Asymmetric total synthesis of (-)-merrilactone A: Use of a bulky protecting group as long-range stereocontrolling element

Article abstract of DOI:10.1002/anie.200601358

(Chemical Equation Presented) Designer elegance: The transannular aldol reaction of a cyclooctene diketone is the key step in this total synthesis of the natural enantiomer of merrilactone A (see scheme). The configuration of the two stereocenters generated in the formation of the central bicyclo[3.3.0]octane framework of the natural product was established using a specially designed bulky protecting group.

Full text of DOI:10.1002/anie.200601358

Angewandte
Chemie
Total Synthesis
of 1 was established on the basis of the modified Mosher
method[4] by using MTPA derivatives at C2 OH. In
addition to its evident architectural complexity, 1 is also
very complex from a stereochemical point of view: Five of its
seven asymmetrically substituted carbon atoms are contigu-
ous and carry only non-hydrogen substituents (C6, C5, C4,
C9, C1). The molecular architecture of 1 is a daunting
challenge for chemical synthesis.
[1]
ꢀ
DOI: 10.1002/anie.200601358
Asymmetric Total Synthesis of (ꢀ)-Merri-
lactone A: Use of a Bulky Protecting Group as
Long-Range Stereocontrolling Element**
Motivated by the important biological activity and
unusual structure of 1, we began a synthetic study, which
resulted in the total synthesis of racemic merrilactone A
((ꢁ )-1) in 2003.[5] To date, two other chemical routes to (ꢁ )-1
have been developed by the Danishefsky and Mehta research
groups,[6,7] and Danishefsky and co-workers have reported an
elegant asymmetric synthesis of a key intermediate en route
to 1.[8] Herein, we report an asymmetric total synthesis of the
natural enantiomer (ꢀ)-1 in which a remote bulky protecting
group is used to control the stereochemistry. This study also
confirmed the assigned absolute configuration for the first
time.
Masayuki Inoue,* Takaaki Sato, and Masahiro Hirama
(ꢀ)-Merrilactone A ((ꢀ)-1, Figure 1), a sesquiterpenoid iso-
lated from Illicium merrillianum in 2000 by Fukuyama and co-
workers,[1,2] has been shown to possess neuroprotective and
In our total synthesis of (ꢁ )-1, the bicyclo[3.3.0]octane
framework was constructed in the form of (ꢁ )-4 through an
intramolecular aldol reaction of the benzyl-protected meso
diketone 2 (Scheme 1).[9,10] Importantly, by controlling the
Figure 1. Structure of (ꢀ)-merrilactone A.
neuritogenic activity in cultures of fetal rat cortical neurons.
Small molecules with these neurotrophic effects are expected
to be useful as metabolically stable alternatives to endoge-
nous neurotrophic factors, and thus as therapeutic agents for
the neurodegeneration associated with Alzheimerꢀs and
Parkinsonꢀs diseases.[3] By X-ray crystallographic analysis
together with extensive NMR spectroscopic studies, 1 was
determined to possess a unique pentacyclic cage structure
comprising a central bicyclo[3.3.0]octane framework, two d-
lactone rings, and an oxetane ring. The absolute configuration
[*] Prof. M. Inoue, Dr. T. Sato, Prof. M. Hirama
Department of Chemistry
Graduate School of Science, Tohoku University
Sendai 980-8578 (Japan)
Fax: (+81)22-795-6566
E-mail: inoue@ykbsc.chem.tohoku.ac.jp
Scheme 1. Diastereoselective transannular aldol reaction of the meso
diketone 2 in the total synthesis of (ꢁ)-1.[5] Bn=benzyl.
Prof. M. Inoue
Research and Analytical Center for Giant Molecules
Graduate School of Science, Tohoku University
and
reaction conditions (using LiN(SiMe3)2 in THF) it was
possible to favor the selective formation of the desired syn
isomer (ꢁ )-4 over that of the anti isomer (ꢁ )-5. The key
intermediate (ꢁ )-4 was then converted into (ꢁ )-1 through a
series of functional-group transformations.
Our plan for preparing enantiomerically pure merrilac-
tone A (ꢀ)-1 was based on the transannular aldol chemistry
described above. Theoretically, exclusive deprotonation of C9
of diketone 2 would lead to the bicyclo[3.3.0]octane system 4
with the absolute configuration found in the natural product
(2!3!4 versus 2!ent-3!ent-4, Scheme 1). We were
intrigued by the possibility of differentiating the deprotona-
PRESTO
Japan Science and Technology Agency
Sendai 980-8578 (Japan)
[**] This research was supported by a Grant-in-Aid for Young Scientists
(A) from the Japan Society for the Promotion of Science (JSPS) and
a SUNBOR Grant awarded to M.I. A fellowship awarded to T.S. by
the JSPS is gratefully acknowledged. We thank Professor Yoshiyasu
Fukuyama (Tokushima Bunri University) for providing NMR spectra
of merrilactone A and its MTPA derivatives, and Dr. Chizuko Kabuto
for crystallographic analyses.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 4843 –4848
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4843

Communications
tion rates at C9 and C3 by utilizing a pseudo-meso substrate:
ꢀ
The attachment of a bulky protecting group at C14 OH
ꢀ
would effectively shield C3 H through a long-range steric
interaction (Scheme 2).[11] To apply the reaction sequence
Scheme 2. Strategy for the asymmetric synthesis of the core bicyclo-
[3.3.0]octane framework of (ꢀ)-merrilactone A from a pseudo-meso
diketone.
developed from (ꢁ )-4 to (ꢁ )-1, this bulky protecting group
needed to be as stable as Bn in a variety of reactions and to be
removed simultaneously with Bn in the final stage of the total
synthesis. Since no existing alcohol protecting group met
these requirements, we designed a new benzyl ether, 2,6-
bis(trifluoromethyl)benzyl (BTB) ether, which has chemically
inert CF3 groups at the two ortho positions to impose a large
steric effect.[12] Thus, our subsidiary goals in the asymmetric
total synthesis were to prepare the differentially protected
diketone 6 and to evaluate the BTB group as a remote
stereocontrolling element for the selective generation of 7
and 8.
To establish the absolute configuration of the two
quaternary carbons atoms (C5, C6) of diketone 6, we planned
a [ +22] photocycloaddition of the chiral tetrasubstituted
olefin 13 (Scheme 3).[13,14] The reduction of 2,3-dimethylma-
leinic anhydride (9) to 10,[15] followed by Wittig olefination
and esterification, afforded methyl ester 11 in 68% yield
(three steps). Chemo- and enantioselective dihydroxylation
of dienone 11 under Sharpless asymmetric dihydroxylation
conditions with the catalyst (DHQ)2PHAL led to enantio-
merically pure 12 (> 99% ee) in 65% yield after one
recrystallization.[16] This hydroxy-g-lactone was protected as
its pivaloate ester 13.[17] The irradiation of 13 in the presence
of cis-1,2-dichloroethylene with a high-pressure mercury
lamp, followed by Zn-promoted dechlorination then afforded
14, the LiAlH4 reduction of which gave cyclobutene 15 (75%)
along with its facial diastereomer (8%). Thus, furanone 13
showed excellent facial discrimination (9.8:1) in the photo-
cycloaddition and acted as a template for the stereoselective
introduction of the two quaternary carbon atoms C5 and
C6.[18] The structure of triol 15 was determined unambigu-
ously by single-crystal X-ray analysis of 23 (Figure 2), the
product of mono-p-bromobenzoylation of 15.
Scheme 3. Synthesis of the pseudo-meso diketone 6; (the principal
reagents are also shown in the Scheme): a) LiAlH(OtBu)3, DME,
ꢀ158C!RT, 85%; b) Ph3PCH3+Brꢀ, tBuOK, 08C!RT, 87%; c) MeI,
K2CO3, THF, 508C, 92%; d) AD-mix-a, tBuOH/H2O (1:1), 08C, 90%,
90% ee; then recrystallization: 65%, >99% ee; e) PivCl, py, DMAP,
CH2Cl2, room temperature, 99%; f) cis-dichloroethylene, CH3CN,
ꢀ208C; g) Zn, Ac2O, toluene, 1208C; h) LiAlH4, Et2O, room temper-
ature, 75% (15, 3 steps), 8% (the facial diastereomer, 3 steps);
i) Me2C(OMe)2, TsOH·H2O, CH2Cl2, room temperature, 81%; j) BnBr,
NaH, THF/DMF (10:1), room temperature; k) THF/3m HCl (5:1),
room temperature, 91% (2 steps); l) Pb(OAc)4, py, CH2Cl2, ꢀ508C;
then DIBAL, ꢀ788C!ꢀ508C, 93% (90% conversion); m) BTBBr, KH,
[18]crown-6, DMF, room temperature; n) OsO4, NMO, tBuOMe/
tBuOH/H2O (2:1:1), room temperature, 94% (89% conversion; 2
steps); o) SO3·Py, iPr2NEt, DMSO, CH2Cl2, ꢀ158C; then allylmagne-
=
sium bromide, ꢀ788C, 78% (21a/21b 2.7:1); p) [(PCy3)2Cl2Ru CHPh],
CH2Cl2, reflux; then Pb(OAc)4, room temperature, 97%. Cy=cyclo-
hexyl, DMAP=4-dimethylaminopyridine, DME=dimethoxyethane,
DIBAL=dibutylaluminum hydride, DMF=N,N-dimethylformamide,
DMSO=dimethyl sulfoxide, NMO=N-methylmorpholine N-oxide,
Piv=pivaloyl, py=pyridine, Ts=p-toluenesulfonyl.
ꢀ
isopropylidene acetal 16, Bn protection of the C12 OH group
of 16 and subsequent removal of the acetonide under acidic
conditions delivered the 1,2-diol 17 (91%, two steps). One-
carbon-atom truncation from 17 to liberate the masked
ꢀ
primary C14 OH functionality was carried out in a single-
flask reaction involving Pb(OAc)4-induced oxidative cleavage
and DIBAL reduction, and led to 18 in 93% yield.[19] The
ꢀ
BTB group was then introduced at C14 OH in 18 to afford
differentially protected 19.
The additional hydroxymethyl group attached to C14 of
15 made it possible to introduce the two benzyl-type protect-
ing groups at the C12 and C14 alcohol groups in a stepwise
fashion (Scheme 3). After triol 15 had been protected as its
By exploiting the pseudo-meso symmetry of 19, diketone
6 was prepared in only three steps through pairwise function-
alization.[20] Olefin 19 was subjected to dihydroxylation to
afford diol 20, the one-pot treatment of which with SO3·py[21]
4844 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4843 –4848

Angewandte
Chemie
mation (8/25) to give the desired bicyclo[3.3.0]octane system
[25] along with smaller amounts of the other three diastereo-
8
mers 25–27. To enhance the formation of 8, the effect of the
counter cation of the amide base was examined (entries 2, 3):
When 6 was treated with NaN(SiMe3)2, enantiomerically pure
8 was isolated in 75% yield after purification by column
chromatography with SiO2. Interestingly, the selectivity of the
same base treatment of 6a, which possesses the less sterically
demanding protecting group 2,6-dichlorobenzyl (DCB; effec-
tive radii: 2.2 Š (CF3) versus 1.7 Š (Cl)),[26] was lower in both
ꢀ
the deprotonation and the C C bond-forming steps (entry 4).
This result clearly suggests that the steric bulk of the ortho
substituents of the phenyl ring has a significant effect on the
selectivity of the reaction. As expected, our BTB group
functioned as a long-range stereocontrolling element for the
aldol reaction.
Figure 2. X-ray crystallographic analysis of compound 23 to confirm
the structure of triol 15.[41]
and an allyl Grignard reagent provided adducts 21a and 21b
in 78% yield.[22] The cis relationship of the olefinic side chains
of 21a and 21b facilitated the subsequent ring-closing meta-
thesis reaction to produce the bicyclo[4.2.0]octyl system 22,[23]
which was subjected in situ to Pb(OAc)4-promoted oxidative
ring expansion[24] to yield the substituted eight-membered
ring 6.
A plausible mechanism for the reaction is shown in
Scheme 5. Molecular modeling (MM2*, MacroModel Ver-
Having established a route to the differentially protected
diketone 6, we undertook the crucial transannular aldol
reaction. To our gratification, the reaction of 6 with LiN-
(SiMe3)2 (Scheme 4, entry 1) exhibited both site-selective
ꢀ
deprotonation (7/24) and diastereoselective C C bond for-
Scheme 5. Plausible mechanism for the diastereoselective transannular
aldol reaction of 6.
sion 8.5)[27] indicated that the eight-membered ring exists as a
mixture of two pseudoenantiomeric conformers 6 and 6’.[28]
Since cis-enolate formation from the eight-membered ring is
energetically more favorable than trans-enolate formation, in
each conformer only one of the two protons (indicated in bold
=
face in Scheme 5) orthogonal to the C O bonds is thought to
be abstracted by the base. The selective deprotonation of 6 to
generate the cis enolate 7 can be explained by effective
insulation of C9 of 6’ by the remote bulky BTB group. After
enolate formation, the severe 1,3-diaxial-like steric interac-
tion between the large BTB-protected oxymethylene group
[29]
ꢀ
and the C7 O bond in 7 would enforce a conformational
flip of the C7–C9 olefin to form 7’, from which the enolate can
react with the ketone at C4 to generate the desired cis-fused
5,5 ring system 8. The proposed mechanism agrees well with
the observation that the bulkier protecting group is more
Scheme 4. Diastereoselective transannular aldol reaction of pseudo-
meso diketones 6 and 6a. The yields of entry 1 are based on recovered
starting material (82% conversion).
Angew. Chem. Int. Ed. 2006, 45, 4843 –4848
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4845

Communications
ꢀ
selective in both the deprotonation and C C bond-forming
steps (Scheme 4, entries 2 and 4).
triflate 36,[34,35] which was converted into the trisubstituted
alkene 37 through palladium-mediated reduction.[36] The
exposure of 37 to Na in NH3 effected both the stereose-
[37]
With the enantiomerically pure bicyclo[3.3.0]octane
framework 8 in hand, we proceeded to synthesize the entire
carboskeleton of (ꢀ)-1 by introduction of the C9 quaternary
center and the C15 methylene group (Scheme 6). a-Epox-
idation of 8 and subsequent florisil treatment produced allylic
alcohol 28, which was converted into enedione 29.[30] An a-
bromoacetal unit was then introduced to afford 30 as a
mixture of diastereomers.[31]
Despite the steric congestion induced around C9 by the
three proximal tetrasubstituted carbon atoms (C4, C5, C6),
radical cyclization of 30 smoothly delivered the 5-exo cyclized
product 31 along with its C11 epimer 32 in 90% combined
yield. Upon treatment with EtOSiMe3 in the presence of
BF3·Et2O, the minor isomer 32 was transformed into the
major isomer 31 in 72% yield. NOESY data obtained for 31
allowed the assignment of configuration and conformation
(Scheme 6): The CH2 group at position 3 was found to be in
spatial proximity to the ethoxy and BTB-protected oxy-
methylene groups. Presumably because of low base accessi-
bility to the sterically crowded C3 center, in the next step the
reaction of 31 with Me3SiOTf and Et3N produced predom-
inantly silyl enol ether 33 through site-selective deprotonation
at the less-shielded C1 center. The C15 methylene group was
introduced by the treatment of 33 with the Eschenmoser
reagent[32] and then with mCPBA to give 34.
lective reduction of the hindered ketone at C7 to the b-
hydroxy group, presumably via 38 through a six-membered
chelate ring, and the removal of both benzyl-type protecting
groups (Bn and BTB) to give lactol 40 along with lactone 39.
The mixture of 39 and 40 was in turn subjected to Fetizon
oxidation[38] to produce the desired bis-g-lactone 41 as a single
isomer. It appears that the reactivity of the hydroxy group at
C12 in 39/40 towards oxidation is higher than that of the
hydroxy groups at C7 and C14 as a
result of its more exposed nature:
Molecular modeling of 40 suggested
that only the C12 hydroxymethyl
group adopts a pseudoequatorial
position, as depicted in Scheme 7.
Thus, the remarkable selectivities of
the reduction (37!40) and oxida-
tion (40!41) steps are governed by
Scheme 7. Conforma-
the intrinsic three-dimensional ori-
entation of the reacting substituents.
Moreover, the stereochemical con-
trolling factor for the aldol reaction,
tion of 40 according to
molecular modeling.
the BTB ether, was shown to be robust under a variety of
reaction conditions up to those for the synthesis of 39, yet was
removed smoothly through Birch reduction.
To complete the total synthesis, the remaining functional-
group manipulations needed to be orchestrated judiciously.
First, acetal 34 was converted quantitatively into g-lactone
35.[33] Hydride addition to the exo alkene of the unsaturated
ketone 35, followed by in situ triflation, afforded the enol
Finally, stereoselective a epoxidation of 41 with dime-
thyldioxirane[39] to give 42 and subsequent acid-mediated
epoxide opening–oxetane formation delivered (ꢀ)-merrilac-
tone A ((ꢀ)-1).[1] Synthetic (ꢀ)-1 exhibited 1H NMR,
13C NMR, IR, and HRMS spectra that were indistinguishable
Scheme 6. Asymmetric total synthesis of (ꢀ)-merrilactone A; (the principal reagents are also shown in the Scheme): a) mCPBA, CH2Cl2; then
florisil, CH2Cl2, room temperature, 75%; b) IBX, DMSO, room temperature, 91%; c) BrCH2Br(OEt), PhNMe2, CH2Cl2, ꢀ788C!RT, 92% (d.r.
4.4:1, 79% conversion); d) Bu3SnH, AIBN, toluene, 858C, 73% (31), 17% (32); e) EtOSiMe3, BF3·Et2O, CH2Cl2, room temperature, 72%;
f) Me3SiOTf, Et3N, CH2Cl2, ꢀ208C; g) Me2NCH2+Iꢀ, CH3CN, room temperature; h) mCPBA, CH2Cl2, room temperature, 64% (3 steps); i) mCPBA,
BF3·Et2O, CH2Cl2, room temperature, 100%; j) LiBH(sBu)3, 2-Tf2N-5-chloropyridine, THF, ꢀ788C, 73%; k) Pd(OAc)2, PPh3, NBu3, HCOOH, DMF,
408C, 92%; l) Na, NH3, THF/EtOH (5:1), ꢀ788C (39/40 1:1.4); m) Ag2CO3 on celite, toluene, 1308C, 41% (2 steps); n) DMDO, CH2Cl2, room
temperature, 91%; o) TsOH·H2O, CH2Cl2, room temperature, 96%. AIBN=N,N-azobisisobutyronitrile, DMDO=dimethyldioxirane, IBX=
o-iodoxybenzoic acid, mCPBA=m-chloroperbenzoic acid, Tf=trifluoromethanesulfonyl, Ts=p-toluenesulfonyl.
4846 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4843 –4848

Angewandte
Chemie
J. Am. Chem. Soc. 1993, 115, 12189; b) L. A. Paquette, F. Geng,
from those of the natural compound. The measured optical
rotation of synthetic (ꢀ)-1 confirmed the absolute config-
uration of the natural product ([a]2D7 = ꢀ15.7 (c = 0.19,
CHCl3); natural (ꢀ)-1: [a]1D8 = ꢀ16.7 (c = 1.10, CHCl3)).[2,40]
In summary, we have completed the asymmetric total
synthesis of (ꢀ)-merrilactone A (1.1% overall yield, 31
steps). The defining step in our synthesis is the diastereose-
lective transannular aldol reaction of 6 to construct the
bicyclo[3.3.0]octane core 8. The selectivity observed in the
construction of the two new stereocenters in 8 is a long-range
effect of the bulky protecting group BTB. Other remarkable
features of our total synthesis include 1) a [2+2] cycloaddition
to install the two contiguous quaternary carbon atoms of 14,
2) efficient pairwise symmetrical functionalization to synthe-
size 6 by taking advantage of the pseudo-meso symmetry,
3) radical cyclization to form the sterically congested C9
quaternary carbon atom of 31, and 4) highly selective
substrate-controlled reactions to introduce three functional
groups: the C15 methylene group of 34, the b-hydroxy group
at C7 of 40, and the C12-containing g-lactone in 41.
J. Am. Chem. Soc. 2002, 124, 9199; c) L. A. Paquette, Eur. J. Org.
Chem. 1998, 1709; d) P. A. Wender, C. R. D. Correia, J. Am.
Chem. Soc. 1987, 109, 2523; e) P. A. Wender, G. G. Gamber,
R. D. Hubbard, L. Zhang, J. Am. Chem. Soc. 2002, 124, 2876;
f) M. Zora, I. Koyuncu, B. Yucel, Tetrahedron Lett. 2000, 41,
7111; g) S. K. Verma, E. B. Fleischer, H. W. Moore, J. Org.
Chem. 2000, 65, 8564; h) D. M. Hodgson, I. D. Cameron, Org.
Lett. 2001, 3, 441; i) K. G. Dongol, R. Wartchow, H. Butenschön,
Eur. J. Org. Chem. 2002, 1972; j) T. Hamura, S. Tsuji, T.
Matsumoto, K. Suzuki, Chem. Lett. 2002, 280.
[11] For recent applications of O-protecting groups as long-range
stereocontrolling elements, see: a) G. Guanti, S. Perrozzi, R.
Riva, Tetrahedron: Asymmetry 2002, 13, 2703; b) I. A. I. Ali,
E. S. H. El Ashry, R. R. Schmidt, Eur. J. Org. Chem. 2003, 4121;
c) M. Inoue, T. Sasaki, S. Hatano, M. Hirama, Angew. Chem.
2004, 116, 6662; Angew. Chem. Int. Ed. 2004, 43, 6500; d) H.
Tokimoto, Y. Fujimoto, K. Fukase, S. Kusumoto, Tetrahedron:
Asymmetry 2005, 16, 441.
[12] It is important from a practical viewpoint that the requisite bulky
benzyl bromide BTBBr (Scheme 3) was prepared in two steps
from commercially available 1,3-bis(trifluoromethyl)benzene
(see the Supporting information).
[13] For reviews on the construction of quaternary stereocenters, see:
a) E. J. Corey, A. Guzman-Perez, Angew. Chem. 1998, 110, 402;
Angew. Chem. Int. Ed. 1998, 37, 388; b) J. Christoffers, A. Mann,
Angew. Chem. 2001, 113, 4725; Angew. Chem. Int. Ed. 2001, 40,
4591; c) I. Denissova, L. Barriault, Tetrahedron 2003, 59, 10105;
d) C. J. Douglas, L. E. Overman, Proc. Natl. Acad. Sci. USA
2004, 101, 5363.
[14] a) R. AlibØs, P. D. March, M. Figueredo, J. Font, M. Racamonde,
A. Rustullet, A. Alvarez-Larena, J. F. Piniella, T. Parella,
Tetrahedron Lett. 2003, 44, 69; b) R. AlibØs, P. D. March, M.
Figueredo, J. Font, M. Racamonde, T. Parella, Org. Lett. 2004, 6,
1449.
[15] T. Harrison, P. L. Myers, G. Pattenden, Tetrahedron 1989, 45,
5247.
[16] a) H. Becker, M. A. Soler, K. B. Sharpless, Tetrahedron 1995, 51,
1345; b) H. C. Kolb, M. S. VanNieuwenhze, K. B. Sharpless,
Chem. Rev. 1994, 94, 2483.
[17] For an alternative synthetic route to a related compound, see: S.
Hanessian, P. J. Murray, J. Org. Chem. 1987, 52, 1170.
[18] For recent reviews on cyclobutane derivatives, see: a) E. Lee-
Ruff, G. Mladenova, Chem. Rev. 2003, 103, 1449; b) J. C.
Namyslo, D. E. Kaufmann, Chem. Rev. 2003, 103, 1485.
[19] The yield of this reaction was highly dependent on the reaction
temperature. When the oxidative cleavage of 17 was performed
at room temperature, the resulting aldehyde underwent facile
electrocyclic ring opening to generate a 2Z,4E dienal as the sole
product. For related reactions, see: F. Binns, R. Hayes, S.
Ingham, S. T. Saengchantara, R. W. Turner, T. W. Wallace,
Tetrahedron 1992, 48, 515.
[20] For reviews on related approaches, including two-directional
synthesis, see: a) C. S. Poss, S. L. Schreiber, Acc. Chem. Res.
1994, 27, 9; b) S. R. Magnuson, Tetrahedron 1995, 51, 2167.
[21] J. R. Parikh, W. E. Doering, J. Am. Chem. Soc. 1967, 89, 5505.
[22] R. E. Ireland, D. W. Norbeck, J. Org. Chem. 1985, 50, 2198.
[23] a) G. C. Fu, S. T. Nguyen, R. H. Grubbs, J. Am. Chem. Soc. 1993,
115, 9856; b) P. Schwab, R. H. Grubbs, J. W. Ziller, J. Am. Chem.
Soc. 1996, 118, 100; c) A. Fürstner, Angew. Chem. 2000, 112,
3140; Angew. Chem. Int. Ed. 2000, 39, 3012; d) T. M. Trnka,
R. H. Grubbs, Acc. Chem. Res. 2001, 34, 18; e) K. C. Nicolaou,
P. G. Bulger, D. Sarlah, Angew. Chem. 2005, 117, 4564; Angew.
Chem. Int. Ed. 2005, 44, 4490.
Received: April 6, 2006
Published online: June 23, 2006
Keywords: aldol reaction · asymmetric synthesis ·
.
medium-ring compounds · protecting groups · terpenoids
[1] a) J.-M. Huang, R. Yokoyama, C.-S. Yang, Y. Fukuyama,
Tetrahedron Lett. 2000, 41, 6111; b) J.-M. Huang, C.-S. Yang,
M. Tanaka, Y. Fukuyama, Tetrahedron 2001, 57, 4691.
[2] The reported optical rotation of merrilactone A ([a]2D1 = + 11.8
(c = 1.20, CH3OH), reference [1a]) was found to be an error.
The correct value for the natural product is: [a]1D8 = ꢀ16.7 (c =
1.10, CH3OH): private communication from Prof. Y. Fukuyama
(Tokushima Bunri University).
[3] For reviews on neurotrophic activity, see: a) F. Hefti, J. Neuro-
biol. 1994, 25, 1418; b) F. Hefti, Annu. Rev. Pharmacol. Toxicol.
1997, 37, 239; c) M. V. Sofroniew, C. L. Howe, W. C. Mobley,
Annu. Rev. Neurosci. 2001, 24, 1217.
[4] I. Ohtani, T. Kusumi, Y. Kashman, H. Kakisawa, J. Am. Chem.
Soc. 1991, 113, 4092.
[5] M. Inoue, T. Sato, M. Hirama, J. Am. Chem. Soc. 2003, 125,
10772.
[6] a) V. B. Birman, S. J. Danishefsky, J. Am. Chem. Soc. 2002, 124,
2080; b) G. Mehta, S. R. Singh, Angew. Chem. 2006, 118, 967;
Angew. Chem. Int. Ed. 2006, 45, 953.
[7] For synthetic studies on merrilactone A, see: a) B.-C. Hong, Y.-J.
Shr, J.-L. Wu, A. K. Gupta, K.-J. Lin, Org. Lett. 2002, 4, 2249;
b) G. Mehta, S. R. Singh, Tetrahedron Lett. 2005, 46, 2079; c) J.
Iriondo-Alberdi, J. E. Perea-Buceta, M. F. Greaney, Org. Lett.
2005, 7, 3969; d) K. Harada, H. Kato, Y. Fukuyama, Tetrahedron
Lett. 2005, 46, 7407.
[8] a) Z. Meng, S. J. Danishefsky, Angew. Chem. 2005, 117, 1535;
Angew. Chem. Int. Ed. 2005, 44, 1511; b) H. Yun, Z. Meng, S. J.
Danishefsky, Heterocycles 2005, 66, 711.
[9] For reviews on the construction of eight-membered rings, see:
a) N. A. Petasis, M. A. Patane, Tetrahedron 1992, 48, 5757; b) G.
Mehta, V. Singh, Chem. Rev. 1999, 99, 881; c) M. E. Maier,
Angew. Chem. 2000, 112, 2153; Angew. Chem. Int. Ed. 2000, 39,
2073; d) L. Yet, Chem. Rev. 2000, 100, 2963.
[10] For selected examples of the synthesis of bicyclo[3.3.0]octane
systems from eight-membered rings, see: a) J. T. Negri, T.
Morwick, J. Doyon, P. D. Wilson, E. R. Hickey, L. A. Paquette,
[24] E. P. Balskus, J. MØndez-Andino, R. M. Arbit, L. A. Paquette, J.
Org. Chem. 2001, 66, 6695.
Angew. Chem. Int. Ed. 2006, 45, 4843 –4848
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4847

Communications
[25] The structure of 8 was determined by NMR spectroscopy on the
basis of HMBC and NOE data, and was later confirmed through
the X-ray crystallographic analysis of 29 (see the Supporting
information).[41]
[26] G. Bott, L. D. Field, S. Sternhell, J. Am. Chem. Soc. 1980, 102,
5618.
[27] F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M.
Lipton, C. Caufield, G. Chang, T. Hendrickson, W. C. Still, J.
Comput. Chem. 1990, 11, 440.
[28] When the temperature was lowered to ꢀ1008C in [ D]THF, the
8
1H NMR signals separated into two sets of peaks corresponding
to two interconverting conformers (2.5:1). The structures of the
conformers could not be determined because of signal overlap.
[29] The formation of a highly strained trans-fused 5,5 ring system
from conformer 7 is unlikely; see: a) S. Chang, D. McNally, S.
Shary-Tehrany, S. M. J. Hickey, R. H. Boyd, J. Am. Chem. Soc.
1970, 92, 3109; b) N. L. Allinger, M. T. Tribble, M. A. Miller,
D. H. Wertz, J. Am. Chem. Soc. 1971, 93, 1637.
[30] M. Frigerio, M. Santagostino, S. Sputore, G. Palmisano, J. Org.
Chem. 1995, 60, 7272.
[31] a) Y. Ueno, K. Chino, M. Watanabe, O. Moriya, M. Okawara, J.
Am. Chem. Soc. 1982, 104, 5564; b) G. Stork, R. Mook, Jr., S. A.
Biller, S. D. Rychnovsky, J. Am. Chem. Soc. 1983, 105, 3741;
c) W. Zhang, Tetrahedron 2001, 57, 7237; d) X. J. Salom-Roig, F.
DØn›s, P. Renaud, Synthesis 2004, 1903; e) G. S. C. Srikanth,
S. L. Castle, Tetrahedron 2005, 61, 10377.
[32] S. Danishefsky, T. Kitahara, R. McKee, P. F. Schuda, J. Am.
Chem. Soc. 1976, 98, 6715.
[33] P. A. Grieco, T. Oguri, Y. Yokoyama, Tetrahedron Lett. 1978,
419.
[34] G. T. Crisp, W. J. Scott, Synthesis 1985, 335.
[35] D. L. Comins, A. Dehghani, Tetrahedron Lett. 1992, 33, 6299.
[36] S. Cacchi, E. Morera, G. Ortar, Tetrahedron Lett. 1984, 25, 4821.
[37] J. W. Huffman, J. T. Charles, J. Am. Chem. Soc. 1968, 90, 6486.
[38] a) M. Fetizon, M. Golfier, C. R. Hebd. Seances Acad. Sci. 1968,
267, 900; b) A. McKillop, D. W. Young, Synthesis 1979, 401.
[39] a) R. W. Murray, Chem. Rev. 1989, 89, 1187; b) W. Adam, R.
Curci, J. O. Edwards, Acc. Chem. Res. 1989, 22, 205.
[40] The absolute structure of (ꢀ)-1 was further supported by
1H NMR spectroscopic data of MTPA ester derivatives of
synthetic and natural (ꢀ)-1 (see the Supporting Information).
[41] CCDC-603839 (23) and CDCC-603840 (29) contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
4848 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4843 –4848