A concise diels-alder strategy for the asymmetric synthesis of (+)-albicanol, (+)-albicanyl acetate, (+)-dihydrodrimenin, and (-)-dihydroisodrimeninol

Article abstract of DOI:10.1021/ol901372m

A short, mild, highly diastereo-, regio-, and stereoselective Diels-Alder strategy has been developed for the asymmetric synthesis of (+)albicanol, (+)-albicanyl acetate, (+)-dihydrodrimenin, and (-)-dihydroisodrimeninol.

Full text of DOI:10.1021/ol901372m

ORGANIC
LETTERS
2009
Vol. 11, No. 15
3178-3181
A Concise Diels-Alder Strategy for the
Asymmetric Synthesis of (+)-Albicanol,
(+)-Albicanyl Acetate,
(+)-Dihydrodrimenin, and
(-)-Dihydroisodrimeninol
Jeff R. Henderson, Masood Parvez, and Brian A. Keay*
Department of Chemistry, UniVersity of Calgary, Calgary, AB, Canada, T2N 1N4
keay@ucalgary.ca
Received April 11, 2009
ABSTRACT
A short, mild, highly diastereo-, regio-, and stereoselective Diels-Alder strategy has been developed for the asymmetric synthesis of (+)-
albicanol, (+)-albicanyl acetate, (+)-dihydrodrimenin, and (-)-dihydroisodrimeninol.
The drimane class of natural products is ubiquitous in nature
(Scheme 1)1 with many of them exhibiting a wide range of
biological activity.2 For example, (+)-albicanol 13a,b and (+)-
albicanyl acetate 23b have exhibited potent piscicidal effects,
while (+)-dihydrodrimenin 33c shows antimicrobial activity
against Photobacterium leiognathi and moderate inhibition
of Na+/K+-ATPase.3d While 3 and 4 have not been synthe-
sized to date, synthetic strategies toward albicanol 1 and
albicanyl acetate 2 have been reported. Initial syntheses of
1 and 2 were racemic,4,5 while more recent endeavors have
involved asymmetric approaches involving (a) kinetic resolu-
tions of suitable synthetic precursors with enzymes6 or
N-Boc-L-proline,7 (b) starting the synthesis with ready
Scheme 1
(1) (a) Jansen, B. J. M.; de Groot, A. J. Nat. Prod. 1991, 8, 319. (b)
Jansen, B. J. M.; de Groot, A. J. Nat. Prod. 1991, 8, 309.
(2) Jansen, B. J. M.; de Groot, A. Nat. Prod. Rep. 2004, 21, 449.
(3) (a) Ohta, Y.; Andersen, N. H.; Liu, C.-B. Tetrahedron 1977, 33,
617. (b) Hellou, J.; Andersen, R. J. Tetrahedron 1982, 38, 1875. (c) Paul,
V. J.; seo, Y.; Cho, K. W.; Rho, J.-R.; Shin, J.; Bergquist, P. R. J. Nat.
Prod. 1997, 60, 1115. (d) To our knowledge, (-)-dihydro-isodrimeninol 4
has not been tested for biological activity: Montagnac, A.; Martin, M.-T.;
Debitus, C.; Pais, M. J. Nat. Prod. 1996, 59, 866.
available natural products (+)-sclareolide8 and (+)-manool,9
and (c) using enantiopure Wieland-Miescher ketone as the
starting material.10 A simple, short, asymmetric route to the
drimane skeleton is still needed.
(4) Armstrong, R. J.; Harris, F. L.; Weiler, L. Can. J. Chem. 1986, 64,
1002
.
(6) (a) Akita, H.; Nozawa, M.; Mitsuda, A.; Ohsawa, H. Tetrahedron:
Asymmetry 2000, 11, 1375. (b) Anilkumar, A. T.; Sudhir, U.; Joly, S.; Nair,
M. S. Tetrahedron 2000, 56, 1899.
(5) Banerjee, A. K.; Correa, J. A.; Laya-Mimo, M. J. Chem. Res. (S)
1998, 710
.
10.1021/ol901372m CCC: $40.75
Published on Web 07/13/2009
 2009 American Chemical Society

The nonasymmetric Diels-Alder (DA) reaction ap-
proaches to 1 using diene 5 (Scheme 1) with various
dienophiles involved in the use of harsh conditions. For
example, high pressures and/or strong Lewis acids,11 high
temperatures,12 or dienophiles containing two electron-
withdrawing groups have been found necessary to effect a
successful DA reaction.11,12 Mayelvaganan et al. reported
the use of AlCl3 in nonasymmetric Lewis acid catalyzed DA
reactions of diene 5 with cis-substituted dienophiles; how-
ever, no other milder aluminum Lewis acids were mentioned
in the paper.13
further as an intermediate to other natural products within
the drimane class.
Given that chiral 1,3-oxazolidin-2-ones act as excellent
auxiliaries in asymmetric DA reactions,19we began our
investigation with the trans-substituted dienophile 9 that
contained (S-3a-cis)-(-)-3,3a,8,8a-tetrahydro-2H-indeno-
[1,2-d]oxazolidin-2-one as the chiral entity. Treatment of a
mixture of 520 and 9 with 1.4 equiv of MeAlCl2 at -25 °C
for 8 h provided a 2:1 mixture of two compounds (Table 1,
Asymmetric variants of the DA reaction involving diene
5 have been limited to the use of high pressures (12-15
kbar) with either quinones containing substrate-bound chiral
auxiliaries14 or chiral Lewis acids.15
Table 1. Initial Diels-Alder Results with Trans-Substituted
Dienophile 6
More recently, the use of allenic dienophiles with 6
resulted in [2 + 2] cycloadditions with the silyl enol ether
in 6 after heating a neat mixture of diene and dienophile at
110 °C for 14 days.16 Subsequent thermolysis of the formed
cyclobutene gave products expected from a [4 + 2] cycload-
dition. While these DA reactions are noteworthy, there is
still a need for an asymmetric variant of this reaction that
proceeds under mild reaction conditions and with high regio-,
stereo-, and diastereoselective control. We herein report a
mild yet highly selective DA reaction of this type that
provides a convenient, high-yielding route to drimanes 1-4.
entry
temp (°C)
conversiona
exo:endob
2:1
1
2
3
-25
0
23
10%
55%
75%
10:1
15:1 [99:1]c
a % conversion determined by 1H NMR. b Determined by 400 MHz 1H
NMR. c Recrystallized from hexanes to give >99:1. R ) (S-3a-cis)-(-)-
3,3a,8,8a-tetrahydro-2H-indeno[1,2-d]-oxazolidin-2-one.
17
Our previous work with DA reactions catalyzed by BCl3
and milder Lewis acids like MeAlCl218 led us to investigate
the synthetic strategy outlined in Scheme 2 toward the
entry 1) in which the major compound 10 crystrallized from
hexanes. The X-ray crystal structure of 1021 (Figure 1)
Scheme 2
drimane family of natural products. If the regio-, endo/exo-,
and facial selectivity were controlled during the DA reaction
with a trans-disubstituted dienophile like 7, then this strategy
could provide a facile route to (+)-dihydrodrimenin (3) via
8. Subsequently, (+)-dihydrodrimenin (3) could be used
(7) Toshima, H.; Oikawa, H.; Toyomasu, T.; Sassa, T. Biosci. Biotechnol.
Biochem. 2001, 65, 1244.
(8) Poigny, S.; Huor, T.; Guyot, M.; Samadi, M. J. Org. Chem. 1999,
64, 9318.
(9) Nakano, T.; Villamizar, J.; Maillo, M. A. J. Chem. Res. (S) 1995,
330.
(10) (a) Shishido, K.; Tokunaga, Y.; Omachi, N.; Hiroya, K.; Fukumoto,
K.; Kametani, T. Chem. Commun. 1989, 1093. (b) Shishido, K.; Tokunaga,
Y.; Omachi, N.; Hiroya, K.; Fukumoto, K.; Kametani, T. J. Chem. Soc.,
Perkin Trans.1 1990, 2481.
Figure 1. X-ray crystal structure of 10.
(11) (a) Engler, T. A.; Sampath, U.; Velde, D. V.; Takusagawa, F.
Tetrahedron 1992, 48, 9399. (b) Engler, T. A.; Naganathan, S. Tetrahedron
Lett. 1986, 27, 1015.
(12) Howell, S. C.; Ley, S. V.; Mahon, M.; Wothington, P. A. Chem.
Commun. 1981, 507.
indicated the major product from the DA reaction with 9
was the exo isomer. It had the regiochemistry in which the
oxazolidinone ring was adjacent to the ring-fused methyl
(13) For examples using olefin dienophiles with AlCl3, see: Mayelva-
ganan, T.; Hadimani, S. B.; Bhat, S. V. Tetrahedron 1997, 53, 2185.
Org. Lett., Vol. 11, No. 15, 2009
3179

group, but the absolute stereochemistry of the decalin ring
was opposite to that required for the synthesis of the drimane
skeleton.
at 0 °C after 12 h (entry 5) that improved to a 12:1 ratio at
23 °C after only 5 h.
While these results were encouraging, the exo:endo ratio was
still lower than that with 9 (Table 1, entry 3). Changing the Bn
group on the oxazolidinone to a phenyl group while using the
-OBn-substituted crotonate dienophile 7e provided a 3:1 ratio
of isomers at 0 °C that did not change upon warming to 23 °C
(entries 7 and 8). Finally, dienophile 7f with an iPr-substituted
oxazolidinone gave the best exo:endo ratio of 20:1 (entry 10)
when the reaction was stirred at 23 °C after only 5 h. This ratio
improved from a 12:1 ratio when the reaction mixture was
stirred at 0 °C for 12 h (entry 9).
A few points are noteworthy from the results presented in
Tables 1 and 2. First, changing the substituents on both the
oxazolidinone and allylic carbon atom within the dienophiles
(7a-e and 9) affects the exo:endo ratio, with the exo isomer
predominating in most examples when the reaction is performed
at higher temperatures. Second, the best exo:endo ratios were
obtained with the OBn-substituted dienophiles 7d-f and 9. To
understand better the observed changes in the exo:endo ratio
and why the OBn-substituted dienophiles gave the best product
ratios, some time studies were performed, and an additional
dienophile was studied.
This single result indicated that the DA reaction between
trans-disubstituted dienophile 9 and 5 was possible under
very mild conditions (mild LA at -25 °C), which is in stark
contrast to the severe reaction conditions previously reported
in DA reactions with 5 (Scheme 1). In addition, the DA
reaction proceeded with very high regio- and facial selectiv-
ity.22 To optimize the exo:endo ratio and increase the %
conversion, the DA reaction was repeated at different
temperatures. We were extremely pleased to find that
performing the reaction at 0 °C for 8 h provided a 10:1 exo:
endo ratio and that this ratio was increased to 15:1 when
the reaction was performed at 23 °C over the same length
of time. One recrystallization of the exo:endo mixtures from
hexanes gave 10 as a pure compound.
To probe the scope and limitations of this reaction and to
attempt to improve the exo:endo ratio further, a variety of
dienophiles and 1,3-oxazolidin-2-ones were used in the DA
reaction with 5 (Table 2). The first set of DA reactions
The DA reaction appears to be under thermodynamic control
under the reaction conditions since more of the exo isomer is
formed at higher temperatures. To provide further evidence for
this, the reaction between 5 and 9 was stopped after 2 h at -25
°C to give a 1:2 ratio in favor of the endo isomer 11. The exo:
endo mixture was purified from the remaining unreacted starting
materials 5 and 9, dissolved in DCM, and retreated with 1.4
equiv of Me2AlCl. After 2 h at 23 °C, the ratio changed to
1.5:1 in favor of the exo isomer 10 which further equilibrated
to give essentially the exo isomer after 18 h25 in addition to a
few unidentified compounds. This experiment, in combination
Table 2. Diels-Alder Results of Diene 5 with Various
Dienophiles and 1,3-Oxazolidin-2-ones
compd R1
R2
temp (°C) time (h) % yielda exo:endob
1
2
3
4
5
6
7
8
9
10
7a
7a
7b
7c
7d
7d
7e
7e
7f
Bn
Bn
Bn Br
Bn OTBS
Bn OBn
Bn OBn
Ph OBn
Ph OBn
iPr OBn
iPr OBn
H
H
23
23
23
23
0
23
0
23
0
5
24
5
5
12
5
12
5
12
87
75
40
nr
62
65
52
52
60
60
1:1
1.5:1
1:1
(14) Carreno, M. C.; Ruano, J. L. G.; Toledo, M. A. Chem.sEur. J.
2000, 6, 288.
na
(15) Knol, J.; Meetsma, A.; Feringa, B. L. Tetrahedron: Asymmetry
1995, 6, 1069.
10:1
12:1
3:1
3:1
12:1
20:1
(16) (a) Jung, M. E.; Murakami, M. Org. Lett. 2006, 8, 5857. (b) Jung,
M. E.; Murakami, M. Org. Lett. 2007, 9, 461. (c) Jung, M. E.; Ho, D. G.
Org. Lett. 2007, 9, 375.
(17) (a) Henderson, J.; Parvez, M.; Keay, B. A. Org. Lett. 2007, 9, 5167.
(b) Lait, S. M.; Parvez, M.; Keay, B. A. Tetrahedron: Asymmetry 2003,
14, 749. (c) Lait, S. M.; Rankic, D. A.; Keay, B. A. Chem. ReV. 2007, 107,
767.
7f
23
5
a Isolated yields. b Determined by 400 MHz NMR.
(18) Hunt, I. R.; Rogers, C.; Woo, S.; Rauk, A.; Keay, B. A. J. Am.
Chem. Soc. 1995, 117, 1049, and references therein.
(19) (a) Evans, D. A.; Chapman, K. T.; Bisaha, J. J. Am. Chem. Soc.
1988, 110, 1238.
(20) Diene 5 was easily prepared from ꢀ-ionone by oxzonolysis followed
by a Peterson olefination. For details, see: (a) Crombie, B. S.; Smith, C.;
Varnavas, C. Z.; Wallace, T. W. J. Chem. Soc., Perkin Trans. 1 2001, 206.
(b) Howell, S. C.; Ley, S. V.; Mahon, M.; Worthington, P. A. Chem.
Commun. 1981, 507.
involved using several Bn-substituted 1,3-oxazolidin-2-ones
(7a-d). Treatment of a mixture of 5 and crotonate 7a23 with
1.4 equiv of MeAlCl2 at 23 °C for 5 h provided a 1:1 exo-
12 to endo-13 mixture (Table 2, entry 1) that could be slightly
improved to a 1.5:1 ratio after stirring the reaction mixture
for 24 h (entry 2). Allylic bromination of 7a afforded 7b24
that when subjected to the same DA reaction conditions with
5 provided a complex mixture of products in which the major
constituents were a 1:1 mixture of exo and endo adducts
albeit in lower yield due to decomposition of the DA products
(entry 3). While the -OTBS-substituted dienophile 7c did
not react with 5 (entry 4), -OBn-substituted olefin 7d
provided a 10:1 mixture of exo:endo adducts in 62% yield
(21) X-ray crystal data for 10: monoclinic P21; a ) 9.847(3) Å, b )
25.533(8) Å, c ) 10.795(2) Å, R ) 90°, ꢀ ) 91.114(18)°, γ ) 90°, V )
2713.6(13) Å3; Z ) 4; R ) 0.0474; Rw ) 0.0977.
(22) Roush et al. have reported successful DA reactions using MeAlCl2
with acyclic (Z)-1,3-dienes using N-acryloyl sultam as a chiral auxiliary.
(a) Roush, W. R.; Limberakis, C.; Kunz, R. K.; Barda, D. A. Org. Lett.
2002, 4, 1543. Interestingly, a similar reaction with an achiral oxazolidinone
did not provide any DA products, see: (b) Roush, W. R.; Barda, D. A.
J. Am. Chem. Soc. 1997, 119, 7402.
(23) The DA reaction of 5 and 7a did not provide any products when
1.4 equiv of Me2AlCl, EtAlCl2, Et2AlCl, or Sc(OTf)3 was used (DCM, 5 h,
rt).
(24) Martinelli, M. J. J. Org. Chem. 1990, 55, 5065.
3180
Org. Lett., Vol. 11, No. 15, 2009

with the results in Tables 1 and 2, conclusively demonstrates
that the DA reaction is under thermodynamic control and that
the exo:endo ratio is governed by both the substituent on the
oxazolidinone and on the dienophile. Changes in either can alter
the exo:endo ratio from 1:1 (Table 1, entry 1, crotonate with
Bn-substituted 1,3-oxazolidin-2-one) to up to >99:1 (9 at 23
°C for 18 h).
The best exo:endo results were provided with the OBn-
substituted crotonate dienophiles 7d-f and 9. To determine if
the benzene ring within this substitutent was in any way
specifically responsible for the higher exo:endo product ratios,
we prepared 14 that contained a cyclohexyl ring in place of
the phenyl ring in 7d but still retained a Bn -substituted 1,3-
oxazolidin-2-one. Treatment of 5 and 14 with 1.4 equiv of
Me2AlCl at 23 °C for 1 h provided a 2:1 exo:endo ratio that
did not change over a 24 h period (Table 3, all entries). It is
18.26 Hydroboration-oxidation of ent-10 afforded trans-fused
decalin 19 that was subjected to a Barton deoxygenation
sequence to provide (+)-dihydrodrimenin (3). Reduction of
(+)-3 with DIBAL-H gave a 10:1 ratio of (-)-dihydroisodri-
meninol (4) and its epimer 4a.
(+)-Dihydrodrimenin (3) was easily converted into (+)-
albicanol (1) and (+)-albicanol acetate (2). Treatment of (+)-3
with thionyl chloride in methanol at 60 °C for 24 h provided
2027 that was easily converted into 21 by treatment with
PhSeSePh and NaBH4 followed by a 30% H2O2 workup.
Reduction of the ester in 21 with LAH gave (+)-1. Finally,
acetylation of (+)-1 gave (+)-albicanol acetate (2).
Scheme 3
Table 3. Diels-Alder Results of Diene 5 with 14
time (h)
% conversiona
15 exo:endob
1
2
3
4
a
1
3
6
82
91
94
100
2:1
2:1
2:1
2:1
24
1
Determined by 400 MHz H NMR.
noteworthy that the DA reaction was not finished after one hour,
as the % conversion gradually increased from 82 to 100% upon
allowing the reaction to stir for 24 h. These results indicate that
the DA reaction was already at its thermodynamic equilibrium
after 1 h but that the reaction was still progressing to completion.
In contrast, OBn-substituted dienophile 7d provided a 12:1 exo:
endo ratio after 5 h at 23 °C. Unfortunately, we do not have a
rationale at this time for why dienophiles with an OBn
substituent provide the best exo:endo ratios but were very
satisfied with the knowledge that the DA is under thermody-
namic control providing the desired isomer in high yield for
the synthesis of some drimane natural products as demonstrated
below.
We have shown that the D-A reaction with 5 and trans-
substituted olefins 7a-g and 9 proceeds with excellent regio-,
stereo-, and facial selectivity in the presence of the mild Lewis
acid MeAlCl2. The DA adduct 10 was easily converted into
four drimane natural products. Work is continuing to apply this
method to the synthesis of more elaborate drimanes and labdane
natural products.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council and the University of Calgary
for financial support.
Debenzylation of ent-10 gave 16 followed by spontaneous
lactone formation/chiral auxiliary cleavage which afforded
tricycle 17. Reduction of the olefin in 17 required Adam’s
catalyst and AcOH, which unfortunately provided cis-decalin
Supporting Information Available: Crystallographic data
of 10, 18, 19, and 3 and experimental procedures for all
reactions. This material is available free of charge via the
Internet at http://pubs.acs.org.
1
(25) The exo isomer was the major product in the H NMR spectrum
with a few very minor unidentified compounds. After 18 h, the endo isomer
could not be detected in the NMR spectrum. See the Supporting Information
for more details.
OL901372M
(26) The structure of 18 was confirmed by X-ray crystal structure
analysis. See the Supporting Information section for more details.
(27) This reaction gave a 5:1 mixture of isomers that were epimeric at
the ester in 20.
Org. Lett., Vol. 11, No. 15, 2009
3181