Amino acid mediated intramolecular asymmetric aldol reaction to construct a new chiral bicyclic enedione containing a seven-membered ring: Remarkable inversion of enantioselectivity compared to the six-membered ring example

Article abstract of DOI:10.1021/jo061824n

A detailed study to assess the enantioselectivity of the amino acid mediated intramolecular asymmetric aldol reaction of 1,3-cycloheptanedione bearing a C-2 methyl substituent has been undertaken. The cyclizations were mediated by a series of L-amino acid

Full text of DOI:10.1021/jo061824n

Amino Acid Mediated Intramolecular Asymmetric Aldol Reaction to
Construct a New Chiral Bicyclic Enedione Containing a
Seven-Membered Ring: Remarkable Inversion of Enantioselectivity
Compared to the Six-Membered Ring Example
Takashi Nagamine,† Kohei Inomata,*,† Yasuyuki Endo,† and Leo A. Paquette*,‡
Tohoku Pharmaceutical UniVersity, 4-4-1 Komatsushima Aoba-ku, Sendai 981-8558 Japan, and EVans
Chemical Laboratories, The Ohio State UniVersity, Columbus, Ohio 43210
inomata@tohoku-pharm.ac.jp
ReceiVed September 4, 2006
A detailed study to assess the enantioselectivity of the amino acid mediated intramolecular asymmetric
aldol reaction of 1,3-cycloheptanedione bearing a C-2 methyl substituent has been undertaken. The
cyclizations were mediated by a series of L-amino acids in the presence of an acid cocatalyst. Strikingly,
the process is characterized by an inversion of enantioselectivity when compared to a similar reaction
involving the 1,3-cyclohexanedione counterpart.
Introduction
In the early to mid 1970s, Hajos and Wiechert independently
developed an asymmetric aldol reaction of trione 1 using a
catalytic amount of L-proline (L-Pro, 2) to provide the bicyclic
enone 3 (Figure 1).1-3 This process has become known as the
Hajos-Parrish-Eder-Sauer-Wiechert reaction. In the inter-
vening years, the bicyclic enone products 3 have often been
used as choice chiral synthons for realizing the total syntheses
FIGURE 1. Hajos-Parrish-Eder-Sauer-Wiechert reaction.
† Tohoku Pharmaceutical University.
of numerous natural products, especially in the area of steroids
and terpenoids.4,5 The reaction catalyzed by phenylalanine (Phe)
to afford 3a has also been reported and shown to be ac-
‡ The Ohio State University.
(1) (a) Hajos, Z. G.; Parrish, D. R. German Patent DE 2102623, 1971.
(b) Eder, U.; Sauer, G.; Wiechert, R. German Patent DE 2014757, 1971.
(c) Eder, U.; Sauer, G.; Wiechert, R. Angew. Chem. 1971, 83, 492. (d)
Hajos, Z. G.; Parrish, D. R. J. Org. Chem. 1973, 38, 3239. (e) Hajos, Z.
G.; Parrish, D. R. J. Org. Chem. 1974, 39, 1615. (f) Micheli, R. A.; Hajos,
Z. G.; Cohen, N.; Parrish, D. R.; Portland, L. A.; Sciamanna, W.; Scott,
M. A.; Wehri, P. A. J. Org. Chem. 1975, 40, 675. (g) Hajos, Z. G.; Parrish,
D. R. Organic Synthesis; Wiley: New York, 1990; Collect. Vol. VII, p
363. (h) Harada, N.; Sugioka, T.; Uda, H.; Kuriki, T. Synthesis 1990, 53.
(i) Rajagopal, D.; Rajagopalan, K.; Swaminathan, S. Tetrahedron: Asym-
metry 1996, 7, 2189. (j) Bui, T.; Barbas, C. F., III. Tetrahedron Lett. 2000,
41, 6951. (k) Rajagopal, D.; Narayanan, R.; Swaminathan, S. Tetrahedron
Lett. 2001, 42, 4887.
(2) (a) Zhong, G.; Hoffmann, T.; Lerner, R. A.; Danishefsky, S.; Barbas,
C. F., III. J. Am. Chem. Soc. 1997, 119, 8131. (b) Fuhshuku, K.; Funa, N.;
Akeboshi, T.; Ohta, H.; Hosomi, H.; Ohba, S.; Sugai, T. J. Org. Chem.
2000, 65, 129. (c) Hioki, H.; Hashimoto, T.; Kodama, M. Tetrahedron:
Asymmetry 2000, 11, 829. (d) Gassama, A.; d’Angelo, J.; Cave´, C.;
Mahuteau, J.; Riche, C. Eur. J. Org. Chem. 2000, 3165. (e) Shimizu, N.;
Akita, H.; Kawamata, T. Tetrahedron: Asymmetry 2002, 13, 2123.
(3) (a) Wieland, P.; Miescher, K. HelV. Chim. Acta 1950, 33, 2215. (b)
Ramachandran, S.; Newman, M. S. Organic Syntheses; Wiley: New York,
1973; Collect. Vol. V, p 456. (c) Gutzwiller, J.; Buchschacher, P.; Fu¨rst,
A. Synthesis 1977, 167. (d) Uma, R.; Swaminathan, S.; Rajagopolan, K.
Tetrahedron Lett. 1984, 25, 5825. (e) Buchschacher, P.; Fu¨rst, A.;
Gutzwiller, J. Organic Syntheses; Wiley: New York, 1990; Collect. Vol.
VII, p 368.
(4) Recent applications of 3a: (a) Renneberg, D.; Pfander, H.; Leumann,
C. J. J. Org. Chem. 2000, 65, 9069. (b) Molander, G. A.; Quirmbach, M.
S.; Silva, L. F., Jr.; Spencer, K. C.; Balsells, J. Org. Lett. 2001, 3, 2257.
(c) Sevillano, L. G.; Melero, C. P.; Caballero, E.; Tome´, F.; Lelie`vre, L.
G.; Geering, K.; Crambert, G.; Carro´n, R.; Medarde, M.; Feliciano, A. S.
J. Med. Chem. 2002, 45, 127. (d) Yang, Z.; Shannon, D.; Truong, V.-L.;
Deslongchamps, P. Org. Lett. 2002, 4, 4693. (e) Lepage, O.; Deslongchamps,
P. J. Org. Chem. 2003, 68, 2183. (f) Lugar, C. W., III; Magee, D.; Adrian,
M. D.; Shetler, P.; Bryant, H. U.; Dodge, J. A. Bioorg. Med. Chem. Lett.
2003, 13, 4281.
10.1021/jo061824n CCC: $37.00 © 2007 American Chemical Society
Published on Web 12/05/2006
J. Org. Chem. 2007, 72, 123-131
123

Nagamine et al.
companied by a lower ee.1c,6 The absolute configurations of the
products were determined to be S when L-amino acids such as
L-Pro or L-Phe were used to mediate the reaction. Quite recently,
Davies disclosed that similar aldol reactions of 1 involving cis-
pentacin, a cyclic â-amino acid, exhibited enantioselectivity
opposite to those promoted by L-amino acids and afforded (R)-3
in high ee.7 This amino acid catalyzed asymmetric transforma-
tion has been widely recognized to consist of an enamine-based
aldol reaction.8,9 Successful extensions have included intermo-
lecular asymmetric reactions such as cross-aldol couplings,
Mannich condensations, and direct R-aminations.10
FIGURE 2. L-Phe-mediated intramolecular aldol reaction of 4.
The Hajos-Parrish-Eder-Sauer-Wiechert reaction has
been extended to prepare bicyclic ketones 511,12 bearing a variety
of substituents (R) such as methyl, ethyl, butenyl, and oxygen-
ated alkyl and by the use of catalytic or stoichiometric amounts
of L-amino acids, especially L-Phe (Figure 2).13-16 Notwith-
standing the variety of substituents, those reactions mediated
by L-amino acids afforded the products 5 whose absolute
configurations at the quaternary carbon were invariably S.
There has been little development of this reaction for the
FIGURE 3. Example of an intramolecular aldol reaction to construct
a 6-7 fused bicycle.
purpose of constructing new ring systems encompassing a seven-
membered or larger carbocycle except for two studies reported
independently by Swaminathan and our group.17,18 Swaminathan
et al. investigated attempts to bring about the L-Pro- or L-Phe-
mediated asymmetric aldol reaction of trione 6 involving a
seven-membered carbocycle, but no optically active product 7
was isolated under diverse reaction conditions. They also
observed that the reaction of trione 6 involving pyrrolidine in
acetic acid (AcOH) afforded racemic 719 in 60-65% yield
(Figure 3).17b This result demonstrates that the possibility exists
for realizing enamine-based asymmetric aldol reactions of 6
when mediated by an amino acid to afford chiral 7.
On the other hand, we reported that the amino acid mediated
diastereomeric intramolecular aldol reaction of trione 8 bearing
a chiral acetonide moiety afforded cyclized products 10 and
11.18 During these studies, we observed that the diastereose-
lectivity of this reaction did not depend on the stereostructure
of the D- or L-amino acid because the same stereoselectivity
was observed irrespective of the absolute configuration of the
catalyst with the exception of Pro derivative 9.20 Although the
reaction proceeded smoothly in the presence of various cyclic
or acyclic amino acids, enone 11 was obtained as the major
product in most cases, and diastereoselectivities observed for
10 and 11 gravitated around 40:60. Clearly, the chirality in the
great majority of the amino acids was not reflected in the
stereostructure of the products. Because we needed to synthesize
10 to develop an ongoing natural product synthesis, these results
were inadequate. The best means for us to prepare the desired
10 was secured in the presence of a stoichiometric amount of
9 with only 16% diastereomeric excess (Figure 4). However,
we achieved the inverted diastereoselectivity when using the
enantiomer of 9 as a mediator.
(5) Recent applications of 3b: (a) Shah, N.; Scanlan, T. S. Bioorg. Med.
Chem. Lett. 2004, 14, 5199. (b) D´ıaz, S.; Gonza´lez, A.; Bradshaw, B.;
Cuesta, J.; Bonjoch, J. J. Org. Chem. 2005, 70, 3749.
(6) Buchschacher, P.; Cassal, J.-M.; Fu¨rst, A.; Meier, W. HelV. Chim.
Acta 1977, 60, 2747.
(7) Davies, S. G.; Sheppard, R. L.; Smith, A. D.; Thomson, J. E. Chem.
Commun. 2005, 3802.
(8) (a) Brown, K. L.; Damm, L.; Dunitz, J. D.; Eschenmoser, A.; Hobi,
R.; Kratky, C. HelV. Chim. Acta 1978, 61, 3108. (b) Agami, C.; Sevestre,
H. J. Chem. Soc., Chem. Commun. 1984, 1385. (c) Agami, C.; Puchot, C.;
Sevestre, H. Tetrahedron Lett. 1986, 27, 1501. (d) Agami, C.; Puchot, C.
J. Mol. Catal. 1986, 38, 341. (e) Agami, C.; Platzer, N.; Sevestre, K. Bull.
Chim. Soc. Fr. 1987, 358. (f) Agami, C.; Platzer, N. Bull. Chim. Soc. Fr.
1998, 499. (g) Gathaergood, N. Aust. J. Chem. 2002, 55, 615.
(9) (a) Bahmanyar, S.; Houk, K. N. J. Am. Chem. Soc. 2001, 123, 11273.
(b) Bahmanyar, S.; Houk, K. N. J. Am. Chem. Soc. 2001, 123, 12911. (c)
Hoang, L.; Bahmanyar, S.; Houk, K. N.; List, B. J. Am. Chem. Soc. 2003,
125, 16. (d) Bahmanyar, S.; Houk, K. N.; Martin, H. J.; List, B. J. Am.
Chem. Soc. 2003, 125, 2475. (e) List, B.; Hoang, L.; Martin, H. J. Proc.
Natl. Acad. Sci. U.S.A. 2004, 101, 5839. (f) List, B. Acc. Chem. Res. 2004,
37, 548. (g) Allemann, C.; Gordillo, R.; Clemente, F. R.; Cheong, P. H.-
Y.; Houk, K. N. Acc. Chem. Res. 2004, 37, 558. (h) Cheong, P. H-Y.; Houk,
K. N. Synthesis 2005, 1533.
(10) For reviews: (a) List, B. Tetrahedron 2002, 58, 5573. (b) Notz,
W.; Tanaka, F.; Barbas, C. F., III. Acc. Chem. Res. 2004, 37, 580. (c)
Berkessel, A.; Gro¨ger, H. Asymmetric Organocatalysis; Wiley-VCH:
Weinheim, 2005. (d) List, B. Chem. Commun. 2006, 819.
(11) Recent applications of 5a: (a) Hagiwara, H.; Sakai, H.; Uchiyama,
T.; Ito, Y.; Morita, N.; Hoshi, T.; Suzuki, T.; Ando, M. J. Chem. Soc.,
Perkin Trans. 1 2002, 583. (b) Arai, N.; Ui, H.; Omura, S.; Kuwajima, I.
Synlett 2005, 1691. (c) Brady, T. P.; Kim, S. H.; Wen, K.; Kim, C.;
Theodorakis, E. A. Chem.-Eur. J. 2005, 11, 7175.
(12) Recent applications of 5b: (a) Katoh, T.; Nakatani, M.; Shikita,
S.; Sampe, R.; Ishiwata, A.; Ohmori, O.; Nakamura, M.; Terashima, S.
Org. Lett. 2001, 3, 2701. (b) Stahl, P.; Kissau, L.; Mazitschek, R.; Huwe,
A.; Furet, P.; Giannis, A.; Waldmann, H. J. Am. Chem. Soc. 2001, 123,
11586. (c) Ling, T.; Poupon, E.; Rueden, E. J.; Theodorakis, E. A. Org.
Lett. 2002, 4, 819. (d) Ling, T.; Rivas, F.; Theodorakis, E. A. Tetrahedron
Lett. 2002, 43, 9019. (e) Ling, T.; Poupon, E.; Rueden, E. J.; Kim, S. H.;
Theodorakis, E. A. J. Am. Chem. Soc. 2002, 124, 12261. (f) Honda, T.;
Favaloro, F. G., Jr.; Janosik, T.; Honda, Y.; Suh, N.; Sporn, M. B.; Gribble,
G. W. Org. Biol. Chem. 2003, 1, 4384. (g) Favaloro, F. G., Jr.; Honda, T.;
Honda, Y.; Gribble, G. W.; Suh, N.; Risingsong, R.; Sporn, M. B. J. Med.
Chem. 2002, 45, 4801. (h) Iwasaki, K.; Nakatani, M.; Inoue, M.; Katoh, T.
Tetrahedron Lett. 2002, 43, 7937. (i) Ling, T.; Rivas, F.; Theodorakis, E.
A. Tetrahedron Lett. 2002, 43, 9019. (j) Guo, Z.; Vangapandu, S.; Nimrod,
A.; Walker, L. A.; Sindelar, R. D. Med. Chem. 2005, 1, 3. (k) Takikawa,
H.; Imamura, Y.; Sasaki, M. Tetrahedron 2006, 62, 39.
Thus, application of the Hajos-Parrish-Eder-Sauer-
Wiechert reaction for constructing 6-7 fused ring systems
(17) (a) Selvarajan, R.; John, J. P.; Narayanan, K. V.; Swaminathan, S.
Tetrahedron 1966, 22, 949. (b) Rajagopal, D.; Narayanan, R.; Swaminathan,
S. Proc. Indian Acad. Sci. 2001, 113, 197. (c) Malathi, R.; Rajagopal, D.;
Hajos, Z. G.; Swaminathan, S. J. Chem. Sci. 2004, 116, 159.
(18) Inomata, K.; Barrague´, M.; Paquette, L. A. J. Org. Chem. 2005,
70, 533.
(19) Ireland, R. E.; Aristoff, P. A.; Hoyng, C. F. J. Org. Chem. 1979,
44, 4318.
(20) Applications of 9 for asymmetric aldol reactions: (a) Itagaki, N.;
Kimura, M.; Sugahara, T.; Iwabuchi, Y. Org. Lett. 2005, 7, 4185. (b) Itagaki,
N.; Sugahara, T.; Iwabuchi, Y. Org. Lett. 2005, 7, 4181. (c) Hayashi, Y.;
Sumiya, T.; Takahashi, J.; Gotoh, H.; Urushima, T.; Shoji, M. Angew. Chem.
Int. Ed. 2006, 45, 958.
(13) Danishefsky, S.; Cain, P. J. Am. Chem. Soc. 1976, 98, 4975.
(14) Banerjee, D. K.; Kasturi, T. R.; Sarkar, A. Proc. Indian Acad. Sci.
1983, 92, 181.
(15) Corey, E. J.; Virgil, S. C. J. Am. Chem. Soc. 1990, 112, 6429.
(16) Shigehisa, H.; Mizutani, T.; Tosaki, S.; Ohshima, T.; Shibasaki,
M. Tetrahedron 2005, 61, 5057.
124 J. Org. Chem., Vol. 72, No. 1, 2007

New Enedione Containing a SeVen-Membered Ring
TABLE 1. Screening of Several L-Amino Acids for Asymmetric
Aldol Reaction of 6a
time, yield,b ee,c,d
entry amino acid
R
h
%
%
1
2
3
4
5
6
7
8
9
10
11
12
13
14e
15
16e
L-Pro (2)
L-Phe
L-Tyr
L-Trp
L-Val
L-Leu
L-Ser
L-homo-Ser
L-Cys
L-Met
-
114
100
95
102
113
144
95
95
120
96
120
120
137
24
18
146
7
86
24
14
44
34
37
56
23
99
47
18
40
66
80
32
6
48
6
20
41
32
9
32
1
53
11
2
CH2Ph
CH2Ph-p-OH
CH2-indol-3-yl
CH(CH3)2
CH2CH(CH3)2
CH2OH
(CH2)2OH
CH2SH
(CH2)2SCH3
FIGURE 4. Diastereomeric intramolecular aldol reaction of 8 to
construct 6-7 fused bicycles.
SCHEME 1a
S-Me-L-Cys CH2SCH3
L-Gln
L-Asn
L-Asp
L-Arg
(CH2)2CONH2
CH2CONH2
CH2CO2H
(CH2)3NHC(dNH)NH2
47
-0.5
-13
22
L-Lys (HCl)f (CH2)3NH2HCl
a Reaction conditions: L-amino acid (1.0 equiv), 1 N HClO4 (0.5 equiv),
DMSO, 90 °C. b Isolated yield. c Determined by HPLC with a chiral
stationary phase. d Opposite enantioselection is denoted by a minus sign.
e 1 N HClO4 was not added. f Monohydrogen chloride salt was used.
a Reaction conditions: (a)∼(c) See ref 21; (d) EVK (2.0 equiv), Triton
B (0.1 equiv), methanol, reflux, 83%.
eventuates differently from those involved in construction of
the 5-6 or 6-6 fused products in Figures 1 and 2. This fascinating
divergence prompted us to scope out the features associated
with construction of a 6-7 fused ring in a highly enantioselective
fashion. Presently, we report our findings associated with
intramolecular asymmetric aldol reactions of 6 to afford chiral
7 by the use of a series of amino acids. Also examined are the
reactions of a substrate of differing ring size to allow direct
comparisons of reactivity and enantioselectivity.
°C (Table 1). Enantiomeric excesses (ee’s) of the resulting enone
7 were determined by HPLC equipped with a chiral stationary
phase column (Chiralpak AS-H, Daicel Chemical Corporation,
Ltd.). The results are compiled in Table 1.
Relevantly, most of the reactions involving L-amino acids
afforded (R)-7 as the major enantiomer. Because the reaction
of 1 and 4 afforded (S)-3 and (S)-5, respectively, in the presence
of L-amino acids, inversion of enantioselectivity must be
operative in those reactions of 6 that involve a seven-membered
ring. The details relating to the determination of the absolute
configuration of (R)-7 are given below (vide infra). L-Pro (2)
was not particularly effective in this reaction. Although a trace
amount of 7 was isolated after 114 h, the ee value was only 6%
(entry 1). Among several amino acids, L-Phe and L-methionine
(L-Met) were particularly suitable, delivering (R)-7 in 86% yield
accompanied by 48% ee (entry 2) and in 99% yield accompanied
by 53% ee, respectively (entry 10). Amino acids bearing other
side chain aromatic substituents such as L-tyrosine (L-Tyr) and
L-tryptophan (L-Trp) afforded 7 in lower yield and ee than L-Phe
(entries 3 and 4). Both L-valine (L-Val) and L-leucine (L-Leu),
which have a branched alkyl side chain, afforded 7 in moderate
yield and ee (entries 5 and 6). In the case of L-serine (L-Ser)
and L-homo-serine (L-homo-Ser) where the heteroatom is more
remote, some beneficial effect on yield and ee was evident
(entries 7 and 8). The same tendency was observed in relation
to L-cysteine (L-Cys), L-Met, and S-methyl-L-cystein (S-Me-L-
Cys, entries 9-11). An amide group on the side chain of the
amino acid also gave rise to elevated levels of enantioselectivity.
L-Asparagine (L-Asn), which has a nitrogen or an oxygen
functionality at the position defined above as suitable, was more
effective than L-glutamine (L-Gln), which has a longer carbon
chain than L-Asn. However, the isolated yields of 7 in both cases
were low or modest (entries 12 and 13). Both acidic and basic
Results and Discussion
Dione 13 was prepared from diethyl adipate (12) according
to a known method.21 Following Michael addition of 13 to
1-penten-3-one (ethyl vinyl ketone, EVK) in the presence of a
catalytic amount of benzyltrimethylammonium hydroxide (Tri-
ton B) in methanol, the trione 617b was isolated in 83% yield
(Scheme 1).
In preliminary experiments, the aldol reaction of 6 in the
presence of L-Pro or L-Phe at room temperature (rt) did not
proceed to afford 7 in several solvents such as methanol,
dimethylsulfoxide (DMSO), acetonitrile (MeCN), and N,N-
dimethylformamide (DMF). When the same reaction mixtures
were heated, very long times were needed to achieve completion,
and complex mixtures of many products accompanied trace
amounts of desired 7. The reaction of 6 was greatly improved
by adding 0.5 equiv of 1 N HClO4 to stimulate the dehydration
that follows the aldol reaction. There ensued the screening of
several commercially available cyclic and acyclic L-amino acids
for the aldol reaction. All of the reactions were carried out under
the same conditions in the presence of a stoichiometric amount
of L-amino acid and 0.5 equiv of 1 N HClO4 in DMSO at 90
(21) Lewicka-Piekut, S.; Okamura, W. H. Synth. Commun. 1980, 10,
415.
J. Org. Chem, Vol. 72, No. 1, 2007 125

Nagamine et al.
SCHEME 2a
TABLE 2. Screening of Several Amino Acids for Asymmetric
Aldol Reaction of 14a
time,
h
yield,b
%
ee,c
%
entry
amino acid
1
2
3
4
5
6
7
8
L-Pro
L-Phe
L-Tyr
L-Trp
132
22
22
22
22
22
22
22
22
22
13
56
70
51
73
74
71
74
41
69
11
91
79
90
81
84
80
84
31
19
L-Val
L-Leu
L-homo-Ser
L-Met
L-Asp
L-His
a Reaction conditions: (a) TMSCN (1.2 equiv), KCN (0.15 equiv), 18-
crown-6 (0.04 equiv), CH2Cl2, 0 °C, 30 min, 92%; (b) LAH (4.0 equiv),
THF, reflux, 13 h, 57%; (c) NaNO2 (3.0 equiv), AcOH/H2O ) 1:1, 0 °C,
3 h; (d) MnO2 (15.0 equiv), CH2Cl2, rt, 96 h, 25% (two steps) for 18, 0.7%
(two steps) for (S)-7.
9d
10
a Reaction conditions: L-amino acid (1.0 equiv), 1 N HClO4 (0.5 equiv),
DMSO, 90 °C. b Isolated yield. c Determined by HPLC equipped with a
chiral stationary phase. d HClO4 was not added.
asymmetric aldol reactions involving 14 in the presence of
several amino acids are compiled. All of these reactions were
carried out in DMSO at 90 °C in the presence of 0.5 equiv of
1 N HClO4 except for entry 9. The absolute configuration of
15 was assigned as S by comparison with the optical rotation
of (S)-15 reported previously.16 More specifically, the optical
rotation of product (S)-15 displayed [R]D +130.0 (CHCl3, 86%
ee) (lit.16 [R]D + 120.7 (CHCl3, 80% ee)). The reaction using
L-Pro (2) afforded (S)-15 in low yield and low ee even after a
reaction time of 132 h. This finding compares favorably with
Pecher’s result.22b Most of the L-amino acids were effective in
this reaction and furnished (S)-15 in the range of 51-74% yield
and 80-91% ee (entries 2-8). The highest ee’s were observed
in the cases using L-Phe or L-Trp and were accompanied with
modest yields. L-Met, which provided good results in the
reaction of 6, also afforded (S)-15 with a little lower ee and
higher yield than the examples involving L-Phe and L-Trp (entry
8). Both acidic and basic amino acids such as L-Asp and
L-histidine (R ) CH2-imidazol-4-yl; L-His) were not effective
and afforded (S)-15 with quite low ee and moderate yield (entries
9 and 10). On the basis of these results, we came to the
conclusion that asymmetric aldol reactions of 14 differ from
those of 6 and that enantioselectivity is marginally, if at all,
dependent on the nature of the amino acid side chain. A variety
of L-amino acids could be used for highly enantioselective aldol
reactions of 14 to afford (S)-15.
amino acids were not suitable (entries 14-16). In the case of
L-aspartic acid (L-Asp), the reaction was carried out without
HClO4 because of the existing carboxylic acid functionality and
was completed in shorter time than the other amino acids, but
no enantioselectivity was observed (entry 14). L-Arginine (L-
Arg), with its existing basic guanidine functionality, revealed
an opposite enantioselectivity to afford (S)-7 as the major
enantiomer accompanied by low ee, high chemical yield, and
the shortest reaction time (entry 15). L-Lysine (L-Lys) mono-
hydrogen chloride salt was also examined, but it was not
effective (entry 16). Although the best ee value was moderate
in our trials, we could obtain optically pure (R)-7 by fractional
recrystallization. In the final analysis, L-Met and L-Phe were
selected as mediators to develop further experiments generating
(R)-7 via the asymmetric intramolecular aldol reaction.
To compare the reactivity offered by a differing ring size,
we also studied the aldol reaction of 14 to provide 15 bearing
a methyl substituent and a 6-6 fused bicyclic structure. Although
many reports utilizing (S)- or (R)-15 for syntheses of natural or
unnatural products have been documented,11,12 enantioselective
syntheses of 15 via amino acid mediated aldol reactions of 14
have consisted of four methods as reported independently by
Yamada, Pecher, Uda, and Swaminathan.22 Pecher described a
route to (S)-15 via the L-Pro-catalyzed aldol reaction of 14 and
obtained (S)-15 in 44-60% yield with 24-41% ee after a
reaction period of 20-34 days.22b This finding suggests that
L-Pro is not a suitable promoter of the reaction. Recently,
Shibasaki reported a catalytic asymmetric synthesis of (S)-15
involving the use of L-Phe and p-toluenesulfonic acid pyridinium
salt (PPTS) in 1.0 equiv of DMSO as solvent.16 Here, (S)-15
was always obtained as the major enantiomer when L-Pro, L-Phe,
or L-Pro derivatives were used. However, there have been no
reports describing reactions promoted by a variety of acyclic
amino acids. For this reason, we next investigated the reaction
of 14 with several acyclic amino acids under conditions similar
to those utilized for 6 (Table 2).
In a manner similar to that reported by us previously, we
next synthesized (S)-7 from (S)-15 with 87% ee to determine
the absolute configuration of the aldol product (R)-7 (Scheme
2). To this end, chemo- and stereoselective cyanohydrin
formation from ketone (S)-15 was undertaken using trimethyl-
silyl cyanide23 in the presence of a catalytic amount of
(22) (a) Hiroi, K.; Yamada, S. Chem. Pharm. Bull. 1975, 23, 1103. (b)
Coisne, J.-M.; Pecher, J. Bull. Soc. Chim. Belg. 1981, 90, 481. (c) Tamai,
Y.; Mizutani, Y.; Uda, H.; Harada, N. J. Chem. Soc., Chem. Commun. 1983,
114. (d) Hagiwara, H.; Uda, H. J. Org. Chem. 1988, 53, 2308.
(23) (a) Evans, D. A.; Truesdale, L. K.; Carroll, G. L. J. Chem. Soc.,
Chem. Commun. 1973, 55.; J. Org. Chem. 1974, 39, 914. (b) Lidy, W.;
Sundermeyer, W. Chem. Ber. 1973, 106, 587.
Trione 14 was prepared from 2-methyl-1,3-cyclohexanedione
by way of a known method.16 In Table 2, the results of
126 J. Org. Chem., Vol. 72, No. 1, 2007

New Enedione Containing a SeVen-Membered Ring
SCHEME 3a
TABLE 3. Solvent Effects of Asymmetric Aldol Reaction of 6
Mediated by L-Meta
temp,
°C
time,
h
yieldb,
%
eec,
%
entry
solvent
1
2
3
4
5
6
DMF
95
75
95
75
95
75
120
120
120
120
120
120
37
61
10
5
51
15
24
40
22
37
7
CH3CN
dioxane
DME
nBuOH
benzene
38
a Reaction and conditions: L-Met (1.0 equiv), 1 N HClO4 (0.5 equiv).
b Isolated yield. c Determined by HPLC with a chiral stationary phase.
of 57% ee in ethanol at 0 °C afforded allylic alcohol 19
stereoselectively in 85% yield as a single product. We did not
observe the corresponding alcohol 20 or diol 21 in this reaction.
This chemo- and stereoselective reduction differs intrinsically
from a similar reaction of 3b or 5 reported independently by
Heathcock, Hagiwara, and Shibasaki.16,22d,27 These works
reported that reduction of 3b or 5 with NaBH4 afforded the
product resulting from stereoselective reduction of the saturated
ketone moiety and not the enone moiety. Alcohol 19 was in
turn reacted with (R)-MTPA chloride [(R)-MTPACl] (22),28
prepared from (S)-MTPA by a known method, in the presence
of 4-dimethylaminopyridine (DMAP) in pyridine and dichlo-
romethane to afford 23 accompanied by a small amount of a
diastereomer at C-4a which originated from (S)-7 contained in
starting (R)-7 with 57% ee. After fractional recrystallization,
of major product 23 as single crystals, X-ray diffraction
demonstrated the absolute configuration of 23 to be 2R,4aR by
relating to the known (S)-configuration of the MTPA compo-
nent. The absolute configuration of starting 7 had therefore been
correctly assigned as R. Furthermore, it was demonstrated that
hydride reduction of the enone functionality of (R)-7 had
occurred stereoselectively from the side opposite the angular
methyl substituent.
At this point, we returned to optimize the reaction conditions
involving 6. Table 3 summarizes those solvent effects involving
the reaction mediated by L-Met. Nonprotic polar solvents such
as DMF and MeCN prolonged the reaction time and afforded
(R)-7 in lower yield and ee than in the DMSO example (entries
1 and 2). Ethereal solvents such as 1,4-dioxane and 1,2-
dimethoxyethane (DME) proved unsuitable (entries 3 and 4).
When n-butanol was used as solvent, a modest yield ac-
companied by the lowest ee from among these trials was
observed (entry 5). On this basis, DMSO was selected as the
solvent for continued optimization.
a Reaction conditions: (a) NaBH4 (0.7 equiv), EtOH, 0 °C, 10 h, 80%;
(b) (R)-MTPACl (22, 1.3 equiv), DMAP (0.05 equiv), pyridine-CH2Cl2, rt,
5 h, 90%.
18-crown-6 and potassium cyanide.24 The single product 16 was
obtained in 92% yield. Although the stereostructure of 16 has
not been determined, we have anticipated that attack of cyanide
occurs from the side opposite that occupied by the angular
methyl substituent to afford the R-oriented isomer. When 16
was subsequently reduced with lithium aluminum hydride in
refluxing tetrahydrofuran (THF), the amino diol was formed
accompanied by 1,2-reduction of the enone moiety. The
diastereoselectivity of allylic alcohol formation in 17 was
determined to be 2.2:1 by integration of 1H NMR spectra
recorded on a CDCl3 solution of unpurified products. The
relevant signals of the amino methylene protons appear at δ
2.79 and δ 2.91, respectively. Diazotization of 17 with sodium
nitrite in aqueous acetic acid25 resulted in the generation of a
number of products. Oxidation of the allylic alcohol functionality
with manganese dioxide simplified matters to the four-
component level. After chromatographic separation, two of these
were identified as the desired enone (S)-7 and the regioisomeric
enone 1826 in 0.7% yield and 25% yield, respectively. We have
not pursued identification of the other two components that were
isolated in reasonable amounts. All of the spectral data for (S)-
7, except for the optical rotation, were identical with those of
enone (R)-7 which was obtained as above from 6. The optical
purity of (S)-7 was determined to be 91% ee by the same HPLC
technique as that described above. Therefore, significant loss
of optical purity had not occurred during the ring expansion
process. The optical rotation of enone (S)-7 was shown to be
[R]D +72.6 (c 0.66, CHCl3, 91% ee). Because the optical
rotation of the aldol product (R)-7 consisted of [R]D -83.4 (c
1.0, CHCl3, >99% ee), the absolute configuration of the aldol
product 7 must be R.
The effect of additives was next probed. The reactions were
carried out in DMSO at 90 °C in the presence of an L- or
D-amino acid with or without an acid such as pyridinium
p-toluenesulfonate (PPTS), p-toluenesulfonic acid (p-TsOH), and
(+)- or (-)-camphorsulfonic acid (CSA) as the additive. The
results are given in Table 4.
To provide added confirmation of the stereostructure of (R)-
7, we also examined by X-ray diffraction the corresponding (S)-
2-methoxy-2-(trifluoromethyl)phenylacetic acid [(S)-MTPA]
ester 23 (Scheme 3). Sodium borohydride reduction of (R)-7
We first examined the effect of HClO4. The opposite
enantioselectivity to afford (S)-7, accompanied with a lower ee
(24) Greenlee, W. J.; Hangauer, D. G. Tetrahedron Lett. 1983, 24, 4559.
(25) Di Maio, G.; Tardella, P. A. Tetrahedron 1966, 22, 2069.
(26) Uma, R.; Rajagopalan, K.; Swaminathan, S. Tetrahedron 1986, 42,
2757.
(27) Dutcher, J. S. D.; Macmillan, J. G.; Heathcock, C. H. J. Org. Chem.
1976, 41, 2663.
(28) Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34,
2543.
J. Org. Chem, Vol. 72, No. 1, 2007 127

Nagamine et al.
TABLE 4. Effects of Additive in the Reaction of 6a
TABLE 5. Effects of the Amount of (+)-CSA for Asymmetric
Intramolecular Aldol Reaction of 6a
time,
h
yield,b
%
ee,c,d
%
entry
amino acid
additive
(+)-CSA,
yieldb
of 7, %
eec,d
of 7, %
recoveryb
of 6, %
entry
equiv
1e
2
3
4
5
L-Met
L-Met
L-Met
L-Met
L-Met
L-Met
L-Met
-
D-Met
D-Met
D-Met
L-Phe
L-Phe
L-Phe
HClO4
-
96
18
120
18
18
15
99
44
41
67
66
82
70
52
40
41
69
86
80
81
53
-14
22
60
59
1e,f
2
0
44
66
82
87
85
91
27
15
-14
26
55
53
54
56
58
10
0
0
0
0
0
0
70
69
H2Of
0.2
0.4
0.5
0.6
0.8
1.0
1.2
PPTS
3
4g
5
p-TsOH
(+)-CSA
(-)-CSA
(+)-CSA
HClO4
PPTS
(+)-CSA
HClO4
PPTS
6
53
54
7
15
6
7
8
8g
9
120
114
19
40
100
13
-1
-25
-55
-57
48
10
11
12e
13
14
a Reaction and conditions: L-Met (1.0 equiv), (+)-CSA (n equiv), DMSO,
90 °C, 21 h. b Isolated yield. c Determined by HPLC with a chiral stationary
phase. d Opposite enantioselection was denoted by a minus sign. e The same
result is shown in Table 4. f Reaction was complete after 18 h. g Reaction
was complete after 15 h.
50
52
(+)-CSA
21
a Reaction conditions: Met or L-Phe (1.0 equiv), additive (0.5 equiv),
DMSO, 90 °C. b Isolated yield. c Determined by HPLC with a chiral
stationary phase. d Opposite enantioselection was denoted by a minus sign.
e The same result is shown in Table 1. f H2O (0.23 mL) was added without
any acids. g Amino acid was not added.
than that in entry 1 in Table 4, was observed in the presence of
L-Met without HClO4 (entry 2). Therefore, HClO4 plays a very
important role in this asymmetric process. Because 1 N HClO4
contains water, the effect of water was next probed. Thus, the
reaction was carried out with the same amount of water as in
entry 1 without HClO4 in the presence of L-Met. Entry 3 shows
that (R)-7 was obtained in moderate yield and lower ee than in
entry 1. In comparison with entries 2 and 3, it was seen that a
proton source, especially an acid, affected the enantioselection.
Next, acidic additives were screened to optimize the reaction.
The addition of PPTS or p-TsOH led to a slight increase in ee
with accompanying shortened reaction times (entries 4 and 5).
When (+)-CSA was used, (R)-7 was obtained in 82% yield and
53% ee (entry 6). Although the addition of (-)-CSA afforded
almost the same result as that in entry 1, the yield was a bit
lower than the example involving (+)-CSA (entry 7). Subse-
quently, reaction was carried out in the presence of D-Met
instead of L-Met to invert the enantioselectivity (entry 9). We
observed the production of (S)-7 under these conditions, but
both yield and ee were lower than those in the case involving
L-Met. Thus, complete inversion of enantioselectivity could not
be realized by the combination of D-Met and HClO4. In the
case of D-Met with PPTS or (+)-CSA, complete inversion of
enantioselectivity was observed with modest yield (entries 10
and 11). To elucidate the effect of CSA on asymmetric
induction, we tried the reaction without amino acid. Aldolization
of 6 by adding only (+)-CSA proceeded slowly to afford 7 after
120 h in 52% yield with no enantioselectivity (entry 8).
Consequently, enantioselection must be due to the chirality of
the amino acid and not to CSA. We also examined the
combination of L-Phe and PPTS or (+)-CSA, but no drastic
effect was observed, except for the shortening of reaction times
(entries 12-14).
FIGURE 5. Effects of the level of (+)-CSA on the asymmetric
intramolecular aldol reaction of 6.
of (+)-CSA, and both parameters reached a plateau at around
85% yield and 55% ee in the range of 0.4-0.8 equiv of CSA
(entries 3-6). Reactivity in the presence of more than 1.0 equiv
of CSA decreased drastically, and starting 6 was recovered even
after 21 h. However, the ee of (R)-7 continued to be around
55% (entry 7). With more than 1.2 equiv of CSA, the ee
decreased sharply. Therefore, the optimal amount of CSA is in
the range from 0.4 to 0.8 equiv.
Finally, we tried to extend this process to a catalytic version.
When the reaction was carried out in the presence of 0.3 equiv
of L-Met and 0.15 equiv of (+)-CSA or PPTS in DMSO at 90
°C, the results compiled in Table 6 were noted. Both reactions
using CSA or PPTS afforded (R)-7 in moderate yield with lower
ee, accompanied by recovery of 6 even after 120 h (entries 1
and 3). Shortened reaction times were a good improvement.
Thus, when the reaction was arrested after 24 h, (R)-7 was
obtained in moderate yield with almost the same ee as that in
the stoichiometric reaction, and the chemical yield based on
the recovery of starting 6 in both cases was improved to 75-
77% (entries 2 and 4). However, the catalytic version of the
reaction was difficult to achieve because of the low turn-over
numbers (TONs) in the attempted reactions. This low TON
meant that L-Met would lose the activity as a reaction mediator
or decompose under these conditions.
We next examined the effect of quantities of (+)-CSA on
the yield and ee of (R)-7. All reactions were carried out with
1.0 equiv of L-Met in DMSO at 90 °C for 21 h in the presence
of (+)-CSA over the range of 0 to 1.2 equiv (see Table 5 and
Figure 5). Higher yields and ee’s than those for the reaction
without any acids were observed upon increasing the amount
In the aldol reaction of 6 to afford a 6-7 fused bicycle, there
exists the possibility of producing eight stereoisomers as a result
of three newly generated stereogenic centers. Because we could
not isolate the initially formed â-hydroxy ketones, dehydration
to afford 7 must occur rapidly. The transition states predicted
128 J. Org. Chem., Vol. 72, No. 1, 2007

New Enedione Containing a SeVen-Membered Ring
FIGURE 6. Proposed transition states for the aldol reaction of 6 to afford all stereoisomers of aldol products.
TABLE 6. Asymmetric Aldol Reaction of 6 in the Presence of a
Catalytic Amount of L-Meta
time,
h
yieldb,c
of 7, %
eed
of 7, %
recoveryb
of 6, %
entry
additive
1
2
3
4
(+)-CSA
(+)-CSA
PPTS
120
24
120
24
40 (48)
43 (77)
41 (40)
43 (75)
35
55
33
56
17
44
33
43
FIGURE 7. Proposed transition states for the aldol reaction of 14.
PPTS
these considerations, the (Z)-trans-anti transition states 24 should
be most favored and afford the 1R,4aR,9aS-aldol product 25,
which would give rise to (R)-7 after dehydration. The existence
of hetero atoms at key positions on the side chain of the amino
acid can help to stabilize the transition state by engaging in a
new coordination to the hydrogen-bonded proton. When no
Brønsted acid was used, inversion of enantioselectivity was
noted with lower ee (Table 4, entry 2). This result means that
the reaction without an acid might proceed through an acyclic
transition state and result in lower ee. A Brønsted acid cocatalyst,
such as HClO4, CSA, and PPTS, may suppress dissociation of
the proton from the carboxylic acid moiety and help to maintain
or stabilize the cyclic transition states. Further investigations
into the effects of an acid cocatalyst and calculations involving
the energy differentiation of our proposed transition states are
currently underway.
a Reaction conditions: L-Met (0.3 equiv), (+)-CSA or PPTS (0.15 equiv),
DMSO, 90 °C. b Isolated yield. c Yields in parentheses were based on the
recovery of starting 6. d Determined by HPLC with a chiral stationary phase.
to reach each of the stereoisomers, which are based on Houk’s
transition model from the reaction of 4a (R ) CH3, n ) 1), are
shown in Figure 6.9g Among these models, all of the syn (syn,
anti: conformation between olefinic and carboxylic acid
moieties) transition states, 28, 30, 32, and 33, are considered to
be significantly higher in energy because of the steric hindrance
caused by repulsion between the methyl enamine and carboxylic
acid moieties in 28 and 32, the seven-membered carbocycle in
30, and the angular methyl substituent in 33. The trans-anti
(cis, trans: cis- or trans-fused carbocycles) transition state 26
appears to be disfavored because of 1,3-diaxial repulsion
between two methyl substituents. The (Z)- or (E)-cis-anti (Z,E:
geometry of the enamine moiety) transition states 34 and 35
also feature steric repulsion between the methyl enamine and
the seven-membered carbocycle. From expectations based on
The favored predicted transition states leading to (S)- or (R)-
15 existing in a 6-6 fused bicycle are shown in Figure 7. All of
the syn- and (E)-trans-anti transition states are expected to be
disfavored as per the reasons described above. On direct
J. Org. Chem, Vol. 72, No. 1, 2007 129

Nagamine et al.
The optical purity of (R)-7 was determined to be 53% ee by HPLC
equipped with a chiral stationary phase. HPLC conditions: Chiral-
pak AS-H, ethanol/hexanes ) 10:90, flow rate 1.0 mL/min, detected
at 254 nM, tR ) 9.5 min for (R)-7, 10.5 min for (S)-7. Optically
pure material was obtained by fractional recrystallization as
follows: compound (R)-7 (1.33 g, 53% ee) was recrystallized from
hexanes/ether (4:1) to afford racemic 7 (0.445 g, 0.27% ee) as
colorless prisms, and the filtrate was concentrated under reduced
pressure to afford (R)-7 (0.885 g, 80% ee), which was recrystallyzed
twice from hexanes/ether (4:1) to afford optically pure (R)-7 (0.134
g, >99% ee) as colorless prisms. For (R)-7: mp 98 °C (hexanes/
FIGURE 8. Transition states for the aldol reaction of 3a as proposed
by Houk.
ether); [R]23 -83.4 (c 1.01, CHCl3, >99% ee); IR (KBr, cm-1
)
D
1
1710, 1664, 1616; H NMR (270 MHz, CDCl3) δ 2.78-2.65 (m,
2 H), 2.62-2.46 (m, 2 H), 2.35-2.28 (m, 1 H), 2.16 (dt, J ) 6.0
Hz, 13.4 Hz, 1 H), 2.05-1.95 (m, 2 H), 1.86 (s, 3 H), 1.76-1.46
(m, 4 H), 1.34 (s, 3 H); 13C NMR (67.5 MHz, CDCl3) δ 213.7,
197.0, 158.9, 132.7, 53.4, 39.9, 32.7, 32.2, 31.2, 28.3, 28.2, 18.2,
10.4; EIMS m/z 206 (M+), 121 (100%); HRMS calcd for C13H18O2
206.1307, obsd 206.1308.
comparison of (Z)-trans-anti-36 with (Z)-cis-anti-37, 37 should
be more favored because of the lack of steric repulsion due to
the axial methyl substituent in 36. Also, there is no serious steric
hindrance as observed in 34 between the methyl enamine moiety
and the six-membered carbocycle because of the smaller six-
membered ring. The reaction should therefore proceed through
transition state 37 to afford (S)-15 preferentially after dehydra-
tion.
In regards to the aldol reaction of 3a to afford a 5-6 fused
bicycle in the presence of L-Phe, Houk reported the calculated
most favored transition states leading to (S)- and (R)-5a to be
39 and 38, respectively (Figure 8).9g Because the energy of
transition state 39 was 5.6 kcal/mol lower than that of 38, the
reaction of 3a afforded (S)-5a preferentially.
Typical Procedure for Amino Acid Mediated Aldol Reaction
of 14. A mixture of 14 (100 mg, 0.48 mmol), L-methionine (71
mg, 0.48 mmol), and 1 N HClO4 (0.24 mL, 0.24 mmol) in DMSO
(1 mL) was stirred at 90 °C for 22 h. After cooling, the mixture
was poured into saturated NaHCO3 solution and extracted with ethyl
acetate (AcOEt). The combined organic layers were washed with
brine and dried (MgSO4). The solvent was removed under reduced
pressure. The residue was chromatographed on silica gel (hexanes/
AcOEt, 5:1) to afford (S)-15 (68 mg, 74%) as a pale yellow oil.
The optical purity of (S)-15 was determined to be 84% ee by HPLC
equipped with a chiral stationary phase. HPLC conditions: Chiral-
pak AS-H, ethanol/hexanes ) 10:90, flow rate 1.0 mL/min, detected
at 254 nM, tR ) 10.9 min for (S)-15, 12.3 min for (R)-15. For
(S)-15: [R]24D +130.0 (c 1.0, CHCl3, 86% ee), lit.16 [R]22D + 120.7
(CHCl3, 80% ee); IR (KBr, cm-1) 1708, 1660, 1609; 1H NMR (270
MHz, CDCl3,) δ 2.88 (dtd, J ) 1.2 Hz, 5.0 Hz, 15.9 Hz, 1 H),
2.69 (ddd, J ) 5.8 Hz, 10.3 Hz, 15.9 Hz, 1 H), 2.57-2.38 (m, 4
H), 2.21-2.02 (m, 3 H), 1.85-1.67 (m, 1 H), 1.81 (d, J ) 1.4 Hz,
3 H), 1.42 (s, 3 H); 13C NMR (67.5 MHz, CDCl3) δ 212.0, 197.5,
158.1, 130.6, 50.5, 37.2, 33.2, 29.5, 27.2, 23.3, 21.4, 11.2; EIMS
m/z 192 (M+), 136 (100%); HRMS calcd for C12H16O2 192.1150,
obsd 192.1149.
Conclusion
In conclusion, we have developed a new intramolecular
asymmetric aldol reaction involving the use of amino acids to
obtain both enantiomers of the 6-7 fused bicyclic enone 7.
During these studies, a crossover in enantioselectivity was
uncovered as a function of the differing ring size. The
inducement of more elevated levels of enantioselectivity is
anticipated. The use of 7 as a new chiral synthon to achieve
total syntheses of selected pharmaceutically important natural
products is currently being probed.
Silylated Cyanohydrin (16). To a stirred solution of (S)-15 (10
g, 5.2 mmol, 87% ee) in CH2Cl2 (10 mL) was added KCN (50 mg,
0.78 mmol), 18-crown-6 (55 mg, 0.21 mmol), and trimethylsilyl
cyanide (0.83 mL, 6.2 mmol) at 0 °C. After being stirred for 30
min at the same temperature, the mixture was diluted with CH2-
Cl2, washed with saturated NaHCO3 solution and brine, then dried
(Na2SO4). The solvent was removed under reduced pressure. The
residue was chromatographed on silica gel (hexanes/ethyl acetate,
20:1 to 5:1) to afford 16 (1.389 g, 92%) as a colorless solid: mp
Experimental Section
2-Methyl-2-(3-oxopentyl)cycloheptane-1,3-dione (6). To a
stirred solution of 1-penten-3-one (1 mL, 10 mmol) and Triton B
(0.21 mL, 0.5 mmol) in methanol (4 mL) was added 2-methylcy-
cloheptane-1,3-dione (13)21 (700 mg, 5 mmol) as a small portion
over 5 min at rt. The mixture was heated to reflux for 1 h. After
cooling, the solvent was removed under reduced pressure and the
residue was dissolved in ether. The ether layer was washed with
saturated NaHCO3 solution and brine, then dried (Na2SO4). The
solvent was removed under reduced pressure, and the residue was
chromatographed on silica gel (hexanes/ethyl acetate, 10:1) to afford
6 (932 mg, 83%) as a pale yellow oil: IR (film, cm-1) 1715, 1694;
1H NMR (270 MHz, CDCl3) δ 2.54-2.40 (m, 4 H), 2.39 (q, J )
7.4 Hz, 2 H), 2.33-2.27 (m, 2 H), 2.11-2.05 (m, 2 H), 1.93-1.83
(m, 4 H), 1.18 (s, 3 H), 1.03 (t, J ) 7.3 Hz, 3 H); 13C NMR (67.5
MHz, CDCl3) δ 212.2, 209.7, 63.4, 41.2, 36.1, 35.8, 27.8, 26.6,
17.5, 7.5; EIMS m/z 224 (M+), 125 (100%); HRMS calcd for
C13H20O3 224.1412, obsd 224.1431.
Typical Procedure for Amino Acid Mediated Aldol Reaction
of 6. A mixture of 6 (400 mg, 1.78 mmol), L-methionine (266 mg,
1.78 mmol), and (+)-CSA (207 mg, 0.89 mmol) in DMSO (4 mL)
was stirred at 90 °C for 15 h. After cooling, the mixture was poured
into saturated NaHCO3 solution and extracted with ethyl acetate
(AcOEt). The combined organic layers were washed with brine
and dried (MgSO4). The solvent was removed under reduced
pressure. The residue was chromatographed on silica gel (hexanes/
AcOEt, 7:1) to afford (R)-7 (300 mg, 82%) as pale yellow crystals.
67-69 °C (hexanes); [R]22D +67.9 (c 1.00, CHCl3); IR (KBr, cm-1
)
1657, 1609; 1H NMR (400 MHz, CDCl3) δ 2.76 (ddd, J ) 2.2 Hz,
4.6 Hz, 15.3 Hz, 1 H), 2.55 (dt, J ) 3.8 Hz, 16.4 Hz, 1 H), 2.41
(dt, J ) 5.1 Hz, 15.4 Hz, 1 H), 2.29 (dt, J ) 4.4 Hz, 13.7 Hz, 1
H), 2.16-1.92 (m, 5 H), 1.87-1.74 (m, 1 H), 1.82 (s, 3 H), 1.23
(s, 3 H), 0.26 (s, 9 H); 13C NMR (100 MHz, CDCl3) δ 197.6, 155.7,
132.3, 120.5, 79.2, 45.2, 33.8, 33.5, 30.9, 25.6, 21.2, 16.7, 11.6,
1.2; EIMS m/z 291 (M+), 291 (100%); HRMS calcd for C16H25-
NO2Si 291.1655, obsd 291.1659. Anal. calcd for C16H25NO2Si: C,
65.93; H, 8.65; N, 4.81. Found: C, 66.10; H, 8.77; N, 4.81.
Hydride Reduction of 16. To a stirred suspension of lithium
aluminum hydride (6.345 g, 167.2 mmol) in THF (240 mL) was
added a solution of 16 (12.169 g, 41.8 mmol) in tetrahydrofuran
(120 mL) at 0 °C, and the mixture was heated to reflux for 13 h.
After cooling, water was carefully introduced, and stirring was
resumed for 2 h at rt. The mixture was filtered through a Celite
pad, and the filtrate was extracted with ethyl acetate. The combined
organic layers were dried (Na2SO4), and the solvent was removed
under reduced pressure. The residue was chromatographed on basic
130 J. Org. Chem., Vol. 72, No. 1, 2007

New Enedione Containing a SeVen-Membered Ring
Al2O3 (ethyl acetate/methanol/NH4OH, 10:1:0.1) to afford a dia-
stereomeric mixture of amino alcohol (5.311 g, 57%) as a pale
yellow oil. Product 17 was used in subsequent reactions without
further purification of diastereomers.
J ) 5.3 Hz, 11.7 Hz, 1 H), 2.21 (dd, J ) 6.7 Hz, 10.7 Hz, 1 H),
2.14-2.05 (m, 1 H), 1.97-1.90 (m, 1 H), 1.90-1.78 (m, 1 H),
1.81 (s, 3 H), 1.73-1.62 (m, 3 H), 1.57-1.37 (m, 4 H), 1.23 (s, 3
H); 13C NMR (100 MHz, CDCl3) δ 217.0, 135.4, 132.6, 70.2, 52.7,
40.0, 31.0, 30.3, 28.5, 28.4, 27.9, 20.6, 13.8; EIMS m/z 208 (M+),
107 (100%); HRMS calcd for C13H20O2 208.1463, obsd 208.1469.
(S)-MTPA Ester (23). To a stirred solution of 19 (678 mg, 3.26
mmol) and (R)-MTPACl28 (1.068 g, 4.24 mmol), which was freshly
prepared from (S)-MTPA, in CH2Cl2 (4 mL) was added pyridine
(4 mL) and DMAP (20 mg, 0.16 mmol) at 0 °C. After being stirred
for 5 h at rt, the mixture was poured into water and extracted with
ethyl acetate. The combined organic layers were washed with brine
and dried (Na2SO4). After solvent removal under reduced pressure,
the residue was chromatographed on silica gel (hexanes/ethyl
acetate, 5:1) to afford 23 as a diastereomeric mixture at C-4a (1.25
g, 90%, dr 4:1). Pure 23 was obtained as colorless prisms after
recrystallization from hexanes/CH2Cl2 (5:2).
Ring Expansion of 17. To a stirred solution of 17 (5.311 g,
23.6 mmol) in acetic acid (25 mL) and water (25 mL) was added
sodium nitrite (4.885 g, 70.8 mmol) at 0 °C. After being stirred for
3 h at the same temperature, the mixture was poured into water
and extracted with ethyl acetate. The combined organic layers were
washed with saturated NaHCO3 solution and brine, then dried (Na2-
SO4). The solvent was removed under reduced pressure. The residue
was dissolved in CH2Cl2 (50 mL), and manganese dioxide (20 g,
230 mmol) was added to the solution at rt. After 48 h of stirring at
rt, more manganese dioxide (10 g, 115 mmol) was added. The
mixture was stirred at rt for a further 48 h and filtered through a
pad of Celite. The filtrate was concentrated under reduced pressure.
The residue was chromatographed on basic alumina (hexanes/ethyl
acetate, 100:1 to 10:1) to afford pure 18 (1.224 g, 25%) and (S)-7
with a trace amount of impurity. The latter was chromatographed
again on silica gel (CH2Cl2/ethyl acetate, 100:1) to afford (S)-7
(35 mg, 0.7%).
For 23: mp 123 °C (hexanes/CH2Cl2); [R]24 -27.2 (c 1.01,
D
1
CHCl3); IR (KBr, cm-1) 1741, 1710, 1656; H NMR (400 MHz,
CDCl3) δ 7.58-7.55 (m, 2 H), 7.43-7.39 (m, 3 H), 5.60 (t, J )
7.2 Hz, 1 H), 3.65-3.55 (m, 3 H), 2.68-2.62 (m, 1 H), 2.54 (dd,
J ) 5.4 Hz, 11.8 Hz, 1 H), 2.28-2.18 (m, 2 H), 1.94-1.72 (m, 4
H), 1.54-1.37 (m, 4 H), 1.48 (s, 3 H), 1.23 (s, 3 H); 13C NMR
(100 MHz, CDCl3) δ 215.6, 166.4, 138.9, 132.3, 129.5, 128.3,
127.6, 127.1, 123.3 (q, JC-F ) 290 Hz, CF3), 84.3 (q, JC-F ) 28
Hz, CCF3), 75.5, 55.4, 52.4, 40.1, 30.6, 30.4, 28.5, 27.9, 24.4, 20.5,
13.8; EIMS m/z 424 (M+), 190 (100%); HRMS calcd for C23H27F3O4
424.1861, obsd 424.1869. Anal. calcd for C23H27F3O4: C, 65.08;
H, 6.41. Found: C, 65.10; H, 6.42.
For 18: colorless oil; [R]23 +261.2 (c 1.01, CHCl3); IR (film,
D
1
cm-1) 1697, 1666, 1620; H NMR (400 MHz, CDCl3) δ 2.91 (d,
J ) 12.4 Hz, 1 H), 2.82 (dt, J ) 4.8 Hz, 13.5 Hz, 1 H), 2.62-2.35
(m, 5 H), 2.24 (d, J ) 12.4 Hz, 1 H), 2.24-2.08 (m, 2 H), 1.78 (s,
3 H), 1.72-1.59 (m, 2 H), 1.27 (s, 3 H); 13C NMR (100 MHz,
CDCl3) δ 210.0, 197.7, 161.7, 131.7, 54.0, 43.6, 37.5, 33.9, 33.8,
27.9, 25.3, 22.8, 11.0; EIMS m/z 206 (M+), 91 (100%); HRMS
calcd for C13H18O2 206.1307, obsd 206.1277.
24
For (S)-7: colorless crystals; mp 98 °C (hexanes/ether); [R]D
+72.6 (c 0.66, CHCl3, 91% ee). The spectroscopic data were
Acknowledgment. We thank Prof. Kendall N. Houk and
Dr. Fernando R. Clemente, University of California, Los
Angeles, for their kind suggestions about the transition states
of amino acid mediated intramolecular aldol reactions. We also
thank Prof. Kentaro Yamaguchi and Dr. Masatoshi Kawahata,
Tokushima Bunri University, Kagawa, for the X-ray diffraction
studies.
identical to those of authentic (R)-7.
Sodium Borohydride Reduction of (R)-7. To a stirred suspen-
sion of (R)-7 (1.892 g, 9.18 mmol, 57% ee) in ethanol (20 mL)
was added NaBH4 (121 mg, 3.21 mmol) in small portions over 5
min at 0 °C. After 3 h of stirring at the same temperature, NaBH4
(121 mg, 3.21 mmol) was added to the mixture again and stirring
was maintained for a further 7 h. The reaction mixture was
quenched by adding acetone (20 mL), and the solvent was removed
under reduced pressure. The residue was dissolved in ethyl acetate,
washed with saturated NaHCO3 solution and brine, and dried (Na2-
SO4). After the solvent was removed under reduced pressure, the
residue was chromatographed on silica gel (hexanes/Et2O, 2:1) to
afford 19 (1.53 g, 80%) as a pale yellow oil: [R]23D -5.3 (c 1.01,
Supporting Information Available: General procedures; the
ORTEP structure and the crystallographic data for compound
23 in CIF format; and spectroscopic data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
1
CHCl3); IR (film, cm-1) 3409, 1705, 1653; H NMR (400 MHz,
CDCl3) δ 4.15 (t, J ) 7.5 Hz, 1 H), 2.72-2.65 (m, 1 H), 2.58 (dd,
JO061824N
J. Org. Chem, Vol. 72, No. 1, 2007 131