Practical resolution of 3-phenyl-2H-azirine-2-methanol at very low temperature by using lipase immobilized on porous ceramic and optimized acylating agent

Article abstract of DOI:10.1246/bcsj.76.1819

A synthetically useful primary alcohol, 3-phenyl-2H-azirine-2-methanol, was resolved efficiently by a reaction at -40 °C with a lipase immobilized on the porous ceramic (Toyonite) and vinyl butanoate in ether (E value of 130 and TTN/h value of 7800).

Full text of DOI:10.1246/bcsj.76.1819

Ó 2003 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 76, 1819–1821 (2003) 1819
Practical Resolution of 3-Phenyl-2H-azirine-2-methanol at Very Low
Temperature by Using Lipase Immobilized on Porous Ceramic and
Optimized Acylating Agent
ꢀ
Takashi Sakai, Akiko Matsuda, Toshinobu Korenaga, and Tadashi Ema
Department of Applied Chemistry, Faculty of Engineering, Okayama University,
3-1-1 Tsushima-naka, Okayama 700-8530
Received March 31, 2003; E-mail: tsakai@cc.okayama-u.ac.jp
A synthetically useful primary alcohol, 3-phenyl-2H-azirine-2-methanol, was resolved efficiently by a reaction at
ꢁ40 C with a lipase immobilized on the porous ceramic (Toyonite) and vinyl butanoate in ether (E value of 130 and
ꢂ
TTN/h value of 7800).
We have recently proposed ‘‘the low-temperature method’’
for enhancing the enantioselectivity in the lipase-catalyzed res-
olutions;1,2 this method was found to be especially effective for
a variety of primary alcohols3 involving 3-phenyl-2H-azirine-
2-methanol (1), a versatile synthetic intermediate.4 Among
many fine-tuning methods for enhancing the enantioselectivi-
ty,5 the low-temperature method is a theoretically reliable
method3 as usual chemical asymmetric reactions. After we
proposed the method, it has been frequently utilized.6 In pre-
vious papers,1,2 we described that Celite-immobilized lipase-
catalyzed resolution of azirine (ꢃ)-1 with vinyl acetate in ether
provement of the E value by simply lowering the reaction tem-
perature is obvious by comparison with the results for the re-
action at 30 C as shown in entry 2.
ꢂ
In order to adjust the two conflicting features, the reaction
rate and the enantioselectivity, the length of the acylating
agent was then examined. Elongation of the acyl chain to vi-
nyl butanoate (3b) (entry 4) dramatically increased the E value
up to 96, which is comparable to the value obtained by using
the Celite-immobilized lipase (entry 1), and gave a sufficient
TTN/h value of 3200. In contrast, a similar reaction using vi-
nyl hexanoate (3c) somewhat lowered the E value (54) and the
reaction rate (entry 5). Interestingly, use of two molar
amounts of 3c (entry 6) not only raised the TTN/h value but
also raised the E value up to 102. Therefore, two molar
amounts of the acylating agents were used in the following re-
actions with longer acylating agents 3d–3g. Further elonga-
tion of the chain to vinyl octanoate (3d), however, suddenly
decreased the E value to 35 (entry 7). In the case of vinyl do-
decanoate (3f), both E value and TTN/h value were slightly
recovered unexpectedly (entry 9). Vinyl 3-phenylpropanoate
(3g), which is a suitable acylating agent for primary alcohols
as reported by Kawasaki et al.,10,11a also gave a good E value
of 71 and an acceptable TTN/h value (entry 10). Vinyl chlo-
roacetate (3h) accelerated the reaction rate markedly, but gave
a low E value (entry 11).
Similar reactions of (ꢃ)-1 using the Toyonite-200M-immo-
bilized lipase and acylating agents 3b and 3c were further
examined. As shown in entry 12, the reaction using the Toyo-
nite-200M-immobilized lipase and 3b further increased both E
value (130) and TTN/h value (7800), which are much higher
than those obtained with the Celite-immobilized lipase (entry
1). Changing from 3b to 3c (entry 13) turned out to decrease
the E value to 77, although TTN/h was the highest (8600). As
the last attempt, the organic bridges were changed to 6-(pro-
panoyloxy)hexylsilanetrioxyl groups, because we have recent-
ly found that this organic bridge gives the results suitable for
the resolution of (ꢃ)-2-hydroxymethyl-1,4-benzodioxane.13
However, it rather decreased the E and TTN/h values for com-
pound 1 (entry 14).
7
ꢂ
at 30 C gave ester (R)-2a and alcohol (S)-1 with an E value
of 17. The reaction rate was roughly estimated by the total
turnover number per hour (TTN/h),8 which was 4700 at 30
ꢂC. In contrast, although lowering the reaction temperature
ꢂ
to ꢁ40 C enhanced the E value up to 99, the TTN/h value
was lowered to 210. Thus, enhancing not only the enantiose-
lectivity but also the activity even at such very low tempera-
tures is an urgent subject for practical application. Here, we
attained the purpose by using both an optimized acylating
agent9–11 and a lipase immobilized on a porous ceramic
(Toyonite)12,13 in the low-temperature reaction of (ꢃ)-1 at
ꢂ
ꢁ40 C.
The porous ceramic support (Toyonite) is a new type of
commercially available inorganic material for immobilization
of enzymes. There are two types: Toyonite D-M and Toyonite
200M,14 on which we immobilized lipase PS. In the present
reactions, the temperature was kept at ꢁ40 ꢂC, the temperature
that gave the highest E value for 1.1 Results of the reactions
using the Toyonite-immobilized lipase and various acylating
agents are summarized in Table 1. As shown in entry 3, use
of the Toyonite D-M-immobilized lipase and 3a remarkably
raised the TTN/h to 4200, 20 times larger than that of the re-
action catalyzed by the Celite-immobilized lipase (entry 1).
The immobilization of the lipase on Toyonite greatly enhanced
the reaction rate, probably due to the high dispersion of lipase
molecules on the highly porous support. However, it lowered
the E value from 99 to 34. The immobilization may restrict
the required conformational flexibility of the lipase.15 Im-
Published on the web September 15, 2003; DOI 10.1246/bcsj.76.1819

1820 Bull. Chem. Soc. Jpn., 76, No. 9 (2003)
Table 1. Toyonite-Immobilized Lipase-Catalyzed Resolution of (ꢃ)-1 at ꢁ40 C
Lipase Reaction at Very Low Temperature
ꢂ
Entry
Acylating agents
LipaseaÞ
Time
/h
% Ee
Conv.
E
TTN/hbÞ
R
(R)-2
(S)-1
1cÞ
2eÞ
3
4
5
6
7
8
9
10
11
12
13
14
CH3
CH3
CH3
3a (1)dÞ
3a (1)
3a (1)
3b (1)
3c (1)
3c (2)
3d (2)
3e (2)
3f (2)
3g (2)
3h (1)
3b (1)
3c (2)
3b (2)
A
B
B
B
B
B
B
B
B
B
B
C
C
D
4.3
0.4
5.2
6.0
97fÞ
63fÞ
86fÞ
96fÞ
95fÞ
91fÞ
90gÞ
91fÞ
89fÞ
95hÞ
67hÞ
97fÞ
93fÞ
95fÞ
46fÞ
52
81
71
33
99
62
63
81
62
37
52
85
58
0.32
0.46
0.49
0.43
0.26
0.52
0.41
0.41
0.47
0.40
0.36
0.35
0.44
0.37
99
7
210
50000
4200
3200
1600
2000
1500
1500
1800
3600
5300
7800
8600
1500
34
96
54
102
35
40
44
71
7
CH3(CH2)2
CH3(CH2)4
CH3(CH2)4
CH3(CH2)6
CH3(CH2)8
CH3(CH2)10
PhCH2CH2
ClCH2
7.3
12.0
11.8
11.0
12.0
8.0
3.0
2.0
2.4
11.0
CH3(CH2)2
CH3(CH2)4
CH3(CH2)2
130
77
68
a) A: Lipase PS on Celite; B: Lipase PS immobilized on Toyonite D-M with 3-(methacryloyloxy)propylsilanetrioxyl
bridges; C: Lipase PS immobilized on Toyonite 200M with 3-(methacryloyloxy)propylsilanetrioxyl bridges; D: Lipase
PS immobilized on Toyonite 200 with 3-(propanoyloxy)hexylsilanetrioxyl bridges.
ꢂ
b) See Ref. 8. c) Reported results in Ref. 1. d) Molar amounts of acylating agent. e) Reaction at 30 C. f)
Analyzed by Chiralcel OB-H. g) Analyzed by Chiralcel AS-H. h) Analyzed by Chiralpak OD-H.
(1.5 g), and the mixture was shaken for 24 h at 25 ꢂC. The result-
In conclusion, these results suggest that the choice of acylat-
ing mixture was filtered, and the Toyonite-immobilized lipase was
dried under vacuum (267 Pa) for 15 h and stored in a desiccator.
Protein wt % was determined to be about 1.3 wt % by the differ-
ence between the amounts of the protein in the aqueous solutions
before and after immobilization.16
ing agent and the use of the Toyonite-immobilized lipase are
critically important in the practical operation of the low-tem-
perature reaction. A matched combination of these techniques
will further enhance the utility of the low-temperature
method.
Typical Lipase-Catalyzed Resolution of 3-Phenyl-2H-azi-
rine-2-methanol ((Æ)-1) (entry 12). A mixture of (ꢃ)-1 (50
mg, 0.34 mmol), Toyonite-200M-immobilized lipase PS (25
mg), and 5 mL of ether was cooled to ꢁ40 ꢂC in a thermo-control-
led cooling bath. Acylating agent 3b (39 mg, 0.34 mmol) was in-
jected quickly to the cooled mixture through a syringe. The reac-
tion progress was monitored by TLC (hexane/ethyl acetate = 2/
1). The mixture was filtered with suction to remove the lipase.
The filtrate was concentrated, and the residual products were sep-
arated by column chromatography (silica-gel 5 g, hexane/ethyl
Experimental
Lipase PS (Amano Enzyme Inc.) was purchased from Wako
Pure Chemical Co. Ltd. Ether was distilled from sodium before
use. Preparative column chromatography was carried out by using
silica gel (Fuji Silysia BW-127 ZH, 100–270 mesh), and thin-
layer chromatography was performed by using precoated silica-
gel plate (Merck 60 PF254, plate length 40 mm). 1H NMR spectra
were measured at 200 MHz. Enantiomeric purities were deter-
mined by HPLC analyses detected at 254 nm with a chiral column
(Chiralpak OB-H, AS-H, or OD-H, Daicel Chemical Industries)
as indicated in Table 1.
Immobilization of Lipase to the Ceramic Support (Toyonite)
with Organic Bridges.13 A suspension of commercially availa-
ble lipase PS immobilized on Celite (15 g) in phosphate buffer (60
mL, pH 7.0, 10 mM) was stirred at room temperature for 3 h. Af-
ter filtration, to the filtrate was added Toyonite (D-M or 200M)
27
acetate = 3/1–1/1). (S)-1: ½ꢀꢄD ¼ þ123 (c 1.00, CHCl3); IR
(KBr) 3333, 1736 cmꢁ1 1H NMR (CDCl3) ꢁ 2.15 (br s, 1H),
;
2.48 (dd, J ¼ 5:1, 2.2 Hz, 1H), 3.69 (dd, J ¼ 10:2, 5.1 Hz, 1H),
3.99 (dd, J ¼ 10:2, 2.2 Hz, 1H), 7.49–7.58 (m, 3H), 7.86–7.90
27
(m, 2H). (R)-2b: ½ꢀꢄD ¼ ꢁ160 (c 0.68, CHCl3); IR (KBr) 1736
cmꢁ1
;
1H NMR (CDCl3) ꢁ 0.90 (t, J ¼ 3:7 Hz, 3H), 1.60 (m,
2H), 2.27 (t, J ¼ 7:4 Hz, 2H), 2.50 (dd, J ¼ 5:0, 3.8 Hz, 1H),
4.20 (dd, J ¼ 11:8, 5.0 Hz, 1H), 4.32 (dd, J ¼ 11:8, 3.8 Hz,

T. Sakai et al.
Bull. Chem. Soc. Jpn., 76, No. 9 (2003) 1821
1H) 7.53–7.65 (m, 3H), 7.87–7.90 (m, 2H). The conditions and
the retention times in HPLC analyses (flow rate = 0.5 mL/min,
otherwise indicated) for determination of the optical purity are
as follows: (S)-1: hexane/i-PrOH = 9/1, flow rate = 0.8 mL/
min, (R) 12.5 min, (S) 18.2 min. (R)-2b: hexane/i-PrOH = 20/
1, (R) 22.7 min, (S) 26.0 min.
2
T. Sakai, T. Kishimoto, Y. Tanaka, T. Ema, and M. Utaka,
Tetrahedron Lett., 39, 7881 (1998).
The temperature effect is an inherent phenomenon in the
3
reactions of both primary and secondary alcohols. However, the
secondary alcohols remarkably decrease the reaction rate at the
low temperatures. Therefore, in a viewpoint of practical utiliza-
tion, this method is useful for primary alcohols, which still have
acceptable rate even at the low temperatures.
Data of HPLC Analyses for (R)-2a, (R)-2c–(R)-2h. (R)-2a:
hexane/i-PrOH = 9/1, detection 254 nm, (R) 19.6 min, (S) 21.8
min. (R)-2c: hexane/i-PrOH = 100/1, (R) 29.2 min, (S) 32.2
min. (R)-2d: hexane/i-PrOH = 300/1, (R) 40.8 min, (S) 43.7
min. (R)-2e: hexane/i-PrOH = 300/1, (R) 33.8 min, (S) 37.6
min. (R)-2f: hexane/i-PrOH = 300/1, (R) 25.4 min, (S) 29.0
min. (R)-2g: hexane/i-PrOH = 100/1, flow rate = 0.7 mL/
min, (S) 54.1 min, (R) 57.8 min. (R)-2h: hexane/i-PrOH = 9/
1, flow rate = 0.7 mL/min, (R) 14.4 min, (S) 15.3 min.
4
Review for azirines: F. Palacios, A. M. Ochoa de Retana,
E. Martines de Marigorta, and J. Manuel de los Santos, Org. Prep.
Proced. Int., 34, 219 (2002).
5
and V. Sunjic, Food Technol. Biotechnol., 37, 215 (1999).
´ ´
Review: E. Ljubovic, M. Majeric-Elenkov, A. Avdagic,
ˇ
´
a) U. T. Bornscheuer and R. J. Kazlauskas, ‘‘Hydrolases in
6
Organic Synthesis,’’ Wiley-VCH, Weinheim (1999). b) ‘‘Enzyme
Catalysis in Organic Synthesis,’’ ed by K. Drauz and H.
Waldmann, Wiley-VCH, Weinheim (2002).
Spectral Data.
(R)-2a: IR (neat) 1742 cmꢁ1 1H NMR
,
(CDCl3) ꢁ 2.03 (s, 3H), 2.49 (dd, J ¼ 5:4, 3.7 Hz, 1H), 4.18
(dd, J ¼ 12:1, 5.4 Hz, 1H), 4.27 (dd, J ¼ 12:1, 3.7 Hz, 1H),
7.52–7.62 (m, 3H), 7.85–7.89 (m, 2H).
7
C.-S. Chen, Y. Fujimoto, G. Girdaukas, and J. C. Sih, J.
Am. Chem. Soc., 104, 7294 (1982).
Total turnover number (TTN) is the term defined as the
1
(R)-2c: IR (neat) 1735 cmꢁ1, H NMR (CDCl3) ꢁ 0.86 (t, J ¼
8
6:5 Hz, 3H), 1.25 (m, 4H), 1.53 (m, 2H), 2.28 (t, J ¼ 7:7 Hz, 2H),
2.49 (dd, J ¼ 7:5, 4.1 Hz, 1H), 4.23 (dd, J ¼ 10:5, 7.5 Hz, 1H),
4.29 (dd, J ¼ 10:5, 4.1 Hz, 1H), 7.56–7.61 (m, 3H), 7.86–7.90
(m, 2H).
number of mole of product formed per mole of enzyme during
the course of a complete reaction. Here, TTN/h shows the total
turnover number per hour.
9
a) T. Ema, S. Maeno, Y. Takaya, T. Sakai, and M. Utaka,
1
(R)-2d: IR (neat) 1755 cmꢁ1, H NMR (CDCl3) ꢁ 0.87 (t, J ¼
J. Org. Chem., 61, 8610 (1996). b) T. Ema, S. Maeno, Y. Takaya,
T. Sakai, and M. Utaka, Tetrahedron: Asymmetry, 7, 625 (1996).
10 M. Kawasaki, M. Goto, S. Kawabata, and T. Kometani,
Tetrahedron: Asymmetry, 12, 585 (2001).
8:0 Hz, 3H), 1.25 (m, 8H), 1.61 (m, 2H), 2.34 (t, J ¼ 6:6 Hz, 2H),
2.50 (dd, J ¼ 6:8, 4.6 Hz, 1H), 4.21 (dd, J ¼ 8:4, 6.8 Hz, 1H),
4.28 (dd, J ¼ 8:4, 4.6 Hz, 1H), 7.51–7.59 (m, 3H), 7.86–7.90
(m, 2H).
11 Recent reports for the enzymatic resolution with various
acylating reagents: a) M. Kawasaki, M. Goto, S. Kawabata, T.
Kodama, and T. Kometani, Tetrahedron Lett., 40, 5223 (1999).
b) K. Hirose, H. Naka, M. Yano, S. Ohashi, K. Naemura, and
Y. Tobe, Tetrahedron: Asymmetry, 11, 1199 (2000). c) Y. Kita,
Y. Takebe, K. Murata, T. Naka, and S. Akai, J. Org. Chem., 65,
83 (2000). d) J. Ottosson and K. Hult, J. Mol. Catal. B: Enzym.,
11, 1025 (2001). e) T. Itoh, E. Akasaki, and Y. Nishimura, Chem.
Lett., 2002, 154. f) S. Akai, T. Naka, T. Fujita, Y. Takebe, T.
Tsujino, and Y. Kita, J. Org. Chem., 67, 411 (2002).
12 a) M. Kamori, T. Hori, Y. Yamashita, Y. Hirose, and Y.
Naoshima, J. Mol. Catal. B: Enzym., 9, 269 (2000). b) K. Kato,
Y. Gong, T. Saito, and H. Kimoto, J. Biosci. Bioeng., 90, 332
(2000). c) T. Sakai, Y. Miki, M. Tsuboi, H. Takeuchi, T. Ema,
K. Uneyama, and M. Utaka, J. Org. Chem., 65, 2740 (2000). d)
M. Suzuki, C. Nagasawa, and T. Sugai, Tetrahedron, 57, 4841
(2001).
13 T. Sakai, K. Hayashi, F. Yano, M. Takami, M. Ino, T.
Korenaga, and T. Ema, Bull. Chem. Soc. Jpn., 76, 1441 (2003).
14 Toyonite D-M has 120 nm of mean pore size and Toyonite
200M has 60 nm mean pore size. Both of them have 3-(2-meth-
acryloyloxy)propylsilanetrioxyl bridges on the surface. They are
commercially available from Toyo Denka Kogyo Co. Ltd.
Toyonite-immobilized lipase is also available from Amano En-
zyme Inc. or Wako Pure Chemical Industries, Ltd. as Lipase
PS-C ‘‘Amano’’ II.
1
(R)-2e: IR (neat) 1755 cmꢁ1, H NMR (CDCl3) ꢁ 0.88 (t, J ¼
6:8 Hz, 3H), 1.26 (m, 12H), 1.62 (m, 2H), 2.38 (t, J ¼ 7:8 Hz,
2H), 2.51 (dd, J ¼ 7:4, 4.0 Hz, 1H), 4.23 (dd, J ¼ 13:2, 7.4 Hz,
1H), 4.27 (dd, J ¼ 13:2, 4.0 Hz, 1H), 7.56–7.59 (m, 3H), 7.86–
7.90 (m, 2H).
(R)-2f: IR (neat) 1759 cmꢁ1 1H NMR (CDCl3) ꢁ 0.90 (t,
,
J ¼ 6:8 Hz, 3H), 1.28 (m, 16H), 1.64 (m, 2H), 2.40 (t, J ¼ 8:0
Hz, 2H), 2.52 (dd, J ¼ 7:4, 4.6 Hz, 1H), 4.14 (dd, J ¼ 16:6, 7.4
Hz, 1H), 4.27 (dd, J ¼ 16:6, 4.6 Hz, 1H), 7.58–7.61 (m, 3H),
7.88–7.93 (m, 2H).
(R)-2g: IR (KBr) 1738 cmꢁ1 1H NMR (CDCl3) ꢁ 2.49 (dd,
,
J ¼ 9:0, 4.5 Hz, 1H), 2.65 (t, J ¼ 8:2 Hz, 2H), 2.90 (t, J ¼ 8:2
Hz, 2H), 4.22 (dd, J ¼ 11:4, 9.0 Hz, 1H), 4.28 (dd, J ¼ 11:4,
4.5 Hz, 1H), 7.55–7.62 (m, 3H), 7.83–7.87 (m, 2H).
,
(R)-2h: IR (neat) 1740 cmꢁ1 1H NMR (CDCl3) ꢁ 2.54 (dd,
J ¼ 5:4, 3.8 Hz, 1H), 4.06 (s, 2H), 4.30 (dd, J ¼ 13:2, 5.4 Hz,
1H), 4.39 (dd, J ¼ 13:2, 3.8 Hz, 1H), 7.53–7.60 (m, 3H), 7.86–
7.91 (m, 2H).
The authors are grateful to Toyo Denka Kogyo Co. Ltd. for
providing us with Toyonite. We also thank the SC-NMR Lab-
oratory of Okayama University for the measurement of
1H NMR spectra. This work was partially supported by a
Grant-in-Aid for Scientific Research from the Ministry of Ed-
ucation, Culture, Sports, Science and Technology.
15 J. M. Palomo, G. F.-Lorente, C. Mateo, M. Fuentes, R.
F.-Lafuente, and J. M. Guisan, Tetrahedron: Asymmetry, 13,
1337 (2002).
References
1 T. Sakai, I. Kawabata, T. Kishimoto, T. Ema, and M.
Utaka, J. Org. Chem., 62, 4906 (1997).
16 M. Bradford, Anal. Biochem., 72, 248 (1976).