Enzymatic Baeyer-Villiger oxidation of benzaldehydes

Article abstract of DOI:10.1002/adsc.200404307

The selectivity of the chemical Baeyer-Villiger oxidation of benzaldehydes depends on steric and electronic factors, the type of oxidizing agent and the reaction conditions. Here we report on the enzymatic Baeyer-Villiger oxidation of fluorobenzaldehydes

Full text of DOI:10.1002/adsc.200404307

FULL PAPERS
Enzymatic Baeyer--Villiger Oxidation of Benzaldehydes
a
a
b
Mari e¨ lle J. H. Moonen, Adrie H. Westphal, Ivonne M. C. M. Rietjens,
a,
Willem J. H. van Berkel *
a
Laboratory of Biochemistry, Wageningen University, Dreijenlaan 3, 6703 HA Wageningen, The Netherlands
Phone: (þ31)-317-482861, fax: (þ31)-317-484801, e-mail: willem.vanberkel@wur.nl
b
Division of Toxicology, Wageningen University, Tuinlaan 5, 6703 HE Wageningen, The Netherlands
Received: October 8, 2004; Revised: March 18, 2005; Accepted: March 19, 2005
Abstract: The selectivity of the chemical Baeyer–Vil- HAPMO-mediated conversion of 2-fluoro-, 3-fluoro-,
liger oxidation of benzaldehydes depends on steric 2,3-difluoro- and 2,4-difluorobenzaldehyde yielded,
and electronic factors, the type of oxidizing agent besides fluorophenols, also minor amounts of fluoro-
and the reaction conditions. Here we report on the benzoic acids. The high preference of HAPMO for
enzymatic Baeyer–Villiger oxidation of fluorobenzal- the production of fluorophenols is in disagreement
dehydes as catalyzed by the flavoprotein 4-hydroxy- with the rule derived from chemical studies that elec-
acetophenone monooxygenase (HAPMO). HAPMO tron-poor benzaldehydes form mainly benzoic acids.
was most active with 4-amino- and 4-hydroxybenzal- This suggests that interactions of the benzaldehyde
dehydes. With these compounds significant substrate substrates with amino acids and/or the flavin cofactor
inhibition occurred. Monofluoro- and difluorobenzal- in the enzyme active site influence the selection of the
dehydes were readily oxidized by HAPMO without migratory group in favor of the phenyl ring.
19
substrate inhibition. F NMR analysis revealed that
-fluoro-, 2,6-difluoro-, 3,4-difluoro-, 2-fluoro-4-hy-
droxy- and 3-fluoro-4-hydroxybenzaldehyde were Keywords: Baeyer–Villiger oxidation; benzaldehyde;
4
1
9
quantitatively converted by HAPMO to the corre- enzymatic synthesis; flavoprotein; F NMR; 4-hy-
sponding fluorophenyl formates. These products droxyacetophenone monooxygenase; substrate inhibi-
spontaneously hydrolyzed to fluorophenols. The tion
Introduction
Newer methods for the chemical synthesis of fluoro-
phenols make use of the Baeyer–Villiger oxidation of ar-
[
8–11]
Fluorinated chemicals are gaining in industrial impor- omatic ketones and benzaldehydes.
In these reac-
tance, with applications in pharmaceuticals, agrochemi- tions a nucleophilic peroxy moiety is incorporated into
[
1]
cals and material products. Among these fluorinated the starting compound yielding a tetrahedral “Criegee”
[
12,13]
chemicals, fluorophenols have attracted much attention intermediate.
This unstable species undergoes a
due to their ability to modify the biological activity of rearrangement involving migration of one of the sub-
molecules. Fluorophenols can serve as synthons for the stituents of the carbonyl carbon and subsequent hetero-
[
2,3]
production of drugs, including enzyme inhibitors
lytic cleavage of the peroxidic bond, yielding either the
[
4]
and receptor antagonists. Catecholamines and amino ester or benzoic acid (Scheme 1). By using an aqueous
18
acids with a radiolabelled [ F]-fluorophenol moiety base, the ester is hydrolyzed to give the desired phenol.
have found applications for the in vivo imaging of amino In general, the migration preference of the Baeyer–
acid metabolism and protein synthesis using positron Villiger rearrangement depends on the steric and elec-
[5]
[6]
[
7]
emission tomography (PET).
tronic properties of the groups attached to the carbonyl
Chemical methods to produce fluorophenols include carbon atom. The migration ability is also influenced by
diazotization/Sandmeyer decomposition with fluoro- the type of oxidizing agent and the reaction condi-
[
8,10,11,13–16]
anilines, diazotization/thermal decomposition with ami- tions.
Since all groups attached to the carbonyl
nophenols, alkaline hydrolysis with substituted fluoro- carbon of acetophenones migrate in preference to the
benzene and F /N gas/liquidphase conversion with phe- methyl group, substituted acetophenones show a strong
2
2
[
17,18]
nol. All these methods display both economic or techno- preference for ester formation.
For other aromatic
logical benefits and disadvantages, and none of them alkyl ketones, migration of the alkyl group depends on
seems to be entirely satisfactory for the industrial pro- its structure, while the migration of the aryl group de-
[
8]
duction of fluorophenols.
pends on the electron-donating or electron-withdrawing
[
8,13]
properties of the substituents.
For benzophenones it
Adv. Synth. Catal. 2005, 347, 1027–1034
DOI: 10.1002/adsc.200404307
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1027

FULL PAPERS
Mari e¨ lle J. H. Moonen et al.
Scheme 1. Baeyer–Villiger oxidation of benzaldehydes.
was established that the product selectivity is strongly responding alcohols through the biooxidative activity
[
33]
dependent on the type and position of the substituents of alcohol dehydrogenases and aryl-alcohol oxidas-
[
8,10,19]
[34–36]
in the phenyl rings.
Benzaldehydes with electron- es.
donating substituent(s) in ortho and para positions
Recently, we described the purification, characteriza-
stimulated migration of the phenyl ring and formation tion and heterologous expression of 4-hydroxyaceto-
of phenyl formates (upper pathway in Scheme 1). On phenone monooxygenase (HAPMO) from Pseudomo-
[
37,38]
the other hand, benzaldehydes lacking electron-donat- nas fluorescens ACB.
This homodimeric flavoen-
[30]
ing substituents and/or possessing electron-withdrawing zyme is a type I Baeyer–Villiger monooxygenase
groups yielded benzoic acids (lower pathway in that is active with a wide range of aromatic ketones.
[38,39]
[9,10]
Scheme 1).
HAPMO also catalyzes the conversion of aliphatic ke-
[
28,29,39]
[39]
Ekaeva et al. reported the Baeyer–Villiger oxidation tones to lactones
and the oxidation of aryl sulfides
18
[9]
of [ F]-fluorobenzaldehydes. Using m-chloroperben- to chiral sulfoxides. Being highly active with aromatic
zoic acid in the presence of trifluoroacetic acid, it was compounds, HAPMO can be discriminated from the ex-
shown that, in contrast to the above rules, fluorobenzal- tensively studied cyclohexanone monooxygenase from
[
28,40–43]
dehydes can be converted to fluorophenols to a signifi- Acinetobacter sp. NICMB 9871.
cant extent. Relatively little is known about the reactivity of HAP-
The chemical Baeyer–Villiger reaction requires the MO with benzaldehydes. Preliminary experiments re-
use of potentially hazardous oxidants. Therefore, efforts vealed that, in line with the chemical Baeyer–Villiger re-
have been made to replace the traditionally used per- action, 4-hydroxybenzaldehyde is converted to 4-hy-
acids with safer oxidants. Most of these strategies rely droxyphenyl formate, which after hydrolysis yields hy-
[
39]
on the use of transition metal catalysts or organocatalyt- droquinone. To study in more detail the selectivity
ic compounds in combination with hydrogen peroxide, of the enzyme-mediated Baeyer–Villiger oxidation of
[
11,16,17,20–27]
peroxy acids or molecular oxygen.
However, aromatic compounds we addressed in this paper the re-
the most challenging “green chemistry” approach for activity and product specificity of HAPMO with fluori-
the selective Baeyer–Villiger oxidation of ketones nated benzaldehydes. Unlike most chemical procedures,
and aldehydes is offered by enzyme-mediated reac- the enzymatic approach gives high yield and selectivity
[28–30]
tions.
towards the sustainable production of fluorophenols.
Fluorobenzaldehydes are the most attractive starting
compounds for the synthesis of fluorophenols because
they are available via sustainable pathways. Several Results
white-rot fungi have a great potential for the biocatalytic
[
31]
production of haloaromatic aldehydes.
Bioconver-
Substrate Specificity
sion experiments with whole cells of Bjerkandera adusta
showed that 2-fluoro-, 3-fluoro- and 4-fluorobenzoic
acid and 2-fluoro- and 4-fluorocinnamic acid were re- Kinetic studies revealed that HAPMO is active with a
[
32]
duced to the corresponding fluorobenzaldehydes.
wide range of substituted benzaldehydes. 4-Aminoben-
Fluorobenzaldehydes can also be obtained from the cor- zaldehyde appeared to be the best HAPMO aldehyde
1028
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1027–1034

Enzymatic Baeyer–Villiger Oxidation of Benzaldehydes
FULL PAPERS
substrate (Table 1). However, with this compound sig- Product Analysis of the Conversion of Fluorinated
nificant substrate inhibition occurred (Fig. 1A). 4-Hy- Benzaldehydes by HAPMO
droxybenzaldehydes were also good HAPMO sub-
strates. With these compounds, weak substrate inhibi- The aromatic products formed in the reactions of HAP-
tion was observed (Table 1). The kinetic parameters es- MO with fluorinated benzaldehydes were analyzed by
1
9
19
timated with 4-hydroxybenzaldehyde slightly deviate
F NMR. Fig. 2A presents, as an example, the
F
from previous data where substrate inhibition was not NMR spectrum recorded after incubation of 2-fluoro-
[
38]
taken into account.
benzaldehyde (ꢀ124.7 ppm) with a limiting amount of
The HAPMO-mediated oxidation of fluorinated ben- NADPH. From this spectrum it is clear that two prod-
zaldehydes was not sensitive to substrate inhibition (Ta- ucts are formed. On the basis of reference compounds,
ble 1, Fig. 1B). As found for the corresponding aceto- the main product (95%) was assigned to 2-fluorophenol
phenones (Table 1), introduction of fluoro substituents (ꢀ142.0 ppm), whereas the minor product (5%) was as-
in the benzaldehyde ring slightly decreased the reaction signed to 2-fluorobenzoic acid (ꢀ120.3 ppm). Under
rate and increased the Michaelis constant to some ex- the conditions applied, no ester formation was observed.
tent. 3-Bromo-4-fluorobenzaldehyde was not a sub-
strate for HAPMO.
Enzymatic incubations with other fluorinated benzal-
dehydes mainly resulted in the production of fluorinated
phenols (Table 2). With 2-fluoro-, 3-fluoro-, 2,3-difluoro-
and 2,4-difluorobenzaldehyde, some fluorinated benzo-
ic acid was also formed. The enzymatic conversions of 2-
Figure 1. Michaelis–Menten kinetics of the 4-hydroxyacetophenone monooxygenase-mediated conversion of substituted ben-
zaldehydes, (A) 4-aminobenzaldehyde, (B) 2-fluorobenzaldehyde.
Table 1. Kinetic parameters of the conversion of substituted benzaldehydes and acetophenones by 4-hydroxyacetophenone
monooxygenase from P. fluorescens ACB.
Substitution
Benzaldehyde
Acetophenone
ꢀ
1
ꢀ1
Kꢀm [mM]
kꢀcat [s
]
K’ [mM]
s
Kꢀm [mM]
kꢀcat [s
]
K’ [mM]
s
parent compound
3.4ꢁ0.5
3.0ꢁ0.4
4.8ꢁ0.4
4.0ꢁ0.1
0.8ꢁ0.6
3.4ꢁ0.3
14.7ꢁ1.4
5.7ꢁ0.7
4.7ꢁ0.8
7.9ꢁ1.0
0.27ꢁ0.02
0.014ꢁ0.004
5.2ꢁ0.3
3.9ꢁ0.1
4.0ꢁ0.1
2.6ꢁ0.0
3.2ꢁ1.6
7.0ꢁ0.3
0.7ꢁ0.0
2.3ꢁ0.1
2.4ꢁ0.1
0.7ꢁ0.0
12.5ꢁ0.4
15.9ꢁ2.4
–
–
–
–
2.0ꢁ0.8
1.6ꢁ0.2
0.7ꢁ0.1
4.6ꢁ1.2
1.0ꢁ0.1
1.0ꢁ0.1
0.7ꢁ0.1
12.4ꢁ0.8
–
–
–
–
[
[
a]
a]
2
3
4
2
3
2
2
2
3
4
4
-fluoro-
-fluoro-
-fluoro-
-fluoro-4-hydroxy-
-fluoro-4-hydroxy-
,3-difluoro-
,4-difluoro-
,6-difluoro-
,4-difluoro-
-hydroxy-
[
b]
[a]
1.0ꢁ0.3
2.0ꢁ0.2
1.8ꢁ2.0
ND
–1 .5ꢁ1.2
–
–
–
3.7ꢁ0.4
4.8ꢁ5.6
ND
–
ND
14.2ꢁ10.2
ND
[
a]
1.7ꢁ0.5
0.5ꢁ0.1
ND
ND
8.3ꢁ0.3
20.4ꢁ3.4
ND
–
ND
ND
4.8ꢁ0.4
0.006ꢁ0.001
0.9ꢁ0.1
-amino-
0.07ꢁ0.02
0.006ꢁ0.003
0.3ꢁ0.2
[
[
a]
[44]
From ref.
b]
[38]
From ref.
ND¼not determined.
Adv. Synth. Catal. 2005, 347, 1027–1034
asc.wiley-vch.de
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1029

FULL PAPERS
Mari e¨ lle J. H. Moonen et al.
19
Figure 2. F NMR analysis of the enzymatic conversion of 2-fluorobenzaldehyde by 4-hydroxyacetophenone monooxygenase.
(
A) Reaction at pH 8.0 with a limiting amount of NADPH to have about 50% substrate left at the end of the incubation
(
50,000 scans). (B) Reaction at pH 6.0 with excess NADPH (5,000 scans). I. S. is the internal standard (4-fluorobenzoic
acid, ꢀ114.2 ppm). See Experimental Section for further details.
Table 2. Product ratio and yield of the conversion of fluorinated benzaldehydes by 4-hydroxyacetophenone monooxygenase
1
9
from P. fluorescens ACB as determined by F NMR. The mean standard error in the concentrations of fluorinated compounds
was about 3%.
[a]
Substrate
Yield [%]
Product ratio
Phenol [%]
Benzoic acid [%]
2
3
4
2
3
2
2
2
3
-fluorobenzaldehyde
-fluorobenzaldehyde
-fluorobenzaldehyde
-fluoro-4-hydroxybenzaldehyde
-fluoro-4-hydroxybenzaldehyde
,3-difluorobenzaldehyde
,4-difluorobenzaldehyde
,6-difluorobenzaldehyde
,4-difluorobenzaldehyde
88
83
75
86
91
25
64
70
31
95
82
100
100
100
73
95
100
100
5
18
0
0
0
27
5
0
0
[
a]
Total yield of phenol and benzoic acid in incubations with stoichiometric amounts of benzaldehyde and NADPH (see Ex-
perimental Section for further details).
þ
fluoro-4-hydroxybenzaldehyde and 3-fluoro-4-hydroxy- generating system consisting of NADP , glucose 6-
benzaldehyde both resulted in the exclusive formation phosphate, glucose 6-phosphate dehydrogenase and
of 2-fluorohydroquinone (Table 2). The chemical shift air-saturating conditions, it was possible to convert mg
values of fluorinated benzaldehydes and their products amounts of 2-fluorobenzaldehyde into 2-fluorophenol
obtained after conversion by HAPMO are listed in Ta- (95%) and 2-fluorobenzoic acid (5%).
ble 3.
2,3-Difluorobenzaldehyde and 3,4-difluorobenzalde-
hyde were rather poor substrates for HAPMO (cf. Ta-
ble 1). Substantial conversion of these compounds was
only obtained at relatively high enzyme concentrations.
Moreover, for high yields large excesses of NADPH and
Factors Governing Product Yield
The product yield of the HAPMO-mediated conversion oxygen were needed, becauseof the competing non-pro-
ꢀ
1
[38]
of fluorinated benzaldehydes was dependent on the type ductive NADPH oxidase activity (k’ ¼0.11 s ).
cat
of substrate, the reaction conditions and the amount of
Additional studies on factors governing product yield
enzyme used. With all substrates tested, no product in- were performed using 2-fluorobenzaldehyde as the
hibition was observed. Monofluorinated benzaldehydes model substrate. Absorption spectral analysis of the en-
were completely converted, providing that excess zymatic conversion of 2-fluorobenzaldehyde pointed at
NADPH and oxygen were supplied. Using an NADPH the initial formation of a compound with a low absorp-
1030
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1027–1034

Enzymatic Baeyer–Villiger Oxidation of Benzaldehydes
FULL PAPERS
19
Table 3. F NMR chemical shifts [ppm] of fluorinated benzaldehydes and their products after conversion by 4-hydroxyaceto-
phenone monooxygenase from P. fluorescens ACB.
Substitution
Benzaldehyde
Benzoic acid
Phenol
[
[
[
a]
a]
a]
2
3
4
2
3
2
2
2
3
-fluoro-
-fluoro-
-fluoro-
-fluoro-4-hydroxy-
-fluoro-4-hydroxy-
,3-difluoro-
,4-difluoro-
,6-difluoro-
,4-difluoro-
ꢀ124.7
ꢀ120.3
ꢀ142.0
ꢀ116.6
ꢀ129.2
ꢀ138.8
ꢀ138.6
ꢀ116.7
ꢀ117.9
ꢀ106.3
ꢀ114.2
ꢀ121.8
ꢀ116.4
ꢀ141.3
ND
ꢀ142.2 (F3) and ꢀ150.1 (F2)
ꢀ101.8 (F4) and ꢀ119.8 (F2)
ꢀ119.6
ꢀ142.8 (F3) and ꢀ46.4 (F2)
ꢀ111.5 (F4) and ꢀ15.1 (F2)
ꢀ120.1
ꢀ143.1 (F3) and ꢀ167.1 (F2)
ꢀ126.3 (F4) and ꢀ137.3 (F2)
ꢀ139.3
[a]
ꢀ131.2 (F4) and ꢀ140.7 (F3)
ꢀ139.0 (F4) and ꢀ42.6 (F3)
ꢀ140.9 (F3) and ꢀ154.3 (F4)
[
a]
[62]
See ref.
tion in the 300–400 nm region (not shown). This absorp- nols can be produced by the chemical Baeyer–Villiger
tion disappeared in time, indicative for the non-enzy- oxidation of fluorobenzophenones, fluoroacetophe-
matic hydrolysis of the intermediate ester product. Sim- nones and fluorobenzaldehydes. With all these com-
ilar transient absorption changes were observed in the pounds, the amount of fluorophenol produced is de-
enzymatic conversion of other fluorobenzaldehydes. pendent on the substituents in the aromatic ring(s),
To get additional insight into ester formation, the HAP- the type of oxidizing agent and the reaction condi-
[
8–10,14]
MO-mediated conversion of 2-fluorobenzaldehyde was tions.
The chemical Baeyer–Villiger oxidation of
studied as a function of pH and at relatively high enzyme 3-bromo-4-fluorobenzaldehyde led to the correspond-
concentration (Table 4). At pH 6.0, the enzymatic reac- ing fluorobenzoic acid as opposed to the desired phe-
1
9
[26]
tion was significantly slower than at pH 8.0, but
F
nol.
With monofluorobenzaldehydes, the chemical
NMR product analysis clearly showed the initial forma- conversion resulted in nearly equal amounts of fluoro-
tion of 2-fluorophenyl formate (ꢀ133.9 ppm, Fig. 2B). phenols and fluorobenzoic acids with a final yield of flu-
[
9]
After 10 min incubation, 80% of the total amount of orophenols of about 25%.
products was the ester which slowly decayed to yield
In a previous study we reported that 4-fluorophenols
2
-fluorophenol. After measuring for 1 h at 78C, nearly can be produced from 4-fluoroacetophenones using
[
44]
half the amount of ester was hydrolyzed to the phenol HAPMO as biocatalyst.
Oxyfunctionalization by
(
Table 4), whereas after 2 h less than 25% of the ester re- HAPMO at pH 6 resulted in the exclusive production
mained. At pH 7.0, 2-fluorobenzaldehyde was more of 4-fluorophenyl acetates. At pH 8.0, the reaction rate
readily converted, but almost no formate ester was de- increased and the fluorinated phenyl acetate products
19
tected by F NMR. Table 4 summarizes the product readily hydrolyzed to the corresponding fluorophe-
[
44]
yields and product ratios of the reactions of HAPMO nols. Here we show that HAPMO is active with a
with 2-fluorobenzaldehyde at different pH values. At wide range of fluorinated benzaldehydes. Under the
all pH values, formation of about 5% of 2-fluoroben- conditions applied (pH 8.0), these reactions mainly re-
zoate was observed.
sulted in the production of fluorophenols. At pH 6.0,
transient formation of the initial ester product could
be observed.
Discussion
The enzymatic conversion of 4-amino-, 4-hydroxy-
and 3-fluoro-4-hydroxybenzaldehyde was sensitive to
Fluorophenols are attractive building blocks for the syn- substrate inhibition. Interestingly, there was a clear cor-
thesis of high-value fluorinated chemicals. Fluorophe- relation between the Michaelis constants of the aromat-
Table 4. Product ratio and yield of the conversion of 2-fluorobenzaldehyde by 4-hydroxyacetophenone monooxygenase from
1
9
P. fluorescens ACB at different pH values. F NMR spectra were recorded for 1 h at 78C. The mean standard error in the con-
centrations of fluorinated compounds was about 3%.
pH
Product yield [%]
Product ratio [%]
2-fluorophenyl formate
2-fluorophenol
2-fluorobenzoate
8
7
6
.0
.0
.0
100
95
57
0
2
53
95
93
42
5
5
5
Adv. Synth. Catal. 2005, 347, 1027–1034
asc.wiley-vch.de
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1031

FULL PAPERS
Mari e¨ lle J. H. Moonen et al.
ic substrates (K’ ) and the degree of substrate inhibition is of particular interest from a preparative point of
m
[53,54]
(
K’ ). This suggests that the inhibition is directly related view because it avoids expensive cofactor recycling.
s
to catalysis and is not caused by binding of the substrate Therefore, with proper upscaling and downstream proc-
[
41,55–58]
at a secondary site. For the related cyclohexanone mon- essing,
HAPMO may develop as an environmen-
þ
ooxygenase, it was reported that NADP remains tally benign alternative for the synthesis of fluorophe-
bound during the entire reaction cycle and that the inter- nols.
action between the reduced enzyme and pyridine nu-
cleotide is needed for the stabilization of the flavin
[
45]
C4a peroxide oxygenation species. Thus, it is unlikely
that the inhibitory substrates are binding at the
NADP(H) binding site. A more plausible explanation
for the observed substrate inhibition is that the aromatic
substrate competes with the aromatic product for bind-
ing to (one of) the enzyme intermediates which are
formed after substrate oxygenation. A similar competi-
tion between substrate and product binding has been ob-
served with flavoprotein aromatic hydroxylases.
The HAPMO-mediated Baeyer–Villiger oxidation of
other fluorobenzaldehydes was not sensitive to sub-
strate inhibition. In line with the above notion, this cose 6-phosphate dehydrogenase (grade I), NADPH and
might be related to a relatively weak binding of these NADP were from Roche.
Experimental Section
Materials
4
-Fluorobenzaldehyde, 3-fluorobenzoic acid and 4-hydroxy-
acetophenone were purchased from Aldrich. 2-Fluorophenol,
-fluorophenol and 4-fluorophenol were from Acros. 2-Fluo-
3
ro-4-hydroxybenzaldehyde was synthesized as described earli-
[
46–48]
[59]
er. All other fluorinated aromatic compounds were obtained
from Fluorochem. Dimethylformamide (DMF) and potassium
phosphate were from Merck. Glucose 6-phosphate, yeast glu-
þ
compounds in the active site.
The HAPMO-mediated conversion of fluorobenzal-
dehydes mainly resulted in the production of fluorophe- Enzyme Purification
nols. The high preference for the formation of phenols is
HAPMO was purified from Pseudomonas fluorescens ACB as
described earlier.
in disagreement with the rule derived from chemical
studies that aromatic substrates with electron-with-
[38]
[
8,10,16]
drawing substituents mainly form benzoic acids.
Because the rearrangement of the tetrahedral Criegee
intermediate (Scheme 1) of the enzymatic reaction is
believed to be governed by the same stereoelectronic ef-
fects as in the non-enzymatic Baeyer–Villiger oxida-
Analytical Methods
The activity of 4-hydroxyacetophenone monooxygenase
(
HAPMO) was routinely determined spectrophotometrically
[28,49,50]
by monitoring the aromatic substrate-stimulated oxidation of
tion,
this suggests that interactions of the benzal-
ꢀ
1
ꢀ1
NADPH at 370 nm (e370 ¼2.7 mM cm ). The standard en-
dehyde substrates with amino acids or the flavin cofac-
tor in the active site influence the selection of the migra-
tory group in favor of the phenyl ring. The first insight
zyme activity was measured at 258C in 1.0 mL air-saturated
5
0 mM potassium phosphate, pH 8.0, containing 250 mM
[38]
NADPH, and 200 mM 4-hydroxyacetophenone. The reac-
into the active site topology of type I Baeyer–Villiger tion was started by the addition of 10 mg HAPMO. Specific ac-
monooxygenases has recently come from the crystal tivities were calculated from initial rate determinations and
structure determination of the uncomplexed form of were corrected for endogenous NADPH oxidase activity.
[
51]
One unit of HAPMO activity is defined as the amount of en-
zyme that catalyzes the oxidation of 1 mmol NADPH per mi-
nute under the standard assay conditions. Steady-state kinetics
were performed under the above conditions with 250 mM
NADPH as the fixed substrate and varying concentrations
phenylacetone monooxygenase. From this structure
and sequence comparison it was deduced that an argi-
nine residue (Arg440) plays a critical role in HAPMO
catalysis. However, a deeper insight into the enzyme-
substrate interaction must await the crystal structure de-
termination of a protein complex.
(
1 mM–20 mM) of aromatic substrate. With slow substrates,
the enzyme concentration was increased ten-fold. The kinetic
In conclusion, we have shown in this paper that HAP-
parameters K’ and k’ were determined from non-linear re-
m
cat
MO catalyzes the conversion of a wide range of benzal- gression analysis using the Michaelis–Menten equation. In
dehydes to the corresponding esters. The latter com- case of substrate inhibition, the kinetic data were treated ac-
pounds spontaneously hydrolyze to the corresponding cording to Eq. (1):
phenols. The enzymatic reactions with fluorobenzalde-
hydes show high selectivity and product yield and are
not sensitive to substrate inhibition. A main advantage
ð1Þ
of the enzymatic Baeyer–Villiger reaction is the use of
a mild oxidant (i.e., the flavin C4a peroxide). Recent
studies have shown that recombinant whole cells ex-
where k’obs is the apparent initial rate observed, k’cat is the appa-
rent maximal rate at saturating substrate concentration, K’ is
m
pressing HAPMO can be used for the biooxidation of the apparent Michaelis constant for the aromatic substrate at
[
52]
prochiral cyclobutanones. Such a whole-cell system fixed values of the other substrates (NADPH and O ) and K’
2
s
1032
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1027–1034

Enzymatic Baeyer–Villiger Oxidation of Benzaldehydes
FULL PAPERS
is the apparent secondary-binding inhibition constant for the Acknowledgements
[60]
aromatic substrate.
We thank Barend van Lagen for assistance in NMR experi-
ments. This work was supported by the Council for Chemical
Sciences of the Netherlands Organization for Scientific Re-
search (CW-NWO), division “Procesvernieuwing voor een
Schoner Milieu”.
Reactions with Fluorinated Benzaldehydes
Stock solutions of aromatic substrates (100 mM) were pre-
pared in DMF. Enzymatic incubation mixtures (2.0 mL) con-
tained 0.6 mM fluorinated benzaldehyde, 6 mM sodium ascor-
bate, and 0.4–1.0 mM NADPH in 50 mM potassium phos-
phate pH 8.0. Ascorbate was added to prevent autooxidation References
of aromatic substrates and products. Reactions were per-
formed for 1 h at 308C. At 10 min intervals, the reaction mix-
tures were saturated with air. Reactions were started by the ad-
dition of 10–100 mg HAPMO and stopped by freezing the sam-
ples in liquid nitrogen.
The enzymatic conversion of 2-fluorobenzaldehyde was
studied at pH 6.0, 7.0 and 8.0 at 308C. The incubation mixture
[1] C. Dong, F. Huang, H. Deng, C. Schaffrath, J. B. Spencer,
D. OꢁHagan, J. H. Naismith, Nature 2004, 427, 561–565.
[2] R. P. Robinson, E. R. Laird, K. M. Donahue, L. L. Lo-
presti-Morrow, P. G. Mitchell, M. R. Reese, L. M.
Reeves, A. I. Rouch, E. J. Stam, S. A. Yocum, Bioorg.
Med. Chem. Lett. 2001, 11, 1211–1213.
(
2.0 mL) contained 0.5 mM 2-fluorobenzaldehyde, 0.6 mM
[
3] A. Tanaka, T. Terasawa, H. Hagihara, N. Ishibe, M. Sa-
wada, Y. Sakuma, M. Hashimoto, H. Takasugi, H. Tana-
ka, J. Med. Chem. 1998, 41, 4408–4420.
NADPH in 50 mM potassium phosphate. Reactions were
started by the addition of 0.7 mg HAPMO. Incubation mix-
tures were exposed to air to have maximal conversion. Record-
[
[
[
4] A. M. Ismaiel, J. De Los Angeles, M. Teitler, S. Ingher,
R. A. Glennon, J. Med. Chem. 1993, 36, 2519–2525.
5] J. T. Konkel, J. Fan, B. Jayachandran, K. L. Kirk, J. Fluo-
rine Chem. 2002, 115, 27–32.
6] W. Vaalburg, H. H. Coenen, C. Crouzel, P. H. Elsinga, B.
Langstrom, C. Lemaire, G. J. Meyer, Int. J. Rad. Appl.
Instrum. [B] 1992, 19, 227–237.
19
ing of F NMR spectra at 78C was initiated after 10 min of in-
cubation.
2
-Fluorobenzaldehyde was also converted by HAPMO us-
ing an NADPH generating system. To that end, 5 mM 2-fluoro-
benzaldehyde was incubated for 2 h at 308C in 50 mM potassi-
um phosphate (pH 8.0) with 0.9 mg HAPMO, 0.25 mM
þ
NADP , 10 mM glucose 6-phosphate, and 7.5 mM ascorbate.
[
7] H. H. Coenen, in: PET studies on amino acid metabolism
and protein synthesis, (Eds.: B. Mazoyer, W. D. Weiss, D.
Comar), Kluwer Academic Publisher, Dordercht, Neth-
erlands, 1993, pp. 109–129.
Thereaction(2.0 mL) was startedby the additionof 0.5 mg glu-
cose 6-phosphate dehydrogenase. At 10 minintervals, the reac-
tion mixture was saturated with air.
[
[
8] L. Conte, M. Napoli, G. P. Gambaretto, A. Guerrato,
F. M. Carlini, J. Fluorine Chem. 1994, 67, 41–45.
9] I. Ekaeva, L. Barre, M.-C. Lasne, F. Gourand, Appl. Ra-
diat. Isot. 1995, 46, 777–782.
1
9
F NMR Product Analysis
19
F NMR experiments were performed on a Bruker DPX 400
[61]
[
10] T. Ludwig, J. Ermert, H. H. Coenen, Nucl. Med. Biol.
2002, 29, 255–262.
[11] A. Corma, V. Fornes, S. Iborra, M. Mifsud, M. Renz, J.
NMR spectrometer, essentially as described elsewhere.
The temperature was 78C. A dedicated 10 mm F NMR pro-
behead was used. The spectral width for the F NMR measure-
ments was 50,000 Hz. The number of data points used for data
acquisition was 32768. Pulse angles of 308 were used. Between
1
9
1
9
Catal. 2004, 221, 67–76.
[12] R. Criegee, Justus Liebigs Ann. Chem. 1948, 560, 127–
1
000 and 66,000 scans were recorded, depending on the con-
135.
centrations of the fluorine-containing compounds and the sig-
nal-to-noise ratio required. The detection limit for an over-
night run (60,000 scans) is 1 mM. H decoupling was achieved
[
13] M. Renz, B. Meunier, Eur. J. Org. Chem. 1999, 4, 737–
50.
7
1
[
[
14] Y. Ogata, Y. Sawaki, J. Org. Chem. 1969, 34, 3985–3991.
15] A. Corma, L. T. Nemeth, M. Renz, S. Valencia, Nature
with a Waltz16 decoupling sequence.
The sample volume was 1.6 mL. For calibration, an insert
2
001, 412, 423–425.
16] G. J. ten Brink, I. W. Arends, R. A. Sheldon, Chem. Rev.
004, 104, 4105–4124.
containing D O and a calibrated amount of 4-fluorobenzoate
2
[
[
[
[
was used which also served as deuterium lock for locking the
magnetic field. For the enzymatic conversion of 4-fluoroben-
2
17] G. R. Krow, in: Organic Reactions, (Ed.: L. A. Paquette),
John Wiley & Sons, New York, 1993, pp. 251–798.
18] S. L. Friess, A. H. Soloway, J. Am. Chem. Soc. 1951, 73,
3968–3972.
19] W. v. E. Doering, L. Speers, J. Am. Chem. Soc. 1950, 72,
5515–5518.
zaldehyde, a 1625 mL sample with addition of 100 mL D O
2
was used without the insert. Concentrations of the various flu-
orinated compounds were calculated by comparison of the in-
19
tegrals of their F NMR resonances with the integral of the 4-
fluorobenzoate resonance. Chemical shifts are reported rela-
19
tive to CFCl . F NMR chemical shift values of the various flu-
3
[
[
20] G. Strukul, Angew. Chem. Int. Ed. 1998, 37, 1198–1209.
21] S. Murahashi, S. Ono, Y. Imada, Angew. Chem. Int. Ed.
2002, 41, 2366–2368.
orine-containing compounds were identified using authentic
fluorinated benzaldehydes, benzoic acids and phenols or previ-
1
9
ously reported resonances. The F NMR spectral data were
corrected for impurities (<5%) of fluorobenzoic acids and flu-
orophenols present in some substrate solutions.
[22] C. Mazzini, J. Lebreton, R. Furstoss, J. Org. Chem. 1996,
61, 8–9.
Adv. Synth. Catal. 2005, 347, 1027–1034
asc.wiley-vch.de
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1033

FULL PAPERS
Mari e¨ lle J. H. Moonen et al.
[
[
[
23] O. Fukuda, S. Sakaguchi, Y. Ishii, Tetrahedron Lett. 2001,
[43] M. T. Reetz, B. Brunner, T. Schneider, F. Schulz, C. M.
Clouthier, M. M. Kayser, Angew. Chem. Int. Ed. 2004,
43, 4075–4078.
[44] M. J. H. Moonen, I. M. C. M. Rietjens, W. J. H. van Ber-
kel, J. Ind. Microbiol. Biotechnol. 2001, 26, 35–42.
[45] D. Sheng, D. P. Ballou, V. Massey, Biochemistry 2001, 40,
11156–11167.
4
2, 3479–3481.
24] C. Bolm, C. Palazzi, G. Francio, W. Leitner, Chem. Com-
mun. 2002, 1588–1589.
25] M. Hudlicky, in: Oxidations in Organic Chemistry, ACS
Monograph 186, American Chemical Society, Washing-
ton, DC, 1990, p. 164.
[
[
[
[
26] A. Adejare, J. Shen, A. M. Ogunbadeniyi, J. Fluorine
[46] M. Husain, B. Entsch, D. P. Ballou, V. Massey, P. J. Chap-
man, J. Biol. Chem. 1980, 255, 4189–4197.
[47] A. P. Jadan, M. J. H. Moonen, S. A. Boeren, L. A. Golov-
leva, I. M. C. M. Rietjens, W. J. H. van Berkel, Adv.
Synth. Catal. 2004, 346, 367–375.
[48] M. G. Taylor, V. Massey, J. Biol. Chem. 1990, 265,
13687–13694.
[49] J. M. Schwab, J. Am. Chem. Soc. 1981, 103, 1876–1878.
[50] J. M. Schwab, W. Li, L. P. Thomas, J. Am. Chem. Soc.
1983, 105, 4800–4804.
[51] E. Malito, A. Alfieri, M. W. Fraaije, A. Mattevi, Proc.
Natl. Acad. Sci. USA 2004, 101, 13157–13162.
[52] M. D. Mihovilovic, P. Kapitan, J. Rydz, F. Rudroff, F. H.
Ogink, M. W. Fraaije, J. Mol. Catal. B: Enzym. 2005, 32,
135–140.
[53] V. Alphand, R. Furstoss, J. Org. Chem. 1992, 57, 1306–
1309.
[54] J. D. Stewart, K. W. Reed, C. A. Martinez, J. Zhu, G.
Chen, M. M. Kayser, J. Am. Chem. Soc. 1998, 120,
3541–3548.
Chem. 2000, 105, 107–109.
27] G. J. ten Brink, J. M. Vis, I. W. Arends, R. A. Sheldon, J.
Org. Chem. 2001, 66, 2429–2433.
28] M. D. Mihovilovic, B. Mꢂller, P. Stanetty, Eur. J. Org.
Chem. 2002, 2002, 3711–3730.
29] M. W. Fraaije, J. Wu, D. P. Heuts, E. W. van Hellemond,
J. H. Spelberg, D. B. Janssen, Appl. Microbiol. Biotech-
nol. 2005, 66, 393–400.
[
[
[
[
[
[
[
30] N. M. Kamerbeek, D. B. Janssen, W. J. H. van Berkel,
M. W. Fraaije, Adv. Synth. Catal. 2003, 345, 667–678.
31] A. Hage, H. E. Schoemaker, J. A. Field, Appl. Microbiol.
Biotechnol. 1999, 52, 834–838.
32] F. R. Lauritsen, A. Lunding, Enz. Microb. Technol. 1998,
2
2, 459–465.
33] W. Kroutil, H. Mang, K. Edegger, K. Faber, Adv. Synth.
Catal. 2004, 346, 125–142.
34] F. Guillen, A. T. Martinez, M. J. Martinez, Eur. J. Bio-
chem. 1992, 209, 603–611.
35] R. H. H. van den Heuvel, C. Laane, W. J. H. van Berkel,
Adv. Synth. Catal. 2001, 343, 515–520.
[55] N. M. Kamerbeek, M. W. Fraaije, D. B. Janssen, Eur. J.
Biochem. 2004, 271, 2107–2116.
[56] A. Z. Walton, J. D. Stewart, Biotechnol. Prog. 2002, 18,
262–268.
36] P. Ferreira, M. Medina, F. Guill e´ n, M. J. Martꢃnez,
W. J. H. van Berkel, A. T. Martꢃnez, Biochem. J. 2005,
in press.
[
37] M. J. H. Moonen, I. M. C. M. Rietjens, W. J. H. van Ber-
kel, in: 13th International Congress on Flavins and Flavo-
proteins, (Eds.: S. Ghisla, P. Kroneck, P. Macheroux, H.
Sund), Agency for Scientific Publication, Berlin,
[57] S. D. Doig, P. J. Avenell, P. A. Bird, P. Gallati, K. S. Land-
er, G. J. Lye, R. Wohlgemuth, J. M. Woodley, Biotechnol.
Prog. 2002, 18, 1039–1046.
[58] I. Hilker, V. Alphand, R. Wohlgemuth, R. Furstoss, Adv.
Synth. Catal. 2004, 346, 203–214.
1
999, pp. 375–378.
[
38] N. M. Kamerbeek, M. J. H. Moonen, J. G. M. van der
Ven, W. J. H. van Berkel, M. W. Fraaije, D. B. Janssen,
Eur. J. Biochem. 2001, 268, 2547–2557.
[59] W. J. H. van Berkel, M. H. M. Eppink, W. J. Middel-
hoven, J. Vervoort, I. M. C. M. Rietjens, FEMS Micro-
biol. Lett. 1994, 121, 207–215.
[
[
[
[
39] N. M. Kamerbeek, A. J. Olsthoorn, M. W. Fraaije, D. B.
Janssen, Appl. Environ. Microbiol. 2003, 69, 419–426.
40] C. T. Walsh, Y.-C. J. Chen, Angew. Chem. 1988, 100,
[60] W. van Berkel, A. Westphal, K. Eschrich, M. Eppink, A.
de Kok, Eur. J. Biochem. 1992, 210, 411–419.
[61] J. Vervoort, P. A. de Jager, J. Steenbergen, I. M. C. M.
Rietjens, Xenobiotica 1990, 20, 657–670.
[62] S. Peelen, I. M. C. M. Rietjens, M. G. Boersma, J. Ver-
voort, Eur. J. Biochem. 1995, 227, 284–291.
3
42–352.
41] V. Alphand, G. Carrea, R. Wohlgemuth, R. Furstoss,
J. M. Woodley, Trends Biotechnol. 2003, 21, 318–323.
42] M. T. Reetz, F. Daligault, B. Brunner, H. Hinrichs, A.
Deege, Angew. Chem. Int. Ed. 2004, 43, 4078–4081.
1034
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 1027–1034