Effect of substituent on the enantioselectivity for lipase-catalyzed kinetic resolution of glycerol derivatives

Article abstract of DOI:10.1016/j.tet.2004.10.107

The lipase-catalyzed transesterifications of various substituted diphenyl 1,2-ketals of glycerol have been investigated. Efficient modification of the substrate structure with bis(4-bromophenyl) ketal was found to enhance the enantioselectivity up to E=57

Full text of DOI:10.1016/j.tet.2004.10.107

Tetrahedron 61 (2005) 693–697
Effect of substituent on the enantioselectivity for lipase-catalyzed
kinetic resolution of glycerol derivatives
*
Yasuhiro Kawanami, Akira Honnma, Kayo Ohta and Naoko Matsumoto
Department of Biochemistry and Food Science, Faculty of Agriculture, Kagawa University, Miki-cho, Kagawa 761-0795, Japan
Received 9 September 2004; accepted 26 October 2004
Available online 19 November 2004
Abstract—The lipase-catalyzed transesterifications of various substituted diphenyl 1,2-ketals of glycerol have been investigated. Efficient
modification of the substrate structure with bis(4-bromophenyl) ketal was found to enhance the enantioselectivity up to EZ57 at 0 8C.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
dioxolane-4-methanol with a variety of 4-substituted phenyl
groups in place of the methyl group could improve the
enantioselectivity (Fig. 1).
Chiral glycerol derivatives are useful as chiral building
blocks for the syntheses of biologically active chiral
compounds such as b-blockers1 and PAF antagonists.2
Although the efficient asymmetric syntheses of glycerol
derivatives were achieved by the lipase-catalyzed trans-
esterification of racemic 3-(aryloxy)-1,2-propanediols with
lipase PS from Pseudomonas cepacia (PCL) and vinyl
acetate,3 the esterification of various 1,2-ketals of glycerol
with succinic anhydride4 and the hydrolysis of carboxylic
esters of the 1,2-ketals of glycerol5 were reported to proceed
with low enantioselectivities. These results are consistent
with the reported finding that the enantioselectivity of PCL
toward primary alcohols is much lower than toward
secondary alcohols.6 However, Sakai et al. described
that the E value for the kinetic resolution of 2,2-dimethyl-
1,3-dioxolane-4-methanol 1 using lipase AK from Pseudo-
monas fluorescens (PFL) increased up to 55 (9 at 30 8C) by
lowering the reaction temperature to K40 8C in i-Pr2O,7
although it needed a large amount of lipase and longer
reaction time. As an alternative approach, we investigated
the optimization of the structures of the glycerol derivatives
by altering the protecting group because we recently found
that the enantioselectivity during transesterification cata-
lyzed by the immobilized PCL significantly depended on
the position of the substituent on the aromatic ring, and the
N-3,5-dimethylbenzyl group was found to transform the
trans-2,5-substituted pyrrolidine derivative into an effi-
ciently-resolved substrate.8 Thus, we expected that 1,3-
Figure 1.
In this paper, we describe the study of the substituent effect
on the enantioselectivity for the lipase-catalyzed trans-
esterification of various 2,2-bis(4-substituted phenyl)-1,3-
dioxolane-4-methanols 2a–f.
2. Results and discussion
The substrates 2a–f were prepared by the cyclic ketal
formation of glycerol and the corresponding 4-substituted
benzophenones in the presence of p-toluenesufonic acid at
reflux using a Dean–Stark apparatus as previously
described5 (Scheme 1). To investigate the solvent effect,
the substrate 2e was selected as a suitable substrate and the
transesterifications catalyzed by lipase PS (PCL) and AK
(PFL) were carried out in various solvents at 25 8C
(Scheme 2). The enantiomeric excess (ee) values of alcohols
2e and the resulting acetates 3e were determined by an
Keywords: Lipase; Transesterification; Kinetic resolution; Glycerol;
Substituent effect.
* Corresponding author. Tel.: C81 87 891 3088; fax: C81 87 891 3021;
e-mail: kawanami@ag.kagawa-u.ac.jp
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.10.107

694
Y. Kawanami et al. / Tetrahedron 61 (2005) 693–697
examined using a series of substrates 2a–f and vinyl acetate
in dry hexane at 25 8C (Table 2).
Although the enantioselectivity of lipase AK for the
substrates 2a–c decreased as the size of the electron-
donating group increased (Entries 1–3), those of the
substrates 2d–f increased as the size of the electron-
withdrawing halogen group on the benzene ring far from
the stereocenter increased (Entries 4–6). On the other hand,
the enantioselectivity of lipase PS for the substrates 2a–c
and 2d–f increased as the size of both substituent groups
increased (Entries 7–12). Among them, the substrate 2f
(RZBr) showed the highest enantioselectivity (EZ36).
Thus, bis(4-bromophenyl) ketal as the protecting group was
found to transform glycerol into an efficiently-resolved
substrate.
Scheme 1. Preparation of glycerol derivatives.
HPLC analysis using a chiral column. The absolute
configurations of the remaining alcohols were determined
to be R-form by comparison of the optical rotations, which
were consistent with the empirical rule for chiral primary
alcohols.6 These results are summarized in Table 1.
Finally, the temperature effect on the enantioselectivity was
examined using lipase PS in dry hexane at 0 8C as shown in
Figure 2. Although the E values of the substrates 2a, 2b, and
2d (RZH, Me, F) did not change at 0 8C, those of substrates
2e and 2f bearing an electron-withdrawing group (RZCl,
Br) increased from 21 to 25 and from 36 to 57, respectively.
On the other hand, the E value decreased for substrate 2c
(RZOMe). This result suggests that the inversion of the
temperature effect might be due to the strong electron-
donating character of the methoxy group in the diphenyl
ketals 2c. It is noted that this effect of the substituent and
temperature is quite opposite to that of the esterification of
2-(4-substituted phenoxy)propionic acids catalyzed by
lipase MY from Candida rugosa (EZ120, the substrate
Acylation of substrate 2e with vinyl butyrate and lipase AK
in i-Pr2O afforded a slightly higher enantioselectivity (EZ
11) than that of 1 (EZ9)7 under the similar reaction
conditions. Use of vinyl acetate as an acylating agent
slightly increased the enantioselectivity (EZ18). Dry
hexane was found to be a better solvent (EZ18) than wet-
hexane for the kinetic resolution of the substrate 2e using
lipase AK. In the case of lipase PS, the similar solvent effect
was observed and dry hexane was also proved to be the
optimal solvent (EZ21) among the solvents examined.
Next, the substituent effect on the enantioselectivity was
Scheme 2. Lipase-catalyzed transesterification of glycerol derivatives.
Table 1. Solvent effect on the enantioselectivity in lipase-catalyzed acylations of 2ea
Entry
Solvent
Lipase
Time (h)
ees(%)b
eep(%)b
Convn.(%)c
E valuec
1
2
3
4
5
6
7
8
i-Pr2Od
i-Pr2O
THF
Toluene
Wet-hexanee
Dry-hexanef
i-Pr2O
AK
AK
AK
AK
AK
AK
PS
PS
PS
PS
PS
8
3.5
7
6
2
2
5
8
61
71
63
48
48
69
33
15
35
76
67
71
79
63
61
44
94
62
56
70
69
83
46
47
30
44
65
58
35
21
33
52
45
11
18
6.0
7.0
7.0
18
5.9
4.1
8.0
10
THF
9
10
11
Toluene
Wet-hexanee
Dry-hexanef
8.5
1.5
4
21
a Reaction conditions: substrate (0.2 mmol), vinyl acetate (0.8 mmol), lipase (60 mg), solvent (2 ml), 25 8C.
b Determined by HPLC analysis using chiralcel OD column. ees; (R)-2, eep; (S)-2-acetate.
9
c Calculated using the equation in Ref. .
d Vinyl butyrate was used.
e Water-saturated hexane.
f
Distilled from sodium.

Y. Kawanami et al. / Tetrahedron 61 (2005) 693–697
695
Table 2. Subustituent effect on the enantioselectivity in lipase-catalyzed acylations of glycerol derivatives 2a–fa
Entry
Substrate, R
Lipase
Time (h)
ees(%)b
eep(%)b
Convn. (%)c
E valuec
1
2
3
4
5
6
7
8
2a, H
AK
AK
AK
AK
AK
AK
PS
PS
PS
PS
PS
1.5
2
4
2
2
2
6
4
4
95
49
65
83
69
86
49
41
67
53
67
81
68
69
60
42
94
77
58
74
83
72
83
87
58
41
52
66
58
53
46
36
45
42
45
48
20
9.1
7.8
5.9
18
22
6.0
10
21
10
21
36
2b, 4-Me
2c, 4-OMe
2d, 4-F
2e, 4-Cl
2f, 4-Br
2a, H
2b, 4-Me
2c, 4-OMe
2d, 4-F
2e, 4-Cl
2f, 4-Br
9
10
11
12
4
4
2
PS
a Reaction conditions: substrate (0.2 mmol), vinyl acetate (0.8 mmol), lipase (60 mg), dry-hexane (2 ml), 25 8C.
b Determined by HPLC analysis using chiralcel OD or OD-H column. ees; (R)-2, eep; (S)-3.
c Calculated using the equation in Ref. 9.
bearing a methoxy group, 57 8C).10 Figure 3 shows the
correlation between the enantioselectivity of 2a–f and the
Hammett sp value,11 implying that the electron-withdraw-
ing character might be the main factor to enhance the
enantioselectivity. Figure 4 shows the correlation between
the enantioselectivity and van der Waals radius.12 The
enantioselectivity of 2b bearing the larger methyl group was
much lower than that of 2e,f (RZCl, Br), which suggests
that the electronic factor play a more important role than the
steric factor in this PCL-catalyzed transesterification of the
glycerol derivatives. On the other hand, a good correlation
between the enantioselectivity and reactivity was observed
as shown in Figure 5. These results suggest that the
hydrophobic interaction between the bromo group and the
active site of PCL allowed the fast-reacting substrate to
strongly bind, leading to the enhancement of both the
reactivity and enantioselectivity.
Schrag et al. reported that the active-site cleft of PCL based
Figure 3. Relation between the enantioselectivity and Hammett sp.
˚
on an X-ray structure analysis was ovoid (10 A!25 A
˚
˚
˚
across) and was about 15 A deep and 4 A wide at the
bottom.12 Figure 6 shows a cutaway view of the active-site
cleft (Figure 2(d) in the literature12) and the supposed
transition-state model for the fast-reacting (S)-bis(4-bromo-
phenyl) ketal of glycerol. Although the detailed computer
modeling of the X-ray crystal structure of PCL and
conformations of the ketal have not yet been performed,
the methylene group of dioxalane seems to nestle into the
smaller hydrophobic pocket and one of the 4-bromophenyl
groups seems to point toward the alternate hydrophobic
pocket.6 Therefore, this overlapping also supported the
previous proposition that the additional hydrophobic
interaction between the bromo group and the active-site
cleft might enhance the reactivity and enantioselectivity.
Figure 4. Relation between the enatioselectivity and van der Waals radius.
Figure 2. Temperature effect.

696
Y. Kawanami et al. / Tetrahedron 61 (2005) 693–697
using a JEOL JMS-SX102A mass spectrometer. The optical
rotations were determined using a JASCO P-1010 polari-
meter. The HPLC analysis was carried out using DAICEL
Chiralcel OD and OD-H columns (0.46!25 cm) with a
Shimadzu LC6A. TLC was carried out on Merck glass
plates precoated with silica gel 60F-254 (0.25 mm) and the
column chromatography was performed using Merck 23-
400 mesh silica gel. All reagents were purchased from
Wako Chemical Co., Tokyo Kasei Kogyo Co., Lancaster
Synthesis, Ltd. or Aldrich Chemical.
4.2. Typical procedure for the synthesis of 1,2-ketal of
glycerols
4.2.1. 2,2-Bis(4-methylphenyl)-1,3-dioxolane-4-methanol
(2b). A mixture of glycerol (0.759 g, 8 mmol), 4,40-
dimethylbenzophenone (1.75 g, 8.2 mmol), and p-toluene-
sulfonic acid (18.2 mg) in toluene (5 ml) was refluxed for
24 h. After cooling, the reaction mixture was quenched with
0.1 M NaOH and extracted with ether. The combined
organic layer was washed with saturated NaHCO3 solution,
dried over MgSO4, and concentrated. The residue was
washed with hexane and chromatographed (hexane/ethyl
acetate, 3:1) to give a viscous oil (1.52 g, 64%); IR (neat)
3465 cmK1; 1H NMR (CDCl3) d 1.80 (1H, br s), 2.30 (3H,
s), 2.32 (3H, s), 3.62 (1H, dd, JZ5.1, 11.7 Hz), 3.78 (1H,
dd, JZ3.4, 11.7 Hz), 3.95 (1H, dd, JZ6.1, 8.1 Hz), 4.00
(1H, t, JZ8.1 Hz), 4.29 (1H, m), 7.10 (2H, d, JZ8.0 Hz),
7.13 (2H, d, JZ7.8 Hz), 7.34 (2H, d, JZ8.0 Hz), 7.40 (2H,
d, JZ7.8 Hz); 13C NMR (CDCl3) d 21.1, 63.2, 66.1, 76.8,
110.2, 125.9, 126.1, 128.8, 137.7, 137.9, 139.1, 139.2; EI
HRMS C18H20O3 (MC) 284.1412. Found 284.1452.
Figure 5. Relation between the enatioselectivity and reactivity. Relative
reaction rate was calculated by comparing each (convn./time) of 2b–f with
that of 2a (RZH).
4.2.2. 2,2-Bis(4-methoxyphenyl)-1,3-dioxolane-4-metha-
nol (2c). Yield 36%; IR (neat) 3460 cmK1 1H NMR
;
(CDCl3) d 1.80 (1H, br s), 3.76–3.82 (2H, m), 3.78 (3H, s),
3.80 (3H, s), 3.96 (1H, dd, JZ6.4, 8.0 Hz), 4.00 (1H, t, JZ
8.0 Hz), 4.30 (1H, m), 6.83 (2H, d, JZ8.4 Hz), 6.87 (2H, d,
JZ8.4 Hz), 7.36 (2H, d, JZ8.4 Hz), 7.42 (2H, d, JZ
8.0 Hz); 13C NMR (CDCl3) d 55.3, 63.3, 66.1, 76.7, 110.2,
113.5, 127.7, 134.4, 159.4; EI HRMS C18H20O5 (MC)
316.1311. Found 316.1325.
Figure 6. Supposed model of the fast-reacting (S)-enantiomer for PCL and
the cutaway view of the active-site cleft of PCL. Larger hydrophobic pocket
is located far away from the cutaway view and alternate hydrophobic
pocket points toward this side.
3. Conclusion
4.2.3. 2,2-Bis(4-fluorophenyl)-1,3-dioxolane-4-methanol
1
(2d). Yield 38%; IR (neat) 3455 cmK1; H NMR (CDCl3)
In summary, we have demonstrated that the efficient
modification of the substrate structure with bis(4-bromo-
phenyl) ketal as the protecting group enhanced the
enantioselectivity of the lipase-catalyzed kinetic resolution
of glycerol derivatives at 0 8C. This substrate-tuning method
by a protecting group would provide an alternative approach
to improve the lipase-catalyzed kinetic resolution of
primary alcohols.
d 1.81 (1H, dd, JZ5.8, 6.8 Hz), 3.66 (1H, ddd, JZ5.1, 6.8,
12.0 Hz), 3.81 (1H, ddd, JZ3.6, 5.8, 11.7 Hz), 3.98 (1H, dd,
JZ6.1, 8.0 Hz), 4.02 (1H, t, JZ8.3 Hz), 4.31 (1H, m), 6.98–
7.10 (4H, m), 7.40–7.51 (4H, m); 13C NMR (CDCl3) d 63.1,
66.3, 76.7, 109.5, 115.1, 115.3, 128.1, 128.2, 161.4, 163.9;
EI HRMS C16H14F2O3 (MC) 292.0911. Found 292.0870.
4.2.4. 2,2-Bis(4-chlorophenyl)-1,3-dioxolane-4-methanol
(2e). Yield 42%; IR (neat) 3460 cmK1; 1H NMR (CDCl3) d
1.76 (1H, dd, JZ5.2, 6.4 Hz), 3.65 (1H, ddd, JZ5.6, 6.4,
12.0 Hz), 3.80 (1H, ddd, JZ4.0, 5.2, 11.6 Hz), 3.98 (1H, dd,
JZ6.1, 8.1 Hz), 4.02 (1H, dd, JZ7.1, 8.1 Hz), 4.31 (1H, m),
7.34–7.25 (4H, m), 7.47–7.37 (4H, m); 13C NMR (CDCl3) d
62.6, 65.9, 76.7, 108.9, 127.0, 128.1, 133.9, 139.8; EI
HRMS C16H14Cl2O3 (MC) 324.0320. Found 324.0304.
4. Experimental
4.1. General
The IR spectra were determined using a JASCO A 302 FT-
1
IR spectrometer. The H NMR and 13C NMR spectra were
recorded at 400 and 100 MHz using a JNM-A400
spectrometer, respectively. The mass spectra were recorded
4.2.5. 2,2-Bis(4-bromophenyl)-1,3-dioxolane-4-methanol
(2f). Yield 43%; IR (neat) 3460 cmK1; 1H NMR (CDCl3) d

Y. Kawanami et al. / Tetrahedron 61 (2005) 693–697
697
1.70 (1H, dd, JZ5.8, 6.8 Hz), 3.65 (1H, ddd, JZ5.1, 6.8,
Acknowledgements
12.0 Hz), 3.80 (1H, ddd, JZ3.6, 5.8, 12.0 Hz), 3.98 (1H, dd,
JZ6.4, 8.3 Hz), 4.02 (1H, dd, JZ7.6, 8.3 Hz), 4.31 (1H, m),
7.40–7.31 (4H, m), 7.49–7.43 (4H, m); 13C NMR (CDCl3) d
63.1, 66.4, 77.2, 109.4, 122.7, 127.8, 128.0, 131.6, 131.7,
140.8; EI HRMS C16H14Br2O3 (MC) 413.9310. Found
413.9329.
We are grateful to the Amano Pharmaceutical Co. for
providing the lipase Amano PS and AK, and Dr. Schrag for
providing a colored reprint.
References and notes
4.3. Typical procedure for lipase-catalyzed tranesteri-
fication of 2a with vinyl acetate
1. Tsuda, Y.; Yoshimoto, K.; Nishikawa, T. Chem. Pharm. Bull.
1981, 29, 3593. Terao, Y.; Murata, M.; Achiwa, K.; Nishio, T.;
Akamatsu, M.; Kamimura, M. Tetrahedron Lett. 1988, 29,
5173. Wang, Y. F.; Wong, C.-H. J. Org. Chem. 1988, 53, 3127.
Chem, C.-S.; Sih, C. J. Angew. Chem., Int. Ed. Engl. 1989, 28,
695.
Vinyl acetate (74 ml, 0.8 mmol) was added to a suspension
of the 1,2-ketal of glycerol 2a (25.6 mg, 0.1 mmol) and
lipase (60 mg) in hexane (2 ml) and the mixture was stirred
at 25 8C for 6 h. The reaction was monitored by TLC. The
reaction mixture was filtered off on celite and washed with
dichloromethane. The filtrate was evaporated under reduced
pressure. Flash column chromatography (hexane/ethyl
acetate 3:2) of the residue afforded the 2a-acetate (HPLC
2. Shiraiwa, M.; Fujita, K.; Yosiwara, H.; Kobayashi, S.; Ohno,
M. J. Synth. Org. Chem. Jpn 1987, 45, 369. Takano, S.;
Yanase, M.; Sekiguchi, Y.; Ogasawara, K. Tetrahedron Lett.
1987, 28, 1783.
with Chiralcel OD, hexane/i-PrOH 97:3, 0.7 ml/minK1
,
3. Thei, F.; Weidner, J.; Ballschuh, S.; Kunath, A.; Schick, H.
J. Org. Chem. 1994, 59, 388.
retention time; 14 and 16 min for (R)- and (S)-enantiomer)
and the alcohol 2a (HPLC with Chiralcel OD, hexane/
i-PrOH 97:3, 0.7 ml/minK1, retention time; 32 and 37 min
for (R)- and (S)-enantiomer). To establish the absolute
configuration of the remaining alcohol 2a, its optical
rotation sign was compared with the reported value; (R)-
(K)-2a; [a]2D1 K11.5 (c, 0.52, MeOH) (lit.,5 [a]2D0 C22.5 (c,
0.36, MeOH) for (S)-isomer). The other absolute configur-
ations of the remaining alcohols 2b-f were assigned as (R)
by comparison of the sign of the optical rotations and the
order of retention times on an HPLC chiral column with that
of (R)-(K)-2a. Moreover, those of the acetates were
confirmed by transformation (10% acetic acid) into (S)-
(C)-1-O-acetylglycerol.13
4. Terao, Y.; Tsuji, K.; Murata, M.; Achiwa, K.; Nishio, T.;
Watanabe, N.; Seto, K. Chem. Pharm. Bull. 1989, 37, 1653.
5. Partali, V.; Melbye, A. G.; Alvik, T.; Anthonsen, T.
Tetrahedron Asymmetry 1992, 3, 65.
6. Weissfloch, A. N. E.; Kazlauskas, R. J. J. Org. Chem. 1995,
60, 6959. Tuomi, W. V.; Kazlauskas, R. J. J. Org. Chem. 1999,
64, 2638.
7. Sakai, T.; Kishimoto, T.; Tanaka, Y.; Ema, T.; Utaka, M.
Tetrahedron Lett. 1998, 39, 7881.
8. Kawanami, Y.; Iizuna, N.; Okano, K. Chem. Lett. 1998, 1231.
Kawanami, Y.; Iizuna, N.; Maekawa, K.; Takahasi, N.;
Kawada, T. Tetrahedron 2001, 57, 3349.
9. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am.
Chem. Soc. 1982, 102, 7294.
For 2b and 2b-acetate, HPLC with Chiralcel OD, hexane/
i-PrOH 97:3, 0.7 ml/minK1; 2c, OD, hexane/i-PrOH
97:3, 0.6 ml/min; 2c-acetate, OD-H, hexane/i-PrOH 97:3,
10. March, J. Advanced Organic Chemistry, Reaction, Mechan-
isms, and Structure, 3rd ed.; Wiley: New York, 1985; p 244.
11. Pauling, L. Nature of Chemical Bond, 3rd ed.; Cornell
University Press: New York, 1960; pp 260–263.
0.7 ml/minK1; 2d, OD, hexane/i-PrOH 96:4, 0.5 ml/minK1
;
2d-acetate, OD-H, hexane/i-PrOH 200:1, 0.4 ml/minK1; 2e,
OD, hexane/i-PrOH 96:4, 0.6 ml/minK1; 2e-acetate, OD-H,
hexane/i-PrOH 200:1, 0.8 ml/minK1; 2f, OD, hexane/
i-PrOH 98:2, 0.6 ml/minK1; 2f-acetate, OD-H, hexane/
12. Schrag, J. D.; Li, Y.; Cygler, M.; Lang, D.; Burgdorf, T.;
Hecht, H.; Schmid, R.; Schomburg, D.; Rydel, T. J.; Oliver,
J. D.; Strickland, L. C.; Dunaway, C. M.; Larson, S. B.; Day,
J.; McPherson, A. Structure 1997, 5, 187.
i-PrOH 200:1, 0.4 ml/minK1
.
13. Baer, E.; Fischer, H. O. L. J. Am. Chem. Soc. 1945, 67, 2031.