Enzymatic synthesis of vanillin

Article abstract of DOI:10.1021/jf010093j

Due to increasing interest in natural vanillin, two enzymatic routes for the synthesis of vanillin were developed. The flavoprotein vanillyl alcohol, oxidase (VAO) acts on a wide range of phenolic compounds and converts both creosol and vanillylamine to vanillin with high yield. The VAO-mediated conversion of creosol proceeds via a two-step process in which the initially formed vanillyl alcohol is further oxidized to vanillin. Catalysis is limited by the formation of an abortive complex between enzyme-bound flavin and creosol. Moreover, in the second step of the process, the conversion of vanillyl alcohol is inhibited by the competitive binding of creosol. The VAO-catalyzed conversion of vanillylamine proceeds efficiently at alkaline Ph values. Vanillylamine is initially converted to a vanillylimine intermediate product, which is hydrolyzed nonenzymatically to vanillin. This route to vanillin has biotechnological potential as the widely available principle of red pepper, capsaicin, can be hydrolyzed enzymatically to vanillylamine.

Full text of DOI:10.1021/jf010093j

2954
J. Agric. Food Chem. 2001, 49, 2954−2958
En zym a tic Syn th esis of Va n illin
Robert H. H. van den Heuvel,† Marco W. Fraaije,‡ Colja Laane,†,§ and Willem J . H. van Berkel*,†
Laboratory of Biochemistry, Wageningen University, Dreijenlaan 3, 6703 HA Wageningen,
The Netherlands, and Laboratory of Biochemistry, Department of Chemistry, University of Groningen,
Nijenborgh 4, 9747 AG Groningen, The Netherlands
Due to increasing interest in natural vanillin, two enzymatic routes for the synthesis of vanillin
were developed. The flavoprotein vanillyl alcohol oxidase (VAO) acts on a wide range of phenolic
compounds and converts both creosol and vanillylamine to vanillin with high yield. The VAO-
mediated conversion of creosol proceeds via a two-step process in which the initially formed vanillyl
alcohol is further oxidized to vanillin. Catalysis is limited by the formation of an abortive complex
between enzyme-bound flavin and creosol. Moreover, in the second step of the process, the conversion
of vanillyl alcohol is inhibited by the competitive binding of creosol. The VAO-catalyzed conversion
of vanillylamine proceeds efficiently at alkaline pH values. Vanillylamine is initially converted to
a vanillylimine intermediate product, which is hydrolyzed nonenzymatically to vanillin. This route
to vanillin has biotechnological potential as the widely available principle of red pepper, capsaicin,
can be hydrolyzed enzymatically to vanillylamine.
Keyw or d s: Capsaicin; creosol; flavoprotein; vanillin; vanillyl alcohol oxidase; vanillylamine
INTRODUCTION
nol) is the principal constituent of clove oil and is an
economically realistic feedstock. VAO catalyzes the
hydroxylation of eugenol to coniferyl alcohol (4-hydroxy-
3-methoxycinnamyl alcohol) (14, 15). This aromatic
alcohol can be further converted, via ferulic acid (4-
hydroxy-3-methoxycinnamic acid), to vanillin by several
microorganisms (1, 12).
Vanillin (4-hydroxy-3-methoxybenzaldehyde) is a
widely used flavor compound in food and personal
products (1), and in high concentrations, the antioxi-
dizing properties of this compound prevent oxidative
damage in mammalian cells (2, 3). By far the dominant
route for vanillin production is chemical synthesis from
lignin, coniferin, the glucoside of coniferyl alcohol (4-
hydroxy-3-methoxycinnamyl alcohol), guaiacol (2-meth-
oxyphenol), or eugenol (4-allyl-2-methoxyphenol) (1).
Natural vanillin supplies <1% of the total demand for
vanillin and is produced from glucovanillin when the
beans of the orchid Vanilla planifolia are submitted to
a multistep curing process. After this curing process,
vanillin is the most abundant component of the bean
at a level of ∼2-3 wt % (4, 5).
VAO can also produce vanillin by catalyzing the
oxidation of vanillyl alcohol (4-hydroxy-3-methoxybenzyl
alcohol), the oxidative demethylation of 2-methoxy-4-
(methoxymethyl)phenol, and the oxidative deamination
of vanillylamine (4-hydroxy-3-methoxybenzylamine) (14)
(Figure 1A-C). Vanillyl alcohol, 2-methoxy-4-(meth-
oxymethyl)phenol, and vanillylamine are not widely
available in nature and thus not of direct interest for
biotechnological applications. However, vanillylamine
can be obtained from the abundant precursor compound
capsaicin (8-methyl-N-vanillyl-6-nonenamide), the pun-
gent principle of hot red pepper, by the cleavage of its
amide bond (16, 17) (Figure 1D). Another potential
feedstock for vanillin is creosol (2-methoxy-p-cresol),
which is the major component in creosote obtained from
heating wood or coal tar (18, 19). On the basis of the
reactivity of VAO with p-cresol it was anticipated that
creosol is oxidized to vanillin (20) (Figure 1E).
With the increasing interest in natural products,
alternative processes are being developed to produce
natural vanillin. The term “natural” can be applied both
in the European Union and in the United States when
the product has been derived from a natural raw
material via biological (e.g., enzymes or whole cells) and/
or mild processing tools (e.g., extraction or distillation)
(6). In recent years, a large number of studies have been
made on natural vanillin biosynthesis using microor-
ganisms or isolated enzymes (7-12). However, these
bioconversions are not yet economically feasible.
The flavoprotein vanillyl alcohol oxidase (VAO; EC
1.1.3.38) from Penicillium simplicissimum (13) is a
versatile biocatalyst that is of particular interest for the
production of vanillin. Eugenol (4-allyl-2-methoxyphe-
In this study, we have investigated the VAO-mediated
synthesis of vanillin from natural feedstocks. We se-
lected creosol, which is the major product in creosote,
and vanillylamine, as this compound can be obtained
from capsaicin.
MATERIALS AND METHODS
* Author whom correspondence should be addressed (tele-
phone +31-317-482861; fax +31-317-484801; e-mail
willem.vanberkel@fad.bc.wau.nl).
Ma ter ia ls. Creosol (2-methoxy-p-cresol), vanillylamine (4-
hydroxy-3-methoxybenzylamine), vanillyl alcohol (4-hydroxy-
3-methoxybenzyl alcohol), and vanillin (4-hydroxy-3-methox-
ybenzaldehyde) were obtained from Aldrich (Steinheim,
Germany), and capsaicin (8-methyl-N-vanillyl-6-nonenamide)
was from Fluka (Buchs, Switzerland). Penicillin G acylase from
Escherichia coli was a kind gift from DSM/Gist (Delft, The
† Wageningen University.
‡ University of Groningen.
§ Present address: DSM Food Specialties Research and
Development, P.O. Box 1, 2600 MA Delft, The Netherlands.
10.1021/jf010093j CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/17/2001

Enzymatic Synthesis of Vanillin
J. Agric. Food Chem., Vol. 49, No. 6, 2001 2955
F igu r e 2. pH dependence of the reaction of VAO with creosol.
The activity measurements were performed in 50 mM succi-
nate/NaOH (pH 5.6-6.9), 50 mM potassium phosphate buffer
(pH 6.1-7.8), 50 mM Tris/H2SO4 (pH 7.9-8.9), and 50 mM
glycine/NaOH (pH 9.1-10.4) at 25 °C using 150 µM creosol
and 0.5 µM VAO.
Ta ble 1. Stea d y-Sta te Kin etic P a r a m eter s for VAO in
Air -Sa tu r a ted 50 m M P ota ssiu m P h osp h a te Bu ffer , p H
7.5, a t 25 °C
k′cat/K′m
substrate
k′cat (s-1
)
K′m (µM) (s-1 mM-1
)
4-(methoxymethyl)phenola
vanillylamine
3.1
55
48
50
56.4
0.42
1.4
0.02
0.07
3.3
creosol
vanillyl alcohola
160
20.6
a
Data from ref 26.
F igu r e 1. VAO-catalyzed conversion of (A) vanillyl alcohol,
(B) 2-methoxy-4-(methoxymethyl)phenol, and (C) vanilly-
lamine to vanillin and two-step enzymatic conversion of (D)
capsaicin and (E) creosol to natural vanillin.
using mixtures of methanol and water containing 1% (v/v)
acetic acid. For monitoring the conversion of creosol to vanillin
a methanol/water ratio of 35:65 (v/v) was used. For monitoring
the conversion of capsaicin to vanillylamine, elution was
started with a linear gradient from 8 to 13% (v/v) methanol
in 20 min followed by an increase to 75% in 3 min, which was
held for 20 min. The VAO-mediated conversion of creosol was
carried out in different buffers at 25 °C. The hydrolysis of
capsaicin by penicillin G acylase was performed in 50 mM
potassium phosphate buffer, pH 7.5, at 25 °C, and the
hydrolysis of capsaicin by rat liver microsomes was performed
in 50 mM potassium phosphate buffer, pH 7.0, at 37 °C.
Chemical hydrolysis of capsaicin was carried out in 10 or 100
mM NaOH at 95 °C. In all reaction mixtures capsaicin was
added as a powder. The reactions were terminated by the
addition of 5% (w/v) trichloroacetic acid and subsequent
centrifugation.
Netherlands). Lipase B from Candida antarctica was obtained
from Novo Nordisk (Bagsværd, Denmark). Carboxypeptidase
A from bovine pancreas, carboxypeptidase Y from yeast,
acylase from pig kidney, and thermolysin from Bacillus
thermoproteolyticus were purchased from Sigma (St. Louis,
MO). All other chemicals were obtained as previously described
(21, 22). Rat liver microsomes were prepared essentially as
described by Rietjens and Vervoort (23).
Exp r ession a n d P u r ifica tion of VAO. E. coli strain TG2
(24) and the plasmid pEMBL19 (Boehringer Mannheim) were
used for expression of the vaoA gene (25). Transformed E. coli
cells were grown in Luria-Bertani medium supplemented
with 75 µg/mL ampicillin and 0.25 mM isopropyl â-D-thio-
galactopyranoside (25). VAO was purified as described to a
purity of >98% (22, 25).
RESULTS
An a lytica l Meth od s. VAO activity was determined at 25
°C by monitoring changes in the absorption spectra of aromatic
products or by oxygen consumption experiments using a Clark
electrode (14). Vanillin production was measured at 340 nm
(ꢀ340 varies from 2.2 mM/cm at pH 5.6 to ꢀ340 ) 23.3 mM/cm at
pH 10.5 with a pKa of 7.2) and vanillylimine production at 390
nm. For enzyme-monitored-turnover experiments, air-satu-
rated enzyme and substrate solutions were mixed, and the
redox state of the flavin prosthetic group was monitored at
439 nm using a Hewlett-Packard HP 8453 diode array spec-
trophotometer. Stopped-flow kinetics were performed with a
Hi-Tech SF-51 apparatus equipped with a Hi-Tech SU-40
spectrophotometer, essentially as described previously (26). In
anaerobic reduction experiments, glucose-containing enzyme
solutions were flushed with oxygen-free argon gas and glucose
oxidase was added to eliminate final traces of oxygen. Fluo-
rescence emission spectra were recorded at 25 °C on an Aminco
SPF-500C spectrofluorometer; the excitation wavelength was
360 nm (20).
Con ver sion of Cr eosol. Table 1 summarizes the
steady-state kinetic parameters of VAO with 4-(meth-
oxymethyl)phenol, vanillyl alcohol, vanillylamine, and
creosol at pH 7.5. 4-(Methoxymethyl)phenol and vanillyl
alcohol are efficient substrates for VAO, whereas both
vanillylamine and creosol react rather slowly with the
enzyme. In contrast to the other substrates, the conver-
sion of creosol proceeds via a two-step enzymatic pro-
cess. In the first step creosol is hydroxylated by VAO to
yield vanillyl alcohol, and in the second step vanillyl
alcohol is oxidized to yield vanillin.
Earlier studies have shown that the reaction of VAO
with 4-(methoxymethyl)phenol and vanillyl alcohol is
optimal around pH 10 (13, 14). However, when the VAO-
mediated conversion of creosol was studied at pH 10,
almost no activity was observed. Upon determining the
pH optimum of the reaction, we found that VAO is most
active with creosol between pH 7 and 8 (Figure 2). This
pH optimum is identical to the optimum of the VAO-
mediated oxidation of p-cresol (20) and clearly distinct
HPLC experiments were performed with an Applied Bio-
systems pump equipped with a Waters 996 photodiode array
detector and a 4.6 × 150 mm Alltima C18 column (Alltech)

2956 J. Agric. Food Chem., Vol. 49, No. 6, 2001
van den Heuvel et al.
F igu r e 3. Observed reduction rates of VAO with various
concentrations of creosol. VAO (2.5 µM) was mixed anaerobi-
cally with creosol (25-500 µM) in 50 mM potassium phosphate
buffer, pH 7.5, at 25 °C. Flavin reduction was followed at 439
nm. The arrow indicates the value found for the reverse
reaction (k-2).
F igu r e 4. HPLC analysis of the VAO-mediated conversion
of creosol. The reaction mixture contained 100 µM creosol and
1 µM VAO in 50 mM potassium phosphate buffer, pH 7.5, and
was incubated at 25 °C. The concentrations of creosol (b),
vanillyl alcohol (9), and vanillin (2) were determined from
reference solutions.
from other VAO-catalyzed oxidations (13). It should be
noted here that the pH optimum of the reaction with
creosol was determined at a substrate concentration of
150 µM. At higher substrate concentrations the overall
conversion rate decreased due to the competitive binding
of creosol, inhibiting the conversion of the intermediate
product vanillyl alcohol.
Table 1 shows that the catalytic efficiency of VAO
with creosol at pH 7.5 is 15-fold lower compared to the
efficiency with vanillyl alcohol. This suggests that the
slow hydroxylation of creosol limits the synthesis of
vanillin. Therefore, we studied the initial enzymatic
conversion of creosol to vanillyl alcohol in more detail.
This reaction can be described by two half-reactions.
First, the enzyme-bound flavin cofactor is reduced by
the substrate, forming a complex between the reduced
enzyme and the p-quinone methide intermediate of
creosol (eq 1). In the second step, the reduced enzyme
is reoxidized by molecular oxygen and water attacks the
intermediate product, forming vanillyl alcohol (eq 2).
that the enzyme-bound flavin was mainly in the reduced
state during turnover at pH 7.5 (66%), again indicating
that the reductive half-reaction does not limit the
turnover rate. Moreover, upon excitation at 360 nm of
the aerobic complex between creosol and VAO, a strong
fluorescence emission with a maximum at 470 nm was
observed. In line with earlier results, this suggests the
formation of an abortive covalent flavin-substrate
adduct, which is stabilized under aerobic conditions (20).
As mentioned previously, the VAO-mediated turnover
of creosol at pH 10 was 10-fold slower than that at pH
7.5 (Figure 2). When VAO and creosol were mixed at
pH 10 under aerobic conditions, the flavin was almost
completely in the reduced state (95%). Moreover, the
fluorescence emission spectrum upon excitation at 360
nm became stronger at pH 10 compared to pH 7.5.
These results clearly support the proposition that the
decreased turnover rate of creosol at high pH values is
caused by the increased stability of the abortive creosol-
flavin adduct.
Eox + creosol yk1z Eox∼creosol yk2z
When the VAO-mediated conversion of creosol was
followed by HPLC, only low concentrations of vanillyl
alcohol were detected at pH 7.5 and 10 (Figure 4). This
is in agreement with the slow conversion of creosol
relative to vanillyl alcohol. At both pH 7.5 and 10 the
final yield of vanillin was 100% when the reaction was
performed at a starting substrate concentration of 100
µM, as determined by HPLC and spectral analysis.
k-1
k-2
Ered∼quinone methide (1)
k3
98
Ered∼quinone methide + O2 + H2O
Eox + vanillyl alcohol + H2O2 (2)
In most VAO-catalyzed reactions studied so far, the
reductive half-reaction limits catalysis (20, 26). When
creosol and VAO were mixed in the stopped-flow spec-
trophotometer under anaerobic conditions, a monopha-
sic decrease in absorbance at 439 nm was observed,
indicative of the reduction of enzyme-bound flavin. The
rate of reduction was dependent on the concentration
of creosol with a dissociation constant Kd ) 159 ( 34
µM at pH 7.5 (Figure 3). At low substrate concentrations
Va n illyla m in e P r od u ction . Capsaicin, the pungent
principle of red pepper, is a cheap feedstock for the
production of vanillylamine. In agreement with earlier
reports (16, 17), we observed that capsaicin can be
metabolized by enzymes from liver via vanillylamine to
the end products vanillyl alcohol and vanillic acid. In
our experiments capsaicin was converted by rat liver
microsomes to vanillylamine at a rate of 1.2 × 10-3 mM/
min/mg of protein. Potentially, the amide bond in
capsaicin can also be hydrolyzed by other hydrolases.
Therefore, several commercially available enzymes were
tested for their capacities to hydrolyze capsaicin. Among
these enzymes (see Materials and Methods) only peni-
cillin G acylase was able to hydrolyze capsaicin at a rate
of 3.5 × 10-6 mM/min/mg. Vanillylamine was detected
as the single aromatic product. For comparison, when
5 mM capsaicin was incubated in 100 mM NaOH at 95
°C, the chemical hydrolysis proceeded with a pseudo-
first-order rate of 2.0 × 10-3/min, whereas in 10 mM
NaOH at 95 °C no hydrolysis occurred.
the reduction rate reached a finite value of 0.16 s-1
,
suggesting that the reduction is a reversible process in
which k-2 ) 0.16 ( 0.01 s-1 (20, 27). The forward
reduction rate (k2) was calculated to be 0.30 ( 0.02 s-1
,
which is 4-fold higher than the turnover rate of VAO
with this substrate. This strongly indicates that the
reductive half-reaction, in analogy to the VAO-mediated
conversion of p-cresol, only partially limits catalysis (20).
When VAO was mixed with creosol under aerobic
conditions, the redox state of the flavin cofactor could
be followed spectrophotometrically at 439 nm. We found

Enzymatic Synthesis of Vanillin
J. Agric. Food Chem., Vol. 49, No. 6, 2001 2957
turnover at both pH 7.5 (80%) and pH 9.7 (>90%),
strongly indicating that, in contrast to the VAO-medi-
ated conversion of creosol, the reductive half-reaction
limits the turnover rate.
When the enzymatic deamination of vanillylamine
was monitored spectrophotometrically, the initial for-
mation of the intermediate product vanillylimine and
its subsequent conversion to vanillin could be observed
at 390 and 340 nm, respectively (Figure 6). The con-
centration of vanillylimine during catalysis was calcu-
lated using the molar absorption coefficient of vanillin
(ꢀ340 ) 22.9 mM/cm at pH 9.0) and the absorbance
changes in time due to vanillylimine and vanillin
formation. The transient accumulation of relatively low
amounts of vanillylimine is in line with the observation
that the enzymatic formation of vanillylimine is the
rate-limiting step in catalysis. For the VAO-mediated
conversion of creosol, the yield of vanillin production
reached almost 100%.
F igu r e 5. pH dependence of the reaction of VAO with
vanillylamine. VAO activity was monitored either by spectral
analysis of vanillin production (b) or oxygen consumption (O).
The reactions were performed in 50 mM potassium phosphate
buffer (pH 7.0-7.8), 50 mM Tris/H2SO4 (pH 7.8-8.9), and 50
mM glycine/NaOH (pH 9.0-10.5) at 25 °C using 1 mM
vanillylamine and 1 µM VAO for spectral studies or 4 µM VAO
for oxygen consumption experiments.
DISCUSSION
In the present study we have addressed the catalytic
potential of VAO for the production of vanillin. A
potentially attractive feedstock for the enzymatic forma-
tion of vanillin is creosol, which can be obtained from
creosote (18, 19). The VAO-mediated conversion of
creosol to vanillin reaches a yield of 100%, but the
conversion rate is rather low due to the formation of a
nonreactive covalent adduct between creosol and the
flavin prosthetic group of VAO. The covalent adduct
slowly decomposes to vanillyl alcohol, thereby determin-
ing the overall turnover rate of VAO. The flavin-creosol
adduct is more stable at basic pH values, hence shifting
the pH optimum of VAO, which is normally around pH
10, to pH 7.5 (13, 26). Moreover, the competitive binding
of creosol inhibited the conversion of the intermediate
product vanillyl alcohol to vanillin. As a consequence,
the optimal conditions for the VAO-catalyzed conversion
of creosol were found to be at pH 7.5 and a substrate
concentration of 150 µM.
A second potentially attractive feedstock for the
enzymatic production of vanillin is capsaicin. This
pungent principle of red pepper can be easily obtained
at low cost and can be hydrolyzed enzymatically to
vanillylamine by a carboxylesterase from liver. This
esterase has been found in the liver of chicken, hog, cow,
and rat (16), and the rat liver enzyme has been purified
and characterized to some extent (17). Capsaicin is not
hydrolyzed by trypsin, peptidase, or aminoacylase (16).
We found that capsaicin is extremely slowly hydrolyzed
by penicillin G acylase from E. coli. The rate of this
enzymatic hydrolysis reaction might be increased by
performing the reaction in organic solvents, as the
solubility of capsaicin would be enhanced.
F igu r e 6. Enzymatic conversion of vanillylamine via vanil-
lylimine to vanillin. The reaction mixture contained 50 mM
glycine/NaOH, pH 9.0, 100 µM vanillylamine, and 0.5 µM VAO
and was incubated at 25 °C. Spectra were taken at regular
time intervals after the addition of VAO. The inset shows the
estimated concentrations of (1) vanillamine, (2) vanillylimine,
and (3) vanillin, respectively.
Con ver sion of Va n illyla m in e. In addition to creo-
sol, vanillylamine obtained from capsaicin is a natural
precursor of vanillin. The VAO-catalyzed conversion of
vanillylamine involves the initial formation of vanil-
lylimine. Subsequently, this intermediate product is
hydrolyzed nonenzymatically to vanillin. When the
conversion of vanillylamine by VAO was studied at pH
7.5, the turnover rate of the enzyme was very low (Table
1). This prompted us to study the pH dependence of the
enzymatic reaction by oxygen consumption experiments
as well as spectral analysis. With both methods we
found similar pH-dependent turnover rates (Figure 5).
The rate of the VAO-mediated conversion of vanilly-
lamine increased dramatically when the pH of the
reaction medium was raised to pH >9. Again, this pH
profile is clearly distinct from that of other VAO-
catalyzed reactions (13, 14). Due to the protein instabil-
ity it was not possible to measure reliable turnover rates
at pH values >10.5. When VAO and vanillylamine were
mixed aerobically, the redox state of the flavin cofactor
could be followed spectrophotometrically. It was found
that the enzyme is mainly in the oxidized state during
Vanillylamine reacted slowly with VAO at neutral pH
values but was efficiently converted between pH 9 and
10.5 with a high vanillin yield. This strong increase in
VAO activity around pH 9 is unusual. As the rate-
limiting step in the conversion of vanillylamine to
vanillin is associated with the formation of the vanil-
lylimine intermediate, the pH dependence of this reac-
tion might be related to the binding of vanillylamine.
For other VAO substrates, it is thought that the initial
formation of the quinone methide intermediate (eq 1)
is facilitated by the preferential binding of the phenolate
form of the substrate (28). For example, when the
substrate analogue isoeugenol binds to VAO, the pKa

2958 J. Agric. Food Chem., Vol. 49, No. 6, 2001
van den Heuvel et al.
(9) Lomascolo, A.; Stentelaire, C.; Asther, M.; Lesage-
Meessen, L. Trends Biotechnol. 1999, 17, 282-289.
(10) Overhage, J .; Priefert, H.; Rabenhorst, J .; Steinbuechel,
A. Appl. Microbiol. Biotechnol. 1999, 52, 820-828.
(11) Li, T.; Rosazza, J . P. Appl. Environ. Microbiol. 2000, 66,
684-687.
(12) Walton, N. J .; Narbad, A.; Faulds, C. B.; Williamson,
G. Curr. Opin. Biotechnol. 2000, 11, 490-496.
(13) de J ong, E.; van Berkel, W. J . H.; van der Zwan, R. P.;
de Bont, J . A. M. Eur. J . Biochem. 1992, 208, 651-657.
(14) Fraaije, M. W.; Veeger, C.; van Berkel, W. J . H. Eur. J .
Biochem. 1995, 234, 271-277.
(15) van Berkel, W. J . H.; Fraaije, M. W.; de J ong, E. Process
for Producing 4-Hydroxycinnamyl Alcohols; 1997,
European Patent Application 071028gB1.
(16) Oi, Y.; Kawada, T.; Watanabe, T.; Iwai, K. J . Agric. Food
Chem. 1992, 40, 467-470.
(17) Park, Y. H.; Lee, S. S. Biochem. Mol. Biol. Int. 1994,
34, 351-360.
(18) Ogata, N.; Baba, T. Res. Commun. Chem. Pathol.
Pharmacol. 1989, 66, 411-423.
(19) Dyreborg, S.; Arvin, E.; Broholm, K. Biodegradation
1996, 7, 191-201.
(20) Fraaije, M. W.; van den Heuvel, R. H. H.; Roelofs, J . C.
A. A.; van Berkel, W. J . H. Eur. J . Biochem. 1998, 253,
712-719.
(21) van den Heuvel, R. H. H.; Fraaije, M. W.; Laane, C.;
van Berkel, W. J . H. J . Bacteriol. 1998, 180, 5646-5651.
(22) van den Heuvel, R. H. H.; Fraaije, M. W.; Mattevi, A.;
van Berkel, W. J . H. J . Biol. Chem. 2000, 275, 14799-
14808.
of the phenol decreases from 9.8 to 5.0 (14). Possibly,
binding of vanillylamine to the enzyme does not stimu-
late phenol deprotonation. Alternatively, the pH effect
might be related to the preferred binding of the pheno-
late form of vanillylamine. On the other hand, the
protonation state of the amide moiety of vanillylamine
might also be of importance for its reactivity.
In conclusion, this paper presents two novel enzy-
matic routes for the biocatalytic production of natural
vanillin. The bienzymatic process from capsaicin to
vanillin is most promising from a biotechnological point
of view, as capsaicin is a widely available compound.
To make this process more feasible, the first enzyme in
the reaction sequence, a carboxylesterase, needs to be
characterized in detail and overexpressed in a suitable
host. The VAO-catalyzed deamination of vanillylamine
is efficient, does not need any external cofactors, and
uses molecular oxygen as a clean and mild oxidant.
Moreover, the enzyme can be easily obtained in large
amounts (25). When the two enzymes can be combined
in a single reaction mixture, one can design a one-pot
reactor to produce natural vanillin. Because the pH
optima of these two enzymatic reactions vary between
pH 8-9 for the carboxylesterase (17) and pH 10-10.5
for VAO, one can assume that the optimum pH condi-
tions for a one-pot system are between pH 9 and 10.
ACKNOWLEDGMENT
We thank Maurice Franssen (Wageningen University)
for fruitful discussions and the gift of lipase B from
Candida antarctica and Marlou van Iersel (Wageningen
University) for the gift of rat liver microsomes. We also
thank Niek Bastiaensen and Rigoberto Gonzales for
technical assistance.
(23) Rietjens, I. M.; Vervoort, J . Xenobiotica 1989, 19, 1297-
1305.
(24) Gibson, T. G. Studies on the Epstein-Barr virus ge-
nome. Ph.D. dissertation, University of Cambridge,
Cambridge, U.K., 1984.
(25) Benen, J . A. E.; Sanchez-Torres, P.; Wagemaker, M. J .
M.; Fraaije, M. W.; van Berkel, W. J . H.; Visser, J . J .
Biol. Chem. 1998, 273, 7865-7872.
LITERATURE CITED
(26) Fraaije, M. W.; van Berkel, W. J . H. J . Biol. Chem. 1997,
272, 18111-18116.
(27) Strickland, S.; Palmer, G.; Massey, V. J . Biol. Chem.
1975, 250, 4048-4052.
(28) Mattevi, A.; Fraaije, M. W.; Mozzarelli, A.; Olivi, L.;
Coda, A.; van Berkel, W. J . H. Struct. Fold. Des. 1997,
5, 907-920.
(1) Rao, S. R.; Ravishankar, G. A. J . Sci. Food Agric. 2000,
80, 289-304.
(2) Aruoma, O. I. Free Radical Res. 1999, 30, 419-427.
(3) Kamat, J . P.; Ghosh, A.; Devasagayam, T. P. Mol. Cell.
Biochem. 2000, 209, 47-53.
(4) Ranadive, A. S.; Krings, U.; Berger, R. G. J . Agric. Food
Chem. 1992, 40, 1922-1924.
(5) Prince, R. C.; Gunson, D. E. Trends Biochem. Sci. 1994,
19, 521.
(6) Lesage-Meessen, L.; Delattre, M.; Haon, M.; Thibault,
J .-F.; Colonna Ceccaldi, B.; Brunerie, P.; Asther, M. J .
Biotechnol. 1996, 50, 107-113.
Received for review J anuary 22, 2001. Revised manuscript
received April 4, 2001. Accepted April 5, 2001. This work was
performed within the framework of the Innovation Oriented
Research Program (IOP) Catalysis of the Dutch Ministry of
Economy Affairs (Project IKA 96005).
(7) Cheetham, P. S. J . Trends Biotechnol. 1993, 11, 478-
488.
(8) Hagedorn, S.; Kaphammer, B. Annu. Rev. Microbiol.
1994, 48, 773-800.
J F010093J