Enzymatic halogenation of the phenolic monoterpenes thymol and carvacrol with chloroperoxidase

Article abstract of DOI:10.1039/c3gc42269k

The conversion of the phenolic monoterpenes thymol and carvacrol into antimicrobials by (electro)-chemoenzymatic halogenation was investigated using a chloroperoxidase (CPO) catalyzed process. The CPO catalysed process enables for the first time the biotechnological production of chlorothymol, chlorocarvacrol and bromothymol as well as a dichlorothymol with high conversion rates, total turnover numbers and space time yields of up to 90%, 164000 and 4.6 mM h ^-1, respectively.

Full text of DOI:10.1039/c3gc42269k

Green Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
Enzymatic halogenation of the phenolic
monoterpenes thymol and carvacrol with
chloroperoxidase†
Cite this: Green Chem., 2014, 16,
104
1
Received 4th November 2013,
Accepted 28th November 2013
a
a
b
a
a
Laura Getrey, Thomas Krieg, Frank Hollmann, Jens Schrader and Dirk Holtmann*
DOI: 10.1039/c3gc42269k
www.rsc.org/greenchem
5
The conversion of the phenolic monoterpenes thymol and carvacrol ions in seawater. Of particular interest are the halogenated
into antimicrobials by (electro)-chemoenzymatic halogenation was monoterpenes, a diverse class of organohalogen marine
investigated using a chloroperoxidase (CPO) catalyzed process. natural products found in only three genera of red macroalgae,
6
The CPO catalysed process enables for the first time the biotech- Plocamium, Portieria, and Ochtodes. Halogenated mono-
nological production of chlorothymol, chlorocarvacrol and bromo- terpenes are pharmacologically active, and some acyclic struc-
7,8
thymol as well as a dichlorothymol with high conversion rates, tures exhibit potent and selective anti-tumor activity.
total turnover numbers and space time yields of up to 90%,
The essential oil of common thyme (Thymus vulgaris) con-
tains 20–54% thymol. Thymol, an antiseptic, is the main
64 000 and 4.6 mM h− , respectively.
1
9
1
active ingredient of various mouthwashes and has also been
shown to be effective against various fungi that commonly
infect toenails. Chlorothymol is a chlorinated phenolic anti-
septic used as an ingredient of preparations for hand and skin
disinfection and topical treatment of fungal infections. It has
also been used in preparations for anorectal disorders, cold
symptoms, and mouth disorders. Chlorothymol has 75 times
higher bactericidal potency, while also possessing low human
toxicity, compared to thymol. A comparison of larvicidal
activity of different monoterpenoids shows that chlorothymol
was found to be the most effective against the common house
mosquito Culex pipiens followed by thymol, carvacrol and
Introduction
Terpenes are a highly diverse class of natural products from
which numerous commercial compounds are derived. Today,
besides their use as flavor and fragrance compounds, the
potential therapeutic properties of terpenes such as anti-
cancer, anti-microbial and anti-inflammatory properties make
them interesting and desirable molecules for the pharma-
ceutical, sanitary, cosmetic, agricultural and food industries.
Monoterpenes in particular represent a cheap and abundantly
available source of chiral substances that can be transformed
1
1
0
cinnamaldehyde. Thymol, chlorothymol as well as the isomer
2
1
1
into valuable bioactive compounds. Regio- or stereoselective
carvacrol are also used as cosmetic preservatives. Amongst
the various halogenating agents established, in situ formed
elementary halogens or hypohalogenites are the environmen-
functionalization of terpenes is one of the main goals of syn-
thetic organic chemistry, which is possible through radical
reactions but is not selective enough to introduce the desired
12
tally least harmful ones. In this respect, enzymatic gene-
ration of the reactive hypohalogenites may be attractive due to
the generally acknowledged mild reaction conditions and
3
functions into those compounds. Beside oxidation, halogena-
tion of the natural precursor allows the formation of more
valuable products. The controlled introduction of halogen sub-
stituents into the structure of naturally occurring mono-
1
3
highly efficient catalysts. Therefore we became interested in
1
4,15
evaluating the peroxidase from Caldriomyces fumago (CPO)
terpenes provides an entry into numerous important synthetic
as a biocatalyst for the in situ generation of hypochlorite to
act as an electrophilic chlorination agent for thymol
4
intermediates.
Halogenated secondary metabolites are
common in marine plants (i.e. seaweeds), while they are rare
(
Scheme 1).
in land plants, due to the abundance of chloride and bromide
a
DECHEMA Research Institute, Theodor-Heuss-Allee 25, 60486 Frankfurt, Germany.
Experimental
E-mail: holtmann@dechema.de; Fax: +49 69 7564 388; Tel: +49 69 7564 610
b
Delft University of Technology, Department of Biotechnology, Julianalaan 136,
The production and purification of chloroperoxidase (CPO)
from C. fumago is described elsewhere. CPO activity was
measured with the MCD assay under standard conditions
2
†
1
628 BL Delft, The Netherlands
Electronic supplementary information (ESI) available. See DOI:
0.1039/c3gc42269k
1
6
1104 | Green Chem., 2014, 16, 1104–1108
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Communication
2
on the applied current I following V = 0.036I + 1.83 (R = 0.975)
between 5 and 45 mA for the saturated GDE.
Results and discussion
In a first set of experiments we evaluated the suitability of CPO
as a thymol halogenation catalyst. For this hydrogen peroxide
was added at intervals to a mixture of thymol and CPO in
citrate buffer. As shown in Fig. 1, already under these arbitra-
rily chosen reaction conditions thymol was smoothly converted
into chlorothymol at significant rates. Overall, a yield of more
than 90% (based on hydrogen peroxide as the limiting com-
ponent) was obtained. Interestingly, no indications of poly-
Scheme 1 Hypothesised chemoenzymatic halogenation of thymol.
(
25 °C, 100 mM citric acid buffer, pH 2.75, 100 μM monochloro-
dimedone, 20 mM NaCl and 2 mM H ). CPO activity was
photometrically detected by measuring absorption at 278 nm
2 2
O
(
thymol) formation (e.g. as a result of CPO-catalysed direct oxi-
1
−1
(
εMCD = 12 200 M− cm ). The concentration of CPO was
dation of the phenolic compound followed by radical chain
−
1
photometrically determined at 400 nm (ε = 75 300 M
19
polymerisation) were found. It is worth mentioning here that
−
1 17
cm ).
in the absence of the biocatalyst (CPO), no product formation
was observed and the substrate was recovered completely.
Interestingly, there was a significant lag time for product
The halogenation of the substrates was investigated in 5 mL
glass reactors. Different concentrations and reaction con-
ditions were evaluated. The reaction was started by adding the
formation observed in all experiments. Published K
CPO to H O are in the range of 0.03–1.14 mM.
2 2
M
-values of
Therefore,
co-substrate H O . In general, the hydrogen peroxide is added
20,21
2
2
in five equal doses after 0, 10, 20, 30 and 40 min of the reac-
tion up to the final concentration. For each reaction time, an
independent experiment was set up and the reaction products
were analyzed. Reaction products were determined with a GC/
MS (GC17A, Shimadzu, Kyoto, Japan), a Valcobond VB-5
column, l = 30 m, ID = 0.25 mm (VICI, Schenkon, Switzerland),
temperature program 80 °C (5 min) then 10 °C min− to
it is possible that the hydrogen peroxide concentration after
the first dose (0.088 mM) is too low to enable full CPO activity.
More probably the lag phase may be caused by a comparably
low rate of the uncatalysed electrophilic halogenation of
thymol by hypochloride. GC analysis of the reaction product
revealed the formation of the ortho- and para-chlorination
product in an approximately 70 : 30 ratio. This product distri-
bution was expected for the chemical halogenation reaction.
Next, we further explored the reaction parameters influencing
the rate and overall yield of the chemoenzymatic thymol halo-
genation. As shown in Table 1, the reaction temperature, pH,
and [NaCl] were varied systematically.
1
2
00 °C, injector temperature 250 °C, GC/MS interface tempera-
ture 290 °C, scan range MS 35–250 m/z. 950 µL of the samples
were mixed with 50 µL of the internal standard (2 g L−
1
1
-octanol in ethanol) and subsequently extracted with 1 mL
butyl acetate. The organic phase was dried with sodium sulfate
and analyzed by GC/MS (retention times: thymol 11.9 min,
para-chlorothymol 12.7 min, ortho-chlorothymol 15.1 min).
Several experiments were performed twice, and the deviation
between the experiments was less than 5.5%.
With regard to substrate conversion and product formation
pH 3.5 was identified as the best condition in the range of per-
formed experiments. Between 25 °C and 37 °C and in a pH
range of 2.8–4 there was only a low impact of temperature and
The reactor for the electro-chemoenzymatic conversion has
1
8
been described in detail previously. Briefly, a gas diffusion
2
electrode (GDE) with an apparent surface area of 5.5 cm was
used as the cathode and platinum was used as the anode. The
reactor casing has three parts, the middle part forming the
flow channel, with an internal volume of 8 mL. The platinum
anode is placed on one side, and the gas diffusion cathode is
placed on the other one. Oxygen from the air can diffuse
through a notch of the casing to the reverse side of the GDE. A
current generator was used to apply suitable currents for the
electrogeneration of H O by oxygen reduction. The reactor
2
2
was applied in the bypass of a 50 mL reservoir and a flow of
0 mL min− was pumped from the reservoir through the
1
3
reactor cell. Experiments were carried out with an untreated
and a pretreated GDE, respectively. Pre-treatment of the GDE
was done by saturation with 5 mM thymol solution for 24 h at
room temperature to prevent adsorption of thymol to the PTFE
Fig. 1 Representative time course of the CPO-initiated chlorination of
thymol. Conditions: citrate buffer 0.1 M, pH 3.5, 10 nM CPO, 0.5 mM
2 2
thymol, 10 mM sodium chloride, 0.088 mM H O added every 10 min
up to 0.44 mM, T = 25 °C. Each data point represents the endpoint of an
layer of the electrode. The terminal voltage V depended linearly independent experiment.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 1104–1108 | 1105

View Article Online
Communication
Green Chemistry
Table 1 Influence of reaction parameters on the product formation of
chlorothymol. Conditions: 100 mM citrate buffer, 10 nM CPO; 0.5 mM
Total turnover number and space time yield can be opti-
mized by high substrate concentrations and equimolar con-
thymol; 0.44 mM H
stepwise addition of H
substrate conversion took place (reaction time 90–150 min)
2
O
2
; 25 °C, enzyme inactivation was prevented by
2 2
centrations of H O . A ttn of 145 000 mol product per mol
2
O
2
, the reaction was stopped until no further
enzyme was measured at equimolar concentrations of thymol
2
2
and H O . The addition of caffeic acid as an antioxidant led
2
2
[
NaCl]/ Product formation
to an improved ttn of 164 000. On the other hand the pro-
duction rate decreased to 0.68 mM h . The measured product
rate/mM h−
1
Conversion/%
pH/− Temperature/°C mM
−1
yield is in the range of the theoretical product yield.
Variation of pH
a
2
3
4
6
.8
.5
25
25
25
25
10
10
10
10
0.33
0.29
0.27
0.02
83
86
73
15
Having set the stage with external addition of hydrogen per-
oxide we advanced to systems enabling its in situ generation
based on catalytic reduction of molecular oxygen. Electro-
b
1
4,23
Variation of temperature
chemical methods
systems
appear attractive compared to established
3
3
3
.5
.5
.5
25
30
37
10
10
10
0.29
0.30
0.29
86
80
75
17,24
as the cathode serves as a reagent-free source of
reducing equivalents and accumulation of co-products in the
reaction mixture can be avoided. Recently, we have demon-
Variation of NaCl concentration
1
3.5
3.5
3.5
3.5
3.5
25
25
25
25
25
1
5
10
25
50
0.19
56
77
86
77
79
2 2
strated that electrochemical generation of H O at a gas
0.39
0.29
0.35
0.24
diffusion electrode (GDE) can be used to develop efficient CPO
1
8
catalyzed processes. Therefore, we used this reactor system to
improve the chlorination of thymol (Scheme 2).
a
b
In a first set of experiments, we investigated the correlation
between H O generation rate and the applied current (citrate
2 2
Experiment was performed twice, the standard deviation was 5.5%.
Experiment was performed twice, the standard deviation was 5.1%.
buffer 1 mM, pH 3.5). Using untreated electrodes a linear
increase of the H O -generation rate with increasing current
2
2
−
1
−2
pH, respectively, on product formation. However, the conver- was observed until 20 mA (0.53 µmol H
sion was improved at low temperatures. In general, the CPO Beyond that value, the H generation rate did not increase
catalyzed chlorination of thymol is less affected by changes of any more, most probably due to O diffusion to the cathode
the temperature and pH in the ranges mentioned before. For a surface becoming overall rate limiting. Using thymol-equili-
technical process no laborious control of these parameters is brated electrodes (vide infra) H -production rates were
needed, and the reaction can be performed at the given room reduced by roughly 30% but showed linear dependency on the
2 2
O min cm ).
2 2
O
2
2 2
O
−
1
−2
2 2
temperature. Sodium chloride had a notable influence on the current applied up to 60 mA (0.7 µmol H O min cm ).
product formation. The highest product formation rate was Overall, we concluded that the GDE-mediated reduction may
measured at 5 mM NaCl and the highest product concen- be a viable approach to provide CPO with appropriate amounts
tration at 25 mM NaCl. The rate of chloride incorporation of H
decreased with increased halogen concentration. At 1 mM minimising H
8% of the chloride was incorporated into chlorothymol. At we coupled the electrogeneration of H O
2 2
2
O
2
to sustain high hypochloride generation rates while
-related inactivation of the biocatalyst. Next,
to the CPO-initiated
2 2
O
2
higher concentrations an increase of the product con- halogenation of thymol. However, using untreated electrodes,
centration occurred, and in parallel the yield on chloride apparent productivities (chlorothymol accumulation rates)
decreased. The halogenation process was optimized by testing were approximately 2.3 times lower than expected from the
different concentration ratios between the monoterpene and
2 2
H O generation rate. At the same time, the mass-balance of
hydrogen peroxide (Table 2).
the reaction was not closed and only 40% of the substrate was
recovered (either as unreacted thymol or product). We sus-
pected adsorption of the reactants to the hydrophobic PTFE
2 2
Table 2 Influence of varying thymol/H O ratios on the chemoenzy-
matic halogenation. Conditions: 100 mM citrate buffer, pH 3.5, 25 mM
sodium chloride, 10 nM CPO, T = 30 °C, the reaction was stopped until
no further substrate conversion took place (reaction time 90–150 min)
Thymol/
mM
H
mM
2
O
2
/
ttn/mol product
mol enzyme−
Product formation
rate/mM h
Product
yield/%
1
−1
0
1
1
1
2
2
2
.5
.0
.0
.0
.5
.5
.5
0.44
0.44
0.7
1
0.44
2.5
2.5
43 000
45 000
69 000
88 000
50 000
145 000
164 000
0.29
0.30
0.28
0.59
0.33
0.78
0.68
86
47
66
90
22
73
78
a
Scheme 2 Electro-chemoenzymatic halogenation of thymol to para-
and ortho-chlorothymol with a ratio of 30% to 70%.
a
22
In the presence of 0.25 mM caffeic acid.
1106 | Green Chem., 2014, 16, 1104–1108
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Communication
Table 3 Halogenation of thymol, chlorothymol and carvacrol was basi-
cally performed like the experiments shown in Fig. 1 (0.1 mM citrate
buffer (pH 3.5, T = 30 °C), substrate concentration 1 mM, 25 mM sodium
2 2
salt of halide, 10 nM CPO, H O was added five times every 10 min to a
final concentration of 1 mM). The products were identified by m/s
spectra without quantification
Substrate
Thymol
Halide
Cl−
Product
Thymol
Br−
Fig. 2 Formation of chlorothymol in an electro-chemoenzymatic reactor
at a thymol-saturated electrode (0.1 M citrate buffer, pH 3.5, 10 nM CPO,
2
.5 mM thymol, 50 mM sodium chloride, continuous production of
F−
Thymol
—
2 2
H O
, T = 30 °C). The insert shows the product formation over time of a
step-wise addition of hydrogen peroxide and the electro-chemoenzymatic
conversion at 45 mA. The kinetics of the reactions are shown in the
ESI.†
Cl−
Chlorothymol
Cl−
membrane of the gas diffusion electrode to account for both Carvacrol
observations. Indeed, repeating the electro-chemoenzymatic
experiments in reactor setups pre-equilibrated with thymol,
improved mass-balances with 66% substrate recovery were
obtained. Nevertheless, a further improvement of the GDE for
hydrophobic substrates is necessary to obtain a closed mass-
harsh reaction conditions and results in unwanted bypro-
duct formation. Therefore it is of great interest to investigate
the biosynthesis of halogenated natural products and the
biotechnological potential of halogenating enzymes. Here,
we demonstrate for the first time the chemoenzymatic pro-
duction of halogenated phenolic monoterpenes thymol and
carvacrol.
balance. Again, the productivity of the overall reaction system
depended on the current applied (Fig. 2). Compared to the pro-
ductivity of the initial system using externally added H O
2
2
2
5
up to 0.78 mM h− ) the productivity of this electro-chemo-
1
(
−
1
enzymatic process was up to 4.6 mM h
.
It is worth mentioning here that throughout the experi-
ments reported here, no dihalogenation product was observed.
Only upon using chlorothymol as the starting material,
gradual accumulation of a further product was observed,
which we identified as dichlorothymol via GC/MS. Due to the
lack of authentic standards, an exact quantification of this
dihalogenation reaction is not possible at present, but – as
very roughly estimated from the GC area – the second
halogenation step proceeded significantly slower than the first
one. This may also explain the very high chemoselectivity in
the reactions observed.
Finally, the product scope of the proposed chemoenzymatic
halogenation was explored to some extent. As shown in
Table 3, thymol bromination was also feasible. Quite expect-
edly, no fluorination was observed. Likewise, carvacrol was
also converted smoothly into bromocarvacrol.
H O is considered to be an environmentally friendly
2
2
chemical, since it leaves no hazardous residues, such as other
oxidants. Electrogeneration of H O is an attractive approach
2
2
since it does not require additional chemicals, and electricity
is readily available.
II
Compared to the current state of the art using Cu -cata-
lysed in situ generation of hypochlorite from chloride and
2
6
molecular oxygen, our method excels especially by the cata-
lyst performance of CPO compared to simple transition metal
salts. Hence, the catalyst loading (0.0004–0.002 mol%) is
orders of magnitude lower than in the chemical counterpart
(
12.5 mol%) while achieving comparable product yields.
Additionally, the milder reaction conditions (RT and aqueous
reaction media vs. 80 °C and acetic acid as a solvent) presum-
ably contribute to higher product quality (and reduced down-
stream processing and purification effort).
Admittedly, at present, the low substrate loadings used
impair preparative applicability. However, further improve-
Conclusions
Halogenated natural products are widely distributed in nature, ments such as the use of cosolvents, two-liquid-phase systems
some of them being potent biologically active substances. or slurry-to-slurry reaction setups are currently being
Incorporation of halogen atoms into drugs is a common strat- implemented in our laboratory. We are confident that these
egy to modify molecules in order to vary their bioactivities and improvements will result in a truly environmentally more
2
5
specificities. Chemical halogenation, however, often requires benign alternative to existing chemical technologies.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 1104–1108 | 1107

View Article Online
Communication
Green Chemistry
1
3 F. Hollmann, I. W. C. E. Arends, K. Buehler, A. Schallmey
and B. Buhler, Green Chem., 2011, 13, 226–265.
Notes and references
1
P. K. Ajikumar, K. Tyo, S. Carlsen, O. Mucha, T. H. Phon 14 M. Hofrichter, R. Ullrich, M. J. Pecyna, C. Liers and
and G. Stephanopoulos, Mol. Pharmaceutics, 2008, 5, 167–
90.
T. Lundell, Appl. Microbiol. Biotechnol., 2010, 87, 871–
897.
1
2
3
4
5
6
C. C. de Carvalho and M. M. da Fonseca, Biotechnol. Adv., 15 M. Hofrichter and R. Ullrich, Appl. Microbiol. Biotechnol.,
006, 24, 134–142. 2006, 71, 276–288.
H. Schewe, M. A. Mirata, D. Holtmann and J. Schrader, 16 B. Kaup, K. Ehrich, M. Pescheck and J. Schrader, Biotech-
Process Biochem., 2011, 46, 1885–1899. nol. Bioeng., 2008, 99, 491–498.
M. S. Yusubo, L. A. Drygunov, A. V. Tkachev and 17 K. Seelbach, M. P. J. van Deurzen, F. van Rantwijk and
V. V. Zhdankin, ARKIVOC, 2005, IV, 179–188. R. A. Sheldon, Biotechnol. Bioeng., 1997, 55, 283–288.
J. J. Polzin, G. L. Rorrer and D. P. Cheney, Biomol. Eng., 18 T. Krieg, S. Huttmann, K.-M. Mangold, J. Schrader and
003, 20, 205–215. D. Holtmann, Green Chem., 2011, 13, 2686–2689.
S. Naylor, F. J. Hanke, L. V. Manes and P. Crews, in 19 F. Hollmann and I. W. C. E. Arends, Polymers, 2012, 4, 759–
Fortschritte der Chemie organischer Naturstoffe/Progress in 793.
the Chemistry of Organic Natural Products, Springer, Vienna, 20 A. Longoria, R. Tinoco and R. Vázquez-Duhalt, Chemo-
983, vol. 44, ch. 3, pp. 189–241. sphere, 2008, 72, 485–490.
R. W. Fuller, J. H. Cardellina, Y. Kato, L. S. Brinen, 21 G. Bayramoğlu, S. Kiralp, M. Yilmaz, L. Toppare and
2
2
1
7
8
9
J. Clardy, K. M. Snader and M. R. Boyd, J. Med. Chem.,
992, 35, 3007–3011.
R. W. Fuller, J. H. Cardellina, J. Jurek, P. J. Scheuer,
M. Y. Arıca, Biochem. Eng. J., 2008, 38, 180–188.
22 C. E. R. Grey, Fabian, Adlercreutz and Patrick, J. Biotechnol.,
2008, 135, 196–201.
1
B. Alvarado-Lindner, M. McGuire, G. N. Gray, J. R. Steiner 23 S. Lütz, K. Vuorilehto and A. Liese, Biotechnol. Bioeng.,
and J. Clardy, J. Med. Chem., 1994, 37, 4407–4411.
S. S. Salih, Br. J. Pharmacol. Toxicol., 2012, 3, 147–150.
2007, 98, 525–534.
24 F. van de Velde, N. D. Lourenço, M. Bakker, F. van Rantwijk
and R. A. Sheldon, Biotechnol. Bioeng., 2000, 69, 286–291.
25 C. Wagner, M. El Omari and G. M. König, J. Nat. Prod.,
2009, 72, 540–553.
1
0 M. A. Radwana, S. R. El-Zemitya, S. A. Mohameda and
S. M. Sherbya, Int. J. Trop. Insect Sci., 2008, 28.
1 A. Andersen, Int. J. Toxicol., 2006, 25, 29–127.
1
1
2 A. Podgoršek, M. Zupan and J. Iskra, Angew. Chem., Int. Ed., 26 L. Menini and E. V. Gusevskaya, Chem. Commun., 2006,
009, 48, 8424–8450. 209–211.
2
1108 | Green Chem., 2014, 16, 1104–1108
This journal is © The Royal Society of Chemistry 2014