Lipase-catalyzed kinetic resolution of racemic seven-membered substituted lactones

Article abstract of DOI:10.1016/S0040-4039(00)01421-0

Optically active 3-, 4-, 5-, or 6-methyl substituted seven-membered lactones were prepared by Candida antarctica lipase (CAL)-catalyzed kinetic resolution of the racemic lactones. Among these lactones, only in the case of 5-methylhexanolide (R)-isomer reacts faster than the other enantiomer, while the rest of lactones show (S)-selectivity. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01421-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 8799–8802
Lipase-catalyzed kinetic resolution of racemic
seven-membered substituted lactones
Kosei Shioji,a,* Ayako Matsuo,a Kentaro Okuma,a Kaoru Nakamurab and
Atsuyoshi Ohnob,†
aDepartment of Chemistry, Faculty of Science, Fukuoka University, Jonan-ku, Fukuoka 814-0180, Japan
bInstitute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan
Received 21 July 2000; revised 16 August 2000; accepted 18 August 2000
Abstract
Optically active 3-, 4-, 5-, or 6-methyl substituted seven-membered lactones were prepared by Candida
antarctica lipase (CAL)-catalyzed kinetic resolution of the racemic lactones. Among these lactones, only
in the case of 5-methylhexanolide (R)-isomer reacts faster than the other enantiomer, while the rest of
lactones show (S)-selectivity. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: enzymes and enzyme reaction; hydrolysis; lactone.
Lipase-catalyzed enantioselective hydrolysis has been known as a method for preparing
optically active five- or six-membered substituted lactones.1 Rousseau and co-workers reported
that enzymatic hydrolysis of g- and d-lactones using commercially available horse liver esterase
(HLE), pig liver esterase (PLE), or pig pancreatic lipase (PPL), gave optically active lactones.2
On the contrary, substituted seven-membered lactones can be utilized for the kinetic resolution
only if they are substituted at the 6-position. Fellous and co-workers have reported the
hydrolysis of 6-substituted hexanolide by esterase (PLE and HLE).3 They have stated that the
stereoselectivity of the reaction was influenced by the bulkiness of the substituents at the
6-position. Thus, although the ethyl derivative showed high selectivity, the methyl analog
afforded poor selectivity. Until now, there has been no report on enantioselective hydrolysis of
seven-membered lactones which have a chiral carbon at the 3-, 4-, or 5-position.4 In this
communication, we report the lipase-catalyzed kinetic resolution of 3-, 4-, 5-, or 6-methyl
substituted seven-membered lactones.
* Corresponding author. E-mail: shioji@fukuoka-u.ac.jp
†
Present address: Department of Applied Sciences, Fukui University of Technology, 3-6-1 Gakuen, Fukui 910-8505,
Japan.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01421-0

8800
The substrates, racemic methylhexanolides (1a–d), were synthesized with Baeyer–Villiger
oxidation of the corresponding methylcyclohexanones.5 Kinetic resolution of the lactones (1a–d)
with Candida antarctica lipase (CAL) was examined in water-saturated diisopropyl ether (IPE)
and the results are summarized in Table 1.
Table 1
CAL catalyzed hydrolysis of methyl substituted seven-membered lactones 1a
Position of methyl group
Conversion of 1 (%)
Ee of recovered 1 (%)b (configuration)
Ec
53
13
79
1a; 3-
1b; 4-
1c; 5-
1d; 6-
47
50
50
50
82 (R)
72 (R)
92 (S)
94 (R)
\100
a Lactone; 8 mg, CAL; 2 mg, in 2 ml of IPE/H2O at 36°C for 24 h.
b Determined by GC (CYCLODEX b 236M 25 m×0.25 mm).
c Calculated from conversion and substrate’s ee value.
Thus, in the reaction of 1b, the (S)-isomer reacted faster than the (R)-isomer.6 In the
resolution of lactone 1d, the lipase also has the (S)-preferency. Previously, PLE-catalyzed
hydrolysis of 1d was reported,3,7 and we have found that the stereoselectivity of CAL is better
than that of PLE. The oxidation of 3-methylcyclohexanone afforded a 1:1 mixture of 1a and 1c.
Attempted isolation of 1a and 1c by distillation, HPLC, and GPC was unsuccessful. Since the
separation of 1a and 1b was difficult, CAL-catalyzed hydrolysis of the mixture was examined.
The reaction afforded the hydroxyhexanoic acid (S)-2a and (R)-2c along with the recovered two
enantiomers, (R)-1a and (S)-1c. The absolute configurations of the recovered 1a and 1c were
determined by comparing their retention times of GC with those of (R)-3- and (R)-5-methylhex-
anolide derived from commercially available (R)-3-methylcyclohexanone.8 The stereoselectivity
in the hydrolysis of 1b using CAL was lower than that of hydrolysis of 1a and 1c. Only the
(R)-isomer of 1c reacts faster than the (S)-isomer while the rest of lactones show (S)-selectivity.
The kinetic resolution with CAL on a preparative scale provided an enantiomerically pure 1.5
g of (R)-1a and (S)-1c, starting from 4.0 g of the mixture of 1a, c (800 ml IPE/H2O, 1.0 g CAL).
The resolution of the mixture of (R)-1a and (S)-1c with PLE afforded (R)-2a and (S)-1c.9
Compounds (R)-1a and (S)-1c were also separated by using extrelut® as a solid acid to give
methyl ester (S)-3c and (R)-1a (Scheme 1).10

8801
Scheme 1. (a) With CAL in Pri2O/H2O at 36°C. (b) With PLE in phosphate buffer (pH 7.4) at 36°C. (c) In the
presence of extrelut® in MeOH at 30°C for 4 h
Thus, the resolution with CAL resulted in the first successful synthesis of optically active 3-,
4-, or 5-methyl substituted lactones in a gram scale. These lactones are of great potential value
as synthons for the synthesis of natural products.
References
1. Theil, F. Chem. Rev. 1995, 95, 2203; Makita, A.; Nihira, T.; Yamada, Y. Tetrahedron Lett. 1987, 28, 805;
Gutman, A. L.; Zuobi, K.; Boitansky, A. Tetrahedron Lett. 1987, 28, 3861.
2. Blanco, L.; Guibe-Jampel, E.; Rousseau, G. Tetrahedron Lett. 1988, 29, 1915.
3. Fellous, R.; Cuvelier, L. L.; Loiseau, M. A.; Sassy, E. Tetrahedron: Asymmetry 1994, 5, 343.
4. Lipase-catalyzed hydrolysis of an autocatalytic oxidation product of cyclic ketones was reported. However, the
stereoselectivity of the reaction was poor. Pchelka, B. K.; Gelo-Pujic, M.; Guibe-Jampel, E. J. Chem. Soc., Perkin
Trans. 1 1998, 2625.
5. The oxidation of 4-methylcyclohexanone afforded 4-methylhexanolide (1b). Yield 65%, 1H NMR (CDCl3)
l=4.31–4.15 (m, 2H), 2.71–2.58 (m, 2H), 1.96–1.85 (m, 2H), 1.82–1.73 (m, 1H), 1.56–1.46 (m, 1H), 1.00 (d,
J=6.4 Hz, 3H); 13C NMR (CDCl3) l=175.78, 69.26, 42.01, 37.23, 29.41, 27.89, 22.17. The oxidation of
2-methylcyclohexanone gave 6-methylhexanolide (1d) contaminated by 2-methylhexanolide (5%). We applied this
mixture to CAL-catalyzed kinetic resolution without separation. The oxidation of 3-methylcyclohexanone gave
1
3-methylhexanolide (1a) and 5-methylhexanolide (1c) (1:1). Total yield 65%, ( )-1a; H NMR (CDCl3) l=4.29–
4.16 (m, 2H), 2.65–2.52 (m, 2H), 1.85–1.76 (m, 3H), 1.74–1.58 (m, 1H), 1.49–1.36 (m, 1H), 1.10 (d J=6.8 Hz,
3H); 13C NMR (CDCl3) l=175.13, 69.26, 42.01, 37.23, 29.41, 27.89, 22.17. ( )-1c; 1H NMR (CDCl3)
l=4.11–3.96 (m, 2H), 2.65–2.57 (m, 2H), 1.94–1.88 (m, 3H), 1.68–1.62 (m, 1H), 1.37–1.36 (m, 1H), 0.97 (d,
J=6.4 Hz, 3H); 13C NMR (CDCl3) l=176.00, 73.99, 36.36, 34.42, 33.85, 21.64, 17.74.
6. The absolute configuration of the unreacted lactone 1b was determined by comparison of the sign of the optical
rotation with that reported. A pure enantiomer (R)-1b was obtained at a conversion of 72%. [h]2D5 +44.9 (c 0.99,
CHCl3), ee>98%. Kayser, M.; Chen, G.; Stewart, J. Synlett 1999, 153; Stewart, J. D.; Reed, K. W.; Martinez, C.
A.; Zhu, J.; Chen, G.; Kayser, M. M. J. Am. Chem. Soc. 1998, 120, 3541.

8802
7. A small amount of 2-methylhexanolide was not hydrolyzed under these reaction conditions.
8. Retention times of GC with those enantiomers of ( )-1a and ( )-1c were 39.6, 42.0, 43.3, and 43.8 min,
respectively. Retention times of GC with (R)-1a and (R)-1c, derived from 3-methylcyclohexanone, were 40.6 and
42.4 min, respectively.
9. A mixture of (R)-1a and (S)-1c (1.0 g) was dissolved in 250 ml of phosphate buffer (pH 7.4), 1.0 g of PLE was
added and the mixture was stirred at 36°C for 3 h. The PLE was removed by filtration with extrelut®. The crude
residue was purified by silica gel chromatography. Elution with CH2Cl2 afforded (S)-1c in 35% yield. [h]2D5 +41.3
(c 0.35, CHCl3). The optical rotation of (S)-1c differed from the literature value in Ref. 9, while the ee (>98%)
and configuration of (S)-1c determined in this experiment were confirmed by GC analysis.
10. Powdered extrelut® (25 g) was added to a solution of (R)-1a and (S)-1c (1.0 g) in CH3OH (1000 ml) at 30°C. The
resulting mixture was stirred for 4 h. After the removal of extrelut® by filtration, the reaction mixture was
concentrated in vacuo. The crude residue was purified by silica gel flash column chromatography. Elution with
CH2Cl2 afforded (R)-1a and methyl ester (S)-3c. (R)-1a; yield 30% (based on the total amount of the mixture).
[h]2D5 −15.1 (c 0.35, CHCl3). (S)-3c was contaminated by (R)-3a (23%).
.