Unprecedented chemo-enzymatic synthesis of stereochemically pure 3-acetoxy-2-methyl-2-vinylcycloalkanones

Article abstract of DOI:10.1016/j.tetlet.2003.12.084

A new approach to 3-acetoxy-2-methyl-2-vinylcyclohexanone and 3-acetoxy-2-methyl-2-vinylcyclopentanone in stereochemically pure state, by means of a combination of yeast-catalyzed reduction and subsequent radical β-fragmentation is described.

Full text of DOI:10.1016/j.tetlet.2003.12.084

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 1763–1767
Unprecedented chemo-enzymatic synthesis of stereochemically
pure 3-acetoxy-2-methyl-2-vinylcycloalkanones
Ken-ichi Fuhshuku, Mina Tomita and Takeshi Sugai*
Department of Chemistry, Keio University, 3-14-1, Hiyoshi, Yokohama 223-8522, Japan
Received 19 November 2003; revised 10 December 2003; accepted 12 December 2003
Abstract—A new approach to 3-acetoxy-2-methyl-2-vinylcyclohexanone and 3-acetoxy-2-methyl-2-vinylcyclopentanone in stereo-
chemically pure state, by means of a combination of yeast-catalyzed reduction and subsequent radical b-fragmentation is
described.
ꢀ 2003 Elsevier Ltd. All rights reserved.
2,2-Disubstituted-3-hydroxycycloalkanones, which are
prepared by bakerꢀs yeast-catalyzed reduction of the
corresponding diketones, are good starting materials for
the construction of a characteristic structural feature
with chiral centers (A, such as stypoldione, Fig. 1) in
natural product synthesis, involving terpenoids, de-
graded carotenoids, steroids, and related substances, in
enantiomerically pure form.1 For example, Mori has
demonstrated a large number of applications on 2,2-
dimethyl-3-hydroxycyclohexanone 2a from 2,2-dimeth-
ylcyclohexane-1,3-dione 1a (Scheme 1).2
callol (Fig. 1). Recently, Nakada and co-workers
reported an elegant approach,3 the preparation of
2-benzyloxymethyl-2-methylcyclohexane-1,3-dione 1b,
based on the Birch reduction of 2,6-dimethoxybenzoic
acid and subsequent transformations, submitted that to
bakerꢀs yeast-catalyzed reduction and successfully
obtained the hydroxyketone 2b0. In turn, another
approach via the alkylation of the precursors, 2-methyl-
cycloalkane-1,3-diones, which apparently seems to be
more general, is somewhat problematic. Toward most
alkyl halides other than methyl iodide, the enolates of
the diketones are prone to undergo O-alkylation even
with ethyl iodide.4 The allyl and propargyl groups meet
this criterion, and the diketones 1c and 1d are obtainable
in acceptable yields, but the subsequent yeast-catalyzed
reductions proceeded in a nonenantiotopic group-
selective manner to give an almost equimolar mixture
of diastereomers (2c+2c0, 2d+2d0, Scheme 1).5;6 We
A problem in this approach is that the range of sub-
strates available to yeast-catalyzed reduction, however,
is rather limited. Needless to say, some advanced com-
pounds bearing substituents convertible to oxygen-con-
taining functional groups at the C-2 position (B) deserve
attention in natural product synthesis such as tripto-
O
OMe
H
HO
HO
HO
H
H
R
O
O
HO
Me
CO2R',
CH2OR'
A R =
B
stypoldione
triptocallol
Figure 1. Characteristic structural feature with chiral centers in natural product synthesis, such as stypoldione and triptocallol.
Keywords: Substituted cyclohexanone; Substituted cyclopentanone; Hydroxyketone; Radical b-fragmentation.
* Corresponding author. Fax: +81-45-566-1697; e-mail: sugai@chem.keio.ac.jp
0040-4039/$ - see front matter ꢀ 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.12.084

1764
K. Fuhshuku et al. / Tetrahedron Letters 45 (2004) 1763–1767
O
OH
OH
baker's yeast
R
R
O
R
O
+
O
1a
b
2a
2a'
R = Me
b
c
d
e
b'
c'
d'
e'
CH2OBn
c
CH2CH CH2
CH2C CH
CH CH2
d
e
OH
O
O
O
O
OH
T. delbrueckii
IFO10921
( )n
( )n
( )n
O
O
O
1f n = 1
2f
4f
5f n = 1
6f 0
3f
0
Scheme 1. Yeast-catalyzed reduction of 2-substituted-2-methylcycloalkane-1,3-diones.
envisioned that 2-methyl-2-vinylcycloalkane-1,3-diones
such as 1e, would be other very attractive substrates for
yeast-catalyzed reduction. The direct introduction of a
vinyl group on this position, however, is extremely dif-
ficult, and only phenyl 2-(trimethylsilyl)ethynyl sulfone
as an electrophile, vinyl cation equivalent, has been
reported.7
In this paper, we report an unprecedented route to
3-acetoxy-2-methyl-2-vinylcyclohexanone 7e and
3-acetoxy-2-methyl-2-vinylcyclopentanone 8e in stereo-
chemically pure state, from these cyclic hemiacetals 5f
and 6f by means of a radical b-fragmentation. Our idea,
which was inspired by Rigbyꢀs transformation,10 is
shown in Scheme 2. If a proper precursor can be pre-
pared and the homolytic cleavage of this O–X bond
takes place (C), the subsequent b-fragmentation will
provide the radical intermediate (D). The oxidation of
this species will give a 2-vinyl substituent in the desired
product. This scheme may be advantageous in that a
part of the original hemiacetal skeleton is incorporated
into the protecting group of the liberated hydroxyl
group, in the form of acetate.
In recent years, we found that a yeast strain, Torulas-
pora delbrueckii IFO10921,8 worked very well on the
triketones 1f and 3f, which are readily available from
2-methylcycloalkane-1,3-diones and methyl vinyl
ketone, to give the stereochemically pure products
(Scheme 1).8;9 For example, from the triketone 1f, a
cyclic hemiacetal 5f was obtained in 60% yield. The
combined X-ray crystallographic structure determina-
tion and chemical transformations proved its structure
to be (1S,3S,6S)-configuration.8
Along this line, the reaction conditions were extensively
examined (Table 1). In our first attempt for the use of
O
OH
O
O
X
O
O
OAc
( )n
( )n
( )n
( )n
O
O
O
O
C
D
5f n = 1
6f
7e n = 1
8e
0
0
Scheme 2. Synthetic plan for 3-acetoxy-2-methyl-2-vinylcycloalkanones.
Table 1. Studies on the reagents available to the b-fragmentation of the hemiacetal 5fa
Entry
Reagent
Oxidant
Yield (%)
12
Recovery (%)
––
1b
2c
3
Pb(OAc)4
Cu(OAc)2ÆH2O
Cu(OAc)2
Cu(OAc)2
PhI(OCOCF3)2
PhI(OAc)2
PhI@O
Decomposition
30
9
10
––
4
Cu(OAc)2
a Pyridine, benzene, reflux, 15 h.
b 38 h.
c CH2Cl2, rt, 1 h.

K. Fuhshuku et al. / Tetrahedron Letters 45 (2004) 1763–1767
Table 2. Conditions for the b-fragmentation with iodobenzene diacetatea
1765
Entry
Substrate
Additive
Solvent
Yield (%)
Recovery (%)
1
5f
5f
5f
5f
5f
5f
5f
5f
5f
6f
––
Benzene
Benzene
Benzene
Benzene
Cyclohexane
ClCH2CH2Cl
DME
No reaction
2
Pyridine
27
9
10
43
10
19
16
34
3
2,20-Bipyridyl
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
I2, hm
4
30
26
16
13
5
6
7
8b
9c
10c
CH2Cl2
Benzene
Benzene
Decomposition
2,6-Lutidine
2,6-Lutidine
39
65
––
––
a PhI(OAc)2, Cu(OAc)2, reflux, 15 h.
b PhI(OAc)2, tungsten lamp (100 W), rt, 10 h.
c 5 mmol scale; PhI(OAc)2 (20 mmol, five times), Cu(OAc)2 (2.0 mmol, five times), 2,6-lutidine (6.0 mmol), benzene (100 mL), reflux, 5 h; for detail, see
Ref. 14.
lead(IV) acetate and copper(II) acetate under Rigbyꢀs
conditions,10 however, only a trace amount of the
desired product 7e was obtained as a very complex
mixture with a number of byproducts. Even after the
extensive investigation on the reaction conditions with
this radical initiator, the yield of 7e remained as low as
12% (entry 1).
for the radical intermediate (D), and the use of cop-
per(II) acetate was necessary (entry 1). As the substitu-
ents for another additive, pyridine (entry 2), which
works as the ligand indispensable for the monomeriza-
tion of the copper complex,13 2,20-bipyridyl and 2,6-
lutidine were compared (entries 3 and 4). The latter
advantageous case was further studied by a change of
the reaction solvent (entries 4–7), and the highest yield
was given in benzene (entry 4). Although the yield of 7e
increased, this reaction produced a considerable amount
of byproducts due to the side reactions, which occurred
under the severe reflux conditions. In turn, a more mild
photochemical initiation and the following trapping
with iodine were attempted (entry 8), expecting to
obtain a primary iodide, which would be led to the vinyl
group by the subsequent dehydrohalogenation. To our
disappointment, the multispots on TLC analysis were
still observed. Among the many products, only an
isolable but very unstable component showed an acetate
of the partial structure by NMR, immediately after the
We turned our attention to the use of organohyperva-
lent iodine reagents, by which a wide range of applica-
tions have been exploited in organic synthesis.11 It has
been known that a combined use of iodobenzene diac-
etate, pyridine, and a catalytic amount of copper(II)
acetate equally worked well instead of the lead(IV)-
mediated decarboxylative radical formation into
alkene.12 Through the studies with several candidates
(entries 2–4), we selected iodobenzene diacetate, and
based on the use of this reagent, the reaction conditions
were further elaborated (Table 2). The excessively used
iodobenzene diacetate itself did not work as the oxidant
OAc
OAc
HO
OH
a
b,c
O
O
O
O
O
7e
9
10
TBSO
O
OTBS
OBn
TBSO
O
OTBS
OBn
h
d,e
11
12
13
14
10
RO
RO
O
f,g
BnO
O
BnO
O
i
Scheme 3. Reagents and conditions: (a) (CH2OH)2, p-TsOH, benzene, reflux (89%); (b) O3, CH2Cl2, Me2S; (c) LiAlH4, Et2O (2 steps 77%); (d)
TBSCl, imidazole, DMF, 60 ꢁC; (e) PPTS, acetone–H2O, 60 ꢁC (2 steps 75%); (f) BnBr, NaH, DMF; (g) p-TsOH, acetone–H2O, 60 ꢁC (2 steps quant);
(h) LHMDS, HMPA, THF, )78 ꢁC, then MeI (74%); (i) LHMDS, HMPA, THF, )78 ꢁC, then MeI (quant).

1766
K. Fuhshuku et al. / Tetrahedron Letters 45 (2004) 1763–1767
2. Mori, K.; Mori, H. Org. Synth. (Coll. Vol.) 1994, 8, 312–
315; Mori, K. Chemoenzymatic Synthesis of Pheromones,
Terpenes, and Other Bioregulators. In Stereoselective
Biocatalysis; Patel, R. N., Ed.; Marcel Dekker: New
York, 2000; pp 59–85.
3. Iwamoto, M.; Kawada, H.; Tanaka, T.; Nakada, M.
Tetrahedron Lett. 2003, 44, 7239–7243.
4. Mori, K.; Fujiwhara, M. Tetrahedron 1988, 44, 343–354.
5. Brooks, D. W.; Mazdiyasni, H.; Grothaus, P. G. J. Org.
Chem. 1987, 52, 3223–3232.
purification. The component rapidly turned red by
releasing iodine. Based on these evidences, we reasoned
that the desirable b-fragmentation certainly occurred to
give an iodide, but due to its unstable nature, further
decomposition predominated. In contrast, we were
pleased that under the optimized conditions14 giving the
highest yield of 7e (39%, entry 9) from 5f, the yield of the
five-membered ring product 8e was remarkably high
(65%, entry 10) from the substrate 6f. In this way, new
stereochemically pure cyclohexanone 7e15 and cyclo-
pentanone 8e16 were obtained.
6. Wei, Z.-L.; Li, Z.-Y.; Lin, G.-Q. Synthesis 2000, 1673–
1676.
7. Ohnuma, T.; Hata, N.; Fujiwara, H.; Ban, Y. J. Org.
Chem. 1982, 47, 4713–4717; Ohnuma, T.; Hata, N.;
Miyachi, N.; Wakamatsu, T.; Ban, Y. Tetrahedron Lett.
1986, 27, 219–222.
8. Fuhshuku, K.; Funa, N.; Akeboshi, T.; Ohta, H.; Hosomi,
H.; Ohba, S.; Sugai, T. J. Org. Chem. 2000, 65, 129–135,
T. delbrueckii IFO10921 is now available as the strain of
NBRC10921 from the NITE Biological Resource Center
(NBRC), 2-5-8, Kazusakamatari, Kisarazu, Chiba 292-
0818, Japan.
9. Fuhshuku, K.; Tomita, M.; Sugai, T. Adv. Synth. Catal.
2003, 345, 766–774.
10. Rigby, J. H.; Warshakoon, N. C.; Payen, A. J. J. Am.
Chem. Soc. 1999, 121, 8237–8245; Rigby, J. H.; Payen, A.
J.; Warshakoon, N. C. Tetrahedron Lett. 2001, 42, 2047–
2049.
We continued the functional group transformations
toward the key synthetic intermediates of B as shown in
Scheme 3. The carbonyl group of 7e was protected as
1,3-dioxolane (9)17 in 89% yield, and the ozonolysis of 9
followed by the reduction provided the diol 1018 in 77%
yield from 9. Both the TBS group (11)19 and the benzyl
group (12)20 were effectively introduced in 75% yield and
quantitative yield from 10, respectively. This bis-benzyl
ether 12 implies the successful approach to the diaste-
reomeric synthetic equivalent of 2b0 reported by Nakada
and co-workers.3 Although these compounds (11 and
12) have three functional groups, all adjacent to the
quaternary chiral center in a very congested manner, the
methylation was successful to provide 13 and 14,
important synthetic intermediates for B, in 75% yield
and quantitative yield as a diastereomeric mixture at the
C-6 position (ca. 1:1 mixture), respectively.
11. Togo, H.; Katohgi, M. Synlett 2001, 565–581; Zhdankin,
V. V.; Stang, P. J. Chem. Rev. 2002, 102, 2523–2584;
Wirth, T., Ed.; Hypervalent Iodine Chemistry. Top. Curr.
Chem. 2003, 224.
ꢀ
12. Concepcion, J. I.; Francisco, C. G.; Freire, R.; Hernandez,
R.; Salazar, J. A.; Suarez, E. J. Org. Chem. 1986, 51, 402–
ꢀ
ꢀ
In conclusion, a new entry to stereochemically pure
forms of highly functionalized cycloalkanone interme-
diates for natural product synthesis, by means of a
combination of yeast-catalyzed reduction and sub-
sequent radical b-fragmentation as the key steps, was
disclosed.
404.
13. Kochi, J. K.; Subramanian, R. V. Inorg. Chem. 1965, 4,
1527–1533; Kochi, J. K.; Subramanian, R. V. J. Am.
Chem. Soc. 1965, 87, 4855–4866; Kochi, J. K.; Bemis, A.;
Jenkins, C. L. J. Am. Chem. Soc. 1968, 90, 4616–
4625.
14. A mixture of benzene (100 mL), Cu(OAc)2 (2.0 mmol), and
2,6-lutidine (6.0 mmol) was stirred for 15 min at 60 ꢁC, and
the hemiacetal (5.0 mmol) was added. The reaction mix-
ture was degassed by the repetitive evacuation and
substitution with argon. The first portion of PhI(OAc)2
(20 mmol) was then added, and the reaction mixture was
stirred under reflux. To this reaction mixture, four
additional portions of PhI(OAc)2 and Cu(OAc)2 (each
20 and 2.0 mmol, respectively) were added in an interval of
every 1 h, and the mixture was stirred under reflux for a
total of 5 h. The disappearance of the starting material was
confirmed by TLC analysis (silica gel, developed with
hexane–EtOAc 2:1). The reaction mixture was filtered
through a pad of silica gel, and the filtrate was concen-
trated in vacuo. The residue was diluted with EtOAc,
successively washed with saturated aqueous Na2S2O3
solution, water, saturated aqueous NaHCO3 solution
and brine, dried over anhydrous Na2SO4, and concen-
trated in vacuo. The residue was purified by silica gel
Acknowledgements
The authors thank Professors Hideo Togo of Chiba
University, Kozo Shishido of Tokushima University for
their helpful discussions on the radical b-fragmentation.
This work was accomplished as the 21st Century COE
Program (KEIO LCC) from the Ministry of Education,
Culture, Sports, Science, and Technology, Japan. This
work was also supported by the collaborated program
of ꢁCREST: Creation of Functions of New Molecules
and Molecular Assembliesꢀ of Japan Science and Tech-
nology Corporation, and we express our sincere thanks
to Professors Keisuke Suzuki, and Takashi Matsumoto
of Tokyo Institute of Technology. Part of this work was
supported by Grant-in-Aid for Scientific Research (no.
14560084).
column chromatography.
19
D
15. 7e: ½aꢀ +65.1ꢁ (c 1.07, CHCl3). IR mmax 1738, 1714, 1634,
1229 cmꢁ1
.
1H NMR (400 MHz, CDCl3) d 1.21 (3H, s),
References and notes
1.56–1.70 (1H, m), 1.83–2.06 (3H, m), 2.06 (3H, s), 2.36
(1H, dddd, J ¼ 1.0, 4.4, 4.4, 14.7 Hz), 2.52 (1H, ddd,
J ¼ 5.9, 11.7, 14.7 Hz), 4.81 (1H, dd, J ¼ 3.9, 9.8 Hz), 5.00
(1H, d, J ¼ 17.6 Hz), 5.22 (1H, d, J ¼ 10.7 Hz), 6.25 (1H,
dd, J ¼ 10.7, 17.6 Hz). 13C NMR (100 MHz, CDCl3) d 20.4,
20.4, 21.1, 26.9, 37.8, 56.6, 78.1, 116.9, 137.4, 170.0, 209.1.
Anal. Found: C, 67.19; H, 8.35. Calcd for C11H16O3: C,
€
1. Servi, S. Synthesis 1990, 1–25; Csuk, R.; Glanzer, B. I.
Chem. Rev. 1991, 91, 49–97; Fuhshuku, K.; Oda, S.; Sugai,
T. Recent Res. Devel. Org. Chem. 2002, 6, 57–74;
Nakamura, K.; Yamanaka, R.; Matsuda, T.; Harada, T.
Tetrahedron: Asymmetry 2003, 14, 2659–2681.

K. Fuhshuku et al. / Tetrahedron Letters 45 (2004) 1763–1767
1767
23
D
67.32; H, 8.22. Based on the GLC analysis [column, TCI
Chiraldex B-PM, 30 m·0.25 mm · 0.125 lm; flow rate
50 mL/min; pressure 180 kPa; oven temperature 120 ꢁC;
an authentic specimen ( )-7e: tR ¼ 26:8 and 27.6 min], the
b-fragmentation product 7e of microbial origin coincided
the peak of 26.8 min, and its ee was determined to be
>99.9%. As, no racemization was observed in this b-
fragmentation process, the absolute configuration of 7e
18. 10: ½aꢀ +13.0ꢁ (c 1.03, CHCl3). IR mmax 3324, 1453, 1190,
1132, 1100, 1044, 1023 cmꢁ1. 1H NMR (400 MHz, CDCl3)
d 1.13 (3H, s), 1.49–1.80 (6H, m), 2.59 (1H, s), 3.13 (1H, s),
3.72–3.78 (2H, m), 3.89–4.03 (5H, m). 13C NMR
(100 MHz, CDCl3) d 16.8, 18.3, 28.7, 29.9, 46.2, 63.9,
65.1, 65.3, 75.1, 112.9. Anal. Found: C, 59.46; H, 8.96.
Calcd for C10H18O4: C, 59.39; H, 8.97.
20
D
19. 11: ½aꢀ +19.8ꢁ (c 0.89, CHCl3). IR mmax 1709, 1472, 1463,
1
was unambiguously confirmed to be (2S,3S).
1255, 1078, 837, 775 cmꢁ1. H NMR (400 MHz, CDCl3) d
22
D
16. 8e: ½aꢀ )25.7ꢁ (c 1.02, CHCl3). IR mmax 1743, 1638,
0.03 (3H, s), 0.03 (3H, s), 0.05 (3H, s), 0.05 (3H, s), 0.86
(9H, s), 0.88 (9H, s), 1.13 (3H, s), 1.62–1.71 (1H, m), 1.80–
1.97 (2H, m), 1.99–2.09 (1H, m), 2.33 (1H, ddd, J ¼ 5.9,
6.4, 14.2 Hz), 2.45 (1H, ddd, J ¼ 5.9, 8.8, 14.6 Hz), 3.76
(1H, d, J ¼ 9.8 Hz), 3.81 (1H, d, J ¼ 9.8 Hz), 3.94 (1H, dd,
J ¼ 2.9, 6.4 Hz). 13C NMR (100 MHz, CDCl3) d )5.5,
)5.2, )5.0, )4.2, 18.1, 18.3, 19.2, 20.5, 25.9, 26.0, 28.8,
38.3, 55.8, 64.4, 75.2, 214.0. Anal. Found: C, 62.31; H,
1242 cmꢁ1
.
1H NMR (400 MHz, CDCl3) d 1.19 (3H, s),
1.95–2.03 (1H, m), 2.07 (3H, s), 2.29–2.38 (2H, m), 2.44–
2.56 (1H, m), 5.11–5.16 (2H, m), 5.25 (1H, d, J ¼ 10.7 Hz),
5.86 (1H, dd, J ¼ 10.7, 17.6 Hz). 13C NMR (100 MHz,
CDCl3) d 20.1, 21.1, 25.5, 34.5, 55.4, 79.2, 117.4, 134.4,
170.1, 215.9. Anal. Found: C, 66.07; H, 7.83. Calcd for
C10H14O3: C, 65.91; H, 7.74. Based on the GLC analysis
[column, Supelco BETA DEXe 225, 30 m · 0.25 mm ·
0.25 lm; flow rate 50 mL/min; pressure 180 kPa; oven
temperature 120 ꢁC; an authentic specimen ( )-8e:
tR ¼ 15:4 and 15.8 min], the b-fragmentation product 8e
of microbial origin coincided the peak of 15.4 min, and its
10.70. Calcd for C20H42O3Si2: C, 62.12; H, 10.95.
20
D
20. 12: ½aꢀ +27.9ꢁ (c 1.07, CHCl3). IR mmax 3087, 3062, 3030,
1705, 1453, 1091, 1076, 736, 696 cmꢁ1
.
1H NMR
(400 MHz, CDCl3) d 1.30 (3H, s), 1.67–1.75 (1H, m),
1.94–2.10 (3H, m), 2.30 (1H, ddd, J ¼ 5.4, 5.4, 14.6 Hz),
2.50 (1H, ddd, J ¼ 5.9, 10.2, 14.6 Hz), 3.68 (1H, d,
J ¼ 9.3 Hz), 3.80 (1H, dd, J ¼ 3.9, 4.4 Hz), 3.85 (1H, d,
J ¼ 9.3 Hz), 4.40 (1H, d, J ¼ 11.7 Hz), 4.47 (1H, d,
J ¼ 12.2 Hz), 4.52 (1H, d, J ¼ 12.2 Hz), 4.57 (1H, d,
J ¼ 11.7 Hz), 7.23–7.33 (10H, m). 13C NMR (100 MHz,
CDCl3) d 20.0, 20.5, 24.0, 37.9, 54.4, 71.3, 71.7, 73.3, 81.6,
127.3, 127.3, 127.3, 128.2, 138.5, 138.5, 213.1. Anal.
Found: C, 78.00; H, 7.78. Calcd for C22H26O3: C, 78.07;
H, 7.74.
ee was determined to be >99.9%.
25
D
17. 9: ½aꢀ +99.5ꢁ (c 1.35, CHCl3). IR mmax 1734, 1653, 1237,
1042 cmꢁ1
.
1H NMR (400 MHz, CDCl3) d 1.05 (3H, s),
1.57–1.87 (6H, m), 2.03 (3H, s), 3.86–3.98 (4H, m), 4.89–
4.98 (1H, m), 5.26 (1H, dd, J ¼ 2.0, 11.2 Hz), 5.29 (1H, dd,
J ¼ 2.0, 17.6 Hz), 6.15 (1H, dd, J ¼ 11.2, 17.6 Hz). 13C
NMR (100 MHz, CDCl3) d 16.3, 19.5, 21.2, 26.6, 30.7, 50.0,
65.1, 65.2, 77.3, 111.6, 116.9, 137.9, 170.4. Anal. Found: C,
65.22; H, 8.55. Calcd for C13H20O4: C, 64.98; H, 8.39.