Lipase-catalyzed kinetic resolution of racemic 1-heteroarylethanols-experimental and QM/MM study

Article abstract of DOI:10.1016/j.tetasy.2008.07.004

The kinetic resolution of racemic 1-(benzothiazol-2-yl)ethanol rac-2a, 1-(benzo[b]thiophen-2-yl)ethanol rac-2b, 1-(benzo[b]furan-3-yl)ethanol rac-2c and 1-(benzo[b]thiophen-3-yl)ethanol rac-2d was studied by enantiomer selective acylation catalyzed by a selection of commercially available and in house produced lipases. Alcoholysis of the corresponding racemic acetates rac-3a-d catalyzed by Candida antarctica lipase B (CaLB) was also investigated. Two racemic 1-heteroarylethanols rac-2a,b were prepared by addition of the corresponding lithiated heteroarylic compounds 1a,b to acetaldehyde, whereas two others, rac-2c,d were synthesized by the addition of MeMgI onto the corresponding heteroaryl-carbaldehydes 1c,d. The high enantiomer selectivities of CaLB in the acylation of racemic 1-heteroarylethanols rac-2a-d allowed the determination of the enantiomeric preference of these enzymatic acetylation reactions by QM/MM [pm3/uff or hf(3-21+g**)/uff] calculations. For acetylation of each of the racemic alcohols rac-2a-d, four possible tetrahedral intermediate states were compared and analyzed.

Full text of DOI:10.1016/j.tetasy.2008.07.004

Tetrahedron: Asymmetry 19 (2008) 1844–1852
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Lipase-catalyzed kinetic resolution of racemic
-heteroarylethanols—experimental and QM/MM study
1
a
b
a
a
b
c
Monica To ßs a , Sarolta Pilbák , Paula Moldovan , Csaba Paizs , Gábor Szatzker , György Szakács ,
b
a,
b,
*
*
Lajos Novák , Florin-Dan Irimie , László Poppe
a
Department of Biochemistry and Biochemical Engineering, Babe ßs -Bolyai University of Cluj-Napoca, Ro-400028 Cluj-Napoca, Arany János 11, Romania
Department of Organic Chemistry and Technology, and Research Group for Alkaloid Chemistry of the Hungarian Academy of Sciences, Budapest University of Technology
b
and Economics, H-1111 Budapest, M u} egyetem rkp. 3, Hungary
c
Department of Applied Biotechnology and Food Science, Budapest University of Technology and Economics, H-1111 Budapest, M u} egyetem rkp. 3, Hungary
a r t i c l e i n f o
a b s t r a c t
Article history:
The kinetic resolution of racemic 1-(benzothiazol-2-yl)ethanol rac-2a, 1-(benzo[b]thiophen-2-yl)ethanol
rac-2b, 1-(benzo[b]furan-3-yl)ethanol rac-2c and 1-(benzo[b]thiophen-3-yl)ethanol rac-2d was studied
by enantiomer selective acylation catalyzed by a selection of commercially available and in house pro-
duced lipases. Alcoholysis of the corresponding racemic acetates rac-3a–d catalyzed by Candida antarctica
lipase B (CaLB) was also investigated. Two racemic 1-heteroarylethanols rac-2a,b were prepared by addi-
tion of the corresponding lithiated heteroarylic compounds 1a,b to acetaldehyde, whereas two others,
rac-2c,d were synthesized by the addition of MeMgI onto the corresponding heteroaryl-carbaldehydes
Received 13 June 2008
Accepted 2 July 2008
Available online 30 July 2008
1
2
c,d. The high enantiomer selectivities of CaLB in the acylation of racemic 1-heteroarylethanols rac-
a–d allowed the determination of the enantiomeric preference of these enzymatic acetylation reactions
by QM/MM [pm3/uff or hf(3-21+g**)/uff] calculations. For acetylation of each of the racemic alcohols rac-
a–d, four possible tetrahedral intermediate states were compared and analyzed.
Ó 2008 Elsevier Ltd. All rights reserved.
2
1
. Introduction
The need for the production of optically active compounds as
heteroaromatic group. For example, the optically active 1-(benzo-
furan-2-yl)ethanol was prepared by enantiotopic selective biore-
duction or by lipase-catalyzed enantiomer selective reactions.
9
10
single enantiomers is growing rapidly. Catalytic enantioselective
synthesis is highly preferred in the preparation of enantiopure
compounds, although classical resolution via diastereoisomeric
salts is still widely employed. Due to their chiral nature, biocata-
lysts are predominantly suited for the production of enantiopure
stereoisomers. Thus, biocatalysis has developed from a research
technology to a widely used manufacturing method in the pharma-
ceutical and fine chemicals industry.1 Biotransformations are
already being used by industry to manufacture various products,
As part of our interest in the development of stereoselective
methods for the preparation of optically active heteroaromatic
compounds, we wanted to follow the study of lipase-catalyzed
kinetic resolutions of further racemic 1-heteroarylethanols 2a–d
by lipase-catalyzed enantiomer selective acylation (Fig. 1). Lipases
usually retain their enantiomeric preference in hydrolysis or alco-
1
–3,6
holysis.
Consequently, such reactions should result in opposite
–3
enantiomeric forms of the enantiomerically enriched 1-heteroaryl-
ethanols 2a–d and 1-heteroarylethyl acetates 3a–d as the kinetic
resolution by acetylation. Therefore, the study of hydrolysis or
alcoholysis of the corresponding racemic 1-heteroarylethyl ace-
tates 3a–d has been also considered.
4
,5
such as drugs, agricultural chemicals, fine chemicals and plastics.
Hydrolases—especially lipases—proved to be versatile biocatalysts
6
for synthetic biotransformations. Lipases, in many cases, show
remarkable chemoselectivity, regioselectivity and enantioselectiv-
ity towards a broad range of substrates. Lipases can retain their
remarkably high stereoselectivity in organic media.6 The stereo-
selectivity of lipases has been utilized to perform kinetic resolu-
Stereoselective chemical methods for the 1-heteroarylethanols
2b–d have already been studied. The enantioselective reduction
of the corresponding ketone with (S)-2-methyl-CBS-oxazaboro-
lidine gave the optically active 1-(benzo[b]thiophen-2-yl)ethanol
6
,7
8
11
tions,
deracemizations and dynamic kinetic resolutions. The
(R)-2b.
The optically active 1-(benzo[b]thiophen-3-yl)ethanol
Zn to the corresponding
low cost of many lipases make them very useful catalysts for the
kinetic resolution of secondary alcohols including those containing
(R)-2d was obtained by addition of (CH
aldehyde 1d catalyzed by a commercially available ClCr(Salen).
3 2
)
12
The (S)-enantiomer of 1-(benzo[b]furan-3-yl)ethanol (S)-2c was
obtained by the reduction of (S)-benzo[b]furan-3-yl)oxirane with
*
Corresponding authors. Tel.: +36 1 4632229; fax: +36 1 4633297 (L.P.).
E-mail address: poppe@mail.bme.hu (L. Poppe).
4
LiAlH or by the stereoselective reduction of 3-acetylbenzofuran
0
957-4166/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2008.07.004

M. To ßs a et al. / Tetrahedron: Asymmetry 19 (2008) 1844–1852
1845
the excellent reactivity and selectivity of CaLB with chiral alcohols
OH
1
8,19
1
) BuLi
were explained by several modelling
and molecular dynam-
Ar
H
2
0,21
Ar
2
) acetaldehyde
ics
studies. Very recently, QM/MM calculations within the
1
a,b
rac-2a,b
lipase of P. cepacia revealed that such calculations performed on
the tetrahedral intermediate (THI) states of the reaction can
explain better the reactivity of the lipase towards the enantiomers
of the racemic alcohol substrates than the X-ray structure co-crys-
tallized with the (S)- or (R)-phosphonate inhibitors mimicking the
OH
MeMgI
Ar
O
Ar
rac-2c,d
1c,d
1
7
ester forms of the acetylated products. As systematic conforma-
tional analysis performed within the CaLB active site at MM level
has already been applied to absolute configuration assignments,22
we proposed to extend this method using the experimentally
determined enantiomer selectivities of CaLB towards the enantio-
mers of 1-heteroarylethanols 2a–d combined with QM/MM calcu-
lations within the CaLB structure for assignment of their absolute
configuration.
N
Ar=
S
S
a
b
O
S
2. Results and discussion
c
d
The preparation of racemic 1-heteroarylethanols rac-2a–d was
straightforward (Fig. 1). The 1-(benzothiazol-2-yl)ethanol rac-2a
and 1-(benzo[b]thiophen-2-yl)ethanol rac-2b were prepared by
the addition of the corresponding lithiated heteroarylic com-
pounds 1a,b to acetaldehyde, whereas 1-(benzo[b]furan-3-yl)etha-
nol rac-2c and 1-(benzo[b]thiophen-3-yl)ethanol rac-2d were
synthesized by the addition of MeMgI onto the corresponding het-
eroaryl-carbaldehydes 1c,d.
O
OH
OH
O
lipase
vinyl acetate Ar
+
+
Ar
Ar
rac-2a-d
(S)-2a-d
(R)-3a-d
O
O
O
O
With the racemic 1-heteroarylethanols rac-2a–d and with a
wide selection of commercial and in house prepared lipases2
in hand, the enantiomer selectivity of the enzyme-catalyzed acyla-
tions of the racemic alcohols rac-2a–d was screened first (Fig. 1,
Tables 1–4). As a result of the screen, a number of biocatalysts
exhibiting high enantiomer selectivity towards the racemic
alcohols rac-2a–d were found (Tables 1–4). As expected, several
commercial lipases such as lipase B from C. antarctica (CaLB,
formulated as Novozym 435), and lipase AK exhibited high
enantiomer selectivity (E >100) towards all 1-heteroarylethanols
rac-2a–d (Tables 1–4). Within the series of the four racemic alco-
hols rac-2a–d, the lipases usually retained their degree of enantio-
mer selectivity. Lipase AY, lipases from Candida rugosa/cyclindracae
and porcine pancreas (PPL) exhibited only moderate enantiomer
selectivities towards the racemic 1-heteroarylethanols rac-2a–d
3,24
OH
lipase
Ar
rac-3a-d
R-OH
Ar
Ar
(S)-3a-d
(R)-2a-d
R= Me, Et, Pr, Bu
Figure 1. Preparation and enzymatic kinetic resolution of racemic 1-heteroar-
ylethanols rac-2a–d.
1
3
with (ꢀ)-DIP-Cl. Although enzymatic kinetic resolution of quite
similar structural derivatives (benzimidazolyl ethanols) is already
1
4
described, no enzymatic methods for the enantioselective prepa-
ration of optically active 1-heteroarylethanols 2a–d have been
found.
Although the absolute configuration of the enantiomers of
11
1
-(benzo[b]thiophen-2-yl)ethanol (R)-2b, 1-(benzo[b]furan-3-yl)-
Table 1
1
3
Enantiomer selective enzymatic acylation of racemic 1-(benzo[d]thiazol-2-yl)ethanol
(rac-2a)
ethanol (S)-2c and (R)-2c and 1-(benzo[b]thiophen-3-yl)ethanol
1
2
(
R)-2d was assigned, in many cases the absolute configurations
Ea
were assumed only on the basis of the known stereopreference
of the reagent or catalyst. As there is no data for the optically active
forms of 1-(benzothiazol-2-yl)ethanol 2a, the absolute configura-
tion of its enantiomers cannot be assigned by specific rotation.
Traditionally, proteins and other large biological molecules
have been out of the reach of electronic structure methods. The
QM/MM computational technique allows the study of large mole-
cules by defining two or three layers within the structure that are
treated at different levels of accuracy. In a typical two-layer
approach, the active site is treated using an electronic structure
method, whilst the rest of the system is modelled with molecular
mechanics. Calibration studies have demonstrated that the result-
ing predictions are essentially equivalent to those that would be
produced by the high accuracy method. For example, in a study,
Enzyme
Time [h]
c [%]
ee (S)-2a [%]
ee (R)-3a [%]
Lipase AK
Novozym 435
Lipase AY
CrL
Lipozyme TL IM
CcL
6
72
24
8
46.1
50.4
38.5
27.9
8.7
41.3
2.2
42
48.5
46.4
28.8
26.4
12.4
15.4
13.0
97.9
98.8
59.5
27.1
0.1
63.5
16.1
78.7
99.4
81.2
32.3
28.4
2.6
98.0
99.1
84.2
87.0
68.5
88.2
85.7
98.9
99.5
98.7
99.2
99.7
99.6
85.2
89.4b
>200
ꢁ200
21
19
5
28
25
26
24
38
38
38
38
38
38
26
31
15
PPL
Lipase PS
Lipase 3b
SSF-9
SSF-23
SSF-97
SSF-101
SSF-117
Lipase F
>200
ꢁ200
>200
>200
ꢁ200
ꢁ200
13
2.1
42.9b
27
a
the excited states of bacteriorhodopsin were calculated using
The degree of enantiomer selectivity (E) was determined from ee(S)-2a and
QM/MM (ONIOM) methods.15
25
ee(R)-3a
.
Due to sensitivity to experimental errors, E values calculated in the
100–300 range are reported as >100, values in the 300–500 range are reported as
Well resolved X-ray crystal structures are available for several
>
200 and values calculated above 500 are given as ꢁ200.
16
lipases, such as Candida antarctica lipase B (CaLB), or lipase of
b
Configurations of the major enantiomers are opposite to those given in the
Burkholderia cepacia (formerly Pseudomonas cepacia).17 Moreover,
headline: (R)-2a and (S)-3a.

1
846
M. To ßs a et al. / Tetrahedron: Asymmetry 19 (2008) 1844–1852
Table 2
Table 4
Enantiomer selective enzymatic acylation of racemic 1-(benzo[b]thiophen-2-yl)eth-
anol rac-2b
Enantiomer selective enzymatic acylation of racemic 1-(benzo[b]thiophen-3-yl)eth-
anol (rac-2d)
Enzyme
Time [h]
c [%]
ee(S)-2b [%]
ee(R)-3b [%]
Ea
Enzyme
Time [h]
c [%]
ee(S)-2d [%]
ee(R)-3d [%]
Ea
Lipase AK
Novozym 435
Lipase AY
CrL
Lipozyme TL IM
CcL
24
24
46
24
46
46
46
46
15
22
22
22
22
22
24
49.6
49.8
47.0
26.5
12.5
40.8
10.6
50.5
49.8
48.2
48.8
46.2
31.3
36.3
9.1
99.2
99.4
60.0
35.9
27.1
50.0
11.1
99.8
98.8
99.7
90.6
69.5
20.0
50.0
97.8
99.3
87.9
82.2
99.4
88.5
91.2
98.9
99.4
99.2
98.8
97.0
96.1
96.1
ꢁ200
ꢁ200
37
15
>200
30
Lipase AK
Novozym 435
Lipase AY
CrL
Lipozyme TL IM
CcL
24
24
46
24
46
46
46
46
15
22
22
22
22
22
24
47.5
42.2
24.1
9.3
8.1
16.0
3.0
20.6
49.8
12.8
12.7
7.1
99.0
77.3
32.6
12.3
10.8
19.1
4.8
27.7
98.8
13.6
12.4
6.7
99.4
95.8
82.8
72.7
98.4
83.5
94.5
99.3
99.5
96.2
98.6
99.1
98.6
96.0
84.6b
ꢁ200
>100
15
7
>100
13
PPL
24
PPL
37
Lipase PS
Lipase 3b
SSF-9
SSF-23
SSF-97
SSF-101
SSF-117
Lipase Fb
ꢁ200
ꢁ200
ꢁ200
ꢁ200
>100
77
Lipase PS
Lipase 3b
SSF-9
SSF-23
SSF-97
SSF-101
SSF-117
Lipase Fb
>200
ꢁ200
59
>100
>100
>100
54
4.4
8.8
6.9
0.5
87
2
9.7
9.6
b
b
b
37.9
7.7
13
a
a
The degree of enantiomer selectivity (E) was determined from ee(S)-2b and
The degree of enantiomer selectivity (E) was determined from ee(S)-2d and
25
25
ee(R)-3b
.
Due to sensitivity to experimental errors, E values calculated in the
ee(R)-3d
.
Due to sensitivity to experimental errors, E values calculated in the
1
00–300 range are reported as >100, values in the 300–500 range are reported as
100–300 range are reported as >100, values in the 300–500 range are reported as
>
200 and values calculated above 500 are given as ꢁ200.
>200 and values calculated above 500 are given as ꢁ200.
b
b
Configurations of the major enantiomers are opposite to those given in the
Configurations of the major enantiomers are opposite to those given in the
headline: (R)-2b and (S)-3b.
headline: (R)-2d and (S)-3d.
based on the assumption that the binding pockets of the lipases
can discriminate between the enantiomers of a secondary alcohol.
The sterically different arrangement of the large substituent and
the medium substituent in the two enantiomers of secondary alco-
hols result as in different binding affinity and thus different rates of
reaction. Usually, the (R)-enantiomer is the fast reacting one. Inter-
estingly, lipase F exhibited an opposite enantiomer preference and
resulted in the formation of the (S)-acetates (S)-3a–d (Tables 1–4).
Next, preparative scale kinetic resolutions of the four racemic 1-
heteroarylethanols rac-2a–d were performed (200 mg, each) with
the most selective immobilized commercial enzyme, Novozym
Table 3
Enantiomer selective enzymatic acylation of racemic 1-(benzo[b]furan-3-yl)ethanol
rac-2c
Ea
Enzyme
Time [h]
c [%]
ee(S)-2c [%]
ee(R)-3c [%]
Lipase AK
Novozym 435
Lipase AY
Lipozyme TL IM
CcL
16
16
40
40
40
24
48
21
21
21
21
21
21
40
47.0
48.1
18.2
5.1
16.6
5.1
97.7
96.2
22.0
8.5
97.5
97.6
92.6
91.1
90.7
97.2
91.5
97.7
96.0
93.7
85.2
97.1
87.4
>200
>200
32
23
25
18.4
17.5
11.2
99.7
42.7
30.6
14.5
10.3
20.8
PPL
75
25
Lipase PS
Lipase 3b
SSF-9
SSF-23
SSF-97
SSF-101
SSF-117
Lipase Fb
7.1
4
35, for full characterization of the optically active products (R)-
49.8
26.0
15.5
7.5
6.3
19.4
4.7
ꢁ200
74
3a–d and (S)-2a–d (Table 5). The high enantiomer selectivity of
this lipase allowed the preparation of the alcohols (S)-2a–d and
acetates (R)-3a–d in high enantiomeric excess (ee 96.5–99.3; Table
41
14
76
18
5).
b
b
4.4
74.6
7
The opposite enantiomeric forms of the optically active prod-
ucts [(S)-3a–d and (R)-2a–d] can be prepared by alcoholysis of
the corresponding racemic acetates rac-3a–d with the same immo-
bilized lipase (Novozym 435, Table 6). The degree of enantioselec-
tivity in the alcoholysis reactions of the racemic acetates rac-3a–d
was systematically tested in four alcohols (methanol, ethanol, pro-
pan-1-ol and butan-1-ol; Table 6). The highest selectivities were
achieved in the alcoholyses of racemic 1-acetoxy-1-benzo[b]thio-
phen-2-yl-ethane rac-3b (E >100 in each alcohol). Considering
a
The degree of enantiomer selectivity (E) was determined from ee(S)-2c and
25
ee(R)-3c
.
Due to sensitivity to experimental errors, E values calculated in the
1
>
00–300 range are reported as >100, values in the 300–500 range are reported as
200 and values calculated above 500 are given as ꢁ200.
b
Configurations of the major enantiomers are opposite to those given in the
headline: (R)-2c and (S)-3c.
(
E between 7 and 75, see Tables 1–4). The enhanced steric
bulkiness of 1-(benzo[b]thiophen-3-yl)ethanol rac-2d compared
to 1-(benzo[b]thiophen-2-yl)ethanol rac-2b only influenced the
reactivity, but not the degree of the selectivity of the lipases
Table 5
Novozym 435 (CaLB) catalyzed kinetic resolution of the heteroaryl alcohols
(Tables 2 and 4). Only Lipozyme TL IM (immobilized form of lipase
(rac-2a–d)
from Thermomyces lanuginosus) behaved differently. This lipase
acted with high enantiomer selectivities on the lipophilic heterocy-
cles (E >100 for rac-2b and rac-2d; Tables 2 and 4). On the other
hand, these high selectivities dropped significantly in the kinetic
resolutions of the more polar heterocycles (E = 5 and 23 for
rac-2a and rac-2c, respectively; Tables 1 and 3).
Ea
24,b
Compound
Yield [%]
ee [%]
½aꢂ
D
(S)-2a
(R)-3a
(S)-2b
44
45
45
46
45
44
46
43
97.2
98.0
99.1
99.3
96.7
97.0
96.5
97.8
>200
ꢀ18.4
+75.5
ꢀ21.2
ꢁ200
>100
>200
(R)-3b
+142.2
ꢀ18.5
+73.2
ꢀ27.1
+78.2
(S)-2c
(R)-3c
(S)-2d
(R)-3d
In addition, a number of our novel preparations from thermo-
2
3
philic fungi (lipase 3b) or solid state fermentation of mesophilic
fungi (SSF-9, 23, 97, 117),24 also exhibited excellent enantiomer
selectivities (E >100, see Tables 1–4).
a
The degree of enantiomer selectivity (E) was determined from ee(S)-2 and
Due to sensitivity to experimental errors, E values calculated in the
00–300 range are reported as >100, values in the 300–500 range are reported as
ee(R)-3 2
.
5
As expected, almost all lipases in this study followed the so-
called Kazlauskas rule,26 which predicted the enantiomeric
preference for lipase-catalyzed kinetic resolutions. The rule is
1
>200.
b
3
(c 1.0, CHCl ).

M. To ßs a et al. / Tetrahedron: Asymmetry 19 (2008) 1844–1852
1847
Table 6
or (S)-alcohol with the acyl enzyme, were used to model the tran-
sition states.
Alcoholysis of racemic 1-heteroarylethyl acetates rac-3a–d with Novozym 435 (CaLB)
Compound
Alcohol
Time [h]
c [%]
ee(S)-3 [%]
ee(R)-2 [%]
Ea
It is known that the mechanism of serine proteases involves the
Asp-His-Ser catalytic triad. It is also generally accepted that a tet-
rahedral intermediate (THI) is formed during the acylation and
deacylation step. Within CaLB, the specific regions important for
the prediction of enantioselectivity were determined with the aid
Methanol
Ethanol
Propan-1-ol
Butan-1-ol
28
28
28
28
19.5
51.6
51.7
42.2
7.3
99.3
87.1
70.5
7.3
97.5
85.7
94.6
1
>100
37
rac-3a
77
2
0
Methanol
Ethanol
Propan-1-ol
Butan-1-ol
48
48
48
48
10.7
49.7
42.6
37.7
9.4
94.6
71.3
65.0
99.3
97.5
97.4
96.8
>100
>100
>100
>100
of molecular dynamics-based methodology. The ‘double-pro-
ton-transfer’ mechanism involves a proton transfer from Ser to
His and a proton transfer from His to Asp at the tetrahedral inter-
mediate. Full enzyme empirical valence bond studies concluded
that Asp in the catalytic triad stays ionized during the reaction,
rac-3b
rac-3c
Methanol
Ethanol
Propan-1-ol
Butan-1-ol
24
24
24
24
1.4
18.1
7.1
1.6
21.8
7.1
96.6
95.6
97.7
97.8
58
55
92
98
2
9,30
the ‘double-proton-transfer’ mechanism.
The analysis indi-
7.1
7.1
cates that the most important factor in the catalysis of serine pro-
teases is the stabilization of the negatively charged tetrahedral
intermediate (THI) by the main-chain N–H dipoles of the ‘oxoan-
ion-hole’. High level calculations on 30 atom models reveal that
Methanol
Ethanol
Propan-1-ol
Butan-1-ol
48
48
48
48
2.5
22.1
17.3
9.5
5.9
21.8
13.0
6.8
75.1
99.3
96.3
96.7
2
>100
60
rac-3d
65
‘
double-proton-transfer’ mechanism is comparable in energy to
a
The degree of enantiomer selectivity (E) was determined from ee(R)-2 and
.
the single proton transfer mechanism, but may not be relevant
for real enzymes where other hydrogen bonds can stabilize the
Asp and are therefore likely to disfavour the proton transfer from
2
5
ee(S)-3
Due to sensitivity to experimental errors, E values calculated in the
00–300 range are reported as >100, values in the 300–500 range are reported as
200.
1
>
3
1
His to Asp. High level calculations revealed that THI’s can be good
3
1
approximations of the transition states. Molecular dynamics and
mechanics were used to find low energy conformations and the
calculated energy difference of the THI’s formed by the interaction
2
8,32
the productivity and selectivity together, ethanol is the most
appropriate solvent for the Novozym 435-catalyzed alcoholyses
of the racemic acetates rac-3a–d (Table 6).
The absolute configuration of the enantiomers of 1-(benzofu-
ran-3-yl)ethanol (S)-2c and (R)-2c and 1-(benzothiophene-3-
yl)ethanol (R)-2d was known. Therefore, configurations of all
the residual enantiomers in the CaLB-catalyzed reactions 3a–d
were assigned as (S) on the basis of the same sign of their specific
rotation and also on the basis of the empirical Kazlauskas rule.
Consequently, an (R)-configuration can be assigned to the products
of the (R)- or (S)-substrate with the acyl enzyme.
A systematic
conformational search was applied in case of acyloxypyrrolinon-
3
3
es where a very good agreement was found between the models
and empirical results.
13
The reactions of the (R)- or (S)-enantiomers of a secondary alco-
hol with the acyl enzyme result in the formation of four possible
THI’s. To explain the stereoselectivity of the reactions of racemic
1-heteroarylethanols rac-2a–d, calculations on the four THI’s of
each alcohol were performed.
First, the THI’s were constructed within the CaLB structure and
the systematic conformation search was performed for each THI by
molecular mechanics within a rigid enzyme environment. Next,
the best five structures were calculated by a three-layer ONIOM
model (pm3/uff/uff; over 1020/1019 atoms). To determine the ste-
reoselectivity for the secondary alcohols 2a–d, the lowest energy
ONIOM models for each of the four possible THI’s were compared
(Table 7). The four possible THI structures for racemic 1-(benzo-
thiazol-2-yl)ethanol rac-2a revealed that the bulky heteroaromatic
part of the substrate always occupies the same binding pocket
whilst the carbonyl oxygen of the acyl enzyme always points to
the so-called oxoanion hole (Fig. 2).
The calculations determined that for each of the 1-heteroary-
lethanols rac-2a–d, the (R)-2-(S)-THI structure is the most fa-
voured (Table 7). Accordingly, the fast reacting acetates from all
racemic alcohols 2a–d have an (R)-configuration.
Amongst the four calculated racemic alcohols, the THI’s of the
more bulky 1-(benzo[b]furan-3-yl)ethanol rac-2c and 1-(benzo[b]-
thiophen-3-yl)ethanol rac-2d were structurally more diverse
(Fig. 3). Here, the bicyclic heteroaromatic substituents were
1
2
3
a–d on a similar basis.
There is no absolute configuration assignment based on exper-
imental data for the fast reacting enantiomers (R)-3a and (R)-3b.
Therefore, it was thought worthy to determine the sense of the
enantiomer selectivities of CaLB towards the enantiomers of 1-het-
eroarylethanols 2a–d based on the structure of CaLB.27 The inter-
pretation of the enantiomer selectivity of CaLB with secondary
alcohols was investigated by docking the two alcohol enantiomers
into the active site. Conformation searches of the covalent tetra-
hedral intermediates (THI) at molecular mechanics level were al-
ready applied for absolute configuration assignments.2 Here, we
extended this method by using more reliable QM/MM calculations
to determine the absolute configurations of the fast reacting enan-
tiomers of 3a–d.
The stereoselectivity towards the enantiomers of racemic 1-het-
eroarylethanols rac-2a–d in CaLB was examined by a systematic
conformation search (CS) at molecular mechanics level within
the rigid enzymic environment followed by QM/MM relaxation.
For the calculations, the THI’s, formed by the reaction of the (R)-
2
8
2
Table 7
Comparison of the THI states of acylation of 1-heteroarylethanols rac-2a–d within the active site of CaLB determined by QM/MM (pm3/uff) method
Substrate
(R)-2-(R)-THIa
E (kcal/mol)
(R)-2-(S)-THIa
E (kcal/mol)
(S)-2-(R)-THIa
E (kcal/mol)
8.4
14.8
10.8
10.7
(S)-2-(S)-THIa
N–OS
E (kcal/mol) d /N–OE (Å)
D
d
N–OS
/
N–OE (Å)
D
d
N–OS
/N–OE (Å)
D
d
N–OS
/
N–OE (Å)
D
rac-2a
rac-2b
rac-2c
rac-2d
20.1
37.5
16.1
21.4
2.65/4.06
2.65/4.10
2.57/2.77
2.63/4.20
0.0
0.0
0.0
0.0
2.55/2.68
2.57/2.66
2.54/2.69
2.55/2.68
2.63/3.17
2.67/3.21
2.64/3.16
2.64/3.15
21.2
35.3
10.0
16.8
2.57/3.93
2.52/3.64
2.53/2.71
2.51/3.62
a
The relative energies (
D
E) and N–O distances between His224-N
s
and THI oxygens (d) for the THI states are shown [dN–OS: His224-N
s
–Ser105–O
3
s
; dN–OE: His224-N –
ester–O (i.e., O of 2)].

1
848
M. To ßs a et al. / Tetrahedron: Asymmetry 19 (2008) 1844–1852
Figure 2. Tetrahedral intermediate (THI) states within the active site of CaLB determined by QM/MM (pm3/uff) calculations for acetylation of racemic 1-(benzothiazol-2-
yl)ethanol rac-2a [A, (S)-2a-(S)-THI; B, (R)-2a-(R)-THI; C, (R)-2a-(S)-THI; D, (S)-2a-(R)-THI].
Figure 3. Comparison of the two lower energy THI states for acylation of racemic 1-(benzo[b]furan-3-yl)ethanol rac-2c and 1-(benzo[b]thiophen-3-yl)ethanol rac-2d within
the active site of CaLB determined by QM/MM (pm3/uff) calculations [A, (S)-2c-(R)-THI and (R)-2c-(S)-THI; B, (S)-2d-(R)-THI and (R)-2d-(S)-THI].
arranged oppositely but larger six-membered benzo parts occupied
almost the same position. In fact, the absolute configurations of
and THI oxygens also indicate that the (R)-2-(S)-THI’s are more fa-
voured (Table 8). The significantly longer N –Oester distances (3.34
vs 2.59 Å, for 2a; 3.23 vs 2.56 Å, for 2b; Table 8) for (S)-2-(R)-THI
s
1
2,13
these alcohols 2c,d
our calculation method.
unambiguously supported the validity of
Table 8
At last, to determine more accurately the absolute configura-
tions of the fast reacting enantiomers of racemic 1-(benzothiazol-
Comparison of the THI states of acylation of 1-heteroarylethanols rac-2a and rac-2b
within the active site of CaLB determined by QM/MM [hf(3-21+g**)/uff] method
2
-yl)ethanol rac-2a and 1-(benzo[b]thiophen-2-yl)ethanol rac-2b,
Substrate
(R)-2-(S)-THIa
E (kcal/mol)
(S)-2-(R)-THIa
N–OS
E (kcal/mol) d /N–OE (Å)
a higher level ONIOM method was applied (Table 8). The higher
level method [hf(3-21+g**)/uff/uff, Table 8] as well as the lower
level calculations (pm3/uff/uff, Table 7) indicated (R)-acetates (R)-
D
d
N–OS
/
N–OE (Å)
D
rac-2a
rac-2b
0.0
0.0
2.75/2.59
2.74/2.56
8.3
6.8
2.55/3.34
2.56/3.23
2
a and (R)-2b as the fast forming enantiomers. The heteroaromatic
substituents of 2a and 2b enantiomers occupy the same position
Fig. 4). In addition to the energy differences (8.3 kcal/mol for 2a
and 6.8 kcal/mol for 2b), the distance data between His224-N
a
Relative energies (
D
E) and N–O distances between His224-N
s
and THI oxygens
(
(
d) for the THI states are shown [dN-OS: His224-N
s
–Ser105–O ; dN–OE: His224-N –
3 s
ester–O (i.e., O of 2)].
s

M. To ßs a et al. / Tetrahedron: Asymmetry 19 (2008) 1844–1852
1849
Figure 4. Comparison of the two lower energy THI states for acylation of 1-(benzothiazol-2-yl)ethanol rac-2a and 1-(benzo[b]thiophen-2-yl)ethanol rac-2b within the active
site of CaLB determined by QM/MM [hf(3-21+g**)/uff] calculations [A, (S)-2a-(R)-THI and (R)-2a-(S)-THI; B, (S)-2b-(R)-THI and (R)-2b-(S)-THI].
3
5
than the (R)-2-(S)-THI states indicate that the energy differences
between the real transition states may be even higher than the ones
between the THI states.
Benzofuran-3-carbaldehyde 1c and benzo[b]thiophene-3-carbal-
3
6
dehyde 1d are known compounds.
4
.1.3. Biocatalysts
Lipase AK, lipase AY, lipase F and lipase PS were obtained from
3
. Conclusion
Aldrich. Lipozyme TL IM and Novozym 435 were products of
Novozymes, Denmark. Lipases from C. rugosa (CrL) and Pseudomo-
nas fluorescens (PfL) were purchased from Fluka. PPL and lipase
from C. cyclindracae (CcL) were obtained from Sigma. Several extra-
cellular hydrolases were isolated as acetone dried supernatant of
shake flask fermentation of thermophilic fungus (Lipase 3b)23 or
air dried preparations from solid state fermentation of mesophilic
An efficient synthesis of both enantiomeric forms of four race-
mic 1-heteroarylethanols rac-2a–d, and their acetates rac-2a–d
has been achieved by enzymatic kinetic resolution. Among several
highly selective commercial and in-house developed lipases,
Novozym 435 has been selected as a catalyst for the acetylation
of racemic alcohols rac-2a–d with vinyl acetate. Whereas the
enzymatic acetylation afforded the (R)-enantiomers of acetates
2
4
fungi (SSF-9 SSF-117).
(
R)-3a–d and (S) enantiomers of the alcohols (S)-2a–d in high
4
.2. Synthesis of racemic 1-heteroarylethanols rac-2a–d
enantiomeric excess, alcoholysis of the racemic acetates rac-3a–d
yielded the opposite enantiomeric forms [(R)-2a–d and (S)-3a–d].
For the determination of the absolute configuration of the fast
4
.2.1. Racemic 1-(benzothiazol-2-yl)ethanol, rac-2a and 1-
(
benzo[b]thiophen-2-yl)ethanol, rac-2b
reacting enantiomers of 2a and 2b,
a new computational
To a stirred solution of the heteroaryl compound 1a or 1b
20 mmol) in anhydrous THF (120 mL) under argon, 2.7 M n-BuLi
methodology based on QM/MM calculations on the tetrahedral
intermediates of the enantiomers within the C. antarctica lipase B
structure has been applied.
(
in heptane (8.15 mL, 1.1 equiv) was added below ꢀ70 °C, and the
mixture was stirred for 1 h at ꢀ78 °C. Then, anhydrous acetalde-
hyde (4.38 g, 4.62 mL, 60 mmol) was added below ꢀ70 °C, and
the solution was allowed to warm to ꢀ10 °C in 2 h. The mixture
4
4
4
. Experimental
4
was quenched with saturated NH Cl solution (30 mL) and diluted
.1. Materials and methods
with water (100 mL). After separation of the forming layers, the
water layer was extracted with Et
O (3 ꢃ 50 mL). The combined or-
ganic layers were washed with water (50 mL) and brine (30 mL),
dried over anhydrous MgSO and concentrated in vacuo. The resi-
due was purified by column chromatography on silica gel using
CH Cl as eluent to yield rac-2a or rac-2b as a white solid.
2
.1.1. Analytical methods
The 1H and 13C NMR spectra were recorded in CDCl
3
solution on
4
a Brucker Avance DPX-300 spectrometer operating at 300 and
5 MHz, respectively. Chemical shifts on the d scale are expressed
7
2
2
in ppm values with TMS as the internal standard. IR spectra were
recorded in KBr or CsBr plates on a Jasco FT-IR spectrometer and
the wavenumbers are reported in cm . GC analyses were made
by an Konik HRGC 4000 B gas chromatograph (carrier gas N ; head
ꢀ1
4.2.1.1. Racemic 1-(benzo[d]thiazol-2-yl)ethanol rac-2a. Yield:
3
7
1
9
2%; mp: 55–56 °C (lit. 54–56 °C ); H NMR: 1.73 (d, 3H,
2
J = 6.6 Hz), 3.31 (s, 1H), 5.29 (q, 1H, J = 6.6 Hz), 7.37–7.52 (m, 2H),
7
6
3
1
M
pressure: 60 psi, injector: 250 °C; FID detector: 250 °C) on a Astec-
Chiral column (30 m ꢃ 0.25 mm, dimethylated-b-cyclodextrin, No.
1
3
.89 (d, 1H, J = 8.0 Hz), 7.99 (d, 1H, J = 8.0 Hz); C NMR: 24.0,
8.4, 121.8, 122.7, 125.1, 126.2, 134.7, 152.5, 177.1; IR (KBr):
212, 2977, 2682, 2364, 1513, 1438, 1365, 1313, 1290, 1241,
176, 1157, 1101, 1072, 1058, 1012, 890, 757, 730, 505; HRMS:
7
77023 G0507-30). EI mass spectra were taken on a VG 7070E
mass spectrometer operating at 70 eV, and are characterized by
m/z (relative intensity %) values. Optical rotations were determined
+
+
found (M calculated for C
H
9 9
NOS): 179.04067 (179.04048);
on a Bellingham-Stanley ADP 220 polarimeter. Preparative vac-
+
uum-chromatography34 was performed on Merck Kieselgel 60
MS: 179(M , 27), 164(16), 136(100), 108(35), 91(9), 82(13),
6
9(28), 63(18), 50(10) 45(35), 39(16).
(
0.063–0.200 lm).
4.2.1.2. Racemic 1-(benzo[b]thiophen-2-yl)ethanol rac-2b.11
Yield: 88% (lit. 84% ); mp: 62 °C (lit. 62–63 °C ); H NMR: 1.66
(d, 3H, J = 6.4 Hz), 2.53 (s, 1H), 5.19 (q, 1H, J = 6.4 Hz), 7.17 (s,
1H), 7.31–7.40 (m, 2H), 7.73 (d, 1H, J = 7.6 Hz), 7.83 (d, 1H,
4
.1.2. Reagents and solvents
Benzothiazol 1a, benzo[b]thiophen 1b, vinyl acetate and the
1
1
38
1
other commercial chemicals and solvents were products of Aldrich
or Fluka. All solvents were purified and dried by standard methods.

1
850
M. To ßs a et al. / Tetrahedron: Asymmetry 19 (2008) 1844–1852
J = 7.6 Hz); 13C NMR: 25.1, 66.8, 119.5, 122.5, 123.5, 124.1, 124.3,
4.3.2. Racemic 1-acetoxy-1-benzo[b]thiophen-2-yl-ethane rac-
3b
1
1
8
C
1
39.3, 139.5, 150.5; IR (KBr): 3301, 2967, 2923, 2358, 1733,
455, 1434, 1369, 1315, 1249, 1203, 1143, 1128, 1066, 985, 937,
Yield: 98%; mp: 30 °C (lit. mp = 30–31 °C40); 1H NMR: 1.72 (d,
3H, J = 6.6 Hz); 2.12 (s, 3H); 6.27 (q, 1H, J = 6.6 Hz); 7.30 (s, 1H),
7.32–7.37 (m, 2H), 7.76 (d, 1H, J = 6.6 Hz), 7.83(d, 1H,
+
+
89, 831, 742, 725, 584; HRMS: M found (M calculated for
+
10
H10OS): 178.04573 (178.04524); MS: 178(M , 37), 163(29),
1
3
35(100), 115(14), 91(52), 69(19), 63(24), 51(16), 43(64), 39(24).
J = 6.6 Hz); C NMR: 21.3, 21.9, 68.2, 121.7, 122.4, 123.7, 124.4,
24.5, 139.2, 139.4, 145.1, 170.2; IR (CsBr): 3058, 2983, 2933,
1
4
1
.2.2. Racemic 1-(benzofuran-3-yl)ethanol rac-2c and
-(benzo[b]thiophen-3-yl)ethanol rac-2d
To a stirred solution of methylmagnesium iodide, prepared
2355, 1741, 1436, 1371, 1236, 1151, 1049, 1018, 937, 831, 748,
+
+
727, 584; HRMS:
M
found (M calculated for
C
H
12 12
O
2
S):
+
220.05597 (220.0558); MS: 220(M , 50), 178(95), 161(85),
135(30), 128(52), 115(41), 89(30), 69(12), 63(17), 51(9),
43(100), 39(12).
from magnesium (7.4 mmol, 0.177 g) and iodomethane (7.4 mmol,
.05 g, 0.460 mL) in Et O (5 mL), a solution of a heteroaryl-3-carb-
aldehyde (1c or 1d, 6.17 mmol) dissolved in Et O (3 mL) was added
1
2
2
slowly at 0 °C under argon. The resulting mixture was stirred at
room temperature overnight. After quenching the reaction mixture
by the slow addition of saturated ammonium chloride solution
4.3.3. Racemic 1-acetoxy-1-benzo[b]furan-3-yl-ethane rac-3c
1
Yield: 96%; semisolid;
H NMR: 1.72 (d, 3H, J = 6.8 Hz),
2.11(s, 3H), 6.23 (q, 1H, J = 6.8 Hz), 7.26–7.37 (m, 2H),
(
8 mL), the organic layer was separated and the aqueous layer ex-
7.52 (d, 1H, J = 6.6 Hz), 7.64 (s, 1H), 7.68 (d, 1H, J = 6.6 Hz); 13
C
tracted with Et O solutions were
dried over anhydrous sodium sulfate. After removing the solvent
by rotatory evaporation, the solid residue was purified by vac-
2
O (2 ꢃ 10 mL). The combined Et
2
NMR: 20.4, 21.3, 64.9, 111.7, 120.4, 121.2, 122.8, 124.6, 125.9,
142.0, 155.6, 170.4; IR (CsBr) 2983, 2933, 2364, 1735, 1454,
+
1371, 1238, 1087, 1049, 1016, 941, 856, 746; HRMS: M found
+
uum-chromatography on silica gel using CH
2
Cl
2
as eluent resulting
(M calculated for
C
12
H
12
O
3
): 204.07848 (204.07864); MS:
+
in the racemic alcohol rac-2c or rac-2d as a colourless semisolid.
204(M , 23), 162(42), 145(51), 115(57), 89(26), 63(26), 51(11),
3(100), 39(19).
4
4
8
5
7
1
2
.2.2.1. Racemic 1-(benzo[b]furan-3-yl)ethanol rac-2c. Yield:
9%; semisolid; H NMR: 1.65 (d, 3H, J = 6.7 Hz), 2.33 (s, 1H),
1
4.3.4. Racemic 1-acetoxy-1-benzo[b]thiophen-3-yl-ethane rac-
.13 (q, 1H, J = 6.7 Hz), 7.24–7.36 (m, 2H), 7.50 (d, 1H, J = 7.4 Hz),
.55 (s, 1H), 7.71 (d, 1H, J = 7.4 Hz); 13C NMR: 23.4, 63.1, 111.6,
20.5, 122.6, 124.5, 125.1, 126.1, 140.8, 155.7; IR (CsBr): 3359,
3d
1
Yield: 98%; semisolid; H NMR: 1.73 (d, 3H, J = 6.6 Hz), 2.13 (s,
3H), 6.35 (q, 1H, J = 6.6 Hz), 7.38–7.43 (m, 2H), 7.45 (s, 1H), 7.86–
+
13
977, 1579, 1452, 1369, 1282, 1074, 1004, 858, 746; HRMS: M
7.91 (m, 2H); C NMR: 20.6; 21.3; 67.4; 122.1; 122.9; 123.3;
+
found (M calculated for C10
10
H O
2
): 162.06838 (162.06808); MS:
62(M , 36), 147(59), 115(18), 91(100), 63(26), 51(13), 43(30),
124.2; 124.5; 136.3; 137.1; 140.7; 170.4; IR (CsBr): 2981, 2933,
+
+
1
3
2358, 1735, 1428, 1369, 1240, 1043, 939, 761, 734; HRMS: M
+
9(22).
found (M calculated for C12
H
12
O
2
S): 220.05563 (220.0558); MS:
20(M , 48), 178(85), 163(43), 161(91), 135(30), 128(54),
+
2
4
.2.2.2. Racemic
1-(benzo[b]thiophen-3-yl)ethanol rac-2d.
115(45), 89(33), 69(11), 63(16), 51(8), 43(100), 39(10).
39
1
Yield: 85%; mp: 55 °C (lit. 53.5–55 °C ); H NMR: 1.65 (d, 3H,
J = 6.6 Hz), 2.37 (s, 1H), 5.24 (q, 1H, J = 6.6 Hz), 7.43 (s, 1H), 7.36–
7
1
2
found (M calculated for C10
1
6
4.4. Lipase-catalyzed acylations of racemic 1-hetero-
arylethanols rac-2a–d
.41 (m, 2H), 7.87–7.92 (m, 2H); 13C NMR: 23.4, 65.7, 121.5,
22.3, 122.9, 124.1, 124.4, 137.3, 140.7, 141.0; IR (CsBr): 3359,
+
973, 2364, 1456, 1427, 1369, 1255, 1122, 1070, 838; HRMS: M
4.4.1. Acylations on an analytical scale
+
H10OS): 178.04552 (178.04524); MS:
To a solution of racemic 1-heteroarylethanol (rac-2a–d, 20 mg)
in vinyl acetate (0.5 mL), lipase (20 mg) was added. The reaction
mixture was shaken at 1000 rpm at room temperature. For GC
analysis, samples were taken directly from the reaction mixture
+
78(M , 49), 163(67), 161(12), 135(100), 128(7), 115(18), 91(55),
9(14), 63(20), 51(12), 43(43), 39(15).
4
2
.3. Chemical acetylation of racemic 1-heteroarylethanols rac-
a–d
2 2
(10 lL, diluted to 200 lL with CH Cl ). Data on reaction time, con-
version and enantiomeric composition of the products [(S)-2a–d
and (R)-3a–d] are presented in Tables 1–4.
To a solution of the racemic 1-heteroarylethanol (rac-2a–d,
5
.61 mmol) in 15 mL of dry CH
2
Cl
2
3
, Et N (6.17 mmol, 0.62 g,
4.4.1.1. GC analysis of 1-(benzo[d]thiazol-2-yl)ethanol 2a and 1-
0
.85 mL), acetic anhydride (6.17 mmol, 0.63 g, 0.58 mL) and DMAP
(benzo[d]thiazol-2-yl)ethyl acetate 3a. GC (110 °C,
t
R /min):
(
20 mg) were added. The resulting mixture was stirred at room
44.89/51.82 [(S)-/(R)-enantiomers of 2a, respectively], 22.3/23.57
[(S)-/(R)-enantiomers of 3a, respectively].
temperature overnight and was quenched with water (15 mL).
The organic layer was separated and dried over anhydrous sodium
sulfate, and the solvent was distilled off by rotatory evaporation.
The residue was purified by vacuum-chromatography on silica
4.4.1.2. GC analysis of 1-(benzo[b]thiophen-2-yl)ethanol 2b and
1-(benzo[b]thiophen-2-yl)ethyl acetate 3b. GC (140 °C, R /min):
t
gel using CH
semisolid.
2
Cl
2
as eluent to give acetate rac-3a–d as a colourless
12.48/12.92 [(R)-/(S)-enantiomers of 2b, respectively], 9.79/10.19
[(S)-/(R)-enantiomers of 3b, respectively].
4
.3.1. Racemic 1-acetoxy-1-benzo[d]thiazol-2-yl-ethane rac-3a
4.4.1.3. GC analysis of 1-(benzo[b]furan-3-yl)ethanol 2c and 1-
(benzo[b]furan-3-yl)ethyl acetate 3c. GC (120 °C, R /min): 23.3/
t
24.73 [(R)-/(S)-enantiomers of 2c, respectively], 14.41/16.32 [(S)-/
(R)-enantiomers of 3c, respectively].
1
Yield: 97%; semisolid; H NMR: 1.78 (d, 3H, J = 6.6 Hz), 2.20 (s,
H), 6.26 (q, 1H, J = 6.6 Hz), 7.38–7.53 (m, 2H), 7.89 (d, 1H,
3
13
J = 8.2 Hz), 8.05 (d, 1H, J = 8.2 Hz); C NMR: 20.8, 21.1, 70.2,
1
2
7
21.7, 123.2, 125.3, 126.2, 134.6, 152.9, 169.8, 171.3; IR (CsBr):
991, 2933, 2358, 1749, 1521, 1438, 1371, 1230, 1060, 937, 759,
4.4.1.4. GC analysis of 1-(benzo[b]thiophen-3-yl)ethanol 2d and
1-(benzo[b]thiophen-3-yl)ethyl acetate 3d. GC (140 °C, R /min):
t
14.6/15.46 [(S)-/(R)-enantiomers of 2d, respectively], 7.94/8.23
[(S)-/(R)-enantiomers of 3d, respectively].
+
+
30; HRMS: M found (M calculated for C11
H11NO
2
S): 221.05131
+
(
221.05105); MS: 221(M , 13), 178(64), 162(15), 136(34),
09(19), 90(4), 69(17), 65(10), 55(5), 43(100), 39(9).
1

M. To ßs a et al. / Tetrahedron: Asymmetry 19 (2008) 1844–1852
1851
4
.4.2. Preparative scale acylation of racemic 1-hetero-
arylethanols rac-2a–d
To solution of racemic 1-heteroarylethanols rac-2a–d
200 mg, each) in vinyl acetate (6 mL) Novozym 435 (200 mg)
4.6.2. QM/MM calculations within the active site of CaLB
All QM/MM optimizations were carried out using the ONIOM
4
2
a
protocol as implemented in GAUSSIAN 03 with the aid of GAUSS-
4
3
(
VIEW for selecting and preparing the calculation levels.
The best three conformations for each THI’s were optimized
using a three-layer ONIOM method (pm3/uff/uff) over 1020/1019
atoms. The high layer (HL) included the atoms of the alcohols
2a–d, the acetate part of THI, the side chains of the catalytic triad
(Ser105, His224, Asp187 + one water), the relevant parts of the res-
idues forming the so-called oxoanion hole (Gln106, Thr40). The HL
was relaxed with the semiempirical PM3 methods. The medium
layer (ML) included Trp104 (+ one water), Asp134, Gln157,
Ile189, Leu278, Ala281 and Ala282. The low layer (LL) consisted
of the remaining residues which were kept fixed during the opti-
mization. The ML and LL were optimized by molecular mechanic
force-field (UFF). In addition to the (pm3/uff/uff) calculations, sev-
eral THI’s of 2a and 2b [(R)-2-S-THI and (S)-2-R-THI] were opti-
mized with Hartree–Fock methods in the HL [hf(3-21+g**)/uff/
uff] as well.
was added and the mixture was stirred at room temperature
overnight. The enzyme was filtered off and washed with acetone
(
2 ꢃ 5 mL). Solvents were distilled off from the combined filtrates,
and the residue was purified by column chromatography (silica
gel, CH Cl
2
2
) resulting in the optically active alcohol (ꢀ)-(S)-2a–d
and acetate (+)-(R)-3a–d fractions as white semisolids. IR, NMR
and MS spectra of the optically active alcohols, (S)-2a–d and
acetates, (R)-3a–d were indistinguishable from those of their
racemates (for spectra of rac-2a–d and rac-3a–d, see Sections
4
.2 and 4.3). Data on yield, enantiomeric composition and specific
rotation of the products [(S)-2a–d and (R)-3a–d] are shown in
Table 5.
4
.5. Lipase-catalyzed alcoholysis of racemic 1-heteroarylethyl
acetates rac-3a–d
To a mixture of racemic 1-heteroarylethyl acetates rac-3a–d
Acknowledgements
(
20 mg, each) in alcohol (0.5 mL), Novozym 435 (20 mg) was
added, and the mixture was shaken at 1000 rpm and room temper-
ature. Samples were analyzed by GC similarly as described in Sec-
tion 4.4.1. Data on solvent, reaction time, conversion and
enantiomeric composition of the products (R)-2a–d and (S)-3a–d
are presented in Table 6.
L.P. thanks the financial support from Hungarian OTKA Founda-
tion (T-048854). The financial support to C.P from the Romanian
Ministry of Education and Research (No. 1480-6/07.04.2006) and
Domus Hungarica Foundation is gratefully acknowledged.
References
4
4
.6. Computational methods
1
2
.
.
Faber, K. Biotransformations in Organic Chemistry, 4th ed.; Springer: Berlin,
004.
Rehm, H.-J.; Reed, G.; Pühler, A.; Stadler, P.; Kelly, D. R., 2nd ed. In
2
.6.1. Systematic conformational search within the active site
Biotechnology: Biotransformations I and II; Wiley-VCH: Weinheim, 1998; Vols.
of CaLB at MM level
8a and 8b.
The calculations were performed within the active site of the
3. Poppe, L.; Novák, L. Selective Biocatalysis: A Synthetic Approach; VCH:
crystal structure CaLB containing Tween80 (T80) as substrate
Weinheim-New York, 1992.
Industrial Biotransformations; Liese, A., Seelbach, K., Wandrey, C., Eds., 2nd ed.;
Wiley-VCH: Weinheim-New York, 2006.
Panke, S.; Held, M.; Wubbolts, M.; Schmid, A.; Hollmann, F.; Park, J. B.; Bühler,
B. Curr. Opin. Biotechnol. 2004, 15, 272–279.
Bornschauer, U. T.; Kazlauskas, R. J. Hydrolases in Organic Synthesis: Regio- and
Stereoselective Biotransformations; Wiley-VCH: Weinheim-New York, 2006.
Ghanem, A.; Aboul-Enein, H. Y. Chirality 2005, 17, 1–15.
mimic (PDB code: 1LBT).27 From this CaLB structure, a part con-
4.
y
taining the residues within 10 Å spheres around the Ser105 O ,
5.
6.
7.
T80 C10 and T80 C20 atoms was cut off as the active site model
for CS and later QM/MM calculations. Next, T80 was removed
and by the Hyperchem41 standard procedure, hydrogens were
added to the amino acid residues and water oxygens. The C- and
N-termini at cutting were completed to neutral aldehyde and ami-
no moieties. During our calculation His224 was protonated and
Asp187 was deprotonated.
The tetrahedral intermediate (THI) states were built in the
active site model by constructing a covalent bond between the
O
8. Turner, N. J. Curr. Opin. Chem. Biol. 2004, 8, 114–119.
9.
Paizs, Cs.; To ßs a, M.; Majdik, C.; Moldovan, P.; Novák, L.; Kolonits, P.; Marcovici,
A.; Irimie, F.-D.; Poppe, L. Tetrahedron: Asymmetry 2003, 14, 1495–1501.
1
1
1
0. Paizs, Cs.; To sß a, M.; Bódai, V.; Szakács, G.; Kmecz, I.; Simándi, B.; Majdik, C.;
Novák, L.; Irimie, F.-D.; Poppe, L. Tetrahedron: Asymmetry 2003, 14, 1943–1949.
1. Yokoyama, Y.; Shiraishi, H.; Tani, Y.; Yokoyama, Y.; Yamaguchi, Y. J. Am. Chem.
Soc. 2003, 125, 7194–7195.
2. Cozzi, P. G.; Kotrusz, P. J. Am. Chem. Soc. 2006, 128, 4940–4941.
c
atom of Ser105 and the carbonyl carbon of the (R)- and (S)-
13. Zaidlewicz, M.; Chechlowska, A.; Prewysz-Kwinto, A.; Wojtczak, A. Heterocycles
001, 55, 569–577.
4. Ravi Kumar, C.; Rachna, S.; Palakodety, R. K. Tetrahedron: Asymmetry 2008, 19,
01–905.
2
enantiomers of 3a–d acetates. In this way, starting structures were
constructed for the four THI states (R)-2-R-THI, (R)-2-S-THI, (S)-2-
R-THI, (S)-2-S-THI for acetylation of each of the racemic alcohols
1
9
15. Vreven, T.; Morokuma, K. Theor. Chem. Acc. 2003, 109, 125–132.
16. (a) Uppenberg, J.; Hansen, M. T.; Patkar, S.; Jones, T. A. Structure 1994, 2, 293–
rac-2a–d. The initial THI’s including C
a
and C
b
(with their hydro-
308; (b) Uppenberg, J.; Patkar, S.; Bergfors, T.; Jones, T. A. J. Mol. Biol. 1994, 235,
gens) of Ser105 (33 atoms for 3b–d and 32 atoms for 3a acetates)
were optimized by the MM+ method of HYPERCHEM41 within the
rigid enzymic environment (including no waters).
790–792; (c) Uppenberg, J.; Öhrner, N.; Norin, M.; Hult, K.; Kleywegt, G. J.;
Patkar, S.; Waagen, V.; Anthonsen, T.; Jones, T. A. Biochemistry 1995, 34, 16838–
16851.
1
7. Luic, M.; Stefanic, Z.; Ceilinger, I.; Hodoscek, M.; Janezic, D.; Lenac, T.; Lescic, A.;
Sepac, D.; Tomic, S. J. Phys. Chem. B 2008, 112, 4876–4883.
8. Haeffner, F.; Norin, T.; Hult, K. Biophys. J. 1998, 74, 1251–1262.
The conformational search (CS) within this active site model
4
1
was performed by the CS module implemented in Hyperchem
1
at molecular mechanics (MM+) level using default settings (MM+
forcefield; gradient: 0.1 kcal/mol; Polak-Ribiere method until
19. Otto, R. T.; Scheib, H.; Bornscheuer, U. T.; Pleiss, J.; Syldatk, C.; Schmid, R. D. J.
Mol. Catal. B 2000, 8, 201–211.
20. Raza, S.; Fransson, L.; Hult, K. Protein Sci. 2001, 10, 329–338.
1. Ottosson, J.; Fransson, L.; Hult, K. Protein Sci. 2002, 11, 1462–1471.
0
.05 kcal/mol RMS gradient or maximum 600 cycles; limits: 300
2
iterations, 150 optimizations, 15 conformations; test options: ‘skip
if atoms are closer than 0.3 Å’). During CS, five torsion angles (start-
ing from C of Ser105) and no rings of the alcohols 2a–d were cho-
a
sen to vary in a rigid enzymic environment (including no waters).
The three lowest energy conformations of each of the 2a–d alco-
hol–acyl enzyme THI systems were investigated further by QM/
MM methods.
22. (a) Szatzker, G.; Pilbák, S.; T o} ke, E. R.; Bódai, V.; Poppe, L. FEBS J. 2005, 272, 114;
b) T o} ke, E. R.; Kolonits, P.; Novák, L.; Poppe, L. Tetrahedron: Asymmetry 2006,
7, 2377–2385.
(
1
2
2
2
3. Bódai, V.; Peredi, R.; Bálint, J.; Egri, G.; Novák, L.; Szakács, G.; Poppe, L. Adv.
Synth. Catal. 2003, 345, 811–818.
4. Nagy, V.; T o} ke, E. R.; Chee Keong, L.; Szatzker, G.; Ibrahim, D.; Che-Omar, I.;
Szakács, G.; Poppe, L. J. Mol. Catal. B, Enzym. 2006, 39, 141–148.
5. Rakels, J. L. L.; Straathof, A. J. J.; Heijnen, J. J. Enzyme Microb. Technol. 1993, 15,
1051–1056.

1
852
M. To ßs a et al. / Tetrahedron: Asymmetry 19 (2008) 1844–1852
2
2
6. Kazlauskas, R. J.; Weissfloch, A. N. E.; Rappaport, A. T.; Cuccia, L. A. J. Org. Chem.
991, 56, 2656–2665.
7. Uppenberg, J.; Ohmer, N.; Norin, M.; Hult, K.; Kleywegt, G. J.; Patkar, S.;
39. Noyce, D. S.; Forsyth, D. A. J. Org. Chem. 1974, 19, 2828–2831.
40. Hill, E. A.; Gross, M. L.; Stasiewicz, M.; Manion, M. J. Am. Chem. Soc. 1969, 91,
7381–7392.
1
Waagen, V.; Anthomen, T.; Jones, T. A. Biochemistry 1995, 34, 16838–
41. HYPERCHEM 7.5, Hypercube, 2003.
1
6851.
42. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G.
A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe,
M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; and Pople,
J. A.; GAUSSIAN 03, Revision B.05, Gaussian, Wallingford, CT, 2003.
2
2
3
8. Cuiper, A. PhD thesis, University of Groningen (NL), 1999.
9. Warshel, A.; Russell, S. J. Am. Chem. Soc. 1986, 108, 6569–6579.
0. Warshel, A.; Naray-Szabó, G.; Sussman, F.; Hwang, J.-K. Biochemistry 1989, 28,
3
629–3637.
3
3
1. Hu, C. H.; Brinck, T.; Hult, K. Int. J. Quantum Chem. 1998, 69, 89–103.
2. Rotticci, D.; Haeffner, F.; Orrenius, C.; Norin, T.; Hult, K. J. Mol. Catal. B 1998, 5,
2
67–272.
3
3. Orrenius, C.; Haeffner, F.; Rotticci, D.; Öhrner, N.; Norin, T.; Hult, K. Biocatal.
Biotransf. 1998, 16, 1–15.
4. Poppe, L.; Novák, L. Magy. Kém. Lapja 1985, 40, 366–369.
5. Zaidlewicz, M.; Chechlowska, A.; Prewysz-Kwinto, A.; Wojtczak, A. Heterocycles
3
3
2
001, 55, 569–577.
3
3
3
6. King, W. J.; Nord, F. F. J. Org. Chem 1948, 13, 635–637.
7. Florio, S.; Capriati, V.; Colli, G. Tetrahedron 1997, 53, 5839–5846.
8. Yokoyama, Y.; Shiraishi, H.; Tani, Y.; Yokoyama, Y. J. Am. Chem. Soc. 2003, 125,
43. GAUSSVIEW, Version 3.09, R. Dennington II, T. Keith, J. Millam, K. Eppinnett, W. L.
Hovell, R. Gilliland, Semichem, Shawnee Mission, KS, 2003.
7
194–7195.