Epoxidation, hydroxylation and aromatization is catalyzed by a peroxygenase from Solanum lycopersicum

Article abstract of DOI:10.1016/j.molcatb.2013.07.001

Plant peroxygenase (PXG) oxidizes unsaturated fatty acids by transferring an oxygen atom of a hydroperoxide to the double bond, thereby providing epoxides. In this work we investigated the potential of a PXG from tomato (Solanum lycopersicum, SlPXG) to catalyze the oxidation of a variety of natural products. A SlPXG gene was cloned from tomato, heterologously expressed in yeast and the membrane bound recombinant SlPXG protein was used as enzyme source. Unsaturated fatty acids, fatty acid derivatives, and terpenes were epoxidized by SlPXG in the presence of various hydroperoxides exclusively at their cis-double bonds. Terpenes with p-menthene skeleton were transformed in different ways depending on their molecular structures. R-(+)- and S-(-)-limonene were converted to R-(+)-limonene-trans-1,2-epoxide (97%) and cis-S-(-)-limonene-1,2- epoxide (88%), respectively whereas α-terpinenewas hydroxylated to cis-1,4-dihydroxy-p-menth-2-ene and γ-terpinene was aromatized to p-cymene. In the last reaction the hydroperoxide served as hydrogen acceptor rather than an oxygen donor. PXG appears to be a versatile biocatalyst able to perform different kinds of oxidation reactions. As no cofactors like NAD(P)H are required and H₂O₂is an environmentally friendly oxidant, PXG enables new applications for the synthesis of fine chemicals from renewable resources.

Full text of DOI:10.1016/j.molcatb.2013.07.001

Journal of Molecular Catalysis B: Enzymatic 96 (2013) 52–60
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Epoxidation, hydroxylation and aromatization is catalyzed by a
peroxygenase from Solanum lycopersicum
Christopher Fuchs, Wilfried Schwab∗
Biotechnology of Natural Products, Technische Universität München, Liesel-Beckmann-Str. 1, 85354 Freising, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Plant peroxygenase (PXG) oxidizes unsaturated fatty acids by transferring an oxygen atom of a hydroper-
oxide to the double bond, thereby providing epoxides. In this work we investigated the potential of
a PXG from tomato (Solanum lycopersicum, SlPXG) to catalyze the oxidation of a variety of natural
products. A SlPXG gene was cloned from tomato, heterologously expressed in yeast and the mem-
brane bound recombinant SlPXG protein was used as enzyme source. Unsaturated fatty acids, fatty acid
derivatives, and terpenes were epoxidized by SlPXG in the presence of various hydroperoxides exclu-
sively at their cis-double bonds. Terpenes with p-menthene skeleton were transformed in different ways
depending on their molecular structures. R-(+)- and S-(−)-limonene were converted to R-(+)-limonene-
trans-1,2-epoxide (97%) and cis-S-(−)-limonene-1,2-epoxide (88%), respectively whereas ␣-terpinene
was hydroxylated to cis-1,4-dihydroxy-p-menth-2-ene and ␥-terpinene was aromatized to p-cymene.
In the last reaction the hydroperoxide served as hydrogen acceptor rather than an oxygen donor. PXG
appears to be a versatile biocatalyst able to perform different kinds of oxidation reactions. As no cofactors
like NAD(P)H are required and H2O2 is an environmentally friendly oxidant, PXG enables new applications
for the synthesis of fine chemicals from renewable resources.
Received 21 May 2013
Received in revised form 1 July 2013
Accepted 2 July 2013
Available online 11 July 2013
Keywords:
Peroxygenase
Enzymatic oxidation
Fatty acid
Terpene
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
were investigated due to their promising biocatalytic applications.
Numerous articles about these unspecific/aromatic peroxygenases
Plant oils and fatty acids are appreciated educts for indus-
trial applications [1,2]. Besides oil and carbohydrate, which are
compounds of the plant’s primary metabolism, the plant sec-
ondary metabolites are an important resource for the industry, too.
Terpenes belong to the largest and most diverse group of plant sec-
ondary compounds. The main sources for terpenes are the essential
oils of coniferous, citrus fruits and terebinth [1,2]. They are used as
aroma, fragrances, natural insecticides, solvents and pharmaceu-
tical active compounds [1]. Terpenes and other natural products
can be used themselves or they can be chemically and/or biotech-
nologically converted to new molecules. Especially the oxidation of
natural products is an important reaction which can be done chem-
ically by heavy metal catalysts or biotechnologically by cell cultures
or enzymes [3–5].
can be found in the literature dealing with physiological, cat-
alytic, phylogenetic and molecular aspects of the enzymes (EC
1.11.2.1) including hydroxylation and epoxidation of terpenes,
alkenes and diverse other substrates [6–8]. However, plant perox-
ygenases (PXG) belong to group EC 1.11.2.3 of the oxidoreductases.
They were first discovered in isolated microsomes from pea, Pisum
sativum, and oxidize substrates by using hydroperoxides as oxidant
[9].
The catalytically active proteins from fungi and plants con-
tain a heme molecule in the active site, which is involved in the
heterolytic cleavage of the hydroperoxide O O bond to form a
ferryl-oxo complex [10–12]. While the hydroperoxide is reduced
to an alcohol, the oxygen atom of the ferryl-oxo complex is trans-
ferred to substrates such as aromatic nitrogen compounds, organic
sulfur compounds or C C double bonds of unsaturated fatty acids.
Consequently, this oxygen transfer leads to hydroxylations, sul-
foxidations or epoxidations, respectively [11–14]. Plant PXG is a
membrane protein of the endoplasmic reticulum (ER) and oil bodies
and binds calcium in a helices E and F (EF)-hand motive, character-
istic for caleosin proteins [11]. In the plant metabolism PXG is part
of the oxylipin pathway. After the action of a lipoxygenase, which
catalyses the dioxygenation of fatty acids to fatty acid hydroper-
oxides, PXG subsequently uses these fatty acid hydroperoxides for
Enzymatic oxidations are generally catalyzed by members of
the enzyme class 1 (oxidoreductases). Peroxygenases are one sub-
class of oxidoreductases and are widespread in the fungal and plant
kingdom. Peroxygenases from fungi (EC 1.11.2.1) are able to cat-
alyze hydroxylations of aromatic compounds and fatty acids and
∗
Corresponding author. Tel.: +49 8161 71 2912; fax: +49 8161 71 2950.
E-mail address: schwab@wzw.tum.de (W. Schwab).
1381-1177/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2013.07.001

C. Fuchs, W. Schwab / Journal of Molecular Catalysis B: Enzymatic 96 (2013) 52–60
53
epoxidation reactions [11,13,14]. Plant PXG is able to transfer one
oxygen atom to the C C double bond of unsaturated fatty acids
(intermolecular transfer) or to the C C double bond of the fatty
acid hydroperoxide (intramolecular transfer). After epoxidation,
epoxide hydrolases can cleave the formed epoxide to yield a diol.
The pathway delivers to a mixture of oxidized fatty acids called
oxylipins. These molecules are monomers for cutin biosynthesis
and phytoalexins which are involved in the defence of the plant
against biotic and abiotic stresses [15–17].
Recently, plant PXG protein was isolated from oat seeds (Avena
sativa) to investigate the epoxidation of a variety of fatty acids and
alkenes for biotechnological purposes [18,19]. The present work
was undertaken to test the application of a recombinant PXG from
tomato (Solanum lycopersicum) for the biotechnological oxidation
of natural lipids like fatty acids and terpenes. A PXG gene was cloned
from S. lycopersicum and the recombinant protein was expressed
in Saccharomyces cerevisiae to provide a reliable enzyme source.
A qualitative and quantitative substrate screening was performed
and products were identified by GC–MS, LC–MS and NMR. SlPXG
converted a number of lipid compounds and catalyzed different
kinds of oxidation reactions.
with microsomal proteins from empty vector yeast cells, with-
out proteins and without H2O2, respectively. The products were
extracted twice with each 500 ␮l CH2Cl2, the organic layer was
separated and dried under a stream of nitrogen. For methylation
of free carboxyl groups, the pellet was solved in 300 ␮l MeOH,
mixed with 150 ml trimethylsilyldiazomethane (2 mol/l) and incu-
bated for 60 min at room temperature. The samples were dried by
Speedvac, re-dissolved in 200 ml n-hexane and analyzed by Trace
GC Ultra gas chromatograph connected to a Trace DSQ mass spec-
trometer (2.12). Samples containing oleyl alcohol 3a were solved
in 60 ␮l MeOH (30%) and analyzed by LC–MS (2.13). The pH opti-
mum was determined by varying the pH values of the reaction in
steps of 1 between pH 4 and 6 in sodium acetate buffer (10 mmol/l,
2% glycerol) and between 7 and 9 in Tris buffer (10 mmol/l, 2%
glycerol), whereas 1a and H2O2 served as substrates. The tempera-
ture optimum was determined by varying the reaction temperature
in steps of 10 ◦C between 0 and 80 ◦C in sodium acetate buffer
(10 mmol/l, pH 6, 2% glycerol). For determination of saturating
curves the substrate concentrations were varied between 0.02 and
2 mmol/l in sodium acetate buffer (10 mmol/l, pH 6, 2% glycerol)
with 50 ␮g microsomal protein containing SlPXG. The values of sat-
urating curves were used for apparent Km value calculation with
excel solver (Microsoft).
2. Experimental
2.6. Screening of other natural products as substrates for SlPXG
2.1. Chemicals
Microsomal protein containing SlPXG (100 ␮g) was incu-
bated with 1 mmol/l substrate (0.5 ␮mol in 10 ␮l methanol) and
2.5 mmol/l H2O2 (1.25 ␮mol) in 500 ␮l sodium acetate buffer
(10 mmol/l, pH 6, 2% glycerol) for 40 min at 40 ◦C. Three controls
were performed with proteins from empty vector yeast cells, with-
out proteins and without H2O2, respectively. The products were
extracted and analyzed by GC–MS as described for fatty acids,
except that the methylation reaction was omitted. The dried pellet
was solved in 100 ␮l MeOH.
All chemicals were used as received from Fluka, Merck,
Sigma–Aldrich and Roth.
2.2. Cloning and expression of recombinant PXG from tomato
The cloning and transformation of the SlPXG gene and cultiva-
tion of Saccharomyces cerevisiae INVSc1 cells (Invitrogen) for the
heterologous protein expression were performed as described [20].
2.3. Protein crude extract of yeast
2.7. Activity assay for the oxidation of fatty acids
The cell pellet from 100 ml culture was washed with 10 ml
breaking buffer (50 mmol/l sodium phosphate buffer, pH 7.4,
1 mmol/l EDTA, 5% glycerol, 1 mmol/l phenylmethylsulphonylfluo-
ride, PMSF). After centrifugation (1500 × g and 4 ◦C) for 5 min the
pellet was re-suspended in 4 ml breaking buffer containing 0.01%
Tween® 20. An equal volume of glass beads (0.25–0.5 mm, Roth,
Karlsruhe, Germany) was added and the suspension was vortexed
for 30 s, followed by 30 s on ice, which was repeated eight times.
The suspension was centrifuged at 2000 × g and 4 ◦C for 5 min. The
pellet was washed twice with 4 ml breaking buffer containing 0.01%
Tween® 20 and the supernatants were combined.
Microsomal protein containing SlPXG (50 ␮g) was incubated
with 0.4 mmol/l fatty acids (0.1 ␮mol in 10 ␮l methanol) and
2 mmol/l H2O2 (0.5 ␮mol) in 0.25 ml sodium acetate buffer
(10 mmol/l, pH 6, 2% glycerol, 0.01% Tween® 20) for 20 min at 40 ◦C.
A 0.1 ml aliquot of this mixture was added to 1.9 ml NADH solu-
tion (final concentration 125 ␮mol/l; in 100 mmol/l Tris-HCl, pH 8;
Sulfite-kit, R-Biopharm AG, Darmstadt, Germany). The OD340 was
measured and 5 ␮l of peroxidase solution (Sulfite-kit) was added.
After 1 h at room temperature the OD340 was measured again. The
amount of H2O2 was calculated from OD340 values as described
in the manual of the kit. The activity of SlPXG was determined by
comparison of the H2O2 levels at the start and end of the SlPXG reac-
tion. Background activity was measured using methanol instead
any substrate and substracted from the enzyme activity.
2.4. Isolation of yeast microsomes
The isolation of the yeast microsomes was done as described
[21] except for the re-suspension buffer which consisted of
100 mmol/l Tris, pH 7.5 and 1 mmol/l PMSF. The different protein
fractions were subjected to SDS-PAGE followed by Coomassie stain-
ing. Enzyme assays were performed with oleic acid 1a and H2O2
to determine the fractions showing the highest activity. Protein
concentration was determined [22].
2.8. Activity assay for the oxidation of terpenes and other natural
products
Microsomal protein containing SlPXG (50 ␮g) were incubated
with 0.1 mmol/l substrate (0.05 ␮mol in 10 ␮l methanol) and
2.5 mmol/l H2O2 (1.25 ␮mol) in 500 ␮l sodium acetate buffer
(10 mmol/l, pH 6, 2% glycerol) for 40 min at 40 ◦C. Products and
remaining substrates were extracted by head space SPME (65 ␮m
polydimethylsiloxane/divinylbenzene, fused silica, SUPELCO Ana-
lytical, Bellefonte, USA) for 20 min. Quantification was performed
by Trace GC Ultra gas chromatograph connected to a Trace DSQ
mass spectrometer (2.12). The conversion of substrate was cal-
culated by a calibration curve created with different substrate
2.5. Screening of fatty acids as substrates for SlPXG
Microsomal protein containing SlPXG (50 ␮g) was incubated
with 2 mmol/l fatty acid (1 ␮mol in 10 ␮l methanol) and 2.5 mmol/l
H2O2 (1.25 ␮mol) in 500 ␮l sodium acetate buffer (10 mmol/l, pH
6, 2% glycerol) for 20 min at 40 ◦C. Three controls were performed

54
C. Fuchs, W. Schwab / Journal of Molecular Catalysis B: Enzymatic 96 (2013) 52–60
concentrations (0.01, 0.05, 0.075 and 0.1 mmol/l) under the same
conditions. The activity of the SlPXG was quantified by the decrease
of the substrate level corrected with an empty vector control.
with a flow rate of 1.1 ml/min. The temperature was maintained
at 100 ◦C for 5 min, then raised to 280 ◦C at a rate of 5 ◦C/min,
and finally held at 280 ◦C for 10 min. The mass spectrometer
parameters were as follows: electron energy 70 eV; emission
current 100 ␮A; ion source temperature 250 ◦C. The mass spec-
tra were measured in the mass range from m/z 50 to 600.
Limonene oxide enantiomers were resolved with a gaschromato-
graph 450-GC (Agilent Technologies, Santa Clara, USA) connected
to a FID. A chiral capillary column (octakis-(2,3-di-O-butyryl-6-O-
tert-butyldimethylsilyl)-␥-cyclodextrin solved in SE 52, 30 m) was
used. The carrier gas was hydrogen with a flow rate of 1 ml/min.
The column temperature was maintained at 40 ◦C for 5 min, then
raised to 210 ◦C at a rate of 1.5 ◦C/min, and finally held at 210 ◦C
for 5 min. The temperature of the FID was 280 ◦C; the combus-
tion gas consisted of synthetic air (300 ml/min) and hydrogen
(30 ml/min). The molecular weights of unknown terpene oxides
were determined with a Hewlett Packard GC HP 5890 Series II (GMI,
Ramsey, USA) connected with sector field mass spectrometer MAT
95S (Finnigan MAT, Bremen, Germany) for chemical ionization. A
DB-5 capillary column (5% diphenyl/95% dimethylsiloxane; 30 m;
film thickness 0.25 ␮m Agilent Technologies, Santa Clara, USA) was
used. The carrier gas was helium with a flow rate of 1 ml/min. The
column temperature was maintained at 35 ◦C for 2 min, then raised
to 185 ◦C at a rate of 5 ◦C/min, then raised to 250 ◦C at a rate of
15 ◦C/min and finally held at 250 ◦C for 10 min. The mass spectrom-
eter parameters were as follows: electron energy 70 eV; ion source
temperature 200 ◦C. Reactant gas was isobutane.
2.9. Identification of unknown oxidation products by NMR
Crude protein extract containing SlPXG (10 mg) was mixed with
substrates (13.6 mg ␣-phellandrene 10a, 13.6 mg ␣-terpinene 11a,
or 50 mg ␣-linolenic acid 6a) in 100 ml sodium acetate buffer
(10 mmol/l, pH 6). The reaction was started by adding 25.5 ␮l H2O2
(9.79 mol/l) and stirred at 40 ◦C for 5 h. The reaction was monitored
by taking samples of 0.5 ml every 30 min, which were extracted
with CH2Cl2 and measured by GC–MS as described above. Addi-
tional amounts of 25.5 ␮l H2O2 (9.79 mol/l) were added every 0.5 h,
and 13.6 mg 10a was added after 1.5 and 3 h. The products of ␣-
terpinene 11a were extracted three times with 50 ml diethyl ether,
whereas the products of ␣-phellandrene 10a and ␣-linolenic acid
6a were extracted with 50 ml CH2Cl2. The organic phases were
dried with sodium sulphate and slowly concentrated with the help
of a Vigreux column at 35 ◦C. The products of 6a were methyl-
ated by trimethylsilyldiazomethane (2.5) and the terpene products
were directly subjected to silica gel chromatography to purify the
oxidation products. A mixture of n-hexane and ethyl acetate (3:2
for terpene products and 8.5:1.5 for products of 6a) was used.
Fractions were monitored by GC–MS. The fractions containing the
purified oxidation product were combined, slowly dried at room
temperature and the residue solved in CDCl3 with 0.03% tetram-
ethylsilane (TMS). 1H and 13C NMR analysis was performed using
a Bruker DMX-400 spectrometer (Bruker, Rheinstetten, Germany).
The spectrometer frequencies were 500 MHz and 125 MHz for the
determination of chemical shifts of 1H and 13C nuclei, respec-
tively. The chemical shifts were determined using TMS as internal
standard in the proton dimension and the carbon signal of CDCl3
in the carbon dimension. The assessment of NMR spectral data was
done using the software MestReNova (www.mestrelab.com).
2.13. LC–MS
An Agilent 1100 HPLC (Agilent Technologies, Waldbronn,
Germany) equipped with a quaternary pump and a variable
wavelength detector, and connected to
a Bruker Daltonics
Esquire3000plus ion trap mass spectrometer (Bruker Daltonics,
Bremen,Germany) was utilized. A C-18 column (150 mm × 2 mm,
particle size 5 ␮m, Phenomenex, Newport Beach, USA) held at
25 ◦C was used. The mobile phase was a mixture of water (A) and
methanol (B) both containing 0.1% formic acid. The flow rate was
0.2 ml/min. After 5 min at 30% B the gradient went from 30% to
100% B in 45 min. Finally, the gradient decreased to 30% B in 10 min.
The detection wavelength was 234 nm. The electrospray ionization
voltage of the mass spectrometer ranged from 4000 V to −500 V.
Nitrogen was used as dry gas at a temperature of 330 ◦C and a flow
rate of 9 l/min. The full scan mass spectra were measured in the
range from m/z 30 to 800 with a scan resolution of 13,000 m/z/s.
The collision gas for the mass spectrometry was helium with a
collision voltage of 1 V. Mass spectra were acquired in both pos-
itive and negative ionization modes. Analyses of LC–MS.data were
performed using the Bruker Daltonics software.
2.10. Chemical oxidation
The oxidation of substrates with m-chloroperoxybenzoic
(mCPBA) acid was performed as described [23].
2.11. Oxidation with different hydroperoxides as oxidant
The oxidation of 1a, S-(−)-limonene 9a, 11a and ␥-terpinene 12a
was done as described (2.5; 2.6), but instead of H2O2 0.75 mmol/l
tert butyl hydroperoxide or 0.5 mmol/l cumene hydroperoxide was
used as oxidant. Incubation period, extraction and measurement
procedure were identical. For saturating curves, 50 ␮g microsomal
protein containing SlPXG was mixed with 0.5 mmol/l 1a (0.25 ␮mol
in 10 ␮l methanol) in 500 ␮l sodium acetate buffer (10 mmol/l, pH
6, 2% glycerol). The concentration of H2O2, tert butyl hydroperox-
ide and cumene hydroperoxide varied from 0.01 to 5 mmol/l, 0.1
to 5 mmol/l and 0.005 to 2 mmol/l, respectively. The mixture was
incubated for 40 min at 40 ◦C and treated as described above for
fatty acids. The amounts of the oxidation products were quantified
by GC–MS and the values used for apparent Km value calculation
with excel solver.
2.14. Mass spectra and NMR data
9,10-Epoxy octadecanoic acid methyl ester 2b. RT 14.9 min. MS
(EI): m/z 312 (M+, 1%), 155 (74), 97 (45), 83 (46), 74 (88), 69 (76),
55 (100).
9,10-Epoxy octadecanol 3b. RT 43.1 min. MS (APCI): m/z 307
(M+Na+), MS/MS 307 (100%), 305 (39).
9,10-Epoxy-12-octadecenoic acid 4b (methylated). RT 11.6 min.
MS (EI): m/z 310 (M+, 1%), 155 (18), 109 (24), 95 (50), 83 (44), 79
(37), 69 (59), 67 (78), 55 (100).
2.12. GC–MS
Fatty acids and terpenes were analyzed with a Finnigan Trace
GC Ultra gas chromatograph (Thermoelectric Corporation, Dreieich,
Germany) connected with a FinniganTrace DSQ (Thermoelec-
tric Corporation). A VF-5MS capillary column (5% diphenyl/95%
dimethylsiloxane; 30 m; film thickness 0.25 ␮m; Agilent Tech-
nologies, Santa Clara, USA) was used. The carrier gas was helium
12,13-Epoxy-9-octadecenoic acid 4c (methylated). RT 11.5 min.
MS (EI): m/z 310 (M+, 1%), 164 (15), 136 (19), 121 (24), 95 (58), 81
(90), 79 (45), 67 (84), 55 (100).
12,13-Epoxy-10-octadecenoic acid 5b (methylated). RT
13.4 min. MS (EI): m/z 310 (M+, 1%), 225 (24), 207 (19), 199
(36), 149 (27), 139 (41), 121 (27), 99 (100), 81 (58), 55 (98).

C. Fuchs, W. Schwab / Journal of Molecular Catalysis B: Enzymatic 96 (2013) 52–60
55
15,16-Epoxy-9,12-octadecadienoic acid 6b (methylated). RT
11.7 min. MS (EI): m/z 308 (M+, 1%), 108 (32), 107 (12), 93 (42), 81
(22), 79 (100), 67 (45), 57 (37), 55 (52); NMR: ıH (500 MHz; CDCl3;
Me4Si) 1.07 (t, J = 7,5, 3H), 1.25–1.41 (m, 8H), 1.5–1.66 (m, 2H), 2.06
(q, J = 6.6, 7.1, 2H), 2.2–2.28 (m, 2H), 2.32 (t, J = 7,5, 2H), 2.40–2.45
(m, 2H), 2.82 (dd, J = 6.9, 7.0, 2H), 2.87–3.01 (m, 2H), 3.68 (s, 3H),
5.31–5.55 (m, 4H); ıC (125 MHz; CDCl3; Me4Si) 10.6 (C18), 21.1
(C17), 26.1 (C3), 26.2 (C14), 27.2 (C11), 27.2 (C8), 29 (C4), 29.1 (C7),
29.2 (C5), 29.6 (C6), 34.1 (C2), 51.4 (C19), 56.5 (C16), 58.3 (C15),
124.2 (C12), 127.3 (C10), 130.5 (C9), 130.8 (C13), 174.2 (C1).
11,12-Epoxy eicosanoic acid 7b (methylated). RT 16.5 min. MS
(EI): m/z 340 (M+ 0.2%), 199 (10), 183 (20), 167 (18), 155 (29), 149
(41), 87 (52), 81 (54), 74 (76), 69 (72), 55 (100).
analysis showed a protein band at 28 kDa representing SlPXG
(calculated molecular weight: 27.9 kDa) (supplementary online
material Figure S1). Enzyme activity assays were performed on dif-
ferent fractions obtained during the preparation of the microsomes.
Considerable enzymatic activity was detected in the microsomal
fraction which was 1.5 times higher than in the crude protein
extract. Furthermore, the activity could be increased up to 2.3 times
by adding the detergent Tween® 20 probably due to the extraction
of the protein from the membrane.
PXG genes from different plant sources have already been
expressed in Escherichia coli, Pichia pastoris and S. cerevisiae to ana-
lyze their biological functions [11,20,24] and a PXG enzyme has
been isolated from oat seed [19] to perform a substrate screening.
Similar to the PXG protein sequences from Arabidopsis thaliana,
Sesamum indicum, A. sativa and Oryza sativa SlPXG contains a trans-
membrane domain which is localized between arginine 91 and
leucine 111 (supplementary online material Figure S2) and explains
the membrane integration and the high PXG enzyme activity in the
microsomal fraction. Like other PXGs, SlPXG has a heme molecule
in the active site and contains conserved histidine residues 69 and
133 which are believed to be involved in heme binding [24] (sup-
plementary online material Figure S2). The heme iron is involved in
the oxygen transfer from a hydroperoxide to the substrate. Exper-
iments with the monoxygenase inhibitor tetcyclacis proved the
important role of the heme in the catalysis of SlPXG [20].
R-(+)-limonene-trans-1,2-epoxide 8b. MS (EI): m/z 152 (M+, 1%),
137 (5), 108 (44), 94 (68), 79 (47), 67 (61), 43 (100).
S-(−)-limonene-cis-1,2-epoxide 9b. MS (EI): m/z 152 (M+, 3%),
137 (30), 119 (15), 109 (40), 93 (42), 81 (34), 79 (41), 67 (77), 43
(100).
1,2-Epoxy-p-menth-5-ene 10b. MS (EI): m/z 152 (M+, 3%), 126
(65), 111 (85), 109 (37), 95 (47), 93 (28), 71 (66), 43 (100); NMR:
ıH (500 MHz; CDCl3; Me4Si) 0.85–0.95 (2d, J = 6.8, 6H), 1.34 (s, 3H),
1.62–1.68 (m, 1H), 1.75–1.82 (m, 2H), 2.05–2.1 (m, 1H), 3.45–3.50
(m, 1H), 5.65–5.72 (m, 2H); ıC (125 MHz; CDCl3; Me4Si) 19.00 (C9,
C10), 25.91 (C3), 30.17 (C7), 31.49 (C8), 42.58 (C4), 69.13 (C1), 73.65
(C2), 132.09 (C6), 134.04 (C5).
5,6-Epoxy-p-menth-1-ene 10c. MS (EI): m/z 152 (M+, 3%), 127
(17), 111 (40), 100 (95), 85 (55), 84 (41), 71 (100), 69 (70); NMR: ıH
(500 MHz; CDCl3; Me4Si) 0.86–1.00 (2d, J = 7.0, 6H), 1.55–1.65 (m,
1H), 1.80 (s, 3H), 2.0–2.05 (m, 1H), 2.12–2.16 (m, 2H), 3.46–3.51 (m,
1H), 3.74 (q, J = 7.0, 1H), 5.46–5.48 (m, 1H); ıC (125 MHz; CDCl3;
Me4Si) 16.61 (C9, C10), 18.49 (C7), 26.39 (C3), 31.77 (C8), 48.10
(C4), 69.37 (C5), 71.05 (C6), 128.75 (C2), 139.13 (C1).
3.2. Epoxidation of fatty acids and fatty acids derivatives
PXGs are enzymes that oxidize unsaturated fatty acids to cis-
configurated epoxides [25,26]. Similarly, recombinant SlPXG was
able to epoxidize unsaturated fatty acids and their derivatives at
their double bonds (Table 1). The pH optimum was determined to
be 6 and the temperature optimum 40 ◦C when using 1a as sub-
strate. The identity of the products was proven by comparison of
the mass spectra with those of reference substances, literature data
and chemically (by mCPBA) synthesized epoxides. One of the oxida-
tion products of ␣-linolenic acid 6a was identified by NMR analysis
(Table 1).
Cis-1,4-Dihydroxydihydroxy-p-menth-2-ene 11b. MS (EI): m/z
168 (M+, 0.5%), 141 (35), 127 (37), 123 (40), 109 (100), 81 (48),
43 (47); NMR: ıH (500 MHz; CDCl3; Me4Si) 0.92–1.00 (2d, J = 6.9
6H), 1.36 (s, 3H), 1.56–1.62 (m, 2H), 1.74–1.80 (m, 2H), 1.85–1.95
(m, 1H), 5.65 (dd, J = 1.5, 10.0, 1H), 5.75 (dd, J = 1.5, 10.0, 1H); ıC
(125 MHz; CDCl3; Me4Si) 16.37 (C10), 17.55 (C9), 27.09 (C5), 29.69
(C7), 33.50 (C6), 37.44 (C8), 67.25 (C1), 71.58 (C4), 133.46 (C3),
135.46 (C2).
1a and 2a were oxidized by SlPXG to 9,10-epoxy octadecanoic
acid 1b and 9,10-epoxy octadecanoic acid methyl ester 2b, respec-
tively, as already described for PXGs from Glycine max [10] and
Avena sativa [18,19,24,26]. In contrast to the enzymatic oxida-
tion of oleic acid, where only the cis-9,10-epoxy stearic acid was
formed, the chemical oxidation by mCPBA resulted in cis-9,10-
epoxy octadecanoic acid (76%) and trans-9,10-epoxy octadecanoic
acid (24%) (supplementary online material Figure S5). The chem-
ical oxidation of methyl oleate yielded a similar distribution of
products. The double bond partly isomerized from cis to trans
configuration during the oxidation by mCPBA. Thus, enzymatic
epoxidation by SlPXG resulted in purer products. Although the oleyl
alcohol 3a turned out to be a poor substrate and exact quantifica-
tion of the product yield was not feasible, the product 9,10-epoxy
octadecanol 3b could be clearly detected. Conjugated trans-10,
cis-12-linoleic acid 5a was only oxidized at the cis-double bond
to yield 12,13-epoxy-10-octadecenoic acid 5b, demonstrating the
preference of SlPXG for cis-double bonds [13,24]. The chemical
oxidation of 5a by mCPBA led to the formation of 5b and a sec-
ond product, which was putatively identified as the 10,11-epoxide.
SlPXG oxidized linoleic acid 4a to the monoepoxides 9,10-epoxy-
12-octadecenoic acid 4b and 12,13-epoxy-9-octadecenoic acid 4c
in a ratio of 1: 1. The diepoxide was not formed. In contrast, PXG
from G. max favoured the attack at the 9,10-double bond (72%)
[13] whereas PXGs from A. sativa and A. thaliana produced both
monoepoxides and the diepoxide [19,27]. Different regioselectivity
of different PXGs was also observed in the case of ␣-linolenic acid
6a. This fatty acid was epoxidized by SlPXG to one main and two
p-Cymene 12b. MS (EI): m/z 134 (M+, 31%), 120 (11), 119 (100),
117 (18), 115 (9), 91 (34), 77 (9), 65 (9).
Nerol-2,3-epoxide 15b. MS (EI): m/z 170 (M+, 0.5%), 121 (17),
109 (72), 95 (26), 93 (31), 82 (31), 69 (100), 67 (56), 41 (85).
3,4-Dihydroxy bisabolol 16b. MS (EI): m/z 256 (M+, 0.5%), 220
(8), 151 (11), 127 (6), 109 (100), 93 (14), 82 (50), 69 (56), 43 (67).
Bisabololoxid B 16c. MS (EI): m/z 238 (M+, 1%), 179 (14), 161
(36), 143 (100), 125 (46), 107 (26), 105 (57), 85 (52), 81 (43), 71
(44).
Cis-jasmone-7,8-epoxide 17b. MS (EI): m/z 180 (M+, 3%), 165
(15), 152 (7), 123 (17), 122 (35), 110 (43), 109 (20), 95 (22), 79
(100), 67 (33), 43 (25), 41 (32).
Cis-stilbene-oxide 18b. MS (EI): m/z 196 (M+, 31%), 195 (41), 178
(22), 167 (100), 165 (35), 152 (24), 105 (42), 90 (52), 89 (69), 77 (28).
3. Results and discussion
3.1. Heterologous expression of SlPXG
A PXG gene was cloned from tomato (S. lycopersicum) and het-
erologously expressed in yeast (S. cerevisiae) [20]. S. cerevisiae was
chosen as expression host because PXG is a membrane protein of
the endoplasmic reticulum (ER) and can be easily isolated with the
yeast microsomes, the relict of the ER after cell lysis [11,20,24].
After protein expression and isolation of microsomes, SDS-PAGE

Table 1
SlPXG catalyzed transformation of fatty acids and fatty acids derivatives.a
Entry
1
Substrate
Product
Product ratio (%)
Product yield (%)b
Activity (nmol/min)b
Km value (mmol/l)c
Identification
100
100
100
15
13
<1
0.74
0.66
<0.1
0.071
0.167
0.05
Reference
Reference
Synthesis
2
3
4
50
50
9.5
9.5
0.46
0.46
0.183
0.163
[38]
[38]
5
6
100
68
23
24
25
1.14
1.18
1.24
0.815
0.21
Synthesis
NMR
7
100
0.078
Synthesis
a
Reaction conditions are detailed in Section 2
Determined by consumption of H2O2
Apparent Km values determined by GC–MS except for 3a which was determined by LC–MS.
b
c

C. Fuchs, W. Schwab / Journal of Molecular Catalysis B: Enzymatic 96 (2013) 52–60
57
side products (68%, 17% and 15%). The main product was purified by
silica gel chromatography and identified by NMR as 15,16-epoxy-
9,12-octadecadienoic acid 6b (see 2.14 for NMR data). However,
PXG from A. sativa [18] and A. thaliana [23] mainly formed the 9,10-
, 15,16-diepoxide (92%). This indicates that different isoforms of
PXGs enable the synthesis of defined epoxides from unsaturated
fatty acids. Absolute stereochemistry of epoxy products formed by
SlPXG was not determined but PXG4 from A. thaliana [27] catalyzed
exclusively the formation of (R),(S)-epoxide enantiomers, which is
the absolute stereochemistry of the epoxides found in planta.
The conversion of fatty acids and the activity of SlPXG were
quantified using a coupled enzyme assay. Unsaturated fatty acids
were oxidized by SlPXG in the presence of H2O2 and the remaining
hydroperoxide was reduced by NADH and NADH-peroxidase. The
NADH quantity was determined photometrically at 340 nm and
used for the calculation of consumed H2O2 (Table 1). While SlPXG
epoxidized 7a and 6a with highest (1.24 and 1.18 nmol/min, respec-
tively) and 1a and 2a with lowest activity (0.74 and 0.66 nmol/min,
respectively), the substrate preference of PXG from A. sativa was
in reverse order; highest activity was determined with 1a and 2a,
while 7a and 6a were converted with 75 and 50% relative activity,
respectively [24].
Fig. 1. Hydroxylation of ␣-terpinene 11a by SlPXG via the hypothetical intermediate
ascaridole.
the hydroxylation of 11a. The p-menthenes ␥-terpinene 12a and
␣-thujene 14a were transformed to p-cymene (12b, 14b) with 71
and 26% yield, respectively (Tables 2 and 3). SlPXG efficiently cat-
alyzed the dehydrogenation of the cyclic monoterpenoid skeleton
of 12a and 14a to the aromatic p-cymene 12b rather than an epox-
idation or hydroxylation reaction. This is the first report showing
that PXGs catalyze an aromatization reaction. The aromatization
of 12a to 12b was first discovered by incubating 12a together
with leaves of thyme (Thymus vulgaris) [33] and is probably cat-
alyzed by cytochrome P450 enzymes in thyme. The p-menthenes,
which possess a 1,4-cyclohexadiene structure, are well known
educts for aromatization reactions. It has been shown that 12a,
1,4-cyclohexadiene and 1-methyl-1,4-cyclohexadiene react easily
with KMnO4 at 0 ◦C to 12b, benzene and toluene, respectively [34].
In the oxidation mechanism of 1,4-cyclohexadiene to benzene a
cyclohexadienyl cation was proposed to be a transition state [35].
Likewise, in the aromatization reaction catalyzed by SlPXG H2O2
seems to act as hydrogen acceptor instead of an oxygen donor, thus
enabling the dehydrogenation of the cyclohexadiene ring. Inter-
action of the heme iron with the single-oxygen donor H2O2 can
lead to the formation of an iron-oxo intermediate which eventually
gets reduced by hydrogen provided by the substrate (Fig. 2). Alter-
natively, it is conceivable that 12a is hydroxylated and then may
aromatize via water elimination. Compound 12b is an important
solvent and fragrance. Furthermore it can be a platform chemical
for other aromatic products [36].
3.3. Oxidation of terpenes and volatile natural products
In addition to unsaturated fatty acids also numerous terpenes
contain cis double bonds and could function as potential substrates
for PXGs. Therefore, the enzymatic oxidation of certain terpenes by
SlPXG was investigated (Tables 2 and 3).
SlPXG stereoselectively epoxidized R-(+)-limonene 8a and S-
(−)-limonene 9a to R-(+)-limonene-trans-1,2-epoxide 8b (97%
diastereomeric purity) and S-(−)-limonene-cis-1,2-epoxide 9b
(88% diastereomeric purity), respectively. The absolute configura-
tion of the products were determined with pure R-(+)-limonene-
trans-1,2-epoxide 8b and chemically enriched S-(−)-limonene-
trans-1,2-epoxide by GC-FID using
a chiral capillary column
(supplementary online material Figures S3 and S4). Although the
conversion of 8a by the SlPXG is lower (39%) than by heavy metal
containing catalysts (75% [28] and 97% [29]) the diastereomeric
excess of the SlPXG product is higher (97%) than that in the chemical
reaction (max. 89% [28] and 93% [29]). Limonene-1,2-epoxides can
be used as fragrances [1,2] or for the synthesis of polycarbonates
[30].
Besides limonene, additional terpenes containing a p-menthene
skeleton were tested as substrates for SlPXG. R,S-␣-phellandrene
10a was epoxidized by SlPXG to at least five products. The molec-
ular mass of every product was determined to be 152 Da by
GC-(CI)–MS. Two pairs of products showed the same mass spectra
but different retention times which let us suppose that these were
diastereomeric forms. Diastereomeric pairs were purified by silica
gel chromatography and the NMR spectra of the different products
showed (2.14) that SlPXG catalyzed the formation of 1,2-epoxy-p-
menth-5-ene 10b (ratio of diastereomers: 26: 13% of total products)
and 5,6-epoxy-p menth-1-ene 10c (26: 17%). The fifth product (18%
of total products) was not identified but the molecular weight of
152 Da implies that an epoxide or ketone was produced.
Terpinolene 13a seemed to be converted to minor amounts of
an epoxide or ketone as was deduced from the mass spectrum of
the product.
The fact that SlPXG did not oxidize geraniol but nerol 15a is in
accordance to the preference of the enzyme for cis double bonds
[13]. Nerol-2,3-epoxide 15b was the only product formed by SlPXG
from 15a (Table 3).
Hydroxylation reactions catalyzed by plant PXGs are known
for indole, phenol and aniline [9]. This kind of oxidation was also
noticed for at least two terpenes (11a and 16a). SlPXG oxidized
␣-terpinene 11a to only one product, which was identified as cis-
1,4-dihydroxy-p-menth-2-ene 11b by NMR and mass spectrometry
(2.14). As the endoperoxide ascaridole is known to be rapidly con-
verted into 11b [31], ascaridole is postulated to be an intermediate
of the hydroxylation of 11a (Fig. 1). The transformation of 11a to
ascaridole has already been demonstrated by an iodide-peroxidase
[32] and supports the involvement of ascaridole as intermediate in
Fig. 2. Postulated reaction mechanism for the aromatization of ␥-terpinene 12a
to p-cymene 12b by SlPXG. Only the heme residue of the enzyme is shown for
simplicity.

58
C. Fuchs, W. Schwab / Journal of Molecular Catalysis B: Enzymatic 96 (2013) 52–60
Table 2
SlPXG catalyzed transformation of terpenes and other natural products.a
Entry
Substrate
Product
Product ratio
(%)
Product yield
(%)b
Activity
Identification
Reference
(nmol/min)b
8
8a
8b
97
88
39
43
38
31
20
22
0.48
0.39
0.25
0.28
9
9a
9b
Reference
10
10a
10b
NMR
10c
NMR
11
11a
12a
13a
11b
100
100
100
47
71
7
0.59
0.88
0.09
NMR
12
12b
Reference
13
Probably epoxide or ketone
a
Reaction conditions are detailed in Section 2.
Determined by consumption of substrate.
b
The sesquiterpene ␣-bisabolol 16a was efficiently oxidized by
16b whereas the 7-hydroxyl group of the 10,11-epoxide may
attack nucleophilically the epoxide at position C10, which results
in a 7,10-epoxide and the 11-hydroxyl group of 16c. In contrast,
biotransformation of 16a by a culture of the fungus Glomerella
cingulate resulted in the production of numerous products [37].
SlPXG (37% residual 16a). The products are 3,4-dihydroxy-bisabolol
16b (40%), bisabololoxide B 16c (35%) and one unidentified com-
pound (25%). The 3,4- and 10,11-epoxide are assumed to be
reaction intermediates [37]. Hydrolysis of the 3,4-epoxide yields

C. Fuchs, W. Schwab / Journal of Molecular Catalysis B: Enzymatic 96 (2013) 52–60
59
Table 3
SlPXG catalyzed transformation of terpenes and other natural productsa
Entry
Substrate
Product
Product ratio
(%)
Product yield
(%)b
Activity
Identification
(nmol/min)b
14
14a
15a
16a
14b
15b
16b
50
26
16
25
0.33
0.20
0.32
Reference
15
16
100
[28]
40
[37]
16c
35
22
0.28
[39]
17
17a
18a
17b
18b
100
100
11
10
0.14
0.12
[40]
18
Reference
a
Reaction conditions are detailed in Section 2.
Determined by consumption of substrate.
b
Compound 16c is a pharmaceutical active ingredient known from
German chamomile (Matricaria chamimilla). Thus, the synthesis
of bisabolol oxides by the SlPXG can open the way for new
drugs.
Furthermore, the oxidation of cis-jasmone 17a resulted in
cis-jasmone-7,8-epoxide 17b, whereas jasmonic acid and methyl
jasmonate were not oxidized by the SlPXG. Cis-stilbene 18a was
transformed to cis-stilbene oxide 18b.
The p-menthenes ␣-pinene, 3-carene and S-(+)-carvone were
not accepted as substrates by SlPXG despite their cis double
bonds. Similarly, ␣-ionone, ␤-caryophyllene and two aromatic
compounds eugenol and isoeugenol were not oxidized by SlPXG.
The calculated SlPXG reactivity ranged from 0.09 to
0.88 nmol/min for the terpene substrates 13a and 12a, respectively.
Although most of the tested p-menthenes and other natural prod-
ucts were epoxidized, it seems that the hydroxylation (entries 11
and 16) and aromatization reactions (entries 12 and 14) proceeded
the fastest.
3.4. Comparison of different hydroperoxides
In addition to H2O2, cumene hydroperoxide and tert butyl
hydroperoxide have been described as co-substrates for plant PXGs
[9,13,18]. Similarly SlPXG catalyzed the epoxidation of 1a and
9a, the hydroxylation of 11a and the aromatization of 12a in
the presence of the three hydroperoxide co-substrates. The three
hydroperoxides produced identical products. The enzymatic reac-
tion was inhibited when the levels of H2O2, cumene and tert butyl
hydroperoxide exceeded 2.5, 0.5, and 0.75 mmol/l, respectively.
The apparent Km value for cumene hydroperoxide was 0.05 mmol/l
using 1a as substrate and corresponded to the value obtained for
a PXG from G. max [13]. The apparent Km values for H2O2 and
tert-butyl hydroperoxide were 0.6, and 0.25 mmol/l, respectively.
SlPXG showed the highest epoxidation activity of 1a with H2O2
as oxidant (relative product amount 100%), whereas cumene and
tert-butyl hydroperoxide were less active oxidants (43% and 32%,
respectively). Unlike the preference of SlPXG, a PXG from A. sativa

60
C. Fuchs, W. Schwab / Journal of Molecular Catalysis B: Enzymatic 96 (2013) 52–60
showed the highest activity when tert-butyl hydroperoxide was
used [18]. H2O2 is a suitable co-substrate for SlPXG because of its
low costs and eco friendliness.
[7] R. Ullrich, J. Nuske, K. Scheibner, J. Spantzel, M. Hofrichter, Appl. Environ. Micro-
biol. 70 (2004) 4575–4581.
[8] A. Gutiérrez, E.D. Babot, René Ullrich, M. Hofrichter, A.T. Martínez, J.C. del Río,
Arch. Biochem. Biophys. 514 (2011) 33–43.
[9] A. Ishimaru, I. Yamazaki, J. Biol. Chem. 252 (1977) 6118–6124.
[10] E. Blée, A.L. Wilcox, L.J. Marnett, F. Schuber, J. Biol. Chem. 268 (1993) 1708–1715.
[11] A. Hanano, M. Burcklen, M. Flenet, A. Ivancich, M. Louwagie, J. Garin, E. Blée, J.
Biol. Chem. 281 (2006) 33140–33151.
[12] X. Wang, S. Peter, M. Kinne, M. Hofrichter, J.T. Groves, J. Am. Chem. Soc. 134
(2012) 12897–12900.
[13] E. Blée, F. Schuber, J. Biol. Chem. 265 (1990) 12887–12894.
[14] E. Blée, J.E. Casida, F. Durst, Biochem. Pharmacol. 34 (1985) 389–390.
[15] E. Blée, F. Schuber, Plant J. 4 (1993) 113–123.
[16] J. Lequeu, M.L. Fauconnier, A. Chammaï, R. Bronner, E. Blée, Plant J. 36 (2003)
155–164.
[17] M. Partridge, D.J. Murphy, Plant Physiol. Biochem. 47 (2009) 796–806.
[18] G.J. Piazza, T.A. Foglia, A. Nun˜ez, J. Am. Oil Chem. Soc. 22 (2000) 217–221.
[19] G.J. Piazza, A. Nun˜ez, T.A. Foglia, J. Mol. Catal. B: Enzym. 21 (2003)
143–151.
[20] J. Aghofack-Nguemezi, C. Fuchs, S.Y. Yeh, F.C. Huang, T. Hoffmann, W. Schwab,
J. Exp. Bot. 62 (2011) 1313–1323.
4. Conclusion
A recombinant PXG from S. lycopersicum (tomato) is able to
oxidize a number of unsaturated cis-configurated natural com-
pounds such as fatty acids and terpenes to yield a variety of
products. The kind of oxidation reaction catalyzed by PXG depends
on the molecular structure of the substrate and includes epoxi-
dation, hydroxylation and aromatization. Thus, PXG provides new
opportunities for the biotechnological production of functionalized
natural products. The results can find application in whole-cell-
production systems and transgenic plants.
[21] D. Eberle, P. Ullmann, D. Werck-Reichhart, M. Petersen, Plant Mol. Biol. 69
(2009) 239–253.
Acknowledgements
[22] M.M. Bradford, Anal. Biochem. 72 (1976) 248–254.
[23] P. Weyerstahl, H. Marschall-Weyerstahl, S. Scholz, Liebigs Ann. Chem. (1986)
1248–1254.
[24] D. Meesapyodsuk, X. Qiu, Plant Physiol. 157 (2011) 454–463.
[25] E. Blée, F. Schuber, Biochem. Biophys. Res. Commun. 173 (1990) 1354–1360.
[26] M. Hamberg, G. Hamberg, Plant Physiol. 110 (1996) 807–815.
[27] E. Blée, M. Flenet, B. Boachon, M.L. Fauconnier, FEBS J. 279 (2012) 3981–3995.
[28] R. Saladino, Tetrahedron 59 (2003) 7403–7408.
M. Wüst (Institute for Nutrition and Food Science, University
of Bonn) and T. Hofmann (Chair of Food Chemistry and Molecular
Sensory Science, Technische Universität München) are thanked for
chiral GC measurements and NMR analysis, respectively. We are
grateful for the financial support provided by the Federal Ministry
of Food, Agriculture and Consumer Protection, Germany.
[29] T.J. Bhattacharjee, J.A. Dines, J. Anderson, J. Catal. 225 (2004) 398–407.
[30] C.M. Byrne, S.D. Allen, E.B. Lobkovsky, G.W. Coates, J. Am. Chem. Soc. 126 (2004)
11404–11405.
Appendix A. Supplementary data
[31] G. Rücker, U. Molls, Arch. Pharm. Pharm. Med. Chem. 313 (1980) 237–243.
[32] M.A. Johnson, R. Croteau, Arch. Biochem. Biophys. 235 (1984) 254–266.
[33] A. Poulose, R. Croteau, Arch. Biochem. Biophys. 187 (1978) 307–314.
[34] C.M. McBride, W. Chrisman, C.E. Harris, B. Singaram, Tetrahedron Lett. 40 (1999)
45–48.
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.
molcatb.2013.07.001.
[35] K. Ikeshita, N. Kihara, M. Sonodaa, A. Ogawaa, Tetrahedron Lett. 48 (2007)
3025–3028.
References
[36] J.H. Clark, E.M. Fitzpatrick, D.J. Macquarrie, L.A. Pfalzgraff, J. Sherwood, Catal.
Today 190 (2012) 144–149.
[37] M. Miyazawa, H. Nankai, H. Kameoka, Phytochemistry 39 (1995) 1077–1080.
[38] T. Ozawa, M. Hayakawa, T. Takamura, S. Sugiyama, K. Suzuki, M. Iwata, F. Taki,
T. Tomita, Biochem. Biophys. Res. Commun. 137 (1986) 1071–1078.
[39] H. Schilcher, L. Novotny, K. Ubik, O. Motl, V. Herout, Arch. Pharm. 309 (1975)
189–196.
[1] W. Schwab, C. Fuchs, F.C. Huang, Eur. J. Lipid Sci. Technol. 115 (2013) 3–8.
[2] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
[3] J.O. Metzger, U. Bornscheuer, Appl. Microbiol. Biotechnol. 71 (2006) 13–22.
[4] J.L.F. Monteiro, C. Veloso, Top. Catal. 27 (2004) 169–180.
[5] U. Schörken, P. Kempers, Eur. J. Lipid Sci. Technol. 111 (2009) 627–645.
[6] E.D. Babot, J.C. Del Río, L. Kalum, A.T. Martínez, A. Gutiérrez, Biotechnol. Bioeng.
(2013), http://dx.doi.org/10.1002/bit.24904.
[40] L. Pinheiro, L.G. de Oliveira, A.J. Marsaioli, J. Mol. Catal. B: Enzym. 60 (2009)
133–137.