Selective hydrolysis of aliphatic dinitriles to monocarboxylic acids by a nitrilase from Arabidopsis thaliana

Article abstract of DOI:10.1055/s-2001-17523

The hydrolysis of a variety of dinitriles including α,ω-dicyanoalkanes 1, β-substituted glutaronitriles 5, and γ-cyanopimelonitrile 7 with a recombinant plant nitrilase from Arabidopsis thaliana, expressed in E. coli, is described. Conversion rate and selectivity of the hydrolysis of dinitriles 1a-f to ω-cyanocarboxylic acids 2a-f depend on the chain length. The enzyme activity markedly increases from malononitrile (1a) to octanedinitrile (1f). The selectivity, however, does not correlate with the rates. Up to a chain length of 6 C-atoms, the cyanocarboxylic acid is the only product, even at complete conversion of the starting material. Pimelonitrile (1e) is hydrolyzed to the cyanocarboxylic acid 2e without formation of diacid (<1%) up to 73% conversion. Glutaronitriles 5a-c were also hydrolyzed to the corresponding cyanobutanoic acids 6a-c with perfect selectivity. The nitrilase hydrolyzes exclusively the primary cyano group of 7 to give 3,5-dicyanoheptanoic acid (8a), whereby the selectivity is slightly reduced compared to the unsubstituted pimelonitrile (1e). If the hydrolysis is terminated at conversions ≤90%, pure 8a can be isolated in 72% yield (92% referred to conversion). After esterification of 8a to the methyl ester 8b, only the 5-cyano group but not the ester function was hydrolyzed enzymatically to give cyanoheptanedioic acid monoester (10).

Full text of DOI:10.1055/s-2001-17523

1866
PAPER
Selective Hydrolysis of Aliphatic Dinitriles to Monocarboxylic Acids by a
Nitrilase from Arabidopsis thaliana1
Selective
Hyd
r
o
a
Aliphatic
n
D
initriles by
z
Nitrilase Effenberger,* Steffen Oßwald2
Institut für Organische Chemie, Universität Stuttgart, Pfaffenwaldring 55, 70569 Stuttgart, Germany
Fax +49(711)6854269; E-mail: franz.effenberger@po.uni-stuttgart.de
Received 10 April 2001; revised 19 June 2001
isophthalonitrile and fumaronitrile,4 whereas aliphatic
a,w-dinitriles such as glutaronitrile were hydro-lyzed ef-
fectively but not selectively to the monoacids.4 The nit-
rilase from Rhodococcus rhodochrous J1 exhibits
Abstract: The hydrolysis of a variety of dinitriles including a,w-di-
cyanoalkanes 1, b-substituted glutaronitriles 5, and g-cyanopimel-
onitrile 7 with a recombinant plant nitrilase from Arabidopsis
thaliana, expressed in E. coli, is described. Conversion rate and se-
lectivity of the hydrolysis of dinitriles 1a–f to w-cyanocarboxylic specificity for dicyanobenzenes, yielding selectively the
corresponding cyanobenzoic acids.5,6 A nitrilase from Ac-
acids 2a–f depend on the chain length. The enzyme activity mark-
edly increases from malononitrile (1a) to octanedinitrile (1f). The
idovorax facilis 72 W having aliphatic nitrilase activity
selectivity, however, does not correlate with the rates. Up to a chain
converted a,w-dinitriles from malononitrile to glutaroni-
length of 6 C-atoms, the cyanocarboxylic acid is the only product,
trile selectively to the corresponding w-cyanocarboxylic
even at complete conversion of the starting material. Pimelonitrile
(1e) is hydrolyzed to the cyanocarboxylic acid 2e without formation
acids in 92–99% yield, depending on the chain length.3
Pétré et al. reported a selective enzymatic hydrolysis par-
ticularly of adiponitrile to 5-cyanovaleric acid, an inter-
mediate for the preparation of e-caprolactam, using the
nitrilase from Comamonas NI1.7
of diacid (<1%) up to 73% conversion. Glutaronitriles 5a–c were
also hydrolyzed to the corresponding cyanobutanoic acids 6a–c
with perfect selectivity. The nitrilase hydrolyzes exclusively the pri-
mary cyano group of 7 to give 3,5-dicyanoheptanoic acid (8a),
whereby the selectivity is slightly reduced compared to the unsub-
stituted pimelonitrile (1e). If the hydrolysis is terminated at conver-
Cloning and overexpression of a nitrilase from Arabidop-
sions £90%, pure 8a can be isolated in 72% yield (92% referred to sis thaliana8 (EC 3.5.5.1) offered for the first time access
conversion). After esterification of 8a to the methyl ester 8b, only
the 5-cyano group but not the ester function was hydrolyzed enzy-
matically to give cyanoheptanedioic acid monoester (10).
to a plant nitrilase in amounts sufficient for synthetic ap-
plications. Investigations of the substrate range of A.
thaliana demonstrated the acceptance of aliphatic nitriles
with longer chain length.2,9 Because, as mentioned above,
Key words: catalysis, enzymes, hydrolyses, nitriles, regioselectivi-
ty
the monohydrolysis of aliphatic dinitriles is of great in-
dustrial interest, we have examined the possibility of hy-
drolyzing
selectively
aliphatic
dinitriles
to
monocarboxylic acids using an Arabidopsis thaliana nit-
rilase.
Introduction
The nitrilase-mediated selective hydrolysis of dinitriles to
cyanocarboxylic acids is of particular synthetic interest,
since a selective chemical hydrolysis of dinitriles is virtu-
ally impossible. While dinitriles are readily available not
only on bench scale but also in technical quantities, the
preparation of compounds bearing both a cyano group and
a carboxyl function is much more difficult. Nevertheless,
cyanocarboxylic acids are of interest for a variety of appli-
cations. For example, w-cyanocarboxylic acids, derived
from aliphatic a,w-dinitriles by nitrilase-catalyzed selec-
tive hydrolysis, could be converted directly to the corre-
sponding five- and six-membered ring lactams after
hydrogenation of the cyano group to the primary amine.3
Selective Hydrolysis of Unsubstituted a,w-Dicyano-
alkanes
a,w-Dicyanoalkanes 1 with varying chain lengths were
hydrolyzed with a nitrilase from Arabidopsis thaliana as
catalyst as outlined in Scheme 1.
All nitrilases used so far in the selective hydrolysis of
dinitriles to monocarboxylic acids were obtained from
bacterial sources. Depending on the growth substrates,
resting cells of Rhodococcus rhodochrous N.C.I.B 11216,
for example, were capable of hydrolyzing selectively
Scheme 1
Both hydrolysis selectivity and hydrolysis rate were deter-
mined by gas chromatography. Thereby, as stated by coin-
jection, diacids 3, which are commercially available,
could be detected at concentrations of 1% referred to the
Synthesis 2001, No. 12, 25 09 2001. Article Identifier:
1437-210X,E;2001,0,12,1866,1872,ftx,en;T02501ss.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0039-7881

PAPER
Selective Hydrolysis of Aliphatic Dinitriles by Nitrilase
1867
initial concentration of 1. The results of enzymatic hydrol- taronitriles were investigated. Substituents in a-position
ysis are summarized in Table 1.
to the cyano group inhibit the A. thaliana catalyzed hy-
drolysis of a cyano group.2,9 The only exception are a-flu-
oroni-triles, which could be hydrolyzed with A. thaliana.1
Therefore the selective hydrolysis of b-substituted glu-
taronitriles 5 has been examined. Whereas b-hydroxyglu-
taronitrile (5b)12 and b-phenylglutaronitrile (5e)13 are
known, the glutaronitriles 5a,c and d had to be prepared
(Scheme 2).
The hydrolysis of dinitriles 1a–d catalyzed by A. thaliana
afforded selectively the corresponding cyanocarboxylic
acids 2a–d. In all cases, the amount of the corresponding
diacid 3a–d is <1% at 100% conversion. Table 1 clearly
shows the dependence of the nitrilase activity on the chain
length. While the dinitriles 1a–c (n = 1–3) were found to
be moderate substrates for the nitrilase, adiponitrile (1d),
pimelonitrile (1e) and octanedinitrile (1f) (n = 4–6) were
converted very quickly to the monoacids 2d–f. The selec-
tivity of A. thaliana, however, is diminished in the hydrol-
ysis of 1e and 1f to the cyanocarboxylic acids 2e and 2f.
The percentage of diacid 3e and 3f amounts to 1.1% at
83% conversion of 1e and 12.5% at 66% conversion of 1f,
respectively (Table 1). It was found that the initial ratio
2f:3f of approximately 5 remained constant in the course
of hydrolysis. The crystalline 7-cyanoheptanoic acid (2f)
could be separated easily from the diacid 3f and was iso-
lated in 61% yield. In order to obtain pure 6-cyanohexano-
ic acid (2e), the hydrolysis of 1e was terminated at 73%
conversion (percentage of diacid 3e <1%). After distilla-
tion, 2e was obtained in 66% yield (91% referred to con-
version).
Table 1 Selective Hydrolysis of a,w-Dinitriles 1 to w-Cyanocar-
boxylic Acids 2 and Diacids 3 Catalyzed by a Nitrilase from Arabi-
dopsis thaliana
Scheme 2
Substrate Rel. Activity Conversion Selectivity Mono- Yield
(%)a
(%)b
100
100
100
100
83
2:3 (%)
>99:<1
>99:<1
>99:<1
>99:<1
98.9:1.1
87.5:12.5
acids (%)
In a Michael reaction comparable to a known method,13
5a was obtained by reacting ethyl cyanoacetate with crot-
ononitrile. The resulting Michael adduct, ethyl 2,4-dicy-
ano-3-methylbutanoate (4), was hydrolyzed in NaOH and
subsequently decarboxylated in boiling pyridine to give
5a in 42% yield. For the preparation of 5c and 5d, b-hy-
droxyglutaronitrile (5b) was either methylated with dime-
thyl sulfate to give 5c in 63% yield or acetylated with
acetic anhydride to afford 5d in 92% yield. The corre-
sponding cyanobutanoic acids 6a–e required as reference
compounds were accessible by selective enzymatic hy-
drolysis of 5a–e. For characterization, the monoacids 6b–
e were transformed into the corresponding methyl esters.
1a
1b
1c
1d
1e
1f
3.2
2a
2b
2c
2d
2e
2f
–
14
38
–
–
123
271
326
88
66
61
66
a Activity referred to butyronitrile.
b Conversion is defined as: conversion = [2] + [3]/[1]0.
The A. thaliana nitrilase-mediated hydrolysis of glu-
taronitriles 5a–e is shown in Scheme 3 and summarized in
Table 2.
The w-cyanocarboxylic acids 2, required as reference
compounds, were prepared as follows: 2b10 and 2c11 were
obtained by reacting b-propiolactone and g-butyrolac-
tone, respectively, with sodium cyanide. The monoacids
2d–f were prepared by enzymatic hydrolysis as described
on a preparative scale.
Selective Hydrolysis of b-Substituted Glutaronitriles
In order to clarify if the hydrolytic selectivity of the nit-
rilase is restricted to a,w-dicyanoalkanes, substituted glu-
Scheme 3
Synthesis 2001, No. 12, 1866–1872 ISSN 0039-7881 © Thieme Stuttgart · New York

1868
F. Effenberger, S. Oßwald
PAPER
Table 2 Selective Hydrolysis of Glutaronitriles 5 to 4-Cyanobu-
tanoic Acid Derivatives 6 Catalyzed by a Nitrilase from Arabidopsis
thaliana with 100% Conversion
Substrate Rel. Activity 4-Cyanobutanoicacids 6
5
(%)a
Selectivity Yield (%)
(%)
5a
5b
5c
5d
5e
2.3
6a
6b
6c
6d
6e
>99
>99
>99
–
53
41b
29b
–
9.7
0.5
<0.1
<0.1
Scheme 4
–
–
a Activity referred to butyronitrile.
b Isolated as methyl ester (see experimental part).
Table 3 Selective Hydrolysis of 3-Cyanopimelonitrile 7 and its De-
rivatives 8a,b Catalyzed by a Nitrilase from Arabidopsis thaliana
Substrate Rel. Activity Conversion Product
Selectivity (%)
As can be seen from Table 2, the b-substituted glutaroni-
triles 5 were found to be poor substrates for A. thaliana.
The nitrilase activity and thus acceptance of compounds 5
as substrates was strongly influenced by the steric demand
of the substituents in b-position. The hydrolysis rates of
5a and 5b are comparable, 5c, however, is converted 19
times slower than 5b, whereas 5d and 5e are not accepted
(%)a
271
167
1.3
(%)
83
1e
7
2e
8a
9a
98.9
96.7
100
77
8a
96
as substrates any more. In all cases, the enzymatic hydrol- 8bb
ysis of the dinitriles 5 proceeded selectively to give only
131
100
10
100
a Activity referred to butyronitrile.
the corresponding 4-cyanobutanoic acids 6.
b Hydrolysis of a mixture 8b/9b.
Taking into account the results from the hydrolysis of
dinitriles 1a–d and 5a–c, A. thaliana nitrilase hydrolyzes
dinitriles but not the corresponding cyanocarboxylic ac-
ids. Due to almost complete deprotonation of the acid
function at pH 8.5, we assume that this chemoselectivity
arises from a repulsion of the carboxylate group and the
hydrophobic binding site of the enzyme.
Like the unsubstituted pimelonitrile (1e), compound 7 is
also an excellent substrate for A. thaliana. Compound 7
was converted significantly faster than the reference com-
pound butyronitrile (Table 3). The specific enzyme activ-
ity was diminished by about 40% compared to 1e,
indicating less influence of the secondary cyano group on
the reaction rate. The selectivity of hydrolysis of 7 is less
than that of 1e with 96.7% of 3,5-dicyanoheptanoic acid
(8a) and 3.3% of diacid 9a formed at 77% conversion.
Compound 8a crystallized when the purity was at least
90%. Terminating the hydrolysis of 7 at conversions be-
low 90% therefore yielded pure 8a after recrystallization
from petroleum ether/ethyl acetate.
The selective hydrolysis of prochiral 3-substituted glut-
arate diesters using hydrolytic enzymes is a common
method to prepare the corresponding chiral monoesters
with high enantiomeric excesses.14 Recently, prochiral 3-
substituted glutaronitrile derivatives were subjected to hy-
drolysis with a nitrile hydratase/amidase system to afford
enantioselectively the optically active cyanobutanoic ac-
ids.15 The enantioselectivity of A. thaliana nitrilase, how-
ever, is relatively low. The cyanobutanoic acids 6a–c
were obtained from 5a–c with enantiomeric excesses of
only 13–60%, neither sufficient for practical purposes nor
for an assignment of the enantiomers.
At higher enzyme concentrations and with longer reaction
times, 7 was hydrolyzed completely to 3-cyanopimelic
acid (9a). Although monoacid 8a thereby formed as inter-
mediate was a markedly poorer substrate, as can be seen
from the enzyme activity (Table 3), it was converted ex-
clusively to 9a. Since, as mentioned above, the carboxy-
late form of 8a dominating at pH 9 was assumed to be
responsible for the selectivity, a mixture of 8a/9a (95:5),
yielded by the enzymatic hydrolysis of 7 at 95% conver-
sion, was esterified with diazomethane17 to give the meth-
yl esters 8b/9b. From this mixture monoester 8b was
hydrolyzed about 100 times faster than 8a (Table 3) ex-
clusively to methyl hydrogen 3-cyanoheptanedioate (10),
while diester 9b remained unchanged under these reaction
conditions (Scheme 4). This result was confirmed using
Selective Hydrolysis of g-Cyanopimelonitrile
The hydrolysis of g-cyanopimelonitrile (7),16 containing
cyano groups on primary and secondary C-atoms, offers
further possibilities to investigate the regiospecificity and
the selectivity of A. thaliana (Scheme 4, Table 3).
Synthesis 2001, No. 12, 1866–1872 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Selective Hydrolysis of Aliphatic Dinitriles by Nitrilase
1869
concd HCl (10 mL). Approximately 100 mL of EtOH were distilled
off, and the aqueous layer was extracted with Et2O (3 ¥ 100 mL).
The combined organic layers were dried (Na2SO4), and concentrat-
ed. Distillation through a Vigreux column under high vacuum gave
28.53 g (71%) of 4 as a colorless oil; bp 101–103 °C/0.001 torr.
1H NMR (250 MHz): d = 1.26–1.38 (m, 6 H, 2 CH3), 2.50–2.73 (m,
3 H, CH, CH2), 3.61 and 3.73 (d, 1 H, J = 4.9 Hz, CH, diastereomer-
ic mixture).
13C NMR (63 MHz): d = 13.99 (OCH2CH3), 16.53, 17.71, 21.72,
22.72 (C-2,3), 31.73, 31.86, 42.50, 42.63 (CH3, C-4), 63.50
(OCH2CH3), 113.97, 114.30 (C-1), 116.92, 117.00 (C-5), 164.45
(CO2Et).
isolated 9b for hydrolysis. Compound 9b could be sepa-
rated from 10 by extraction with base. In the reaction se-
quence shown (Scheme 4), cyanoheptanedioic acid
monoester 10 was isolated in 74% total yield without ne-
cessity to isolate and purify the intermediates. In no case,
hydrolysis of the secondary cyano group was observed.
The enantioselectivity of A. thaliana nitrilase in the hy-
drolysis of 7 is even lower than in the hydrolysis of 5a–c,
yielding 8a and 10 with <10% ee.
The experimental results clearly demonstrate that the A.
thaliana nitrilase exhibited not only regiospecificity in the
case of dinitrile but also selectivity for cyano groups in
presence of ester functions. Comparable selectivity has
been described for the nitrilase from Rhodococcus ATCC
39484, which converts benzonitrile but not methyl ben-
zoate.18
Anal. Calcd for C9H12N2O2: C, 59.99; H, 6.71; N, 15.55. Found: C,
60.15; H, 6.78; N, 15.51.
b-Methylglutaronitrile (5a)
Butanoate 4 (17.16 g, 95.2 mmol) was slowly added dropwise to
ice-cold aq NaOH (5%, 100 mL). After stirring for 75 min at r.t., the
mixture was acidified with concd HCl (ice-cooling) and extracted
with EtOAc (3 ¥ 100 mL). The combined extracts were dried
(MgSO4), and concentrated. The crude product was dissolved in py-
ridine (100 mL) and heated at reflux for 4 h. The solution was con-
centrated, and the residue was taken up in Et2O and washed with a
0.1 M solution of HCl. The organic layer was dried (Na2SO4), and
concentrated. Distillation through a Vigreux column under vacuum
gave 4.33 g (42%) of 5a as a colorless oil; bp 133–134 °C/15 torr.
Conclusion
In summary, among the nitrile-hydrolyzing enzymes in-
vestigated so far the nitrilase from Arabidopsis thaliana
described shows the highest selectivity in the hydrolysis
of aliphatic a,w-dini-triles to cyanocarboxylic acids
known today. Thereby the enzyme activity raises with in-
creasing chain length of the dinitrile. Moreover, this en-
zyme differentiates chemically non-equivalent cyano
groups, hydrolyzing exclusively the cyano groups on pri-
mary C-atoms. In contrast to many other nitrilases this en-
zyme is stable under the applied reaction conditions.
1H NMR (250 MHz): d = 1.26 (d, 3 H, J = 6.6 Hz, CH3), 2.21–2.41
(m, 1 H, CH), 2.49 (d, 4 H, J = 6.1 Hz, 2 CH2).
13C NMR (63 MHz): d = 19.02 (CH3), 23.58 (C-3), 28.25 (C-2,4),
117.13 (CN).
Anal. Calcd for C6H8N2: C, 66.64; H, 7.46; N, 25.90. Found: C,
66.62; H, 7.70; N, 25.84.
The A. thaliana nitrilase-catalyzed hydrolysis, starting
from readily available di- and trinitriles, therefore opens
access to a wide range of cyanocarboxylic and dicyano-
carboxylic acids, which cannot be obtained by chemical
hydrolysis. Since cyanocarboxylic acids are of interest for
the preparation of other important compounds such as
amino carboxylic acids and the corresponding lactams,
this new method might have interesting synthetic poten-
tial.
b-Methoxyglutaronitrile (5c)
To a suspension of NaH (0.7 g, 29.2 mmol) in anhyd THF (40 mL)
at 0 °C under inert gas atmosphere was slowly added dropwise a so-
lution of 5b12 (3.0 g, 27.2 mmol) in anhyd THF (10 mL) followed
by addition of neat dimethyl sulfate (3.61 g, 28.6 mmol), and the
mixture was stirred for 15 min at 0 °C. After stirring for 6 h at r.t.,
the mixture was poured onto ice (100 mL)/1 M solution of HCl (50
mL), and extracted with Et2O (3 ¥ 100 mL). The combined extracts
were dried (Na2SO4), and concentrated. The residue was chromato-
graphed on silica gel with petroleum ether–EtOAc (80:20) to give
2.13 g (63%) of 5c.
1H NMR (250 MHz): d = 2.73 (d, 4 H, J = 5.6 Hz, CH2), 3.50 (s, 3
Melting points were determined on a Büchi SMP-20 and are uncor-
rected. Unless otherwise stated, 1H NMR spectra were recorded on
a Bruker AC 250 F (250 MHz) and ARX 500 (500 MHz) in CDCl3
with TMS as internal standard. Chromatography was performed us-
ing silica gel S (Riedel-de Haën), grain size 0.032–0.063 mm. GC
separations were conducted using capillary glass columns (20 m ¥
0.32 mm) with OV 1701, carrier gas hydrogen. GC determination
of enantiomeric excess was performed with a Chiraldex G-TA
(ICT) column (30 m ¥ 0.32 mm), carrier gas hydrogen. All solvents
were dried and distilled. The recombinant nitrilase from Arabidop-
sis thaliana was obtained after overexpression in Escherichia coli
as described elsewhere.8,9 Petroleum ether used refers to the fraction
boiling at 30-75°C.
H, OCH3), 3.82 (quin, 1 H, CH).
13C NMR (63 MHz): d = 22.50 (C-2,4), 58.13 (OCH3), 72.65 (C-3),
115.79 (CN).
Anal. Calcd for C6H8N2O: C, 58.05; H, 6.50; N, 22.57. Found: C,
57.91; H, 6.60; N, 22.57.
b-Acetoxyglutaronitrile (5d)
A solution of 5b12 (2.0 g, 18.2 mmol), Ac2O (3.71 g, 36.3 mmol)
and pyridine (2.15 g, 27.2 mmol) in CH2Cl2 (25 mL) was heated at
60 °C for 19 h. The reaction mixture was then taken up in Et2O (100
mL), washed with a 1 M solution of HCl (2 ¥ 25 mL) followed by
aq sat. solution of NaHCO3 (2 ¥ 25 mL), dried (Na2SO4), and con-
centrated. The residue was chromatographed on silica gel with pe-
troleum ether–EtOAc (50:50) to give 2.53 g (92%) of 5d.
Ethyl 2,4-Dicyano-3-methylbutanoate (4)
To sodium (1.03 g, 44.8 mmol) in anhyd EtOH (150 mL) was added
portionwise ethyl cyanoacetate (32.88 g, 290.7 mmol). Crotononi-
trile (15.0 g, 223.6 mmol) was slowly added dropwise (ice-cooling),
and the reaction mixture was stirred at r.t. for 2 h. After heating at
60 °C for 5 h, the mixture was poured onto the same volume of ice/
1H NMR (250 MHz): d = 2.73 (d, 4 H, J = 5.6 Hz, CH2), 3.50 (s, 3
H, CH3CO), 3.82 (quin, 1 H, CH).
13C NMR (63 MHz): d = 20.63 (CH3CO), 22.27 (C-2,4), 64.24 (C-
3), 114.87 (CN), 169.50 (CH3CO).
Synthesis 2001, No. 12, 1866–1872 ISSN 0039-7881 © Thieme Stuttgart · New York

1870
F. Effenberger, S. Oßwald
PAPER
Enzymatic Hydrolysis of Dinitriles 1 and 5 on a Preparative
Anal. Calcd for C7H8N2O2: C, 55.26; H, 5.30; N, 18.41. Found: C,
55.27; H, 5.48; N, 18.33.
Scale; General Procedure
To a solution of crude enzyme in Tris/HCl buffer (70 mM, pH 8.5)
at r.t. was added the respective dinitrile 1 or 5 (Table 4), and the
mixture was stirred for the time given in Table 4. Then the mixture
was acidified to pH 4 with a 5 M solution of HCl and extracted with
Et2O. The combined organic layers were extracted twice with 10%
(v/v) sat. aq NaHCO3 solution. The aqueous layer was washed with
the double volume of Et2O, acidified to pH 4 and extracted with
Et2O. The combined extracts were dried (Na2SO4), and concentrat-
ed. Crude cyanocarboxylic acids 2d, 2e, and 6a were purified by
distillation under vacuum, 2f by recrystallization from petroleum
ether/EtOAc. Compounds 6b and 6c were reacted with
diazomethane17 to give the corresponding methyl esters and chro-
matographed on silica gel with petroleum ether–EtOAc (60:40 and
70:30, respectively). For physical data see Table 5.
Crude Extract Preparation and Enzymatic Hydrolysis of
Dinitriles 1, 5, and 7 with a Nitrilase from Arabidopsis thaliana;
General Procedure
Cells of E. coli, grown in LB medium supplemented with ampillicin
and kanamycin at 30 °C and induced with isopropyl-b-D-thiogalac-
topyranoside,9 were separated from the nutrient medium by centrif-
ugation for 30 min at 4 °C (5700¥g) and washed with Tris/HCl
buffer (70 mM, pH 8.5). The pellet was resuspended in Tris/HCl
buffer (100 mL/10 g wet weight), sonicated with ice-cooling (3 ¥ 5
min), and centrifuged for 40 min (186000¥g). The supernatant con-
taining the crude enzyme was diluted with Tris/HCl buffer, pH 8.5
(to reach 15–40% conversion after 2–4 h), and a solution of the re-
spective dinitrile 1, 5, 7 in MeOH was added (10 mM end concen-
tration; the concentration of the stock solution must be as high as the
added volume and should not exceed 100 mL/5 mL reaction solu-
tion). After stirring for 2–4 h, samples of 1 mL volume were taken
and analyzed as described below.
3,5-Dicyanohexanoic Acid (8a)
Compound 7 (4.0 g, 27.2 mmol) was hydrolyzed as described above
with nitrilase (39.2 U) in Tris/HCl buffer (200 mL, 70 mM, pH 8.5).
After 9.5 h at 78% conversion of 7, the reaction was terminated. Re-
crystallization from petroleum ether–EtOAc (95:5) gave 3.25 g
(72%) of 8a as white needles; mp 45–46 °C.
1H NMR (250 MHz): d = 1.86–2.09 (m, 4 H, 2 CH2), 2.51–2.70 (m,
4 H, 2 CH2), 2.72–2.98 (m, 1 H, CH), 8.60 (br s, 1 H, CO2H).
Determination of Conversion by Gas Chromatography
The sample taken was acidified with a 5 M HCl solution (50 mL) and
extracted with Et2O (5 mL). After centrifugation (2000¥g) and cool-
ing at -30 °C for 30 min to freeze the aqueous layer, the organic
layer was decanted and a 0.2 M solution of diazomethane in Et2O17
was added. The excess of diazomethane and Et2O was distilled off
under vacuum. The residue was taken up in Et2O (1 mL), and the
conversion was determined by GC from the ratio area of methyl es-
ter to area of nitrile.
13C NMR (63 MHz): d = 15.34, 26.64, 28.09, 30.12, 30.96 (C-
2,3,4,5,6), 117.89, 119.59 (CN), 177.25 (CO2H).
Anal. Calcd for C8H10N2O2: C, 57.82; H, 6.07; N, 16.86. Found: C,
57.66; H, 6.15; N, 16.81.
Determination of the Enzyme Activity
Methyl 3-Cyanoheptane-1,7-dioate (9b)
To 5 mL of a solution of the enzyme in Tris/HCl buffer (70 mM, pH
8.5), warmed at 35 °C for 10 min, was added a 0.25 M solution of
3-phenylpropionitrile in MeOH (50 mL), and the mixture was stirred
at 35 °C for 1 h. A sample of 1 mL was worked up and analyzed by
GC as described for the determination of conversion. For the cali-
bration of peak areas 3-phenylpropionitrile (181.5 mg) and 3-phe-
nylpropionic acid (205.3 mg) were dissolved in MeOH/buffer (10
mL/990 mL). Amounts of 5 mL of each solution were mixed, and 5
mL from this mixture added to 5 mL of a solution of 3-phenylpro-
pionitrile. Samples of 1 mL from the respective dilution were treat-
ed and analyzed by GC as described. The conversion was calculated
from the ratio of peak areas of nitrile and acid considering the cali-
bration factor. One unit is defined as 1 mmol conversion per minute.
Compound 7 (0.22 g, 1.5 mmol) was hydrolyzed as described above
with nitrilase (78.4 U) in Tris/HCl buffer (25 mL, 70 mM, pH 8.5).
After 72 h at 100% conversion of 7 and 74% conversion of 8a, the
reaction was terminated. The mixture of 8a/9a was separated by
chromatography on silica gel with petroleum ether–EtOAc (85:15),
and reaction of 9a with diazomethane gave 0.18 g (57%) of 9b.
1H NMR (250 MHz): d = 1.89–2.06 (m, 4 H, 2 CH2), 2.44–2.66 (m,
4 H, 2 CH2), 2.76–2.88 (m, 1 H, CH), 3.71 (s, 6 H, 2 CO2CH3).
13C NMR (63 MHz): d = 27.32, 30.35, 31.19 (C-2,3,4), 51.95
(CO2CH3), 120.80 (CN), 172.52 (CO2CH3).
Anal. Calcd for C10H15NO4: C, 56.33; H, 7.09; N, 6.57. Found: C,
56.29; H, 7.13; N, 6.49.
Table 4 Preparative Enzymatic Hydrolysis of Dinitriles 1 and 5 to Cyanocarboxylic Acids 2 and 6
Substrates
Nitrilase
(U)
Buffer
(mL)
250
500
250
50
Time
(h)
Conversion
Products
2, 6
2d
1, 5
1d
1e
g (mmol)
3.78 (35.0)
6.11 (50.0)
2.72 (20.0)
1.00 (9.2)
1.00 (9.1)
0.25 (2.0)
(%)
100
73
Yield (g)
3.92
27.7
55.4
27.7
68.0
13.1
68.0
19
6.75
8.5
48
2e
4.67
1f
94
2f
1.91
5a
5b
5c
92
6a
0.62
50
50
97
6b
0.53a
0.09a
50
48
83
6c
a Isolated as methyl ester.
Synthesis 2001, No. 12, 1866–1872 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Selective Hydrolysis of Aliphatic Dinitriles by Nitrilase
1871
Table 5 Physical and Spectroscopic Data of Cyanocarboxylic Acids 2 and 6
Producta Appearance
Bp
(°C/0.001
torr)
1H NMR (CDCl3/TMS,
500 MHz)
d, J (Hz)
13C NMR (CDCl3/TMS,
125.8 MHz)
d
2d
2eb
2f
colorless oil
colorless oil
101–103
1.70–1.83 (m, 4 H, 2 CH2), 2.36–2.47 (m, 4
H, 2 CH2), 10.28 (br s, 1 H, CO2H)
17.00 (C-5), 23.58, 24.69 (C-3,4), 33.05 (C-
2), 119.30 (CN), 179.05 (CO2H)
98–99
1.54–1.76 (m, 6 H, 3 CH2), 2.34–2.42 (m, 4
H, 2 CH2), 10.32 (br s, 1 H, CO2H)
17.01 (C-6), 23.78, 25.07, 28.03 (C-3,4,5),
33.63 (C-2), 119.54 (CN), 179.54 (CO2H)
colorless crystals 38–39c
1.33–1.48 (m, 4 H, 2 CH2), 1.60–1.67 (m, 4
17.46 (C-7), 24.64, 25.51, 28.53, 28.66 (C-
H, 2 CH2), 2.29–2.38 (m, 4 H, 2 CH2), 11.50 3,4,5,6), 34.21 (C-2), 120.06 (CN), 180.32
(br s, 1 H, CO2H) (CO2H)
6a
colorless crystals 85–87
41–43c
1.00 (d, 3 H, J = 6.3, CH2), 2.09–2.63 (m, 5 H, 18.77 (CH3), 22.80 (C-4), 26.85 (C-2),
CH, 2 CH2), 12.39 (br s, 1 H, CO2H)
39.04 (C-3), 119.21 (CN), 172.76 (CO2H)
6bd
-
2.63–2.72 (m, 4 H, 2 CH2), 3.51 (br s, 1 H,
25.05 (C-4), 39.76 (C-2), 52.21 (CO2CH3),
OH), 3.75 (s, 3 H, CO2CH3), 4.33–4.39 (m, 1 64.11 (C-3), 116.92 (CN), 172.01 (CO2CH3)
H, CH)
6cb,d
-
2.56–2.78 (m, 4 H, 2 CH2), 3.43 (s, 3 H,
22.49 (C-4), 38.35 (C-2), 52.07 (CO2CH3),
OCH3), 3.72 (s, 3 H, CO2CH3), 3.87–3.97 (m, 57.78 (OCH3), 73.29 (C-3), 116.94 (CN),
1 H, CH
170.66 (CO2CH3)
a Satisfactory microanlyses or HRMS obtained: C 0.30, H 0.10, N 0.28.
b 250 MHz spectrum.
c Mp.
d Characterized as methyl ester.
Methyl Hydrogen 3-Cyanoheptane-1,7-dioate (10)
References
Compound 7 (1.77 g, 12.0 mmol) was hydrolyzed as described
above with nitrilase (39.2 U) in Tris/HCl buffer (100 mL, 70 mM,
pH 8.5). After 17 h at 95% conversion of 7, the reaction was termi-
nated. The mixture of 8a/9a was reacted with diazomethane. The re-
sulting mixture of methyl esters 8b/9b was hydrolyzed with nit-
rilase (39.2 U) in Tris/HCl buffer (100 mL, 70 mM, pH 8.5) for 48
h. Workup as described above for the enzymatic hydrolysis gave
1.77 g (74%) of 10 as a colorless oil.
(1) Enzyme catalyzed reactions, Part 41. Part 40: Effenberger,
F.; Oßwald, S. Tetrahedron: Asymmetry 2001, 12, 279.
(2) Oßwald, S. Ph.D. Dissertation; University of Stuttgart:
Germany, 2000.
(3) Gavagan, J. E.; Fager, S. K.; Fallon, R. D.; Folsom, P. W.;
Herkes, F. E.; Eisenberg, A.; Hann, E. C.; DiCosimo, R. J.
Org. Chem. 1998, 63, 4792.
1H NMR (500 MHz): d = 1.87–2.02 (m, 4 H, 2 CH2), 2.49–2.68 (m,
4 H, 2 CH2), 2.81–2.87 (m, 1 H, CH), 3.75 (s, 3 H, CO2CH3), 8.40
(br s, 1 H, CO2H).
(4) (a) Bengis-Garber, C.; Gutman, A. L. Tetrahedron Lett.
1988, 29, 2589. (b) Bengis-Garber, C.; Gutman, A. L. Appl.
Microbiol. Biotechnol. 1989, 32, 11.
(5) Kobayashi, M.; Nagasawa, T.; Yamada, H. Eur. J. Biochem.
1989, 182, 349.
(6) Kobayashi, M.; Nagasawa, T.; Yamada, H. Appl. Microbiol.
Biotechnol. 1988, 29, 231.
(7) Cerbelaud, E.; Bontoux, M.; Foray, F.; Faucher, D.; Lévy-
Schil, S.; Thibaut, D.; Soubrier, F.; Crouzet, J.; Pétré, D. D.
Ind. Chem. Libr. 1996, 8, 189.
(8) Bartling, D.; Seedorf, M.; Mithöfer, A.; Weiler, F. W. Eur.
J. Biochem. 1992, 205, 417.
13C NMR (125.8 MHz): d = 26.97, 27.24, 30.13, 30.24, 31.13 (C-
2,3,4,5,6), 52.00 (CO2CH3), 120.64 (CN), 172.61 (CO2CH3),
177.58 (CO2H).
HRMS (CI, 70 eV): m/z Calcd for C9H14NO4 (MH+): 200.09228.
Found: 200.09247.
MS (CI): m/z (%) = 399.2 (3) [2 M + H]+, 381.1 (4) [2 M - H2O +
H]+, 200.1 (43) [MH]+, 182.1 (100) [MH - H2O]+.
(9) Oßwald, S.; Wajant, H.; Effenberger, F., in preparation.
(10) Gresham, T. L.; Jansen, J. E.; Shaver, F. W.; Frederick, M.
R.; Fiedorek, F. T.; Bankert, R. A.; Gregory, J. T.; Beears,
W. L. J. Am. Chem. Soc. 1952, 74, 1323.
(11) Pichat, L.; Mizon, J.; Herbert, M. Bull. Soc. Chim. Fr. 1963,
1792.
Acknowledgement
We would like to thank the Fonds der Chemischen Industrie for fi-
nancial support, and Dr. A. Baro for preparing the manuscript.
(12) Johnson, F.; Panella, J. P.; Carlson, A. A. J. Org. Chem.
1962, 27, 2241.
(13) Ghosez, J. Bull. Soc. Chim. Belg. 1932, 41, 477.
Synthesis 2001, No. 12, 1866–1872 ISSN 0039-7881 © Thieme Stuttgart · New York

1872
F. Effenberger, S. Oßwald
PAPER
(14) (a) Lam, L. K. P.; Lui, R. A. H. F.; Jones, J. B. J. Org. Chem.
1986, 51, 2047. (b) Roy, R.; Rey, A. W. Tetrahedron Lett.
1987, 28, 4935. (c) Nakada, M.; Kobayashi, S.; Ohno, M.;
Iwasaki, S.; Okuda, S. Tetrahedron Lett. 1988, 29, 3951.
(d) Santaniello, E.; Chiari, M.; Ferraboschi, P.; Trave, S. J.
Org. Chem. 1988, 53, 1567. (e) Boland, W.; Frößl, C.;
Lorenz, M. Synthesis 1991, 1049.
(15) Maddrell, S. J.; Turner, N. J.; Kerridge, A.; Willetts, A. J.;
Crosby, J. Tetrahedron Lett. 1996, 37, 6001.
(16) Zahn, H.; Schäfer, P. Chem. Ber. 1959, 92, 736.
(17) deBoer, T. J.; Backer, H. J. Org. Synth. Coll.Vol. 4, 1963,
250.
(18) Stevenson, D. E.; Feng, R.; Dumas, F.; Groleau, D.; Mihoc,
A.; Storer, A. C. Biotechnol. Appl. Biochem. 1992, 15, 283.
Synthesis 2001, No. 12, 1866–1872 ISSN 0039-7881 © Thieme Stuttgart · New York