Selective oxidative decarboxylation of amino acids to produce industrially relevant nitriles by vanadium chloroperoxidase

Article abstract of DOI:10.1002/cssc.201200098

Industrial nitriles from biomass: Vanadium-chloroperoxidase is successfully used to transform selectively glutamic acid into 3-cyanopropanoic acid, a key intermediate for the synthesis of bio-succinonitrile and bio-acrylonitrile, by using a catalytic amount of a halide salt. This clean oxidative decarboxylation can be applied to mixtures of amino acids obtained from plant waste streams, leading to easily separable nitriles. Copyright

Full text of DOI:10.1002/cssc.201200098

DOI: 10.1002/cssc.201200098
Selective Oxidative Decarboxylation of Amino Acids to Produce Industrially
Relevant Nitriles by Vanadium Chloroperoxidase
[
a, b]
[b]
[b]
[c]
[b]
Andrada But,
Jꢀrꢁme Le Nꢁtre,* Elinor L. Scott,* Ron Wever, and Johan P. M. Sanders
[
5]
With growing ecological concerns, fluctuating oil prices and se-
curity of supply, there have been many developments towards
the production of biofuels. More recently, the chemical indus-
try has also started to expand towards biobased production
routes for chemicals and materials. Alternative renewable re-
sources such as carbohydrates, lignocellulosic biomass, glycer-
aminobutyric acid (GABA), N-methylpyrrolidone (NMP), N-vi-
[6]
[7]
[8]
nylpyrrolidone (NVP),
succinonitrile,
and acrylonitrile.
Techno-economic and environmental assessments have shown
the potential and points for further development in the use of
[9]
glutamic acid as a platform chemical. The main intermediate
[7]
[8]
in the syntheses of succinonitrile and acrylonitrile from glu-
tamic acid is 3-cyanopropanoic acid (1) (Figure 1). It is obtained
by an oxidative decarboxylation reaction using sodium hypo-
chlorite, and life-cycle assessment and comparison to petro-
chemical-based syntheses have shown that this step is detri-
[1]
ol, and fatty acids have become a main focus of research.
In addition to conventional biobased feedstocks, research is
also focusing on the transformation of protein-rich waste
streams into bulk chemicals. Amino acids may be considered
as an economically and energetically attractive raw material for
nitrogen-containing chemicals.
[9]
mental.
Their functionality reduces the
need of energy-intensive pro-
cesses to functionalize hydrocar-
bons with ammonia (or other re-
agents) and offers potential to
[
2]
circumvent process steps. From
the expected global rise in bio-
[
3a]
fuels by 2020, and considering
the European Union target of
Figure 1. Biobased syntheses of succinonitrile and acrylonitrile from glutamic acid via 3-cyanopropanoic acid
[7–9]
(
1).
1
0% of the total transportation
[
3b]
market as biofuels by 2020, an
additional 100 million metric tonnes of protein will be pro-
duced globally, corresponding to ca. 5 million tonnes of each
amino acid. Not all amino acids are required for nutritional pur-
poses. The isolation of non-essential amino acids, such as glu-
tamic acid, could lead to their use without compromising feed
or food value of the remaining composition. Glutamic acid is
the most abundant amino acid in a number of biofuel rest
Herein we report for the first time the conversion of gluta-
mic acid to 3-cyanopropanoic acid (1) in high yield and selec-
tivity using a vanadium chloroperoxidase enzyme (VCPO) and
hydrogen peroxide. The reaction proceeds with a catalytic
amount of sodium bromide and constitutes a general and
more sustainable route to industrial nitriles from biomass due
to the elimination of sodium hypochlorite.
[4]
streams, such as dried distillers grains and solubles (DDGS),
A large number of halogenated compounds are produced in
[10]
and could be considered as an interesting raw material for the
production of industrial chemicals. Glutamic acid has been
shown to be a versatile starting material for the synthesis of g-
nature, and the first halogenating enzyme, or haloperoxidase
[11]
(HPO), was discovered in 1966 by Hager et al. Two types of
haloperoxidase co-exist in nature, depending on their prosthet-
ic group which can be either an iron–heme complex or a vana-
dium oxide moiety. In the presence of hydrogen peroxide and
[a] A. But
a halogen source they produce a halogenating intermediate or
“
Politehnica” University of Timisoara
+
“
X ” [Scheme 1, Equation (1)], which reacts with organic mole-
Industrial Chemistry and Environmental Engineering Faculty
Piata Victoriei 2, 300006 Timisoara (Romania)
cules to form the corresponding halogenated molecule
[
b] A. But, Dr. J. Le Nꢀtre, Dr. E. L. Scott, Prof. J. P. M. Sanders
Chair of Biobased Commodity Chemicals
Wageningen University
PO Box 17, 6700AA Wageningen (The Netherlands)
Fax: (+31)317-483011
E-mail: jerome.lenotre@wur.nl
elinor.scott@wur.nl
[c] Prof. R. Wever
Van’t Hoff Institute for Molecular Sciences
University of Amsterdam
PO Box 94157, 1090 GD Amsterdam (The Netherlands)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201200098.
Scheme 1. General scheme for the halogenation reaction with haloperoxi-
dase enzymes (HPOs).
ChemSusChem 0000, 00, 1 – 4
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

[
Scheme 1, Equation (2)]. A reaction may also occur between
of phenylalanine (2), 2 mm of sodium bromide, and 18 nm of
VCPO. Hydrogen peroxide (1.2 mm) was added every 15 min
over a period of 5 h to maintain a low concentration and avoid
+
X
and hydrogen peroxide which result in the formation of
singlet oxygen [Scheme 1, Equation (3)].
[14]
In our study we used a vanadium chloroperoxidase (VCPO)
obtained from the terrestrial fungus Curvularia inaequalis that
singlet oxygen formation. After screening the reaction at pH
values from 5.0 to 6.2, a pH of 5.6 was selected at which the
best conversion of phenylalanine (2) to phenylacetonitrile (3)
was obtained with a minimum of degradation of the products
(see Supporting Information).
[
12]
has been fully characterized and can be produced on a large
[
13]
scale. This enzyme has a number of interesting properties,
such as a high thermostability (908C), and can be stored at
À208C over several years without activity loss. It also shows
a very high resistance towards oxidants such as hydrogen per-
Figure 2 shows that a complete conversion of phenylalanine
(2) was observed after 4 h of reaction and phenylacetonitrile
(3) was formed as the major product, together with phenylace-
taldehyde (4) in a 3:1 ratio. Phenylacetonitrile production
[
14]
oxide (up to 100 mm) and singlet oxygen, is stable in the
[
15]
presence of organic solvents,
enzyme remains fully stable for
5000 turnovers. The Km in bro-
and during conversions the
2
mide oxidation is unusually
small (less than 10 mm), which
means that the enzyme is fully
functional at low bromide con-
centration.
Although both heme and va-
nadium peroxidases have been
[
16]
used in organic syntheses, to
halogenate or oxidize com-
pounds there is only one report
on the use of a heme containing
bromoperoxidase in the transfor-
mation of amino acids to nitriles
via an oxidative decarboxylation
[
17]
reaction.
studied were converted into
their corresponding nitriles.
The amino acids
However due to a deamination
reaction, aldehydes were ob-
tained as side products and the
Figure 2. Phenylacetonitrile (3) and phenylacetaldehyde (4) formation from phenylalanine (2).
ratio nitrile/aldehyde never exceeded 3:1. In the reaction
mechanism a hypobromite ion is formed that reacts with the
carboxylic function of the amino acid, inducing a decarboxyla-
seemed to reach its maximum after 4 h, and the maximum
concentration of phenylacetaldehyde (4) was obtained after
3 h. The results suggest that in the following two hours a deg-
radation of phenylacetaldehyde (4) occurred, probably due to
the presence of an excess of oxidant, where phenylacetonitrile
(3) is not affected. Thus about 10% of the phenylalanine is
lost. However, phenylacetonitrile (3) was produced in 70%
yield.
+
tion. A second “Br ” reacts with the primary amine and the ni-
trile function is formed. If the reaction is catalytic in bromide,
two equivalents of oxidant are needed to reach complete con-
version.
The bromoperoxidase activity of the VCPO enzyme used in
this article has been well-studied, with turnover values ranging
Because VCPO enzymes can also oxidize chloride, this halide
(5 mm) was also tested in the reaction instead of bromide.
However only 38% of phenylalanine (2) was converted after
5 h, probably due to the slower rate of oxidation of chloride
À1
[12a,18]
from 6 to 250 s depending upon pH.
First, phenylala-
nine (2) was chosen as a test substrate to be able to follow the
reaction by HPLC analysis using UV detection and optimize the
reaction conditions (Scheme 2). The reaction was performed at
room temperature in citrate buffer at pH 5.6 containing 5 mm
[12a]
by the enzyme.
Furthermore, only phenylacetaldehyde (4)
was formed and the selectivity of the reaction was completely
inverted. This suggests that the mechanism of the re-
action is dependent on the nature of the halide used.
We did not investigate this further but the produc-
tion of phenylacetaldehyde (4) from phenylalanine
may be an interesting option because it is a valuable
product in the fragrance industry.
Notably, phenylacetonitrile (3) and phenylacetalde-
hyde (4) are poorly miscible with water and therefore
the samples were mixed with a solution of acetoni-
Scheme 2. Enzymatic oxidative decarboxylation of phenylalanine (2).
&2
&
www.chemsuschem.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 4
ÝÝ
These are not the final page numbers!

trile/methanol (1:1) to obtain homogeneous samples prior to
analysis (see Supporting Information). This property is, howev-
er, an advantage in the isolation of the products. Phase separa-
tion or continuous extraction from an aqueous reaction mix-
ture is in principle easily carried out. Indeed it was possible to
continuously extract the aqueous reaction phase with different
solvents, such as toluene, hexane, and ethyl acetate. Ethyl ace-
tate gave the best results: it did not affect the rate of
the reaction, slightly increased the selectivity towards
phenylacetonitrile (3) (up to 83%), and especially the
degradation of phenylacetaldehyde (4) did not occur.
The reaction conditions described above were
then used in the conversion of glutamic acid at
pH 5.6 (Scheme 3). The pH dependence was also
studied and, as for phenylalanine (2), an optimal pH
value of 5.6 to 5.8 was found (see Supporting Infor-
mation).
Scheme 3. Enzymatic oxidative decarboxylation of glutamic acid into 3-cya-
nopropanoic acid (1).
As depicted in Figure 3, the rate of conversion of
glutamic acid was similar to that of phenylalanine (2)
and reached completion after 5 h. However, the se-
lectivity was better, with only the nitrile product, 3-
cyanopropanoic acid (1), formed and no product
degradation. Thus the VCPO enzyme can transform
Figure 3. 3-Cyanopropanoic acid (1) formation from glutamic acid.
glutamic acid into 3-cyanopropanoic acid (1) in com-
plete conversion and high selectivity. Chloride was
also tested instead of bromide, but the yield of 3-cyanopropa-
noic acid (1) reached only 50% after 5 h reaction, which is con-
sistent with our previous conclusion on the choice of halide af-
fecting reaction.
degradation of phenylacetaldehyde (4) was observed after 3 h
of reaction.
This suggests that our oxidative decarboxylation system
may be applied to a mixture of amino acids without any losses
of reactivity and selectivity. Depending on the substrate range,
it is conceivable that this enzymatic process could be applied
to more complex mixtures of amino acids such as hydrolysate
of protein-rich waste streams. Immobilizing the enzyme on
In our efforts to valorize protein-rich waste streams for the
production of bulk chemicals, we often encounter problems in
extracting amino acids from water and their separation prior
to modification to bulk chemicals. In this respect the oxidative
decarboxylation reaction by vanadium chloroperoxidase may
be a very effective tool to perform, in a single step, the trans-
formation of mixtures of amino acids into nitriles and the ex-
traction of the hydrophobic
[19,20]
a solid support
may allow recyclability and scale-up.
In conclusion, vanadium chloroperoxidase appears as an effi-
cient and a selective tool to transform glutamic acid into 3-cya-
products from the aqueous layer
by phase separation. The water-
insoluble nitriles could then be
isolated by crystallization or dis-
tillation. To validate this idea we
applied the haloperoxidase oxi-
dative system to an equimolar
mixture of phenylalanine (2) and
glutamic acid (Figure 4).
After 4 h of reaction both
amino acids were fully converted
and the ratio of products ob-
tained fit the individual reactions
seen with each amino acid. Glu-
tamic acid was transformed into
3-cyanopropanoic acid (1) with-
out degradation; phenylacetoni-
trile (3) and phenylacetaldehyde
(
4) were produced from phenyla-
Figure 4. Products formation of the enzymatic oxidative decarboxylation performed on a mixture of phenylalanine
lanine (2) with a 3:1 ratio and (2) and glutamic acid.
ChemSusChem 0000, 00, 1 – 4
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
3
&
ÞÞ
These are not the final page numbers!

nopropanoic acid (1). Compared to our previous studies, in
which bleach was used to perform the reaction, this enzymatic
process appears more sustainable and is likely to be more eco-
nomically viable. Also during the process bromide is regenerat-
ed and only a low concentration is needed. Thus, it may be
possible to develop a suitable biobased route to succinonitrile
and acrylonitrile. More efforts are currently in progress in our
laboratory to improve this transformation and to study its eco-
logical and economic aspects. In addition the oxidative decar-
boxylation catalyzed via formation of HOBr by the stable vana-
dium chloroperoxidase may be used to convert a mixture of
amino acids. For example the amino acids present in the hy-
drolysate of protein-rich waste streams could be transformed
into nitriles of industrial interest. These are easily extracted
from water and can be separated by conventional purification
techniques, avoiding the difficulty of isolating individual amino
acids from aqueous protein hydrolysate.
Chem. 2010, 12, 539–554; f) J. Zakzeski, P. C. A. Bruijnincx, A. L. Jonger-
ius, B. M. Weckhuysen, Chem. Rev. 2010, 110, 3552–3599.
2] E. Scott, F. Peter, J. Sanders, Appl. Microbiol. Biotechnol. 2007, 75, 751–
[
762.
[3] a) http://www.iea.org/press/pressdetail.asp?PRESS_REL_ID=411 and re-
lated report: http://www.iea.org/papers/2011/biofuels_roadmap.pdf;
b) Directive 2009/28/EC of the European Parliament and The Council of
rd
April the 23 2009, Official Journal of the European Union 2009, L140/
16.
[4] a) T. M. Lammens, Ph.D. Thesis, Wageningen University, 2011; b) V.
Deike, J. D. Mol, Raw Material Compendium, Novus International, Brus-
sels, 1996; c) M. J. Spiehs, M. H. Whitney, G. C. Shurson, J. Anim. Sci.
2
002, 80, 2639–2645; d) Y. Kim, N. S. Mosier, R. Hendrickson, T. Ezeji, H.
Blaschek, B. Dien, M. Cotta, D. Dale, M. R. Ladisch, Bioresour. Technol.
2008, 99, 5165–5176.
[5] T. M. Lammens, D. De Biase, M. C. R. Franssen, E. L. Scott, J. P. M. Sand-
ers, Green Chem. 2009, 11, 1562–1567.
6] T. M. Lammens, M. C. R. Franssen, E. L. Scott, J. P. M. Sanders, Green
Chem. 2010, 12, 1430–1436.
[
[7] T. M. Lammens, J. Le Nꢁtre, M. C. R. Franssen, E. L. Scott, J. P. M. Sanders,
ChemSusChem 2011, 4, 785–791.
[
[
8] J. Le Nꢁtre, E. L. Scott, M. C. R. Franssen, J. P. M. Sanders, Green Chem.
011, 13, 807–809.
2
9] a) T. M. Lammens, S. Gangarapu, M. C. R. Franssen, E. L. Scott, J. P. M.
Sanders, Biofuels Bioprod. Biorefin. 2012, 6, 177–187; b) T. M. Lammens,
J. Potting, J. P. M. Sanders, I. J. M. de Boer, Environ. Sci. Technol. 2011, 45,
8521–8528.
10] G. W. Gribble, Naturally Occurring Organohalogen Compounds—A Com-
prehensive Update, Springer, 2010.
Experimental Section
Typical procedure for the VCPO-mediated oxidative decarboxylation
of phenylalanine: In a 2 mL vial a solution containing 5 mm of phe-
nylalanine, 2 mm of NaBr, 18 nm of VCPO, and 20 mm citrate buffer
at the desired pH was stirred at 300 rpm at room temperature
[
[
[
11] L. P. Hager, D. R. Morris, F. S. Brown, H. Eberwein, J. Biol. Chem. 1966,
(
(
218C). To this reaction mixture, small aliquots (2 mLꢃ15) of H O2
24 mm) were added at intervals of 20 min. To prevent evaporation
2
2
41, 1769–1777.
12] a) J. W. P. M. van Schijndel, E. G. M. Vollenbroek, E. G. M. R. Wever, Bio-
chim. Biophys. Acta Protein Struct. Mol. Enzymol. 1993, 1161, 249–256;
b) W. Hemrika, R. Wever, Handbook of Metalloproteins 2001, 1416–
1428; c) W. Hemrika, R. Renirie, S. Macedo-Ribeiro, A. Messerschmidt, R.
Wever, J. Biol. Chem. 1999, 274, 23820–23827.
during the reaction the vials were closed with caps, the gas pres-
sure was equilibrated when the covers were opened for the addi-
tion of hydrogen peroxide. The reaction was stopped by adding
Na S O to the mixture, 6 mL of acetonitrile/methanol (1:1, v/v)
2
2
3
were added to homogenize the reaction mixture, and samples
were taken for UHPLC analysis.
[13] P. Barnett, D. H. Hondmann, L. H. Simons, P. F. ter Steeg, R. Wever (Unile-
ver), Int. Patent Appl. WO95/27046, 1995.
[
[
[
14] R. Renirie, C. Pierlot, J.-M. Aubry, A. F. Hartog, H. E. Schoemaker, P. L. Al-
sters, R. Wever, Adv. Synth. Catal. 2003, 345, 849–858.
15] J. W. P. M. van Schijndel, P. Barnett, J. Roelse, E. G. M. Vollenbroek, R.
Wever, Eur. J. Biochem. 1994, 225, 151–157.
Acknowledgements
16] a) D. Wischang, O. Brꢄcher, J. Hartung, Coord. Chem. Rev. 2011, 255,
We are grateful to the Erasmus Programme for financial support
2204–2217; b) A. Butler, Coord. Chem. Rev. 1999, 187, 17–35; c) D. Wi-
(
A.B.) and to Prof. Dr. Francisc Peter from the University of Timi-
schang, J. Hartung, Tetrahedron 2011, 67, 4048–4054; d) J. S. Martinez,
G. L. Carroll, R. A. Tschirret-Guth, G. Altenhoff, R. D. Little, A. Butler, J.
Am. Chem. Soc. 2001, 123, 3289–3294; e) J. N. Carter-Franklin, J. D. Par-
rish, R. A. Tschirret-Guth, R. D. Little, A. Butler, J. Am. Chem. Soc. 2003,
soara. We also thank Susan Witte and Jana Kralova for their sup-
ports with HPLC and UHPLC analyses, and Maurice Franssen and
Rokus Renirie for their helpful discussions.
1
25, 3688–3689.
[
[
17] M. Nieder, L. Hager, Arch. Biochem. Biophys. 1985, 240, 121–127.
18] Z. Hasan, R. Renirie, R. Kerkman, H. J. Ruijssenaars, A. F. Hartog, R.
Wever, J. Biol. Chem. 2006, 281, 9738–9744.
Keywords: amino acids · biomass · enzyme catalysis · halides ·
oxidation
[19] M. Hartmann, C. Streb, J. Porous Mater. 2006, 13, 347–352.
[
20] D. Wischang, J. Hartung, T. Hahn, R. Ulber, T. Stumpf, C. Fecher-Trost,
[
1] a) A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411–2502; b) P.
Gallezot, Chem. Soc. Rev. 2012, 41, 1538–1558; c) J. van Haveren, E. L.
Scott, J. Sanders, Biofuels Bioprod. Biorefin. 2008, 2, 41–57; d) C. H.
Christensen, J. Rass-Hansen, C. C. Marsden, E. Taarning, K. Egeblad,
ChemSusChem 2008, 1, 283–289; e) J. J. Bozell, G. R. Petersen, Green
Green Chem. 2011, 13, 102–108.
Received: February 8, 2012
Published online on && &&, 0000
&4
&
www.chemsuschem.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 4
ÝÝ
These are not the final page numbers!

COMMUNICATIONS
A. But, J. Le Nꢀtre,* E. L. Scott,* R. Wever,
J. P. M. Sanders
&& – &&
Selective Oxidative Decarboxylation of
Amino Acids to Produce Industrially
Relevant Nitriles by Vanadium
Chloroperoxidase
Industrial nitriles from biomass: Vana-
dium-chloroperoxidase is successfully
used to transform selectively glutamic
acid into 3-cyanopropanoic acid, a key
intermediate for the synthesis of bio-
succinonitrile and bio-acrylonitrile, by
using a catalytic amount of a halide salt.
This clean oxidative decarboxylation can
be applied to mixtures of amino acids
obtained from plant waste streams,
leading to easily separable nitriles.
ChemSusChem 0000, 00, 1 – 4
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
&
5
&
ÞÞ
These are not the final page numbers!