Highly enantioselective synthesis of arylaliphatic tertiary alcohols using mutants of an esterase from Bacillus subtilis

Article abstract of DOI:10.1002/adsc.200600641

The kinetic resolution of a series of acetates of arylaliphatic tertiary alcohols was studied using recombinant esterase variants from Bacillus subtilis (BS2) expressed in E. coli. Highest enantioselectivities (E > 100) were achieved in the synthesis of 1

Full text of DOI:10.1002/adsc.200600641

FULL PAPERS
DOI: 10.1002/adsc.200600641
Highly Enantioselective Synthesis of Arylaliphatic Tertiary
Alcohols using Mutants of an Esterase from Bacillus subtilis
Robert Kourist,a Sebastian Bartsch,a and Uwe T. Bornscheuera,*
a
Institute of Biochemistry, Department of Biotechnology and Enzyme Catalysis, Greifswald University,
Felix-Hausdorff-Str. 4, 17487 Greifswald, Germany
Phone: (+49)-3834-86-4367; fax: (+49)-3834-86-80066; e-mail: uwe.bornscheuer@uni-greifswald.de
Received: December 16, 2006
Abstract: The kinetic resolution of a series of ace- chain substantially lower conversions and/or enantio-
tates of arylaliphatic tertiary alcohols was studied selectivity were observed, which also varied with the
using recombinant esterase variants from Bacillus BS2 variant used. Thus, small changes in the sub-
subtilis (BS2) expressed in E. coli. Highest enantiose- strate structure and point mutations in the esterase
lectivities (E>100) were achieved in the synthesis of had a remarkable influence on both activity and
1,1,1-trifluoro-2-phenylbut-3-yn-2-ol and three para- enantioselectivity.
substituted analogues using BS2 mutant G105A.
With mutant E188D only two compounds were con-
verted with E>100. For a thiophene analogue or Keywords: enantioselectivity; enzyme catalysis; hy-
compounds with small variations in the aliphatic drolases; steric hindrance; tertiary alcohols
Introduction
drolysis.[7] Unfortunately, the enantioselectivity of all
GGG(A)X-hydrolases investigated so far has been
Optically active tertiary alcohols are an important very low. For instance, tertiary a-acetylenic acetate
class of organic compounds with valuable applications esters could only be resolved by Candida rugosa
in organic synthesis.[1,2] For instance, the tertiary a- lipase with low enantioselectivity.[8,9]
acetylenic alcohols 2-phenylbut-3-yn-1-ol (1a) and
For an esterase from Bacillus subtilis (BS2) ex-
1,1,1-trifluoro-2-phenylbut-1-yn-3-ol (2a) were recent- pressed recombinantly in E. coli, we determined an
ly applied for the synthesis of A2A receptor antago- E=3 towards 2-phenylbut-3-yn-1-yl acetate (1)
nists that were shown to be orally active in a mouse (Scheme 1).[6] In a molecular modelling study a
catalepsy model.[3] In addition, fluorine-containing op- mutant was predicted, that catalysed the hydrolysis of
tically active alcohols are also of great interest due to 1 at a 6-fold increased enantioselectivity (E=19).[10]
their potential use as ferroelectric liquid crystals and This value represented a notable increase but was still
drugs.[4]
too low for preparative applications. One reason for
Carboxylester hydrolases (lipases, EC 3.1.1.3, ester- the low enantioselectivity is a non-enzymatic and
ases, EC 3.1.1.1) represent a class of versatile biocata- therefore not enantioselective autohydrolysis of 1 via
lysts for the preparation of enantiopure compounds, an SN1 mechanism.[8] This autohydrolysis could be
especially of optically pure secondary alcohols and to suppressed by addition of the water-miscible co-sol-
a smaller extent also for the resolution of primary al- vent DMSO, resulting in an increased Evalue in the
cohols and carboxylic acids.[5] In contrast, tertiary al- kinetic resolution of 1 (E=54 at 20% v/v DMSO).[11]
cohols (TAs) are not accepted as substrates by almost A further molecular modelling study identified anoth-
all carboxylester hydrolases due to their sterically de- er mutant with a substantially increased enantioselec-
manding structure. We recently identified that a tivity. This BS2-mutant E188D hydrolysed 1 with an
GGG(A)X-motif located in the active site region de- enantioselectivity of E=45 (with 20% v/v DMSO)[11]
.
termines activity towards tertiary alcohols and that li- A valuable alternative to the addition of a cosolvent
pases or esterases with the much more common GX- is the use of the trifluoromethyl analogue 2, which is
motif are inactive.[6] The GGG(A)X-motif is located stable in aqueous solution due to the electron-with-
in the oxyanion hole pocket, which is required for the drawing effect of the CF3 group. This compound was
stabilisation of the oxyanion in the tetrahedral inter- resolved with excellent selectivity (E>100) by both
mediate formed during the catalytic cycle of ester hy- BS2 mutants.[11]
Adv. Synth. Catal. 2007, 349, 1393 – 1398
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1393

Robert Kourist et al.
FULL PAPERS
Scheme 1. Enzymatic kinetic resolution of arylaliphatic tertiary alcohol acetates catalysed by BS2 esterase mutants G105A
and E188D.
In this paper, we have extended these studies to a took place. Introduction of a fluoro-substituent in the
series of the tertiary alcohol acetates 2–11 using both m-position (5) rather than the p-position (4) resulted
BS2 mutants G105A and E188D in order to identify in a considerably decreased activity and lower enan-
more substrates for this kinetic resolution and to get tioselectivity and the double-substituted compound 6
further insights into the substrate range and selectivi- was not converted at all. Also, the presence of a thio-
ty of these esterases variants.
phene instead of a phenyl ring lowered the Evalue to
only E=4. Whereas mutant E188D showed similar
activity and enantioselectivity as G105A in the resolu-
tion of 2,[11] it was much less active and selective to-
wards the series of acetates studied in this work. In
Results and Discussion
The tertiary alcohols 2a–11a were synthesised by contrast to G105A, only compound 4 was converted
Grignard reactions from the corresponding ketones at high enantioselectivity, but the conversion was low.
and ethynylmagnesium bromide.[8] Subsequent acety- The thiophene derivative 9 was hydrolysed with high
lation in the presence of dimethylaminopyridine conversion (60%) but at only moderate enantioselec-
yielded the corresponding acetates 2–11. For all com- tivity (E=11). Interestingly, the a-acetylenic group
pounds except 10a a method for the separation of seems to be essential for catalysis as neither the
their enantiomers by GC analysis could be established methyl-substituted acetylenic acetate 10 nor the
(see Experimental Section). Next, the acetates were double bond derivative 11 were converted by both es-
subjected to analytical-scale kinetic resolutions using terase mutants. This confirms previous reports using
both BS2 esterase mutants (E188D or G105 A) at Candida rugosa lipase.[8,9]
pH 7.5, 378C for 15 min in the presence of 10%
Next, preparative 100 mg-scale resolutions of ace-
(G105A) or 20% (E188D) DMSO (Scheme 1).
tates 4, 7 and 8 and a 500 mg-scale resolution of 2
These experiments revealed that the activity and were performed with mutant G105A in order to con-
enantioselectivity of both mutants differed considera- firm the results from the analytical scale experiments
bly with variation of the substrate structure (Table 1). (Table 2).
Mutant G105A showed excellent enantioselectivity
The absolute configuration of several products was
(E>100) towards 2 and all p-substituted substrates 4, inferred from the known enantiopreference of BS2
7 and 8. In contrast, all other compounds were hardly G105A towards the analogue substrate 1.[6] Thus, for
converted at all or a substantial decrease in enantiose- 2 ,4 ,7 and 8, an (S)-preference of the enzyme was as-
lectivity (44% conversion, but E=6 for acetate 3) sumed. In the case of 3, 5, 6 and 9–11, however, the
1394
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1393 – 1398

Highly Enantioselective Synthesis of Arylaliphatic Tertiary Alcohols
FULL PAPERS
Table 1. Results of analytical scale kinetic resolutions of acetates 2–11.
Substrate
Mutant
Units[a]
Enantiomeric excess
[% eeS][b] [% eeP][b]
Conversion
E
G
A
[%][c]
[ ][c]
2
3
4
5
6
7
8
9
10
11
2
3
4
5
6
7
8
G105A
G105A
G105A
G105A
G105A
G105A
G105A
G105A
G105A
G105A
E188D
E188D
E188D
E188D
E188D
E188D
E188D
E188D
E188D
E188D
12
12
12
12
12
12
20
12
20
20
10
10
10
10
10
10
10
10
20
20
94
45
67
96
57
99
83
n.d.
99
>99
60
32
n.d.
94
n.d.
99
n.d.
n.d.
99
n.d.
59
n.d.
n.d.
49
44
40
8
<1
44
36
10
4
<1
51
<1
16
1
<1
4
<1
60
<1
<1
>100
6
>100
12
n.d.
>100
>100
4
n.d.
n.d.
>100
n.d.
>100
n.d.
n.d.
n.d.
n.d.
11
7
n.d.[d]
77
57
7
n.d.
n.d.
99
n.d.
19
n.d.
n.d.
4
n.d.
88
9
10
11
n.d.
n.d.
n.d.
n.d.
[a]
The specific activity of BS2 G105A in the hydrolysis of pNPA was 0.8 UmgÀ1 lyophilisate and 0.3 UmgÀ1 for BS2 E188D.
Determined by chiral GC analyses.
Conversion after 15 min, Evalues were calculated from eeS and eeP.
n.d.: not determined.
[b]
[c]
[d]
Table 2. Results of BS2 G105A-catalysed preparative scale
kinetic resolutions of acetates 2, 4, 7 and 8.
Substrate Acetate
Alcohol
Conversion E
[%][a]
[%
[%][a]
[%
[%][c]
[ ][c]
eeS][b]
eeP][b]
2
4
7
8
40
39
47
43
95
80
>99
97
49
21
39
48
95
>99
97
51
44
51
50
>100
>100
>100
>100
>99
[a]
[b]
[c]
Isolated yield.
Determined by chiral GC analysis.
Conversion after 15 min, Evalues were calculated from
eeS and eeP.
lower enantioselectivity and/or activity of the BS2
mutants did not allow a tentative assignment of the
absolute configuration.
Molecular modelling was used to get an insight into
the substrate specificities and enantioselectivities of
the esterase mutants studied. Glycine 105 occupies a
position close to the entrance of the active site of es-
terase BS2. A substitution by alanine decreases the
available space for the larger substituent of the tetra-
hedral intermediate of the slower reacting enantiomer
Figure 1. View of the active site of BS2 esterase. Ser189 and
His399 from the catalytic site and the amino acids bearing
the mutations Gly105 and Glu188 are highlighted.
(Figure 1). Hence, this might explain the observed in- explained by steric hindrance. Glu188 is neighbouring
crease in enantioselectivity.[10] However, the effect of the catalytically active serine (Figure 1) and the in-
a substitution of glutamine 188 by aspartate cannot be crease in enantioselectivity for 1 and 2 compared to
Adv. Synth. Catal. 2007, 349, 1393 – 1398
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1395

Robert Kourist et al.
FULL PAPERS
AHCTREUNG
the BS2 wild-type was attributed to an involvement of
this amino acid in the hydrogen bonding network of
the catalytic site during catalysis.[11] However, the
model cannot explain why E188D did not convert
substrates 3, 5, 7 and 8 towards which G105A is still
active.
as a colourless oil after bulb-to-bulb distillation; yield: 3.9 g
1
(18 mmol, 67%). H NMR: d=2.5 (1H, s, CCH), 2.6 (1H, s,
OH), 3.1 (2H, dd. CH2), 7.3 (5H, m, Ar-H); 13C NMR: d=
40.6 (s), 71.7 (q, J=32 Hz), 77.3 (s), 78.3 (s), 123.1 (q, J=
278 Hz), 127.6, 128.2, 131.1, 133.1; MS (EI): m/z=214 (M+),
196, 156, 91, 65.
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-(4’-fluorophenyl)-but-3-yn-2-ol
(4a): Obtained as a colourless oil after bulb-to-bulb distilla-
1
Conclusions
tion; yield: 2.8 g (13 mmol, 50%). H NMR: d=2.8 (1H, s,
CCH), 3.4 (1H, s, OH) 7.1, 7,7 (4H, m, Ar-H); 13C NMR:
d=72.4 (q, J=33 Hz), 76.9 (s), 79.3 (s), 115.4 (d, J=21 Hz),
123.0 (q, J=284 Hz), 129.2, (d, J=8 Hz), 130.5 (s), 163.5 (d,
J=249 Hz); MS (EI): m/z=218 (M+), 201, 149, 123, 95, 53.
In summary, we could show in this work that several
acetates of a-acetylenic tertiary alcohol could be re-
solved by BS2 esterase mutants in excellent yields
and optical purities. However, the substrate range of
these esterases appears to be rather narrow, as small
variations in the substrate structure resulted in consid-
erable losses of enzyme activity and/or enantioselec-
tivity.
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-(3’-fluorophenyl)-but-3-yn-2-ol
(5a): Obtained as a yellowish oil after bulb-to-bulb distilla-
1
tion; yield: 4.5 g (21 mmol, 80%). H NMR: d=2.8 (1H, s,
CH), 3.5 (1H, s, OH), 7.1 (1H, m, Ar-H), 7.5 (3H, m, Ar-
H); 13C NMR: d=72.5 (q, J=33 Hz), 76.9 (s),79.0 (s), 114.5
(d, J=24 Hz), 116.7 (d, J=21 Hz), 122.9 (s), 123.0 (q, J=
286 Hz), 137.13 (d, J=8 Hz), 162.5 (d, J=247 Hz); MS (EI):
m/z=218 (M+), 200, 149, 123, 53.
Experimental Section
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-(3’,4’-difluorophenyl)-but-3-yn-2-ol
(6a): Obtained as a colourless oil after bulb-to-bulb distilla-
1
General Remarks
tion; yield: 860 mg (3.6 mmol, 88%). H NMR: d=2.9 (1H,
s, CH), 3.4 (1H, s, OH), 7.1, 7.6 (3H, m, Ar-H); 13C NMR:
d=72.0 (q, J=34 Hz), 77.7 (s), 78.8 (s), 116.7 (d, J=14 Hz),
117.3 (d, J=12 Hz), 123.6 (s), 124.6 (d, J=285 Hz), 131.6
(s), 150.2 (dd, J=22 Hz, J=250 Hz), 151.4 (dd, J=16 Hz,
J=254 Hz); MS (EI): m/z=236 (M+), 219, 197, 187, 167.
All chemicals were purchased from Fluka (Buchs, Switzer-
land), Sigma (Steinheim, Germany) and Merck (Darmstadt,
Germany), unless stated otherwise. NMR spectroscopy ex-
periments were performed on an ARX300 (300.13 MHz for
1H and 75.5 MHz for 13C), a Bruker Avance II (600.7 MHz
for 1H and 150.1 MHz for 13C) or a Bruker Avance 600
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-(4’-chlorophenyl)-but-3-yn-2-ol
(7a): Obtained as a yellowish oil after bulb-to-bulb distilla-
device (600.13 MHz for H and 150.92 MHz for 13C, Bruker,
1
1
tion; yield: 2.2 g (9.4 mmol, 39%). H NMR: d=2.8 (1H, s,
Karlsruhe, Germany), using the d scale (ppm) for chemical
shifts; 13C NMR spectra were edited using DEPT tech-
niques. Optical rotations were recorded on a Polarimat
(Carl Zeiss, Jena). Mass spectra were recorded on a QP2010
GC-MS device (electron impact, 70 eV, Shimadzu, Japan).
CCH), 3.4 (1H, s, OH), 7.4, 7.7 (4H, m, Ar-H); 13C NMR:
d=72.5, (q, J=32 Hz), 74.6 (s), 79.1 (s), 121.9 (q, J=
285 Hz), 128.3 (s), 128.6 (s), 133.2 (s), 135.9 (s); MS (EI): m/
z=234 (M+), 217, 195, 165, 53.
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-(4’-methylphenyl)-but-3-yn-2-ol
(8a): Obtained as a colourless oil after bulb-to-bulb distilla-
1
tion; yield: 720 mg (3.4 mmol 63%). H NMR: d=2.4 (3H,
Molecular Modelling
s, CH3), 2.8 (1H, s, OH), 3.0 (1H, s, CCH), 7.2, 7.6 (5H, m,
Ar-H); 13C NMR: d=21.2 (s), 72.7 (q, J=33 Hz), 77.0 (s),
79.6 (s), 123.2 (q, J=285 Hz), 125.9 (s), 129.0 (s), 131.6 (s),
139.7 (s); MS (EI): m/z=214 (M+), 145, 115, 91, 53.
Molecular modelling of BS2 esterase was performed as de-
scribed.[11]
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-thiophenbut-3-yn-2-ol (9a): Ob-
General Procedure for the Synthesis of Tertiary
Alcohols
tained as a colourless oil after bulb-to-bulb distillation;
1
yield: 2.6 g (13 mmol, 47%). H NMR: d=2.8 (1H, s, OH),
3.3 (1H, s, CH), 7.0 [1H, m, C(CH)S], 7.4 [2H, m,
SC(CH)]; 13C NMR: d=76.6 (s), 76.8 (q, J=285 Hz), 78.7
(s), 122.6 (q, J=285 Hz), 125.5 (s), 126.9 (s), 128.2 (s), 137.9
(s); MS (EI): m/z=206 (M+), 189, 167, 137, 53.
Tertiary alcohols were prepared by adapting a previously re-
ported procedure.[8]
A
solution of 30 mmol ketone
(1.0 equiv.) in tetrahydrofuran (20 mL, THF) was added to a
refluxing solution of ethynylmagnesium bromide (1.2
equivs.) in THF (70 mL). The mixture was refluxed until the
ketone was consumed (control via thin layer chromatogra-
phy). Water (20 mL) was added to the cooled reaction mix-
ture and the product was extracted into diethyl ether (3”
100 mL), washed with saturated brine (100 mL) and evapo-
rated under reduced pressure. The residue was distilled or
purified by column chromatography.
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-phenylpent-3-yn-2-ol (10a): Ob-
tained as a colourless oil after bulb-to-bulb distillation;
1
yield: 3.9 g (18 mmol, 63%). H NMR: d=2.0 (3H, s, CH3),
2.9 (1H, s, OH), 7.4, 7.7 (5H, m, Ar-H); 13C NMR: d=3.7
(s), 72.9 (q, J=32 Hz), 75.4 (s), 85.3 (s), 123.4 (q, J=
285 Hz), 127.2 (s), 128.1 (s), 129.1 (s), 135.6 (s); MS (EI): m/
z=214 (M+), 195, 177, 164, 145, 67.
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-phenylbut-3-en-2-ol (11a): Ob-
ACHTREUNG(R,S)-1,1,1-Trifluoro-2-phenylbut-3-yn-2-ol (2a): Obtained
tained as a colourless oil after column chromatography;
yield: 4.0 g (19.8 mmol, 70%). 1H NMR: d=2.7 (1H, s,
OH), 5.6 (2H, dd, CH2), 6.5 (1H, dd, CH), 7.4, 7.6 (5H, m,
as a colourless oil after bulb-to-bulb distillation; yield: 4.4 g
(22 mmol, 76%). The spectroscopic data matched literature
data.[11]
1396
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1393 – 1398

Highly Enantioselective Synthesis of Arylaliphatic Tertiary Alcohols
FULL PAPERS
Ar-H); 13C NMR: d=77.5 (q, J=29 Hz), 118.3 (s), 124.7 (q,
J=283 Hz), 126.9 (s), 128.0 (s), 128.5 (s), 135.6 (s), 136.9 (s);
MS (EI): m/z=202 (M+), 165, 133, 55.
2.1 (3H, s, CH3), 2.3 (3H, s, CH3), 2.9 (1H, s, CCH), 7.2, 7.5
(4H, m, Ar-H); 13C NMR: d=21.2 (s), 21.3 (s), 75.3 (s), 76.7
(q, J=32 Hz), 78.6 (s), 122.1 (q, J=284 Hz), 127.8 (s), 129.0
(s), 129.4 (s), 139.8 (s), 166.9 (s); MS (EI): m/z=256 (M+),
241, 213, 197, 43.
General Procedure for the Synthesis of Tertiary
Alcohol Acetates
ACHTRE(UNG R,S)-1,1,1-Trifluoro-2-thiophenbut-3-yn-2-yl acetate (9):
Obtained as a colourless oil after column chromatography;
yield: 656 mg, 2.6 mmol (65%) 1H NMR: d=2.1 (3H, s,
CH3), 2.9 (1H, s, CCH), 7.0 [1H, m, C(CH)S], 7.4 [2H, m,
SC(CH)]; 13C NMR: d=21.3 (s), 73.9 (q, J=34 Hz), 74.8 (s),
78.3 (s), 125.3 (q, J=283 Hz), 126.9 (s), 128.0 (s), 129.2 (s),
135.3 (s), 166.8 (s); MS (EI): m/z=248 (M+), 228, 206, 189,
43.
To 4.5 mmol of the alcohol and dimethylaminopyridine
(6 equivs.) in dry dichloromethane (60 mL), acetyl chloride
(4 equivs.) was added dropwise and the solution was stirred
overnight. The mixture was washed with 1M HCl (3”
100 mL) and distilled water (3”100 mL). The organic layer
was dried over anhydrous Na2SO4 before the solvent was re-
moved under reduced pressure. For purification column
chromatography was performed.
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-phenylpent-3-yn-2-yl acetate (10):
Obtained as a colourless oil after column chromatography;
yield: 746 mg (2.9 mmol, 71%). 1H NMR: d=2.0 (3H, s,
CC-CH3), 2.2 (3H, s, CH3), 2.9 (1H, s, OH), 7.4, 7.6 (5H, m,
Ar-H); 13C NMR: d=3.8 (s), 21.4 (s), 71.0 (s), 77.8 (q, J=
32), 87.3 (s), 122.4 (q, J=284), 127.3 (s), 128.0 (s), 129.4 (s),
133.4 (s), 166.8 (s); MS (EI): m/z=256 (M+), 241, 213, 197,
43.
ACHTREUNG(R,S)-1,1,1-Trifluoro-2-phenylbut-3-yn-2-yl acetate (2):
Obtained as a colourless oil after column chromatography;
yield: 0.92 g (3.8 mmol, 82%). The spectroscopic data
matched literature data.[11]
ACHTREUNG(R,S)-1,1,1-Trifluoro-2-benzylbut-3-yn-2-yl acetate (3):
Obtained as a colourless oil after column chromatography;
AHCTRE(UNG R,S)-1,1,1-Trifluoro-2-Phenylbut-3-en-2-yl acetate (11):
1
yield: 652.3 mg (3.0 mmol, 24%). H NMR: d=2.1 (3H, s,
Obtained as a colourless oil after column chromatography;
yield: 547 mg (2.3 mmol, 47%). 1H NMR: d=2.2 (3H, s,
CH3), 5.6 (2H, m, CH2), 6.6 (1H, dd, CH), 7.4 (5H, m, Ar-
H); 13C NMR: d=21.9 (s), 83.2 (q, J=32 Hz), 123.6 (q, J=
286 Hz), 126.5 (s), 127.2 (s), 128.6 (s), 129.2 (s), 130.9 (s),
134.2 (s), 167.6 (s); MS (EI): m/z=244 (M+), 202, 184, 165,
43.
CH3), 2.8 (1H, s, CCH), 3.4 (2H, dd, CH2), 7.2 (5H, m, Ar-
H); 13C NMR: d=21.5 (s), 39.7 (s), 75.0 (s), 75.2 (q, J=
31 Hz), 79.2 (s), 122.6 (q, J=288 Hz), 127.6 (s), 128.1 (s),
130.9 (s), 132.8 (s), 167.6 (s); MS (EI): m/z=214
(M+ÀCH3CO), 196, 146, 91, 43.
ACHTREUNG(R,S)-1,1,1-Trifluoro-2-(4’-fluorophenyl)-but-3-yn-2-yl ace-
tate (4): Obtained as a yellow oil after column chromatogra-
1
phy; yield: 1.0 g (3.8 mmol, 84%). H NMR: d=2.2 (3H, s,
CH3), 2.9 (1H, s, CCH), 7.1, 7.6 (4H, m, Ar-H); 13C NMR: General Procedure for Esterase-Catalyzed Small-
d=21.3 (s), 74.9 (s), 76.3 (q, J=34 Hz), 79.0 (s), 115.5 (d,
J=22 Hz), 122.1 (q, J=284 Hz),128.3 (s), 129.0 (d, J=8 Hz),
163.6 (d, J=249 Hz), 166.8 (s); MS (EI): m/z=260 (M+),
217, 201, 189, 162, 43.
Scale Resolutions
To a stirred solution of acetate (25 mM) in phosphate buffer
(100 mM, pH 7.5) and the appropriate amount of cosolvent
DMSO [10–20% (v/v)], the appropriate amount of esterase
solution was added to a total volume of 1 mL. The reaction
mixture was stirred in a thermoshaker (Eppendorf, Ham-
burg, Germany) at 378C for 15 min. The reaction mixture
was extracted twice with 400 mL dichloromethane, the com-
bined organic layers were dried over anhydrous sodium sul-
phate and the organic solvent was removed under nitrogen.
Enantioselectivity and conversion were calculated according
to Chen et al.[12]
ACHTREUNG(R,S)-1,1,1-Trifluoro-2-(3’-fluorophenyl)-but-3-yn-2-ace-
tate (5): Obtained as a colourless oil after column chroma-
tography; yield: 1.18 g (4.9 mmol, 97%). 1H NMR: d=2.1
(3H, s, CH3), 2.9 (1H, s, CH), 7.1, 7.4 (4H, m, Ar-H);
13C NMR: d=21.2 (s), 74.7 (s), 76.2 (q, J=32 Hz), 79.0 (s),
114.4 (d, J=26 Hz), 117.0 (d, J=21 Hz), 122.6 (s) 130.1 (d,
J=8 Hz), 182.7 (d, J=247 Hz), 166.78; MS (EI): m/z=245
(M+ÀCH3), 225, 218, 201, 198, 43.
ACHTREUNG(R,S)-1,1,1-Trifluoro-2-(3’,4’-difluorophenyl)-but-3-yn-2-yl
acetate (6): Obtained as a colourless oil after column chro-
1
matography; yield: 326 mg (1.2 mmol, 63%). H NMR: d=
General Procedure for Esterase-Catalyzed Kinetic
Resolution on a Preparative Scale
2.2 (3H, s, CH3), 3.0 (1H, s, CH), 7.2, 7.5 (3H, m, Ar-H);
13C NMR: d=21.1 (s), 74.4 (s), 76.2 (q, J=29 Hz), 79.3 (s),
116.7 (d, J=20 Hz), 117.5 (d, J=18 Hz), 123.6 (s), 124.6 (q,
J=286 Hz), 129.4 (s), 150.3 (dd, J=11 Hz, J=256 Hz), 151.3
(dd, J=8 Hz, J=248 Hz), 166.7 (s). MS (EI): m/z=278
(M+), 243, 236, 219, 43.
BS2 G105A lyophilisate with an activity of 0.8 U/mg (1500
U for 2 and 300 U for 4, 7 and 8, respectively) was added to
a stirred solution of ester (2.3 mmol for 2, 0.4 mmol for 4, 7
and 8) in 10% DMSO (v/v) and phosphate buffer (100 mM,
pH 7.5) to a total volume of 100 mL for 2 and 20 mL for 4, 7
and 8, respectively. The reaction mixture was stirred for
15 min at 308C. The reaction mixture was extracted twice
with 10 mL dichloromethane. The organic layers were com-
bined and washed with brine (2”10 mL), water (2”10 mL)
and dried over anhydrous NaHSO4. The solvent was re-
moved under reduced pressure. Purification was performed
by column chromatography.
ACHTREUNG(R,S)-1,1,1-Trifluoro-2-(4’-chlorophenyl)-but-3-yn-2-yl
acetate (7): Obtained as a yellow oil after column chroma-
tography; yield: 1.0 g (3.6 mmol, 84%). 1H NMR: d=2.1
(3H, s, CH3), 2.9 (1H, s, CCH), 7.4, 7.6 (4H, m, Ar-H);
13C NMR: d=21.1 (s), 74.7 (s), 76.6 (q, J=32 Hz), 79.1 (s)
121.9 (q, J=284), 128.3 (s), 128.8 (s), 130.9 (s), 136.1 (s),
166.7 (s); MS (EI): m/z=276 (M+), 233, 217, 198, 43.
ACHTREUNG(R,S)-1,1,1-Trifluoro-2-(4’-methylphenyl)-but-3-yn-2-yl
acetate (8): Obtained as a colourless oil after column chro-
matography; yield: 284 mg (1.1 mmol, 81%). H NMR: d=
(À)-1,1,1-Trifluoro-2-phenyl-but-3-yn-2-ol (2a): Yield:
1
50% (227 mg); 95% ee; [a]2D1: À3.38 (CHCl3, c 1.8).
Adv. Synth. Catal. 2007, 349, 1393 – 1398
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1397

Robert Kourist et al.
FULL PAPERS
Table 3. Details of chiral GC analyses.
Compound Tcolumn [8C][a] Column Retention time [min]
(+)-1,1,1-Trifluoro-2-phenyl-but-3-yn-2-yl acetate (2):
Yield: 40% (220 mg); 95% ee; [a]21: +32.68 (CHCl3, c 1.2).
(À)-1,1,1-Trifluoro-2-(4’-fluorophDenyl)-but-3-yn-2-ol (4a):
Yield: 21% (18 mg); 99% ee; [a]2D1: À6.78 (CHCl3, c 1.8).
(À)-1,1,1-Trifluoro-2-(4’-fluorophenyl)-but-3-yn-2-yl ace-
tate (4): Yield: 39% (40 mg); 80% ee; [a]2D1: À2.9 (CHCl3, c
1.4).
(+)-1,1,1-Trifluoro-2-(4’-chlorophenyl)-but-3-yn-2-ol (7a):
Yield: 39% (36 mg); 97% ee; [a]2D1: +4.68 (CHCl3, c 2.6).
(À)-1,1,1-Trifluoro-2-(4’-chlorophenyl)-but-3-yn-2-yl ace-
tate (7): Yield: 47% (51 mg);>99% ee; [a]2D1: À35.78
(CHCl3, c 1.1).
(+)-1,1,1-Trifluoro-2-(4’-methylphenyl)-but-3-yn-2-ol (8a):
Yield: 48% (41 mg); >99% ee; [a]2D1: +27.68 (CHCl3, c 2.5).
(À)-1,1,1-Trifluoro-2-(4’-methylphenyl)-but-3-yn-2-yl ace-
tate (8): Yield: 43% (44 mg); 97% ee; [a]2D1: À1.68 (CHCl3, c
1.5).
2a
2
3a
3
4a
4
5a
5
6a
6
7a
7
8a
8
9a
9
10a
10
11a
11
P1
P1
120
110
P1
P1
P1
P1
P1
P1
P2
P2
P1
P1
P3
P3
P1
-
A
A
B
C
A
A
A
A
A
A
A
A
A
A
C
C
C
-
29.6 (S)
15.2 (S)
11.8
31.1 (R)
15.8 (R)
14.2
17.3
18.8
31.1 (R)
15.2 (S)
33.7
16.0
34.7
36.0 (S)
16.1 (R)
37.4
17.3
40.1
13.8
14.7
28.0 (S)
14.3 (R)
35.1 (R)
21.5 (S)
35.9
16.3
30.3
n.d.
14.5
31.9 (R)
15.0 (S)
49.4 (S)
22.5 (R)
37.3
17.5
31.3
n.d.
15.1
Production of BS2 Esterase Mutants
BS2 esterase mutants G105A and E188D were produced as
described[11] by expression in E. coli using a rhamnose induc-
tion system. The esterase mutants were isolated by cell dis-
ruption and the supernatant was directly used for biocataly-
sis. Esterase activity was determined spectrophotometrically
by hydrolysis of p-nitrophenol acetate (pNPA, 10 mM in
DMSO) in sodium phosphate buffer (100 mM, pH 7.5). p-
Nitrophenol released was quantified at 410 nm (e=15”
103 MÀ1 cmÀ1). One Unit (U) of activity was defined as the
amount of enzyme releasing 1 mmol p-nitrophenol per min
under assay conditions.
100
100
C
C
7.9
8.7
[a]
P1: temperature program: 15 min 1008C – 48C minÀ1
–
1208C; P2: temperature program: 15 min 1258C – 48C
minÀ1 – 1458C; P3: temperature program: 15 min 908C –
48C minÀ1 – 1158C.
References
[1] D. OꢂHagan, N. A. Zaidi, R. B. Lamont, Tetrahedron:
Asymmetry 1993, 4, 1703–1708.
[2] R. Riveiros, D. Rodriguez, J. P. Sestelo, L. A. Sarande-
Chiral GC Analyses
GC analyses were carried with chiral columns. Hydrodexꢁ-
b-3B [heptakis-(2,6-di-O-methyl-3-O-pentyl-b-cyclodextrin)]
was used on a GC-MS QP2010 (Shimadzu, Japan, Column
A) and on a GC-14 A gas chromatograph (Shimadzu,
Tokyo, Japan, Column C), the column heptakis-(2,3-di-O-
acetyl-6-O-tert-butyldimethylsilyl)-b-cyclodextrin, provided
by Prof. Kçnig (University of Hamburg, Germany) was used
on a GC-14 A gas chromatograph (Shimadzu, Tokyo, Japan,
column B). Details are given in Table 3.
ses, Org. Lett. 2006, 8, 1403–1406.
[3] G. Yao, S. Haque, L. Sha, G. Kumaravel, J. Wang, T. M.
Engber, E. T. Whalley, P. R. Conlon, H. X. Chang,
W. F. Kiesman, R. C. Petter, Bioorg. Med. Chem. Lett.
2005, 15, 511–515.
[4] F. M. D. Ismail, J. Fluor. Chem. 2002, 118, 27–33.
[5] U. T. Bornscheuer, R. J. Kazlauskas, Hydrolases in Or-
ganic Chemistry, 2nd edn., Wiley-VCH Verlag GmbH,
Weinheim, 2005.
The absolute configuration of 2, 4, 7, and 8 was inferred
from the known enantiopreference of BS2 G105 A towards
the similar substrate 3-phenylbut-1-yn-3-yl acetate.[10]
[6] E. Henke, J. Pleiss, U. T. Bornscheuer, Angew. Chem.
Int. Ed. 2002, 41, 3211–3213.
[7] A. Ordentlich, D. Barak, C. Kronman, N. Ariel, Y.
Segall, B. Velan, A. Shafferman, J. Biol. Chem. 1998,
273, 19509–19517.
[8] D. OꢂHagan, N. A. Zaidi, J. Chem. Soc., Perkin Trans. 1
1992, 947–949.
[9] D. OꢂHagan, N. A. Zaidi, Tetrahedron: Asymmetry
Acknowledgements
We thank Anita Gollin for support in chemical syntheses and
Dr. Bettina Appel and Dr. Markus Kindermann (all Institute
of Biochemistry, Greifswald University) for helpful discus-
sions about NMR-spectroscopy. Financial support by the
Deutscher Akademischer Austauschdienst (DAAD, Germa-
ny) Deutsche Bundesstiftung Umwelt (DBU, Osnabrück,
Germany, AZ13162) is gratefully acknowledged.
1994, 5, 1111–1118.
[10] E. Henke, U. T. Bornscheuer, R. D. Schmid, J. Pleiss,
ChemBioChem. 2003, 4, 485–493.
[11] B. Heinze, R. Kourist, L. Fransson, K. Hult, U. T. Born-
scheuer, Prot. Eng. Des. Sel. 2007, 20, 125–131.
[12] C. S. Chen, Y. Fujimoto, G. Girdaukas, C. J. Sih, J. Am.
Chem. Soc. 1982, 104, 7294–7299.
1398
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 1393 – 1398