being quenched with satd NaHCO3 aq and extracted with
EtOAc. The extract was washed with brine, dried (MgSO4), and
concentrated in vacuo. The residue was purified by silica gel column
chromatography (hexane/EtOAc¼10:1e3:1) to give 103 mg (86%) of
heptenyl]tetrahydropyran-2-yl}acetaldehyde (30). To
a
stirred
mixture of 29 (70.0 mg, 0.143 mmol) in CH2Cl2 (0.75 mL)/1 M
phosphate buffer (pH 7.0, 0.25 mL) was added DDQ (67.5 mg,
0.288 mol) at room temperature. After 5 h, additional DDQ
(66.7 mg, 0.285 mmol) was added, and the mixture was stirred for
3 h. The mixture was extracted with CH2Cl2, and the extract was
washed with brine, dried (MgSO4), and concentrated in vacuo. The
residue was purified by silica gel column chromatography (hexane/
D15
27 as a pale yellow oil. [
a]
ꢁ5.69 (c 1.37, CHCl3); IR: nmax 3453 (br
m), 1513 (m), 1248 (s), 1088 (s), 834 (s); 1H NMR (400 MHz):
d 0.057
(3H, s), 0.064 (3H, s), 0.89 (9H, s), 1.17 (3H, d, J¼6.1 Hz), 1.34e1.62
(5H, m), 1.62e1.80 (3H, m), 1.89 (1H, ddt, J¼13.5, 6.5, 3.7 Hz),
1.98e2.12 (3H, m), 2.70 (1H, br t, J¼5.1 Hz, OH), 3.48 (1H, sex,
J¼6.1 Hz), 3.71e3.83 (3H, m), 3.80 (3H, s), 3.93 (1H, ddd, J¼10.2, 4.1,
4.1 Hz), 4.14e4.21 (1H, m), 4.37 (1H, d, J¼11.3 Hz), 4.49 (1H, d,
J¼11.3 Hz), 5.43 (1H, dd, J¼15.7, 5.9 Hz), 5.65 (1H, dtd, J¼15.7, 6.5,
0.8 Hz), 6.85e6.89 (2H, m), 7.24e7.28 (2H, m); 13C NMR (100 MHz):
EtOAc¼5:1e3:1) to give 42.2 mg (80%) of 30 as a yellow oil. [
þ49.1 (c 1.02, CHCl3); IR: nmax 3414 (br m), 2734 (w), 1728 (s), 1253
(s), 1091 (vs); 1H NMR (400 MHz):
0.050 (3H, s), 0.064 (3H, s), 0.87
a]
D18
d
(9H, s), 1.18 (3H, d, J¼6.1 Hz), 1.35e1.62 (7H, m), 1.71e1.78 (1H, m),
1.96e2.08 (3H, m), 2.48 (1H, ddd, J¼16.2, 8.6, 3.1 Hz), 2.77 (1H, ddd,
J¼16.2, 3.9, 1.8 Hz), 3.35 (1H, ddd, J¼10.0, 9.0, 4.5 Hz), 3.71 (1H, ddd,
J¼8.8, 8.8, 3.9 Hz), 3.75e3.87 (2H, m), 5.43 (1H, dd, J¼15.5, 5.8 Hz),
5.64 (1H, dtd, J¼15.5, 6.6, 0.8 Hz), 9.77e9.80 (1H, m); 13C NMR
d
ꢁ4.9, ꢁ4.7, 18.1, 19.6, 25.0, 25.8 (3C), 27.5, 28.1, 29.2, 32.3, 36.1,
55.2, 61.8, 68.6, 69.9, 70.2, 74.3, 75.9, 113.7 (2C), 129.2 (2C), 129.8,
131.1, 132.9, 158.9; HRMS (EI): m/z calcd for C28H48O5Si, 492.3271;
found, 492.3273 (Mþ).
(100 MHz):
d
ꢁ4.8, ꢁ4.0, 17.9, 23.5, 25.1, 25.7 (3C), 31.3, 32.2, 33.2,
38.7, 46.6, 67.9, 70.8, 77.8, 77.8, 130.1, 131.9, 201.9; HRMS (FAB): m/z
calcd for C20H39O4Si, 371.2618; found, 371.2619 ([MþH]þ).
4.1.2. {(2S,3S,6R)-3-tert-Butyldimethylsilyloxy-6-[(S)-6-(4-methox-
ybenzyloxy)-1-heptenyl]tetrahydropyran-2-yl}acetaldehyde (28). To
a stirred suspension of 27 (95.5 mg, 0.194 mmol) and NaHCO3
(66.4 mg, 0.790 mmol) in CH2Cl2 (2 mL) was added DesseMartin’s
periodinane (167 mg, 0.394 mmol) at room temperature. After 5 h,
the mixture was quenched with Na2S2O3 aq and extracted with
CH2Cl2. The extract was washed with brine, dried (MgSO4), and
concentrated in vacuo. The residue was purified by silica gel column
chromatography (hexane/EtOAc¼10:1) to give 91.8 mg (97%) of 28
4.1.5. {(2R,3S,6R)-3-tert-Butyldimethylsilyloxy-6-[(S)-6-hydroxy-1-
heptenyl]tetrahydropyran-2-yl}acetic acid (31). To a stirred solution
of 30 (41.6 mg, 0.112 mmol) and 2-methyl-2-butene (1.5 mL) in t-
BuOH (1.5 mL) was added to a freshly prepared mixture of NaClO2
(55.8 mg, 0.494 mmol) and 20% NaH2PO4 aq (2.7 mL) at 0 ꢀC. After
being stirred at 0 ꢀC for 7 h in the dark, the mixture was quenched
with satd NH4Cl aq and extracted with EtOAc. The extract was

T. Nagasawa et al. / Tetrahedron 67 (2011) 2882e2888
2887
washed with brine, dried (MgSO4), and concentrated in vacuo. The
residue was purified by silica gel column chromatography (hexane/
EtOAc¼2:1e1:1) to give 41.9 mg (97%) of 31 as a white solid. Mp
J¼15.6, 13.2 Hz), 3.59 (1H, br s), 4.23e4.29 (1H, m), 4.27 (1H, br s),
4.93e5.00 (1H, m), 5.72 (1H, ddd, J¼15.1, 9.3, 3.4 Hz), 5.81 (1H, ddd,
J¼15.1, 8.8, 2.0 Hz); 13C NMR (125 MHz):
d 18.6, 21.7, 22.0, 23.7, 31.1,
82.5e83.5 ꢀC; [
a
]
þ68 (c 0.92, CHCl3); IR: nmax 3375 (br w), 3293
32.2, 40.5, 66.8, 71.2, 71.5, 74.0, 132.1, 137.1, 170.0; HRMS (EI): m/z
D20
(br w), 2666 (br w), 1703 (s), 1095 (s); 1H NMR (400 MHz):
d
0.063
calcd for C14H22O4, 252.1518; found, 252.1522 (Mþ).
(6H, s), 0.87 (9H, s), 1.18 (3H, d, J¼6.3 Hz), 1.36e1.60 (6H, m),
1.70e1.77 (1H, m), 1.98e2.09 (3H, m), 2.44 (1H, dd, J¼15.7, 9.0 Hz),
2.85 (1H, dd, J¼15.7, 3.1 Hz), 3.30e3.38 (1H, m), 3.61 (1H, ddd,
J¼8.9, 8.9, 3.2 Hz), 3.75e3.89 (2H, m), 5.44 (1H, dd, J¼15.4, 5.9 Hz),
Acknowledgements
This work was financially supported, in part, by a Grant-in-Aid
for Scientific Research (B) from the Ministry of Education, Culture,
Sports, Science and Technology, Japan (No. 22380064).
5.66 (1H, dt, J¼15.4, 6.7 Hz); 13C NMR (100 MHz):
ꢁ4.8, ꢁ4.0, 17.9,
d
23.4, 25.0, 25.7 (3C), 31.1, 32.1, 33.1, 37.7, 38.6, 68.0, 70.6, 78.0, 78.7,
130.0, 132.4, 175.2; HRMS (FAB): m/z calcd for C20H39O5Si,
387.2567; found, 387.2564 ([MþH]þ).
Supplementary data
4.1.6. (1R,5S,11R,14S)-14-tert-Butyldimethylsilyloxy-5-methyl-4,15-
dioxabicyclo[9.3.1]pentadec-9-en-3-one (32). To a stirred solution
Supplementary data contains the 1H and 13C NMR spectral data
of compounds 27e33, and 1. Supplementary data related to this
article can be found online at doi:10.1016/j.tet.2011.02.061. These
data include MOL files and InChiKeys of the most important com-
pounds described in this article.
of 31 (9.0 mg, 23
mmol) and Et3N (20
m
L, 144
mmol) in THF (0.5 mL)
was added 2,4,6-trichlorobenzoyl chloride (19.0
m
L, 118 mol) at
m
0 ꢀC, and the mixture was stirred at room temperature for 1 h. The
mixture was diluted with toluene (3 mL) and stirred for an addi-
tional 0.5 h. The resulting mixture was added dropwise to a solu-
References and notes
tion of DMAP (43.5 mg, 356 m
mol) in toluene (25 mL) at 80 ꢀC over
7 h. The mixture was cooled to room temperature and then diluted
with EtOAc. The resulting solution was successively washed with
0.5 M HCl aq, satd NaHCO3 aq, and brine, dried (MgSO4), and
concentrated in vacuo. The residue was purified by preparative
TLC (Merck silica gel 60 F254, 0.25 mm thick, hexane/EtOAc¼5:1)
to give 2.6 mg (30%) of 32 and 3.2 mg (19%) of macrodiolide 33 as
1. Kito, K.; Ookura, R.; Yoshida, S.; Namikoshi, M.; Ooi, T.; Kusumi, T. Org. Lett.
2008, 10, 225e228.
2. Nagasawa, T.; Kuwahara, S. Org. Lett. 2009, 11, 761e764.
3. Hande, S. M.; Uenishi, J. Tetrahedron Lett. 2009, 50, 189e192.
4. Ookura, R.; Kito, K.; Saito, Y.; Kusumi, T.; Ooi, T. Chem. Lett. 2009, 38, 384.
5. For the only example of a natural fourteen-membered macrolide containing
a
2,6-cis-substituted tetrahydropyran ring as a nonhemiacetal form, see:
Youngsaye, W.; Lowe, J. T.; Pohlki, F.; Ralifo, P.; Panek, J. S. Angew. Chem., Int. Ed.
2007, 46, 9211e9214 and a reference cited therein.
6. For the only example of a natural fourteen-membered macrolides containing
a 2,6-trans-substituted tetrahydropyran ring as a nonhemiacetal form, see:
Shinonaga, H.; Kawamura, Y.; Ikeda, A.; Aoki, M.; Sakai, N.; Fujimoto, N.; Ka-
washima, A. Tetrahedron Lett. 2009, 50, 108e110.
D20
colorless oils. 32: [
a
]
ꢁ30 (c 0.20, CHCl3); IR: nmax 1733 (s), 1660
(w), 1065 (s), 1029 (s); 1H NMR (400 MHz):
d
0.047 (3H, s), 0.053
(3H, s), 0.90 (9H, s), 1.21 (3H, d, J¼6.7 Hz), 1.30e1.38 (1H, m),
1.46e1.61 (3H, m), 1.70e1.82 (1H, m), 1.86e1.97 (2H, m), 2.06e2.16
(1H, m), 2.21e2.31 (2H, m), 2.38 (1H, dd, J¼14.5, 4.5 Hz), 2.53 (1H,
dd, J¼14.5, 12.7 Hz), 3.53 (1H, dt, J¼4.7, 2.9 Hz), 4.08 (1H, dm,
J¼12.7 Hz), 4.14e4.20 (1H, m), 4.93e5.02 (1H, m), 5.73 (1H, ddd,
J¼15.3, 8.8, 3.3 Hz), 5.79 (1H, ddd, J¼15.3, 7.7, 1.4 Hz); 13C NMR
7. (a) Nagasawa, T.; Kuwahara, S. Tetrahedron Lett. 2010, 51, 875e877; (b) Díaz-
ꢀ
Oltra, S.; Angulo-Pachon, C. A.; Murga, J.; Carda, M.; Marco, J. A. J. Org. Chem.
2010, 75, 1775e1778; (c) Fuwa, H.; Yamaguchi, H.; Sasaki, M. Org. Lett. 2010, 12,
1848e1851; (d) Sabitha, G.; Reddy, D. V.; Rao, A. S.; Yadav, J. S. Tetrahedron Lett.
2010, 51, 4195e4198; (e) Fuwa, H.; Yamaguchi, H.; Sasaki, M. Tetrahedron 2010,
(100 MHz):
d
ꢁ4.8, ꢁ4.7, 18.1, 18.5, 23.2, 24.1, 24.2, 25.8 (3C), 30.9,
ꢀ
66, 7492e7503; (f) Díaz-Oltra, S.; Angulo-Pachon, C. A.; Murga, J.; Falomir, E.;
Carda, M.; Marco, J. A. Chem.dEur. J. 2011, 17, 675e688; (g) Kanematsu, M.;
Yoshida, M.; Shishido, K. Angew. Chem., Int. Ed. 2011, 50, 1e4.
31.8, 40.8, 68.0, 71.4, 72.0, 74.9, 132.8, 136.9, 170.7; HRMS (FAB): m/
z calcd for C20H37O4Si, 369.2461; found, 369.2465 ([MþH]þ). 33:
ꢀ
8. (a) Díaz-Oltra, S.; Angulo-Pachon, C. A.; Kneeteman, M. N.; Murga, J.; Carda, M.;
D23
[
a]
þ49 (c 0.39, CHCl3); IR: nmax 1731 (s), 1186 (m), 1130 (m),
Marco, J. A. Tetrahedron Lett. 2009, 50, 3783e3785; (b) Liu, J.; Xu, K.; He, J.;
Zhang, L.; Pan, X.; She, X. J. Org. Chem. 2009, 74, 5063e5066; (c) Nagasawa, T.;
Kuwahara, S. Biosci. Biotechnol. Biochem. 2009, 73, 1893e1894; (d) Hendrix, A. J.
M.; Jennings, M. P. Tetrahedron Lett. 2010, 51, 4260e4262.
1089 (s); 1H NMR (400 MHz):
d 0.048 (6H, s), 0.059 (6H, s), 0.88
(18H, s), 1.20 (6H, d, J¼6.3 Hz), 1.31e1.50 (8H, m), 1.51e1.62 (4H,
m), 1.71e1.78 (2H, m), 1.83e1.95 (2H, m), 1.95e2.12 (4H, m), 2.27
(2H, dd, J¼13.7, 10.8 Hz), 2.80 (2H, dd, J¼13.7, 2.3 Hz), 3.31 (2H,
ddd, J¼10.2, 9.2, 4.7 Hz), 3.62 (2H, ddm, J¼8.6, 2.3 Hz), 3.73e3.80
(2H, m), 5.01e5.11 (2H, m), 5.38 (2H, dd, J¼15.5, 5.3 Hz), 5.57 (2H,
9. (a) Panarese, J. D.; Waters, S. P. Org. Lett. 2009, 11, 5086e5088; (b) Kanematsu,
M.; Yoshida, M.; Shishido, K. Tetrahedron Lett. 2011, 52, 1372e1374.
10. We did not examine the acid-promoted epimerization of 8 due to the presence
of the acid-labile TBS protecting group at the C4 position.
11. Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull. Chem. Soc. Jpn.
1979, 52, 1989e1993.
12. (a) Shiina, I.; Kubota, M.; Ibuka, R. Tetrahedron Lett. 2002, 43, 7535e7539; (b)
Shiina, I.; Kubota, M.; Oshiumi, H.; Hashizume, M. J. Org. Chem. 2004, 69,
1822e1830.
13. The reduction of 18 with NaBH4 was non-stereoselective, affording aspergillide
B (the C4 epimer of 17) in 50% yield (two steps) in addition to the desired
product 17 (45% yield).
dd, J¼15.5, 6.5 Hz); 13C NMR (100 MHz):
d
ꢁ4.8 (2C), ꢁ4.0 (2C),
17.9 (2C), 20.3 (2C), 24.6 (2C), 25.7 (6C), 31.4 (2C), 32.2 (2C), 33.3
(2C), 35.9 (2C), 38.7 (2C), 70.2 (2C), 70.7 (2C), 77.3 (2C), 80.1 (2C),
130.0 (2C), 131.1 (2C), 172.3 (2C); HRMS (EI): m/z calcd for
C40H72O8Si2, 736.4766; found, 736.4771 (Mþ).
14. A recent report by Shishido and co-workers revealed that the treatment of the
C4-epimer of 15 (compound 32 in Scheme 5) prepared by their own synthetic
route epimerized into its thermodynamically more stable C3-epimer (com-
pound 3-epi-32 in Fig. 2) in 94% yield by treating with KH and 18-crown-6 in
THF at 0 ꢀC for 30 min. This may suggest another possibility that 3-epi-15 might
have been formed from the undetectable macrocyclization product 15 via
a retro-oxy-Michael/oxy-Michael process. For details, see Ref. 7g.
15. Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277e7287.
16. (a) Corey, E. J.; Nicolaou, K. C. J. Am. Chem. Soc. 1974, 96, 5614e5616; (b) Gerlach,
H.; Thalmann, A. Helv. Chim. Acta 1974, 57, 2661e2663; (c) Parenty, A.; Moreau,
X.; Campagne, J.-M. Chem. Rev. 2006, 106, 911e939.
4.1.7. (1R,5S,11R,14S)-14-Hydroxy-5-methyl-4,15-dioxabicyclo[9.3.1]
pentadec-9-en-3-one (1). To a stirred solution of 32 (1.9 mg,
5.2
15
m
mol) in THF (0.1 mL) was added TBAF (1 M solution in THF,
m
L, 15 mol) at room temperature, and the mixture was stirred
m
for 2 h before being diluted with EtOAc. The resulting solution was
successively washed with satd NH4Cl aq and brine, dried (MgSO4),
and concentrated in vacuo. The residue was purified by silica gel
column chromatography (hexane/EtOAc¼3:1e1:3) to give 1.0 mg
17. The transformation (23/24) was effected in 20% yield by Fuwa and co-workers
in their unified total synthesis of aspergillides A and B (Refs. 7c and 7e).
18. Massi, A.; Nuzzi, A.; Dondoni, A. J. Org. Chem. 2007, 72, 10279e10282 This
(76%) of 1 as a white solid. Mp 64.5e65.5 ꢀC; [
a
]
ꢁ59 (c 0.13,
D23
CHCl3) [lit.1 [
1665 (w), 1011 (s), 976 (s); 1H NMR:
a
]
ꢁ59.5 (c 0.45, CHCl3)]; IR: nmax 3495 (m), 1735 (s),
D27
method was originally developed for the epimerization of
a-C-glycosylmethyl
d
1.21 (3H, d, J¼6.8 Hz),
aldehydes with l-proline into the corresponding -isomers. Taking into account
b
1.38e1.44 (1H, m), 1.48e1.57 (2H, m), 1.70e1.76 (1H, m), 1.79e1.87
(1H, m), 1.89e1.99 (2H, m), 2.08e2.17 (1H, m), 2.17e2.26 (2H, m),
2.28e2.34 (1H, m), 2.40 (1H, dd, J¼15.6, 4.4 Hz), 2.65 (1H, dd,
the absolute stereochemistry of d-sugars employed as the substrates by Massi
and co-workers, we employed d-proline as the first choice. The use of l-proline,
however, gave almost the same result as that of d-proline..

2888
T. Nagasawa et al. / Tetrahedron 67 (2011) 2882e2888
19. For examples of related epimerization reactions using catalysts other than pro-
line, see: (a) Tatsuta, K.; Suzuki, Y.; Toriumi, T.; Furuya, Y.; Hosokawa, S. Tetra-
hedron Lett. 2007, 48, 8018e8021; (b) Hinkle, R. J.; Lian, Y.; Litvinas, N. D.; Jenkins,
A. T.; Burnette, D. C. Tetrahedron 2005, 61, 11679e11685; (c) Wang, Z.; Shao, H.;
Lacroix, E.; Wu, S.-H.; Jennings, H. J.; Zou, W. J. Org. Chem. 2003, 68, 8097e8105;
20. (a) Mukaiyama, T. Angew. Chem., Int. Ed. Engl. 1979, 18, 707e721; (b) Smith, A.
B., III; Dong, S.; Brenneman, J. B.; Fox, R. J. J. Am. Chem. Soc. 2009, 131,
12109e12111.
21. The m-bromobenzoate of 1 analogous to 32 was revealed to exist in a confor-
mation with the C3-, C4-, and C7-substituents axially oriented by its X-ray
crystallographic analysis, and the MOM-protected congener (24 in Scheme 4) of
32 was also shown to adopt essentially the same conformation as the m-bro-
mobenzoate. For details, see Refs. 4 and 7e.
(d) Shao, H.; Wang, Z.; Lacroix, E.; Wu, S.-H.; Jennings, H. J.; Zou, W. J.ˇ Am. Chem.
Soc. 2002, 124, 2130e2131; (e) Michelet, V.; Adiey, K.; Bulic, B.; Genet, J.-P.; Du-
jardin, G.; Rossignol, S.; Brown, E.; Toupet, L. Eur. J. Org. Chem. 1999, 2885e2892.