Enzymatic kinetic resolution of racemic ketones catalyzed by Baeyer-Villiger monooxygenases

Article abstract of DOI:10.1016/j.tetasy.2007.05.033

A set of racemic cyclic and linear ketones, as well as 2-phenylpropionaldehyde, were tested as substrates in the enzymatic Baeyer-Villiger oxidation catalyzed by two Baeyer-Villiger monooxygenases: phenylacetone monooxygenase (PAMO) and 4-hydroxyacetophenone monooxygenase (HAPMO). Excellent enantioselectivites (E > 200) can be obtained in the kinetic resolution processes depending on the substrate structure and the reaction conditions. The parameters affecting the biocatalytic properties of these enzymes were also studied, in order to establish a deeper understanding of these novel biocatalysts.

Full text of DOI:10.1016/j.tetasy.2007.05.033

Tetrahedron: Asymmetry 18 (2007) 1338–1344
Enzymatic kinetic resolution of racemic ketones catalyzed
by Baeyer–Villiger monooxygenases
a
a
b
a,
*
Cristina Rodr ´ı guez, Gonzalo de Gonzalo, Marco W. Fraaije and Vicente Gotor
a
Departamento de Qu ı´ mica Org a´ nica e Inorg a´ nica, Instituto Universitario de Biotecnolog ı´ a de Asturias, Universidad de Oviedo,
3006 Oviedo, Asturias, Spain
Laboratory of Biochemistry, Groningen Biomolecular Sciences and Biotechnology Institute, University of Groningen, Nijenborgh 4,
747 AG Groningen, The Netherlands
3
b
9
Received 10 April 2007; accepted 29 May 2007
Abstract—A set of racemic cyclic and linear ketones, as well as 2-phenylpropionaldehyde, were tested as substrates in the enzymatic Bae-
yer–Villiger oxidation catalyzed by two Baeyer–Villiger monooxygenases: phenylacetone monooxygenase (PAMO) and 4-hydroxyace-
tophenone monooxygenase (HAPMO). Excellent enantioselectivites (E > 200) can be obtained in the kinetic resolution processes
depending on the substrate structure and the reaction conditions. The parameters affecting the biocatalytic properties of these enzymes
were also studied, in order to establish a deeper understanding of these novel biocatalysts.
ꢀ
2007 Elsevier Ltd. All rights reserved.
1
. Introduction
Baeyer–Villiger monooxygenases (BVMOs) represent a
class of enzymes widely used in the oxidation of linear
and cyclic ketones. BVMOs are nicotinamide cofactor
dependent flavoproteins that catalyze a set of oxidative
reactions, in addition to the nucleophilic oxidation of the
carbonyl group and boron atoms, electrophilic oxygen-
ations of different heteroatoms, such as sulfur, nitrogen,
Nowadays, the Baeyer–Villiger oxidation of ketones is one
of the key reactions in synthetic processes. The transfor-
1
mation of ketones into the corresponding esters and lac-
tones, valuable intermediates in organic chemistry and
frequently used precursors in enantioselective synthesis,
can be carried out chemically by employing peroxides or
peracids. While these oxidants are very effective, they also
represent toxic and labile reagents. To avoid the use of
these reactive substances, transition metal catalysts and
4
selenium and phosphorous are also catalyzed. Over the
last few decades, a large number of these biocatalysts have
been discovered but only a few BVMOs were employed and
studied for synthetic purposes. The favorite enzyme for
BVMO-mediated catalysis was cyclohexanone monooxy-
genase from Acinetobacter calcoaceticus NCIMB 9871, a
versatile enzyme, which is able to oxidize a wide range of
2
organocatalytic compounds have been developed, which
use hydrogen peroxide or oxygen as milder oxidants. How-
ever, these novel methodologies in most cases have demon-
strated a lack of selectivity. In recent years, the enzymatic
approach to perform Baeyer–Villiger reactions has been
5
substrates.
3
successfully employed. By applying these novel biocata-
Due to advances in biochemistry and the availability of
new (recombinant) enzymes, BVMOs have received
lytic methods, optically active lactones and esters can be
obtained in processes carried out in water and worked with
under mild and environmental friendly conditions, with
high regio- and/or enantioselectivities. All of these advan-
tages make the biocatalytic approach most appealing for
the industrial application of the Baeyer–Villiger oxidation.
6
increasing interest in recent years. Clear examples of this
trend are the newly cloned and overexpressed BVMOs
7
phenylacetone monooxygenase (PAMO) and 4-hydroxy-
8
acetophenone monooxygenase (HAPMO).
PAMO (EC 1.14.13.92) is a monomeric and thermostable
NADPH-dependent BVMO isolated from Thermobifida
fusca. Previous studies reported phenylacetone as its best
substrate, but PAMO is also able to oxidize aliphatic
*
Corresponding author. Tel./fax: +34 985 103448; e-mail: vgs@fq.
uniovi.es
0957-4166/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.05.033

C. Rodr ı´ guez et al. / Tetrahedron: Asymmetry 18 (2007) 1338–1344
1339
ketones, sulfides, racemic sulfoxides, amines and the boron
atom. This biocatalyst can even perform oxidations when
Table 1. BVMO-catalyzed oxidation of 4-methylcyclohexanone to (S)-4-
methyl-6-hexanolactone
a
9
working in non-conventional reaction media (mixtures of
aqueous buffer–water miscible or immiscible organic cos-
olvent), which is accompanied by interesting changes in
its biocatalytic properties (e.g., improvement or reversal
O
O
BVMO/ Buffer
O
G6P/G6PDH/NADP+
T (ºC)/ 250 rpm
1
0
of enantioselectivity). PAMO is the first BVMO for
which the crystal structure has been elucidated by X-ray
analysis, which has provided details on the catalyst perfor-
1a
(
S
)-1b
b
b
1
1
Entry
Enzyme
pH
T (ꢁC)
c
(%)
ee (%)
mance at an atomic level. HAPMO (EC 1.14.13.84),
obtained from Pseudomonas fluorescens ACB, is a homo-
dimeric NADPH-dependent BVMO, that is, primarily
active with a wide range of aromatic ketones, especially
when presenting an electron-donating group at the para-
1
2
3
4
PAMO
PAMO
PAMO
PAMO
PAMO
HAPMO
HAPMO
HAPMO
8
9
9
10
9
8
30
20
30
30
30
20
20
30
61
6
15
19
8
61
9
14
—
40
32
28
38
—
44
42
c
5
1
2
position. This enzyme can also oxidize aliphatic ketones
6
7
8
1
3
and also catalyze enantioselective sulfoxidations.
9
9
Herein we report the Baeyer–Villiger oxidation of different
racemic carbonylic compounds catalyzed by the above-
mentioned BVMOs, in order to obtain enantiopure ketones
and corresponding chiral lactones or esters.
a
ꢀ1
Reaction time, 48 h and substrate concentration: 2 g L . For other
details, see Section 4.
Determined by GC.
b
c
Reaction performed in 1% MeOH as cosolvent.
2
. Results and discussion
Finally, the oxidation of phenylcyclohexanones was inves-
tigated, aided by the fact that PAMO and HAPMO are pri-
marily active on aromatic ketones. After long incubation
times, no reaction was observed with either (± )-2-phenyl-
or 4-phenylcyclohexanone. This may be due to the low sol-
ubility of these substrates in aqueous buffer.
2
.1. Enzymatic oxidation of cyclic ketones
The enzymatic Baeyer–Villiger oxidations of ketones to the
corresponding esters and lactones by recombinant PAMO
and HAPMO were coupled to an auxiliary enzymatic
reaction in order to regenerate the nicotinamide cofactor
NADPH. As the NADPH recycling system, glucose-
6
(
-phosphate and glucose-6-phosphate dehydrogenase
2.2. Oxidation of 2-phenylpropionaldehyde catalyzed by
PAMO and HAPMO
G6PDH) were employed.14
Our first efforts were devoted to studying the Baeyer Vil-
liger oxidation of a set of cyclic ketones. Both (± )-2-meth-
ylcyclopentanone and (± )-2-methylcyclohexanone were
oxidized by PAMO and HAPMO, but in processes with
low selectivity (data not shown). Different reaction condi-
tions were employed, by modifying the pH and tempera-
ture, leading in all cases to the (S)-lactone, with only E
After these results, we focused on the enzymatic resolution
of different aromatic a-substituted carbonyl compounds
(aldehyde 2a and ketones 3a–6a), catalyzed by both PAMO
and HAPMO. In previous reports, it has been described
that phenylacetone was the best PAMO substrate and
could also be easily oxidized by HAPMO. It was observed
that for this set of compounds, higher conversions were
measured using PAMO. This is in agreement with the phys-
iological substrates for both enzymes: PAMO is primarily
active with phenylacetones, while HAPMO preferably
oxidizes acetophenones. Racemic ketones (± )-3a–6a were
1
5
values lower than 3. Oxidations conducted with PAMO
were generally faster than those performed with HAPMO.
Both BVMOs were also tested in the Baeyer–Villiger oxida-
tion of a prochiral ketone, 4-methylcyclohexanone 1a, as
shown in Table 1. Compound (S)-1b was obtained with
low to moderate conversions and enantiomeric excesses
depending on the conditions employed. Increasing the pH
and temperature led to a higher enzymatic activity, with
only a slight loss in selectivity. Compound (S)-1b can be
obtained with ee = 40% by performing the oxidation with
PAMO at pH 9 and 20 ꢁC (entry 2) or with ee = 44% by
carrying out the Baeyer–Villiger oxidation of 1a employing
HAPMO under the same conditions (entry 7). It was
observed that the addition of 1% MeOH to the reaction
medium when working at pH 9 and 30 ꢁC, can slightly
improve the PAMO selectivity with a loss in conversion
1
6
prepared as shown in Scheme 1, starting from the corre-
sponding commercially available phenylacetones 3–6.
Due to the structure of substrates 2a–6a, presenting a
hydrogen atom with a strong acidic character, we expected
that spontaneous racemization could occur via a keto–enol
tautomerization. Nevertheless, the enantiomeric excess of
optically active compounds (R)-2a–6a when submitted to
different reaction conditions, did not decrease after 24 h,
even when carrying out the processes at high pHs (pH 9–
10) and temperatures (40 ꢁC). This excludes the possibility
of performing dynamic kinetic resolutions of phenylalde-
hyde and phenylketones, as described in a previous paper
entry 5), as observed previously for this enzyme.10b Higher
for a similar substrate. Only PAMO- or HAPMO-cata-
17
(
contents of methanol resulted in no oxidation after long
reaction times.
lyzed kinetic resolution of the racemic carbonyl com-
pounds was observed.

1340
C. Rodr ı´ guez et al. / Tetrahedron: Asymmetry 18 (2007) 1338–1344
R1
O
3
-6
2
+
-
R I / Bu4N HSO4
CH2Cl2 / NaOH (2.0 M)
R2
R2
R2
O
R1
BVMO/ T (ºC)/ 250 rpm
R1
+
R1
O
O
O
(
±)-3a-6a
O2 + H+
(
R)-3a-6a
(S)-3b-6b
NADP+
NADPH
+
+ H O
2
D-Gluconate-6-P
D-Glucose-6-P
G6PDH
1
2
3
4
5
6
a: R =Et, R =Me
1
2
a: R =Me, R = Me
1
2
a: R =Me, R = Et
1
2
a: R = Et, R = Et
Scheme 1. Enzymatic resolution of different substituted phenylketones catalyzed by the Baeyer–Villiger monooxygenases PAMO and HAPMO.
First, the enzymatic Baeyer–Villiger oxidation of commer-
cial 2-phenylpropionaldehyde (± )-2a catalyzed by PAMO
and HAPMO was studied (Table 2). When working under
the same reaction conditions, higher selectivities and activ-
ities were obtained when employing PAMO, but for both
biocatalysts, moderate to low E-values were achieved.
Oxidation of 2a led in all cases to the formation of the
two effects in the enzymatic resolution of (± )-2a: (i) an
important increase in the enzyme activity: whereas at pH
6 there was no oxidation (entry 1), (S)-2b can be obtained
with 50% conversion at pH 10, and (ii) a slight decrease in
the E-values, especially at pH values higher than 9. This
correlation between the pH and enantioselective behaviour
of PAMO has been thoroughly studied in a recent publica-
tion and can be attributed to (de)protonation of specific
enzyme active-site moieties. By increasing the tempera-
ture when performing the PAMO-catalyzed oxidations at
pH 8, higher enzyme activities (at 40 ꢁC, it was possible
to obtain a c = 28% after 4 h) were accompanied by a loss
in E (entries 7 and 8).
(
S)-1-phenylethyl formate 2b, leaving the aldehyde with
1
8
an (R)-configuration.
The effect of pH when conducting the oxidations with
PAMO at 20 ꢁC was studied (entries 1–6). It was observed
that by increasing the pH of the Tris/HCl buffer, there were
a
Table 2. Enzymatic oxidation of (± )-2-phenylpropionaldehyde (± )-2a employing novel BVMOs
H
(
(
R)-2a
O
BVMO/ Buffer
H
+
G6P/G6PDH/NADP+
T (ºC)/ 250 rpm
O
O
S)-2b
O
H
(
±)-2a
b
b
c
Entry
BVMO
pH
T (ꢁC)
ee (%) (R)-2a
ee (%) (S)-2b
c (%)
E
1
2
3
4
5
6
7
8
9
0
1
PAMO
PAMO
PAMO
PAMO
PAMO
PAMO
PAMO
PAMO
HAPMO
HAPMO
HAPMO
6
7
8
9
9.5
10
8
8
8
9
9
20
20
20
20
20
20
30
40
20
20
30
—
9
50
62
65
76
53
30
9
—
92
88
85
81
76
83
75
88
77
83
61
8
—
26
25
23
19
17
20
9
36
42
44
50
38
28
9
d
17
9
10
1
1
12
14
13
15
a
b
c
ꢀ1
Reactions stopped after 8 h for PAMO and 30 h for HAPMO. [2a] = 2 g L . For other details, see Section 4.
Determined by GC.
Conversion, c = ee
s
s p
/(ee + ee ).
d
Reaction stopped after 4 h.

C. Rodr ı´ guez et al. / Tetrahedron: Asymmetry 18 (2007) 1338–1344
1341
The enzymatic Baeyer–Villiger oxidation of (± )-2a employ-
ing HAPMO exhibited the highest enantioselectivity
formed at low pH (6 and 7, entries 1 and 2, respectively),
there was an important loss in the conversions measured
(c < 10% after 1 h).
(
E = 17, entry 9) at pH 8 and 20 ꢁC. An increase in either
pH (entry 10) or temperature (entry 11) led to higher con-
versions and less selective resolutions.
As PAMO has been isolated from a thermophilic micro-
organism, the optimal temperature in terms of activity
and selectivity for the Baeyer–Villiger oxidation of (± )-3a
was analyzed. It should be noted that the oxidations are
performed in a double enzymatic system, so the tempera-
ture can also affect glucose-6-phosphate dehydrogenase.
High selectivities were achieved in the range from 15 to
50 ꢁC. Only when the oxidation was carried out at 60 ꢁC,
was there a slight decrease in the E-value (E = 148, entry
13). PAMO activity increased by working at higher temper-
atures, reaching a maximum value at 40 ꢁC (c = 20% after
30 min, entry 11). Increasing the temperature from this
value resulted in a negative effect on the PAMO-catalyzed
resolution, obtaining lower activities. It is worth noting
that oxidations with this biocatalyst can even be performed
at 60 ꢁC (c = 11% after 2 h, entry 13), demonstrating the
ability of PAMO to work even under drastic conditions.
2.3. PAMO and HAPMO catalyzed kinetic resolution of
phenylketones
The enzymatic oxidation of racemic 2-phenylpentan-3-one
(
± )-3a catalyzed by PAMO at pH 8 and 20 ꢁC (Table 3,
entry 4) led to (S)-3b, while the unreacted (R)-enantiomer
remained, in a highly selective process (E > 200) with a
c = 19% after 1 h.
The pH effect on this kinetic resolution was analyzed,
which revealed a similar trend as observed for 2-phenylpro-
pionaldehyde. The resolution of (± )-3a was carried out
with high selectivities until pH 9, enabling production of
enantiopure (R)-3a and (S)-3b depending on the conver-
sion. When working under more basic Tris/HCl buffer
media, there was a loss in selectivity, but even at pH 10 a
good enantioselectivity can be achieved (E = 111, entry
Baeyer–Villiger oxidation of (± )-3a with HAPMO at pH 8
and 20 ꢁC (Table 4, entry 4) also resulted in an excellent E-
value (E = 165), in a process with a conversion close to
50% (c = 46%) after 14 h. The (S)-enantiomer was mainly
7
). Enzyme activity became higher by increasing the pH
value. The highest conversion was observed at pH 10,
c = 32% after 1 h). Contrarily, when oxidations were per-
(
a
Table 3. Effect of pH and temperature in the enzymatic resolution of (± )-2-phenylpentan-3-one catalyzed by PAMO
b
b
c
Entry
pH
T (ꢁC)
Time (h)
ee (%) (R)-3a
ee (S)-3b (%)
c (%)
E
1
2
3
4
5
6
7
8
9
0
1
2
3
6
7
7.5
8
8.5
9
20
20
20
20
20
20
20
20
15
30
40
50
60
1
1
1
1
1
1
1
1
3
6
99
99
99
99
99
99
98
97
99
99
99
98
98
3
6
>200
>200
>200
>200
>200
>200
155
13
24
22
26
35
56
13
27
24
24
11
12
19
18
21
27
32
12
21
20
20
11
9.5
10
111
8
8
8
8
8
1
>200
>200
>200
193
1
1
1
1
0.75
0.5
0.75
2
148
a
b
c
ꢀ1
[
3a] = 2 g L . For other reaction details, see Section 4.
Determined by GC.
Conversion, c = ee /(ee
s
s p
+ ee ).
a
Table 4. HAPMO catalyzed oxidation of (± )-3a
b
b
c
Entry
pH
T (ꢁC)
Time (h)
ee (%) (R)-7a
ee (S)-7b (%)
c (%)
E
1
2
3
4
5
6
7
8
9
6
7
7.5
8
9
10
8
8
8
8
8
20
20
20
20
20
20
10
15
30
40
50
48
48
24
14
14
10
14
14
14
24
24
—
—
25
83
95
91
93
98
76
9
—
—
98
97
87
83
98
97
97
98
96
61
61
20
46
53
52
48
50
44
9
—
—
146
165
52
35
>200
>200
155
120
52
1
1
0
1
4
4
a
b
c
ꢀ1
[
3a] = 2 g L . For other details, see Section 4.
Determined by GC.
Conversion, c = ee /(ee
s
s p
+ ee ).

1342
C. Rodr ı´ guez et al. / Tetrahedron: Asymmetry 18 (2007) 1338–1344
ꢀ
1
oxidized in all tested conditions. The effect of pH on this
enzyme for the kinetic resolution of (± )-3a was analyzed.
No reaction was observed at pH 6 and 7, even after 48 h.
When performing the reactions at pH 7.5, a high E value
was obtained in a slow process (c = 20% after 24 h). The
highest conversions for the enzymatic resolutions were
measured at pH 9 and 10 (entries 5 and 6) but in these con-
ditions there was an important loss in HAPMO selectivity
reaching a maximum at 8 g L . After this value, the reac-
tion rate decreased, but the biocatalyst was still active.
PAMO was still able to oxidize (± )-3a at concentrations
ꢀ
1
higher than 20 g L . A very similar substrate concentra-
tion-activity profile (albeit with reduced reaction rate val-
ues) was achieved for HAPMO. This enzyme exhibited a
maximum reaction rate at 4 g L while higher concentra-
tions caused significant deactivation of the biocatalyst. At
ꢀ1
ꢀ
1
(
E = 55–32 for pH 9–10, respectively). The temperature
ketone concentrations higher than 10 g L
no (S)-3b
was also analyzed as an important parameter in the enzy-
matic resolution of (± )-3a. When performing the Baeyer–
Villiger oxidations at 10 and 15 ꢁC, high enantioselectivities
and conversions were obtained (entries 7 and 8). Both the
activity and the E-value became lower by increasing the
temperature. This effect was especially observed at 50 ꢁC,
as shown in entry 11: the oxidation took place with only
c = 4% after 24 h with good enantioselectivity (E = 52).
formation was observed even after long reaction times. A
decrease in the selectivity was also observed for HAPMO
when performing the oxidations at concentrations higher
than 4 g L , indicating that this enzyme is more sensitive
to substrate concentration than PAMO.
ꢀ1
Finally, the racemic phenylketones (± )-3-phenylbutan-2-
one 4a, (± )-3-phenylpentan-2-one 5a and (± )-4-phenyl-
hexan-3-one 6a, presenting a close structure to (± )-2-
phenylpentan-3-one, were also resolved with excellent
selectivities by both enzymes, when the Baeyer–Villiger oxi-
dations were performed at 20 ꢁC and pH 8 (Table 5). (S)-
Esters and (R)-ketones were obtained; the processes carried
out with PAMO, were faster and more selective than those
performed with HAPMO.
The effect of the 3a concentration on the biocatalytic prop-
erties of the two BVMOs was evaluated, as shown in Fig-
ure 1. When the oxidations were performed employing
PAMO, no changes in enantioselectivity were observed un-
ꢀ1
til 12 g L of ketone. For higher values, a slight decrease
in the enantioselectivity was observed (E = 137 for
ꢀ
1
2
0 g L ). In order to compare the activity of the enzyme,
the reaction rate (expressed as grams of (± )-3a consumed
ꢀ
1
per L of solution h ) was defined. When the oxidation
was performed with PAMO, this parameter increased by
working at higher 2-phenylpentan-3-one concentrations,
3
. Conclusion
Two newly cloned and overexpressed Baeyer–Villiger mon-
ooxygenases have been employed for the enzymatic resolu-
tion of a set of carbonyl compounds. Methyl cycloketones
were oxidized by PAMO and HAPMO in processes with
low to moderate selectivities. No reactivity was exhibited
by these two enzymes in the oxidation of phenylcyclohexa-
nones. BVMO-catalyzed oxidation of 2-phenylpropional-
dehyde led to (S)-1-phenylethyl formate and the (R)-
aldehyde with moderate enantioselectivities, depending on
the conditions employed. Much better results can be
achieved by using linear phenylketones as substrates, espe-
cially when the kinetic resolutions were performed by
PAMO. (S)-Esters were obtained, leaving the ketones with
an (R)-configuration in processes with excellent selectivi-
ties. PAMO can perform oxidations at substrate concentra-
2
2
1
1
50
00
50
00
0.6
0
0
0
0
0
.5
.4
.3
.2
.1
E
5
0
0
0.0
0
5
10
15
20
g/ L 2-Phenylpentan-3-one
ꢀ1
tions close to 20 g L with high selectivity, while HAPMO
seems to be more sensitive to the ketone concentration. The
pH effect on the kinetic resolution of the phenylaldehyde
and the phenylketones was studied, revealing that by
increasing the pH of the medium, higher activities and
Figure 1. Effect of substrate concentration in the selectivity (green line)
and reaction rate (red line), expressed as g of ketone consumed per L of
solution and per hour of PAMO (solid line) and HAPMO (dashed line)
when performing the reactions at pH 8 and 20 ꢁC.
a
Table 5. BVMO catalyzed oxidation of substrates (± )-4a–6a
b
b
c
Ketone
BVMO
Time (h)
ee (R)-ket (%)
ee (S)-ester (%)
c (%)
E
4
4
5
5
6
6
a
a
a
a
a
a
PAMO
HAPMO
PAMO
HAPMO
PAMO
HAPMO
1
14
1
14
1
36
43
45
78
98
98
98
97
99
96
95
90
27
30
32
45
51
52
188
137
>200
117
179
87
14
a
b
c
ꢀ1
[
4a–6a] = 2 g L . For more reaction details, see Section 4.
Determined by GC.
Conversion, c = ee /(ee
s
s p
+ ee ).

C. Rodr ı´ guez et al. / Tetrahedron: Asymmetry 18 (2007) 1338–1344
1343
lower enantioselectivities could be obtained for both
PAMO and HAPMO. It was also found that PAMO is
an effective catalyst at higher temperatures (40 ꢁC), with
it able to catalyze Baeyer–Villiger reactions at 60 ꢁC. The
best results were achieved by employing HAPMO at low
temperatures (10–20 ꢁC).
1.0 bar N ) for chiral determinations. For all the analyses,
the injector temperature was 225 ꢁC and the FID tempera-
ture was 250 ꢁC.
2
The absolute configuration of chiral methyl cycloketones
and lactones and formate 2b was determined by compari-
son of the GC chromatograms with the patterns described
1
9c,20
in previous experiments for the known configurations.
4. Experimental
For esters 3b–6b the absolute configuration was established
by comparison with an authentic sample, prepared from
the chemical acylation of the corresponding commercial
chiral alcohols.
4
.1. General
Recombinant histidine-tagged phenylacetone mono-
oxygenase and recombinant 4-hydroxyacetophenone
monooxygenase were overexpressed and purified according
to previously described methods. The oxidation reac-
tions were performed using the purified enzymes. One unit
of Baeyer–Villiger monooxygenase oxidizes 1.0 lmol of 2-
phenylpentan-3-one to the ester, per minute at pH 8 and
4.2. General procedure for the BVMO-catalyzed oxidation
of cyclic ketones, linear ketones (± ±-3a–6a and aldehyde (± ±-
7
,8
2a
In a typical experiment, the substrates (10–15 mM) were
dissolved in a Tris/HCl buffer (50 mM, different pH,
2
0 ꢁC, in the presence of NADPH. Glucose-6-phosphate
1
.0 mL), containing glucose-6-phosphate (1.5 equiv), glu-
cose-6-phosphate dehydrogenase (10.0 units), NADPH
0.02 mM) and 1.0 unit of phenylacetone monooxygenase
dehydrogenase from Leuconostoc mesenteroides was
obtained from Fluka-BioChemika. Glucose-6-phosphate
and NADPH was purchased by Sigma–Aldrich.
(
or 4-hydroxyacetophenone monooxygenase. The mixtures
were shaken at 250 rpm in a rotatory shaker at tempera-
tures selected for the times established. The reactions were
then stopped, extracted with dichloromethane or ethyl ace-
tate (3 · 0.5 mL), dried over Na SO and analyzed by gas
All the starting ketones and 2-phenylpropionaldehyde, as
well as all other reagents and solvents were of the highest
quality grade available, supplied by Sigma–Aldrich–
Fluka–Fluka. The corresponding racemic compounds
2
4
chromatography in order to determine the conversion
and enantiomeric excesses of the (S)-esters and the corre-
sponding enantiomer of the remaining substrates. Control
experiments in the absence of enzyme were performed for
all substrates, not observing the reaction after long times.
(
± )-3a–6a were prepared according to the literature, using
either methyl or ethyl iodide and NaOH in a biphasic med-
1
6
ium (46–60% yields) and exhibit physical and spectral
properties in accordance with those reported. Racemic lac-
tones and formate (± )-2b were synthesized with high yields
from the corresponding ketones employing m-CPBA in
CH Cl . Racemic esters (± )-3b–6b were prepared by the
4
.3. General procedure for the HAPMO-catalyzed oxidation
at multimilligram scale of (± ±-2-phenylpentan-3-one and (± ±-
-phenylhexan-3-one
2
2
chemical acylation of commercial 1-phenylethanol or 1-
phenylpropanol (yields higher than 80%), which cause the
direct oxidation with m-CPBA or H O to give low yields
4
2
2
Racemic ketones 3a or 6a (0.5 mmol) were dissolved in a
TRIS/HCl buffer (50 mM, pH 9.0 for 3a and 8.0 for 6a,
and side products. All compounds synthesized exhibit
physical and spectral properties in accordance with those
2
0 mL) containing glucose-6-phosphate (1.5 mmol), glu-
cose-6-phosphate dehydrogenase (10.0 units), NADPH
0.02 mM) and HAPMO (1 unit). The mixtures were sha-
1
6,19
reported.
(
Flash chromatography was performed using Merck silica
gel 60 (230–400 mesh). IR spectra were recorded on a Per-
kin–Elmer 1720-X infrared Fourier transform spectropho-
tometer using KBr pellets. Optical rotations were measured
using a Perkin–Elmer 241 polarimeter and are quoted in
ken at 20 ꢁC and 250 rpm in a rotatory shaker for 60 h
for compound 3a and 72 h for ketone 6a. Once finished,
the reactions were extracted with ethyl acetate
(
4 · 15 mL) and the combined organic layers were dried
ꢀ
1
2
ꢀ1
1
13
over Na SO , filtered and evaporated under reduced pres-
2
4
units of 10 deg cm g . H NMR, C NMR and DEPT
spectra were recorded with TMS (tetramethylsilane) as the
sure. The crude residues were purified by flash chromato-
graphy on silica gel, with hexane/diethyl ether 8:2 to
afford: (R)-3a (colourless oil, 48.6 mg, 60% yield) and (S)-
3b (colourless oil, 26.0 mg, 29% yield); (R)-6a (colourless
oil, 45.8 mg, 52% yield) and (S)-6b (colourless oil,
37.2 mg, 39% yield).
1
internal standard with Bruker AC-300 ( H, 300.13 MHz
C, 75.4 MHz) and Bruker AC-300-DPX ( H,
00.13 MHz and C: 75.4 MHz) spectrometers. The chem-
1
3
1
and
1
3
3
ical shift values (d) are given in ppm and the coupling
+
constants (J) in Hertz (Hz). ESI using a HP1100
chromatograph mass detector or EI with a Finigan MAT
4
.3.1. (S±-4-Methyl-6-hexanelactone, 1b. Determination
of the ee by GC analysis: RtbDEXse, 50 ꢁC (5 min),
ꢁC/min, 200 ꢁC (10 min). t (S) 37.2 min; t (R) 38.0 min.
9
5 spectrometer was used to record mass spectra (MS).
3
GC analyses were performed on a Hewlett Packard 6890
Series II chromatograph equipped with the following
R
R
columns:
Scientific
Agilent
Technologies
HP-1
4.3.2. (R±-2-Phenylpropionaldehyde, 2a. Determination of
the ee by GC analysis: RtbDEXse, 70 ꢁC (5 min), 3 ꢁC/
min, 200 ꢁC (5 min). t (R) 24.9 min; t (S) 25.4 min.
(
30 m · 0.25 mm · 0.25 lm, 1.0 bar N ) for achiral analy-
2
ses and Restek RtbDEXse (30 m · 0.25 mm · 0.25 lm,
R
R

1344
C. Rodr ı´ guez et al. / Tetrahedron: Asymmetry 18 (2007) 1338–1344
(
(
S±-1-Phenylethyl formate, 2b: same GC conditions: tR
Biotechnol. 2003, 21, 318–323; (e) Kirschner, A.; Bornscheuer,
U. T. Angew. Chem., Int. Ed. 2006, 45, 7004–7006.
. (a) Colonna, S.; Gaggero, N.; Carrea, G.; Pasta, P. Chem.
Commun. 1998, 415–416; (b) Ottolina, G.; Bianchi, S.;
Belloni, B.; Carrea, G.; Danieli, B. Tetrahedron Lett. 1999,
S) 26.9 min; t (R) 27.3 min.
R
5
4
.3.3. (R±-2-Phenylpentan-3-one, 3a. Determination of the
ee by GC analysis: RtbDEXse, 70 ꢁC (5 min), 3 ꢁC/min,
00 ꢁC (5 min). (R) 28.1 min; (S) 28.4 min.
aꢁ ¼ ꢀ76:4 (c 1.20, CHCl ), ee 95%. (S±-1-Phenylethyl
4
0, 8483–8486; (c) Sheng, D.; Ballou, D. P.; Massey, V.
2
½
t
t
R
R
Biochemistry 2001, 40, 11156–11167; (d) Ottolina, G.; de
Gonzalo, G.; Carrea, G.; Danieli, B. Adv. Synth. Catal. 2005,
345, 1035–1040.
2
5
D
3
propionate, 3b: same GC conditions: t (S) 28.9 min; t
R
R
(
R) 29.3 min.
6. (a) Fraaije, M. W.; Kamerbeek, N. M.; van Berkel, W. J. H.;
Janssen, D. B. FEBS Lett. 2002, 518, 43–47; (b) Fraaije, M.
W.; Kamerbeek, N. M.; Heidekamp, A. J.; Fortin, R. J. Biol.
Chem. 2004, 279, 3354–3360; (c) Reetz, M. T.; Daligault, F.;
Brunner, B.; Hinrichs, H.; Deege, A. Angew. Chem., Int. Ed.
4
.3.4. (R±-3-Phenylbutan-2-one, 4a. Determination of the
ee by GC analysis: RtbDEXse, 70 ꢁC (5 min), 1 ꢁC/min,
1
1
4
20 ꢁC (5 min). t (R) 44.3 min; tR (S) 46.4 min. (S±-
-Phenylethyl acetate, 4b: same GC conditions: t (S)
R
2
004, 43, 4078–4081; (d) Clouthier, C. M.; Kayser, M. M.
R
Tetrahedron: Asymmetry 2006, 17, 2649–2653; (e) Clouthier,
C. M.; Kayser, M. M.; Reetz, M. T. J. Org. Chem. 2006, 71,
2.9 min; t (R) 50.5 min.
R
8
431–8437.
4
.3.5. (R±-3-Phenylpentan-2-one, 5a. Determination of the
7
. Fraaije, M. W.; Wu, J.; Heuts, D. P. H. M.; van Hellemond,
E. W.; Lutje Spelberg, J. H.; Janssen, D. B. Appl. Microbiol.
Biotechnol. 2005, 66, 393–400.
ee by GC analysis: RtbDEXse, 110 ꢁC Isotherm. t (S)
2
R
1.4 min; t (R) 22.3 min. (S±-1-Phenylpropyl acetate, 5b:
R
same GC conditions: t (S) 23.4 min; t (R) 26.0 min.
8. Kamerbeek, N. M.; Moonen, M. J. H.; van der Ven, J. G. M.;
van Berkel, W. J. H.; Fraaije, M. W.; Janssen, D. B. Eur. J.
Biochem. 2001, 268, 2547–2557.
R
R
4
.3.6. (R±-4-Phenylhexan-3-one, 6a. Determination of the
ee by GC analysis: RtbDEXse, 90 ꢁC (30 min), 3 ꢁC/min,
00 ꢁC (5 min). (R) 47.3 min; (S) 48.4 min.
aꢁ ¼ ꢀ61:2 (c 0.75, CHCl ), ee 98%. (S±-1-Phenylpropyl
9
. (a) Bocola, M.; Schulz, F.; Leca, F.; Vogel, A.; Fraaije, M.
W.; Reetz, M. T. Adv. Synth. Catal. 2005, 347, 979–986; (b)
de Gonzalo, G.; Torres Pazmi n˜ o, D. E.; Ottolina, G.; Fraaije,
M. W.; Carrea, G. Tetrahedron: Asymmetry 2005, 16, 3007–
2
½
t
t
R
R
2
D
5
3
propionate, 6b: same GC conditions: t (S) 45.5 min; t
R
R
3
083.
25
(
R) 47.9 min. ½aꢁD ¼ ꢀ41:7 (c 0.83, CHCl ), ee 90%.
3
10. (a) Schulz, F.; Leca, F.; Hollman, F.; Reetz, M. T. Belstein J.
Org. Chem. 2005, 1, 10; (b) de Gonzalo, G.; Ottolina, G.;
Zambianchi, F.; Fraaije, M. W.; Carrea, G. J. Mol. Catal. B:
Enzym. 2006, 39, 91–97.
Acknowledgements
1
1. Malito, E.; Alfieri, A.; Fraaije, M. W.; Mattevi, A. Proc. Natl.
Acad. Sci. U.S.A 2004, 101, 13157–13162.
Cristina Rodr ´ı guez is financed by Universidad de Oviedo.
Dr. Gonzalo de Gonzalo thanks the Ministerio de Edu-
caci o´ n y Ciencia (MEC) of Spain for his financial support.
We thank Daniel E. Pazmi n˜ o Torres for his technical assis-
tance. This work was supported by the MEC (Project
CTQ-04-04185).
1
2. (a) Guti e´ rrez, M. C.; Alphand, V.; Furstoss, R. J. Mol. Catal.
B: Enzym. 2003, 21, 231–238; (b) Kamerbeek, N. M.;
Oltshoorn, A. J. J.; Fraaije, M. W.; Janssen, D. B. Appl.
Environ. Microbiol. 2003, 419–426; (c) Moonen, M. J. H.;
Westphal, A. H.; Rietjens, I. M. C. M.; van Berkel, W. J. H.
Adv. Synth. Catal. 2005, 347, 1027–1034; (d) Mihovilovic, M.
D.; Kapitan, P.; Rydz, J.; Rudroff, F.; Ogink, F.; Fraaije, M.
W. J. Mol. Catal. B: Enzym. 2005, 32, 135–140.
References
13. de Gonzalo, G.; Torres Pazmi n˜ o, D. E.; Ottolina, G.; Fraaije,
M. W.; Carrea, G. Tetrahedron: Asymmetry 2006, 17, 130–135.
14. Wong, C.-H.; Whitesides, G. M. J. Am. Chem. Soc. 1981,
103, 4890–4899.
15. The enantiomeric ratio E is generally accepted as an adequate
parameter to quantify the enantioselectivity in biochemical
kinetic resolutions. See: Chen, C.-S.; Fujimoto, Y.; Girdaus-
kas, G.; Sih, C. J. J. Am. Chem. Soc. 1982, 104, 7294–7299.
16. Fry, A. J.; Bujanauskas, J. P. J. Org. Chem. 1978, 43, 3157–
3163.
1
2
. (a) Krow, G. R. Org. React. 1993, 43, 251–296; (b) Renz, M.;
Meunier, B. Eur. J. Org. Chem. 1999, 4, 737–750; (c) Ten
Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. E. Chem. Rev.
2
004, 104, 4105–4123.
. (a) Murahashi, S.-I.; Onoi, S.; Imada, Y. Angew. Chem., Int.
Ed. 2002, 41, 2366–2368; (b) Watanabe, A.; Uchida, T.; Ito,
K.; Katsuki, T. Tetrahedron Lett. 2002, 43, 4481–4485; (c)
Punniyamurthy, T.; Velusamy, S.; Iqbal, J. Chem. Rev. 2005,
1
05, 2329–2363.
17. Berezina, N.; Alphand, V.; Furstoss, R. Tetrahedron: Asym-
metry 2002, 13, 1953–1955.
18. Zambianchi, F.; Fraaije, M. W.; Carrea, G.; de Gonzalo, G.;
Rodr ´ı guez, C.; Gotor, V.; Ottolina, G. Adv. Synth. Catal.
2007, 349, 1327–1331.
19. (a) Rahtke, M. W.; Vogiazoglou, D. J. Org. Chem. 1987, 52,
3697–3698; (b) Enders, D.; Eichenauer, H.; Baus, U.;
Schubert, H.; Kremer, K. A. M. Tetrahedron 1984, 40,
1345–1359; (c) Sørbye, K.; Tautermann, C.; Carlsen, P.;
Fiksdahl, A. Tetrahedron: Asymmetry 1998, 9, 681–689; (d)
Birman, V. B.; Jiang, H. Org. Lett. 2005, 7, 3445–3447; (e)
Berkessel, A.; Sebastian-Ibarz, M. L.; Mueller, T. N. Angew.
Chem., Int. Ed. 2006, 45, 6567–6570.
3
. (a) Faber, K. Biotransformations in Organic Chemistry;
Springer: Berlin-Heidelberg, 2004, pp 242–249; (b) van
Berkel, W. J. H.; Kamerbeek, N. M.; Fraaije, M. W. J.
Biotechnol. 2006, 124, 670–689; (c) Mihovilovic, M. D. Curr.
Org. Chem. 2006, 10, 1265–1287; (d) Urlacher, V. D.; Schmid,
R. D. Curr. Opin. Chem. Biol. 2006, 10, 156–161; (e)
Hollman, F.; Hofstetter, K.; Schmid, A. Trends Biotechnol.
2
006, 24, 163–171.
. (a) Fraaije, M. W.; Mattevi, A. Trends Biochem. Sci. 2000, 25,
26–132; (b) Mihovilovic, M. D.; M u¨ ller, B.; Stanetty, P. Eur.
4
1
J. Org. Chem. 2002, 3711–3720; (c) Kamerbeek, N. M.;
Janssen, D. B.; van Berkel, W. J. H.; Fraaije, M. W. Adv.
Synth. Catal. 2003, 345, 667–678; (d) Alphand, V.; Carrea,
G.; Wohlgemuth, R.; Furstoss, R.; Woodley, J. M. Trends
20. Kyte, B. G.; Rouvi e` re, P.; Cheng, Q.; Stewart, J. D. J. Org.
Chem. 2004, 69, 12–17.