Oxidations catalyzed by phenylacetone monooxygenase from Thermobifida fusca

Article abstract of DOI:10.1016/j.tetasy.2005.08.004

Several organic sulfides, ketones and other organic systems have been tested as substrates in oxidation reactions catalyzed by the recently discovered phenylacetone monooxygenase from Thermobifida fusca. The biocatalytic properties of this Baeyer-Villiger monooxygenase have been studied, revealing reactivity with a large range of sulfides and ketones. Oxidations of several sulfoxides, an amine and an organoboron compound were also observed. The enzyme is able to oxidize a number of sulfides with excellent enantioselectivity, demonstrating the catalytic potential of this novel biocatalyst.

Full text of DOI:10.1016/j.tetasy.2005.08.004

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 3077–3083
Oxidations catalyzed by phenylacetone monooxygenase
from Thermobifida fusca
Gonzalo de Gonzalo,a,* Daniel E. Torres Pazmino,b Gianluca Ottolina,a
˜
Marco W. Fraaijeb and Giacomo Carreaa
aIstituto di Chimica del Riconoscimento Molecolare, CNR, via Mario Bianco 9, 20131 Milano, Italy
bLaboratory of Biochemistry, Groningen Biomolecular Sciences and Biotechnology Institute, University of Groningen,
Nijenborgh 4, 9747 AGGroningen, The Netherlands
Received 14 July 2005; accepted 1 August 2005
Available online 9 September 2005
Abstract—Several organic sulfides, ketones and other organic systems have been tested as substrates in oxidation reactions catalyzed
by the recently discovered phenylacetone monooxygenase from Thermobifida fusca. The biocatalytic properties of this Baeyer–
Villiger monooxygenase have been studied, revealing reactivity with a large range of sulfides and ketones. Oxidations of several
sulfoxides, an amine and an organoboron compound were also observed. The enzyme is able to oxidize a number of sulfides with
excellent enantioselectivity, demonstrating the catalytic potential of this novel biocatalyst.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
1.14.13.22) had been extensively studied and applied in
Baeyer–Villiger reactions and other selective oxidation
processes.4
In recent years, the use of Baeyer–Villiger monooxyge-
nases (BVMOs) has been shown to be an excellent meth-
odology in Baeyer–Villiger reactions, sulfoxidations,
amine oxidations and epoxidations.1 In many of these
reactions, conversions occur with high enantio- and/or
regioselectivity, while using environmentally friendly
conditions. In general, the observed selectivities are dif-
ficult to achieve by chemical means.
The recent recognition of a protein sequence motif that
can be used to identify BVMOs has enabled mining of
the genome database. Using this sequence motif, a large
number of putative BVMO genes can be annotated.5 Via
this approach of enzyme discovery, a novel BVMO has
been obtained from Thermobifida fusca. The initial char-
acterization of this biocatalyst has shown that it repre-
Products that can be obtained using BVMOs are of
great value in organic chemistry. Recently, mono- and
polycyclic lactones have received considerable attention
as therapeutic agents and as intermediates in pharma-
ceutical synthesis.2 Enantiomerically pure sulfoxides
have significant commercial value as chiral synthons,
particularly in the synthesis of biologically active mole-
cules, while they can also serve as versatile chiral auxil-
iaries in asymmetric transformations.3
sents
a
thermostable and monomeric enzyme,
containing FAD as a cofactor and being NADPH
dependent.6 Previous studies reported that the best sub-
strate was phenylacetone and therefore the enzyme was
named phenylacetone monooxygenase (PAMO; EC
1.14.13.92). It was also demonstrated that the enzyme
is able to oxidize other aromatic and aliphatic ketones
and organic sulfides. PAMO represents the first BVMO
of which the X-ray structure has been solved.7 The avail-
ability of this structure will facilitate future mechanistic
and enzyme redesign studies.
Only a limited number of BVMOs are available in
recombinant form. In fact, until a few years ago,
only cyclohexanone monooxygenase (CHMO; EC
The main goal of this work was to explore the synthetic
repertoire of phenylacetone monooxygenase, focusing
on the biocatalytic asymmetric oxidation of aromatic
sulfides and other systems in order to obtain the corre-
sponding products with high enantiomeric excesses.
*
Corresponding author. Tel.: +39 022 8500021; fax: +39 022
8901239; e-mail: gonzalo.calvo@icrm.cnr.it
0957-4166/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.08.004

3078
G. de Gonzalo et al. / Tetrahedron: Asymmetry 16 (2005) 3077–3083
Furthermore, to increase the knowledge on the biocata-
lytic potential of this novel oxidative enzyme, the kinetic
parameters of a set of ketones, sulfides, sulfoxides and
tertiary amines in PAMO catalyzed oxidations were also
determined.
pH 9. The results obtained in the PAMO catalyzed oxi-
dation of different aromatic sulfides are summarized in
Table 1.
With alkyl phenyl sulfides 1–4, low to moderate enantio-
meric excesses were obtained. Oxidation of thioanisole
led to the formation of (R)-1a with almost complete con-
version after 24 h, but with moderate enantiomeric ex-
cess (ee = 44%). Sulfides with a longer alkyl chain 2–3
yielded the (S)-sulfoxides instead of the (R) ones, with
lower conversions and enantiomeric excess compared
to substrate 1. Conversion of cyclopropyl phenyl sulfide
4 led to the preferential formation of the (R)-enantio-
mer, with an enantiomeric excess of 48%, close to that
obtained with methyl group. No oxidation was observed
when a carboxylic acid group was present in the sulfide
structure 5.
2. Results and discussion
2.1. Synthetic applications of PAMO
The oxidation of a set of organic sulfides 1–23 to the
corresponding sulfoxides 1a–23a by recombinant histi-
dine-tagged phenylacetone monooxygenase6 was cou-
pled to an ancillary enzymatic reaction in order to
regenerate NADPH (Scheme 1). As a NADPH regener-
ation system, glucose-6-phosphate and glucose-6-phos-
phate dehydrogenase (G6PDH) were employed. All
oxidations were carried out in a Tris/HCl buffer at
As described before,6 the oxidation of 6 was not selective,
leading to (R)-6a with low enantiomeric excess (ee =
10%) and 68% conversion after 24 h. The enzyme also
showed very low selectivity with the other p-tolyl sulfides
tested (compounds 7–9; ee of the products 617%).
R2
R2
*
PAMO
S
S
n
n
O
The introduction of a methoxy group as a para-substitu-
ent (compound 10) slightly decreased the (R)-selectivity
(ee = 25%) compared to 44% for thioanisole 1. Surpris-
ingly, when a strong electron withdrawing group was
located at the para-position 11, the enantiomeric excess
of the obtained sulfoxide (R)-11a was higher (ee = 76%)
than the rest of phenyl sulfides previously examined.
Despite its bulkiness, methyl 2-naphthyl sulfide 12 was
also converted by PAMO to yield (S)-12a with a similar
selectivity compared to thioanisole.
+ O2 + H+
+ H2O
NADP+
NADPH
(R) or (S)-1a-23a
1-23
D-gluconate-6P + H+
D-glucose-6P
G6PDH
n: 0, 1, 2, 3.
R2: Alkyl chain
Scheme 1. General procedure for the enzymatic oxidation of organic
sulfides.
Table 1. Phenylacetone monooxygenase catalyzed oxidation of aromatic sulfidesa
Compound
Structure
Time (h)
Conv.b (%)
eeb (%)
Configuration
1
2
C6H5–S–CH3
C6H5–S–CH2CH3
C6H5–S–propyl
24
24
24
24
48
24
24
48
24
24
24
24
30
6
6
8
18
18
6
6
8
6
8
94
79
56
67
63
68
31
25
12
47
27
22
18
29
36
36
32
21
30
21
27
23
21
44
33
21
48
—
10
17
63
63
25
76
41
63
94
98
41
28
17
48
66
80
83
70
R
S
3
S
4
C6H5–S–cyclopropyl
C6H5–S–CH2COOH
p-CH3–C6H4–S–CH3
p-CH3–C6H4–S–(CH2)2Cl
p-CH3–C6H4–S–octyl
p-CH3–C6H4–S–CH2COOEt
p-CH3O–C6H4–S–CH3
p-O2N–C6H4–S–CH3
2-Naphthyl–S–CH3
C6H5CH2–S–S–CH2C6H5
C6H5CH2–S–CH3
C6H5–CH2–S–CH2–CH3
C6H5CH2–S–isopropyl
C6H5CH2–S–butyl
C6H5CH2–S–isopentyl
C6H5CH2–S–CH2COOMe
C6H5CH2–S–(CH2)2OAc
C6H5(CH2)2–S–CH3
C6H5(CH2)2–S–CH2CH3
C6H5(CH2)3–S–CH3
R
5
—
R
6
7
S
8
—
—
R
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
R
S
—
S
S
R
R
R
ND
ND
R
ND
R
ND: not determined.
a For reaction details see Section 4.
b Conversion and enantiomeric excess determined by HPLC.

G. de Gonzalo et al. / Tetrahedron: Asymmetry 16 (2005) 3077–3083
3079
The alkyl benzyl sulfides possessing a small alkyl chain
appeared to be very good PAMO substrates in terms
of selectivity. The oxidations carried out with com-
pounds 14 and 15 led to the formation of the (S)-sulfox-
ides in high enantiomeric excesses (ee = 94–98%) and
good conversions (nearly 35% after 6 h). It is interesting
to note that the structure of these sulfides is very similar
to phenylacetone, the best PAMO substrate reported so
far. By contrast, when the size of the alkyl chain for this
kind of sulfide was larger than the ethyl group (isopro-
pyl, butyl and isopentyl), the conversion and the enan-
tiomeric excess of the sulfoxides decreased, as shown
in Table 1 (16–18), lower values being obtained as the
alkyl group increased. For chain lengths longer than
ethyl, a change from (S)- to the (R)-configuration of
the products was observed. Introduction of an ester
group in the alkyl moiety of benzyl sulfides 19–20 gave
sulfoxides 19a–20a with moderate enantiomeric ex-
cesses. Conversions for these reactions were very similar
to those obtained with methyl or ethyl groups.
ring, as shown for compound 23. Finally, the presence of
the sulfur atom next to the aromatic ring 1 seemed to have
a marked negative effect on PAMO enantioselectivity.
Previous studies carried out with CHMO revealed that
the oxidation of the sulfoxide products to the corre-
sponding sulfones was very slow and could not be
exploited for kinetic resolution purposes.4b Instead, an
increase in the enantiomeric excess of the sulfoxide
products as a function of time was observed in some
of the PAMO catalyzed oxidations. This indicated that
the enzyme was not only able to catalyze the asymmetric
oxidation from sulfide to sulfoxide, but also the kinetic
resolution of the sulfoxide to the sulfone, as exemplified
in Scheme 2 for ( )-15a.
S
S
.
.
.
.
O
O
PAMO
(S)-15a
(S)-15a
Sulfides possessing the sulfur moiety further away from
the aromatic ring were also converted with good selec-
tivities. (R)-Methyl 2-phenylethyl sulfoxide (R)-21a
was obtained with 80% ee and 27% conversion after
8 h using 21 as substrate. Furthermore, similar behavior
was observed for the ethyl derivative 22 as shown in
Table 1 (ee = 83%). When methyl 3-phenylpropyl sulfide
23 was used as PAMO substrate, moderate enantiomeric
excess was achieved (ee = 70%) with 21% conversion
after 8 h.
S
S
O
.
O
O
.
(R)-15a
Scheme 2. PAMO catalyzed kinetic resolution for ( )-15a.
Table 2 summarizes the results obtained in the kinetic
resolution of a set of sulfoxides. Oxidation of both
( )-1a and ( )-11a occurred, but did not show any
selectivity. More interesting results were obtained with
the benzyl sulfoxides ( )-14a and ( )-15a, where the
enantioselectivities (E),8 especially in the case of the
ethyl derivate (E = 110), were very high. This allowed
the recovery of (S)-15a with high enantiomeric excess
at conversions near 50% in short reaction times. Instead,
when the isopropyl or methylcarboxymethyl benzyl der-
ivates (16a and 19a) were used, the reactions were slower
and the selectivities very low. Finally, the (S)-enantio-
mer of sulfoxide ( )-21a was selectively oxidized
(E = 57) to sulfone, leaving the (R)-enantiomer behind
(ee = 95%). The high enantiomeric excess of compounds
(S)-14a, (S)-15a and (R)-21a (Table 1) was the result of a
combination of the asymmetric oxidation of the starting
sulfides and, in part, of the resolution process of the sulf-
oxides formed.
Besides, the compounds listed in Table 1, some disul-
fides with bulky chains, such as p-tolyl disulfide or thi-
antrene, were found not to be oxidized by PAMO,
presumably because of steric hindrance. A nonaromatic
disulfide, such as 1,3-dithiane, was also examined as an
enzyme substrate. However, after 48 h only the racemic
monosulfoxide was obtained with low conversion
(c = 18%).
Based on the obtained results, we can deduce, in terms of
PAMO enantioselectivity for different sulfur positions in
aromatic methyl sulfides, that the benzyl structure 14 is
the one preferred by PAMO. The 2-phenylethyl group
21 also led to good enantiomeric excesses. The enantio-
meric excess of the sulfoxide obtained decreased when
the sulfur atom was still further away from the aromatic
Table 2. Kinetic resolution of racemic sulfoxides catalyzed by PAMO at 25 ꢁCa
Compound
Structure
Time (h)
Conv.b (%)
eeb (%)
Ec
Configuration
(
(
(
(
(
(
(
)-1a
C6H5–SO–CH3
p-O2N–C6H4–SO–CH3
C6H5CH2–SO–CH3
C6H5CH2–SO–CH2CH3
C6H5CH2–SO–isopropyl
C6H5CH2–SO–CH2COOMe
C6H5(CH2)2–SO–CH3
24
24
4
4
6
52
21
39
49
25
30
51
63
63
60
93
30
24
95
ꢀ1
ꢀ1
53
110
17
—
—
S
)-11a
)-14a
)-15a
)-16a
)-19a
)-21a
S
R
4
8
4.3
57
ND
R
ND: not determined.
a Reactions details described in Section 4.
b Conversion and optical purity determined by HPLC.
c Enantiomeric ratio, E = ln[(1 ꢁ c)(1 ꢁ ees)]/ln[(1 ꢁ c)(1 + ees)].

3080
G. de Gonzalo et al. / Tetrahedron: Asymmetry 16 (2005) 3077–3083
The kinetic resolution of racemic sulfoxides by bacterial
dimethyl sulfoxide reductases has recently been de-
scribed with moderate to good enantioselectivities being
found depending on the sulfoxide.9 The resolution for
these biocatalysts was based on the selective reduction
of the starting sulfoxide to sulfides, not through their
oxidation, as we described herein using PAMO.
lectivity (ee = 82%) into (R)-1-acetoxy-phenylacetone
(R)-29 in a relatively fast process (c = 88% after 1.5 h
reaction). Hydrolysis of this ester would yield (R)-1-
hydroxy-1-phenylacetone (R)-30. The production of
(R)-29 from 28 by an enzymatic Baeyer–Villiger oxida-
tion shows the potential of PAMO to convert prochiral
phenyldiketones.
PAMO has also been tested in the enzymatic oxidation
of organic compounds possessing heteroatoms different
from sulfur. Tertiary amine N-oxides play an important
role in chiral catalysis and in biological processes.10
These compounds are prepared by the oxidation of the
corresponding amines and have been synthesized by bio-
catalytic methods using CHMO.11 Herein, N,N-dimeth-
ylbenzylamine 24 was oxidized by PAMO to the
corresponding N-oxide 24a with 73% conversion after
48 h. When a bulky tertiary amine such as (S)-(ꢁ)-nico-
tine (S)-25 was subjected to PAMO oxidation, the unal-
tered starting material was fully recovered after 2 days
reaction.
2.2. Determination of the kinetic parameters of PAMO
To obtain a better understanding on the catalytic effi-
ciency of PAMO, the steady-state kinetic parameters
of ketones 28, 31–39, sulfides 1, 14, 15, 21 and their cor-
responding ( )-sulfoxides 1a, 14a, 15a and 21a were
determined using isolated PAMO (Table 3).
The maximal catalytic rate (kcat) was found to be
remarkably similar for all substrates (1.2–3.6 sꢁ1). Only
for some substrates, the exact value for kcat could not be
obtained due to solubility problems. More variation was
found for the KM values suggesting differences in
substrate affinity, while the rate of catalysis is probably
restricted by a common substrate-independent kinetic
step. The oxidation of the sulfur atom at the a-position
from the phenyl ring 1 occurred with a low catalytic effi-
ciency (kcat/KM) due to a high KM. A shift of the sulfur
atom to the b- or c-position 14, 15 and 21 resulted in
a 30- to 50-fold increase in catalytic efficiency. This
increase coincided with an increased enantioselectivity.
The racemic sulfoxides turned out to be rather poor sub-
strates due to relatively high KM values. It is interesting
to note that sulfoxide 15a not only displayed a relatively
high catalytic efficiency but also showed the highest
enantioselectivity. The presence of the sulfoxide moiety
also seemed to negatively influence the apparent affinity
of the enzyme for sulfoxides in comparison with the
corresponding sulfides, while the catalytic rate hardly
altered. Surprisingly, the sulfoxides showed a higher
conversion in shorter times than the corresponding sul-
fides, while the catalytic efficiency of the sulfoxides was
significant lower. A reason for this observation might be
that the produced sulfoxides inhibited the conversion of
the sulfides in a greater extent than the sulfones inhibit-
ing the oxidation of the sulfoxides. This latter statement
could however not be true for thioanisole 1, which
showed a higher conversion than 1a with a lower cata-
lytic efficiency.
From the literature, it is known that organoboron com-
pounds can also be oxidized by BVMOs.12a We have
tested one organoboron compound, phenylboronic acid
26, as a PAMO substrate which led to the formation of
phenol 27 (c = 11% after 24 h). The same product can be
formed by chemical oxidation.12b This adds another
type of oxidative reactivity to the broad catalytic reper-
toire of PAMO.
Finally, PAMO was used as a biocatalyst in the oxida-
tion of 3-phenylpenta-2,4-dione 28. Conversion of this
diketone can lead to the formation of an interesting
pharmaceutical intermediate, (R)-phenylacetylcarbinol
(R)-30 (Scheme 3), a well-known precursor in the syn-
thesis of ephedrine and pseudoephedrine.13 Currently,
this compound is produced in an enzymatic process on
industrial scale, using pyruvate and benzaldehyde as
starting compounds.14 As shown in Scheme 3, diketone
28 is indeed converted by PAMO with a good enantiose-
O
O
O
O
PAMO/ O2
O
G6P / G6PDH
NADPH
Comparison of the steady-state kinetic parameters of
ketones 31 and 33 indicated that the presence of an elec-
tron-withdrawing fluorine group at the para-position of
phenylacetone 33 enhanced the catalytic efficiency of the
enzyme by increasing the rate of catalysis. When fluorine
atoms were positioned next to the carbonylic function
(34), a 50-fold lower catalytic efficiency was observed
when compared with phenylacetone, resulting from an
increase in KM. As shown above, PAMO also converted
the prochiral diketone 28. Steady-state analysis showed
that this substrate was oxidized with a reasonable cata-
lytic rate while displaying a relatively large KM, suggest-
ing a low affinity. The apparent affinity decreased even
further when larger and bulkier groups 35–37 were
introduced, resulting in a very poor catalytic efficiency.
28
(R)-29
Hydrolysis
O
HO
Ephedrine
Pseudoephedrine
(R)-30
Scheme 3. Biooxidation of 3-phenylpenta-2,4-dione 28.

G. de Gonzalo et al. / Tetrahedron: Asymmetry 16 (2005) 3077–3083
Table 3. Kinetics parameters in PAMO oxidation for sulfides, sulfoxides and ketones
3081
Num.
Compound
KM (mM)
kcat (sꢁ1
)
kcat/KM (Mꢁ1 sꢁ1
)
1
C6H5–S–CH3
C6H5CH2–S–CH3
P2.5a
1.6
P0.12
1.8
2.3
1.5
1.9
1.2
2.7
1.3
1.4
1.9
1.8
3.6
2.3
ꢂ47
1100
1800
2000
96
51
310
68
200
32,000
5000
65,000
580
14
15
21
1a
14a
15a
21a
28
31
32
33
34
35
C6H5CH2–S–CH2CH3
C6H5(CH2)2–S–CH3
C6H5–SO–CH3
C6H5CH2–SO–CH3
C6H5CH2–SO–CH2CH3
C6H5(CH2)2–SO–CH3
CH3COCH–C6H5–COCH3
C6H5CH2COCH3
C6H5CH2CH2COCH3
p-F–C6H4CH2COCH3
C6H5CH2COCF3
C6H5CH2COCH2CH2CH2CH3
1.3
0.75
19.5
23.7
8.7
19
6.9
b
0.059
0.36
0.056
4.0
b
P2.5a
P0.03
ꢂ10
36
P5.0a
P0.06
ꢂ10
O
O
37
38
P10.0a
P0.08
ꢂ8
p-HO–C6H4CH2CH2COCH3
8.9
3.6
410
a Due to limited solubilities of the compounds, the substrate concentration could not be increased beyond the indicated values.
b Values taken from Ref. 6.
Other substrates such as 2-phenylcycloheptanone and 2-
indanone also showed a low catalytic efficiency (28 and
18 Mꢁ1 sꢁ1, respectively) indicating that PAMO has dif-
ficulties accepting bulky aromatic ketones. Besides
phenylacetone derivatives, benzylacetone 38 was also in-
cluded in this study in order to compare with benzyl-
acetone 32. The electron donating hydroxyl group at
para-position resulted in a ꢂ10-fold lower catalytic
efficiency. When also taking into account the above
mentioned effect of a fluorine substituent, this indicates
that the enzyme prefers electron withdrawing para-
substituents. However, the effect of the hydroxyl group
could also reflect steric hindrance.
substrate 3-phenylpenta-2,4-dione 28, yielding mainly
(R)-1-acetoxy-phenylacetone (R)-29 (ee = 82%). In
addition, it was also found that nitrogen and boron oxi-
dations can be catalyzed by PAMO. The broad sub-
strate specificity and reactivity makes this newly
discovered biocatalyst a valuable tool for performing
selective oxidation reactions. The kinetic analysis
revealed efficient oxidation of either the sulfur atom or
the carbon atom from the carbonyl group at the b- or
c-position from the phenyl ring, whereas oxidation at
the a-position was either rather poor or did not occur
at all.6 The presence of the oxygen atom in the sulfox-
ides seemed to negatively influence the apparent affinity
of the enzyme for these compounds in comparison
with their corresponding sulfides. Apart from this, the
introduction of bulky groups in phenylacetone deriva-
tives 35–37 resulted in a large decrease of catalytic
efficiency, indicating that PAMO has difficulties accept-
ing bulky aromatic ketones. It was also established
that PAMO prefers electron withdrawing para-substitu-
ents in the aromatic moiety in order to obtain high
catalytic efficiencies. As described by Bocola et al.,
the substrate acceptance of PAMO can be increased
even further by the creation of various enzyme
mutants.6b
Finally, steady-state kinetic studies were also performed
with N,N-benzyldimethylamine 24 as substrate, which
was previously shown to be converted to its correspond-
ing N-oxide. However, by monitoring the consumption
of NADPH in time, a catalytic rate of only 0.03 sꢁ1
was measured.
3. Conclusion
Herein, the substrate acceptance and enatioselectivity of
PAMO was explored. This has revealed that the enzyme
is able to oxidize a wide range of sulfides and sulfoxides
with varying degrees of selectivity, depending on the
substrate structure. Also, the enantiopreference of the
enzyme is substrate dependent, ranging from 98% ee
for the (S)-configuration for benzyl ethyl sulfoxide 15a
to 80% ee for the (R)-configuration for methyl 2-phenyl-
ethyl sulfoxide 21a. Furthermore, the enzyme can selec-
tively oxidize benzyl- and 2-phenylethyl sulfoxides
(E >50). PAMO is also able to convert the prochiral
4. Experimental
4.1. General
Recombinant histidine-tagged phenylacetone monooxy-
genase was overexpressed and purified according to
previously described methods.6 Oxidation reactions
were performed using the purified enzyme. One unit of

3082
G. de Gonzalo et al. / Tetrahedron: Asymmetry 16 (2005) 3077–3083
phenylacetone monooxygenase oxidizes 1.0 lmol of
thioanisole 1 to 1a per minute at pH 9 and 25 ꢁC in
the presence of NADPH. Glucose-6-phosphate dehy-
drogenase from Leuconostoc mesenteroides was obtained
from Fluka-BioChemika. Glucose-6-phosphate and
NADPH were purchased by Sigma–Aldrich.
4.2. Typical procedure for the enzymatic oxidation of
substrates
Substrates (20 mM, except for substrate 28, 2.5 mM)
were dissolved in a Tris/HCl buffer (50 mM, pH 9.0,
pH 7.5 for 28, 1.0 mL), containing glucose-6-phosphate
(40 mM, 2 equiv), glucose-6-phosphate dehydrogenase
(10.0 units), NADPH (0.02 mM), acetanilide (0.02 mg)
as internal standard and 1.0 unit of phenylacetone
monooxygenase. The mixture was shaken at 250 rpm
and 25 ꢁC in a rotatory shaker for the times established.
The reactions were then stopped, worked up by extrac-
tion with dichloromethane (3 · 0.5 mL), dried over
Na2SO4 and analyzed by chiral HPLC in order to deter-
mine the conversion and the enantiomeric excesses of
the sulfoxides. The conversion and enantiomeric excess
for compound (R)-29 were determined by means of
GC. Control experiments in the absence of enzyme were
performed for all substrates tested, not observing reac-
tion after long times.
Sulfides 1, 2, 4, 6, thiantrene, p-tolyldisulfide, 1,3-dithi-
ano; racemic sulfoxides ( )-1a, ( )-6a, (ꢁ)-nicotine 25,
phenylboronic acid 26 and ketones 31–35, 37 and 38
were purchased from Sigma–Aldrich–Fluka. Phenyl sul-
fides 11–12, benzyl sulfides 14–15 and amine 24 were
products from Lancaster. Sulfides 5, 10 and ketone 36
were from Acros-Organics. Diketone 28 was purchased
from TCI Europe. (R)-1-Hydroxy-1-phenylacetone 30
was a kind gift from Dr. M. Breuer (BASF). Com-
pounds 3,15 7,16 8,17 9,18 13,15 16,15 17,15 18,19 19,20
20,21 21,19 22,22 23,19 ( )-2923 and ( )-3023 were pre-
pared according to the literature. Sulfoxides were pre-
pared by chemical oxidation from the corresponding
sulfides and exhibit physical and spectral properties in
accord with those reported.15,17–21,24 N,N-dimethyl benz-
ylamine N-oxide 24a and cis-(S)-(ꢁ)-nicotine N-10-oxide
25a were obtained by chemical oxidation with 30%
H2O2.11,25 All other reagents and solvents were of
the highest quality grade available, supplied by Sigma–
Aldrich–Fluka.
4.3. General procedure for the enzymatic oxidations at
multimilligram scale of sulfides 3, 4, 11 and 16
The sulfides (0.33 mmol for 3 and 4; 0.29 mmol for 11
and 0.30 mmol for 16) were dissolved in a Tris/HCl buf-
fer (50 mM, pH 9.0, 25 mL) containing glucose-6-phos-
phate (1.5 equiv), glucose-6-phosphate dehydrogenase
(2.5 units), NADPH (0.02 mM), acetanilide (0.5 mg)
and phenylacetone monooxygenase (0.25 units). The
mixtures were shaken at 25 ꢁC and 250 rpm for 48 h
for substrates 4 and 16, 60 h for sulfide 3 and 72 h
for compound 11. Once finished, the reactions were
extracted with dichloromethane (3 · 25 mL) and the
combined organic layers dried over Na2SO4, filtered
and evaporated under reduced pressure. The residues
were purified by flash chromatography on silica gel
with petroleum ether–ethyl acetate (9:1) to afford the
corresponding sulfoxides: (S)-3a (colorless oil, 23.2 mg,
42% yield) (S)-4a (colorless oil, 17.6 mg, 32% yield),
(R)-11a (yellow pale solid, 26.2 mg, 49% yield) and
(S)-16a (yellow pale oil, 25.8 mg, 48% yield).
IR spectra were recorded on a Jasco FTIR 610. Optical
rotations were determined on a Perkin–Elmer 141
polarimeter. Chemical reactions were monitored by
analytical TLC, performed on Merck silica gel 60 F254
plates and visualized by UV irradiation. Flash chroma-
tography was carried with silica gel 60 (70–230 mesh,
Merck). 1H and 13C NMR spectra at 300 and
72.5 MHz were recorded on a Bruker AC-300. Mass
spectra were performed on a GC-MS-EI (Finnigan-
Thermo). Chiral HPLC analyses were performed on a
Jasco HPLC instrument (model 880-PU pump, model
870-UV/vis detector) equipped with a Chiralcel OD
(Daicel) or a Chiralcel OB (Daicel) chiral column.
Acetanilide was used as internal standard to determine
the conversion of the oxidation processes. Retention
times of the chiral samples were in accordance with
those of the purified racemic ones. Chiral and achiral
GC analyses were performed on a Shimadzu GC17
4.3.1. (S)-Phenyl propyl sulfoxide. Determination of ee
by HPLC analysis: Chiralcel OB, petroleum ether–i-pro-
panol (85:15), 1.0 mL/min, 254 nm, tR 13.4 (S) and 29.3
25
instrument equipped with
a
FID-detector and
a
(R) min, ½aŠ ¼ ꢁ51.8 (c 0.88, CHCl3), ee 21%.
D
Chiraldex G-TA column (Alltech, 30 m · 0.25 mm ·
0.125 mm) or a HP1 column (Agilent, 30 m · 0.25
mm · 0.25 mm), respectively. The kinetic measurements
were carried out with a PerkinElmer Lambda Bio40
spectrophotometer.
4.3.2. (S)-Cyclopropyl phenyl sulfoxide. Determination
of ee by HPLC analysis: Chiralcel OD, petroleum ether–
i-propanol (95:5), 1.0 mL/min, 254 nm, tR 13.1 (R) and
25
17.3 (S) min, ½aŠ ¼ þ56.6 (c 0.71, acetone), ee 48%.
D
Unless otherwise stated, the absolute configurations of
chiral sulfoxides were established by comparison of the
HPLC chromatograms with the patterns described in
previous experiments for the known configurations.
For sulfoxides 3a,19 4a,23a 11a23e and 16a,15 the absolute
configuration was established by comparison of the spe-
cific rotation measured with the ones reported. The con-
figuration of (R)-29 was established by a comparison
with an authentic sample prepared from chemical acetyl-
ation of (R)-30.
4.3.3. (R)-Methyl p-nitrophenyl sulfoxide. Determina-
tion of ee by HPLC analysis: Chiralcel OB petroleum
ether–i-propanol (75:25), 1.0 mL/min, 254 nm, tR 37.9
25
(S) and 51.0 (R) min, ½aŠ ¼ þ84.1 (c 1.30, CHCl3), ee
D
76%.
4.3.4. (R)-Benzyl isopropyl sulfoxide. Determination of
ee by HPLC analysis: Chiralcel OD, petroleum ether–i-
propanol (95:5), 1.0 mL/min, 254 nm, tR 26.8 (S) and
25
28.9 (R) min, ½aŠ ¼ þ53.7 (c 1.29, EtOH), ee 41%.
D

G. de Gonzalo et al. / Tetrahedron: Asymmetry 16 (2005) 3077–3083
3083
Janssen, D. B. J. Biol. Chem. 2004, 279, 3354–3360; (d)
Mihovilovic, M. D.; Kapitan, P.; Rydz, J.; Rudroff, F.;
Ogink, F. H.; Fraaije, M. W. J. Mol. Catal. B: Enzym.
2005, 32, 135–140.
4.4. Procedure for the determination of the kinetic
parameters
For the determination of the steady-state kinetic para-
meters of phenylacetone monooxygenase with various
ketones, sulfides and amines, the enzyme activity was
determined by monitoring the decrease in NADPH
concentration at 340 nm (e340 = 6.22 mMꢁ1 cmꢁ1) or
370 nm (e370 = 2.7 mMꢁ1 cmꢁ1). A reaction mixture of
1.0 mL usually contained 50 mM Tris/HCl, pH 7.5,
100 lM NADPH, 1% (v/v) DMSO and 0.5 lM PAMO.
The presence of 1% DMSO resulted in only a slight
decrease in PAMO activity (<1%), while a higher
solubility of certain compounds could be obtained.
The steady-state kinetic parameters were determined at
30 ꢁC using air-saturated buffers.
6. (a) 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; (b) Bocola,
M.; Schulz, F.; Leca, F.; Vogel, A.; Fraaije, M. W.; Reetz,
M. T. Adv. Synth. Catal. 2005, 347, 979–986.
7. Malito, E.; Alfieri, A.; Fraaije, M. W.; Mattevi, A. Proc.
Natl. Acad. Sci. U.S.A. 2004, 101, 13157–13162.
8. The enantiomeric ratio E is generally accepted as an
adequate parameter to quantify the enantioselectivity in
biochemical kinetic resolutions. See: Chen, C.-S.; Fuji-
moto, Y.; Girdauskas, G.; Sih, C. J. J. Am. Chem. Soc.
1982, 104, 7294–7299.
9. Luckarift, H. R.; Dalton, H.; Sharma, N. D.; Boyd, D. R.;
Holt, R. A. Appl. Microb. Biotechnol. 2004, 65, 678–685.
10. (a) Lindigkeit, R.; Biller, A.; Buch, M.; Schiebel, H. M.;
´
Boppre, M.; Hartmann, T. Eur. J. Biochem. 1997, 245,
626–636; (b) Nakajiama, M.; Saito, M.; Shiro; Hashimoto,
S.-I. J. Am. Chem. Soc. 1998, 120, 2511–2512.
11. Ottolina, G.; Bianchi, S.; Belloni, B.; Carrea, G.; Danieli,
B. Tetrahedron Lett. 1999, 40, 8343–8346.
12. (a) Walsh, C. T.; Chen, Y.-C. J. Angew. Chem., Int. Ed.
Engl. 1988, 27, 333–343; (b) Kuivila, H. G.; Armour, A.
G. J. Am. Chem. Soc. 1957, 79, 5659–5662.
13. Shin, H. S.; Rogers, P. L. Biotechnol. Bioeng. 1996, 49, 52–
62.
Acknowledgements
We thank CERC3 for funding and Dr. Ivan Lavandera
for his technical assistance. COST D25/0005/03 is grate-
fully acknowledged.
References
14. (a) Leksawasdi, N.; Chow, Y. Y. S.; Breuer, M.; Hauer,
B.; Rosche, B.; Rogers, P. L. J. Biotechnol. 2004, 111, 179–
189; (b) Rosche, B.; Breuer, M.; Hauer, B.; Rogers, P. L.
J. Biotechnol. 2005, 115, 91–99.
15. Holland, H. L.; Rand, C. G.; Viski, P.; Brown, F. M. Can.
J. Chem. 1991, 69, 1989–1993.
1. (a) Alphand, V.; Furstoss, R.. Enzyme Catalysis in
Organic Synthesis In Baeyer–Villiger Oxidations; Drauz,
K., Waldmann, H., Eds.; VCH Publishers: Weinheim,
1995; pp 745–772; (b) Willets, A. Trends. Biotechnol. 1997,
15, 55–62; (c) Mihovilovic, M. D.; Muller, B.; Stanetty, P.
¨
16. Bird, R.; Stirling, J. M. J. Chem. Soc., Perkin Trans. 2
1973, 1221–1226.
17. Colonna, S.; Gaggero, N.; Manfredi, A. Biochemistry
1990, 29, 10465–10468.
18. Truce, W. E.; Steltenkamp, R. J. J. Org. Chem. 1962, 27,
2816–2820.
19. Holland, H. L.; Brown, F. M.; Larsen, B. G. Bioorg. Med.
Chem. 1994, 2, 647–652.
Eur. J. Org. Chem. 2002, 3711–3730; (d) Alphand, V.;
Carrea, G.; Wohlgemuth, R.; Furstoss, R.; Carrea, G.
Trends Biotechnol. 2003, 21, 318–321; (e) Kamerbeek, N.
M.; Janssen, D. B.; van Berkel, J. H.; Fraaije, M. W. Adv.
Synth. Catal. 2003, 345, 667–678.
2. (a) Roberts, S. M.; Wan, P. W. H. J. Mol. Catal. B:
Enzym. 1998, 4, 111–136; (b) Ottolina, G.; de Gonzalo, G.;
Carrea, G.; Danieli, B. Adv. Synth. Catal. 2005, 347, 1035–
1040.
20. Jouen, C.; Lasne, M. C.; Pommelet, J. C. Tetrahedron
Lett. 1996, 14, 2413–2416.
´
3. (a) Fernandez, I.; Khiar, N. Chem. Rev. 2003, 3651–3705;
21. Gasparrini, F.; Giovannoli, M.; Misiti, D.; Natile, G.;
Palmieri, G. Tetrahedron 1983, 39, 3181–3184.
22. Fehnel, E. A.; Carmack, M. J. Am. Chem. Soc. 1949, 71,
84–93.
23. Kanemoto, S.; Makoto, S.; Yoshioka, H. Tetrahedron
Lett. 1987, 28, 6313–6316.
24. (a) Davis, F. A.; Reddy, R. T.; Weismiller, M. C. J. Am.
Chem. Soc. 1989, 111, 5964–5965; (b) Colonna, S.;
Gaggero, N.; Casella, L.; Carrea, G.; Pasta, P. Tetra-
hedron: Asymmetry 1992, 3, 95–106; (c) Secundo, F.;
Carrea, G.; Dallavalle, S.; Franzosi, G. Tetrahedron:
Asymmetry 1993, 4, 1981–1982; (d) Pasta, P.; Carrea, G.;
Holland, H. L.; Dallavalle, S. Tetrahedron: Asymmetry
1995, 6, 933–936; (e) Brunel, J.-M.; Diter, P.; Duetsch, M.;
Kagan, H. B. J. Org. Chem. 1995, 60, 8086–8088; (f)
Yabuuchi, T.; Kusumi, T. J. Am. Chem. Soc. 1999, 121,
10646–10647.
(b) Legros, J.; Dehli, J. R.; Bolm, C. Adv. Synth. Catal.
2005, 347, 19–31.
4. (a) Ottolina, G.; Carrea, G.; Colonna, S.; Ruckemann, A.
¨
Tetrahedron: Asymmetry 1996, 7, 1123–1136; (b) Colonna,
S.; Gaggero, N.; Pasta, P.; Ottolina, G. Chem. Commun.
1996, 2303–2307; (c) Mihovilovic, M. D.; Chen, G.; Wang,
S.; Kyte, B.; Rochon, F.; Kayser, M. M.; Stewart, J. D. J.
Org. Chem. 2001, 66, 733–738; (d) Sheng, D.; Ballou, D.
P.; Massey, V. Biochemistry 2001, 40, 1156–1167; (e)
Zambianchi, F.; Pasta, P.; Carrea, G.; Colonna, S.;
Gaggero, N.; Woodley, J. M. Biotechnol. Bioeng. 2002,
78, 489–496; (f) Hilker, I.; Alphand, V.; Wohlgemuth, R.;
Furstoss, R. Adv. Synth. Catal. 2004, 346, 203–214.
5. (a) 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; (b) Kamerbeek,
N. M.; Olsthoorn, A. J. J.; Fraaije, M. W.; Janssen, D. B.
Appl. Environ. Microbiol. 2003, 419–426; (c) Fraaije, M.
W.; Kamerbeek, M. N.; Heidekamp, A. J.; Fortin;
25. Crooks, P. A.; Pool, W. F.; Damani, L. A.; Winkle, S. A.
Chem. Ind. 1988, 95–96.